Evaluating DInSAR for measuring seasonal elevation dynamics at the Vissátvuopmi palsa mire complex in Northern Sweden

Abstract

During recent years the palsa mires of Fennoscandia have reported to be undergoing extensive lateral degradation, and in a warming environment the palsa mires of Sweden among others are threatened to thaw completely by the end of this century. Since subarctic peatlands account for 30% of the global soil organic carbon storage, small degradation changes in these areas globally, can potentially change the greenhouse gas emissions on a worldwide scale. Therefore it is important to predict future changes in these peatlands. The aim of this thesis was to evaluate use of Sentinel-1 data and the DInSAR method to create a Digital Elevation Model (DEM) and to quantify seasonal displacement dynamics over individual palsas in the Vissátvuopmi palsa mire in northern Sweden. The results showed that creation of a DEM was not possible over the area. The seasonal displacement dynamics over the two individual palsas in this study varied from a minimum of -0.03 cm to a maximum of 3.30 cm over the snow-free months for each of the years from 2019 to 2021. There were no consistent patterns in seasonal displacement dynamics, mainly due to the high variation between the displacements values observed. When analyzed in relation to temperature and precipitation, the best relationship with displacement was with mean temperature (R2 = 0.5238), and showed a negative trend. SAR data are highly sensitive to precipitation and snow, which limits the useful data to only two and a half months per year (from July to mid-September) in a best case scenario. It was difficult to use the DInSAR method for the initial purpose of this thesis, however, other analysis approaches (e.g. ISBAS method) with InSAR data may be useful in future studies. For a more detailed investigation of palsa elevation changes and degradation, I would suggest other higher spatial resolution data instead, such as DEMs created from drone imagery data.