Mapping the Bed Topography and Ice Thickness of Lomonosovfonna Ice Cap (Svalbard) Using Radar Data and Inverse Ice-Dynamical Modelling

Sammanfattning

Under the current climate warming, Svalbard is one of the fastest warming places in the world due to Arctic amplification and the glaciers experience mass loss. As well as having an important role in both marine and terrestrial ecosystems, glaciers are major terrestrial freshwater reservoirs and contributors to sea level rise. Therefore, estimates of ice volume, ice thickness distribution and bed topography are important to make predictions of their future evolution. Ice thickness and bed elevation measurements of most glaciers are sparse compared to the number of datasets describing glacier surfaces. Therefore, applying inverse methods to surface observation data enables the inference of bed conditions and can be used for estimating ice thickness and bed topography. In this thesis, the higher-order Instructed Glacier Model is used to infer distributed ice thickness and bed topography maps for Lomonosovfonna ice cap and the five connected tidewater glaciers with a spatial resolution of 100 m. A total ice volume of 567.5 km3 was estimated and 1.35 mm SLE was calculated from the ice located above sea level. A total of 179 future lakes were predicted from the inferred bed topography with a total volume of approximately 50 km3 for the 149 lakes with surface elevation above sea level. Three new ground penetrating radar transects were processed and combined with other existing ice thickness data and used for both optimization and validation of the inversion. To see how sensitive the model is to surface velocity input data, two versions were tested, one from 2013-2015 and one from 2017-2018. The result was further used for a comparison to existing ice thickness products and shows the overall best fit to the observed ice thickness data with a small bias and good fit for both small and large thicknesses.