Investigations of the DInSAR derived grounding line migration in Antarctica induced by ocean tides

Kurzfassung

A grounding line is the boundary between ice resting on a bedrock and floating ice. The migration of the glacier grounding line is an important indicator of ice sheet stability in a warming climate. A retreat of the grounding line location (GLL) observed over long periods (years to decades) might indicate processes related to ice thinning at the surface or underneath of a glacier. Yet, grounding line in nature is not static in position and it is moving also at short time scale namely back and forth with the ocean tidal cycle. Due to the presence of ice shelves, about 76% of the coastline of Antarctica comprises grounding lines, along which the glacial ice experiences tidal flexure. The objective of this thesis is to summarize and analyze the observations of short-term changes of GLL and ocean tides at specific selected sites in Antarctica, and thus eventually find out how ocean tides contributes to short-term GLL migration. In this study, time series of grounding lines were manually delineated from Sentinel-1 A/B double difference interferograms acquired in 2020. Ocean tide level in each corresponding period was computed from the CATS2008 tide model with air pressure correction from NCEP reanalysis data. To support the analysis, bedrock topography and slope information was extracted from BEDMAP2 bed elevation map at the delineated GLL. In order to perform a quantitative analysis, novel algorithms were developed for generating a reference GLL and calculating GLL migration. The reference GLL generation algorithm generates a concave hull polygon around the GLL datasets, and then perpendicular lines (normals), intersecting GLL segments, along the medial axis of the polygon. Averaging the intersecting points of each normal, a reference line is created by connecting all averaged points. The migration of GLL was estimated using three methods: point-line method, box method and normals outputted when generating the reference GLL. The reference GLL generating algorithm is robust and the performance was satisfactory. The estimation of GLL migration with the different methods was in good agreement, which indicates they have similar level of robustness. In the investigated areas the estimated shortterm displacement was in the order of kilometers, which is the same as the long-term migration. This result indicates the importance of averaging the GLL, as a robust representation of the GLL in a certain period, when quantifying the long-term migration and drawing conclusions about grounding line retreat and ice thinning. Although significant short-term migration is observed, the observations considered in this study are insufficient to explain the relationship between ocean tides and GLL migration. Therefore, for future studies, considering additional information, in particularly assimilating numerical model like the elastic beam model, which contains information on the physical processes related to ice bending, becomes necessary in order to reveal the mechanism of the tidal induced GLL migration underneath the ice.