Mapping newly emerging islands and nunataks in Greenland improves existing bed elevation datasets

While great improvements in our understanding of the subglacial landscape has occurred in recent years, the majority of the land beneath the Greenland Ice Sheet is still unmapped. With this study we map newly emerging land masses using Landsat 8 and Planet Scope satellite imagery. By including new islands, nunataks, and ice-contact outcrops in the current bed elevation model BedMachine we are able to improve the reliability of both pro glacial bathymetry as well as subglacial topography in all areas where new land is emerging.

The previous official Greenland wide mapping occurred in 1978-1987 and was done on the basis of aerial photographs recorded at scale 1:150.000. Here, we manually update new island - and ice contact bedrock outcrops using false color pan-sharpened Landsat 8 (15m) from 2019 and verifying our results with Planet Scope satellite images (3m) from 2019 and 2021. With the mapping of newly emerged islands and bedrock outcrops, existing bathymetric and ice thickness products can be updated. As existing models (eg. BedMachineV3) is limited by the available input, it is common to see large assumed ice sheet thicknesses where nunataks are now exposed. Likewise, many newly mapped islands are appearing in places where fjord depths were expected to be several hundreds of meters.

While traditional methods fcor collecting bedrock elevations below the ice and ocean surfaces are associated with extremely high logistical costs, our approach can in a quick and affordable manner update existing med models with valuable data. This addition will result in more accurate ice flow and fjord circulation models, which will ultimately give us better predictions of future sea-level rise. We argue that with the ongoing retreat and downwasting, these systematic mapping efforts should ideally take place on a biannual basis.