Revealing the former bed of Thwaites Glacier using sea-floor bathymetry: implications for warm-water routing and bed controls on ice flow and buttressing

Abstract. The geometry of the sea floor immediately beyond Antarctica's marine-terminating glaciers is a fundamental control on warm-water routing, but it also describes former topographic pinning points that have been important for ice-shelf buttressing. Unfortunately, this information is often lacking due to the inaccessibility of these areas for survey, leading to modelled or interpolated bathymetries being used as boundary conditions in numerical modelling simulations. At Thwaites Glacier (TG) this critical data gap was addressed in 2019 during the first cruise of the International Thwaites Glacier Collaboration (ITGC) project. We present more than 2000 km 2 of new multibeam echo-sounder (MBES) data acquired in exceptional sea-ice conditions immediately offshore TG, and we update existing bathymetric compilations. The cross-sectional areas of sea-floor troughs are under-predicted by up to 40 % or are not resolved at all where MBES data are missing, suggesting that calculations of trough capacity, and thus oceanic heat flux, may be significantly underestimated. Spatial variations in the morphology of topographic highs, known to be former pinning points for the floating ice shelf of TG, indicate differences in bed composition that are supported by landform evidence. We discuss links to ice dynamics for an overriding ice mass including a potential positive feedback mechanism where erosion of soft erodible highs may lead to ice-shelf ungrounding even with little or no ice thinning. Analyses of bed roughnesses and basal drag contributions show that the sea-floor bathymetry in front of TG is an analogue for extant bed areas. Ice flow over the sea-floor troughs and ridges would have been affected by similarly high basal drag to that acting at the grounding zone today. We conclude that more can certainly be gleaned from these 3D bathymetric datasets regarding the likely spatial variability of bed roughness and bed composition types underneath TG. This work also addresses the requirements of recent numerical ice-sheet and ocean modelling studies that have recognised the need for accurate and high-resolution bathymetry to determine warm-water routing to the grounding zone and, ultimately, for predicting glacier retreat behaviour.