Improved Antarctic Sub-ice Bed Topography from Ice Surface Morphology and Velocity

Abstract:

The subglacial topography of Antarctica is critical input data for ice sheet models and for understanding the geological, geomorphic and glacial history of the continent. Recent compilations of direct measurements using radar and gravity, published as Bedmap (version 2 is current), have improved the coverage, resolution and visualization of the sub-ice-sheet bedrock, but many regions remain poorly mapped. Here we present a new technique that uses continent-wide ice sheet surface morphology from satellite imagery data (a series of MODIS mosaics), surface topography, and surface ice flow velocity to estimate the elevation of the Antarctic bed. The analysis is based upon an inversion of the driving stress equation with corrections for temperature and plasticity. We test a number of versions of the input data sets at various resolutions to find data and scaling that best replicate well-mapped areas of Bedmap2, or match Radio Echo Sounding data not incorporated in Bedmap. Results comparing Bedmap2 cells over East Antarctica to a modelled output based upon the MOA mosaic and ice velocity, show a 0.81 r² correlation with a standard deviation of 898m. The technique works best in areas of moderate flow and thin to moderate ice thickness. In such areas, especially where present RES data is lacking or sparse, the model provides hitherto unrevealed detail. These data could be used to supplement existing Bedmap models, improve geomorphological knowledge of the continent, target areas for detailed field survey, and inform field survey design. The Recovery Glacier region, highlighted below, is an example of such an area.