A simple inverse method for the distribution of basal sliding coefficients under ice sheets, applied to Antarctica

Variations in intrinsic bed conditions that affect basal sliding, such as the distribution of deformable sediment versus hard bedrock, are important boundary conditions for large-scale ice-sheet models, but are hard to observe and remain largely uncertain below the modern Greenland and Antarctic ice sheets. Here a very simple model-based method is described for deducing the modern spatial distribution of basal sliding coefficients. The model is run forward in time, and the basal sliding coefficient at each grid point is periodically increased or decreased depending on whether the local ice surface elevation is too high or too low compared to observed in areas of unfrozen bed. The method considerably reduces large-scale errors in Antarctic ice elevation, from several 100s to several 10s of meters in most regions. Remaining ice elevation errors over mountain ranges such as the Transantarctics are further improved by parameterizing the possible effect of sub-grid topography in the basal sliding law, representing sliding in deep valleys. Results are compared with modern velocity data, and various sensitivity tests are described in Appendices