28 July 2014

Parameterization of Basal Friction Near Grounding Lines in a One-Dimensional Ice Sheet Model


Useful prediction of sea-level rise depends on accurate modeling of the Antarctic ice sheet.  The marine-based West Antarctic Ice Sheet, whose bed lies mostly below sea level, drains most of its grounded ice to the ocean via ice streams. The resolution of a marine ice-sheet model is dictated by the resolution needed to simulate migration of the ice-stream grounding line, where the ice is thin enough to begin floating. When the sub-glacial hydrological system near the grounding line is well connected to the ocean, the basal water pressure increases, reducing friction and allowing the ice to slide faster. This study focuses on how the connection between sub-glacial hydrology and the ocean affects ice-stream dynamics and changes the required model resolution. DOE-funded researchers developed and applied a one-dimensional, depth-integrated ice-sheet model in which connectivity is introduced in the basal-friction law via a parameter that varies between 0 (no connectivity) and 1 (full connectivity). Some simulations included a numerical grounding-line parameterization that is known to reduce numerical errors. A highly accurate Chebyshev numerical solution was used as a benchmark to evaluate the model performance. The results show that strong ocean connectivity not only speeds up ice flow near the grounding line, but also decreases the model error and the need for high resolution near the grounding line.  With strong connectivity, grid resolution of about 1 km is sufficient to accurately model grounding-line migration. Without this connectivity (or a grounding-line parameterization), fixed-grid models typically need a resolution of 200 m or less, implying much greater computational cost.  If these results extend to three-dimensional models, the impact could be significant. Adding a physically plausible parameterization of ocean connectivity to these models could give comparable accuracy at greatly reduced cost.



This work was supported by the Earth System Modeling and Scientific Discovery through Advanced Computing (SciDAC) programs funded by the US Department of Energy, Office of Science, Biological and Environmental Research and Advanced Scientific Computing Research. The Los Alamos National Laboratory is operated by the DOE National Nuclear Security Administration under Contract DE-AC52-06NA25396. The authors would like to thank Christian Schoof, Matt Hoffman, Steve Price and Ed Bueler for fruitful conversation. The authors also thank Rupert Gladstone, Frank Pattyn, an anonymous reviewer, and the editor, Olivier Gagliardini, for detailed comments that have improved the paper.