MELANGE-PISM

Sea level rise is a major concern for many coastal communities worldwide. The Greenland Ice Sheet is expected to contribute significantly to sea level rise this century and beyond. About half of Greenland's contribution comes from calving outlet glaciers. Iceberg calving is mainly suppressed in winter when the proglacial sea ice mélange is thick and stiff, and begins in spring after the mélange breaks up. Therefore, mélange is an important factor in Greenland's contribution to sea level rise. Until now, however, only one ice sheet model is capable of explicitly modeling mélange. In the interest of comparability of model results, it is important that more ice sheet models can also model mélange. The Parallel Ice Sheet Model (PISM), used and co-developed by our group, is a well-established ice sheet model with a large user base worldwide. In a warming climate, we expect increased iceberg calving to feed proglacial mélange. At the same time, higher temperatures also lead to increased melting of the mélange. We want to investigate whether these two opposing processes lead to an overall increase in mélange, which would help stabilize glaciers, or to a decrease in mélange, which would further destabilize already retreating glaciers. In our proposed research project, we plan to implement a mélange model in our ice sheet model PISM and analyze the role of mélange for the ice loss from a Greenland glacier in a warmer climate. We intend to pursue three work packages: First, we plan to implement a mélange model component in PISM, following the example of the existing mélange model within an ice sheet model. Second, we intend to constrain the parameters and test the model in two different idealized settings, a rectangular channel and a synthetic marine ice sheet. Finally, we plan to apply our new mélange model to a regional simulation of Jakobshavn Isbrae, one of the largest outlet glaciers on the west coast of the Greenland Ice Sheet. We intend to first perform a historical spinup using an ensemble study with a Bayesian approach, and then perform future simulations for this century.