Amundsen and Bellingshausen Seas simulation with optimized ocean, sea ice, and thermodynamic ice shelf model parameters

Recent studies suggest that the thickness of Winter Water (WW), that is, water with potential temperature below approximate to-1 degrees C located below Antarctic Surface Water and above Circumpolar Deep Water (CDW) is critical in determining the ice shelf melt rate, especially for the Pine Island Glacier (PIG). Existing model studies, however, misrepresent WW thickness and properties in the Amundsen Sea (AS). Here, we adjust a small number of model parameters in a regional Amundsen and Bellingshausen Seas configuration of the Massachusetts Institute of Technology general circulation model in order to reproduce properties and thickness of WW and CDW close to observations, with significant improvement for WW compared to previous studies. The cost, which is defined as weighted model-data difference squared, is reduced by 23%. Although a previous modeling study points out that the local surface heat loss upstream from Pine Island Polynya could be the reason for the observed 2012 PIG melt decline and WW thickening, they did not show WW freshening, which was observed at the same time. Model sensitivity experiments for surface heat loss, PIG melt rate, and precipitation fail to replicate WW freshening concurrent with PIG melt decline, implying that these processes cannot fully explain the observed PIG melt decrease.