Mesoscale Modeling of the Atmosphere over Antarctic Sea Ice: A Late-Autumn Case Study

Abstract

Abstract Atmospheric flow over Antarctic sea ice was simulated applying a polar version of the fifth-generation Pennsylvania State University–National Center for Atmospheric Research Mesoscale Model (Polar MM5). The simulation period in late autumn lasted for 48 h, starting as northerly warm airflow over the Weddell Sea ice cover and turning to a southwesterly cold-air outbreak. The model results were validated against atmospheric pressure and wind and air temperature observations made by five buoys drifting with the sea ice. Four different satellite-derived sea ice concentration datasets were applied to provide lower boundary conditions for Polar MM5. During the period of the cold-air outbreak, the modeled air temperatures were highly sensitive to the sea ice concentration: the largest differences in the modeled 2-m air temperature reached 13°C. The experiments applying sea ice concentration data based on the bootstrap and Arctic Radiation and Turbulence Interaction Study (ARTIST) algorithms yielded the best agreement with observations. The cumulative fetch over open water correlated with the bias of the modeled air temperature. The sea ice concentration data affected the simulated air temperature in the lower atmospheric boundary layer, but above it the temperature and wind fields were more strongly controlled by the boundary layer scheme applied in Polar MM5. Analysis nudging applying four-dimensional data assimilation had a positive effect on the pressure and wind fields but negative or no effect on the air temperature fields. The results suggest that applying a sea ice model to update sea ice fields frequently throughout atmospheric model simulations will likely lead to important improvements in forecasts.