Atmospheric Boundary Layer over shelf water Polynyas in the Weddell Sea, Antarctica

The boundary layer over shelf water polynyas in the western Weddell Sea, Antarctic, is investigated by observations and model simulations. A shelf water polynya is forced by persistent wind or ocean current displacement of sea ice on the shore of an ice shelf. Along the east coast of the Antarctic Peninsula shelf water polynyas can open adjacent to the Ronne and Larsen ice shelf. The overarching purpose of this project is to increase the knowledge of atmospheric boundary layer processes and their parameterization in the area of shelf water polynyas in the Weddell Sea in the austral summer. For a better understanding of the physical processes in polynya areas we took observations of the sea ice parameters and of the atmospheric turbulent boundary layer during various offshore wind conditions. Observations were conducted with two aircraft in the scope of the joint airborne campaign JASPER, which was a collaboration of the British Antarctic Survey (BAS) and the German Alfred Wegener Institute for Polar and Marine Research (AWI). The airborne observations were carried out in the Antarctic summer months February and March.

The project had two main objectives: first, to compare and verify the boundary layer measurements of both aircraft against each other and second, to study the local polynya effects in the western Weddell Sea ice area. Airborne observations in the boundary layer give the possibility to resolve even small-sized polynyas, which normally cannot be resolved by common satellite measurements. The wind, temperature, and turbulent flux observations of both aircraft showed a good agreement. We investigated radiation, energy, and momentum fluxes, and the vertical boundary layer structure along cross-sections. Furthermore, vertical profiles of atmospheric variables were obtained over polynyas in different stages of their development and under various offshore wind conditions. The data analysis showed that differences in the polynya width and thickness of the nilas cover strongly influenced the turbulent fluxes and structure of the atmospheric boundary layer. Moreover, the observations clarify the influence of polynyas on the energy budget in Weddell Sea ice areas and their role in ice production: The Ronne Polynya region showed a large variability of fluxes along the horizontal flight tracks. Observations orthogonal to the polynya showed the internal boundary layer development starting at the Ronne Ice Shelf and developing over the polynya. We observed a strong decrease in surface and air temperature with distance from the Ronne Ice Shelf front and a decrease in heat fluxes with an increase in the sea surface albedo. The turbulent heat flux over the Ronne Polynya area was in general relatively large and with positive direction. This can be explained mainly by the cold and dry katabatic air which was advected form the Ronne Ice Shelf over the relatively warm surface of the Ronne Polynya area. There, we observed large areas of new sea ice production, although the campaign was carried out during summer. In the south western part of the Weddell Sea we observed, in contrast to the Ronne Polynya region, a case where relatively warm air was advected from the Larsen ice shelf towards the sea and smaller polynyas opened adjacent to the ice shelf. Such warm air can developed during Föhn events over the mountains of the Antarctic Peninsula. As the air descends on the leeward side of the mountain range of the Antarctic Peninsula the air is warmed by adiabatic compression. Under such warm air advection, where the air temperatures were above the freezing point, we observed a different boundary layer structure with relatively small fluxes over the polynya area and hardly any new sea ice production.

In a next step, we studied the spatial development of the atmospheric boundary layer by modelling the flow over a polynya with the non hydrostatic meso-scale model METRAS. The model results were validated against the observations of the two research aircraft. The validation of the model against the observations revealed that both mean and turbulent quantities can be well modelled when the transfer coefficients of heat and momentum agree well with observations