Marginal Thermobaric Stability in the Ice-Covered Upper Ocean over Maud Rise

Abstract

Temperature (T) and salinity (S) profiles from the central Weddell Sea near the Maud Rise seamount measured during the 1994 Antarctic Zone Flux Experiment (ANZFLUX) have been analyzed for stability with respect to the thermobaricity, that is, the pressure dependence of thermal expansion rate. For many T?S profiles in the region ??, the difference between actual density (including the pressure contribution) and density of a water column with uniform temperature and salinity equal to that of the mixed layer, exhibits a maximum in the upper ocean within tens of meters of the mixed layer?pycnocline interface. Following work by K. Akitomo, if the mixed layer were to deepen and increase in density so that the ?? maximum coincided with the base of the mixed layer, the system would be thermobarically unstable and would overturn catastrophically. Thermobaric convection differs from convection driven by surface buoyancy flux (cooling and/or freezing) because once started, the production of turbulent mixing energy is derived from the water column instead of the surface, an important distinction in ice-covered oceans. A stability criterion is developed that considers the total sensible heat and latent heat of freezing required to drive a given T?S profile to thermobaric instability, and is mapped in the Maud Rise region. A simple upper-ocean model, combined with enthalpy conservation at the ice?water interface and driven by surface stress and ice heat conduction observed with a drifting buoy cluster left in place after the ANZFLUX manned drift stations, is used to assess the susceptibility of observed profiles to thermobaric instability as the winter advanced. In the model, roughly one quarter of the profiles become unstable by the end of August, and it is argued that this may account for extensive polynya-like features that appeared in satellite microwave imagery over Maud Rise in August 1994, shortly after completion of the ANZFLUX Maud Rise drift.