Open-ocean convection in the Greenland Sea: preconditioning through a mesoscale chimney and detectability in SAR imagery studied with a hierarchy of nested numerical models
Androsov, Alexey; Rubino, Angelo; Romeiser, Roland; Sein, Dmitry V.
Aspects of the hydrodynamics of the Greenland Sea were investigated through a hierarchy of nested numerical models. The simulations were particularly conceived to study, under realistic conditions, the hydrodynamics induced by the presence of a convectively generated oceanic mesoscale chimney as well as its "long-term" influence on the local convective activity. To this purpose, a very high resolution, fully non-hydrostatic 3D model capable of simulating submesoscale convective vertical plumes was nested into an ocean-ice, regional hydrostatic 3D model which was initialised and forced through the global, coupled atmosphere ocean 3D REMO/MPI-OM model. In the central part of the Greenland Sea, the hydrological structure of an observed, convectively generated oceanic mesoscale chimney and a corresponding reconstructed velocity field were imposed as a part of the forcing for the non-hydrostatic numerical model. Two different, "short-term" realistic scenarios were simulated corresponding, respectively, to episodes characterized by a strong mean oceanic heat loss and by a weak mean oceanic heat gain in the central Greenland Sea. In order to evaluate the role played by mesoscale convective chimneys in promoting preconditioning to open-ocean deep-penetrating convection, two "long-term" simulations of the hydrodynamics of the Greenland Sea were performed using the same model hierarchy and the forcing as described above. The two runs differed merely in that only in one of them the hydrological and velocity structure of a convective chimney were inserted in the central Greenland Sea as a part of the forcing. The dependence of simulated surface convergence patterns on grid step in the central Greenland Sea was also investigated in order to assess the capability of numerical models of predicting the detectability of convective events in synthetic aperture radar imagery.