The two-layer skirted island

Abstract

The flow around a planetary scale island in a baroclinic ocean is examined when the island possesses a topographic skirt representing a steep continental slope and the ocean is modeled as a two-layer system in order to examine the role of stratification in the circulation. The study extends an earlier barotropic model of similar geometry and forcing to focus on the degree to which the topography, limited here to the lower of the two layers, affects the circulation and to what degree the circulation is shielded by stratification from the topographic effects noted in the simpler barotropic model.

As in the barotropic model, the topography is steep enough to produce closed, ambient potential vorticity contours over the topography in the lower layer providing free "highways" for the deep flow in the presence of small forcing by the wind-driven upper layer flow. The flow is very weak outside the region of closed contours but can become of the same order, if somewhat smaller, as the upper layer flow on those contours in the presence of even weak coupling to the upper layer.

A series of models, analytical and numerical, are studied. Linear theory is applied to two configurations. The first consists of a long, meridionally oriented island with a topographic skirt in the lower layer. The lower layer flow is driven by a hypothesized frictional coupling between the two layers that depends on the circulation of the upper layer velocity on a circuit defined by the closed potential vorticity contours of the lower layer. The largest part of the driving flow is identical on both sides of the island and cancels in the contour integration. The major part of the residual forcing comes from relatively small but effective forcing on the semi-circular tips of the topographic skirt. A circular island with a topographic skirt is also examined in which the coupling to the upper layer is stronger all around the island. Even in this case there is a delicate balance of the forcing of the lower layer on each side of the island. In all cases the flow on closed potential vorticity contours in the lower layer is much weaker than in the barotropic model but much stronger than in the flat region of the lower layer.

A sequence of numerical calculations that both check and extend the analytic linear theory is presented demonstrating the subtlety of the force balances. Further nonlinear, eddy-containing experiments give a preview of the direction of future work.