CSIRO Marine and Atmospheric Research
Past Seminars

Tuesday 7 October at 11.30 am (Tas time)

CSIRO Auditorium, Hobart

Andrew Kiss
Research School of Earth Sciences,
Australian National University

How potential vorticity constraints can force separation of
western boundary currents

It is often difficult to obtain realistic western boundary current (WBC) separation in ocean circulation models, and a better understanding of the underlying physics of this process is needed. A variety of features have been shown to influence or induce separation (eg a collision with another western boundary current, a change in sign of the wind stress curl, outcropping of isopycnals or an abrupt change in bottom topography or boundary shape). However I will show that separation also occurs in a simple barotropic single-gyre model lacking these features. Identifying the cause of this apparently unprovoked separation is of interest because the mechanism which operates in this "lowest common denominator" model may also apply in more complex and realistic models.

I will present an explanation for WBC separation in this model, based on potential vorticity (PV) considerations. Fluid parcels have their PV reduced by anticyclonic wind stress in the ocean interior, and this PV is then recovered in the WBC via lateral and bottom friction. The bulk of the WBC is anticyclonic, but under no-slip boundary conditions there is also a strongly cyclonic region against the western boundary. This relative vorticity distorts the PV contours in the WBC, delaying PV recovery in the anticyclonic region and producing early recovery in the cyclonic sublayer. When the relative vorticity is large, flow in the cyclonic sublayer acquires more PV than was lost in the interior, forcing a change in the WBC outflow structure to allow this excess PV to be dissipated. It is demonstrated that this dissipation must involve a change in sign of the Laplacian viscous term (even when bottom friction is also present), and that this in turn creates a PV structure which guides the outer WBC offshore as a separated jet. An adverse streamwise pressure gradient is also produced in the separation region. Under free-slip boundary conditions, the cyclonic sublayer is absent and this mechanism does not operate, resulting in a very different WBC outflow structure. A numerical model must correctly reproduce the vorticity magnitude in the cyclonic sublayer to capture separation by this mechanism - this requires that the horizontal viscosity and grid spacing are sufficiently small.

[Back to Seminars]

Legal Notice and Disclaimer | Copyright