Collision of boundary currents: beyond a steady state

Abstract

Listen

Translate article

Along-shore migration of a western boundary current separation is investigated with a fully nonlinear primitive equation numerical model on an ƒ-plane. We consider separation due to a collision with an opposing current. It is known that, for such a case, stationary collision and separation is possible only for boundary currents with “balanced” transports, i.e. equal near-wall depths. We first compute the stationary flow and then change the transport of the opposing current and investigate the resulting time-dependent flow. Our numerical results demonstrate two important aspects. First, we have found that the transport of a poleward flowing inertial western boundary current is controlled by its interaction with an opposing current. Although there is presently no observational evidence of such a counter-intuitive situation, this unusual effect may be important for actual boundary currents. Secondly, our results suggest that, after a rapid period of geostrophic adjustment, the flow evolution can be described by the path equation for the separated current. An excellent agreement is found between the earlier analytical predictions based on integrated balances and the path equation theory (Lebedev and Nof, 1996) and our present numerical simulations. For instance, the previously predicted analytical migration formula C≈(g′H) 1 2 (D 1 2 − D 2 2) 24D 2 3 2 [where H is the undisturbed depth of the main poleward flowing current; D 1 is its depth near the wall (non-dimensionalized by H); and D 2 (< D 1) is the opposing current near-wall (non-dimensional) depth] is clearly verified by our numerical experiments. Application of the model to the South Atlantic confluence is discussed. It is suggested that the observed migrations of the separation latitude may be caused by the changes of the Brazil and Malvinas current transports.