PROCESSES COUPLING THE UPPER AND DEEP OCEAN

IN BAROCLINC FRONTS OVER TOPOGRAPHY

(Sutyrin, G., I. Ginis, and S. A. Frolov, 2001: Equilibration of baroclinic meanders and deep eddies in a Gulf-Stream-type jet over a sloping topography, J. Phys. Oceanogr. In print. You can download the paper and figures: paper,fig1,fig2,fig3,fig4a,fig4b,fig5,fig6,fig7,fig8,fig9,fig10,fig11,fig12b,fig12b…)

Our goal in this project is to achieve deeper understanding of physical processes of vertical coupling over topography and apply it to the development of more physically accurate model initialization and data assimilation techniques for ocean nowcasting and short and medium range forecasting.

Our primary task is to develop reliable predictive methods for the Gulf Stream meanders and eddies over the continental slope/rise.

We seek to answer the following three fundamental questions:

How are Gulf Stream meanders downstream of Cape Hatteras modified over the continental slope?

What are the effects of dynamical coupling between amplifying meanders and associated deep eddies?

Our approach combines theoretical investigation with advanced numerical modeling to gain understanding and efficient representation of the most important physical processes in the littoral zone. In collaboration with the Dr. Watts’ (URI) project we use analytical and numerical models for analyses of the SYNOP Central Array data set.

Our principal findings as follows:

1) The growth rate of baroclinic instability and group velocity in a Gulf Stream-type jet is only slightly affected by a relatively weak slope. However, the nonlinear evolution of meanders and associated ring detachment processes depend strongly upon the ocean depth and topographic slope. Nonlinear equilibration of meanders is efficiently controlled by topography via constraining the development of deep eddies. The main equilibration mechanism is a homogenization of the lower-layer potential vorticity by deep eddies over sloping bottom.

2) During periods of energetic meandering, the horizontal redistribution of potential vorticity by deep eddies over topography results in the formation of deep recirculations which stabilize the Gulf Stream. The later explains observed periods of low meandering after periods of energetic meandering. The deep recirculations are strongly asymmetric due to the vertical shift of the potential vorticity gradient in the initial state and more intense deep ocean cyclogenesis than anticyclogenesis during the meandering. The later is consistent with the SYNOP observations.

We have analyzed the effects of a topographic slope on the growth of instabilities in a baroclinic jet using the  primitive equation Princeton Ocean Model adapted to the Gulf Stream region. An unperturbed jet is prescribed as a potential vorticity (PV) front in the upper thermocline overlying intermediate layers with weak  PV gradients and a quiescent bottom layer over a positive  cross-stream continental slope.  We have carried out a series of numerical experiments with the same initial conditions over slope and flat bottoms on the beta-plane and on the f-plane.

The  two figures below illustrate some of our results. Two numerical experiments are compared here, in which a small local perturbation induces the growth of an instability which amplifies as it propagates downstream. Fig. 1 shows the temperature at the depth 250 m (red contours are 12 and 14 degree C isotherms) which characterizes thermocline meanders, superimposed by deep velocity field below the thermocline at day 60. In the upper panel the bottom is flat while in the lower panel the bottom slope is 0.002. The swirling motion in deep eddies is nearly barotropic providing an important feedback for meander growth and ring shedding. Meanders growing over a flat bottom are able to pinch off, resembling warm- and cold-core rings, while in the presence of the bottom slope the meander amplitudes saturate, with no ring shedding.

Fig. 1. Temperature at 250 m and Deep Velocity Field at Day 60.

Stability properties of the jet are strongly linked to potential vorticity (PV) which is mostly conserved in fluid parcels. The most essential for baroclinic instability of the jet is the lower layer PV shown in Fig. 2. The panels on the right show zonally averaged deep PV profiles at t = 0 (solid line) and t = 60 days (dashed). In both cases initially there are zones of negative PV gradient favorable for baroclinic instability. A dramatic difference in PV between flat and sloping bottom is seen after development of deep eddies: in the case of slope bottom the negative PV gradient in the lower layer is almost eliminated that stabilizes the jet.

Fig. 2 Deep Layer Potential Vorticity at Day 60.

Important physical mechanism revealed: A topographic slope modifies the development of deep eddies and causes the phase shift of deep eddies in the direction of the upper layer troughs/crests, thus limiting growth of the meanders. This phase-locking of the meanders with deep eddies underneath agrees qualitatively with the observational data at the SYNOP Central array (Watts et al, 1995).

Data assimilation and model initialization for modeling baroclinic jets

We have constructed a multi-layer intermediate model capable of accurately representing the baroclinic structure of the upper ocean in the Gulf Stream using the PV approach. One of the most important advantages of the PV approach is that it can reproduce the observed structural changes to the velocity and density fields between crests and troughs in steep meanders. Based on this model we have developed a new data assimilation scheme for both initialization and data assimilation of dynamically consistent density and velocity fields in a baroclinic jet.

We have demonstrated that the realistic initialization of the baroclinic jet structure (e.g. the Gulf Stream) in a primitive equation model is imperative if realistic simulations of the jet evolution as well as the strengths and structure of the associated deep eddies have a chance to be achieved.

Nevertheless, forecasting a meandering jet remains a formidable challenge. We believe that a bottleneck is related to our inability to properly initialize the deep eddies underneath the jet and our limited understanding of vertical coupling mechanisms and the interaction of the deep eddies with topography.