The kinetic energy in the general circulation of the ocean is forced mainly by surface wind stress at large space and long time scales, yet it must be viscously dissipated at very small scales in equilibrium. Large- and mesoscale circulations typically satisfy a momentum-balanced dynamics (e.g., geostrophy and its generalizations). Yet balanced turbulent cascades are very inefficient in energy dissipation. This leads to the fundamental question of how the necessary dissipation is accomplished for the general circulation in equilibrium. Some of the dissipation undoubtedly occurs within turbulent boundary layers near the surface and bottom, and some occurs at the bottom through generation of internal gravity waves by large-scale flow over topography. A difficulty with viewing these boundary routes as sufficient is that an implausibly efficient spatial energy flux towards the boundaries, effected by balanced currents, would be required to continuously deplete the interior energy reservoirs. We investigate an alternative, more local route through interior, turbulent cascade dynamics of the circulation. In Oceanic General Circulation Models (GCMs), the local route to dissipation is implied by the use of ad hoc horizontal ad vertical eddy viscosities to parameterize this cascade, and the associated interior dissipation rate is a substantial fraction of the volume-integrated total.

There are explicitly specifiable limits to the regime of validity for balanced dynamics that are violated sometimes for the circulation. Violation of these limits leads to energy transfer at intermediate spatial scales from balanced to unbalanced motions, including inertia-gravity waves. Unbalanced turbulent cascades are much more efficient in their energy dissipation and spatial radiation causing energy redistribution. Thus, the important bottleneck in the local route for general-circulation dissipation is loss of balance and its evolutionary consequences. This bottleneck is the focus of our research.

Recently, we have demonstrated the occurrence of several unbalanced, linear instabilities near the limits of balanced validity. The next step is to determine their finite-amplitude evolutionary behavior, involving cascade to dissipation. We study the dynamical behavior of these instabilities using solutions of the non-hydrostatic Boussinesq Equations. The unbalanced instabilities are analyzed as they reach finite amplitudes and eventually lead to a turbulent regime. Since we hypothesize that a crucial step en route to dissipation is instabilities of mesoscale structures to unbalanced motions, we have devised a new method to partition the flow into balanced and unbalanced parts. On this basis we are able to say if and how much the interior dissipation of the large scale flow is enhanced over the expectation from geostrophic turbulence theory.