Bathymetry-aware mesoscale eddy parameterizations

Mesoscale eddy buoyancy fluxes across continental slopes profoundly modulate the boundary current dynamics and shelf-ocean exchanges, but have yet to be appropriately parameterized via the Gent-McWilliams (GM) scheme in predictive ocean models. In this talk, I will present a suite of recently developed, “bathymetry-aware’’ parameterizations of mesoscale eddy buoyancy fluxes across continental slopes. These parameterizations are theoretically-based, account for the transition between continental shelf and open ocean regions, and are examined both diagnostically against eddy-resolving simulations and prognostically using non-eddying simulations. Numerical implementation of the “bathymetry-aware” parameterizations relies either on a modified subgrid-scale eddy energy budget or on simple neural networks serving to predict the mesoscale eddy energy online. In the case of an upwelling frontal system, the predicted eddy energy coupled with the parameterized eddy buoyancy fluxes could further inform strong cross-slope eddy momentum fluxes, which are consistent with the “Neptune effect” and capable of driving an undercurrent running counter to the surface flow. These “bathymetry-aware” parameterizations thus provide a recipe of GM variants for use in predictive ocean models with continental slopes and highlight the potential of machine learning techniques to augment physics-based mesoscale eddy parameterization schemes.