The intellectual merits of this work involve clarification of the unique dynamics near oceanic boundaries and quantification of their effects on essential oceanic processes like boundary stress, mesoscale dissipation and diapycnal mixing followed by adiabatic isopycnal dispersion. The primary unsolved dynamical problem of physical oceanography involves the 'dissipative' closure of western boundary currents, which is the primary focus of this project.
The problem of Topographic Control of the Gulf Stream with particular application to its separation is being addressed. This project is built on the hypothesis that boundaries exert fundamental controls on the ocean circulation in general and the Gulf Stream in particular. These processes are poorly represented or absent from current climate models. The main focus of the proposed research is on sub-inertial excitation of inertia-gravity waves and activation of the sub-mesoscale by boundary processes, with a view towards their potential vorticity and lateral stress implications. The driving hypothesis is that these processes dominate Gulf Stream separation and downstream development. This is a critical aspect of ocean circulation that all current models struggle with, regardless of resolution. The result is the introduction of strong biases in the model North Atlantic that is likely to affect climate projection on the decadal time scale. Observations also suggest an important role for topography in promoting mixing.
Broader impacts: Preparing for climate change is the most pressing problem currently facing society. This project is addressing a significant shortcoming of the oceanic component of coupled climate models. The mesoscale and boundary currents are the dominant kinematic features of the ocean and are controlled in all existing models by parameterizations. This project is designed to build into these parameterizations physically based models of boundary interactions, thereby enhancing the fidelity of climate prediction and eliminating a major bias found in current ocean circulation models. Since the new parameterizations will be implemented in the Community Climate System Model, they will be widely available to the general climate community.
The major goals of this project are to understand the mechanisms that control the separation of the Gulf Stream from the US east coast at Cape Hatteras and to translate that understanding into improvements in their representation in coarse resolution (order 1 degree horizontal resolution) ocean models used in coupled climate simulations. Because climate simulations require long model integrations, coarse resolution coupled models remain the workhorse, i.e., routinely used in climate simulations. All such models exhibit a number of biases, i.e., persistent deviations of model solutions from observations. Arguably the most notable and troubling of such biases occurs in the North Atlantic due to erroneous Gulf Stream separation and incorrect path of the subsequent North Atlantic Current, resulting in large temperature and salinity biases. Our hypothesis is that inaccurate representations of the topographic interactions on the US eastern seaboard are major contributors to these biases. We performed a thorough inter-comparison of the Gulf Stream and North Atlantic Current simulations from a diverse set of coarse, fine, and ultra-fine resolution models, ranging from 1 degree to 1/36 degree horizontal resolution. At coarse resolution, the separation details of the Gulf Stream were not sensitive to the details of either the resolved bottom topography or parameterization of viscous processes. At finer resolution, topographic features do impact the Gulf Stream dynamics, but only locally. Such topographic details remain unresolved in coarse resolution models. A major finding of our project is that despite the differences in the structure of the modeled Gulf Stream and North Atlantic Current system between the coarse resolution and the higher resolution model simulations, the leading order barotropic vorticity balances are in agreement – which was not expected a priori. Specifically, the leading order balance is between the negative input of the wind stress curl and the positive bottom pressure torques within the subtropical gyre. This inviscid balance holds across all our model simulations, indicating a robust feature. More regionally within the Gulf Stream area, the primary balance is between bottom pressure torque and planetary vorticity advection. Nonlinear torques, integrated over the Gulf Stream region, are small. An important conclusion is that the coarse resolution models do capture the correct, i.e., as in the fine resolution models, leading order vorticity balances.
