Turbulent dissipation rates in the Southern Ocean: Lessons learned from observations in a mixing hot spot

SOFINE, the Southern Ocean FINE structure project, is a study of the role and relative importance of internal waves in the dynamics of the Antarctic Circumpolar Current (ACC) and the Southern Ocean Meridional Overturning Circulation (MOC) in a Southern Ocean mixing hot spot.  The hot spot of choice is a standing meander of the current at the northern Kerguelen Plateau slope.  This is a region that is potentially of special significance to closing the Southern Ocean overturning circulation and the ACC momentum budget as it presents both a large-scale topographic obstacle and small-scale topographic roughness in the path of multiple ACC jets. As such, it is a likely site for both enhanced adiabatic and diabatic mixing processes, as well as lee wave and internal tide generation. The relative importance of internal waves and diabatic processes to eddy transports in the Southern Ocean interior has significant implications for the dynamics of the Southern Ocean MOC in the context of zonal mean theory, the modeling of mixing processes, the sensitivity of the MOC to changes in climatic forcing, and a range of important biogeochemical and paleo-oceanographic issues.

The SOFINE study is pursued through the analysis of a suite of observations on internal wave and turbulent scales.  These include a full-depth fine structure and microstructure survey of the region, mooring measurements surveying properties of the internal wave field on the lee side of where an ACC frontal jet passes over a small-scale ridge, and profiles of upper-ocean thermohaline and shear fine structure gathered by a swarm of EM APEX floats.  The aims of the analysis presented here are first to characterize the magnitude and distribution of turbulent dissipation and mixing observed, and second to understand possible mechanisms underpinning the observed distribution in the context of internal wave generation and evolution in the region.

Results of this analysis show first that the turbulent energy dissipation and mixing rates observed are highly spatially variable.  Systematic structure with depth and proximity to rough topography suggest a link with the local internal wave field, which can be characterized as consisting of near-inertial waves propagating from the surface downwards and higher frequency internal waves sourced at the bottom propagating upwards, both being modified by the presence of a background shear as they propagate.  The calculation of various timescales and the exercise of ray tracing give insight into the dominant processes controlling the local internal wave life cycle, and highlight several very important roles the mean flow plays in modulating the generation and evolution of internal waves that eventually produce the dissipation observed. These include the generation of upward propagating lee waves through interaction with the bottom topography, the advection of wave packets, the modification of wave packets' characteristics as they propagate through the background shear, and wave-mean flow interactions, some of which must be considered as three-dimensional.

These results potentially have several significant implications for Southern Ocean overturning and ACC dynamics, as well as for parameterizations of turbulent dissipation and mixing.  First, given that SOFINE demonstrates that turbulent dissipation and mixing in the Southern Ocean interior is, in places, non-negligible, the implications of neglecting or misrepresenting interior diabatic processes in our theoretical descriptions and numerical models of Southern Ocean circulation should be carefully considered.  Second, given that SOFINE shows that turbulent dissipation and mixing are not uniform in space with systematic horizontal and depth dependent structure, the impact of mixing heterogeneity and this systematic structure on Southern Ocean dynamics may also be important to consider.  Finally, given that SOFINE highlights the important roles that the mean flow can play in modulating where turbulent dissipation takes place, SOFINE suggests that first fine scale parameterizations, parameterizations of turbulent dissipation due to breaking internal waves based on a model of wave-wave interaction, may require modification in regions of large mean flow, and second that internal wave dissipation may not be local to internal wave generation.  Mean flow effects, including advection and wave-mean flow interactions, may be important ingredients in an internal wave parameterization of turbulent dissipation and mixing in strongly advective regions like the ACC.