Finescale Parameterizations of Turbulent Dissipation

Kurt L. Polzin School of Oceanography, University of Washington, Seattle, Washington

Search for other papers by Kurt L. Polzin in
Current site
Google Scholar
PubMed
Close
,
John M. Toole Department of Physical 0ceanography, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts

Search for other papers by John M. Toole in
Current site
Google Scholar
PubMed
Close
, and
Raymond W. Schmitt Department of Physical 0ceanography, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts

Search for other papers by Raymond W. Schmitt in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

Fine- and microstructure data from a free fall profiler are analysed to test models that relate the turbulent dissipation rate (ε) to characteristics of the internal wave field. The data were obtained from several distinct internal wave environments, yielding considerably more range in stratification and wave properties than has been previously available. Observations from the ocean interior with negligible large-scale flow were examined to address the buoyancy scaling of ε. These data exhibited a factor of 140 range in squared buoyancy frequency (N2) with depth and uniform internal wave characteristics, consistent with the Garrettt–Munk spectrum. The magnitude of ε and its variation with N(ε ∼ N2) was best described by the dynamical model of Henyey et al. A second dynamical model, by McComas and Muller, predicted an appropriate buoyancy scaling but overestimated the observed dissipation rates. Two kinematical dissipation parameterizations predicted buoyancy scalings of N3/2; these are shown to be inconsistent with the observations.

Data from wave fields that depart from the canonical GM description are also examined and interpreted with reference to the dynamical models. The measurements came from a warm core ring dominated by strong near-inertial shears, a region of steep topography exhibiting high-frequency internal wave characteristics, and a midocean regime dominated al large wavelengths by an internal tide. Of the dissipation predictions examined, those of the Henyey et al. model in which εN−2 scales as E2, where E is the nondimensional spectral shear level, were most consistent with observations. Nevertheless, the predictions for these cases exhibited departures from the observations by more than an order of magnitude. For the present data, these discrepancies appeared most sensitive to the distribution of internal wave frequency, inferred here from the ratio of shear spectral level to that for strain. Application of a frequency-based correction to the Henyey et al. model returned dissipation values consistent with observed estimates to within a factor of 2.

These results indicate that the kinetic energy dissipation rate (and attendant turbulent mixing) is small for the background Garrett and Munk internal wave conditions (0.25εN−2∼0.7 × 10−5 m2 s−1). Dissipation and mixing become large when wave shear spectral levels are elevated, particularly by high-frequency waves. Thus, internal wave reflection/generation at sleep topographic features appear promising candidates for achieving enhanced dissipation and strong diapycnal mixing in the deep ocean that appears required by box models and advection–diffusion balances.

Abstract

Fine- and microstructure data from a free fall profiler are analysed to test models that relate the turbulent dissipation rate (ε) to characteristics of the internal wave field. The data were obtained from several distinct internal wave environments, yielding considerably more range in stratification and wave properties than has been previously available. Observations from the ocean interior with negligible large-scale flow were examined to address the buoyancy scaling of ε. These data exhibited a factor of 140 range in squared buoyancy frequency (N2) with depth and uniform internal wave characteristics, consistent with the Garrettt–Munk spectrum. The magnitude of ε and its variation with N(ε ∼ N2) was best described by the dynamical model of Henyey et al. A second dynamical model, by McComas and Muller, predicted an appropriate buoyancy scaling but overestimated the observed dissipation rates. Two kinematical dissipation parameterizations predicted buoyancy scalings of N3/2; these are shown to be inconsistent with the observations.

Data from wave fields that depart from the canonical GM description are also examined and interpreted with reference to the dynamical models. The measurements came from a warm core ring dominated by strong near-inertial shears, a region of steep topography exhibiting high-frequency internal wave characteristics, and a midocean regime dominated al large wavelengths by an internal tide. Of the dissipation predictions examined, those of the Henyey et al. model in which εN−2 scales as E2, where E is the nondimensional spectral shear level, were most consistent with observations. Nevertheless, the predictions for these cases exhibited departures from the observations by more than an order of magnitude. For the present data, these discrepancies appeared most sensitive to the distribution of internal wave frequency, inferred here from the ratio of shear spectral level to that for strain. Application of a frequency-based correction to the Henyey et al. model returned dissipation values consistent with observed estimates to within a factor of 2.

These results indicate that the kinetic energy dissipation rate (and attendant turbulent mixing) is small for the background Garrett and Munk internal wave conditions (0.25εN−2∼0.7 × 10−5 m2 s−1). Dissipation and mixing become large when wave shear spectral levels are elevated, particularly by high-frequency waves. Thus, internal wave reflection/generation at sleep topographic features appear promising candidates for achieving enhanced dissipation and strong diapycnal mixing in the deep ocean that appears required by box models and advection–diffusion balances.

Save