Studies on parameterizations of turbulent mixing in the deep ocean

Abstract

The global distribution of turbulent mixing in the deep ocean is essential for understanding the global overturning circulation. Currently, however, this requires reliance on a technique referred to as parameterization. This method aims to estimate the strength of turbulent mixing using large-scale observable physical quantities on a global scale. However, there are several problems with the existing turbulence parameterization methods. The parameterization of turbulent dissipation rates in the deep ocean relies on both fine-scale velocity shear and density strain information to measure the distortion of internal wave spectra. However, the application of the single-wave approximation to broadband internal wave spectra leads to overestimation of turbulent dissipation rates. In particular, in regions dominated by near-inertial gravity waves, where this approximation is effective, turbulent dissipation rates are overestimated. Furthermore, the determination of the turbulent mixing intensity from the estimated dissipation rate conventionally assumes an efficiency of approximately 20%, but this efficiency varies depending on the driving mechanisms and developmental stages of turbulent events. In particular, in areas where sporadic, intense turbulent events are induced by convective instability in high-density seawater overflow regions, the mixing efficiency can sharply increase to 50%. In this paper, the existing problems with turbulence parameterization are reviewed, and potential solutions are explored.