Thomas Decloedt (University of Hawaii at Manoa)
Small-scale turbulence and the oceanic density structure
Over the past quarter century, observations have made it abundantly clear that the spatial distribution of diapycnal mixing is highly heterogeneous in the deep ocean. Various preliminary mixing parameterizations taking the spatial heterogeneity into account have been developed over the past decade and their implementation into OGCMs demonstrates the large-scale ocean circulation is sensitive to details of the spatial distribution of diapycnal mixing. We here compare the spatial distributions implied by two mixing parameterizations to distributions inferred from hydrographic inversions and argue that the differences are sufficiently large that present-day mixing parameterizations should be viewed as tools for sensitivity analysis, not as realistic representations of mixing in the deep ocean. Far more observations of small-scale turbulence are required to improve our understanding of what determines the geographical distribution and magnitude of diapycnal mixing in the deep ocean. The Thorpe scale methodology has enabled the inference of turbulence from standard CTD, ARGO and MMP data, in particular permitting the identification of mixing hotspots. However, analysis of microstucture data from a profiler equipped with a pumped CTD system indicates that the Thorpe scale methodology as currently applied is of dubious quantitative value. We discuss a possible enhancement of the technique that incorporates not justvdensity but also shear profiles in an effort to improve its correlation with TKE dissipation rates derived from microstructure observations.