L3: Diagnosing and parameterizing the effects of eddies

Principal investigators: Dr. Kerstin Jochumsen (Universität Hamburg), Dr. Alexa Griesel (Universität Hamburg)

Oceanic eddies, fluctuations on scales on the order of one km to hundreds of km, derive their energy primarily from baroclinic instability processes. It is common to parameterize this loss of available potential energy of the balanced flow in ocean models by an additional eddy-driven advection velocity, or, equivalently a skew diffusivity. In addition, eddies also mix tracers along isopycnals, which is implemented as along-isopycnal diffusion with isopycnal diffusivity. Eddy diffusivities currently are specified in coarse ocean models without connection to the energy budget, and, more fundamentally, it is unclear to what extent, where and on what scales the downgradient eddy-diffusion model is appropriate at all. Rotational components of the eddy fluxes associated with the advective terms in the eddy variance equation are generally large, so that production and dissipation of eddy energy do not balance locally.

Dispersion statistics from Lagrangian numerical particles deployed in the Southern Ocean of an eddying ocean model. Left: Lagrangian diffusivities (upper panel) and Lagrangian velocity autocovariance (lower panel) as a function of time lag at three depth levels for a region southwest of Australia. Right: Lagrangian integral time scale (upper panel) and Lagrangian diffusivity (lower panel) diagnosed from the numerical particle dispersion in bins as a function of depth and longitude along the main jets of the Antarctic Circumpolar Current. The black line is the steering level depth where enhanced mixing is expected from linear theory (from Griesel et al., 2015).

Our main questions are:

  • What is the spatial and temporal variation of eddy-mean flow interactions and scale dependence of eddy diffusivities on a global scale and in one key observational region of the ocean, the Benguela upwelling Current?
  • What are the limits and validities of the eddy diffusion model and consequences for energy-consistent parameterizations of eddy-mean flow interactions across a range of eddy scales?

To address these questions we will use Lagrangian particle statistics in both eddying ocean models and observations to assess the spatial and temporal variability of eddy diffusivities, eddy-mean flow interactions, and submesoscale processes, and to quantify dispersion regimes on different scales to assess the eddy-diffusion model and get insight into energy cascades. In-situ observations and a Lagrangian drifter experiment at the upwelling front in the Benguela current system will  serve as the major observational reference to study the lifetime of meso- and submesoscale features, as well as their dissipation in the interior ocean. Results from the analyses of observations and eddying ocean models will be used to construct energy-consistent parameterizations that use prognostic equations of eddy kinetic energy, taking non-localness into account.

Exemplary sea surface temperature from satellite data off Namibia (MODIS; spatial resolution 4~km) on 5th of January 2012. The blue and green colours clearly show the cold coastal upwelling region as well as the inhomogeneous front at the western boundary of the upwelling system. White colour marks regions which are covered by clouds (no data). Orange stars mark the locations of Walvis Bay (WB) and Luederitz (L).
Snapshot of Sea surface height in January 2000 in the eddying Parallel Ocean Program (POP) model in the Namibian upwelling zone showing cold-core eddies propagating westward encountering warm-core eddies that have propagated from the Agulhas region. Also shown are some of the surface trajectories of the numerical floats.
No guests available.
No reports available.