Large Eddy Simulation

LES Model Description

The LES model, developed by Deardorff (1980) and Moeng (1984) for the atmospheric boundary layer, dynamically solves for unresolved (subgrid) turbulent kinetic energy (TKE). Several adaptations of this model are currently used to model oceanic boundary layers, including high-latitude deep convection in response to storm forcing (Harcourt et al., 2001), and forced convection in more shallow free-slip ocean boundary layers (e.g. Skyllingstad and Denbo, 1995; Wang et al, 1996).

LES Detailed Description

Preliminary Work Completed

Several LES runs have been carried out to test the stability of the upper ocean under conditions observed during ANZFLUX at certain times when ocean observations were not made. In general, they agree with the LTC results of McPhee (2000), showing the upper ocean becoming unstable to a process of thermobaric parcel detrainment, but at an earlier point in time. These simulations do not yet account for cabbeling effects, as this term was omitted in the preliminary simulations for comparison with LTC results.

5Mb GIF animation

Two of the first simulations are presented on a separate document. Simulation 3 was performed with a domain of 128x128x64 grid points, extended to 128x128x168 after one simulated day. The time step is 4s and the grid resolution 6m. The LES was initialized from averaged typical profiles measured in 1994 on year days 216 and 217 by the Turbulent Instrument Cluster and CTD, and forced with time-dependent surface stress deduced from observations.
Potential temperature is shown on 2 vertical sections and two isothermal surfaces at 0°C (above) and 0.4°C (below) representative of the pycnocline. The animation (5Mb) illustrates the evolution of the plumes into a major convective cell that spreads when it reaches a level of neutral buoyancy near 300m depth.

Simulation 3 expanded analysis

The Proposal

We propose that this preliminary work on LES model development be verified through quantitative comparisons with existing data sets from polar boundary layers below ice, and subsequently be used to make predictions about upper ocean instabilities for use in motivating and designing future experiments in the Weddell Sea.

Plans for model-data comparison and experiment design

Maintained by Eun and Myung Park