A study of diapycnal mixing in the Southern Ocean using a tracer release experiment and numerical models

The Diapycnal and Isopycnal Mixing Experiment in the Southern Ocean (DIMES) includes a tracer release experiment and microstructure programme with the aims of diagnosing the strength and variability of mixing in the Southern Ocean. Here numerical models are used to advect and diffuse a tracer in the Antarctic Circumpolar Current, beginning in the Southeast Pacific and progressing through Drake Passage, and model outputs are then compared with observations from the DIMES tracer. The prescribed diapycnal diffusivity fields within the models are varied between different model runs, and the model parameters are optimised using a cost function to give the best fit to the observations. A simple 2D model with dimensions of along-stream distance and depth yields estimates for diapycnal diffusivity neutral density surface onto which the tracer was released. A more complex 3D model using an offline version of the MITgcm with time-evolving observation-based velocities from the SatGEM product yields similar estimates for the Pacific and Drake Passage, respectively. Point microstructure dissipation measurements collected as part of DIMES are used to construct three-dimensional diffusivity fields which are then used in conjunction with the 3D model to test whether the mixing rates inferred from microstructure and the tracer measurements are consistent with one another. Good agreement is found in the Southeast Pacific, but in Drake Passage, where both topography and current field becomes more heterogeneous, the microstructure estimates are 5 times too low to account for the time and spatially averaged mixing implied by the tracer. By contrast, model diffusivities constructed using predicted rates of lee wave generation from modified linear theory predict the along-stream variation in tracer vertical profile widths reasonably well throughout the model domain, but do not capture the across-stream variation