Estimation of Horizontal Turbulent Diffusivity from Deep Argo Float Displacements
|Author(s)||Sévellec Florian1, Colin De Verdière Alain2, Kolodziejczyk Nicolas2|
|Affiliation(s)||1 : Laboratoire d’Océanographie Physique et Spatiale, Univ-Brest CNRS IRD Ifremer, Brest, France Ocean and Earth Sciences, University of Southampton, Southampton, UK
2 : Laboratoire d’Océanographie Physique et Spatiale, Univ-Brest CNRS IRD Ifremer, Brest, France
|Source||Journal Of Physical Oceanography (0022-3670) (American Meteorological Society), 2022-07 , Vol. 52 , N. 7 , P. 1509-1529|
|WOS© Times Cited||1|
|Keyword(s)||Ocean, Diffusion, Dispersion, Lagrangian circulation, transport, Mesoscale processes, In situ oceanic observations|
We use an analog method, based on displacements of Argo floats at their parking depth (nominally located around 1,000 dbar) from ANDRO dataset, to compute continuous, likely trajectories and estimate the Lagrangian dispersion. From this, we find that the horizontal diffusivity coefficient has a median value around 500 m2 s−1 but is highly variable in space: reaching values from 100 m2 s−1 in gyre interior to 40,000 m2 s−1 in a few specific locations (in the Zapiola gyre and in the Agulhas Current retroflection). Our analysis suggests that the closure for diffusivity is proportional to Eddy Kinetic Energy (or square of turbulent velocity) rather than (absolute) turbulent velocity. It is associated to a typical turbulent time scale of 4 to 5.5 days, which is noticeably quite constant over the entire globe, especially away from coherent intense currents. The diffusion is anisotropic in coherent intense currents and around the equator, with a primary direction of diffusion consistent with the primary direction of horizontal velocity variance. These observationally based horizontal diffusivity estimations, and the suggested Eddy Kinetic Energy closure, can be used for constraining, testing, and validating eddy turbulence parameterization.