LES study of vertical eddy diffusivity estimation in bottom boundary layers

Yoshinobu Wakata

Research output: Contribution to journalArticle

Abstract

Osborn's method is commonly used to obtain diffusion coefficient K from the turbulent dissipation rate e. This method is the relational expression K=Γε/N2. The dissipation flux coefficient Γ is often set to a constant value of 0.2, but this study of LES revealed that γ varies greatly vertically in the bottom boundary layer because of the influence of the seabed. Consequently, the eddy diffusion coefficient is overestimated in the lower part of the bottom boundary, but it is slightly underestimated in the upper part. Therefore, Osborn's method with constant Γ cannot give the correct diffusivity. Furthermore, even if treating Γ as a function of flux Richardson number Rf as defined originally by Osborn, the estimation is underestimated by the advection effect because of the influence of spatial nonuniformity. Energy budget analysis revealed that this defect can be improved using the extended flux Richardson number, which can be estimated by multiplying Rf using a constant correction factor. Furthermore, we proposed two alternative estimation methods. For the first method, which estimates the relation between Rf and the gradient Richardson number Rg, Γ can be expressed with Rg instead of Rf with a correction factor. We can estimate the reasonable diffusivity if we have current data supplementary to obtain . For the second method, Γ can be expressed as a similarity function of the height above the bottom normalized by the Ozmidov scale. This method can provide an acceptable estimate of diffusivity without current data for several circumstances.

Original languageEnglish
Pages (from-to)1903-1920
Number of pages18
JournalJournal of Physical Oceanography
Volume48
Issue number8
DOIs
Publication statusPublished - Aug 1 2018

All Science Journal Classification (ASJC) codes

  • Oceanography

Fingerprint Dive into the research topics of 'LES study of vertical eddy diffusivity estimation in bottom boundary layers'. Together they form a unique fingerprint.

  • Cite this