An Estimation of Buoyancy Flux for a Mixture of Turbulence and Double Diffusion

Abstract

Microstructure measurements were made in the Mixed Water Region of the Oyashio/Kuroshio/Tsugaru currents system where both turbulence and double diffusion are involved in mixing. While intense turbulence is observed near the front between the Oyashio and the Tsugaru Current, double diffusion occupies a noticeable fraction in both the Tsugaru Water and the Mixed Water between the Oyashio and the Kuroshio. After determining a criterion to distinguish double diffusion from turbulence, vertical diffusivities and buoyancy fluxes are estimated using microstructure data. When turbulence is weak, double diffusion is observed around temperature and salinity anomalies, partly due to interleaving, and dominates the buoyancy flux. Vertical diffusivities due to double diffusion are parameterized as a function of the 10-m-scale density ratio. The 10-m-scale diffusivity estimates are consistent with the microstructure data when an appropriate criterion to reproduce a probability density function for the Turner angle is applied. A weighted-average diffusivity model is proposed to account properly for turbulence and double diffusion simultaneously.

Evidence of enhanced double‐diffusive convection below the main stream of the Kuroshio Extension

“…Corresponding dissipation rates for microscale thermal variance are as large as (red curve in Figure e) with relatively low turbulent kinetic energy dissipation rates, 10 −8 Wkg −1 (black curve in Figure e). The large thermal variance dissipation with weak turbulence indicates the onset of double‐diffusion [ Schmitt et al ., ; Inoue et al ., ]. These results are similar to the previous observations in frontal regions near a Meddy [ Armi et al ., ], and Azores Front [ Georgi and Schmitt , ].…”

“…In the same region, Inoue et al . [] directly confirmed the occurrence of double diffusion when turbulence was weak, and parameterized diffusivities due to double diffusion, using microstructure measurements. Schmitt [] points out that widespread patterns with density ratio in the surface mixed layer, increasing to in the midlatitude thermocline, are suggestive of ubiquitous double‐diffusion occurrence.…”

Double‐diffusive layers in the Adriatic Sea

“… Inoue et al [2007] formulated a series of concurrent conditions to distinguish episodes of DDC from those of shear‐driven conventional turbulence. For the DL type of DDC, the shear microstructure was classified as non‐turbulent when the buoyancy Reynolds number = <20, and DL is presumed to occur when the inverse of the density ratio R ρ (defined as , with α = 0.216·10 −3 °C −1 and β = 0.77·10 −3 PSU −1 ) is 1.0–3.0, the temperature gradient dissipation χ > 5 · 10 −9 °C s −1 and the buoyancy flux ratio γ > 1, determined from the relationship between the observed mixing efficiency and density ratio as defined by Inoue et al [2007]. At GS2, across the interfaces (e.g., the one roughly bracketed between 11 and 14 m), we have Δ T = 0.67 °C and Δ S = 0.33 PSU.…”

“…At GS2 then, using we obtain = 1.97 · 10 −5 m 2 s −1 and K S = 4.01 · 10 −6 m 2 s −1 . When using other parameterizations, such as those suggested by Kelley [1984] and Inoue et al [2007] (or equivalently, Kelley [1990]), results for range between 1.62 · 10 −5 m 2 s −1 and 8.50 · 10 −6 m 2 s −1 . These values appear to be consistent with the diffusivities presented by Kelley [1984, 1990] and while the peak values in measured are O (10 −3 ), the average value for the water column is about 2 · 10 −5 m 2 s −1 in good agreement with the value found above.…”

“…roughly 100), the growth rate of SF is usually not large until R ρ approaches 1. As in the case of DL, in order to detect whether we are witnessing a SF episode, Inoue et al [2007] proposed the following series of conditions: < 20, 1 < R ρ < 3, χ > 5 · 10 −9 °C s −1 and 0 < γ < 1. In our measurements = 18, R ρ = 1.12, χ > 5 · 10 −9 °C s −1 and γ = 0.06.…”