The interactions between turbulence and flames in premixed reacting flows are studied for a broad range of turbulence intensities by analyzing scalar (reactant mass-fraction) gradient, vorticity, and strain rate fields. The analysis is based on fully compressible, three-dimensional numerical simulations of H 2-air combustion in an unconfined domain. For low turbulence intensities, a flame reconstruction method based on the scalar gradient shows that the internal flame structure is similar to that of a laminar flame, while the magnitudes of the vorticity and strain rate are suppressed by heat release and there is substantial anisotropy in the orientation of intense vortical structures. As the turbulence intensity increases, the local flame orientation becomes increasingly isotropic, and the flame preheat zone is substantially broadened. There is, however, relatively little broadening of the reaction zone, even for high intensities. At high turbulence intensities, the vorticity and strain rate are only weakly affected by the flame, and their interactions with the scalar gradient are similar to those found in nonreacting turbulence. A decomposition of the total strain rate into components due to turbulence and the flame shows that vorticity suppression depends on the relative alignment between vorticity and the flame surface normal. This effect is used to explain the anisotropy of intense vortices at low intensities. The decomposition also reveals the separate effects of turbulent and dilatational straining on the flame width. V

[1] Large eddy simulations of the Craik-Leibovich equations are used to assess the effect of misaligned Stokes drift and wind direction on Langmuir cells in the ocean mixed layer. Misalignments from 0 to 135 are examined and Langmuir turbulence structures are evident in all cases. The Stokes drift is modeled using a broadband empirical spectrum, and cases with and without the Coriolis effect, wind waves, and an initial mixed layer are examined. The expected scaling for the vertical velocity variance is recovered in the aligned simulations and is adapted here to the misaligned cases. The adjusted scaling projects the friction velocity (aligned with the wind stress) into the dominant axial direction of the Langmuir cells. The turbulent Langmuir number is generalized through a similar projection into the axial direction of the Langmuir cells, which reduces its value in realistic conditions. For known Langmuir cell orientations, the strength of Langmuir turbulence for misaligned cases can be estimated using the projected Langmuir number. A prediction for the angle between the wind stress and cell direction is obtained using the law of the wall; this prediction only requires the wind stress, Stokes drift, and boundary layer depth. Conditional analyses show that, with increasing misalignment, the typically antisymmetric Langmuir cell pairs become asymmetric. This asymmetry is due, in part, to the advection by cross cell flow of vorticity from one vortex tube onto the other, and in part due to an asymmetry induced by the stretching of vertical vorticity into cross cell vorticity.

The interactions between boundary layer turbulence, including Langmuir turbulence, and submesoscale processes in the oceanic mixed layer are described using large-eddy simulations of the spindown of a temperature front in the presence of submesoscale eddies, winds, and waves. The simulations solve the surface-wave-averaged Boussinesq equations with Stokes drift wave forcing at a resolution that is sufficiently fine to capture small-scale Langmuir turbulence. A simulation without Stokes drift forcing is also performed for comparison. Spatial and spectral properties of temperature, velocity, and vorticity fields are described, and these fields are scale decomposed in order to examine multiscale fluxes of momentum and buoyancy. Buoyancy flux results indicate that Langmuir turbulence counters the restratifying effects of submesoscale eddies, leading to small-scale vertical transport and mixing that is 4 times greater than in the simulations without Stokes drift forcing. The observed fluxes are also shown to be in good agreement with results from an asymptotic analysis of the surface-wave-averaged, or Craik–Leibovich, equations. Regions of potential instability in the flow are identified using Richardson and Rossby numbers, and it is found that mixed gravitational/symmetric instabilities are nearly twice as prevalent when Langmuir turbulence is present, in contrast to simulations without Stokes drift forcing, which are dominated by symmetric instabilities. Mixed layer depth calculations based on potential vorticity and temperature show that the mixed layer is up to 2 times deeper in the presence of Langmuir turbulence. Differences between measures of the mixed layer depth based on potential vorticity and temperature are smaller in the simulations with Stokes drift forcing, indicating a reduced incidence of symmetric instabilities in the presence of Langmuir turbulence.