Stably Stratified Interfacial-Layer Turbulence

The interfacial layer between the turbulent mixed layer and the non-turbulent air aloft is an important but not well understood region of the atmospheric boundary layer. The most important dynamical process of the interfacial layer is entrainment, by which the turbulent boundary layer grows by mixing air from aloft into the boundary layer, causing the turbulent layer to deepen. The dominant influence on turbulence within the interfacial layer is the stable stratification induced by the capping inversion. We use a series of 26 high-resolution large-eddy simulation (LES) runs to test our hypothesis that interfacial-layer turbulence should behave like other stably stratified turbulence. The simulations range from neutral, inversion-capped runs to free-convective cases, with varying capping inversion strengths and geostrophic wind profiles.

Results from the 1968 Kansas field experiments show that spectra from the stable surface layer, when properly scaled, collapse onto universal curves, the wavelength at the spectral peak scaling with the Monin-Obukhov length L. Other researchers have verified this with numerous observations of the nocturnal PBL. Both the shapes of our interfacial-layer spectra and the locations of their peaks agree with the previous results from nocturnal PBLs.

We hypothesize that the parameters governing the interfacial-layer structure are the local vertical velocity variance, the local potential temperature gradient, and the buoyancy parameter, from which we can define a "buoyancy" length scale. The nondimensional eddy diffusivities, variances, structure-function parameters, and dissipation and destruction rates within the interfacial layer agree with the forms observed in the nocturnal boundary layer.

Interfacial-layer turbulence can be driven almost entirely by the transport of TKE from the mixed layer, whereas classic stable turbulence is driven by production from local shear and has negligible transport. We find that for a given turbulence intensity, the interfacial layer (when properly nondimensionalized) can behave like classic stable boundary layer flows. This observation may lead to improved models of the interfacial layer and of entrainment. It may also have strong implications for remote sensing of the interfacial layer, since in stable conditions measurement of the structure-function parameters can directly yield the turbulent fluxes.