Quantifying the transition from Large-eddy to Mesoscale boundary-layer simulations under Different Forcings

Operational high-resolution numerical weather prediction models are now able to partially resolve turbulent motions due to increasingly available computing resources. This has raised the question whether existing parameterizations are able to represent physical processes over this range of grid resolutions, as their initial assumptions may be no longer valid. In this study, a series of large eddy simulations (LES) under different atmospheric and surface forcings are conducted in order to examine the behavior of resolved turbulence over the grey zone. At first, the characteristics of the upscaled reference LES fields, constructed using the coarse-graining method, are examined in comparison with the actual simulations varying grid resolution. Next, differences between constructed and actual fields are used to establish the impact of the sub-grid parameterization, as turbulent kinetic energy (TKE) becomes more unresolved at intermediate scales towards lower resolution domains. It is shown that, in the presence of shear, turbulence becomes more resolvable as convective cells are organized into larger structures. However, in shear-induced LES, resolved TKE is damped at higher resolutions compared to buoyancy driven runs, leading to the fastest decay of resolved TKE. In fact, all LES are unable to resolve the turbulent field above the Δx/zi=0.6 limit. Decreasing numerical dissipation through the sub-grid scheme leads to the increase of resolved turbulence in conjunction with the coarse-grained results, but fails to reproduce the theoretical transition pattern. Finally, some preliminary similarity relationships are developed using the evolution of resolved TKE through the grey zone in the middle of the boundary layer.