Modeling scale variant stratified three-dimensional turbulent fluxes in the atmospheric surface layer

Abstract:

Accurate representation of the heterogeneous surface boundary layer is essential for numerical weather prediction with sub-kilometer grid spacing. Second-moment turbulent models are widely used for parameterizing the planetary boundary layer (PBL) in numerical weather prediction models such as the Weather Research and Forecasting (WRF) model. The most common parameterizations follow Mellor and Yamada (1982), and typically only account for the contribution of the vertical divergences of the vertical turbulent fluxes. The horizontal tendencies are parameterized based on a Smagorinsky scheme that is an approach devised for numerical stability. Although the latter is a successful approach at coarse resolutions (e.g., grid-size dx ~ 12 – 2 km), the influence of horizontal gradients becomes more critical as resolution increases (< 1 km). A full three-dimensional PBL scheme (3DPBL) was implemented in WRF recently to reconcile the representation of the vertical and horizontal turbulent mixing in the surface layer. The 3DPBL parameterization was coupled to a diagnostic model of the three-dimensional second-order turbulence properties of the flow in the surface layer. Several adjustments, including a modified length-scale, were introduced to capture flow anisotropy and stability conditions. Here, the near-surface diagnostic variables are analyzed and compared against data from the Weather Forecast Improvement Project II (WFIPII) for different weather regimes and using different grid resolutions to examine stability- and scale-dependent behavior. The impact of integrated horizontal mixing is compared with the commonly used PBL parameterization approaches.