Development of roughness models for multiscale topography in neutral and stable atmospheric boundary layers

The atmospheric boundary layer (ABL) is governed by the interaction of the wind system with surface topography. Very often, topography has a multiscale structure, with scales ranging from hundreds of meters down to submillimeter level. The representation of multiscale roughness in simulations remains a major challenge, limiting predictive capabilities in application areas such as wind energy, pollutant dispersion, etc. To date, homogeneous roughness models are used that are simply superimposed on resolved terrain features. However, in recent years, it was found that roughness elements can lead to secondary motions that extend into the outer layer of the boundary layer, resulting in significant horizontal inhomogeneity. Only recently, a theoretical framework was introduced (Meyers, Ganapathisubramani & Cal 2019) that relates roughness length scales to the decay of secondary motions in the outer layer. This enables the formulation of roughness models that include effects of secondary motion. We aim at developing such models for large-eddy simulations of the ABL, and extending understanding towards non-neutral stratification. Whereas studies to date are limited to neutral cases, we study for the first time the effect of multiscale roughness in stable stratification by setting up an experiment in the stratified wind tunnel of Portland State University. We also perform a suite of direct numerical simulations, allowing for the parametrization of wind veer and low-level jets.