A review of ABL modelling in RANS simulations: Inlet conditions and turbulence models
Emanuele Bombardi, Alessandro Gambale, Alessandro Parente
Abstract
While Reynolds-averaged Navier–Stokes (RANS) simulations are widely used for modelling atmospheric boundary layer (ABL) flows, the combined impact of turbulence models and inlet conditions remains insufficiently understood. The present article provides guidance in modelling ABL flows for both open regions and urban contexts, emphasising the necessity of establishing fully developed equilibrium profiles as the initial flow state. We provide a comparison of turbulence models and boundary conditions from existing literature to complement the theoretical analysis through a quantitative assessment methodology based on curve-matching metrics. This evaluation spans three wind tunnel-scale scenarios of increasing complexity: an empty fetch, a single building, and an urban array configuration, providing concrete empirical evidence of model performance under identical conditions. The analysis of multiple boundary conditions reveals that SST k - ω model achieves superior performance, though its effectiveness depends on the chosen inlet conditions. In contrast, standard k - ɛ and k - ω turbulence models demonstrate lower accuracy in predicting ABL flows.