CFD simulation of snow transport over flat, uniformly rough, open terrain: Impact of physical and computational parameters

In the past, computational fluid dynamics (CFD) simulations have been successfully applied for the prediction of snow drift around buildings and on building roofs. A literature study indicates that a wide range of influential computational and physical parameters exist for snow drifting predictions in CFD, while the impact of these parameters is unclear, resulting in a lack of available CFD simulation guidelines. Therefore, this study presents a systematic and generic analysis with emphasis on the fundamentals of snow transport prediction techniques. Snow transport over flat, uniformly rough, open terrain, including snow saltation and snow suspension is successfully simulated using CFD and the results are compared to field measurements of snow concentrations for validation. This paper investigates the impact of grid resolution, falling velocity of snow, turbulent Schmidt number, threshold friction velocity of snow and turbulence model on the CFD simulation results. The results show that the falling velocity and the turbulent Schmidt number have the largest impact. A slight change in the falling velocity or the turbulent Schmidt number significantly impacts the predicted snow concentration in the air. In addition, the turbulence model mainly affects the turbulent kinetic energy, another key factor that influences the numerical predictions of snow transport.