Abstract. In alpine regions, wind-induced snow transport strongly influences the spatio-temporal evolution of the snow cover throughout the winter season. To gain understanding on the complex processes that drive the redistribution of snow, a new numerical model is developed. It directly couples the detailed snowpack model Crocus with the atmospheric model Meso-NH. Meso-NH/Crocus simulates snow transport in saltation and in turbulent suspension and includes the sublimation of suspended snow particles. The coupled model is evaluated against data collected around the experimental site of Col du Lac Blanc (2720 m a.s.l., French Alps). First, 1-D simulations show that a detailed representation of the first metres of the atmosphere is required to reproduce strong gradients of blowing snow concentration and compute mass exchange between the snowpack and the atmosphere. Secondly, 3-D simulations of a blowing snow event without concurrent snowfall have been carried out. Results show that the model captures the main structures of atmospheric flow in alpine terrain. However, at 50 m grid spacing, the model reproduces only the patterns of snow erosion and deposition at the ridge scale and misses smaller scale patterns observed by terrestrial laser scanning. When activated, the sublimation of suspended snow particles causes a reduction of deposited snow mass of 5.3% over the calculation domain. Total sublimation (surface + blowing snow) is three times higher than surface sublimation in a simulation neglecting blowing snow sublimation.