The main objective of this study is to improve high-resolution seasonal snow cover predictions for large river basins in the Alps. Key to these improvements are the coupling of physically based models for both atmospheric and snow cover processes to better resolve processes that act at the snow - atmosphere interface. We propose a one-way and two-way coupling of atmospheric models COSMO and ICAR to a multi-layer snow model (FSM). The coupling between these models will allow us to: i) develop a model for preferential deposition of snowfall, ii) assess snow albedo-temperature feedbacks iii) to assess local heat advection effects on snow melt dynamics. Numerical studies will be complemented by comprehensive field experiments that will be designed to measure boundary layer dynamics over snow driving sub-grid variability of mean meteorological variables. Overall, the project will lead to a better quantitative description of wind-driven coupling processes over snow in large mountain regions.