The local snow distribution in complex alpine terrain is strongly influenced by small-scale terrain-flow-precipitation interaction processes such as local orographic precipitation enhancement and preferential deposition. To improve avalanche forecasting, prediction of seasonal snow water resources and flood prevention, it is important to quantitatively understand these processes. Our large-eddy configuration of the Weather Research and Forecasting model (WRF) shows that a horizontal grid spacing of ≤ 50 m is required to resolve local orographic precipitation, lee-side flow separation, and thereby preferential deposition. Two case studies in the upper Dischma valley (Davos, Switzerland) demonstrate that at this resolution precipitation patterns across mountain ridges have a strong spatial and temporal variability which can be explained by atmospheric humidity and stability conditions in agreement with theoretical concepts. Based on our case study, the overall effect of terrain-flow-precipitation interactions may increase snow accumulation on the leeward side of the mountain ridge by about 26-28% with respect to snow accumulation on the windward side. Using a simple method to distinguish preferential deposition and orographic precipitation processes, results indicate that cloud dynamics and mean advection may locally lead to a 20 % increase of precipitation on the leeward side compared to the windward side. Preferential deposition accounts for a precipitation increase of up to 10 % on the leeward side of mountain ridges. Filtered snow depth measurements show a certain agreement with the modeled precipitation distribution. In future work, a full validation will require the detailed simulation of snow transport and snow depth measurements over a larger area.