Forcing snow cover models with high resolution numerical weather prediction models has provided new insight into the spatial distribution and temporal evolution of, for instance, snow depth and snow water equivalent in mountainous terrain. Tough, of particular interest to avalanche forecasting services are spatial patterns of snow instability. However, it is presently unclear what level of model complexity is required to obtain snow instability information such that the locations where the danger is most pronounced can be identified. Alpine3D is a spatially distributed (surface), threedimensional (atmospheric) model for analyzing and predicting dynamics of snow-dominated surface processes in mountainous topography. We forced Alpine3D with gridded meteorological data for the region of Davos, Switzerland (~ 20 km x 20 km; 100 m grid spacing) with the aim to identify spatial patterns of potential instability with regard to slope elevation and aspects. We simulated the snow cover using SNOWPACK - implemented in Alpine 3D - to identify spatial patterns across the model domain for two winters between 2016 and 2018. The winter season 2016-2017 was a below-average winter in terms of snow depth and hence persistent weak layers existed throughout the entire study region. In contrast, the winter season 2017-2018 showed above-average snow depth with few critical weak layers. General patterns of the two winters were reproduced by the simulations. Despite the high computational costs spatially distributed model chains show promising potential for supporting operational avalanche forecasting.