Snow precipitation in mountains surrounded by semi-arid regions represents an important reservoir of fresh water during the melting season. The snow cover helps to compensate for the scarce precipitation that occurs during their long summer droughts. Knowing the phenomenology that leads to winter precipitation and snow at these areas becomes even more relevant in a context of climate change. Precipitation in Sierra de Guadarrama, a medium size mountain range in the middle of the Iberian Plateau, is the main source of fresh water for millions of inhabitants living under its area of influence, for an active industry and for agriculture and farming. In addition, scarce but heavy snow events affect logistics, transport and security in an area with abundant ground and air traffic. This work analyses the links between large scale atmospheric patterns and the complex winter precipitation and snow cover dynamics observed at local scale. Applying principal component analysis and K-means clustering on geopotential height field, a set of circulation weather types are obtained. The contribution of each circulation weather type to precipitation, snow and heavy snow events is analysed, and favouring conditions leading to snowfalls are identified. Results from this work can be useful as a framework for future modelling exercises, statistical downscaling of climate change scenarios, or even for the development of early warning systems.