Snowfall is an important precipitation system in Beijing. In this paper, observation data combined with the mesoscale numerical simulation model WRF are used to analyze microphysical processes of two different snowfall cases in northern mountain area of Beijing. The results shows that: (1) Because of the great differences in weather situation and water vapor transport during these two snowfall processes, the microphysical mechanism of these two snowfall has great differences. In period I of November 5, 2015, as a result of strong water vapor transport, there is more liquid water in cloud. Deposition and rimming processes are main formation processes for snow particles. Deposition and coalescence are main formation processes for snowfall cases in January 24 and period II of November 5 for the weak water vapor transport. (2) In period I of November 5, there is ice (9%), snow (72%), cloud liquid water (6%) and rain water (12%) in the air. The snow particles grow with deposition (78%) and rimming (20%) processes. In January 24, there is ice (28%) and snow (72%) in the air, while the period II of November 5 has similar portion: ice (11%) and snow (88%). Snow particles grow through deposition and coalescence in January 24 and period II of November 5. (3)Updraft has great influence of ice, snow grow process because it brings sufficient water vapor. The vertical motion and water vapor makes more snow for strong deposition, rimming and coalescence processes.