In temperate forests, changes in snow depth associated with global climate change could affect soil biogeochemical cycles; however, the seasonal effects of snow depth on soil microbial biomass and enzyme activity and the underlying mechanisms remain unclear. Hence, a snow depth manipulation experiment that included three treatments was performed from October 2014 to October 2015 in a temperate forest in Northeast China. The soil temperatures, moisture content, chemical properties, microbial biomass, enzyme activity, litter decomposition rate and snow depth were measured at six stages. The results showed that snow-removal increased the contents of soil DOC and DON during the deep freezing stage, the thawing stage and the following snow-free season, while the reverse patterns were observed in the snow-addition treatment. The soil microbial biomass decreased with decreasing snow depth during the onset of the freezing stage and deep freezing stage due to the insulation provided by the thick snowpack; however, it decreased with decreasing snow depth in the thawing stage due to the low amount of snowmelt water under thin or no snowpack conditions. The soil enzyme activity increased under deep snowpack but decreased when snowpack was absent. Compared with the control, the snow-addition treatment increased soil invertase and urease activities by more than 4.97% and 13.97%, while the snow-removal treatment decreased the corresponding values by more than 8.09% and 20.54%, respectively. Additionally, the controlling factors of soil microbial biomass and enzyme activity also varied between the snow-covered and snow-free seasons. In conclusion, the soil microbial biomass and enzyme activity in this temperate forest were significantly influenced by snowpack, although their directions and drivers were affected by snowpack depth and season. We should pay further attention to the effect of snow depth on ecological soil processes, not only in the snow-covered season but also in the snow-free season.