In recent years, with the rapid development of cities in China, air pollution has become increasingly serious as the primary problem of urban ecological and environmental damage. Therefore, exploring whether plant communities in urban green spaces can reduce the concentration of atmospheric fine particles and their changing characteristics has become the focus of widespread attention by the public and scholars. This paper selects Northeast Forestry University in the cold city of Harbin as the research object. By measuring the concentration of PM_2.5 and investigating the plant community on campus, the effects of different plant communities on the reduction of PM_2.5 concentration, the temporal variation of PM_2.5 concentration and the relationship between PM_2.5 concentration and temperature and air relative humidity were analyzed quantitatively. The results show that: (1) the diurnal variation of PM_2.5 concentration shows a "double peak and single valley" pattern, with higher morning and evening; the seasonal variation presents the lowest PM_2.5 concentration in summer and the highest PM_2.5 concentration in autumn. (2) Plant communities with different structures have slightly different effects on the reduction of atmospheric fine particles. The reduction rate of PM_2.5 in arbor-shrub-grass greenbelt is 30.30%, which is the best reduction effect; the reduction rate of PM_2.5 of arbor and shrub was 14.30% and 7.77%, and the reduction effect was poor. (3) PM_2.5 concentration has a negative correlation with temperature, and a positive correlation with air relative humidity.