Fine particulate matter (PM_2.5) is one of the major contaminants of the urban air pollution in China, which seriously threatens the health of urban residents and affects the sustainability of urban development. There are two natural ways to remove it, namely dry deposition and wet deposition, in which dry deposition plays a leading role, and the process of dry deposition is inseparable from urban forest. At present, the research on the reduction effect of urban forest on air pollutants is mainly to estimate and compare its dust retention from the perspective of different tree species, different community structures and different landscape types on a small scale, and pays less attention to its ratio to the total air pollution, which may affect the judgment of urban forest dust retention service capacity. Therefore, using the core algorithm of air pollution dry deposition module in UFORE (urban forest effect) model and taking 2015 as an example, this paper estimated the reduction of PM_2.5 in the atmosphere and its ratio to the total amount of pollutants in the air, i.e. dust retention rate in China. The results show that:(1) in 2015, the areas with high daily dust retention per unit green area in major cities in China were mainly concentrated in North China, East China and Northeast China, such as 30.47mg/m² in Beijing, 24.63mg/m² in Suzhou, and 28.55mg/m² in Shenyang. (2) The dust retention effect of urban forest has obviously seasonal effect. The dust retention amount in winter is significantly higher than that in summer, and the dust retention rate in summer is higher than that in winter. (3) Although the total amount of dust retention is large, the dust retention of urban forest to urban air pollutants accounts for less than 4% of the total amount of air pollution, highlighting the necessity of objectively understanding the dust retention function of urban green space. The research results of this paper can provide some scientific supports for urban air pollution prevention and ecological livable city construction.