Abstract: Statistical analysis was synthetically conducted using conventional meteorological observation data, hourly meteorological automatic station data, hourly AQI and PM_2.5 concentration. The purpose was to investigate the relationship between meteorological factors in the boundary layer and PM_2.5 concentrations in Hebei Province. The results indicate that there is a significant difference between the boundary layer meteorological factors during clean and heavily polluted weather. The relevant conclusions are as follows:1) The conditions of sea level pressure lower than 1030 hPa, 24 h pressure variation below -3.0--2.0 hPa, surface relative humidity over 60% and dew point temperature higher than -10 ℃ are conducive for heavily polluted weather during winter. In contrast, the conditions of sea level pressure higher than 1040 hPa, 24 h pressure variation above 4.0 hPa, surface relative humidity under 40% and dew point temperature lower than -10 ℃ are beneficial for clean weather. Prevailing wind under clean weather is relevant to cold airflows in winter. In polluted weather, prevailing wind direction varies regionally, and the frequency of thin wind (lower than 3.0 m/s) in the boundary layer is above 90%. 2) Moderate and above magnitude precipitation has comparatively obvious removal effects, demonstrating the following rules:moderate-magnitude precipitation shows a rapid removal rate of PM_2.5 (around 2 h), but with a short duration (15 h on average) of excellent air quality. Heavy and above magnitude precipitation shows a high removal rate of PM_2.5 (about 67.8%), and with relatively long duration (about 27 h) of excellent air quality. 3) Removal of wind shows a more significant effect on PM_2.5 compared to precipitation. Relatively strong southerly wind significantly improves air quality but with a short duration (around 16 h). Systematic, northerly wind above 3.0 m/s shows the most significant removal rate, longest duration (32 h in average) of excellent air quality and most complete improvement of air quality (removal rate is about 85.1%). The research shows that PM_2.5 is closely related to the meteorological factors in the boundary layer, and there are significant distinctions in removal rates on PM_2.5 among different levels of wind and precipitation.