In order to explore the boundary layer characteristics, the multi-source observation data such as automatic meteorological station data, environmental hourly concentration data, microwave radiation data, wind profile radar, aerosol lidar and WRF-Chem source tagged method were used to analyze a typical fog-haze episode in Tianjin from December 7 to 10, 2019. The results showed that: The fog-haze process can be clearly divided into four stages of fog formation, fog and haze alternation, haze, and haze dissipation. Fog-haze weather was closely related to atmospheric temperature stratification, with the generation of temperature inversion, the maximum growth rate of relative humidity and liquid water content reached 13.44%/h and 0.013g/(m3·h), respectively, showing an explosive growth up to 92% of the relative humidity, and the microwave radiation data can better predict the formation of fog. In the stage of fog and haze alternation, fog weather changed the structure of the boundary layer. The atmosphere in the fog layer was in a neutral state, being relatively conducive to the pollutant diffusion in the fog area. The high value of PM2.5 concentration mainly occurred below 400m, and the continuous inversion at the fog top inhibited the diffusion of pollutants to the upper atmosphere, resulting in the aggravation of pollutant concentration in the fog area and an increase in surface PM2.5 up to 135~223µg/m3 near to ground, maintaining moderate-severe pollution. There was a good relationship between fog-haze weather and vertical wind field. In fog and haze alternation stage, there were two situations, low wind speed and high wind speed (the southwesterly wind brought abundant water vapor), which are beneficial to the maintenance of fog. The wind speed was 6~12m/s above the fog top inversion and 1~2m/s in the fog layer. The existence of fog was not conducive to the improvement of air quality near the ground. During the fog-haze process, the local emissions contributed 36.1% to surface PM2.5 concentration in Tianjin, and the contribution of regional transport was 63.9%, demonstrating obvious regional transport characteristics of the whole pollutant process. © 2022 Chinese Society for Environmental Sciences. All rights reserved.