Abstract: The planetary boundary layer (PBL) is an important parameter in the studies of environment and climate and it affects air pollution and evolution of atmospheric boundary layer. We analyze PBL heights retrieved using wavelet covariance transform (WCT) method from lidar backscatter profiles data from May 2011 to April 2012 and evaluate the performance of this method. PBL heights are validated using observations of one radiosonde on the ground and airborne instruments for the purpose to optimize the parameters required for lidar detection using the WCT method under different atmospheric conditions. The result shows that the PBL heights have a better agreement between lidar and airborne retrievals than between lidar and radiosonde retrievals with a correlation coefficient of 0.88. Lidar appears to slightly overestimate PBL height. If the vertical resolution of lidar is 30 m, the step length and threshold for Beijing area are 210 m and 0.05, respectively; the step length and threshold are 135 m and 0.05 respectively for the vertical lidar resolution of 15 m. PBL height shows a diurnal variation with an average maximum value of 1.3 km starting from 1400 BT (Beijing time) and lasts for three to four hours in the summer, while the average maximum value of 1.08 km lasts for about two hours in the winter. The result suggests that the direct radiation reaching the ground is reduced by cloud, which depresses the evolution of turbulence and PBL and results in lower PBL height. We also apply the PBL heights retrieved from lidar to the analysis of relationship between the concentration of PM2.5 and meteorological elements for the one-year period. The result shows that the correlation coefficient between PBL height and the concentration of PM2.5 is -0.340.