Meteorological conditions are critical in the transport and final fate of air pollutants. In this paper, meteorological data simulated by three different planetary boundary layer (PBL) schemes in the Weather Research and Forecasting (WRF) model for the period January 27-February 2, 2005, were used to drive the Community Multi-scale Air Quality (CMAQ) model. The results show that discrepancies in the turbulent closure method employed by the various PBL schemes in the WRF cause a significant impact on the wind and the temperature fields in the PBL and lead to differences in pollutant concentrations simulated by the CMAQ model. Compared with monitored pollutant concentration, the CMAQ model has been determined as suitable for obtaining high spatial and temporal resolution pollutant characteristics in complex terrain areas. In addition, the results of WRF simulation with the Asymmetric Convective Model version 2 (ACM2) PBL scheme coupled with CMAQ yields more reasonable results than those of other two schemes. The local and non-local combined ACM2 scheme represents actual meteorological conditions more accurately due to its reasonable turbulence transport mechanism. Thus, usage of the same vertical turbulent mixing scheme in a coupled meteorological and air quality model would be advantageous. This tight coupling would improve mass conservation because air density and pollutants mix in the same manner, eliminating the risk of accruing mass error in mixing ratios of trace species in air mass.