Simulation of surface ozone pollution in the Central Gulf Coast region during summer synoptic condition using WRF/Chem air quality model

Abstract

WRF/Chem, a fully coupled meteorology–chemistry model, was used for the simulation of surface ozone pollution over the Central Gulf Coast region in Southeast United States of America (USA). Two ozone episodes during June 8–11, 2006 and July 18–22, 2006 characterized with hourly mixing ratios of 60–100ppbv, were selected for the study. Suite of sensitivity experiments were conducted with three different planetary boundary layer (PBL) schemes and three land surface models (LSM). The results indicate that Yonsei–University (YSU) PBL scheme in combination with NOAH and SOIL LSMs produce better simulations of both the meteorological and chemical species than others. YSU PBL scheme in combination with NOAH LSM had slightly better simulation than with SOIL scheme. Spatial comparison with observations showed that YSUNOAH experiment well simulated the diurnal mean ozone mixing ratio, timing of diurnal cycle as well as range in ozone mixing ratio at most monitoring stations with an overall correlation of 0.726, bias of –1.55ppbv, mean absolute error of 8.11ppbv and root mean square error of 14.5ppbv; and with an underestimation of 7ppbv in the daytime peak ozone and about 8% in the daily average ozone. Model produced 1–hr, and 8–hr average ozone values were well correlated with corresponding observed means. The minor underestimation of daytime ozone is attributed to the slight underestimation of air temperature which tend to slow–down the ozone production and overestimation of wind speeds which transport the produced ozone at a faster rate. Simulated mean horizontal and vertical flow patterns suggest the role of the horizontal transport and the PBL diffusion in the development of high ozone during the episode. Overall, the model is found to perform reasonably well to simulate the ozone and other precursor pollutants with good correlations and low error metrics. Thus the study demonstrates the potential of WRF/Chem model for air quality prediction in coastal environments.