Evaluating WRF-CAMx offshore ozone simulation during TRACER-AQ 2021 in Houston, TX

Air quality models have limited confidence in their performance over water due to the lack of offshore ozone observations. From mid-August to September 2021, offshore ozone monitoring was deployed in Galveston Bay and the Gulf of Mexico in support of the TRACER-AQ (TRacking Aerosol Convection ExpeRiment - Air Quality) campaign, which provides a good opportunity for evaluating air quality models' performance over water. In this study, we conducted photochemical simulations in September 2021 using CAMx (Comprehensive Air Quality Model with Extensions) model driven by WRF (Weather Research and Forecasting) meteorology. Two high-ozone episodes, 09/06-09/11 and 09/23-09/26, were selected to evaluate the capability of the model in reproducing the observations from different aspects. The model generally captures spatial distributions of ozone but shows a high positive bias over the water during the first episode compared with the mobile boat data at the surface, which indicates the offshore transport of pollutants may not be captured properly. Vertically, the simulated ozone is lower below 1 km over Galveston Bay on most days in comparison to the ozonesonde observations, which is partly attributed to the misrepresentation of the marine boundary layer in the model. The model also missed a high ozone layer at ~2km for both episodes as presented by the ozonesonde profiles, which implies the model may underpredict the contributions of long-range transport if down mixing happens.