Using Model Predictions, Ground-Based Observations, and Satellites to Evaluate Ozone Exceedances over the US Northeast Corridor

Tropospheric ozone forms the main component of smog. It is generated from precursors like nitric oxides (NO and NO2) and Volatile Organic Compounds (VOCs) that react with each other while exposed to sunlight. People in zones affected by ozone exceedances suffer from an increased risk of long-term respiratory damage, making air quality prediction models vitally important for public health. We chose the Weather Research and Forecasting Model coupled with Chemistry (WRF-Chem) to evaluate during the summer of 2023 using EPA AirNow ground monitoring data and Tropospheric Ozone Lidar NETwork (TOLNET) measurements deployed in the Northeastern US during 2023. We examined the period of July 27^th to 29^th to compare the model predictions of surface ozone levels with AirNow readings as well as compare 8-km vertical ozone profiles from the model with lidar curtains from four TOLNET sites. In addition, we use August 9^th measurements from the recently launched Tropospheric Emissions: Monitoring of Pollution (TEMPO) sensor (the closest available period with sufficiently low cloud fraction over the Northeast corridor) to compare satellite-observed column abundances with modeled NO2 and HCHO columns for the same timeframe. Our findings suggest that, due to overestimates in modelled NO2 against what is seen in satellite columns, the model overestimates surface ozone during high-ozone events, but it is still capable of accurately capturing the vertical distribution of tropospheric ozone observed by lidar. These results emphasize the necessity of multiple sources of field data when assessing the predictive accuracy of ozone variables in atmospheric models.