EXPLORING URBAN TROPOSPHERIC OZONE AND ITS PRECURSORS USING AIRBORNE OBSERVATIONS AND CHEMICAL MODELING

Abstract

Tropospheric ozone, a harmful trace gas in the lower atmosphere, poses a serious threat to human health. This dissertation integrates data from airborne observations, ground-based monitoring networks, and satellite retrievals to characterize ozone and its precursors.Using observations from the Long Island Sound Tropospheric Ozone Study (LISTOS) and a 0-D box model, I test the sensitivity of ozone production (PO3) to ozone precursors. I find that PO3 is greater in the morning than the afternoon due to larger concentrations of NO2 and volatile organic compounds (VOC) in the morning. I further compare PO3 calculations using the near explicit Master Chemical Mechanism (MCMv3.3.1) and Carbon Bond 6 revision 2 (CB6r2) for a more direct link to regulatory air quality models and find that modeled PO3 is 20% greater using MCMv3.3.1 compared to CB6r2. This implies the regulatory air quality models may be underestimating ozone exceedances. The Pandora Global Network (PGN) is a system of ground-based spectrometers reporting continuous daytime column formaldehyde (HCHO) and NO2. Using data from the Atmospheric Emissions and Reactions Observed from Megacities to Marine Areas (AEROMMA) airborne campaign in the summer of 2023, I evaluate the performance of select Pandora monitors relative to in situ airborne observations. I find that Pandora HCHO columns agree with in situ integrated columns, however Pandoras do not capture the vertical shape of HCHO where the Pandora is biased high near the surface and low near the top of the boundary layer. Ten Pandoras in NYC capture day to day variability of HCHO as well as spatial gradients across the region. The mean NYC Pandora HCHO correlates well with mean Tropospheric Emissions Monitoring of Pollution (TEMPO) HCHO columns of a similar domain on clear sky days. Southeast Asia experiences poor air quality driven by a wide range of natural and anthropogenic emission sources, complicating interpretation of observed HCHO. I find that HCHO sources in Thailand during Airborne and Satellite Investigation of Asian Air Quality (ASIA-AQ) are driven primarily by biogenic emissions and biomass burning, with contributions from urban activity in the southern region near Bangkok. A clearer understanding of HCHO sources is therefore essential for interpreting hourly satellite column observations. While machine learning models can predict HCHO when constrained with VOC observations and other species, they rely on species that are more strongly correlated with HCHO, rather than those responsible for the formation of HCHO using a box model.