QUANTITATIVE ANALYSIS OF THE ATMOSPHERIC OXIDATION OF ISOPRENE USING MODELS AND MEASUREMENTS: IMPACTS ON SURFACE OZONE

The oxidation of isoprene – one of the most abundant volatile organic compounds (VOCs) in our atmosphere – significantly impacts the formation of surface ozone, which is detrimental to public health. Computer models simulate the complex relationships between ozone and VOCs like isoprene and are used to guide policy decisions directed at improving ozone. However, uncertainties in the emissions and chemistry of isoprene limit the accuracy of modeled ozone. This body of work comprises a quantitative analysis of atmospheric isoprene oxidation that strives to identify and improve such uncertainties through the combination of models with measurements. Measurements used in this work mainly comprise in situ observations from the Southeast Nexus (SENEX) aircraft campaign, which sampled atmospheric composition across the isoprene-rich summertime Southeast US. I have prepared two models – the Framework for 0-D Atmospheric Modeling (F0AM) and the Comprehensive Air Quality Model with Extensions (CAMx) – to drive simulations of atmospheric isoprene oxidation, which are evaluated against observations from SENEX. Using F0AM, a photochemical box model, I demonstrate that several commonly-used mechanisms significantly underestimate measured mixing ratios of formaldehyde, a high-yield product of isoprene oxidation, by 0.5–1 ppb across a wide range of NOx conditions. The consistent underestimation of formaldehyde suggests a deficit of VOC oxidation among all considered mechanisms. Although the cause for this deficit remains elusive, I provide recommendations for improving the simulated production of formaldehyde upon isoprene oxidation in the Carbon Bond version 6 revision 2 (CB6r2) mechanism, commonly used for air quality modeling. Using CAMx, a three-dimensional chemical transport model, I produce a standard air quality modeling scenario that simulates atmospheric composition across the continental US for the summer of 2013. Evaluation of this scenario reveals that the emissions of isoprene from the Biogenic Emissions Inventory System (BEIS) are underestimated in the Southeast US by at least 40%. Finally, implementation of improvements in the emissions and chemistry of isoprene within the CAMx modeling framework increases the net photochemical production of surface ozone by up to 0.5 ppb hr−1 and shifts surface ozone production regimes more NOx-limited, relative to the standard platform for regional air quality modeling