Emissions and Chemistry of Reactive Nitrogen in Wildfire Plumes

Abstract

Wildfires are an important source of reactive nitrogen species to the atmosphere, accounting for approximately 25% of global annual nitrogen oxides (NOx). Emissions of wildfires are highly variable depending on factors such as fuel types and burning conditions, yet in situ measurements and quantification of primary emissions from open wildfires have been scarce. This work presents detailed observations of reactive nitrogen emissions and chemistry within wildfire plumes sampled during the Western Wildfire Experiment for Cloud chemistry, Aerosol absorption and Nitrogen (WE-CAN) aircraft campaign with unprecedented comprehensive measurements of over 250 wildfire plume transects. In this dissertation, I systematically investigate the complexities of reactive nitrogen chemistry in authentic smoke, including radical sources, photochemical evolutions, and secondary pollutant formation. I demonstrate that the emission ratios of nitrous acid (HONO) from fires may be significantly underestimated in previous studies where measurements were not close enough to fire sources, highlighting the critical need to update the values in regional air quality models. I show that HONO is the most important primary radical source in fresh wildfire plumes, contributing over 90% of hydrogen oxidies (HOx ≡ OH + HO2) production in plumes with age shorter than an hour. I then evaluate potential drivers of variability in HONO emissions across the range of fires sampled, and find a dependence on modified combustion efficiency and fuel nitrogen proxy. With the observational constraints, I examine the rapid daytime post-emission changes of reactive nitrogen using a 0-D photochemical box model. The model underpredicts the loss of NOx in fires with high NOx emissions, and I show that current model mechanisms likely miss out on a suite of oxidized organic nitrogen species such as alkyl and acyl peroxynitrates in fire plumes, consistent with a suite of organic nitrogen compounds measured by chemical ionization mass spectrometry. I find HONO mixing ratios in aged smoke are systematically higher than expected from known chemical reactions, and conducted extensive correlation analysis to identify potential secondary sources of HONO. From an ideal case study of sufficiently aged smoke, I show that the missing HONO production could be predicted by an empirical multilinear regression of two candidate mechanisms – the photolysis of particle nitrate and the aerosol heterogeneous uptake of NO2. The relationship could also be extended to all aged smoke detected in the campaign with good consistency.