Quantifying and Understanding Reactive Nitrogen Emissions and Chemistry from Wildfires Through Remote Sensing and Modeling

Abstract

Wildfires constitute a substantial source of reactive nitrogen to the atmosphere, contributing 11% to 26% of global annual nitrogen oxide (NOx = NO + NO2) emissions to the atmosphere. While NOx is regularly included in biomass burning emission inventories, it is rare for nitrous acid (HONO) to be included. Lacking this source of reactive nitrogen may be detrimental to our representation and understanding of wildfire smoke chemistry, as at least 50% of hydroxyl radicals (OH) from wildfires have been estimated to originate from HONO. This work characterizes and constrains NOx and HONO emissions from wildfires using ultraviolet and visible differential optical absorption spectroscopy (DOAS) from the Fire Influence on Regional to Global Environments and Air Quality (FIREX-AQ) aircraft campaign and from the Sentinel-5P TROPOspheric Monitoring Instrument (TROPOMI).In this dissertation, I show high-resolution airborne retrievals of HONO and NO2 from the Sheridan Fire during FIREX-AQ and calculate the emission rate and lifetime of both chemical species over the course of a daily burn using 1-dimensional line densities. To reproduce the 1-dimensional line densities of HONO and NO2, an emissions and transport model required Gaussian emissions that scaled with the diurnal fire radiative power (FRP) profile. I find HONO and NO2 lifetimes of 15 to 27 minutes and 15 to 45 minutes, respectively. Additionally, I find HONO and NO2 emission rates of 0.6 to 1.35 à 1025 molec/s and 2.0 to 5.5 à 1025 molec/s respectively. TROPOMI became operational in 2018, making it available to observe the extreme Australian wildfire season during the austral summer of 2019-2020, termed the Australian Black Summer. Using these satellite observations, I further show that the ratio of HONO to NO2 increases linearly with fire radiative power (FRP) with a slope of 1.0 (±0.2) à 10-3 MW-1 and an R2 over 0.98 and is dependent on the fuel that is burning. This suggests that the calculation of HONO emissions in FRP-based biomass burning emission inventories should scale with FRP.