Atmospheric Chemistry and Physics (Jul 2020)

The nitrogen budget of laboratory-simulated western US wildfires during the FIREX 2016 Fire Lab study

  • J. M. Roberts,
  • C. E. Stockwell,
  • C. E. Stockwell,
  • R. J. Yokelson,
  • J. de Gouw,
  • J. de Gouw,
  • J. de Gouw,
  • Y. Liu,
  • V. Selimovic,
  • A. R. Koss,
  • A. R. Koss,
  • A. R. Koss,
  • A. R. Koss,
  • K. Sekimoto,
  • K. Sekimoto,
  • K. Sekimoto,
  • M. M. Coggon,
  • M. M. Coggon,
  • B. Yuan,
  • B. Yuan,
  • B. Yuan,
  • K. J. Zarzana,
  • K. J. Zarzana,
  • K. J. Zarzana,
  • S. S. Brown,
  • C. Santin,
  • S. H. Doerr,
  • C. Warneke,
  • C. Warneke

DOI
https://doi.org/10.5194/acp-20-8807-2020
Journal volume & issue
Vol. 20
pp. 8807 – 8826

Abstract

Read online

Reactive nitrogen (Nr, defined as all nitrogen-containing compounds except for N2 and N2O) is one of the most important classes of compounds emitted from wildfire, as Nr impacts both atmospheric oxidation processes and particle formation chemistry. In addition, several Nr compounds can contribute to health impacts from wildfires. Understanding the impacts of wildfire on the atmosphere requires a thorough description of Nr emissions. Total reactive nitrogen was measured by catalytic conversion to NO and detection by NO–O3 chemiluminescence together with individual Nr species during a series of laboratory fires of fuels characteristic of western US wildfires, conducted as part of the FIREX Fire Lab 2016 study. Data from 75 stack fires were analyzed to examine the systematics of nitrogen emissions. The measured Nr ∕ total-carbon ratios averaged 0.37 % for fuels characteristic of western North America, and these gas-phase emissions were compared with fuel and residue N∕C ratios and mass to estimate that a mean (±SD) of 0.68 (±0.14) of fuel nitrogen was emitted as N2 and N2O. The Nr detected as speciated individual compounds included the following: nitric oxide (NO), nitrogen dioxide (NO2), nitrous acid (HONO), isocyanic acid (HNCO), hydrogen cyanide (HCN), ammonia (NH3), and 44 nitrogen-containing volatile organic compounds (NVOCs). The sum of these measured individual Nr compounds averaged 84.8 (±9.8) % relative to the total Nr, and much of the 15.2 % “unaccounted” Nr is expected to be particle-bound species, not included in this analysis. A number of key species, e.g., HNCO, HCN, and HONO, were confirmed not to correlate with only flaming or with only smoldering combustion when using modified combustion efficiency, MCE=CO2/(CO+CO2), as a rough indicator. However, the systematic variations in the abundance of these species relative to other nitrogen-containing species were successfully modeled using positive matrix factorization (PMF). Three distinct factors were found for the emissions from combined coniferous fuels: a combustion factor (Comb-N) (800–1200 ∘C) with emissions of the inorganic compounds NO, NO2, and HONO, and a minor contribution from organic nitro compounds (R-NO2); a high-temperature pyrolysis factor (HT-N) (500–800 ∘C) with emissions of HNCO, HCN, and nitriles; and a low-temperature pyrolysis factor (LT-N) (<500 ∘C) with mostly ammonia and NVOCs. The temperature ranges specified are based on known combustion and pyrolysis chemistry considerations. The mix of emissions in the PMF factors from chaparral fuels (manzanita and chamise) had a slightly different composition: the Comb-N factor was also mostly NO, with small amounts of HNCO, HONO, and NH3; the HT-N factor was dominated by NO2 and had HONO, HCN, and HNCO; and the LT-N factor was mostly NH3 with a slight amount of NO contributing. In both cases, the Comb-N factor correlated best with CO2 emission, while the HT-N factors from coniferous fuels correlated closely with the high-temperature VOC factors recently reported by Sekimoto et al. (2018), and the LT-N had some correspondence to the LT-VOC factors. As a consequence, CO2 is recommended as a marker for combustion Nr emissions, HCN is recommended as a marker for HT-N emissions, and the family NH3 ∕ particle ammonium is recommended as a marker for LT-N emissions.