Using data gathered from a specially equipped jet that spent a month flying through and studying wildfire plumes, scientists have a better understanding now of how wildfire smoke impacts air quality. 

Crucially, they found a mechanism for predicting the production of the pollutant ozone—which, at ground level, can create poor breathing conditions and also harm ecosystems. Also, the team found that mixing wildfire smoke with urban pollution ramps up the production of ozone, meaning that wildfires upwind of cities are a recipe for air quality problems.

"Of course it is well known that wildfires lower air quality. But it's important to understand the chemical and physical mechanisms by which they do so that we can more effectively forecast how individual fires will impact the communities downwind of them," said Paul O. Wennberg, R. Stanton Avery Professor of Atmospheric Chemistry and Environmental Science and Engineering at the California Institute of Technology and a corresponding author of a paper published today in the journal Science Advances.

The paper draws on data collected through the NASA/NOAA FIREX-AQ field campaign, which spent a month flying missions out of Boise, Idaho, during the summer of 2019. Riding in a DC-8 that had been converted into a flying laboratory, scientists flew through smoke plumes and gathered information from instruments mounted on the plane. The research team included scientists from Caltech, CIRES, CU Boulder, NOAA, NASA, and other institutions.

"Smoke is difficult to study," said study lead author Lu Xu, who completed the work as a staff scientist at Caltech and is now a CIRES scientist working in NOAA's Chemical Sciences Laboratory. "The chemistry and mixing evolve at scales much smaller than the resolution typically available by remote sensing. Also, it is operationally difficult to sample even from aircraft given the lack of visibility and challenges with air traffic control. FIREX-AQ overcame most of these barriers with a very well instrumented aircraft with flight planning that was highly coordinated with the local air traffic control and the incident control officials fighting the fires." 

The FIREX-AQ campaign collected unprecedented detailed information about the evolving chemistry in the wildfire plumes. This included key observations that explain the chemistry that produces ozone in these fire plumes. The findings from this analysis help researchers understand why fire plumes have widely diverse chemistry: just downwind of the fire, ozone production is vigorous, but the rate of its formation slows as the plume mixes with ambient air. Initially, the plume is primed with a chemical, nitrous acid (HONO, also written as HNO₂) that, once emitted into the atmosphere, is converted by sunlight to hydroxyl radical (OH) and nitric oxide (NO)—both key ingredients that drive the formation of ozone from volatile organic compounds (VOCs). 

Xu, Wennberg, and their colleagues then pored over the data and demonstrated that the production of ozone could be predicted as a function the amount of HONO emitted from the fire, the amount of VOCs, and the amount of NO. As the fire plume runs out of NO and HONO, the chemistry slows to a standstill, but, importantly, the fuel for ozone formation—the VOCs—remains elevated. As such, when these fire plumes mix into urban atmospheres rich in nitric oxide—which is produced by fossil fuel combustion, for example, from cars and trucks—ozone formation will begin again, impacting air quality.

"Wildfires increase regional ozone across the western United States during fire season. The episodic nature of wildfires can result in more severe impacts on areas in close proximity to wildfires, as we in the U.S. West frequently experienced in the past few years," Xu said.

Next, the team plans to keep mining the FIREX-AQ data for additional insights. "There are a lot more discoveries about fire chemistry and air quality still to be found in the information gathered through this project," Wennberg said.

This story was adapted from a news story by Caltech communications.