The Rim Fire burns near Buck Meadows, California.Max Whittaker / Reuters

It was a hot Monday in August 2013, just before dawn. A team of scientists stood on the tarmac in Houston. They had the keys to an unusual aircraft: a NASA-owned DC-8 jumbo jet with a laboratory inside it.

They made for a large and diverse group: chemists and geoscientists, meteorologists and climatologists, engineers and pilots. Their primary mission took them over patches of woods across the U.S. Southwest, where they measured the kind of chemicals off-gassed by trees.

But on that particular morning they took the plane slightly north, toward the Sierra Nevadas, where an enormous wildfire—known as the Rim Fire—was chewing through ancient pine forest. And after a couple of hours of flying, once they got to the fire, they drove the plane straight into its enormous plume of smoke—and then they did it again, and again.

This was not the first time they had flown the plane through a smoke cloud. NASA’s DC-8 had made two earlier trips to different wildfires in Oregon and Washington that summer. It measured towers of soot and ash close to the ground and more than 10 miles in the air, from the bottom of the troposphere to the lower reaches of the stratosphere.

“The thing that you’re always surprised by when you fly through the fire is that everything that you can measure, you see enhanced,” said Greg Huey, an atmospheric chemist at Georgia Tech and one of the leaders of the mission. “You just get big signals for everything in one of these fires.”

And they got one big signal in particular. The initial results of the three flights were published this month in the Journal of Geophysical Research: Atmospheres. The research could dramatically revise scientists’ understanding of how much pollution is caused by wildfires.

Specifically, Huey and his colleagues found that wildfires seem to emit three times as much of the smallest type of particulate matter—PM1, particulate matter less than 1 micrometer thick—as is commonly estimated. This would be enough to notably worsen air quality in nearby cities.

However, they also suggest a solution. When the NASA plane observed prescribed fires—that is, a wildfire that had been lit and controlled by local authorities—air pollution fell significantly. Prescribed burns have previously been found to reduce wildfire risk; this research suggests they might be healthier, too.

NASA’s DC-8 flying laboratory has been all over the world, including to Mount Vinson, the tallest peak in Antarctica. (NASA)

Particulate matter is the deadliest form of air pollution. It’s not a specific chemical, like ozone or nitrous oxide, but any kind of tiny dust that seeps into the lungs, where it can lead to higher blood pressure, heart attacks, and lung cancer. Particulate matter that measures 2.5 micrometers across, or PM2.5, is one of the most tightly regulated and most common forms of air pollution. Less is known about the health effects of PM1.

Huey limited his warning about PM1 to speculation. “The smaller particles can probably deposit into the lungs better—anything that deposits to the lungs more efficiently is probably an issue,” he told me. “Smaller aerosol tends to follow the gas into your lungs pretty easily.”

The study is “high-quality and important work,” said Sean Raffuse, a fire and atmospheric scientist at the University of California, Davis, who was not connected to the research. Raffuse led work on the 2011 National Emissions Inventory, which provided the baseline estimates of particulate-matter pollution from wildfires.

“This is strong evidence that our PM estimates are low for at least a subset of wildfires, probably many of them. The models we use to estimate wildfire emissions depend on this kind of research, as they themselves are based on very limited information,” he told me in an email. While recent experimental work has suggested that these estimates might be low, especially for particulate matter, it has not predicted this great a disparity.

“Perhaps the biggest weakness [in this research] is that the researchers are extrapolating measurements from three wildfires across the entire western U.S.,” Raffuse noted. “One of them, the Rim Fire, was one of the largest in California history, so it might not be representative of wildfires in, say, Colorado. The landscape, fuels, and burning conditions vary dramatically even within single fires, and certainly across the U.S.”

Huey and his colleagues also produced detailed inventories of the trace gases and aerosols released by the fires. What stands out about those numbers is the intensity: Because wildfires burn everything in their path, but do it incompletely and inconsistently, they produce an especially dirty and complicated chemical pattern.

This won’t be the last paper to emerge from the flights. The team flew every two to three days for the duration of the six-week mission. And they covered a lot of range. “When you’re flying out of Houston, boy you can hit a lot of things in the continental U.S.,” Huey said.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.