Larger, more frequent boreal forest fires threaten legacy carbon stores

Aerial view of fires burning in Northwest Territories, Canada in 2014.

Aerial view of fires burning in Northwest Territories, Canada in 2014.

Photo courtesy of NASA/Peter Griffith

Pools of old carbon in the soil of boreal forests historically safe from combustion are being released by more frequent and larger wildfires, a team led by researchers at Northern Arizona University (NAU) announced in Nature this week. As the climate of these forests in the Northwest Territories of Canada becomes warmer and drier and prone to more frequent fires, the combustion of “legacy carbon,” as the research team calls it, has the potential to shift the global carbon cycle, as boreal forests that have acted as carbon sinks for millennia become sources of atmospheric carbon.

The research team, which included collaborators from NAU, the Government of the Northwest Territories (NWT), four Canadian universities, University of Alaska – Fairbanks, and Woodwell Climate Research Center (formerly Woods Hole Research Center), found that legacy carbon remained protected from combustion in older stands where the historical fire return interval persisted. But in stands younger than 60 years old, legacy carbon burned.

“By defining and analyzing ‘legacy carbon,’ this paper offers a new way to think about long-sequestered carbon stocks in boreal forests and how vulnerable they are to being burned during increasingly frequent and severe wildfires,” said Brendan Rogers, Assistant Scientist at Woodwell Climate and co-author of the study. “This tool helps us understand when burning goes ‘outside the norm’ from a historical perspective and begins to combust carbon stocks that survived past fires. Second, this study further emphasizes why more frequent burning in the boreal forest is bad from a climate perspective.”

“In older stands that burn, this carbon is protected by thick organic soils,” said Xanthe Walker, lead author and postdoctoral researcher at the Center for Ecosystem Science and Society (Ecoss) at Northern Arizona University. “But in younger stands that burn, the soil does not have time to re-accumulate after the previous fire, making legacy carbon vulnerable to burning. This pattern could shift boreal forests into a new domain of carbon cycling, where they become a carbon source instead of a sink.”

For Walker and senior author Michelle Mack, who first identified these protected pools of carbon in fire scars in Alaska, a series of mega-fires in the Northwest Territories in 2014 offered an opportunity to ask whether legacy carbon was being combusted, and where. “These were large and severe fires, and we thought: this is when and where it would burn,” Walker said.

“We know that there is really old carbon in these soils—carbon that is hundreds to thousands of years old, carbon that is irreplaceable,” said NAU’s Mack, who worked with Walker and NAU collaborators Chris Ebert, Scott Goetz and Ted Schuur on the study. “As high-intensity fires begin to burn these stores, we think of this as ‘mining’ soil carbon. Globally, boreal forests store about one-third of terrestrial carbon, primarily in soils, so transferring that carbon from soil to the atmosphere could be a powerful accelerating feedback to climate warming.”

To ensure they sampled a full spectrum of burned areas, Steve Cumming of Laval University in Quebec identified plots in the burn area with maps derived from remote sensing of forest composition. Then Walker and a large field campaign of researchers and students from Canadian universities hiked into the burn areas to sample the soil at 211 of these plots.

“Carbon accumulates in these soils like tree rings, with the newest carbon at the surface and the oldest carbon at the bottom,” said Mack. “We thought we could use this layering to see how far back in time, in the history of the forest, fires were burning.”

To estimate the age of the carbon in the soil, the team used radiocarbon dating, which measures the abundance of 14C carbon isotope in a sample of soil. Carbon dating allows researchers to measure the amount of 14C and gather clues about how long certain carbon stores have been in the soil. This technique requires a high-precision measurement made by relatively few labs in the world; in early 2020, the Center for Ecosystem Science and Society will install a Mini Carbon Dating System at NAU.

In nearly half (45%) of the young stands the researchers sampled, legacy carbon burned. And while the amount of legacy C didn’t alter total carbon emitted from these fires, the pattern the researchers identified has global implications for future climate scenarios. “The frequency of boreal forest fires is projected to increase even more with expected climate warming and drying, and as a result total burned area is expected to increase 130-350% by mid-century,” the authors write, expanding the proportion of young forests vulnerable to burning and loss of legacy carbon.

“The 2014 wildfires were unlike anything the Northwest Territories had experienced, and after them, the Government of the NWT sought to understand the impacts of extreme wildfire on the lands and people of the NWT so they could better plan and adapt,” said Jennifer Baltzer, a researcher at Wilfrid Laurier University in Ontario and co-author of the Nature study. “This research, with their help and partnership, has really advanced our understanding of these fires and the tremendous impact extreme wildfire years have on globally critical stores of carbon.”

“Older forests dominate the landscape, so it’s good news that legacy carbon is protected in these stands,” Walker said. “Younger stands often act as living fire breaks because the fuels that ignite and spread fire have not accumulated yet, and this is also good news for protecting legacy carbon. But as the climate continues to warm, we may see new conditions in which young stands burn and carry fire.”


This project was a collaborative effort primarily supported by funding awarded to Michelle Mack by the NASA Arctic Boreal Vulnerability Experiment (ABoVE) and a National Science Foundation (NSF) Division of Environmental Biology (DEB) Rapid grant. Additional funding was provided by Natural Sciences and Engineering Research Council (NSERC) Discovery Grants awarded to Jill Johnstone and Merritt Turetsky, the Northwest Territories Cumulative Impacts Monitoring Program awarded to Jennifer Baltzer, an NSERC postdoctoral fellowship awarded to Nicola Day, and Polar Knowledge Canada’s Northern Science Training Program awarded to Canadian field assistants. Logistical support was provided by the Government of the Northwest Territories – Wilfrid Laurier University Partnership Agreement.

Research area