Northern Wildfires Release Ancient Carbon from Deep Peat Soils, Study Finds

Climate scientists have long known that wildfires release carbon. What new research is revealing is that the fires burning through boreal forests and peatlands are going much deeper than models assumed — reaching carbon that has been locked in the ground for hundreds, sometimes thousands of years. That distinction matters enormously. When a forest fire burns surface vegetation, the carbon released was recently absorbed from the atmosphere. When it burns deep peat, it's unlocking an ancient carbon debt that the Earth had effectively put in long-term storage. The climate accounting changes significantly when you factor in how much of that ancient store is now being combusted.

Why Peat Is Not Like Other Soil

Peat forms in waterlogged conditions where plant material decomposes extremely slowly. Instead of breaking down and releasing its carbon back into the atmosphere through normal biological decomposition, the organic matter accumulates layer by layer over millennia. A single peat deposit can represent thousands of years of accumulated carbon — a biological record and a carbon bank simultaneously. Northern peatlands, concentrated in Canada, Russia, Scandinavia, and Alaska, collectively hold an estimated 600 billion tons of carbon, more than all the world's forests combined.

That storage is stable under normal conditions — cold temperatures, high moisture, and saturated soils keep the peat intact. But the conditions that maintained those deposits are changing. The Arctic and subarctic regions are warming at roughly twice the global average rate. Permafrost that underlies much of the peatland is thawing. Summers are getting longer and drier. And the fires that have always been part of the boreal ecosystem are burning more frequently, more intensely, and — critically — deeper into the soil than fire behavior models were designed to anticipate.

What the New Research Actually Found

The study used radiocarbon dating to analyze the age of carbon released by boreal wildfires — a method that can determine whether emitted carbon came from recently living plants or from ancient organic material. The results showed that a substantially larger fraction of wildfire emissions came from old, deep peat than existing models accounted for. Some of the carbon being released had been sequestered for centuries or longer.

This finding has direct implications for how wildfire emissions are estimated. Current climate models typically assume that fires burn mostly surface vegetation and shallow organic layers. If fires are routinely reaching deeper peat deposits, the total carbon released per burned hectare is higher than models calculate — and the warming feedback from those fires is more powerful than projections suggest. The study doesn't just identify a data gap; it points toward a systematic underestimation baked into the tools that climate scientists and policymakers use to track and project emissions.

The Feedback Loop That Makes This a Self-Amplifying Problem

What makes deep peat combustion particularly concerning from a climate perspective is the feedback dynamic it creates. Fires warm the atmosphere by releasing carbon. A warmer atmosphere dries peatlands, thaws permafrost, and extends fire seasons. Drier peatlands burn more deeply in subsequent fires, releasing more ancient carbon. The cycle reinforces itself, and each turn of the loop potentially accelerates the next.

Permafrost thaw compounds the problem independently of fire. As permafrost degrades, it destabilizes the frozen peat above it, making it more susceptible to burning and to aerobic decomposition — which releases carbon even without combustion. Researchers studying northern carbon budgets are increasingly concerned that these systems, which have functioned as net carbon sinks for most of human history, are transitioning toward becoming net carbon sources. The wildfires are one pathway for that transition; thaw-driven decomposition is another.

Why Climate Models Have Underestimated This

Climate models are built from historical data and parameterized to reflect observed relationships between variables. For most of the period when sophisticated climate modeling developed, boreal fires were not burning as deeply into peat as they are now — so the models were calibrated against behavior that has since changed. This is a known challenge with climate projections: the models are better at capturing processes that have been directly observed than at projecting how those processes shift as conditions move beyond historical norms.

The researchers are calling for updates to fire emission factors in major carbon accounting frameworks — the parameters that translate area burned into estimated emissions. If the correction is substantial, it would affect not just scientific projections but also national emissions inventories that countries submit under international climate agreements. Canada and Russia, which host the world's largest boreal forest and peatland systems, would see their reported wildfire emissions revised upward.

The Scale of What's at Stake

Northern peatlands represent a carbon store that took thousands of years to accumulate. Fire seasons in boreal regions have been expanding — Canada's 2023 fire season was by some measures the most destructive on record, burning tens of millions of hectares. If a meaningful fraction of that area included deep peat combustion releasing ancient carbon, the actual climate impact of that single season was considerably larger than official estimates captured.

The study doesn't offer an easy solution — you can't put ancient carbon back into peat once it's been burned. What it does is sharpen the scientific community's understanding of how much is being lost and why the trajectory matters. Protecting northern peatlands from fire, where possible, and slowing the warming that drives deeper burn conditions is now an even more urgent priority than existing climate frameworks fully reflect. The numbers need to be updated, and so do the policies built on them.