Tropical peatlands burned more in the 20th century than at any point in the previous 2,000 years

Tropical peatlands store more carbon below the ground than all the world's forest biomass combined. When they burn, that carbon goes into the atmosphere. And according to a new study spanning two millennia of fire history, they are burning at a rate that has no precedent in at least 2,000 years.

What charcoal records reveal

Researchers led by Dr. Yuwan Wang at the University of Exeter analyzed charcoal preserved in peat deposits across Central and South America, Africa, Southeast Asia, and Australasia. Charcoal particles accumulate in peat layers over centuries, creating a natural archive of fire frequency and intensity that can be read like tree rings - each layer corresponding to a period in history.

The record showed a clear pattern. For more than 1,000 years, wildfire activity in tropical peatlands had been declining, tracking changes in global temperatures and natural climate variability. Cooler periods meant fewer fires. The trend was steady and unremarkable.

Then, in the 20th century, the pattern broke. Fire activity surged to levels without precedent in the 2,000-year record.

Southeast Asia and Australasia: where drainage meets fire

The increase was not global. It was concentrated in Southeast Asia and Australasia - regions where peatlands have been extensively drained for agriculture, logged, and converted to palm oil plantations and other commercial uses.

Drainage is the critical mechanism. Intact tropical peatlands are waterlogged. Wet peat does not burn easily. But when peat swamps are drained - channels dug to lower the water table for farming - the surface layers dry out and become vulnerable to ignition. A landscape that evolved to stay wet and fire-resistant becomes, through human engineering, a tinderbox.

Once ignited, peat fires are exceptionally difficult to extinguish. They can burn underground for weeks or months, spreading through subsurface organic layers even when surface flames are suppressed. The 2015 Indonesian peat fires released more carbon dioxide in three months than Germany emits in an entire year.

South America and Africa: a warning, not a reprieve

In less accessible peatland regions across South America and Africa, the charcoal records showed no comparable increase in fire activity during the 20th century. These areas have remained relatively undisturbed - fewer roads, less agricultural conversion, lower population density around peatlands.

But Wang warned that this pattern may not hold. "These regions could experience more wildfires too as population density increases and commercial agriculture and infrastructure expands," she said. The Southeast Asian trajectory offers a preview of what happens when development reaches peatlands: fires follow drainage, and once the cycle starts, the carbon consequences are enormous.

The carbon math

The stakes are not abstract. Tropical peatlands cover a relatively small area - perhaps 4% of the tropics - but store a disproportionate share of terrestrial carbon. When peat burns, it releases carbon that took thousands of years to accumulate. The emissions are functionally irreversible on any human timescale, because peat formation is extremely slow.

"To avoid large carbon emissions that further contribute to global warming we urgently need to protect these carbon-dense ecosystems," Wang said. She pointed to peatland conservation, sustainable resource management, and ecosystem restoration as necessary responses, while acknowledging that these require collaboration among multiple stakeholders and must operate at sufficiently large scale to matter.

The limits of the record

Charcoal records capture fire occurrence and relative intensity but cannot distinguish between natural and human-caused ignitions with certainty. The regional pattern - fires increasing only where human land use has been most intensive - strongly implicates human activity, but the charcoal itself does not carry a signature of causation.

The records also have varying temporal resolution across sites. Some peat cores provide annual or decadal resolution; others are coarser. The 20th-century spike is consistent across the Southeast Asian and Australasian sites, but the exact timing and magnitude of change vary by location.

The study, published in Global Change Biology, does not model future fire scenarios. But the historical trajectory is suggestive. If drainage and land conversion continue in currently undisturbed tropical peatlands, the 20th-century spike in Southeast Asia may be a preview rather than an anomaly.