Old-Growth Forests Store Far More Carbon Than We Thought

The soil beneath an old-growth boreal forest is not especially dramatic to look at. Dark, damp, replete with the slow chemistry of centuries: a mat of needle-litter, then the hummus layer, then the mineral soil going down a metre or more, streaked brown and grey. Unremarkable, you might think. Wrong. A study published today in Science has found that the top metre of soil in Sweden’s primary forests holds roughly 64% of all carbon in the ecosystem, more carbon on its own than the combined total of living trees, dead wood, and soil in a managed forest.

That finding overturns assumptions that have shaped both forest policy and global carbon accounting for decades. Boreal forests, which wrap around the planet’s northern latitudes in a belt wider than the Amazon basin, absorb roughly 30% of humanity’s annual carbon dioxide output.

Scientists have generally assumed that converting old-growth boreal forest to industrial timber production involved a meaningful but bounded carbon penalty: some lost biomass, some disrupted soil, offset at least partially by carbon locked into wood products, the timber frame of a house, say, or engineered flooring. The new research, led by Didac Pascual and Anders Ahlström at Lund University and Rob Jackson at Stanford, suggests that picture was badly wrong, and that the penalty has been underestimated by a factor of between 3 and 8.

The numbers, bluntly stated, are startling. Primary forests store about 72% more carbon per hectare than managed secondary forests when you credit managed forests for all carbon in harvested wood products. Strip out that credit and the gap widens to 83%.

The study is the most comprehensive accounting of Swedish old-growth carbon ever attempted. Because no national map of primary forest existed, Ahlström and colleagues spent close to a decade first identifying forests that had been little affected, or not affected at all, by direct human activity, then conducting extensive fieldwork: nearly 220 soil pits dug to a metre’s depth, measurements taken at more than 200 plots across the country, combined with decades of national forest and soil inventory data. The total difference in carbon storage works out at around 9.9 kilograms of carbon per square metre, a figure somewhere between 2.7 and 8 times larger than what the Global Carbon Project’s models currently report.

“The most surprising result is the large amounts of carbon stored in the soil of old-growth forests,” says Ahlström. “It is the same amount as all the carbon in managed forests, trees, dead wood, and soil, combined.”

What drives this difference? Partly it is simply time: old-growth forests are old, and have been accumulating organic material in their soils for centuries without interruption. But it may also reflect something about the biology of intact forest ecosystems. Jackson, a professor of Earth system science at Stanford, notes that industrial forestry involves furrow-cutting, drainage ditches, and prescribed burns that may disrupt soil structure in ways that persist long after logging ceases. There could be a microbial dimension too. Jackson is now collaborating with Stanford biologist Kabir Peay to look at whether old-growth forests harbour a more diverse and carbon-retentive community of fungi and bacteria in their soils. “Our goal is to understand what makes the fungi and bacteria in these old-growth forests unique,” says Peay, with an eye toward eventually transplanting those properties to recovering managed forests, without having to wait centuries for natural old-growth to develop.

The climate bookkeeping implications are uncomfortable. Most pathways to net-zero emissions assume increased use of northern forests for bioenergy production, burning biomass to generate electricity while theoretically allowing regrown trees to reabsorb the carbon. That logic depends on managed forests storing roughly as much carbon as the primary forests they replaced. They don’t. “The loss of soil carbon through industrial management is persistent and shocking,” says Jackson. Pascual puts the policy problem plainly: the carbon stored in wood products, mostly paper and bioenergy where carbon returns to the atmosphere within years, is nowhere near enough to compensate for what is lost from dead wood, living trees, and above all from soil.

There is a detection problem too, one that has let boreal deforestation escape the kind of scrutiny routinely applied to tropical forests. Satellite sensors can easily tell oil palm from primary rainforest. In boreal regions, spruce, pine, and birch dominate both old-growth and managed stands; from above, they look much the same. Between 2003 and 2019, Sweden lost unprotected old-growth forest to clear-cutting at roughly 1.4% per year, about six times the current rate of primary forest loss in the Brazilian Amazon. It passed largely unnoticed.

The study is specific to Sweden, and Ahlström is careful to note that the intensive management regime common there may mean the results don’t translate directly to less-intensively managed boreal regions in Canada or Russia. More field data from those places is needed.

Still, the scale of what this study suggests is hard to shrug off. Restoring Sweden’s managed forests to the carbon density of its remaining primary forests would, on these numbers, require keeping roughly 8 billion tonnes of carbon dioxide out of the atmosphere. That is equivalent to around two centuries of the country’s current fossil fuel emissions. Protecting what little old-growth remains is likely the easier part. The harder question is what happens to all the managed forest that has already been cut.

DOI / Source: https://doi.org/10.1126/science.adz8554