From space, the boreal forest appears as a near-continuous pine-green band stretching across the Northern Hemisphere, just beneath the Arctic — from Europe through Russia and Asia, and again across Alaska and Canada.

Up close, the forest resolves into a patchwork of species. Conifers like spruce, pine, and fir dominate, while deciduous trees such as birch, aspen, and poplar appear in warmer regions. It is easy to imagine the boreal as distant, austere, and resilient: rows of looming trees growing slowly over long winters and short summers.

The numbers reinforce that sense of scale and endurance. The boreal covers roughly 17% of Earth’s land surface and stores about one-third of the carbon held in forests worldwide. While some of this carbon is stored in the boreal’s large trees, much of it lies below ground in soils, where cold, waterlogged conditions slow how quickly fungi and bacteria decompose organic matter, allowing it to accumulate over centuries.

Today, industry and forest managers are converting large areas of this forest into managed stands, harvesting trees for lumber and wood-based bioenergy. Many climate models aiming for a carbon-neutral future not only predict but also rely on increased use of boreal forests as a substitute for fossil fuels. As a result, even in low-emissions scenarios, we will continue to transform intact forests into managed plantations.

New research, however, shows that preservation, rather than management, may be a more effective strategy: Drawing on new field measurements across primary forests, combined with Swedish national data spanning thousands of plots, scientists from Lund University and Stanford found that intact boreal forests in Sweden store 72% more carbon than managed ones.

Driven largely by differences in soil carbon, the gap far exceeds earlier estimates and suggests that these forests play a much larger role in sequestering carbon than scientists have realized.

Two types of forests

The researchers compared two major types of forest in Sweden — primary forests and secondary managed forests.

Primary forests consist of relatively intact areas that haven’t experienced direct human impacts, such as logging. These include both old-growth stands and forests with younger trees that have experienced natural disturbances, like wildfires. In contrast, foresters actively manage secondary forests for timber production. In Sweden, they typically clear-cut these forests every 60 to 120 years, then replant them or allow them to regenerate naturally. These operations often involve intensive practices such as soil scarification, drainage, thinning, and prescribed burning.

The researchers set out to quantify the “land carbon storage” of each type of forest, which includes carbon stored in live trees, deadwood, and soils. For secondary forests, they also estimated the carbon stored in the products we create from harvested wood, such as paper and lumber, as well as additional deadwood left behind after harvest, like stumps and roots. Including these data in the secondary forest carbon calculation allowed them to make a fair comparison between managed and primary forests.

They relied on existing data from national Swedish databases, which include extensive measurements of vegetation, deadwood, and soils across thousands of boreal plots. Because these datasets contain relatively few primary forests, the researchers also conducted targeted field measurements across more than 200 plots over three years. Working across Sweden, they mapped primary forests and collected new field data over three years, compiling a first-of-its-kind, comprehensive dataset from intact boreal forests. 

To accurately compare the carbon storage between the forests, they conducted a paired analysis, matching and comparing each primary forest with its nearest managed counterpart within 50 km. This approach helped control for environmental differences such as climate, soil type, and elevation that influence carbon storage. They also used models to estimate how much carbon secondary forests would store if they had remained unmanaged, allowing them to extend their analysis across the broader landscape rather than only paired sites.

Intact boreal forests hold far more carbon

Across Sweden, primary forests store about 72% more carbon than managed forests, a gap that is 2.2 to 8 times greater than previous estimates. The carbon lost from converting these ecosystems is equivalent to more than Sweden’s cumulative emissions over the past 200 years. The difference jumps to 83% when the researchers exclude the carbon stored in wood products like paper and lumber. 

Most of the carbon lies below ground: About 64% is stored in soils, compared to roughly 30% in live trees and the remainder in deadwood. Soil carbon also accounts for most of the gap between forest types: Primary forests contain about 68% more soil carbon to a depth of 60 cm than managed forests, and more than double the carbon in the top organic layer.

The lead author, Didac Pascual, a postdoctoral scholar at Lund University, told Stanford that “we didn’t know what to expect from the soils. We learned that primary forests stored more carbon in their soil alone than managed forests do in trees, dead wood, and soils combined.”

While harvesting trees clearly reduces carbon stored in live biomass, the mechanisms driving soil carbon loss are less intuitive. Pascual and his co-authors point to three main pathways: changes in how much carbon enters the soil, how it is processed by microbial communities, and how much carbon is lost through physical processes.

Logging can affect all three pathways. First, it can reduce carbon inputs by removing biomass and reducing deadwood and plant litter, or the raw material that feeds soil carbon. Second, it can increase decomposition rates. Practices such as drainage and mechanical soil preparation disturb the soil, alter microbial activity, and expose previously protected carbon to oxygen, allowing it to break down more quickly. Third, management can increase carbon losses through transport, such as erosion or leaching, especially when soils are disturbed or drained.

Recovery also takes time: Soils can require decades or longer to rebuild the carbon stocks and microbial communities found in undisturbed forests. Research is underway to understand what drives carbon storage in the soils of these forests, including whether the old-growth forests have a unique assemblage of microbes and fungi in their soil. This knowledge may help researchers stimulate similar mechanisms in other forests, without having to wait decades for the forests and their microbial communities to develop. 

From Sweden to the broader boreal

The researchers caution that these findings come from a regional study and may not fully apply to other boreal forests, such as those in Canada and Russia, where management is often less intensive. Extending these results will require a better tracking of logging practices and an understanding of how specific practices, including drainage and mechanical soil preparation, affect carbon storage.

Even so, the findings offer insight into what may happen if other boreal forests undergo similar management and raise a broader question: If managed forests store substantially less carbon than old-growth systems, do the benefits of bioenergy outweigh the costs, or would preserving these ecosystems provide greater climate benefits?

A central tension lies at the core of this question: forests play a critical role in storing carbon, yet societies need their materials and energy. Resolving this tension will require a clearer understanding of how management practices shape carbon storage, particularly belowground. and, as the researchers note, on whether societies adopt lower-emission alternatives and reduce material and energy use.

Nature-based solutions

The implications of these findings also highlight a well-known blind spot in climate policy: So-called “nature-based solutions” to climate change — whether restoring wetlands, protecting forests, or allowing ecosystems to recover — often receive far less funding and attention than technological approaches, despite evidence that they can be both effective and cost-efficient. For example, one estimate finds that just 4% of tracked climate finance goes to nature-based solutions, compared with more than 89% directed toward energy transition. At the same time, assessments from the United Nations Environment Programme indicate that investment in nature-based solutions would need to more than double by 2030 to meet climate and biodiversity targets. Part of the challenge is visibility and timing. It is relatively straightforward to track energy production from solar infrastructure, but far more difficult to quantify changes in soil carbon or microbial processes.

Nature-based solutions also require, at least in part, a willingness to step back and allow natural processes to perform their magic, even if we don’t fully understand them. That, at least, seems to be true of boreal forests, where the best way to “manage” them in support of carbon neutrality may not be to harvest them more intensively, but to leave more of them alone.