Rare is the forest untouched by man. Whether logging or clearing land for agriculture, the bulk of the world's forests have fallen to crops, cattle or younger trees. According to some estimates, less than 10 percent of forests worldwide can be considered old growth, or undisturbed for more than a century. And that is not just a tragedy for the plants and animals that require mature forests—it is also a tragedy for the world's climate, according to a study published today in Nature.

Laborious research in the 1960s by the late pioneering U.S. ecologist Eugene Odum seemed to indicate that forests achieve a balance between the amount of carbon dioxide (CO2) absorbed by growing trees and plants and the amount of CO2 released back into the atmosphere by the decomposition of dead plant matter.

But it seems that old forests may be more efficient than previously believed. Biologist Sebastiaan Luyssaert of the University of Antwerp in Belgium and his colleagues surveyed all the existing measurements of how much carbon is absorbed and released from old-growth forests (exclusively in temperate and boreal forests due to a lack of extensive data on tropical forests). Their findings, Luyssaert says: "old-growth forests continued to accumulate carbon."

In fact, not only do old trees continue to store carbon in their wood, forest soils also appear to be actively capturing carbon over time, although direct observations of this process are lacking. All told, by Luyssaert's calculations the relatively small remaining stands of old-growth forests in the U.S. Pacific Northwest as well as Canada and Russia consume "8 to 20 percent of the global terrestrial carbon sink," or roughly 440.9 million tons (0.4 gigatonnes) of carbon per year.

That is not even close to enough to balance the 1.8 billion tons (1.6 gigatonnes) released into the atmosphere by deforestation or crop-clearing. But it remains important—if unrecognized—in the present battle to combat climate change. Luyssaert suggests that credit—and money—should be given to protect such old-growth forests under carbon trading schemes and other economic mechanisms to combat climate change.

"Any kind of existing program that gives credit to reforestation could give credits to forest preservation," such as the carbon offsets based on tree planting, he says. "Instead of investing the money in a new forest, it could as well be used to protect an old forest."

But the case for old forests as carbon sinks is not airtight. The measurements used by Luyssaert rely on the flux of CO2 levels over the forest, but this kind of metric can be skewed by young stands of trees within an old-growth forest or an increase in growth as a result of higher atmospheric carbon dioxide levels, according to forest ecologist Mark Harmon of Oregon State University in Corvallis, who was not involved in the study.

"To really test this, one would need a far better data set that had different ages in the same system: that is very young, mature, old-growth and super old-growth in each system," he says. But "older forests should not be written off as places to store more carbon. Even if they aren't taking up more carbon, their harvest releases a great deal."

It remains unclear whether tropical forests, such as those of the Amazon or Congo, produce the same effect, due to much faster decomposition of dead plant matter in these climes. But preliminary results suggest they do. "The data that are available show that, like the boreal and temperate forests, tropical old-growth forests also continue to take up and sequester carbon," says forest scientist Eugenie Euskirchen of the University of Alaska Fairbanks, who was not involved in this research.

Protecting old-growth temperate and subpolar forests might prove a boon to the fight against global warming, also because of the soils they currently shade. "Many old boreal forests tend to be underlain by permafrost soils, which can contain many times more carbon than that stored in the vegetation," Euskirchen notes. Melting those soils is an ongoing climate calamity.