Year after year, the tundra covering the landscape of the far north has pulled carbon dioxide out of the atmosphere. Tiny plants absorb it during the intense summers, when they bloom in round-the-clock sunlight. The rest of the year, frozen soils lock that carbon dioxide away.
But that may be about to change.
If current warming trends continue, even cold tundra soil will soon release more carbon in the winter than tundra plants remove in summer, changing the vast northern landscape from a carbon sink into a producer of atmospheric carbon dioxide, new field research from Interior Alaska shows.
Those findings come from a pair of new studies based on field work at a test site near Healy that looked at year-round processes in the active layer of soil that lies atop permafrost. The Healy site was established by Northern Arizona? University's Ted Schuur, a widely recognized tundra expert who co-authored both studies.
Past studies of tundra carbon releases have focused on summer permafrost thaw and the potential for ancient carbon, sequestered there for millennia, to be released.
But this time researchers looked at active soils instead and took field measurements of carbon emissions from the tundra during winter, something rarely done because of challenges of working in the harsh conditions.
One study, led by Elizabeth Webb of the University of Florida and published in the Journal of Geophysical Research: Biogeosciences, took measurements at the test site for field seasons from 2008 to 2013. The project used snow fences and enhanced piles of snow to better insulate some of the test plots during the winter, thus holding soil temperatures higher than those at untreated control plots. Temperatures of the soil at the plots with more snow insulation were, on average, 1.5 degrees Celsius higher than at the control sites.
The project monitored soil temperatures in the various plots and, using a variety of methods, measured the flow of carbon into and out of the soil.
The warmed-up plots saw more vegetation growth in summer and absorbed more atmospheric carbon than the control plots, the study found. "The warmer plots are definitely greener and there's more plant activity, definitely, in the summer," Webb said.
But during the winter, the artificially warmed plots released even more carbon dioxide -- more than offsetting the work of the summer plants.
"When we compared the summer with the winter and looked at it over then entire year, there was a net loss," Webb said.
It is normal for the soil's active layer to release carbon dioxide in winter and the periods of late fall and early spring when plant photosynthesis is minimal, she said. That's because even at below-freezing temperatures, microbes in the soil are active, consuming the carbon deposited there during the summer and releasing it back into the atmosphere. Those microbes become more active -- and release more carbon dioxide -- as temperatures rise, even by a small amount.
The study found that soil from warmed plots released 9 to 36 percent more carbon than that from the control plots.
For the second study, published in the journal Nature Climate Change, a team of Chinese and American researchers analyzed genomes to quantify the diversity of microbes within the active soil layer. They found much more diversity at the artificially warmed plots -- even in winter, when soil temperatures were below freezing -- suggesting that carbon-dioxide-producing microbes will proliferate as winter temperatures increase.
The increase in diversity appeared to happen very quickly, said Vladimir Romanovsky, head of the permafrost laboratory at the University of Alaska Fairbanks Geophysical Institute. He was not involved in either study but is familiar with the work by Schuur's group and is well known for his own permafrost and tundra research.
The bottom line of both studies, Romanovsky said, is that the shallow active soil layer will eventually change from its current role as an absorber of atmospheric carbon to a contributor instead. The shift will likely happen over decades, not quickly, he said.
"It's not a bomb. It's just a shift in this kind of equilibrium," he said. "No explosion, but slowly and steadily."
Romanovsky had one quibble with the Webb-led study. It was based on changes in only temperature, he said, and did not account for variations in water conditions in the soil. Depending on soil types, water can remain in liquid form at temperatures as low as minus-5 degrees Celsius. Water freezes more quickly in coarser soils, but more slowly if soils are very fine-grained. A more detailed analysis would include the role of soil type and water, Romanovsky said.
Still, it's clear that surface soils have been warming over the past decades.
Freeze-up in the active layer has progressively gotten later in the year, he said. On the North Slope, the active layer used to freeze reliably in fall -- as early as October, but at least by November, he said. Now, "The freeze-up happens two months later than it was happening 20 years ago," he said.
A similar trend is happening in the Fairbanks region, which is having a late soil freeze this year. The area's active soil layer typically freezes in January, but this year about half of his test sites there were still unfrozen in late February. "I guess I can call it -- it is a record," he said.