A study led by the University of Florida's Michelle Mack contradicts the widely held assumption that Arctic tundra experiencing the effects of global warming might lock up carbon through increased plant production. The findings, published this week in Nature, provide the first experimental evidence that even though plant growth increases, the tundra habitat as a whole suffers a net carbon loss.

That should hasten the effects of global warming, the authors note, as fragile high-latitude ecosystems are particularly vulnerable to climate change, but hold the potential to contribute to its effects.

Since 1980, climate biologists such as study coauthor Terry Chapin have been enriching plots of moist Arctic tundra (MAT) with a combination of nitrogen and phosphorus. One widely held prediction is that warmer MAT soils will experience an increase in nutrient availability as a result of the faster decomposition of organic material, and it is this process they aim to simulate.

"We all had a lot of unproven assumptions about how nutrient release might affect ecosystems," Chapin, who is based at the Institute of Arctic Biology, University of Alaska, Fairbanks, told The Scientist, "but we believed it to be the single most crucial factor in determining the carbon balance in response to global warming."

Sure enough, adding fertilizer made plants grow better, even in the Arctic. Plant production doubled over the duration of the 20-year experiment, just as predicted. However, Mack's study went further by measuring the amount of carbon locked away not only in plant growth, but also in the soil itself. And it was in this moist, cold soil that the startling discovery lay buried: over time, increased nutrient flow resulted in a net loss of carbon. "This finding was a total surprise," Chapin said. "It completely changes our expectations of how global warming will affect carbon sequestration."

As Wendy Loya and Paul Grogan explain in their accompanying News and Views article, the conventional wisdom held that global warming would lead to increased nutrient availability. That, in turn, should assist plant growth and so store away carbon, which scientists had hoped would put the brakes on further warming. But by breaking with convention and studying what happened in the soil as well, Mack and her colleagues have ignited a new debate among polar biologists and ecologists on how high-latitude ecosystems will contribute to climate change.

"The one factor that we thought would counteract carbon loss actually stimulated it," said Mack, who currently works at the University of Florida's Department of Botany. "These results should affect climate change forecasts—they challenge our conceptual model of how tundra ecosystems will respond to climate warming."

Sarah Hobbie, a climate ecologist at the Department of Ecology, Evolution, and Behavior, University of Minnesota, said that while there were some caveats to the study, the findings have value. "The levels of nitrogen and phosphorus used exceeded those that we'd expect to occur with global warming," she told The Scientist. "However, the experiment's real value lies in what it tells us about the direction that MAT ecosystems might be heading, rather than the magnitude of the change."

But where did all the carbon go? "The only plausible explanation is more rapid plant decomposition," said Hobbie, who was not involved in the study.

Chapin agreed. "What our results show is that the release of nitrogen enhances decomposition and carbon release more strongly than it stimulates plant production," he said. The key, he said, may lie in increased microbial activity, which leads to more rapid breakdown of organic matter in the soil.

However, Mack's findings do not demonstrate a causal link between nutrient addition and increased microbial activity, which Hobbie said must be the focus of future experiments. "If we're to understand how tundra ecosystems will respond to climate change, we really must get to the bottom of why decomposition rates increased in this study."