Sagebrush, by comparison, is less active at photosynthesis and it produces far less plant material each year. As the sagebrush spread, Harte began to observe a disturbing trend: The carbon that was bound up in the soil continued to be released through the work of microbes, but now it was no longer being replaced. The fine balance of the meadow's carbon cycle had been upended, and in the first five years of heating, fully two-thirds of a pound of carbon per square yard was lost-about as much as in a pint of gasoline. The meadow, once at carbon equilibrium, had become, in a few short years, a carbon emitter. In the parlance of climate science, the phenomena at work in Harte's meadow are known as "positive feedbacks." A small increase in temperature leads to an increase of CO2, which in turn leads to even higher temperatures and yet more CO2. Warming begets warming. Harte and others now believe that such feedbacks are the likely engine of the carbon dioxide shifts recorded in the Vostok ice core. A subtle warming event-a slight increase in the sun's energy due to a change in the earth's orbit-served as the spark for climate change, but feedbacks did the rest.

It is hard to fathom that ecological changes could drive such massive climate shifts, but in fact the earth's soils harbor four times as much easily releasable carbon as resides in the atmosphere, and the oceans contain nearly 50 times as much. "Release even a fraction of that," Harte says, "and the effect would be enormous." Warm one meadow and nothing much happens to the global climate, but warm ten thousand meadows-and forests and savannahs and swamps-and, well, look out.

As we talk in his office, Harte's voice becomes increasingly animated and he interrupts himself several times and leaps from his chair to look for articles that will offer more detail on one point or another. He is plainly excited by questions of science, and even on a topic as grave as climate change a kind of breathless enthusiasm can overtake his words. At moments it seems possible to imagine that all of this is just an intellectual exercise. But then Harte brings me to his point.

The frightening thing, he says, is that, for lack of understanding, biological feedbacks like those in the meadow are not factored into our current global warming models. While the models anticipate increases in atmospheric carbon dioxide from future fossil fuel emissions and from some feedbacks like increased water vapor and lower ice reflectivity, they don't account for the carbon that will be unleashed from the ground and the ocean as the world warms. If they did, Harte says, the upper-limit increase in global temperature by 2050 would not be 8 degrees Fahrenheit, as is currently projected, but closer to 12 degrees. "In other words," Harte says, "we're underestimating the magnitude of future warming."

It is a startling assertion, and some climate scientists will undoubtedly take issue with Harte's projections; despite a broad consensus in the scientific community that global warming is happening, temperature forecasting remains an area of considerable disagreement. But Harte has not pulled these numbers out of a hat; he's based his projections on the data uncovered at Vostok, our best record of the climate's past changes. Nor can he be dismissed as some wild-eyed radical. A highly regarded scientist, he's served on six National Academy of Sciences committees and has published dozens of articles in top peer-reviewed journals.

If Harte is right, and our climate models are off by even a few degrees, the implications are serious. An extra warming of four degrees sounds insignificant; we would barely notice such a change in our day-to-day lives. But as a global average, even such a seemingly small increase would have far-reaching consequences. At the planetary level, a little warming goes a long way. At the depths of the last ice age, 20,000 years ago, when nearly a third of the earth was covered by ice, the average global temperature was just 12 degrees Fahrenheit cooler than it is today.

	page 3	|	5