Plants and soils may worsen global warming.

Call it the revenge of nature. Two new ecological studies suggest that plants and soils could exacerbate global warming in the next century by releasing vast reserves of carbon dioxide ([CO.sub.2]) that they have kept locked away for millennia.

Previous studies of vegetation patterns have indicated the opposite: that plants should eventually ameliorate global warming by growing vigorously and sopping up some of the [CO.sub.2] pollution now accumulating in the atmosphere. But such analyses have focused on what happens once the world has warmed, not on the transition period. A simple modeling study now indicates that because plants and soils cannot keep pace with climatic change, they will substantially boost [CO.sub.2] concentrations in the atmosphere over the next 50 to 100 years, report Thomas M. Smith and Herman H. Shugart of the University of Virginia in Charlottesville. They detail their findings in 'the Feb. 11 NATURE.

The [CO.sub.2] release forecasted by the two researchers may already have started in the Arctic, according to a separate study reported in the same issue.

To estimate how vegetation and soils will respond to global warming, Smith and Shugart started with general circulation models that simulate how greenhouse gas emissions will alter the climate. By matching climatic patterns with known plant limitations, the researchers produced maps showing the locations of tundra, forests, savannas, and other types of "life zones," They compared a life-zone map for current conditions with a map representing a climate with double the amount of [CO.sub.2]. Using crude estimates for how long it takes life zones to replace each other, the two ecologists calculated how much [CO.sub.2] the land surface could store as vegetation patterns shift.

The study shows that transitions that release [CO.sub.2] take place much faster than those that store the gas. For instance, forests convert rapidly to grasslands through dieback or fire, which liberates [CO.sub.2]. But it takes centuries for [CO.sub.2]-storing tundra to replace polar deserts, because species must migrate long distances.

Land changes could boost [CO.sub.2] levels by up to a third of the present concentration, the study indicates. While they have little faith in the exact numbers in the study Smith and Shugart believe their qualitative results have significance because the same conclusions emerge when they use other numbers.

Evidence collected from the tundra of northern Alaska suggests that global warming may already have spurred the land there to start releasing [CO.sub.2], report Waiter C. Oechel of San Diego State University and his colleagues.

Oechel's group set up airtight chambers along a 200-mile stretch in northern Alaska to measure gases absorbed and released by growing vegetation and degrading organic matter.

Since the end of the last ice age, the tundra has stored [CO.sub.2] by building up thick layers of peat. However, Oechel's measurements in the last decade indicate that tundra along Alaska's North Slope has started to release [CO.sub.2] - because microbes are consuming peat faster than it can grow,

Oechel believes the shift happened quite recently. In the early 1970s, measurements made at Barrow showed the tundra absorbing [CO.sub.2]. When Oechel and his colleagues remeasured that site, they found the tundra releasing [CO.sub.2].

Temperatures in northern Alaska have risen in recent decades and may have precipitated the change measured by Oechel by drying the tundra and stimulating microbes in the peat. "I personally feel we're seeing the first effects of greenhouse warming," he says. "But even if that's not the case, it gives us indications of how ecosystems will perform when and if that warming occurs."

If tundra across the Arctic were releasing as much [CO.sub.2] as Oechel measured in northern Alaska, it would produce roughly 5 percent of the amount that humans emit through burning coal, gas, and oil. Oechel plans to make measurements this summer in Russia.

Jonathan T. Overpeck of the National Oceanic and Atmospheric Administration in Boulder, Colo., calls the pair of new studies a one-two punch: "Anyone who is going to say you can't believe the modeling stuff because it's so oversimplified better wake up when they see the numbers coming from the tundra."

Yet some tundra scientists remain unconvinced that the tundra has stopped storing [CO.sub.2]. "I wouldn't put any significant money of my own down to say that it has changed much," says Donald Schell of the University of Alaska in Fairbanks.

- R. Monastersky