Is there a vent in the global greenhouse?

Satellite observations suggest that when surface temperatures of the western tropical Pacific warm up, the area of heat-trapping cirrus clouds surrounding low-altitude rainstorms decreases. It's an atmospheric phenomenon that some researchers think could ease the greenhouse effect.

The researchers who discovered this effect liken it to a thermostatically controlled vent that releases energy into space when Earth's temperatures build up. If confirmed by additional research, this heat-venting process could help reduce estimates of future global warming, they say. It hadn't shown up in global-climate models tested by the scientists.

Led by Richard S. Lindzen, a meteorologist at the Massachusetts Institute of Technology, the researchers analyzed 20 months of high-resolution satellite data obtained for tropical latitudes from Australia and southern Japan eastward to near the Hawaiian Islands.

For every 1[degrees]C rise in ocean temperature beneath a cloudy region, the ratio of high-flying cirrus clouds to lower, rain-generating cumulus clouds dropped by about 22 percent, report Lindzen and his colleagues in the March BULLETIN OF THE AMERICAN METEOROLOGICAL SOCIETY.

Cumulus clouds often beget cirrus clouds, Lindzen explains. As warm, moist air rises inside cumulus clouds, some of that moisture condenses and falls as rain. The droplets that don't become rain are lofted even higher, where they freeze to form cirrus clouds. Lindzen and his colleagues propose that higher ocean temperatures make the formation of raindrops more efficient, which leaves less moisture to form the heat-trapping cirrus clouds.

Cirrus clouds don't block much of the sunlight failing on Earth, Lindzen explains. However, they're especially good at trapping the longer-wavelength infrared radiation emitted by Earth's surface, which contributes to the greenhouse effect. When there are fewer cirrus clouds, more heat energy can radiate back into space, cooling the planet.

Computerized climate models indicate that global average temperatures would rise between 1.5[degrees] and 4[degrees]C if the concentration of carbon dioxide in the atmosphere were to double. Lindzen says that climate modelers' preliminary calculations now show that the cooling effect his team has observed might trim those increases to range between 0.64[degrees] and 1.6[degrees]C.

This difference between current climate simulations and Lindzen's results is "disturbing," says Byron A. Boville, an atmospheric scientist at the National Center for Atmospheric Research in Boulder, Colo. Boville notes, however, that estimating the ratio of cloud types in weather systems is one of the weakest areas of climate models. "It's really tough to model clouds," he says, because researchers don't fully understand the physical processes behind precipitation and cloud formation.

"Climate models are really deficient in estimating how much moisture gets rained out versus how much goes into clouds," concurs Roy W. Spencer, an atmospheric scientist at NASA's Marshall Space Flight Center in Huntsville, Ala. Therefore, scientists can't always make good estimates of clouds' long-term climate effects, including global warming, Spencer adds.

Researchers need detailed knowledge of what's going on inside individual clouds to improve climate models, Spencer says. Even simple factors in simulations, such as the size of raindrops, can significantly affect a climate model's calculations.

"None of these studies ever seems to settle the issue," laments Spencer. "They're just pieces of the puzzle." Now, the new cloud-driven cooling effect observed by Lindzen's team may make the climate puzzle a little more complicated.