Summer in south Florida seems to bring out as many towering thunderheads as it does bottles of sunscreen. For tourists, the storms can quickly douse beach plans. But for scientists, they generate a type of cloud that lies at the heart of one of the biggest mysteries in climate forecasting. These tropical cirrus clouds do double duty: They can trap heat like a blanket or reflect sunlight back into space like a mirror. No one has yet figured out which effect dominates and whether cirrus clouds represent a natural brake on global warming.
Now, two years after they wrapped up the largest airborne assault ever on these clouds, scientists are closing in on an answer. Although more tests will be needed beyond Florida, the data already are yielding insights that promise a major overhaul in the way climate forecasters treat cirrus formations.
"We've got some amazing results that no one anticipated," says Anthony Del Genio, a climate modeler with NASA's Goddard Institute for Space Studies at Columbia University in New York. "It's humbling to find out how often you're wrong."
Among the surprises: the possible presence of natural "antifreeze" at 50,000 feet, the key role of distant pollution, and smaller than expected ice crystals, which dramatically bolster both the mirror and blanket effects.
At first glance, spending up to $20 million on flying six aircraft, enlisting three satellites, deploying three sensor-laden ground stations, and tapping 450 scientists might seem like overkill for a research target as tenuous as tropical cirrus clouds.
Researchers beg to differ.
"When you look down on the earth from space, especially over the tropics, you see these gigantic cirrus anvils" topping thunderheads there, notes Owen Toon, a climate scientist at the University of Colorado at Boulder. These cirrus clouds play a significant role in determining the amount of sunlight the planet reflects back into space. Without the clouds, the light would reach the surface to be absorbed and reradiated as heat. "So you want to know how bright the cirrus are and what controls that brightness," he says. The brightness tells scientists how much light is being reflected back into space. The size of the cloud's ice crystals plays a key role in setting its brightness.
More important is the story the cirrus clouds tell about water vapor - the atmosphere's most abundant greenhouse gas - high in the atmosphere.
When humans add carbon dioxide to the atmosphere, only about 25 percent of the atmosphere's warming comes from the CO2, Dr. Toon says. The rest comes from related effects. The most important of these effects is the addition of water vapor to the atmosphere as temperatures warm and seawater evaporates and builds convective thunderheads in response.
Water vapor is "a huge amplifier of the greenhouse effect," he adds. And the portions of the atmosphere the research team studied - the upper troposphere - is "where the lever is."
If warming placed increasing amounts of water vapor in the upper troposphere, warming would be expected to continue.
On the other hand, "if something up there worked [in] the other direction and stabilized things, that would keep the greenhouse effect from growing," he says. If thunderheads grow more vigorously as the climate warms, for example, this could accelerate the downdrafts surrounding the thunderheads. The souped-up downdrafts might pull increasing amounts of moisture out of the upper troposphere, offsetting warming.
The month-long field project, known as CRYSTAL-FACE, highlights the interplay of real-world measurements and computer simulations in climate research.
One of the most fundamental questions surrounds the size of the ice crystals that make up tropical cirrus clouds. A team led by Timothy Garrett at the University of Utah found that the ice crystals in anvil cirrus over south Florida are smaller and reflect light more effectively than most models assume. The results suggest that when the clouds are thick as they first form over the top of a thunderhead, they reflect substantially more light back into space than models currently show.
But the results represent a climatological two-edged sword: As the cirrus clouds disperse and thin, they turn from sun reflector to blanket - trapping as much as three times more heat than models show. Modeling studies, if not another field project, should help spot the points at which the switch from reflector to insulator takes place. The question now, Dr. Garrett says, is whether these results are representative of tropical cirrus in general.
Two other sets of results hint at the role pollution might play in helping or hindering cirrus formation.
A team led by Ann Fridlind, an atmospheric scientist at NASA's Ames Research Center in Mountain View, Calif., found evidence that pollutants from distant sources may have a profound effect on anvil-cirrus formation. One series of thunderstorms coincided with the arrival of a plume of dust that blew across the Atlantic. By analyzing ice samples, the team found that this dust dominated the nuclei of crystals, and not "seed" material from directly beneath the cloud. The results, she says, indicate that distant sources of pollution may have a greater effect on cirrus formation than local sources. The results "just blew us away," Del Genio recalls. It defied conventional wisdom.
Another set of results implied that nitric acid, which occurs naturally and as a byproduct of burning fossil fuels, may retard the formation of anvil-cirrus ice crystals.
Ordinarily, water at anvil altitudes is chilled far below the freezing point. But it's still too warm for the tiny droplets to freeze spontaneously. Supply a "seed" and the "supercooled" water flash-freezes to the particle. A team led by Ru-Shan Gao, with the National Oceanic and Atmospheric Administration's Aeronomy Laboratory in Boulder, Colo., found that in anvil cirrus and in cirrus clouds generated by aircraft contrails, the humidity of the surrounding air was unexpectedly high. This supersaturated air implied that ice crystals were not forming as readily as expected, given the abundance of seeds around for flash-freezing.
Drawing from modeling and lab studies, the team suggests that the air was supersaturated because the ice crystals carried nitric acid, which acted like a mild antifreeze. The result: more water present as vapor - an effective greenhouse gas - than otherwise might be the case.
Yet as these and other results work their way into the debate over the roles of tropical anvil cirrus clouds and water vapor in climate change, the Florida results may have limited value, some researchers acknowledge. Florida's storms form over land, where the rising air is warmer and convection is more vigorous than over the open ocean. Moreover, researchers say, aerosol particles might not be as prevalent over the tropical Pacific as they were over Florida.
Next year, scientists plan a similar set of experiments over the Pacific off Costa Rica. That location, says Dr. Fridlind, should resemble more closely the western tropical Pacific, the globe's most important heat sink, than does south Florida.
Still, the Florida experiment gave researchers a solid crack at these enigmatic clouds. And it helped test approaches for coordinating such a large field experiment, yielding lessons that researchers say they will apply next summer.
• In the time it takes to read this sentence, 2,000 thunderstorms will have flashed and boomed somewhere in the world.
• Countries at or near the equator experience the bulk of the world's thunderstorms year-round.
• Thunderstorms dominate the southeastern United States - especially Florida - up to 90 days per year, mostly in the summer.
• The average life of a thunderstorm is 20 to 30 minutes, and it can stretch 10 to 15 miles in diameter.
• Rain does not determine where lightning strikes; that hair-raising crackle can hit 10 miles away from precipitation.
• The phenomenal energy produced by lightning heats the surrounding air to 50,000 degrees F. - creating a high-pressure shock wave that results in the sonic wave known as thunder.
Sources: NOAA, University of Denver