Herding Hurricanes

It can't be done yet, but one scientist's computer simulations point the way to tempering fierce weather in the future

Everybody talks about the weather, but nobody does anything about it. Whether or not Samuel L. Clemens ever uttered those exact words, changing the weather is an age-old dream. Egyptian priests, medieval sorcerers, and Native American shamans all had incantations and dances devoted to it. And in Mark Twain's day, the Midwest was crawling with charlatans who claimed they could make it rain.

Today, a handful of scientists are taking the idea further. One of them is Ross N. Hoffman, vice-president for research at Atmospheric & Environmental Research Inc. (AER), a weather consulting firm in Lexington, Mass. He predicts that one day it will be possible to tame hurricanes, or at least to steer them away from cities, so they don't wreak havoc on the scale of Hurricane Katrina.

In fact, Hoffman has already achieved that feat -- in computer simulations. Using data from Andrew and Iniki, two major storms in 1992, he was able to show that relatively minor atmospheric adjustments could make a huge difference. With Iniki, which slammed into the Hawaiian island of Kauai that September, Hoffman diverted the simulated storm 60 miles to the west by injecting a rise in air temperature of less than two degrees Celsius six hours before landfall. With Andrew, it took an increase of up to three degrees to keep dangerous winds away from Florida's coast. There is currently no feasible way to make such adjustments, but that hasn't stopped Hoffman from exploring the idea in virtual reality.

Hoffman's research dates from the late 1970s, when he was working on his 1980 PhD in meteorology at Massachusetts Institute of Technology. His mentor was Edward N. Lorenz, the meteorologist who popularized the so-called butterfly effect in chaos theory with his 1972 talk, "Does the flap of a butterfly's wings in Brazil set off a tornado in Texas?" (For a decade before that speech, Lorenz' analogy had been a "flap of a seagull's wings.")

"Once I learned about chaos theory," Hoffman recalls, "I realized it might be possible to make a tiny change at one particular time and it would grow and expand to yield substantial effects." He wanted to base his dissertation on the potential for using chaos theory in weather control but was advised against it. Weather modification was -- and still is -- a touchy subject. So it was unlikely he could raise enough funds to run all the simulations needed to begin uncovering the rules that govern the seeming chaos of weather systems.

After MIT, Hoffman did weather research at NASA, then joined AER in 1983. But his dream didn't fade. In 2001 he heard about the NASA Institute for Advanced Concepts, which is funded by NASA but serves as an independent hatchery for way-out ideas. The institute coughed up $550,000, and Hoffman began searching for those flapping-butterfly changes. "With Iniki, we tried to find the smallest possible perturbations so that six hours later, the hurricane would end up 100 kilometers off to the west," he says. "You don't want to make huge changes. That would take too much energy."

With today's science, Hoffman's scheme for injecting enough heat to raise a hurricane's temperature by even the smidgen in his simulations is pure science fiction. He envisions orbiting satellites with arrays of giant solar heat collectors that would beam energy down to earth. Hoffman admits that such devices are unlikely to be launched for decades, although solar-power satellites are perennial favorites at NASA and the European Space Agency.

In an October, 2004, cover story in Scientific American, Hoffman laid out some intriguing alternatives to orbiting satellites. One is sprinkling rainmaking chemicals on the clouds surrounding a hurricane's eye. By forcing the eyewall to drop some of the warm water that nurtures the storm, it might be possible to deflate menacing hurricanes.

A tactic like this has actually been tested. For 10 years, starting in 1961, the U.S. Navy and federal weather researchers seeded four hurricanes -- not the eyes, but the northeast quadrants, which pack the strongest winds. Each hurricane was seeded for two days. After four of the missions, wind speeds reportedly dropped by 10% to 30%. The other four missions were fruitless. Years later, in 1983, advances in hurricane research indicated that seeding the outer regions was doomed from the get-go, and the project, dubbed Stormfury, was terminated.

Undaunted, scientists have now cooked up a new strategy: Prevent a hurricane from soaking up water and energy by coating the ocean's surface. Creating an oil slick by dumping thousands of gallons of biodegradable vegetable oil on the ocean has been proposed by Damian R. Wilson, a researcher at the Met Office, Britain's weather service. When MIT researchers ran some experiments to test the idea, they found that waves break up the slick into small pools, so it probably wouldn't be effective in the heavy seas of an oncoming hurricane.

But Hoffman asserts that "things like spraying the surface are worth pursuing." His MIT friends may have missed a key aspect, he argues: "A hurricane sucks in air, so it may be possible to use a surface coating to dry out the air at some distance from a hurricane."

Among weather professionals and amateurs, there's no shortage of suggestions for disrupting or "curtaining" a hurricane. They range from dropping a nuclear bomb to cooling the ocean by towing icebergs into the tropics. The nuke-it solution is sheer folly because it would produce radioactive rain. And cooling the ocean probably isn't possible, either. "It would require a great deal of energy, because water has a high capacity to absorb heat before it begins to warm up," says Hoffman. Besides, a hurricane packs 10 times the energy of the atom bombs that ended World War II, so all the icebergs in the North Atlantic might not do the trick.

While Hoffman believes that engineers will eventually learn how to manage hurricanes, he and many others argue that it should not be attempted until the physics are better understood. Without that, trying to steer a storm like Katrina away from New Orleans might cause it to slam into Florida instead. "If you mitigate a storm in one place, how do you know you're not going to make it worse for somebody else?" asks Bernard J. "Ben" Eastlund, CEO of Eastlund Scientific Enterprises Corp. in San Diego, who has been doing research on modifying severe weather for 20 years. At the same time, Eastlund notes, it's likely to cost $300 billion to clean up after Katrina -- just one storm. "Doesn't that tell you it's time to fund a serious effort to understand the weather and find some handles to deal with this?"

By Otis Port in New York

    Before it's here, it's on the Bloomberg Terminal.