Can hurricanes be created and controlled with technology?

Benchmarks: October 13, 1947: A disaster with Project Cirrus
by Meg Marquardt
Monday, June 20, 2016


Bernard Vonnegut and Vincent Schaefer examine preserved snowflakes on a sheet of glass in 1951. Credit: Schenectady Museum & Suits-Bueche Planetarium.

Two days after clipping Cuba and Florida, a hurricane was drifting out into the Atlantic. All predictions had it remaining at sea without further landfall, making it the perfect test subject for the newly minted Project Cirrus, a U.S. government- backed project bent on discovering a way to disable deadly hurricanes. The researchers planned to seed the hurricane’s clouds with dry ice, hoping that the ice would interact with the clouds and disrupt the cyclone’s internal structure, thus weakening it. So on Oct. 13, 1947, a plane flew over the storm and dumped 80 kilograms of dry ice into the storm’s swirling clouds.

What happened next was a worstcase scenario: Instead of dissipating, the storm furiously swung nearly 130 degrees to the west and smashed into Georgia, where it caused $2 million worth of damage. Threats of lawsuits soon followed, with Georgia residents blaming the government for the devastation. Project Cirrus was all-but shut down before it truly began, and any research into weather manipulation was repudiated for decades.

The road to Project Cirrus began with Vincent Schaefer, a meteorologist for General Electric who was trying to control an entirely different weatherrelated phenomenon: airplane icing. Supercooled droplets of water posed an enormous risk for planes. These droplets don’t naturally freeze in clouds unless temperatures reach below minus 40 degrees Celsius. Before that point, they need something around which to freeze — like the metal of an aircraft. When supercooled droplets hit an airplane’s wing, they immediately solidify, covering the wings in a dangerous layer of ice. In 1946, Schaefer discovered that adding an already-frozen element to clouds, such as dry ice or silver iodide, could induce supercooled water to solidify into ice crystals that would ping off metal, thus removing the threat of wing icing. In November 1946, Schaefer dumped dry ice from an airplane into cloud cover and proved his concept by starting a snowfall in Pittsfield, Mass.

It wasn’t long until the experiments at General Electric garnered attention from the government. Together with GE, the various branches of the government and military formed a coalition — headed by GE scientist and Nobel laureate Irving Langmuir — called Project Cirrus, which was to investigate the possibilities of cloud seeding, an early form of geoengineering, for a host of potential uses, like producing rain in arid regions.

But the plans didn’t stop with simply making rain. Project Cirrus quickly set its sights on a bigger obstacle: weakening the destructive force of a hurricane. If an injection of dry ice could disrupt the structure of the storm, the project’s scientists reasoned, then perhaps the hurricane would dissipate. “Right after World War II, science could do anything,” says Hugh Willoughby, an atmospheric scientist at Florida International University in Miami. “No one had thought through what would happen if you seeded a hurricane. They didn’t know about the cloud structure in a hurricane. It was just kind of, ‘Let’s do it!'”

And so they did it. Project Cirrus scientists dropped 80 kilograms of dry ice from a B-17 bomber into the clouds at the western edge of the Oct. 13 hurricane, which was supposed to be heading harmlessly out to sea. Scientists aboard another B-17 took photographs to measure any change the dry ice might cause in the clouds. Though the images did show some distortion in the cloud structure, the results of the experiment were less than conclusive about the potency of cloud seeding in this case.

What was conclusive, however, was that the storm had taken a sudden westward turn and made landfall just outside of Savannah, Ga. To the average person, it looked like Project Cirrus was to blame. Time magazine reported, “In Savannah last week, Southern blood bubbled toward the boiling point. A Miami weatherman had hinted that last month’s disastrous hurricane might have been not an act of God, but just a low Yankee trick.” Fearing lawsuits, GE pulled its support of the project. But before any legal actions could be carried out, other meteorologists pointed out that previous hurricanes had taken nearly identical paths. Later investigations that showed the storm had begun its westward turn before the cloud seeding took place fully cleared Project Cirrus of any liability.

In fact, recent studies on the amount of power unleashed by a hurricane make the attempt to modify one with only 80 kilograms of dry ice appear laughable. Bob Sheets, former director of the National Hurricane Center in Miami, wrote in his book “Hurricane Watch” that the heat energy a hurricane releases each day can be compared to the energy released by 400 20-megaton hydrogen bombs. However, such research is only a few years old, so it could not be used in defense of the failed attempt by Project Cirrus.

“[The hurricane] made landfall in the U.S. almost certainly not because of seeding,” Willoughby says. “But Project Cirrus poisoned the well of cloud seeding forever for hurricanes.”

But Project Cirrus did not entirely stop the government’s interest in weather modification. In 1962, various branches of the government, including the Navy and National Weather Bureau, commissioned Project Stormfury, Cirrus' descendant, and began a new wave of hurricane cloud seeding projects. All seeding with Stormfury was carried out only if a storm met a long list of criteria, including a less than 10 percent chance of hitting land within a day of seeding even if the storm made a sudden turn like the Oct.13 hurricane. The project lasted more than 20 years, but Stormfury never managed to dismantle a hurricane. In 1983, Stormfury was canceled, and the government has not attempted to disrupt a hurricane with cloud seeding since then.