Civil War Union general Edward Powers noticed that whenever there was battle, rain often followed. In his 1871 book War and the Weather, Powers proposed that the chaos of battle could disrupt the atmosphere enough to trigger rainfall and urged Congress to take the idea seriously. “If lightning and thunder and rain have been brought on by the agency of man, when bloodshed and slaughter were only intended, this surely can be done without these latter concomitants,” he wrote.

By 1891, Powers’ vision took shape in a rainmaking expedition to West Texas—the first U.S. government-funded attempt to influence the weather—led by Robert G. Dyrenforth, a patent lawyer. Armed with $7,000 in funding, their team arrived with dynamite, kites to carry the charges skyward and 10-foot balloons. They conducted three rounds of atmospheric explosions. The rainfall that followed was minimal, but Dyrenforth promptly proclaimed the experiments a success.

Others were less certain. Smithsonian meteorologist George E. Curtis, a member of the team, remained unconvinced. Writing later in the journal Nature, he dismissed the effort, stating plainly that “these experiments have not afforded any scientific standing to the theory that rainstorms can be produced by concussions.”

In the 1890s, M. Albert Stiger—mayor of Windisch-Feistritz, Austria—set out to stop the hailstorms ravaging his region’s vineyards. After years of experimentation, Stiger devised a vertical cannon fashioned from thick sheet metal, mounted on timber and flared like a megaphone. When fired, it discharged a thunderous smoke ring aimed high into the clouds—intended to deter hail from forming. 

In 1896, Stiger’s six-cannon trial coincided with a year without hail. By the following year, 30 cannons dotted the landscape and Windisch-Feistritz was spared again—while neighboring regions were pummeled with hail.

Soon, the hail cannon craze swept across Europe, with thousands of cannons deployed across Austria, France, Spain, Hungary and Italy. But skeptics remained. In 1903, the Italian government launched a formal test: 222 cannons set over two test zones with controlled observation lasting two years. The results? Both areas were hit by hail. The cannons were declared a failure.

Despite the doubts, hail cannons are still in use today, guarding vineyards in Bordeaux, France and the Prosecco-producing hills of Valdobbiadene, Italy.

“By far the most significant effort to influence the weather have been the attempts at making it rain, usually by ‘seeding’ clouds with salt (usually silver iodide),” says Chunglin Kwa, a professor in social and behavioral sciences at the University of Amsterdam who has studied weather modification.

Those efforts started in General Electric’s Research Laboratory in Schenectady, New York, where American chemist Irving Langmuir switched from working on applications like gas-filled incandescent light bulbs to investigating rain clouds. 

Working off the idea that clouds are filled with supercooled water droplets (colder than 0°C) and that most rain begins as ice crystals, Langmuir and GE machinist Vincent Schaefer began running experiments. Using an open-top home freezer, they breathed into the chamber to introduce moisture and initiate precipitation. But the water droplets remained suspended, never forming ice heavy enough to fall; their buoyancy in the air stronger than gravity.

Then, on a hot day in July 1946, as the freezer struggled to stay cold, Langmuir dropped in a chunk of dry ice. Almost instantly, ice crystals began to form, growing into snowflakes that settled at the bottom. One of Langmuir’s assistants, Bernard Vonnegut (brother of author Kurt Vonnegut), proposed silver iodide as an even more effective agent for initiating precipitation. Even trace amounts triggered snowfall in cold-chamber tests.

In 1947, Project Cirrus—a collaboration between the Army Signal Corps, the Office of Naval Research, the Air Force and General Electric, with Langmuir and Schaefer as advisors—was launched and the silver iodide theory was applied to the skies.

On a dry July morning in 1949, Langmuir launched a silver iodide generator near Albuquerque, New Mexico. At 8:30 a.m., a cloud began forming 25 miles away, producing heavy rain. Soon, another cloud followed suit, also releasing rain. Langmuir claimed both storms were the result of the seeding—but many didn’t agree.

Of Langmuir’s cloud seeding efforts, Francis W. Reichelderfer, chief of the U.S. Weather Bureau told Time magazine, “I feel quite sure that in many cases the rain was due to natural causes.”

Project Cirrus ambiguously ended in 1952. However, it paved the way for Project Stormfury a decade later, expanding the concept of seeding clouds with silver iodide to a much larger scale.

A multimillion-dollar initiative by the Navy and the Department of Commerce, it sought to defuse hurricanes by bombarding the eyewall—the compact ring surrounding a hurricane’s center that harbors the most violent winds and intense rainfall. Pilots flew DC-6 and Lockheed P-3 Orion military aircraft, sometimes in formations of up to 10, firing canisters of silver iodide out of Gatling guns and dropping 130-pound silver iodide bombs deep into the storms.

Although “the goal of human control of hurricanes was captivating and seemed to be physically attainable in the beginning,” wrote Stormfury scientist Hugh Willoughby in 1985, “observational evidence indicates that seeding in hurricanes would be ineffective.”

Project Stormfury was terminated in 1983, and as Kwa notes, ultimately demonstrated that hurricanes couldn’t be tamed.

As the Olympic Games Beijing 2008 approached, China’s meteorological engineers, equipped with aircraft, artillery and over a thousand rockets, set out to do what no Olympic host had done before: control the weather. With the world watching and the open-air opening ceremony scheduled for August—one of Beijing’s wettest months—the threat of rain led the government to invest $100 million in weather modification.

At 9:35 p.m., as the Beijing Municipal Meteorological Bureau (BMB) issued a yellow alert for thunder showers, its cloud-seeding operations were already underway. Rockets packed with silver iodide and dry ice were launched to trigger early precipitation before reaching the capital.

Guo Hu, who leads the BMB, declared in China Daily: “We fired a total of 1,104 rain dispersal rockets from 21 sites in the city between 4:00 p.m. and 11:39 p.m. on Friday, which successfully intercepted a stretch of rain belt from moving towards the stadium.”

As the four-hour ceremony dazzled viewers across the globe, the skies above remained dry. Nearby Baoding, however, received 100 millimeters of rain that same evening. Coincidence or redirection? “Chances are it was coincidence,” says Kwa. “We may never be absolutely certain, but it just is not likely it worked. And if there was an effect, the rain would have been negligible—less than 1 mm.”