Last month, a study funded by the National Science Foundation tracked for the first time how the technology works in nature. The evidence for cloud seeding has been scarce, but recent research has encouraged officials and companies desperate to increase the amount of water in rivers and reservoirs.
In Colorado alone, more than 100 cloud seeding machines are set up in mountainside back yards, fields and meadows. Some older versions of the contraptions look like large tin cans perched on top of a propane tanks. New ones are large metal boxes festooned with solar panels, weather sensors and a slim tower.
Their goal is the same: to “seed” clouds with particles of silver iodide, a compound that freezing water vapor easily attaches to. That combination makes ice crystals form, which eventually become snowflakes.
Colorado’s program, which costs $1 million a year, is paid for not just by the state, ski resorts and local water users but also by water districts as far away as Los Angeles that want to increase snowmelt into the Colorado River, which sustains more than 30 million people across the Southwest. Most of the river basin is experiencing a drought.
“Everyone starts to get nervous when there’s no snow in Colorado,” said Joe Busto, the scientist who oversees Colorado’s cloud seeding program.
Major urban water districts in Arizona, California and Nevada have funded cloud seeding in the Rocky Mountains for more than 10 years and are now close to signing an agreement with officials in Colorado, New Mexico, Utah and Wyoming to split the cost of nine more years of seeding.
Cloud seeding is a relatively cheap tool for bulking up the water supply in Lake Mead and other reservoirs, said Mohammed Mahmoud, a senior policy analyst for the Central Arizona Water Conservation District.
Yet it’s hard to tell how much additional precipitation cloud seeding creates. The processworks only when there are freezing, moist clouds in the air. And the technology can be controversial.
“The whole thing is propaganda,” said Jamie Kouba, 32, a farmer from Regent, N.D., who argues that cloud seeding is decreasing rainfall in his area, rather than increasing it. He’s organizing local farmers in a campaign against the practice.
A 20th-century technology
Cloud seeding machines generate smoke that floats into the air like incense. Some state programs rely on ground-based machines. Others use airplanes to drop flares that generate silver iodide smoke into clouds, or to fly into a storm with flares strapped to their wings.
The recent study, which was conducted in Idaho and published in the Proceedings of the National Academy of Sciences, was the first to show real-world observations of silver iodide forming ice crystals inside clouds and falling out as snow.
Such research has increased interest in cloud seeding, particularly among private companies and utilities, said Neil Brackin, president of Weather Modification, a North Dakota company that does cloud seeding.
Other recent studies have used computer modeling to estimate the increase in snowfall from cloud seeding. A 2014 study across two Wyoming mountain ranges found that cloud seeding could increase snowfall by 5 to 15 percent — but only when the right conditions for seeding were met, or during 30 percent of snow events.
Relatively small increases still matter. “People in the western United States — we’re always water-stressed out here,” said Frank McDonough, an atmospheric scientist in Nevada who oversees the cloud seeding program at the Desert Research Institute, part of the state university system. Along the Colorado River, more water is promised to people than is available.
Nevada’s cloud seeding program can increase the snowpack by up to 10 percent, McDonough said.
That translates into 80,000 more acre-feet a year of water, enough to sustain about 150,000 households.
Still, he said, cloud seeding programs are difficult to evaluate. “Ten percent of additional snowfall is within the natural variation of storms.”
Idaho Power, which serves customers in Idaho and Oregon, has been using cloud seeding to boost the volume of water moving through its hydroelectric dams since 2003. The company’s representatives say the $3 million seeding program they oversee — which is partly funded by the state and other water users — generates billions of gallons of additional water for much less than 1 percent of the company’s operational budget.
Still, Busto warns that cloud seeding isn’t a cure for drought. Take this winter, which has been too warm and too dry for seeding. “We have not been able to run our cloud seeding machines because there’s been no storms coming through,” he said; the machines have only run a handful of times so far.
Busto said he has been confronted by people who are worried that exposure to silver iodide will make them sick. But silver is a naturally occurring element that is not inherently harmful, he said. In the 2014 Wyoming study, scientists found that seeding added some silver iodide to the surrounding water and soil but far too little to pose a known threat to human health.
North Dakota’s Kouba is one of the skeptics. He has compiled state rainfall data going back to the 1970s and concluded that cloud seeding has decreased rainfall, particularly in counties that are downwind of seeding operations. “I’m in a downwind county, and we have lost considerably,” he said.
North Dakota officials say there’s no evidence that cloud seeding caused last year’s drought. If anything, cloud seeding in one place leads to more precipitation downwind, not less, said Darin Langerud, who oversees cloud seeding for the North Dakota Atmospheric Resource Board.
Busto said actual science fiction — such as “Geostorm,” a 2017 movie about malfunctioning climate control satellites — fuels misperceptions about what states are trying to do.
“Every time something like that comes out, it feeds more conspiracy theories,” he said. “I’m just trying to make more snowflakes at a ski resort.”