Artificial insemination of clouds

Why Los Angeles messes with

storms in the Rocky Mountains

by Allen Best

But does it work? That’s always been the question about cloud seeding, the technology pioneered in 1946—coincidentally with the brother of novelist Kurt Vonnegut as a principal figure – and now widely deployed across the West.

In Colorado alone, nearly $1 million was spent by cities, water districts and ski areas to seed clouds between Winter Park and Telluride. Most see the operations as a low-risk bet on augmenting snowfall in the so-so years. In drought years, typically there are few clouds to seed. In snowy years, it’s not needed.

If even modestly successful, the cost is low – just $1 per acre-foot, according to an operator who seeds clouds near Crested Butte, Colo.

A new study issued in December by the U.S. Bureau of Reclamation more cautiously estimates costs ranging from $30 to $60 per acre-feet. At that rate, says the study, cloud-seeding could be one of the least expensive ways to narrow the gap between supplies of water in the Colorado River Basin and growing demands. Dramatically more expensive would be desalting brackish groundwater in Arizona, which the study estimated at $650 per acre-foot, or ocean water near Los Angeles, at $2,100 per acre-foot.

A ground-based cloud-seeding generator near Eagle, Colo., seeks to disperse silver iodide into clouds blowing toward Beaver Creek and Vail.Photo/Allen Best
A ground-based cloud-seeding generator near Eagle, Colo., seeks to disperse silver iodide into clouds blowing toward Beaver Creek and Vail.
Photo/Allen Best

But even the federal government’s estimate for cloud seeding has almost no empirical basis, says a scientist involved with cloud-seeding research. “It’s pretty back of the envelope, more qualitative (than quantitative),” says Dan Breed, of the National Center for Atmospheric Research in Boulder, Colo.

Not enough pure research has been completed yet to assign a hard figure to the effects of cloud-seeding, says Breed.

Breed also remains skeptical that cloud-seeding could produce 300,000 acre-feet per year, on average, as the Bureau of Reclamation’s study estimates. That’s about as much water as is diverted from the Western Slope of Colorado to the Front Range in a poor year.

Eureka in Schenectady

The hallelujah moment for cloud-seeding occurred at the General Electric Research Laboratory in Schenectady, N.Y. There, Bernard Vonnegut, a graduate of the Massachusetts Institute of Technology, found that silver iodide could serve as the nuclei, or seeds, around which water droplets condense.

Silver iodide remains the most common agent used to seed clouds. The chemical is put in a solution of acetone and combusted. The heat lifts the aerosol into upslope winds, which carry it toward targeted areas. Airplanes allow more effective seeding, but at much greater expense.

The longest-running program in the West occurs in the upper San Joaquin River Basin of California. There, Southern California Edison seeds clouds in a program that began in 1950. Long-standing programs have also existed in Nevada and Utah.

Colorado’s longest steadiest program began during the drought winter of 1976-77, with Vail Mountain as the intended recipient. The program has continued with just temporary lapses ever since.

Other programs in Colorado began after the dire drought of 2002. The efforts coalesced in the winter of 2011-2012 in a partnership of cities, ski areas and water districts pooling money to seed clouds to augment watersheds from Aspen to Grand Lake at elevations above 8,500 feet. Ski areas pay for efforts only through January, although other efforts continue through February and sometimes into March. Total cost of the program is $293,000.

Sites of cloud-seeding  generators in Colorado during the 2012-2013 winter.
Sites of cloud-seeding generators in Colorado during the 2012-2013 winter.

A parallel program uses two remotely operated generators to cloud-seed the Winter Park ski area from November through March.

But does it work?

The National Academy of Sciences, in a 2003 report, threw water on weather modification efforts altogether, although conceding “strong suggestions of positive seeding effects” from winter snowstorms, unlike summer storms.

Cloud-seeding advocates were indignant. Atmospheric modeling is extremely difficult, they pointed out, and the scientific establishment had not demanded the same high bar or proof for global warming theory.

Arlen W. Huggins, of the Desert Research Institute in Reno, Nev., points to work in Utah’s Wasatch Plateau as providing “some of the best physical evidence of seeding-induced precipitation increases.” He also credits research conducted in recent decades on Colorado’s Grand Mesa and in California’s Sierra Nevada.

But for randomized experiments, some of the best work —and only work – was done in the 1960s, in the area between Vail and Breckenridge. That work was supervised by Lew Grant, an atmospheric scientist who grew up in Oklahoma during the Dust Bowl. He followed that project with additional work at the research lab atop the Steamboat ski area in the late 1970s and early 1980s.

After seeding clouds with silver iodide there, Grant quit his work in weather modification and applied his energy full time to growing vegetables on a farm north of Fort Collins, Colo., without aid of chemicals. Later, the term “organic” came into vogue to describe that type of farming.

As for weather modification experiments, the federal government by then had gotten out of the game.

Randomly in Wyoming

Enter Wyoming’s State Legislature, which in 2006 appropriated money to begin what Huggins calls the “first fully randomized wintertime cloud seeding experiment to be conducted in the U. S. in more than 30 years.”

The experiments in Wyoming, as those in Colorado in the 1960s, are independent of any practical seeding efforts. The goal, says Barry Lawrence, of the Wyoming Water Development Commission, is to determine what role cloud-seeding may have in the state’s long-term water strategy.

Huggins says the Wyoming study stands on the shoulders of previous experiments, which defined the periods of storms that are seedable. “This greatly reduces the variability in precipitation among the randomized cases,” he explains. “From a scientific perspective, the study directly addresses the National Academy’s recommendation that further research was necessary to prove the efficacy of cloud seeding.”

The National Center for Atmospheric Research was chosen to design and oversee the study, and Breed was given responsibility as project manager and scientist.

Central to the experiment are two parallel mountain ranges, the Sierra Madre and Medicine Bow, set 40 to 50 miles apart, located southwest of Laramie, near the Colorado border. When storms carrying the appropriate level of precipitation approach, cloud-seeding generators are set off randomly, spewing silver iodide as high as 3,000 feet for as long as four hours.wyoming-satellite-image-m_wx_mod_3_13

Only one range is seeded per storm. The other mountain range serves as a control, with snowfall measured in both ranges after the storm.

“The randomization is needed because you’re looking for a small signal in a fairly large natural variability,” says Breed.

To gauge the effectiveness of seeding, however, weather conditions in the two mountain ranges must be similar. Only then can the apples be compared fairly.

Each mountain range has eight ground-based generators. Another 10 generators are located in the Wind River Range, between Pinedale and Lander, although the study there lacks the same randomized design.

One problem in the Wyoming research has been lack of clouds to seed. “We’ve had several drought winters, and last winter was just terrible,” says Breed. The winter had 15 storms suitable for seeding, far short of his hope for 30 to 40 per winter.

Conversely, the snowpack grew so rapidly two years ago that seeding operations had to be suspended.

For effective seeding, clouds must be cold, at least 25 degrees Fahrenheit, and wet. Even by late February, temperatures often rise too high for effective seeding, says Breed.

Therein lies a fundamental problem with cloud-seeding as a grand strategy for filling Lake Powell and other reservoirs in the arid Southwest. In drought years, the technique has little effect, and in wet years it can’t be done. Between the two extremes during the narrow band of mid-winter lies potential, but is it enough to make that much difference?

Can cloud-seeding help fill reservoirs of the Southwest, including Dillon? Evidence suggests it's a tool, but with only limited value. Photo/Allen Best
Can cloud-seeding help fill reservoirs of the Southwest, including Dillon (shown here in 2002)? Evidence suggests it’s a tool, but with only limited value. Photo/Allen Best

That is what the research in Wyoming intends to find out, but the answer has remained elusive. To have enough storms to achieve statistical significance, the program had to be extended longer than was originally planned, with a corresponding ballooning of the cost, now at $13.5 million. That extension produced some editorial harrumphing by the state’s leading newspaper, the Casper Star-Tribune, but legislators agreed to work through the 2013-2014 winter. A final report is expected in 2015.

So far, says Breed, results give “strong indications of positive effects” of seeding clouds, “but it’s still too early to say that with statistical confidence.”

Much is at stake, says Huggins. “It has the potential of putting the whole cloud seeding package together; i.e. confirming cloud seeding can increase wintertime precipitation and that this increase can be translated as a benefit to water resources.”

In other words, cloud-seeding would become more bankable.

Where LA puts its money

Cloud-seeding already has been proven effective enough to the satisfaction of 66 programs already in place in the West.

A major funder for several of these programs in the Metropolitan Water District, which delivers water to 18 million people in the Los Angeles metropolitan area. The agency has spent $28,500 per year for the last eight winters for cloud-seeding efforts in the lower-basin states. But lower-basin water agencies spend $425,000 annually in the Colorado River headwaters.

“Traditional water projects cost $400 to $500 per acre-foot. We are looking at $10 to $20 per acre-foot (for water produced by seeding clouds),” explains Metropolitan Water’s Tom Ryan. “We wouldn’t be spending that sort of money if we didn’t think it was effective and valuable.”

Colorado’s Gunnison River Basin gets some of that money. Los Angeles and other water providers in lower-basins states foot roughly half the bill on a program that is capped at $95,000, less if there are insufficient clouds to seed.

The most substantial local commitment comes from the Upper Gunnison River Water Conservation District, this year at $26,500. “The cost is relatively modest for the return on investment,” says the district’s general manager, Frank Kugel. “Even if the program produces only 10 percent of the amount projected, it would result in very inexpensive water.”

Mt. Crested Butte, a municipality, also chips in $3,000, and Gunnison County $10,000.

Cloud-seeders emphasize that you can’t make snow from nothing. You need clouds.

“If you’re going to have 50 percent of snowfall naturally, and you get a 10 percent increase from cloud seeding, that would still result in a snowpack 55 percent of average,” explained Don Griffith, president of North American Weather Consultants, the firm that has seeded the clouds near Crested Butte for the last decade.

“There’s still a drought. It’s just going to be a little less dry than it would be naturally,” he told the Crested Butte News earlier this winter.

An elevated cloud-seeding generator in the Sierra Nevada of California.Photo courtesy of Desert Research Institute.
An elevated cloud-seeding generator in the Sierra Nevada of California.
Photo courtesy of Desert Research Institute.

Ski towns like more certainty of snow for obvious reasons. So would Las Vegas. With just four inches of natural precipitation per year, the gambling mecca is largely at the mercy of water flowing from upstream. The Southern Nevada Water Authority has already been actively engaged in discussions about funding desalination and many other techniques, mostly very expensive, for augmenting existing water supplies in the Colorado River Basin. Las Vegas is interested in any and every strategy to produce more water.

Cloud-seeding can probably be one of those strategies, says Breed, the NCAR scientist supervising the Wyoming experiment.

Breed cautions against expecting too much from cloud-seeding in Wyoming. “Every little bit helps. It’s not going to be huge (increase), but it will be a little bit.” And, he adds, cloud-seeding operations can be ramped up relatively quickly in years ahead, unlike some other efforts, such as desalination or, less clearly, imports of water into the basin from other sources, such as the Mississippi River.

Second fiddle to conservation

Indeed, that’s the bottom line of the Colorado River Basin Supply and Demand Study, which was issued in December. The Colorado River water has not reached the Pacific Ocean since the late 1990s, and only intermittently since the 1960s. The report examines a whole host of supply augmentation strategies – including, importing water to existing users from other sources, including a pipeline from the Mississippi River at Memphis to the Front Range of Colorado.

But even if pipelines to deliver water into Colorado or Utah from other, more water-plentiful basins could be done profitably, said Eric Kuhn, general manager of the Colorado River Water Conservation District, they’re still unlikely. At a water conference in Gunnison last summer, he was skeptical about any grand, silver-bullet solutions.

Breed also warned against expecting too much. “It is one small tool,” said Breed. “Conservation will be much more important. Cloud-seeding is important, but it is very small compared to all the other tools. It (produced water) is not a huge amount.”

But does it work? Yes, he thinks so.

This story was amended on May 7 to correct the money spent by the Metropolitan Water District for  lower-basin cloud-seeding efforts. The correct figure, according to a district spokesman, is $28,500 per year for the last eight years. 

This originally was published in the March issue of Mountain Town News, a newsmagazine distributed to subscribers. To inquire about rates and advertising opportunities, please contact me at [email protected]


Allen Best