Wyoming just spent $14 million and the better part of 10 years on a rigorous scientific experiment to evaluate whether it's possible to get extra snow from winter storm clouds through cloud seeding. The conclusion? The final results were thin: There was a 3 percent increase in precipitation, but a 28 percent probability that the cloud seeding had nothing to do with it.
Given the results of this and other winter weather-modification studies, the U.S. Bureau of Reclamation remains unimpressed. "As such," said the agency in a draft analysis released in February, "the 'proof' the scientific community has been seeking for many decades is still not in hand."
Proof in science requires a 95 percent probability of causality. But this standard is extremely difficult to achieve in complex atmospheric processes. Climate scientists, for example, mostly resort to asterisk-laden words such as "likely" to indicate lower levels of probability.
Cloud seeders, in turn, have labored under clouds of suspicion. Some of this is from decades of over-reaching claims made by commercial operators. Yet some meteorologists have also grumbled that cloud seeding is expected to deliver levels of probability that are not required when it comes to the science of climate change.
In Wyoming, in the realm of public policy, this lack of definitiveness has worked in ironic ways. Elected officials there have been moving to expand cloud seeding even though they lack convincing proof that it works. At the same time, many elected officials refuse to accept the existence of global warming, claiming lingering uncertainty in the science.
From the start, Wyoming's cloud-seeding experiment was designed to ensure scientific rigor. Parallel mountain ranges southwest of Laramie, just north of the Colorado border, constituted the Wyoming laboratory. Propane was burned to loft silver iodide from ground-based generators into the clouds passing over the Sierra Nevada and Medicine Bow ranges. In the experiment, 154 storms during six winters had the temperatures needed for effective seeding, but only 118 developed adequate moisture content. And of those, 18 were tossed out because of contamination problems.
Dan Breed, project scientist with the National Center for Atmospheric Research, which designed and oversaw the Wyoming experiment, said that failing to achieve a 95 percent confidence level in results is not unusual in cloud seeding studies. The fundamental problem, he says, involves the difficulty of measuring atmospheric processes.
The challenge inherent in the complexity of the data has prevented most climate scientists from directly linking specific weather events, such as the September 2013 floods in Colorado, to rising global temperatures, or even to the 3 to 5 percent observed moisture in the global atmosphere.
"When it comes to the atmosphere, there are just too many variables, and that variability just keeps rearing its ugly head when it comes to cloud seeding," says Breed. "Even in this case, where we tried to make things as homogeneous as possible to reduce that variability, variability still kind of hurt us."
Breed thinks research might better be invested in understanding the interaction in the atmosphere of wind, temperature and precipitation. For example, how likely is it that silver iodide or other seeding agents released from the ground will get into the clouds? True understanding of atmospheric processes, says Breed, has mostly come from observations instead of experiments—because of that same variability.
This lack of certainty does not necessarily kill the prospects of cloud seeding, as is demonstrated by the continued interest of Wyoming legislators in funding projects. In the Colorado River Basin, cities and water districts seized upon the modeled projections of 5 to 15 percent snowpack augmentation as justification for continued or even expanded operations. Already, metropolitan Los Angeles, the Central Arizona Project and the Southern Nevada Water Authority, among others, pay for seeding clouds in Colorado, the source of half the water in the Colorado River, and last year they also paid to seed clouds in Wyoming, in the Green River drainage.
The Colorado River is notoriously strapped in its capacity to meet all of the wants and maybe even the needs of the millions of people who depend on it. River flows have declined 20 percent in the 21st century as compared to the last century. Though Breed won't say that cloud seeding doesn't necessarily work, he doesn't see it as a game-changer for the Colorado River. Though cloud seeding can be a fairly straightforward, quick and inexpensive way to produce more water, its gains are marginal. It is not, he adds, a magic bullet: "It won't solve the problem."
Allen Best is a contributor to Writers on the Range, a column service of High Country News (hcn.org). He writes in the Denver area of Colorado.