Cloud seeding can increase precipitation by roughly 5 to 15% under the right conditions, though results vary widely depending on cloud type, temperature, and geography. It is not a way to create rain from a clear sky. The technique works by adding particles to existing clouds to help water droplets or ice crystals grow large enough to fall as precipitation. After nearly 80 years of use, the scientific consensus is that cloud seeding works in specific, limited circumstances, but measuring exactly how well it works remains genuinely difficult.
How Cloud Seeding Works
There are two main approaches. Glaciogenic seeding targets cold clouds that contain supercooled water, meaning liquid water that exists below freezing but hasn’t turned to ice yet. Operators release silver iodide particles or dry ice into these clouds, which trigger ice crystal formation. Those crystals grow and eventually fall as snow or rain. This is the most common form of cloud seeding and has been used since the late 1940s, when researchers first demonstrated that silver iodide and dry ice could nucleate ice in laboratory clouds.
The second approach, hygroscopic seeding, works on warmer liquid clouds. Large salt particles are released near the cloud base, where they speed up the natural process of small droplets merging into larger ones heavy enough to fall. This method is newer and less extensively studied, but several trials have reported positive results in tropical and subtropical regions.
What the Evidence Actually Shows
The World Meteorological Organization’s official position is that “there is statistical evidence, and physical evidence from observations, of precipitation enhancement” from glaciogenic seeding of certain cloud types, particularly winter clouds that form over mountains and some frontal weather systems containing supercooled liquid water. The organization also notes that results depend heavily on natural cloud characteristics, and that more research is needed to pin down exactly which conditions produce the best outcomes.
The fundamental challenge in proving effectiveness is the counterfactual problem: you can never know exactly how much it would have rained if you hadn’t seeded. Researchers use statistical comparisons between seeded and unseeded periods, physical measurements of ice crystal formation inside clouds, and radar tracking of precipitation. Each method has limitations. A 10% increase in snowfall over a mountain range during a winter storm is real water, but it’s small enough that natural variability can easily mask or mimic it in any single event.
That said, multiple independent programs across decades have converged on similar numbers. Winter orographic seeding, where silver iodide is released upwind of mountain ranges to boost snowpack, consistently shows precipitation increases in the 5 to 15% range. Some convective (thunderstorm) seeding programs have reported larger increases, but with greater uncertainty.
Conditions That Must Be Met
Cloud seeding only works when the atmosphere is already close to producing precipitation on its own. The Desert Research Institute, which operates seeding programs across the western United States, requires several conditions before initiating operations. Clouds must cover at least 50% of the target area and be deep enough that their bases sit at or below the highest mountain peaks. Supercooled liquid water must be present. And the temperature near mountaintop level needs to reach negative 5°C or colder, though operations can begin at negative 3°C if colder temperatures are forecast within a few hours.
If any of these conditions are missing, seeding won’t produce meaningful results. Clear skies, thin clouds, or clouds that are too warm simply don’t respond to the technique. This is why cloud seeding is a supplement to natural weather patterns, not a replacement for them. Drought-stricken regions with no incoming moisture can’t seed their way to rain.
Cost Compared to Other Water Sources
Where conditions are favorable, cloud seeding is remarkably cheap. Utah estimates the additional water produced by its winter seeding program costs $5 to $10 per acre-foot. For comparison, the state’s Demand Management Pilot Program pays farmers $390 per acre-foot to leave their land unirrigated so more water stays in the Colorado River. Desalination typically costs $1,000 to $2,500 per acre-foot depending on the facility. Even if cloud seeding’s actual yield is at the low end of estimates, the economics are compelling for water-scarce regions that have the right cloud conditions.
Does It Steal Rain From Neighbors?
One of the most common concerns about cloud seeding is that it simply redirects rainfall, benefiting the target area at the expense of regions downwind. The intuition makes sense: if you squeeze more water out of a cloud, there should be less moisture left for the next area in its path. But the evidence points in the opposite direction.
A comprehensive assessment of “extra area” effects found that precipitation changes beyond the target zone were uniformly positive, with increases of 5 to 15% extending up to a couple hundred kilometers from the seeding area. The likely explanation is that seeding doesn’t wring clouds dry. It accelerates processes that were already underway, and the dynamic effects of releasing latent heat (when water vapor turns to ice or liquid) can actually invigorate cloud systems rather than deplete them. Cloud seeding, in other words, typically benefits both the target area and its neighbors.
Scale of Global Operations
Cloud seeding is not experimental in the way most people assume. More than 50 countries have used some form of weather modification. China operates the world’s largest program, now covering more than 50% of the country’s land area, primarily to increase rainfall but also to suppress it in certain areas. The United States has active programs across at least eight western states, focused mainly on boosting winter snowpack in mountain watersheds that supply water to cities and farms downstream. The United Arab Emirates, Australia, India, and several other countries maintain their own programs.
The scale of China’s effort is particularly striking. Its weather modification infrastructure includes thousands of ground-based generators, aircraft, and even rocket launchers designed to deliver silver iodide into clouds. Whether results at that scale match the 5 to 15% increases seen in more targeted programs is an open question, and independent verification of China’s claims remains limited.
Environmental Concerns With Silver Iodide
Silver iodide is the most widely used seeding agent, and its environmental footprint has received scrutiny. Silver is toxic to soil microorganisms and can inhibit bacterial enzymes. In aquatic environments, marine algae can accumulate silver at concentrations 13,000 to 66,000 times higher than surrounding water. However, studies of actual food chains show little evidence of biomagnification, meaning silver doesn’t systematically concentrate as it moves up from algae to invertebrates to fish.
Precipitation from seeded clouds contains silver concentrations of roughly 10 to 4,500 nanograms per liter, compared to 0 to 20 nanograms per liter in unseeded precipitation. While those numbers sound dramatic in relative terms, the absolute concentrations remain very low. The Agency for Toxic Substances and Disease Registry notes that silver released through cloud seeding “is not expected to contribute significant amounts to water.” Total silver iodide use from cloud seeding was estimated at around 3,100 kilograms per year in the early 1970s, a small fraction of atmospheric silver compared to ore processing, fossil fuel combustion, and waste incineration. Still, localized accumulation near frequently seeded watersheds is a reasonable concern that deserves ongoing monitoring, particularly in sensitive alpine ecosystems.
Why Certainty Remains Elusive
After decades of research and operational use, cloud seeding occupies an unusual scientific space. It clearly does something. Physical measurements confirm that silver iodide triggers ice formation in supercooled clouds. Radar observations show precipitation forming in seeded cloud regions. Multiple randomized experiments have found statistically significant increases. Yet the magnitude of the effect in any given program is hard to nail down precisely, because weather is inherently variable and the signal is modest compared to the noise.
The practical takeaway is that cloud seeding is a real but limited tool. It won’t solve water crises or end droughts. It can, however, add a meaningful percentage to snowpack and rainfall in regions that already receive moisture, at a fraction of the cost of other water supply options. For water managers in the right geography, that incremental gain is worth pursuing.