Cloud Seeding for Precipitation Enhancement

The Physics of Supercooled Clouds

Water does not always freeze at zero degrees Celsius. In the clean, relatively particle-free air of many clouds, liquid droplets can persist down to temperatures of minus ten, minus twenty, or even minus forty degrees Celsius before freezing spontaneously. This is because ice formation, like most phase transitions, needs a starting point, a process called nucleation. Pure water droplets lack the microscopic surfaces or impurities that let ice crystals begin growing, so they remain in a metastable liquid state known as supercooled water. Clouds dominated by supercooled droplets are common in winter storms, mountain wave clouds, and the upper portions of many convective systems. These clouds are the primary target of glaciogenic seeding, because they are, in a sense, primed and waiting for a trigger. Without enough natural ice-forming particles, called ice nuclei, drifting through the cloud, precipitation formation can be inefficient or delayed. The atmosphere does contain natural ice nuclei, mineral dust, biological particles, and other aerosols, but their concentration varies enormously by location, season, and air mass. In clean maritime air masses or certain continental conditions, a shortage of effective ice nuclei can leave a cloud holding large amounts of liquid water that never converts into falling precipitation. Understanding this gap between the water a cloud holds and the water it actually releases is central to the entire rationale for seeding. The technique does not create water from nothing; it attempts to more efficiently convert water that is already present in the cloud into precipitation that reaches the ground, and only in clouds where that inefficiency genuinely exists.

Glaciogenic Seeding: Silver Iodide and Dry Ice

The most established form of weather modification is glaciogenic seeding, which targets clouds containing supercooled liquid water. Silver iodide is the workhorse material because its crystal lattice structure closely resembles that of ordinary ice, close enough that ice can begin forming directly on its surface at temperatures as warm as minus four to minus six degrees Celsius, a process called heterogeneous nucleation. Silver iodide particles are typically generated by burning acetone-based silver iodide solutions in ground-based generators or dropping pyrotechnic flares from aircraft, producing enormous numbers of microscopic particles that disperse through the target cloud. Dry ice, solid carbon dioxide, works through a completely different mechanism. Rather than providing a surface for ice to grow on, dry ice pellets are extremely cold, roughly minus seventy-eight degrees Celsius, and cause intense localized cooling as they fall through a cloud. This extreme chilling can cause droplets in the immediate vicinity to freeze spontaneously without needing any nucleating surface at all, a process sometimes called homogeneous or near-homogeneous freezing. Once ice crystals exist by either method, the Wegener-Bergeron-Findeisen process takes over. At a given sub-zero temperature, the saturation vapor pressure over ice is lower than the saturation vapor pressure over liquid water. This difference means the air can be simultaneously slightly supersaturated with respect to ice while being undersaturated with respect to the surrounding liquid droplets. Water vapor therefore diffuses toward and deposits onto the ice crystals, while the liquid droplets evaporate to replenish that vapor. The ice crystals grow at the direct expense of the droplets, often quite rapidly, until they are heavy enough to fall as snow, or as rain if they melt on the way down.

Hygroscopic Seeding: Working with Warm Clouds

Not all rain-producing clouds are cold enough to contain supercooled droplets. Many tropical and warm-season clouds produce precipitation entirely through a liquid-only process called the collision-coalescence mechanism, where droplets of different sizes collide and merge as they move through the cloud, gradually growing large enough to fall as rain. Hygroscopic seeding targets these warm clouds by introducing salt-based particles, commonly potassium chloride, sodium chloride, or calcium chloride, often released as flares from aircraft flying through the lower portions of a cloud's updraft. These salts are hygroscopic, meaning they readily attract and absorb water vapor from their surroundings. When dispersed into a cloud, the salt particles rapidly grow into unusually large droplets compared to the cloud's natural droplet population. Because collision-coalescence efficiency depends heavily on having a range of droplet sizes, these oversized seeded droplets act as effective collectors, sweeping up smaller droplets as they fall and accelerating the growth of raindrops that would otherwise take longer to form naturally, or might not form at all before the cloud dissipates. This approach is particularly relevant in regions like parts of Africa, South America, and Southeast Asia, where warm convective clouds dominate and glaciogenic techniques are simply inapplicable because temperatures never drop low enough to sustain the ice-based process. Hygroscopic seeding research has a shorter and, in some ways, less consistent history than silver iodide work, and questions remain about the optimal particle size distribution and release timing needed to reliably enhance coalescence without producing particles too large or too few to have a meaningful effect.

Historical Programs and Operational Practice

Weather modification has been attempted since the late 1940s, when Vincent Schaefer and Bernard Vonnegut at General Electric first demonstrated that dry ice and later silver iodide could trigger ice formation in supercooled clouds. Since then, cloud seeding programs have operated in dozens of countries, targeting winter snowpack in mountain ranges to boost water supplies, attempting to increase rainfall over agricultural regions, and in some cases trying to suppress hail damage to crops. In the western United States, states including Colorado, Utah, Idaho, and Wyoming run winter orographic seeding programs, releasing silver iodide from ground generators positioned upwind of mountain ranges, aiming to increase snowpack that later feeds into reservoirs. China has operated one of the largest weather modification programs in the world for decades, using aircraft, ground generators, and anti-aircraft-style rockets to seed clouds ahead of major events and during droughts. Numerous other countries, including the United Arab Emirates, Thailand, and Australia, have run or continue to run operational programs. These programs are typically justified using a combination of physical reasoning, the atmospheric conditions were favorable for seeding, and statistical comparisons between seeded and historical unseeded periods. Operational programs generally lack the rigorous controls of scientific experiments, since it would require withholding a potentially beneficial intervention from areas experiencing water scarcity purely for the sake of a clean control group, a tradeoff that operators and funders are often unwilling to accept, even though it limits what can be scientifically concluded from the results.

The Persistent Uncertainty in Measuring Effectiveness

Despite roughly eight decades of research, the central scientific question, how much extra precipitation does seeding actually produce, remains genuinely unresolved for most operational contexts. The fundamental difficulty is a counterfactual problem: it is impossible to know exactly how much precipitation a specific seeded cloud would have produced had it not been seeded. Well-designed randomized experiments, where suitable clouds are randomly assigned to be seeded or left alone and outcomes compared statistically, have produced some of the more credible positive results. The Wyoming Weather Modification Pilot Program and a study over Idaho's Payette Basin, both using aircraft with specialized instruments to directly observe seeded ice crystal formation and track resulting snowfall, are frequently cited as producing relatively strong physical evidence of a measurable seeding effect in specific orographic winter storm conditions, generally estimated in the range of a few percent to around fifteen percent additional precipitation in favorable cases. However, many other trials, including a large body of older experiments from the mid-twentieth century, produced inconclusive, inconsistent, or statistically insignificant results, and some well-designed randomized experiments found no detectable effect at all. Reviews by bodies such as the World Meteorological Organization and the National Academies of Sciences have repeatedly concluded that while the underlying physical mechanisms are sound and can be directly observed in ideal conditions, translating that into a reliable, quantified precipitation increase across the variety of clouds, terrains, and storm types encountered in everyday operational programs remains scientifically unresolved. Confounding factors abound: natural variability in precipitation is enormous, suitable target clouds are not always correctly identified in advance, seeding material does not always reach the intended part of the cloud, and separating a seeding signal from natural noise statistically requires far more repeated trials than most programs ever conduct. Operational claims of large percentage increases in water supply should be treated with considerable caution unless backed by the kind of randomized, instrumented, physically verified studies that remain the exception rather than the rule.

Frequently asked questions

Try it live

Everything above runs in your browser — open Cloud Seeding for Precipitation Enhancement and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.