The Science of Making Rain

[This is a transcript of the video embedded below]


Wouldn’t it be great if we could control the weather? I am sure people have thought about this ever since there’ve been people having thoughts. But what are scientists thinking about this today? In this video we’ll look at the best understood case of weather control, that’s making rain by seeding clouds. How is cloud seeding supposed to work? Does it work? And if it works, is it a good idea? That’s what we’ll talk about today.

First things first, what is cloud seeding? Cloud seeding is a method for increasing precipitation, which is a fancy word for water that falls off the sky in any form: rain, snow, hail and so on. One seeds a cloud by spraying small particles into it, which encourages the cloud to shed precipitation. At least that’s the idea. Cloud seeding does not actually create new clouds. It’s just a method to get water out of already existing clouds. So you can’t use it to turn a desert into a forest – the water needs to be in the air already.

Cloud seeding was discovered, as so many things, accidentally. In nineteen-fourty-six a man named Vincent Schaefer was studying clouds in a box in his laboratory, but it was too warm for his experiment to work. So he put dry ice into his cloud box, that’s carbon dioxide frozen at about minus eighty degrees Celsius. He then observed that small grains of dry ice would rapidly grow to the size of snowflakes.

Schaefer realized this happened because the water in the clouds was supercooled, that means below freezing point, but still liquid. This is an energetically unstable state. If one introduces tiny amounts of crystals into a supercooled cloud, the water droplets will attach to the crystals immediately and freeze, so the crystals grow quickly until they are heavy enough to fall down. Schaefer saw this happening when sprinkles of solid dry ice fell into his box. He had seeded the first cloud. In the following years he’d go on to test various methods of cloud seeding.

Today scientists distinguish two different ways of seeding clouds, either by growing ice crystals, as Schaefer did, that’s called Glaciogenic seeding. Or by growing water droplets, which is called hygroscopic seeding.

How does it work?

The method that Schaefer used is today more specifically called the “Glaciogenic static mode”, static because it doesn’t rely on circulation within the cloud. There’s also a Glaciogenic dynamic mode which works somewhat differently.

In the dynamic mode, one exploits that the conversion of the supercoooled water into ice releases heat, and that heat creates an updraft. This allows the seeds to reach more water droplets, so the cloud grows, and eventually more snow falls. One of the substances commonly used for this is silver iodide, though there are a number of different organic and inorganic substances that have proved to work.

For hygroscopic seeding one uses particles that can absorb water that serve as condensation seeds to turn water vapor into large drops that become rain. The substances used for this are typically some type of salt.

How do you do it?

Seeding clouds in a box in the laboratory is one thing, seeding a real cloud another thing entirely. To seed a real cloud, one either uses airplanes that spray the seeding particles directly into the cloud, or targets the cloud with a rocket which gives off the particles, or one uses a ground-based generator that releases the particles slowly mixed with hot air, that rises up into the atmosphere. They do this for example in Colorado, and other winter tourism areas, and claim that it can lead to several inches more snow.

But does it work?

It’s difficult to test if cloud seeding actually works. The issue is, as I said, seeding doesn’t actually create clouds, it just encourages clouds to release snow or rain at a particular time and place. But how do you know if it wouldn’t have rained anyway?

After Schaefer’s original work in the nineteen-fifties, the United States launched a research program on cloud seeding, and so did several other countries including the UK, Canada, India, and Australia. But evidence that cloud seeding works didn’t come by for a long time, and so, in the late nineteen-eighties, funding into this research area drastically declined. That didn’t deter people from trying to seed clouds though. Despite the absence of evidence quite a few winter sport areas used cloud seeding in an attempt to increase snow fall.

But beginning around the turn of the millennium, interest in cloud seeding was revived by several well-funded studies in the United States, Australia, Japan, and China, for just to name a few. Quite possibly this interest was driven by the increasing risk of drought due to climate change. And today, scientists have much better technology to figure out whether cloud seeding works, and so, the new studies could finally deliver evidence that it does work.

Some of the most convincing studies used radar measurements to detect ice crystals in clouds after a plane went through and distributed the seeds. This was done for example in a 2011 study in Australia and also in a 2018 study in the northern part of the United States.

These radar measurements are a direct signature of seeding, glaciogenic seeding in this case. The researchers can tell that the ice crystals are caused by the seeding because the crystals that appear in the radar signal replicate the trajectory of the seeding plane, downwind.

From the radar measurements they can also tell that the concentration of ice crystals is two to three orders of magnitude larger than those in neighboring, not-seeded areas. And, they know that the newly formed ice-crystals grow, because the amount of radar signal that’s reflected depends on the size of the particle.

This and similar studies also contained several cross checks. For example, they seeded some areas of the clouds with particles that are known to grow ice crystals and others with particles that aren’t expected to do that. And they detected ice formation only for the particles that act as seeds. They also checked that the resulting snowfall is really the one that came from the seeding. One can do this by analyzing the snow for traces of the substance used for seeding.

Besides this, there are also about a dozen studies that evaluated statistically if there changes in precipitation from the glaciogenic static seeding. These come from research programs in the United States, Australia, and Japan. To get statistics, they monitor the unseeded areas surrounding the seeded region as an estimation of the natural precipitation. It’s not a perfect method of course, but done often enough and for long enough periods, it gives a reasonable assessment for the increase of precipitation due to seeding.

These studies typically found an increase in precipitation around 15% and estimated the probability that this increase happened just coincidentally with 5%.

So, at least for the seeding of ice crystals, there is now pretty solid evidence that it works better than a rain dance. For the other types of seeding it’s still unclear whether it’s efficient.

Please check the information below the video for references to the papers.

The world’s biggest weather modification program is China’s. The Chinese government employs an estimated 35,000 people to this end already, and in December 2020 they announced they’ll increase investments into their weather modification program five-fold.

Now, as we have seen, cloud seeding isn’t terribly efficient and for it to work, the clouds have to be already there in the first place. Nevertheless, there’s an obvious worry here. If some countries can go and make clouds rain off over their territory, that might leave less water for neighboring countries.

And the bad news is, there aren’t currently any international laws regulating this. Most countries have regulations for what you are allowed to spray into the air or how much, but cloud seeding is mostly legal. There is an international convention, the Environmental Modification Convention, that seventy-eight states have signed, which prohibits “the military and hostile use of environmental modification techniques.” But this can’t in any clear way be applied to cloud seeding.

I think that now that we know cloud seeding does work, we should think about how to regulate it, before someone actually gets good at it. Controlling the weather is an ancient dream, but, thanks to Vincent Schaefer, maybe it won’t remain a dream forever. When he died in 1993, his obituary in the New York Times said “He was hailed as the first person to actually do something about the weather and not just talk about it”.