Cold War Nuclear Testing May Have Caused Extra Rain Around the World

Photo: Ethan Miller / Getty

A new study has found that nuclear radiation during Cold War-era weapons testing could have induced significant short-term changes in the amount of rainfall far across the globe..


I’ve said it before and I’ll say it again: Weather is just an extremely complex physics experiment, with tons of variables that are hard of keep track of. In the case of rain, small water vapor droplets in the atmosphere collide and stick together, growing into clouds and eventually falling from the sky. But radiation can modify the electrical charges of the droplets, influencing how much they stick together.

Weapons testing “was a substantial perturbation in the atmospheric sense,” the study’s first author, R. Giles Harrison from the Department of Meteorology at the University of Reading in the UK, told Gizmodo.

The researchers compiled data from various sources, including the annual average of the atomic isotope strontium-90 in the atmosphere as recorded by the High Altitude Sampling Program; the amount of charged atoms produced near Earth’s surface during the late 1950s and early 1960s, and the electrical current between the air and Earth taken in London during the same period. All of these showed obvious increases in radioactivity and electrical activity during the early 1960s, when there were a number of atmospheric nuclear tests around the world. While these tests didn’t take place near the United Kingdom, the resulting radiation spread throughout the atmosphere.

But did that increase in electrical activity affect rainfall? The team gathered cloud and rainfall data taken from the Lerwick observatory in Shetland, Scotland in the United Kingdom. They observed in the data that clouds were significantly thicker during this period and that when it did rain, there was 24 percent more precipitation, according to the paper published in Physical Review Letters.

While the researchers couldn’t nail down the exact mechanism, they posited that more radioactivity and more electric charge in water droplets affected how the droplets merged together and grew. Perhaps it’s not such a surprise that radiation from weapons testing could cause impacts from far away—for example, steel produced around the world after World War II contains more radioactive isotopes than pre-WWII steel as a result of weapons testing.

Roelof Bruintjes, project scientist at the National Center for Atmospheric Research who was not involved in the new study, told Gizmodo that the paper certainly presented an interesting concept from a reputable group and that there’s likely truth to it. But he wouldn’t say it was definitive.


“It shows a certain trend that needs to be explored further,” he said. The paper relies on lots of disparate sources of data from six decades ago. It’s possible there was just a weird, coincidental weather fluke at the same time as the nuclear testing. But, he said, it certainly calls for more research in the area.

After all, we already know that particles released by plants and humans can impact the growth of clouds. But, said Bruintjes, there’s still a whole lot we don’t know about the cloud-forming process. Harrison told Gizmodo he hopes that further experiments with weather balloons or other instruments can study droplets coalescing into clouds in more detail.


Ultimately, understanding the effect of radiation or electric charge on cloud formation could be important for scientists researching geoengineering or other technology to address the effects of climate change. But, said Bruintjes, we need to understand single clouds before we can begin to talk about global-scale projects to alter weather.

Former Gizmodo physics writer and founder of Birdmodo, now a science communicator specializing in quantum computing and birds


Times up, time to leave!

You’re spot on, weather is hard!

Some time ago I found an interesting graphic that shows how much climate modelling has improved over the years. Going back just 30 years to 1990 the typical models used a spatial grid of around 500x500km, by 2007 that was down to around 110x110km. The main issue is the compute needed to run the models so as tech improved the models got bigger and yet could still complete a model in reasonable time.

This of course is only one layer, there is also a vertical axis needed as well, this started out around 11km per cell and has got tighter over the years as well. Then on top of that you have to add the temporal dimension also. It is a monumental amount of data points any way you slice it and there are so many things that can easily mess up the sims during a run.

And then, even when all this is working right you still get localized weather events caused by topography etc that this will miss entirely because the resolution is still too crude. I’m totally amazed they get things as close as they do.