If another country tested nuclear weapons, here’s how we’d know

President Donald Trump has called for the United States to test its nuclear weapons for the first time in three decades. But Trump’s statements about testing — in particular, whether other nations are still doing it — have confused experts in nuclear explosion monitoring. Those experts agree that the main nuclear powers have not performed a major, explosive test of a nuclear weapon since the 1990s, when the Comprehensive Nuclear-Test-Ban Treaty established a norm banning all nuclear explosions.

In an Oct. 31 interview on CBS’s 60 Minutes, Trump stated: “Russia’s testing, and China’s testing, but they don’t talk about it.” When challenged on this claim, he doubled down. “You don’t necessarily know where they’re testing. They test way underground where people don’t know exactly what’s happening with the test.”

But scientists can precisely track nuclear explosions, and they have done so many times. The last country to test nuclear weapons was North Korea, from 2006 to 2017. Those nuclear explosions, which took place underground at the Punggye-ri testing site, were swiftly identified based on the tremors they produced and their release of radioactive isotopes — varieties of chemical elements produced in the blasts.

Seismologist Thorne Lay of the University of California, Santa Cruz has been involved with nuclear monitoring research for decades. Science News spoke with Lay to clarify what we know about nuclear testing around the globe. This interview has been edited for length and clarity.

SN: How would we know if other countries were doing explosive testing of nuclear weapons?

Lay: For many decades, there have been technologies developed to monitor remotely whether an explosion has been conducted in the air, under land or in the ocean — anywhere on the Earth. This is primarily listening for the explosion. It will generate signals that propagate through air, through water, through rock. Those technologies have been around a long time, but they’ve gotten better and better over time.

For the last 20 years, there’s an international monitoring system which the U.S. is part of. It operates seismic stations; hydroacoustic stations, which are listening to the sound waves in the oceans; infrasound stations, which are listening to sound waves in the atmosphere; and radiochemical testers that are scooping up some air and looking for the presence of any isotopes. It’s a worldwide system with hundreds of stations. That supplements existing seismic networks; all seismic networks could detect ground vibrations.

SN: What’s the smallest explosion that scientists can detect?

Lay: It depends a lot on where it is. That will determine how close it is to the nearest seismic stations that we might be operating for monitoring a hostile country. But a rule of thumb is that we can monitor everywhere in the world down to about magnitude 4 [the equivalent of a minor earthquake] with the occurrence of an underground explosion and be able to identify it as an explosion and not an earthquake. That corresponds to about a kiloton explosion [equivalent to 1,000 metric tons of TNT] which is one-fifteenth the size of the Hiroshima bomb.

But if you’re in a place that has a history of nuclear testing, where they have done hundreds of explosions, we have calibrated those sites very well, so they could not hide a very tiny event. We can focus all the seismic listening devices on that area and get down to much less than a kiloton.

SN: Is there any way to hide an underground nuclear explosion?

Lay: The only strategies that have ever reduced the sound significantly are to set it off inside a big cavity, which has to be very big because the explosion will release a lot of energy. Not many places in the world have huge holes in the ground in which you could try to muffle the sound from the nuclear test. It doesn’t completely muffle the sound, but it can reduce it by an order of magnitude or even a little bit larger. That means it will be detected at the closest stations, but not farther away. That’s an inefficient way of suppressing. And that just goes to the general acceptance in the research community that we have not missed nuclear explosions.

SN: How do we know if an explosion is nuclear versus another type of explosion?

Lay: The confirmation that the events are nuclear rather than just chemical is based on their size — if they’re above a magnitude 6, it’s very difficult to set up chemical blasts that big — and from testing for radiochemical isotopes that are released solely during nuclear explosions. If it’s underground, it requires that the material penetrate to the surface. And that’s very common, that some cracking above the explosion will allow gases to escape, and they can bring up radon and other uniquely identifying nuclear products that then can be detected.

SN: It’s not always clear if Trump is referring to explosive nuclear testing or tests of non-nuclear components of the weapons. What do you make of that?

Lay: It’s difficult to say anything useful when it’s all in such an ambiguous context. But what I can say, if they were testing weapons in the traditional sense of setting off nuclear tests and designing weapons and improving them, we would know about that. If we were detecting little magnitude 5 earthquakes in these test sites, we would suspect those were nuclear tests, and that is certainly not happening.