Los Alamos National Laboratory

There’s the kaboom. Researchers at Los Alamos National Laboratory lower equipment underground for the September 1992 “Divider” nuclear test, the last such test that the United States conducted.

Satellites Could Detect Nuclear Tests

SAN FRANCISCO, CALIFORNIA—Even when they're underground, nuclear tests can be detected in the skies—and as a result, global satellite networks could become a powerful new tool in the arsenal of weapons to help detect clandestine underground nuclear explosions, a team of scientists reported here today at the fall meeting of the American Geophysical Union.

The International Monitoring System (IMS), established by the Comprehensive Nuclear-Test-Ban Treaty, has a number of different ears to the ground to detect clandestine nuclear weapons testing: seismic networks that listen for terrestrial shock waves, hydroacoustic networks that scan the oceans for sound waves, and radionuclide networks to sniff out radioactive particles that nuclear explosions produce. But those methods may miss clandestine explosions. Now, Jihye Park, a postdoctoral researcher at Ohio State University (OSU), Columbus, and her colleagues suggest adding another tool to the IMS's arsenal. This one would involve looking up—into the ionosphere, the portion of the upper atmosphere that is ionized by solar radiation.

A nuclear explosion—even when detonated underground—sends up a giant electromagnetic pulse that ripples through Earth's ionosphere. That effect—known as a traveling ionospheric disturbance, or TID—should theoretically be detectable with technologies that are sensitive to changes in the ionosphere, such as global satellite networks and radio telescopes. In fact, there have been proposals to use GPS in detonation detection since about 1979, says Andreas Persbo, executive director of the Verification Research, Training and Information Centre in London, who was not involved in the present study.

But a lot of different sources can produce TIDs, including earthquakes and major storms. So is it possible to uniquely identify an underground nuclear explosion (UNE) among the many factors disturbing the ionosphere's fluctuating swirls of particles? Today at the meeting, Park; Dorota Grejner-Brzezinska, an OSU professor of geodetic and geoinformation engineering; and colleagues announced that they have developed a methodology to do just that.

The key insight, Grejner-Brzezinska says, occurred when she and her team were trying to figure out how to improve the positioning data from the Global Navigation Satellite System, which can be thrown off by ionospheric disturbances between the satellites and their ground-based receiving stations. They wanted to remove those disturbances, or noise. But one man's noise, they realized, is another man's signal: Those very disturbances might offer clues to their sources.

Park and her colleagues previously demonstrated that it was possible to identify a UNE by its ionospheric fingerprint, in a study published in Geophysical Research Letters in 2011. The target in that case was North Korea's 25 May 2009 UNE: Park and her team found a unique TID that also pinpointed the location of the explosion to within about 4 kilometers of its seismically determined epicenter. They saw this TID pattern in data from 11 different Global Navigation Satellite System stations—astronomically unlikely for a random event.

In the current study, the team analyzed signals that GPS stations received after two 20-kiloton UNE tests the United States conducted in 1992. The two tests were part of a series of eight UNEs conducted from 1991 to 1992 at a dusty Department of Energy reservation 100 kilometers northwest of Las Vegas, Nevada. Researchers had begun testing nuclear devices at the Nevada Test Site in 1951; this latest series of blasts was codenamed Operation Julin, and the final two tests of the series—dubbed Hunters Trophy and Divider—took place on 18 September and 23 September, respectively. Those two also became the last nuclear tests the United States conducted before President George H. W. Bush signed a law imposing a moratorium on all nuclear weapons testing, on 2 October 1992. (The 1963 Limited Test Ban Treaty had already banned all but underground tests.) In 1996, the United States, Russia, the United Kingdom, France, and China signed the Comprehensive Nuclear-Test-Ban Treaty—but the United States has yet to ratify it.

The team came up with a relatively simple algorithm to find the signal within the noise. They first removed the effect of distortions from changes to the diurnal cycle and from the changing geometry of the satellites themselves. Then they converted the ionospheric delay between satellites and stations into a "total electron content" in the TIDs. From all of these data, they came up with a profile for the TIDs: their amplitude, their frequency, and how quickly they traveled through the ionosphere. That same profile appeared in multiple stations—and as a result, based on where the stations were and how long it took that fingerprinted signal to arrive, the team was also able to pinpoint the location of the original signal—the Hunters Trophy blast. They devised a similar algorithm to identify and characterize the Divider blast using GPS data.

The Very Large Array (VLA) of radio telescopes, located near Socorro, New Mexico, used a similar algorithm to come up with a similar result. VLA measures correlations between signals from pairs of antennas to reconstruct images of the sky, as though they were one single, giant telescope—and so VLA, too, is sensitive to ionospheric fluctuations. After performing similar calculations, the team announced, VLA found a fingerprint for the Hunters Trophy TID that strongly resembled that of the GPS signal.

One advantage to using GPS detection, Grejner-Brzezinska notes, is that the infrastructure is already in place and available globally at no cost. But a lot of kinks still must be ironed out to make the method operational, the team acknowledges—particularly sensitivity limits, and how well the method can distinguish between different point sources, such as earthquakes and UNEs.

"It's always good to have additional discriminants," says Paul Richards, a seismologist at Columbia University's Lamont-Doherty Earth Observatory in Palisades, New York, who was not involved in the study. "But we do have a number of effective ones already." Richards emphasizes the uncertainties still in the method but says that with further progress “at some point we can make comparisons with what is already available and then see if these techniques add a useful arrow to the quiver.”

In any case, politics could still overshadow any advantages the new technique might offer. "In my mind, the method shows promise," Persbo says. "Whether or not the method will be included in formal CTBT monitoring, however, is an open question. The treaty leaves the door open, but I don't think that there is much appetite amongst member states to discuss the formal incorporation of new technologies until the treaty shows signs of entering into force."