Helping inspectors locate and identify underground nuclear tests

“The work is novel in part because of how we did it by injecting gases into an old UNE cavity and then using computer models informed by the experiment to extend our understanding of how xenon gas evolves following the UNE,” said Charles Carrigan, LLNL geoscientist and lead author of a paper appearing in the 16 March edition of the journal Nature-Scientific Reports.

Using computer models developed by LLNL physicist Yunwei Sun, the team showed that including thermally driven migration of telltale gases from the explosion cavity or chimney may substantially shorten their arrival times at the surface when compared to migration of gases caused only by atmospheric pressure fluctuations or barometric pumping. Previous research has focused on barometric pumping as the primary subsurface gas migration mechanism.

“From monitoring gases coming to the surface during the course of our pressurized field experiment, we also found that background radon gas levels were anomalously high (10 to 15 times normal) at the surface over the explosion cavity,” Carrigan said.

The research indicates that the weak subsurface pressurization mimicking the thermal drive following the explosion enhanced the amount of radon that was captured. This suggests that radon anomalies could be potential indicators of hidden or clandestine UNEs that are otherwise difficult to detect during an on-site inspection.

Additionally, the simulations showed that the explosion cavity or chimney behaves something like a leaky chemical reactor or pressure cooker. The gases migrating away from the cooker change the overall chemical makeup (isotopic ratios) of the gases left behind in the cooker or reactor, which continues to make new gases. The team modeled the evolution of these gases out to several months following a UNE.

“The 2013 underground nuclear explosion carried out in North Korea has allowed us some validation of our model of explosion-gas evolution,” Carrigan said. “We find that the gases detected almost two months afterward in Russia are best matched by our evolutionary model for the mixture of different xenon isotopes when we assume a range of yields that is consistent with seismic estimates, less than 10 kilotons, for that event. This is a cool result as no one has suggested that isotopic ratios should depend on nuclear yield.”

The research also may have applications in monitoring other heated or pressurized subsurface regimes such as in situ coal gasification, deep sequestration of supercritical CO₂ and nuclear waste disposal.

LLNL notes that this work was performed as part of the multi-laboratory Underground Nuclear Explosion Signatures Experiment, which is supported by the National Nuclear Security Administration’s Office of Proliferation Detection.

— Read more in Charles Carrigan et al., “Delayed signatures of underground nuclear explosions,” Scientific Reports 6 (16 March 2016): 23032 (doi:10.1038/srep23032)