In an extraordinarily brief statement, the Director of the National Intelligence Office announced that the United States has confirmed that North Korea’s large explosion last week was nuclear. How do they know and why did it take them so long to confirm?
Apparently, the North Koreans had announced the test ahead of time to the Chinese and the Russians but the first physical evidence that something had happened in North Korea was seismic signals indicating a large explosion. (Some of the seismic recording stations are operated by the Comprehensive Test Ban Treaty Office and are designed to verify compliance with the Comprehensive Test Ban Treaty, which has not been ratified by the United States.) The seismic data indicated a fairly small test (by nuclear standards), less than a kiloton, probably only a half kiloton. Some reports cite as low as 200 tons.
When we get down to half a kiloton, 500 tons, of high explosive (specifically TNT) equivalent, then it is technically conceivable that the explosion could have actually been conventional, not nuclear. Given the North Korean penchant for bluff, could they have actually set off half a kiloton of conventional explosives? The United States is planning a conventional test, Divine Strake, that will simulate nuclear effects and it will be over 700 tons of ammonium nitrate and fuel oil explosive, equivalent to almost 600 tons of TNT. A semi-trailer truck can carry 25 tons of cargo, so the 700 tons for Devine Strake, more than enough to simulate the North Korean test, would be 35 truckloads. That is a lot of ammonium nitrate, but not the sort of activity that would necessarily jump out at intelligence analysts monitoring North Korea. Seismic signals can distinguish explosions from small earthquakes but cannot tell the difference between conventional and nuclear explosions. So North Korea could have been faking it.
This explanation seems increasingly unlikely in light of recent reports that radioactivity consistent with a nuclear weapon has been detected downwind of the North Korean test site. This could nail the test as nuclear. It is not easy to seal up a nuclear explosion and we don’t even know how hard the North Koreans tried. If they wanted to minimize radioactive venting, they would have plugged the tunnel with a backfill of soil or clay and might have included concrete barriers. These might have ruptured or leaked during the test. If this happened, then quite an array of radioactive materials might have made it out and could have been detected. In addition to leakage, there is seepage. Even with a successful immediate seal of the test, gases might seep out through porous rock or cracks in the rock. According to my geological map of North Korea, the reported test site is in granite. Granite is not permeable but gases might leak out through difficult-to-detect cracks.
After the test, air samples taken down wind could be collected, returned to a laboratory and tested for even tiny traces of radioactivity. A leak would result in many telltale radioactive particles that would even give us some information about the design of the bomb. A seep most likely would give much less information, since the sample might be largely limited to gases, like xenon-133 with a five day half life. Xenon-133 might come from a nuclear fuel reprocessing facility. If that possibility could be eliminated, then a nuclear test is the most plausible remaining explanation. The DNI statement gave no details, referring simply to radioactive debris.
Anonymous sources in the government apparently have reported that the bomb was plutonium. I have no special inside information so I can only speculate, but this probably implies that there was a leak, not a seep, and we got a good radioactive sample. (Another point in the cited New York Times article, that use of plutonium indicates that the North Korean uranium enrichment program is not developed, is difficult to understand. When the United States tested a plutonium bomb in 1945, the test indicated nothing at all about the highly advanced state of its uranium enrichment program.)