The Tunnels at Punggye-ri: An Alternative View

In recent weeks, a number of news outlets have reported that North Korea may be readying for another nuclear weapons explosion at its test site. The stories resulted from a remark by the South Korean Minister of Defense, who told lawmakers in Seoul that North Korea had finished preparations for another test at its Punggye-ri nuclear test site—although he added that no test was imminent.

The Minister’s statement matches closely the reports that have appeared at 38 North—one in October 2013 noting the presence of new tunnel entrances, followed by reports in December 2013 and February 2014 showing a significant acceleration in excavation of one of the new sites. (North Korea appears to be digging into the same mountain on the north side of its test site—usually called the West Portal area—where it conducted its 2009 and 2013 nuclear tests, while stopping excavation at another mountain south of the site—the South Portal—where digging is either done or on hold.)

We generally think of North Korea as digging tunnels and then conducting a single test in each tunnel in 2006, 2009, and 2013. The pattern of excavation at Punggye-ri, however, raises a disquieting possibility. What if North Korea’s recent excavations are not for new tunnels that will be used only once, but represent an effort to transform the mountains north and south of the site into complexes that could allow it to conduct multiple tests—two or more—in drifts off a single main tunnel with multiple entrances.

If and when North Korea has a steady supply of fissile material in the form of highly enriched uranium from facilities at Yongbyon (and who knows where else) as well as a steady but smaller supply of plutonium from its reactivated five megawatt reactor, might the North prepare to conduct nuclear explosions on a much more regular basis?

After North Korea’s first nuclear test in 2006 under a mountain to the east of the test complex, Pyongyang began digging a tunnel under a different mountain to the north, referred to as the West Portal. After its second nuclear test in 2009, the North Koreans began digging a tunnel into a third mountain to the south of the site—but there was a surprise. North Korea’s third nuclear test occurred under the same mountain to the north as the 2009 detonation. Now, imagery reveals a new entrance again at the northern mountain, while the tunnels or tunnel complex under the southern mountain appear complete.

It is possible that each entrance into the West Portal is for a separate tunnel that runs parallel to the others. But an alternate hypothesis suggested by patterns of US, Russian and Chinese underground nuclear testing is that, rather than parallel tunnels, North Korea may be conducting tests in drifts that branch off a main tunnel. This is how those three countries conducted underground nuclear tests. Here are diagrams from “P-tunnel” at the Nevada Test Site (now known as the Nevada National Security site).[1]

Images are from: (left) Stuart Black, (Nevada Test site) National Emission Standards for Hazardous Air Pollutants Submittal—1993, DOE/NV/11432-139, pp. C-4; (right) Final Environmental Impact Statement for the Continued Operation of the Pantex Plant and Associated Storage of Nuclear Weapon Components EIS-0225.

P-tunnel, for example, had two entrances, as well as a ventilation drift and a machine shop, to service what is really a nuclear testing complex where multiple explosions could be conducted. Russia and China, too, conducted multiple tests using drifts off the same main tunnel. Instead of looking like Figure A, North Korea’s tunneling might look more like Figure B. (The diagrams are purely notional, based on P-tunnel’s geometry.)

Figures by Jeffrey Lewis.

If North Korea is not digging parallel tunnels, but is rather opening multiple entrances to the same main tunnel, then it is probably planning to conduct more than one additional test under the north mountain (West Portal), in addition to the two that have already been conducted there. The continuing excavation at the south mountain looks rather different in this light. If the South Portal is a tunnel complex, rather than simply two or three tunnels, North Korea might have plans to conduct more than two or three additional detonations in that area alone.

If his hypothesis proves correct, it would represent a change in how we view North Korea’s nuclear weapons testing program. Tom Schelling once told me that, for North Korea’s first few nuclear explosions, the term “test” was perhaps less accurate than “demonstration.” Schelling’s insight was that North Korea’s nuclear explosions were political events to demonstrate capabilities or perhaps Pyongyang’s resolve, not technically-driven events intended to result in successively better capabilities. Of course, that is just one wonk’s opinion and it is difficult to discern the balance between the need to demonstrate and technical goals in Pyongyang’s calculus. But it is clear that the pace of testing to date appears to have been constrained—either by political pressure, a lack of fissile material or both.

What would happen if North Korea were to acquire an ample supply of fissile material and political pressures were to subside? Pyongyang may soon have ample material if it does not already. And, although China may have pressured North Korea to refrain from nuclear tests in past years, its influence on Pyongyang seems to have waned. In November 2012, a Chinese delegation arrived in Pyongyang carrying “a letter from China’s new leader, Xi Jinping, which is said to have contained a simple message: Do not launch a ballistic missile.” KCNA released a picture of the Chinese handing over the letter. Then, the next day, KCNA announced the Unha launch.[2] Pyongyang also executed Beijing’s preferred interlocutor, Jang Song Thaek, for various crimes including selling “coal and other precious underground resources at random”—a veiled reference to his business dealings in China. As of right now, Beijing’s track record when it comes to pressuring Pyongyang isn’t so hot.

If the North does step up its nuclear testing program, in the near-term, Pyongyang will probably seek to continue developing smaller nuclear weapons that can arm the country’s ballistic missiles. They claim to have developed a miniaturized device, but we do not know how small these weapons are. Perhaps further miniaturization is needed to arm more than Scud or Nodong missiles; certainly this is the case for longer-range delivery systems.

Thermonuclear weapons—something the Chinese were able to achieve within six tests—are a worrisome possibility. But the good news is that Punggye-ri appears unsuitable for very large explosions above more than tens of kilotons in yield. This is because the mountains can only contain so much explosive power as well as the fact that a major cross-country railway runs relatively near to the test site, probably causing concern about the impact of large explosions on the rail tunnels. North Korea would probably need another test site to conduct very large nuclear tests in vertical boreholes—something we have yet to see.

The bad news is that our current North Korea policy of malign neglect may be more costly than we imagine. The current rate of one nuclear test every three or four years is unpleasant, but manageable. We muster outrage over these tests long enough to send Pyongyang a sternly worded letter and maybe sanction some iPads, before returning to years-long periods of ignoring them. After all, historically, North Korea—perhaps constrained by a small stockpile of fissile material—couldn’t stage nuclear temper tantrums all that often.

However, this may be about to change. North Korea may soon have access to regular amounts of fissile material if it doesn’t already and could be preparing Punggye-ri for a number of nuclear tests if the above analysis is correct. What if North Korea conducts a nuclear test, or even two, on an annual basis? Have we really considered the implications of an increase in the frequency of tests for our current policy of strategic patience? Have we prepared our allies for such a change? The answer is, of course, “No, we will cross that bridge when we come to it.”

Jeffrey Lewis is Director of the East Asia Nonproliferation Program at the James Martin Center for Nonproliferation Studies (CNS), Monterey Institute of International Studies. 

[1] Although it would be preferable from a radiation safety perspective, to drill all the drifts first, the United States and other countries have opened into test chambers to recover samples and refurbished used tunnels, including excavating new drifts following a first nuclear test. See, for example, discussions about radiation exposure among workers who “went in after these particular recovery operations took place when the tunnel was deemed to be radiologically safe so they can go back into digging new drifts, put in new experiments and fix up the tunnels in preparatory for the next event.” [sic] Available at:

[2] This episode is recounted in: Jeffrey Lewis, “Pit or Get Off the Shot: Is North Korea Going to Flight Test the Musudan or Not?” 38 North, May 16, 2013,

Stay informed about our latest
news, publications, & uploads:
I'm interested in...
38 North: News and Analysis on North Korea