Kangson Uranium Enrichment Plant Unzipped

There’s a nasty habit now of the worst case scenario regarding North Korea’s nuclear programme turning out to be correct, and we’ve just got ourselves another instalment.

Ankit Panda at The Diplomat has published an article, based on work at the Middlebury Institute for International Studies, revealing the probable location and history of North Korea’s original uranium enrichment plant known as Kangson. It is most likely located in the industrial city of Chollima. The geolocation is given as 38.957, 125.611.

For further particulars, if you have not done so already, I highly recommend reading the Panda article and listening to the ArmsControlWonk podcast on the matter. Be mindful of the probable location and the probable history. I suspected Chollima after I learnt of the name “Kangson” rather than “Kangsong” and I suspect that North Korea, unlike Iran, decided to go urban-industrial for its first clandestine enrichment plant. In a draft of this post here I had a sentence speculating to this effect, but I had deleted it prior to publication.

Let me first remark on the location, but we should note that the operating timeline suggests that the amount of fissile material and so nuclear weapons in the North Korean arsenal is at or perhaps even beyond the worst case scenario of 60 warheads (North Korea has likely a third if not more enrichment plants).

The Kangson enrichment plant is situated out in the open, as it were, in an urban-industrial setting. This is interesting with respect to another site that played a role in the diplomatic crisis over North Korea’s nuclear activities, namely Kumchang-ri. In the late 1990s a hole in the ground at Kumchang-ri became a central concern as some in the US intelligence community, focused upon the DIA, believed that hole in mountainous terrain was consistent with the early development stages of a clandestine nuclear reactor much like the reactor at Yongbyon and/or a plutonium reprocessing facility.

For a cool $300 million Pyongyang allowed the US to inspect the area twice (May 1999 and May 2000). The inspectors reached the conclusion that the site was not consistent with use for any nuclear related activity.

Perhaps the North Koreans learnt something from this little episode, however. That is, clandestine nuclear facilities should not be dug under the ground or into mountains. They should be built out in the open in an urban-industrial setting just like the enrichment facility at Yongbyon. David Sanger, whose sources came from those in the IC promoting Kumchang-ri as being nuclear related, writing for The New York Times stated at the time

Kumchangri was spotted easily from American satellites: Thousands of North Korean military workers swarm over it every day, excavating a huge area under a mountain. The shape of the excavation led experts to conclude tentatively that the North’s intention was to re-create the installations at Yongbyon, which are under round-the-clock inspection by the international agency

Consider the difference between Kangson and Iran’s enrichment facilities at Natanz (33.724598, 51.723965) and Fordow (34.885649, 50.99669). Both are located away from an urban-industrial centre and the cascade halls have been dug under the ground. That is why, in part, both times the Iranians were caught developing undeclared uranium enrichment facilities (the Iranian MEK opposition group did not discover Natanz). Kangson, however, clearly was a different story. Indeed, there are some North Korean peculiarities associated with Kangson that set it apart from its urban-industrial surrounds, and absent those peculiarities it is interesting to reflect whether its probable location would have been located even in the knowledge that it was somewhere to be found in Chollima.

The Kumchang-ri allegations were part of a political ploy by those opposed to the 1994 Agreed Framework, but those who promoted the Kumchang-ri story were too smart by half. They may well have had something to do with both the location and the set up of North Korea’s first uranium enrichment plant to the degree that they may have taught North Korea a lesson, one not heeded by the Iranians.

This suggests that a third, clandestine, enrichment facility, if not more, is or are located in urban-industrial settings in North Korea. Where it or they could be is a question of educated guesswork, but if all roads lead to Rome then you’d expect all roads to lead to UF6.

The history is extremely interesting. Both brilliant deduction and empirical analysis seems to suggest that Kangson began construction in 2002 and was completed in 2003. Note that’s not long after the final inspection at Kumchang-ri. Kangson is the first weapons related fissile production facility not to have been discovered prior to production commencing, and it is the first weapons related fissile material production facility to operate for a significant time with its location unknown to external actors. You might want to let that sink in. A stated in my previous post linked above, we are witness to the coming of the nightmare scenario opened up the Zippe machine.

The analysts at Middlebury, based on floorspace, estimate that Kangson might have an enrichment capacity 20% greater than the enrichment plant at Yongbyon. Based on the discovery of North Korea’s illicit centrifuge procurement activities following the outing of the AQ Khan network conservative opponents of the Agreed Framework in the Bush administration alleged that North Korea had a secret enrichment facility. We know that the US came to know of Kangson in 2010. That means that the conservative allegations were both based on an inference from that procurement activity and that this inference was correct.

But that inference was drawn for political as much as technical reasons, if not more for political reasons. Those drawing that inference did not know that North Korea had a clandestine enrichment facility as that inference was based on an extrapolation from procurement activities. In 2002 and 2003 there was a lot happening. This was the time of the lead up to and the conduct of the Iraq war, partly seen as an exemplary action enshrining the doctrine of preventive counter-proliferation. The attempts of the neoconservatives in the Bush administration to scuttle the Agreed Framework from day one of the Bush administration were well known. In the late 1990s the United States still maintained sanctions it was supposed to repeal under the Agreed Framework, meaning that North Korea’s declared nuclear activities were constrained but Pyongyang was still sanctioned.

In 2002 Iran went to Natanz and North Korea went to Chollima. Iran got busted, and North Korea didn’t (at first) but perhaps the Bush Doctrine had something to do with both.

One must be very careful of the dominant interpretation of Kangson, namely that for North Korea the enrichment plant at Yongbyon was always a bargaining chip to be negotiated away in a reciprocal process of denuclearisation of the Korean peninsula whilst Kangson was envisaged to continue operating clandestinely. The historian is always mindful of looking at the past with respect to contemporary concerns, processes, attitudes, and opinions. Following the latest leaked intelligence assessments, from the DIA for the most part, regarding North Korean deception and the Singapore Summit it may be too easy to figure this as a generalised North Korean MO that has obtained from day one.

It’s bit like an experimental physicist discovering something new and then the interpretation of experimentalists regarding its wider significance becomes dominant because the experimentalists discovered it. Those who report on this based on leaks from the IC, and those who have done brilliant sterling work on this have provided an interpretation that fits the generalised MO hypothesis above. However, one should not uncritically accept the interpretation of those who have publicly outed Kangson just because they outed it. Discovery and interpretation are not necessarily the same thing.

I’m not so sure about the generalised MO hypothesis. It could be right, but I don’t think North Korea’s nuclear programme is as discrete as most suppose. I tend to think that there’s quite a bit of symbiosis between North Korean nuclear activities on the ground and the modalities of its relations with the US and US wider actions than many are prepared to grant. I think the 2002-2003 starting point for Kangson might be reflective of this. On this view it is not a case of the neoconservatives being right because of a discrete generalised North Korean MO but rather it is a case of the neoconservatives being right because Kangson is something they helped to build through a type of warped symbiosis of action and reaction at work.

One final interesting point bearing in mind is highly speculative. In 2006 North Korea tested its first nuclear weapon. Initially it was reported to be a fizzle, although Chinese reports suggested the yield that was sought was 4KT rather than a nominal 20KT for a first generation device so making the test within the parameters of a successful one. In 2017 North Korea tested a thermonuclear weapon with a yield at least, if not more, of 250KT. The first test vented radionuclides and the analysis of the ratios of the fission products Xe-133 and Xe-135 demonstrated that the fissile material used was plutonium, it was widely reported at the time. The fissile core was not comprised of weapons grade uranium nor was it a composite pit. The 2017 test, though of much higher yield, did not vent radionuclides, and if it did their discovery was too far removed in time for nuclear forensics suggesting that containment of the explosion was successful. Some supposed that the first test had to have a plutonium core because North Korea was years away from producing weapons grade uranium.

From 2003 to 2006 Kangson was likely producing weapons grade uranium. North Korea was producing weapons grade uranium for years before the 2006 test rather than being years away in 2006 from producing weapons grade uranium. Perhaps North Korea deliberately allowed the venting of radionuclides in 2006 to fool us into thinking that its fissile material production programme was centred upon plutonium.