At the June 2018 Singapore Summit, North Korea agreed to the goal of “complete denuclearization” in exchange for “security guarantees” by the United States, including an end to enmity … The two sides seem to have settled on the phrase “complete denuclearization.” For the purposes of this analysis, this is taken to include the key nuclear weapon–related obligations agreed on in the 1992 Joint Declaration of South and North Korea on the Denuclearization of the Korean Peninsula, namely to “not test, manufacture, produce, receive, possess, store, deploy or use nuclear weapons” and that these commitments would be verified.

In March 2018, North Korea announced a moratorium on nuclear weapons and ballistic missile testing. …Moving forward, eliminating North Korea’s nuclear weapons program and related facilities will need a freeze on current weapon-related activities; an agreed baseline of current stockpiles of nuclear weapons, fissile materials, ballistic missiles, and key components; and verified reductions of these stockpiles and downsizing of North Korea’s weapons complex….We assume that a new framework agreement would contain provisions similar to those in some other arms-control agreements, under which the parties agree not to interfere with specified remote-monitoring techniques or use concealment measures intended to obstruct verification.

Since North Korea’s withdrawal from the Nuclear Non-Proliferation Treaty (NPT) in 2003, there have been essentially no international inspection efforts in North Korea. At the same time, North Korea has expanded the scale and complexity of its nuclear weapons program. On the basis of information available via open sources, it is not clear how many nuclear weapons North Korea possesses today, of what kind (including possibly thermonuclear weapons), and whether they use plutonium or highly enriched uranium (HEU) or both as fissile material. Nor is there reliable information on its ballistic missile capabilities. To establish a basis for moving forward, North Korea could add to its freeze on nuclear weapon and ballistic missile tests a freeze on fissile material production. This can be verified primarily through agreed-on nonintrusive provisions.

In the case of plutonium, satellite imagery can be sufficient to confirm the operational status of reactors in North Korea. Imagery can be used to observe heat signatures, vapor plumes, cooling water discharges, and other activities near the reactor . All these indicators would provide good evidence for a suspension of plutonium production at Yongbyon nuclear reactor in North Korea. Regional krypton-85 monitoring, ideally with a small number of detectors placed around the Yongbyon site, could confirm that remaining spent fuel is not reprocessed . There are also simple measures to permanently disable the Yongbyon reactor—for example, by blowing boron dust through the core’s cooling channels—but North Korea may not agree to such actions until the later stages of the denuclearization process.

The situation with regard to uranium enrichment is more difficult. It may be possible to confirm remotely the shutdown status of the Yongbyon enrichment plant and a possible second plant suspected to be at Kangson—for example, by monitoring vehicle traffic, including shipments of uranium hexafluoride (UF6) cylinders entering and leaving the sites, or by monitoring signatures related to electricity supply.

Rather than shut them down, North Korea may prefer to use its enrichment plants for production of low-enriched uranium for its experimental light-water reactor (30 MW-electric). If this or other civilian reactors are allowed to operate, then International Atomic Energy Agency (IAEA) safeguards could be applied to these plants as well as to the feed and product materials associated with them, as happens with civilian uranium enrichment plants in all non-nuclear weapon states and also in some nuclear weapon states. In this case, verification could include unattended measurement systems confirming the nonproduction of HEU, but it would also include onsite inspections. Even if North Korea ended all nuclear activities, IAEA safeguards would still be required to detect possible efforts at reconstitution of its nuclear weapons program.

One major concern is the existence of undeclared nuclear facilities, especially uranium enrichment plants beyond that at Yongbyon and suspected at Kangson. This is a proliferation concern in all states and not limited to North Korea, however…

With a freeze as a starting point, declarations of current fissile material and nuclear warhead inventories would be important for measuring progress toward denuclearization. These initial declarations could be relatively simple. Ideally, as a transparency measure, they could be made public. In the case of nuclear warheads, a declaration could include the total number of warheads in North Korea’s stockpile, perhaps listed by type, and the number of additional warhead components in storage; in the case of fissile material, a declaration could include acquisitions, losses, and removals, including the aggregate amount of material consumed in tests, and the current inventory of plutonium and highly enriched uranium, ideally also specifying the respective plutonium-239 and uranium-235 contents. More detailed declarations could follow at a later stage of the process.

There is a precedent for fissile material declarations. In May 2008, North Korea declared its plutonium inventory, often reported as 37 kg and backed up by 18,000 pages of operating records. At the time, the United States estimated that North Korea had produced a total of 40 to 50 kg of plutonium, raising concerns that the declaration may be incomplete. U.S. negotiators requested access to the Yongbyon reactor to confirm total plutonium production through use of nuclear archaeological techniques, in which the isotopic ratios of trace impurities in graphite samples are analyzed. At that time, North Korea refused.  Nuclear archaeology techniques for graphite-moderated reactors are now well established and would be sufficient to narrow down the uncertainty in plutonium production to a few kilograms, possibly to less than one weapon-equivalent. North Korea may or may not agree to these procedures early on in the denuclearization process, but every effort must be made to preserve the reactor core and relevant operating records so that such an analysis can be conducted when circumstances permit.

Reconstructing uranium enrichment activities is more challenging. Perhaps the best option would be to reconstruct North Korea’s history of uranium supply and use. Such an effort would assess uranium production at North Korean mines, uranium purification, UF6 production, and enrichment. This would involve auditing the records for internal consistency. Reports of North Korean uranium ore grade suggest that it takes 300 to 400 tons of ore to extract 1 ton of uranium. This means that up to 2000 tons of ore are required to make 25 kg of weapon-grade HEU or 5 kg of weapon-grade plutonium, the typical amounts used in a nuclear weapon. The review of records from the different plants could be complemented with forensic analysis of tailings at the mines and depleted uranium in cylinders at known enrichment plants. It also may be possible to examine North Korea’s centrifuge-plant equipment and reconstruct the amount of uranium processed in these plants and respective HEU output.

It will take years to conclude that undeclared stockpiles of materials and warheads do not exist, even if North Korea fully cooperates...

For safety reasons, as former Los Alamos National Laboratory Director Siegfried Hecker and colleagues recently observed, “shipping the North’s nuclear weapons out of the country is naïve and dangerous. The weapons must be disassembled by the people who assembled them.”…A third option would be for North Korea to gradually reduce the size of its weapons complex without revealing where exactly nuclear weapons and long-range ballistic missiles remain. An estimate in 2014 suggested about 90 nuclear weapon and missile sites of potential interest…

Excerpts from Alexander Glaser and Zia Mian, Denuclearizing North Korea: A verified, phased approach, Science, Sept. 7, 2018