David Albright offers an estimate of North Korea’s nuclear weapons that is not too far from mine. As he notes, any estimates have uncertainties upon uncertainties.
President Donald Trump’s bluster at North Korea has died down, but it could start up again at any time. Since North Korea’s nuclear weapons, or the threat of them, figure in the situation, we now need the best estimates possible.
Although Albright’s estimates are only slightly higher than mine, there are other considerations that I took into account implicitly. This post makes those considerations explicit.
Albright reasonably assumes
- 33 kilograms of separated plutonium.
- 175-645 kilograms of weapon-grade uranium, 175 kilograms in the case of one centrifuge plant and 645 kilograms in the case of two centrifuge plants.
Uncertainties that he cites are
- Whether weapon cores are plutonium, uranium, or composite.
- Whether any North Korean weapon design is small enough to fit on their rockets.
- How far North Korea has developed thermonuclear technology. He assesses that fully thermonuclear weapons are unlikely.
Most estimates, including Albright’s, make a simplifying assumption: number of nukes now = total amount of fissile material divided by fissile material per nuke. Albright uses 70% of the estimated amount of fissile material because of materials holdup and related items. That’s reasonable, and something that should be considered by others. Estimates that don’t consider this are likely to be high.
We also need to consider the history of the program. With little known about North Korea’s program, this can be done only in a general way. We know they started out with small amounts of fissile material and gradually increased it. The Manhattan Project faced similar restrictions: early samples of plutonium were so precious that floorboards were ripped up and extracted after a spill.
We do not know North Korea’s design for a nuclear weapon, except in the most general ways. It seems likely that they had some design information early in their program, possibly the design that A.Q. Khan sold. Another uncertainty is how much fissile material is used in their weapons. I won’t consider that in this post.
North Korean rhetoric, the apparent yields of their five nuclear tests so far, and their missile development suggest that they are working on a compact design. It is possible they are working with more than one design.
Let’s say they had a basic design in the early 2000s, when they were ramping up uranium enrichment. They had a small plutonium stash. Testing would be necessary to prove the design and their manufacturing techniques. Manufacturing can cause weapons failures in many ways – for example, initiator timing and imperfections in the conventional explosives or fissile parts. Some testing of these factors is possible without a full-up nuclear test, but ultimately all the parts must be tested together.
With an amount of fissile material sufficient for maybe a half-dozen bombs, North Korean planners had some decisions to make. Those planners would likely have included weapons scientists and high-ranking officials.
- Build one or two bombs from a known design that uses more fissile material?
- Build bombs before or after the known design is tested?
- Wait to build bombs until a reliable compact missile-mounted design is tested?
Even without nuclear weapons, North Korea has a significant deterrent in its conventional armaments ranged against Seoul. Having a nuclear bomb ready to use would confer prestige, but would add little to that deterrent. In any case, it would be foolish to advertise the existence of only one or two bombs. Even the United States bluffed in 1945 after using its two existing nuclear weapons; others were in preparation, but not ready to use.
A nuclear test proves that a nation can make nuclear bombs. Nuclear tests have a distinctive seismic signature, and that signature can be traced to North Korea. A test would be a declaration that North Korea has nuclear weapons. No need to say how many. But each test uses one bomb’s worth of fissile material.
For the North Korean weapons designers, a nuclear test would prove a design. For all those reasons, in 2006, North Korea tested. They also kept producing uranium-235. Plutonium production, which can be gauged by overhead photos of their reactor and processing plant, seems to have proceeded in fits and starts. They continued to increase their stock of fissile material.
If it were up to me, which it clearly wasn’t, and if one design were being modified into a smaller equivalent, I might build one or two once I felt the base design was reliable. Most analysts feel the 2006 test was relatively unsuccessful, so probably no weapons were built then. More likely, that decision would have come after the May 2009 test. The four years between the second and third (February 2013) tests suggest that decision-makers were relatively satisfied with the design, which would have allowed them to turn to developing their manufacturing capabilities and an improved design. As the amounts of fissile material increased, it may have made sense to build a few more weapons of the base design, although material would have been held back for anticipated weapons of the improved design..
The February 2013 test may have been simply to demonstrate that Kim Jong Un, newly in power, had control over the nuclear program and that the rest of the world should not mess with him. Or it could have been that the weapons designers felt they had the basics well in hand and were ready to move on to next steps, which might have been further reducing size or adding in a thermonuclear component, or both.
Fissile materials were still limited, and choices would have had to have been made among base design, testing, and the new design. Two more tests suggest that the modifications were being perfected. The recent reports of North Korea’s attempted sale of lithium-6 suggest that this isotope is part of those modifications. Or it may mean that North Korea separated that isotope and then decided not to use it in their weapons.
A sixth test would suggest that the second design is still being refined. Activity at the test site has not invariably led to another test. So we wait.
This analysis is very speculative. It is consistent with what little information we have about North Korea’s testing, but other scenarios also may be. However, it seems likely that not all North Korea’s usable fissile material is in the form of weapons. Thus, most estimates of numbers of North Korean weapons are likely to be high.
And an EMP attack? With developing designs and limited fissile material? Not hardly.
North Korea probably has fewer than a half-dozen nuclear weapons of an early design. They may be in the early stages of manufacturing a more advanced design, or they may need another test to prove that design.
Given the Trump administration’s focus on North Korea, we need the best estimates we can make.
Photo is of Kim Jong Un and what they claim to be one of their nuclear weapons.
Nuclear Diner 