North Korea Tests an H-bomb

Aaron Stein 0:02

As a personal trainer would say, get comfortable being uncomfortable.

Jeffrey Lewis 0:05

Just as long as we don't have to feel the thermonuclear burn. You are listening to the Arms Control Wonk podcast, the leading podcast on arms control, disarmament, and nonproliferation. I'm Jeffrey Lewis, the founding publisher of the Arms Control Wonk blog and a scholar at the Middlebury Institute of International Studies at Monterey.

Aaron 0:27

And I'm Aaron Stein, a senior fellow at the Atlantic Council's Rapid Career Center for the Middle East. Hey, Aaron, what do you want to talk about today? I was thinking we could talk about your sort of quasi-darkness on Twitter and social media over the weekend. Where the hell were you?

Jeffrey 0:44

I tried to go to the beach.

Aaron 0:46

Don't you always go to the beach? Whenever I see your pictures pop up on social media, you're on that beach in Carmel.

Jeffrey 0:51

So this was different because, one, it was 100 degrees, which it never is here, and no one has any air conditioning, so we all headed to the beach. And by the way, poor John Wolfstall was trying to troll me about this. I know it's like, oh, you don't have air conditioning. And it's like, so for two days, it's roughly as hot as it is in DC all summer.

And I got to spend it laying on the world's most beautiful beach. But I turned my phone off. And what happens when I turn my phone back on and try to pick up a little, little Chinese food at Tommy's walk? Pictures of thermonuclear weapons.

Aaron 1:23

That's right. You know, I was in Turkey for the past 10 days for the annual pilgrimage to the in-laws' house. And, you know, for the exorbitant rates of data abroad, I came back within Wi-Fi range to find my Twitter feed full of pictures of the silver peanut. The peanut of death. Oh, no, we're going to have to do an emergency pod.

But luckily I'm back stateside, so we could do it here.

Other voice 1:46

So friend of the pod, Eric Gomez, had a lovely story from the same where he and his girlfriend got engaged about 10 minutes before the North Koreans dropped their announcement. Or I guess before everyone started dropping their announcements about the North Koreans. And it made it apparently very difficult to take pictures afterwards because his phone was just getting so many notifications that it was difficult to actually wield it as a camera.

Jeffrey 2:14

Now, as someone who had his wedding anniversary yesterday and also had it disrupted, he's got to learn to turn the phone off. It's a prerequisite. Or airplane mode. There we go. The camera still works in airplane mode. Oh, yeah. Smart, smart, smart. Actually, Eric Gomez drove down to see me and joked on Twitter that he had completed the holy trinity of meeting all three of us in person.

This is correct. What was funny was the city of Monterey retweeted his tweet because he was like, I made the pilgrimage to see arms control long. And they retweeted it like I'm a legitimate tourist attraction.

Aaron 2:53

That is hilarious. There's not much else going on in Monterey. It's kind of a sleepy place. I guess not. But, you know.

Jeffrey 3:02

Yeah, so pictures, and then it went kaboom. So I actually loved that because I, you know, as soon as I saw the pictures, I thought to myself, like, okay, so we're going to have this period where people pretend that the peanut of death is filled with styrofoam peanuts.

Aaron 3:23

Yeah, or real peanuts, whatever.

Jeffrey 3:24

Yeah, and it's not real, and it can't be real, and so they unleashed one. It's like, ah, it's pretty big.

Aaron 3:33

So as soon as it goes, I was in Wi-Fi range, and it went kablooey, and I saw the initial estimates. I said, holy fecal material. Like, this is an order of magnitude larger than the other ones.

Jeffrey 3:44

Yeah, I was actually trying to rush my kids to bed, and so the first thing I did to calculate yield was I grabbed the wrong... I grabbed the wrong equation, so I used the Nevada test site equation. I used rock poor propagation as opposed to rock good propagation, which added a couple hundred kilotons. I'm still higher, I think, than other people, which we've got to talk about.

Aaron 4:15

Yeah, well, let's back up. We should talk about how big it was. Let's talk about how big it was and then why that matters. I think you wrote about it in your foreign policy piece today.

Jeffrey 4:24

Yeah, so there are really two questions. There's how big is the underlying explosion, but then there's how big is the seismic wave that is measured.

Aaron 4:33

Right.

Jeffrey 4:34

And, you know, there's actually right now quite a bit of range. For the seismic wave, which is, it's not the Richter scale. They don't use the Richter scale anymore. I guess maybe some people do, but it's like the Richter scale, just like for Earth, and it's the scale now people use also for earthquakes. So, you know, if you know an earthquake is a 6.3, USGS measured this thing as a 6.3.

Aaron 4:59

Mm-hmm.

Jeffrey 5:00

I think the Chinese got the same number, too. Chinese got the same number. I guess it's the moment magnitude scale. The Chinese got the same number. The Japanese came in at 6.1. And then way at the bottom, we had Norsar at 5.8 and the CTBTO at 5.8. But the CTBTO has now gone all the way up to 6.

Aaron 5:20

Okay. I mean, if you want to just map it out there, somewhere between the 5.8 and 6.2.

Jeffrey 5:27

Well, the Russians came in at 6.4, I think.

Aaron 5:29

Oh, I actually didn't know the Russians had released their test data.

Jeffrey 5:34

Yeah, it's very tempting to just dismiss the Russians as always an outlier on this stuff.

Aaron 5:41

Yeah, we keep wanting to do a pod about how the Russians are an outlier on this stuff, but world events intervene.

Jeffrey 5:47

I mean, it's hard, but I think if you... If you average all of those together, 1, 2, 3, 4, 5, 6, right, you get 6.2 or something.

Aaron 5:59

Yeah, and then I think for listeners, you know, the difference is that, you know, by each tenth of a point, this logarithmic... Yeah, it's the problem.

Jeffrey 6:10

Yeah, it's a problem.

Aaron 6:12

So the second number after the point actually matters a lot because you have an order of magnitude larger per, I guess, per decimal point.

Jeffrey 6:23

And, you know, it can be really hard. I think Stephen Gibbons, who is, along with Paul Richards, one of our favorite seismologists, and seismology is really a dismal field. I can't... It is so useful and so valuable, and yet I can't bear to it. He actually posted, there's a book on seismic monitoring of nuclear tests by Alan Douglas.

And there's this passage where a seismologist sort of talks about how you pick the first motion, right, from the signal. Like you say, this is where the signal starts. It's not just noise.

Aaron 7:07

And it's not easy to do. I can't claim to have any expertise on this, but I can imagine it's all that easy. Although it seems easy because we just have pictures of the scale moving back and forth.

Jeffrey 7:16

Yeah, but you have to figure out what's that actual first vibration.

Aaron 7:22

Right, and I think for the second part of this, when you estimate yield, it's not just about the size of the seismic wave, but also assumptions about the test site.

Jeffrey 7:30

But the point is, it's not weird for the CTBT to come in at 5.8 and then go up to 5.9 and then ultimately to 5.6. The numbers jump around a little bit. I think the first USGS estimate was 5.2, and then they jumped it all the way up to 6.3. Indeed, it was lower, and they moved it up to it, and it matched the Chinese one.

So, I mean, there's kind of an underlying question of what number you want to use. You can either use 6 from the CTBT. Well, I don't know. The ones I'm okay with are 6.1 if you're just going to average everything, and I'm still okay with USGS and the Chinese both coming in at 6.3. Gibbons still feels like 5.8 is a good number.

Wow. You know, I mean, they're competent, so, you know, like, it could be that other people are mismeasuring it. Like, you know, it's not. But I just, I feel like it's, you know, somewhere in the sixes. Yeah.

Aaron 8:26

But, you know, like, that's a lot of uncertainty. It is uncertainty, and so it could account for differing estimates in yield.

Jeffrey 8:36

Yeah, I think we'll talk about it. Ankit Panda has given us what the classified U.S. estimate of the yield is, and I think you can reverse engineer the parameters that they're using. Right, we will. But the problem is you have to figure out, once you have the seismic signal, then you have to estimate the yield of the explosion that produced that seismic signal.

I'm waving my finger up and down like a little seismic scale, which I'm sure is really useful for all the people listening to the pod. We can visualize it. Yeah.

Aaron 9:05

Yeah, your finger is moving up and down in front of your window that overlooks the ocean.

Jeffrey 9:10

Yeah, although, oh no, I can see the ocean. We're not fogged in today. I didn't even look. Yeah, so that gets really tricky because the, you know, there is this very indirect relationship between the, well, not indirect, what's the right word? Mediated relationship between the size of the explosion and the seismic signal it produces. So as I said earlier, I had used the, it depends what kind of geology the test site is.

In this case, North Korea's hard rock. It depends whether the waves propagate well or poorly. And as I said at the early, at the onset, I took the propagate poorly and I should have taken propagate well. And then it also depends... Well, there's a kind of debate about this, but how deeply it's buried. Right. There's a school of thought that says these North Korean tests are bigger than they seem because they're under so much rock that the rock is suppressing the signature, which I don't actually fully understand the physical mechanism.

I think it's that it's reflecting energy, but I'm not entirely sure. We should actually... I don't know. Should we have a seismologist on to go over all this stuff? We could ask all our dumb seismology questions.

Aaron 10:30

That'd be good. It's been a while since we've done bringing a guest from the outside where you and I talk to them.

Jeffrey 10:35

Yeah, we should totally do that. And then it also kind of depends where it happened because I think there is some chance that they did this inside one of the cavities created by a prior nuclear test. And that might dampen the signal just a little bit because it means it wouldn't be quite as well coupled to the surrounding environment.

Aaron 10:55

I think... What's interesting is part of your analysis and sort of the team you've built over there in Monterey is that you guys have modeled the test site. And I think, was it a year ago or something along those lines where you did a 3D model where you think this current explosion was and therefore it adds a little meat to your own sort of assumptions about the size of the yield?

Jeffrey 11:16

Yeah, so we took this space-based topography data collected by a Japanese sensor on an American satellite and we built a 3D model of the test site. Where can they find that?

Aaron 11:28

Oh, on the NTI.

Jeffrey 11:29

It's in the show notes downstairs. In the show notes on the Slack channel and on the NTI website. Yeah. But we, then we, actually using Steve Gibbons' work, we kind of, I don't know if you'd say double checked it, but he had done it. Using just the seismic signatures. But we then asked the question, you know, is hit, do his estimates fit inside our mountain without, you know venting because they would have to be buried a certain amount and we, they have to move together.

And so we actually found that, you know, his, his estimates were very consistent with our model of the physicality of the test site. But what we ended up noticing were two things. One is that, well, we noticed more than two things, but the two things that are relevant here are, one, those tests are really deeply buried under a lot of rock.

Yeah. Like a lot. And so the equations that we traditionally use, often they don't, they just assume that the test is buried to the appropriate level. You know, like, that they don't overbury it. But because these mountains just sort of soar up, you know, these tests are all really over-contained. And so that might be depressing them. But then the other thing is we noticed you could do a couple hundred kiloton tests in one of the tunnels.

So, you know, we were like, hey, this thing is sized for a thermonuclear weapon. Like, pay attention.

Aaron 13:01

Then it went kaboom. Then it went kaboom, yeah. And so... Based upon what we know, I mean, I think they're, you can, I mean, I think they've been out there too, is that, you know, the range is somewhere in the low hundreds of kilotons, and I think up into the, I mean, close to a megaton.

Jeffrey 13:18

Yeah, I think the best way to say it right now is to say a few hundred kilotons. Yeah. You know, and if you're at all familiar with this, you know, the difference between 100 and 300 on a logarithmic scale, given the uncertainties involved, is like... Like, I think somebody at one point said to me about an earlier test, like, well, James Acton thinks it's 10 and you think it's 25.

I was like, yeah, we agree. Yeah, they're pretty close. It's like, that's all measurement error. Like, if it turns out to be 10, I won't be embarrassed. And if it turns out to be 25, I won't be proud. Like, that's all whatever. So I think a few hundred kilotons is the best way to say it, because you're really talking about orders of magnitude.

It's probably not a megaton. But this gets to this really interesting question. According to Ankit, the U.S. estimate is 140 kilotons. That's what I've seen. What that, to me, looks like is two things. Is a magnitude of the body wave at about 6.1. It's actually 6.06. Using the simple equation for hard rock in a region of good propagation.

See, I just did it again. Yeah.

Aaron 14:41

Well, that's how you got the first one in the Megaton.

Jeffrey 14:43

I know. Twitter doesn't lie. Twitter doesn't lie. Twitter... Yeah, it doesn't. Yeah, that's right. It's easy to do. And that, a 6.06, so, you know, sort of maybe they're averaging them, or maybe AFTAC has a slightly different number than USGS, or maybe they're just going off the CTBTO number or whatever. Like, that gets you to 140 kill times.

The problem with that, if I'm right about the way they're doing it, is they're not taking into account how deeply buried that sucker is. Right. So you're saying that it's probably... So there's a different equation that I use. Right. I use... There's an equation which a couple of... I believe they are... At least one of them is a Chinese seismologist who works in the U.S.

I don't think he's Chinese-American. I think he's Chinese-Chinese, but I actually don't know that for sure. These two seismologists, they... Working from someone else's publication, there is a calibration of the test site using chemical explosions. And they added the depth correction error, or the term for correcting for depth. And so I tend to use their equation. Okay.

Because at least nominally it's calibrated, and it takes into account this really weird phenomenon, which is these things are super deeply buried, and that gives you a much higher number, right? So that same 6.06 gets you to, you know, like 320 kilotons. That same 6.06 gets you to almost 300 kilotons. If you assume that it's 6.3 like USGS, that gets you up to like 500 kilotons.

I think it's probably not 500, but you can see how much of a difference that makes.

Aaron 16:55

Yeah, you can see where ranges come in. So I like a few hundred kilotons. Right. And I think what's interesting is that in your 3D model, What you guys inferred was that the site was built to contain a thermonuclear explosion.

Jeffrey 17:17

Right. I mean, we were really struck that when I look at these things in three-dimensional space, we can see that the very first test was done at a strange part of the test site, kind of not near the main support area. Like it was very much a kind of one-off, maybe we'll never do this again sort of event.

Mm-hmm. Which, you know, is actually not unusual. I mean, that's certainly what Trinity was like. You know, it's like, oh, let's find some desert, and then it's like, oh, no, we need, like, a proper test site. But tests two through five, and now six... Certainly two through five were all done in the same mountain off a single entrance tunnel which is a big fight I had at with 38 North for like years.

They kept asserting that the North Koreans were digging parallel tunnels into the mountain. And I kept saying like, no, it's like a U S or a Chinese test site where there's a single main tunnel in their branches. And they finally let me publish a, a piece under the insulting title and alternate view. Like, like I'm the damn hippie, but you know, whatever.

But the thing that really struck us when we laid it out in three-dimensional space was that, one, there was a lot of room for more tests. They had selected a test site that was not a one-off. This was a place you could do a lot of tests. And yeah, the tests could be much, much bigger.

Aaron 18:34

So what do you account then for the second seismic event that came about the potential for the cave-in?

Jeffrey 18:42

Yeah, I mean, that's what the USGS says it was. They think it was a cave-in associated with the test. And Cheryl Rofer thinks the same thing. And, you know, it's always good to agree with Cheryl because she's really smart.

Aaron 18:57

She's a friend of the pod and is very smart.

Jeffrey 19:00

I was saying, like, Paul Richards came in with a 250 kiloton estimate with error of a factor of two. And somebody asked me what I thought. It's like, I always like to be within Paul's margin of error. Like, I don't want to get crosswise with Paul. And I particularly don't want to get too crosswise with Cheryl, although it's always fun to argue with her.

But they're pretty smart. So I, you know, I think that there is probably a cave-in and a subsidence crater. So we're going to actually try to get, we'll get optical data, but we're also going to try to get synthetic aperture radar data and then see if we can spot the crater.

Aaron 19:36

Yeah, so for people who don't follow the stuff in as much detail, what does the crater tell you? And then what do you think that means for the U.S. to, you know, secondary detection?

Jeffrey 19:49

Right, so when the explosion goes, it creates a cavity. And if the cavity is small enough, that cavity will sustain itself, right? It'll just be a cavity. In the mountain. In the mountain, right? Sometimes those cavities are so large... That they don't... They can't hold themselves up, and so they collapse. And so you get a little crater on the surface where the rock has sort of fallen down into the hole.

It's like a sinkhole. Yeah, it's like a sinkhole. I mean... And you can actually see that really clearly. There's one at... There's a really nice example of one at Nevaeh Zemlya in Russia. And, you know, if we get lucky... Really, really lucky that subsidence crater will be directly above where the explosion was, and we'll be able to use it to estimate the location of this explosion. But also, because the relative locations are much more accurate than the absolute locations, it'll help us place the other explosions as well. Right, exactly. and then, but we, we got like, we got to get that data and we got to do that work.

Aaron 21:01

Yeah. So let's talk about what it is. I mean, the peanut suggests it's a two-stage thermonuclear weapon. Let's describe the peanut, too, in case people haven't seen it. Okay, so right.

Jeffrey 21:12

So before the test, the North Greens released pictures of what they said was a two-stage thermonuclear weapon. And Kim Jong-un was posing with the weapon and members of the Nuclear Weapons Institute, which, by the way, we think we found. It's not necessarily where the picture was taken, but we do think we know where the Nuclear Weapons Institute is, which is kind of cool, but for another pot.

And it looks like a big peanut. Right? Yeah, it's like an hourglass. And that shape is because it has two stages. Each nut is a stage. The bigger nut is the primary, right? That's the fission device. And then the smaller nut is probably a spherical secondary. And the purpose of a stage thermonuclear weapon is to use the radiation from the primary to implode the secondary.

So you get some fusion. And then you get... And a bigger boom. And then you get a whole bunch of secondary fission, which is actually quite a significant contributor to yield. So, you know... That's what they look like. You know, we actually don't see them. I mean, this is, you know, like one of the hard problems is North Koreans release really explicit images.

And even people who know what U.S. ones look like often don't have a, you know, a list of comparative objects in their head. Right. Because rarely do almost never do countries release this kind of information. So, like, I don't know what a Soviet thermonuclear weapon looks like. I don't know what a Chinese thermonuclear weapon looks like. They just don't release those pictures.

But, you know, I remember when you remember the whole Wen Ho Lee nonsense.

Aaron 22:53

No, because I'm not a DPRK expert.

Jeffrey 22:54

Well, he's a Chinese-American scientist at Los Alamos who got arrested for allegedly giving secrets to the Chinese and or Taiwanese.

Aaron 23:03

Oh, yes, I do actually remember this, yes.

Jeffrey 23:04

And then while he did plead guilty to mishandling classified data, I think he left with an apology from the judge because it was clear there was a fair amount of railroading based on what I think in retrospect will turn out to be racism. But whatever you think of that case, when it was revealed that the W88 was shaped like a peanut, and that the primary was larger than the secondary, and that the, the secondary was went forward in the warhead because it was narrower, like that kind of information, which is on display in this image, caused people to freak out that it was suddenly in the public domain.

Aaron 23:42

Yeah, and then when this one went off, you know, and then they showed us the Peanut of Death, the immediate comparison that came out, at least on feeds that I follow, was to the WADA.

Jeffrey 23:49

Yeah, and it's... Just the shape. Yeah, and it's... I mean, it's bigger than that, but it's definitely... You know, it's the same... I mean, you know, we knew all these things, right? Spherical secondaries, and you've got this kind of tapered radiation case that's form-fitted to the two circles. W88 has an oblate primary, which I don't think this one does, but you know.

So it's not the W88 it's not as efficient not as cool not as fancy but it's I mean they definitely showed what was plausibly a stage thermonuclear weapon and by the way I think we now know we wondered about this with the disco ball of death the plug the big the big object sort of sticking into the disco ball that's a black that's an external neutron generator I think

Aaron 24:42

Yeah, yeah, yeah. I agree. But the U.S. intelligence estimate, does this come from UNKIT or is this out there already? I can't remember, but they call it an advanced nuclear design? It's from everyone.

Jeffrey 24:55

Everyone. Pretty much all of the officials who gave interviews to reporters used that very specific phrase, advanced nuclear design. And UNKIT is the one who's gotten them to answer it. It's because their estimate is on the low side. Not necessarily wrong, but it's one of the lower ones. It's 140 kilotons. At that level, you can't exclude that it was just boosted.

Aaron 25:18

Right, and maybe Cheryl makes a return to the podcast this time because she has some speculation that perhaps what they showed was just a mock-up. The peanut was a mock-up, and it's not exactly what they tested.

Jeffrey 25:31

Yeah, which, you know, you can't exclude. I mean, there's the methodologist in me, the epistemologist, thinks that they're doing it the right way. They're saying it's advanced nuclear design, and at 120 kilotons, it could be the thing they showed, but it could be just a boosted weapon. A boosted weapon? Sorry, yes, a nuclear weapon using deuterium tritium gas to increase the efficiency of the explosion and therefore increase the yield.

But not two-stage. But not two-stage. Not that whole radiation implosion setup I just described. I... Think it probably is the two-stage object, and we can discuss why, but you don't actually know because they just show you a picture of the device, and then you get the ground motion. Like, you don't, there's, there's not a verifiable chain of custody in which the device, you know, is under camera, moved into the test site. And you know, well, maybe we should dare Kim Jong-un, like he seems to take this stuff personally, so show us Kim. Yeah, Jesus, you know, I mean, I'm, I'm worried about Juche Bird.

Aaron 26:34

Yeah, show us. It'd be pretty explicit what people call you out.

Jeffrey 26:39

Yeah, when they put the thermonuclear weapon on the ICBM and fire it out to sea, I'd really rather they not do that.

Aaron 26:44

Yeah, but we'll get to that.

Jeffrey 26:46

Anyways, yeah, go ahead. I'd prefer we just take them at their word now. I have a nightmare scenario, but anyways, yeah, go ahead. So, you know, the reason I believe it's actually the stage... I believe it is likely to be the stage thermonuclear weapon for three reasons. None of these are conclusive. One is you'd still have to use a fair amount of material to boost that device up to 140 kilotons.

And I just don't think the North Koreans want to use that much material. But they could. So that's not impossible. It just seems inconsistent with past North Korean behavior to me. The second thing is I'm still tempted by the slightly higher yield numbers because I want to correct for depth.

Aaron 27:43

And back to the original conversation we had.

Jeffrey 27:45

Yeah. And I mean, and I think this, this, it just gets less plausible if you get up over 140 kilotons, you know, when you get up to two or 300 kilotons, you, it still could be boosted. I'm not saying it's not, but it just, you know, again, you're talking more material. So I, you know, and I feel like, you know, it's a little bit like, you know, in some, some game where you have to like run the table, like for, For, I think, this to just be boosted, the yield's got to be on the low end, and the North Koreans have to have deviated from their past practice of being conservative in the use of fissile material, you know, and just, I don't know, seems less likely to me.

Doesn't seem impossible. And then there's the technical piece, which is, you know, the reason the Chinese had trouble going to a, I'm still not sure I want to call this thing miniaturized, but Missile-deliverable thermonuclear device, certainly something this small, is they had trouble with gas boosting. That's where they struggled. You can see in the atmospheric tests they did, they were instrumenting them for, I guess, neutronics would be the term.

They wanted to see the neutron flux. So I actually think that the argument that the North Koreans can't do this technically... As an empirical matter, like the thing that they would have struggled with, you know, they're at least visually conveying that they've solved that problem. You know, whether they really have or not, obviously, it's hard to say.

But I just, you know, I think the Chinese had problems with, you know, not much test data and not great supercomputers. And I think that just that's a little bit easier today than it was 50 years ago.

Aaron 29:36

Yeah, but I think it's also important, just as you said up top, is we're going off the pictures. It's really impossible to know because we aren't crawling down into that test site.

Jeffrey 29:49

Yeah, that's right. I mean, there's a fine line between what I would say is denialism, which I find a lot of people are slipping into, But also then, you know, the concern on my end, because I don't think I suffered denialism, but, you know, like, you also still want to be skeptical of things that you see, right?

You don't want to lose your ability to look at things skeptically. I think Cheryl's, like, right in the sweet spot of that. So, you know, I think, you know... We should have her on the pod and let her make the argument rather than put words in her damn mouth. But, you know, I think it's a reasonable question.

And I understand why the intelligence community is just saying advanced nuclear device and staying agnostic about just precisely what it is. I would just set that expectation to say it's not crazy to imagine that it's precisely what the North Koreans say it is.

Aaron 30:37

Yeah, I mean, I think we've taken a habit on this pod over time as we've tracked each development, this just being the latest, is that what we can infer from actions and then their statements usually has a whiff of truth to it, even though they're grandiose and their propaganda is so hilarious at most times because of how over-the-top it is.

But it's not beyond the realm of possibility. In fact, this is basically the first test of a two-stage device. I think it's something that we've talked about before that's also important just to keep in the back of your mind. It's like the conceptions of nuclear weapons development. This is their sixth test. And because they didn't go for the U.S.-Manhattan-style approach.

They just tried to skip ahead a few steps.

Jeffrey 31:34

Well, yeah. I mean, it's not... It's not crazy. If you put the Chinese nuclear program side by side with the North Korean program, it's not insane, right? Because, I mean, this will be hard for people to hold in their head, but the first Chinese nuclear test was basically a copy of Fat Man. And figure the North Koreans say, like, no, we know that's going to work.

We're not going to waste the material. We'll skip that. Well, the second Chinese nuclear test is an airdropped weapon that's small enough to be carried by, well, for them it was an aircraft, but, you know, was also small enough for a missile. Well, that's kind of what the first North Korea test is, right? Except for the first North Korean test doesn't work because, as the intelligence community had said, if you skip that first step, you may discover some things you don't like about your design, right?

So the Chinese go from one to two, no problem. The North Koreans skip to two, and it's like, oh, yeah, that doesn't work right, right? So then the Chinese case is weird because for number three... They shift over to a type of thermonuclear device called a layer cake. Yeah. And the North Koreans are just, they're not going to do that.

That's not interesting to them, right? So they're going to skip that. And for the fourth test, they're dealing with this annoying American, like, well, I probably can't survive a missile. So they put it on a missile, right? I think the North Koreans kind of go down the slightly different route during this period with the second and third tests, where I think for the second test, they're probably going back and fixing the problems with Kim Won.

Aaron 33:31

Mm-hmm.

Jeffrey 33:32

And then maybe with, so that's maybe Kim two figures out what went wrong with Kim one and maybe Kim three although it's still not great. Then Kim three comes in and Kim three you know, that, that device maybe makes a slight weight reduction. Cause Kim one, I think was according to effector would have been like a thousand kilograms and the next device was supposed to be 500.

So I figured they retest the thousand kilogram device. They get it right. And then with Kim three, they go down to 500 kilograms. Okay, so then the Chinese on their end are like all about thermonuclear weapons. So 5, 6, and 7, if memory serves me, it's like a thermonuclear principles test, 6 is the real thermonuclear test, and then 7 is another thermonuclear test that doesn't work.

That corresponds with the test the North Koreans just did, which is Kim 6.

Other voice 34:22

Mm-hmm. Mm-hmm.

Jeffrey 34:25

I think Kim 4 and 5, you know, 4 is, they say that's a hydrogen bomb test, so it's either a Kim 6 that didn't work, Or it was like the fifth Chinese test, which was a principles test. And, you know, they were really clear that KIM-5 was their standardized device to go on their warheads. So, you know, like if you look at it, if you look at the first six Chinese tests and the first six North Korean tests, they kind of have analogs.

It's just that the North Koreans moved quicker to looking at composite pits and boosting. Maybe this is going to have to be cut out because this is too rambly. But I just think when you set them side by side, the North Koreans are doing the same thing as the Chinese did in their first... Maybe the things the Chinese did in their first, say, 12 tests.

Because they're skipping some steps and just taking on a little bit more risk. And in some cases, skipping things that are irrelevant.

Aaron 35:28

Yeah, and that... Maybe Scott should cut all that out. I'll just make a chart. No, no. I mean, one thing that gives me... that it lends credence to the North Korean claims is that side-by-side they're demonstrating missile capabilities. So it seems like what they are trying to design, at least what they are testing, would be intended for the missiles that they are literally testing side-by-side or within weeks or months of the nuclear tests.

And they're coming so fast now.

Jeffrey 35:57

Well, that's the other reason, you know, I mean, maybe I'm getting cranky in my old age. But, you know, I felt like we were really making this case that a thermonuclear weapons test was coming. And I'm vaguely annoyed that all the press coverage is no one saw a thermonuclear weapons test coming. So it's like, what? You know, I had this big...

It was on Amanpour in CNN, and I was like, they're going to test an ICBM and a thermonuclear weapon soon. That's going to happen this year. And then they tested an ICBM and a thermonuclear weapon. And, you know, the North Koreans, when they tested that ICBM, said it was designed to carry a large-sized, heavy nuclear weapon. That was a new phrase, right?

That was a phrase that said, not the thing we tested last September.

Aaron 36:51

You know, and I think it's important to point out that even if, you know, what they're showing us in the peanut of death is not what they tested, and it's, you know, closer to Cheryl's point, is that this is a mature weapons program. And that, you know, the peanut of death will be coming. Do you know what I mean?

If that makes sense.

Jeffrey 37:10

Well, right. Because people are acting like boosting is like, oh, it's only boosted. Like, no, boosting is a fancy freaking thing to do. Oh, absolutely. The Chinese figured out thermonuclear weapons before they figured out boosting. They did it in the opposite order, which is they were just taking their unboosted fission devices and using them to set off their thermonuclear devices, which is why they were too big for missiles.

That's why they had so much problems. Then they turned around and went to boosting and struggled with that. So, you know, if it's true that this is just a boosted device, like that should still scare the pants off you because it means the North Koreans have done the harder of the two things.

Aaron 37:56

Right. And I was trying to explain this to my wife who works in the private sector in marketing and is not an avid listener of our show. But, you know, she's like, so why do you care? Why do you want this stuff? And, you know, it's all about size and weight, right? You know, you want to be able to pull these things down.

And get less weight, use less fissile material, and make your devices more efficient because, you know, you can begin to do more fancy stuff with your missiles.

Jeffrey 38:21

Right, and, you know, I think that the one thing I do believe is that the North Koreans are still a little bit fissile material constrained. I mean, I don't fully understand why they haven't built out their plutonium production capability, but, yeah, I think that's still an issue for them.

Aaron 38:36

So, again, somebody who doesn't... That's something that keeps bumping out in my head, too, is... What's the upper limit on their plutonium production capacity?

Jeffrey 38:45

Well, you know, what one person said, which got my attention, was, you know, the reactor they're building next to Yongbyon, which they say is an experimental light water reactor. Like, do we actually know that? Like, they say it's an experimental light water reactor and not a copy of Yongbyon. But then again, wouldn't you say that? No, this is totally proliferation resistant.

No, no, no, this is fine. You don't need to attack this, you know? Like, maybe it's not. Like, maybe it's another gas graphite. When they just put a dome on it because it, you know, looked friendlier. I mean, I don't know. They were building one in Syria. I mean, they seem to be able to build these things.

I mean, but anyway, I mean... Yeah, right. Why is there no other nuclear reactor in North Korea? It, to me, is a little bit of a mystery. And you don't want to go... You don't want to get sucked down to the conspiracy route where you're like, there must be 12. Like, you know, maybe there aren't, but like...

It's freaking weird.

Aaron 39:37

I guess what we're saying is for a mature program like we're describing here, where, you know, this is a serious mother-effing program, you don't want to run up against constraints on your fizzle material. Yeah. So there's got to be more. I literally don't understand why they have to.

Jeffrey 39:48

I mean, you don't want to get conspiratorial, but where's it coming from? Yeah. Well, you know... Yeah. I mean, it's weird, right? Yeah, it is weird.

Aaron 39:56

So my deep, dark... So I was... Thinking on my 11 hour airplane flight home, you know, my son was a total terror for the first five hours and he fell asleep. And so I had some peace.

Jeffrey 40:10

Did people look at you like you were a bad parent? Like this must clearly be your fault.

Aaron 40:15

No, luckily, you know, Turkish Airlines are full of so many young children that we weren't the only ones. But, like, my kid was, like, the kid on that flight. And it was, like, the first time.

Jeffrey 40:25

There's nothing you can do. Like, that's just how it goes, man. We were, like, wrestling with him at one point. But, like, anyways. Like, I always get annoyed when people, like, get pissy because, like, there is nobody who is more stressed out about that screaming, crying child than the parent who is desperately trying to do something about it.

Aaron 40:41

Yeah, yeah, yeah, exactly. Exactly. So I had this deep, dark sort of I'm worried that they're going to mount a warhead on top of a missile and fire it over Japan.

Jeffrey 40:57

Yes! I am too.

Aaron 40:57

Well, over Japan. It sort of just all clicked as I was looking at that fly-along map, you know, where you are in the world. And I was like, holy shit. I was like, that's what I would do.

Jeffrey 41:08

That's my joke is Juche Bird. Because we did it. We put a nuclear weapon on a missile on a submarine and we fired it. And the Chinese in 1966, with their fourth test, after we kept telling them, like, oh, you may have a nuclear device, but it's not small enough for a missile. The Chinese were like, oh, oh, really?

So here, they put a live nuclear weapon on a live missile and fired it across their country.

Aaron 41:32

Yeah, I'm worried they're going to fire it out to sea over Japan. That's exactly what I would do.

Jeffrey 41:35

And, you know, every time I mention this to Americans, they're like, well, we would. You know, and like what? Start a nuclear war? You would start a nuclear war over some dead fish? We couldn't do anything about it. Couldn't do a thing. Couldn't do a thing. Nikki Haley could give a really impassioned, angry speech about how the North Koreans are asking for it before we decide to sanction some more iPads.

Like, come on. Like, we have this idea that we, like, dominate this crisis situation, but, like, we don't.

Aaron 42:07

We don't have any response options anymore. No. So, when I slightly disagree with you is that you think people are still in the denial mode. I think people are in the panic mode. Not a better place to be, but okay. You know, I can't tell you how many questions I get, like, well, now what? You know, as if there's an answer.

And my response is always, well, look, I'm not a DPRK expert. I was like, email Jeffrey. I was like, but I was like, there's nothing to do, you know, unless you deal with them as a, you know, not as a peer. I mean, that's not the right word, but, you know, as a fellow nuclear power person. Who at least has a demonstrated nominal capability to put a warhead on the eastern and western seaboards of the United States.

So I've got a happy note.

Jeffrey 42:59

Yeah. So yes, I totally agree. There's nothing to be done. You know, like... Well, there is. It's diplomacy. It's the nasty job of talking to Kim Jong-un. All right. But let me get to my happy note. I'm going to get there. There's nothing we're going to say that's going to get them to give up their thermonuclear-armed ICBMs.

Like, that part's over. I mean, when we spent the past two years screaming that if we didn't get a freeze before they tested a thermonuclear weapon, the window would close. Like, that window is now closed. Like, that wasn't hyperbole. I wasn't trying to, like, create a false sense of panic in order to encourage a negotiation. Like, that window has closed.

The North Koreans are never, ever, ever, ever, ever, ever giving up their thermonuclear weapons any more than the Chinese are or the Soviet Russians slash Russians are or the French are or the British are or we are or the Indians or the... Like, that's just not going to happen. Mm-hmm. But the happy note, right, because I studied the Chinese nuclear program, and by the way, this would have been impossible to explain to the Johnson administration, but the way it played out is...

And I'm trying to imagine somebody like trying to tell Johnson this in the Oval Office. Well, you know, good news. The Chinese have tested a thermonuclear weapon or, you know, a nuclear weapon and then a thermonuclear. Like, why? Why is that good news? And the reality of it is that the people associated with the nuclear weapons program in China were also the people who were associated with the notion that China should have better relations with the United States.

Like, we have this idea that nuclear weapons are bad and the North Koreans are bad, and so the people who must build nuclear weapons in North Korea are probably double bad. But, like, we actually don't know that. You know, there's every chance that the people who build nuclear weapons in North Korea, instead of being crazy ideological party hacks, are pragmatic technical experts who would be delighted to see North Korea have a more open economy.

Like, we just don't know. And so while we have this idea that this is this terrible development, and it is because I actually think the North Koreans are going to stage a bunch of provocations. It's going to be a really rough couple of years. That part's going to suck. You know, it may well be that a North Korea that feels like it has an adequate deterrent, you know, over the next 10 or 20 years might be easier to live with.

Aaron 45:22

Are you making the case for some funding for some track two dialogues in Beijing? I think that's what you're doing.

Jeffrey 45:28

I think I'm onto your game here. No, I don't want any part of that. I can't even. Don't go to Pyongyang.

Aaron 45:35

No, I can't.

Jeffrey 45:36

Yeah, well, there's that too. Yikes. Don't go there. No, first of all, I don't want to go back to Beijing. I used to love China, but I can't deal with the smog and the lack of food safety. Can we do this in London? No, yes, I am making the case, and they should give that money to Mark Fitzpatrick at IISS, and he can host the meeting in London, and I'll go to that.

Well, that's a happy note. More money for Mark Fitzpatrick. Yeah, the great Fitzpatrick, who's been a little bit quiet, I think, on Twitter lately. I don't know why. Yeah, yeah, well... But it doesn't have to be bad, right? Like, it sucks that we're in this situation, and that's probably a reduced level of security, but, you know, oftentimes the... You know, we're now entering this period where the question is, what are the North Koreans going to use this deterrent for?

And if they use it to stage more conventional provocations, that's a bad thing, but it's equally possible that, you know, a different group of people have been empowered by this, and, you know... I don't know. Maybe Kim Jong-Sik is a, like, you know, nice dude. Like, I just don't know. We don't know these people, and so we don't necessarily know what the future holds.

Aaron 46:44

As a personal trainer would say, get comfortable being uncomfortable.

Jeffrey 46:50

Just as long as we don't have to feel the thermonuclear burn.

Aaron 46:53

That's right. Yeah.

Jeffrey 46:59

So, Scott, how are things going over there on the Slack channel? Are they any closer to getting a yield estimate on this thing?

Other voice 47:05

Oh, I can't reveal the secrets of the Slack channels and our workings, but if listeners would like to be a part of our community and go play around with the equations that we're working on, doing estimates, looking at the same data, or just having conversations about episodes and this sort of content, they should head over to patreon.com slash acwpodcast and check out our Patreon page there.

For $3 a month if you are a student or $5 a month if you are not, you can come join us on the Slack channel. We're working towards reaching enough money to get satellite imagery every month to disseminate to all Slack channel members. It's going to be really awesome. And currently, we have a meetup planned for this week for the DC area listeners.

So head on over to the Slack channel to find the details for that meetup.

Jeffrey 47:52

Yeah, you kick us a few bucks, and then you wait a day or two. You get very impatient. You send me an annoyed email, and then Scott hooks you right up.

Other voice 47:57

Yeah, it'll happen. Don't worry, guys. It's like a 24-hour turnaround on some. It depends. I have to do them all by hand, so it just depends on when I'm doing that batch.

Jeffrey 48:05

Yeah, I'm just teasing because it's like a standard MO.

Other voice 48:08

It's like, hey, I signed up for the Patreon.

Jeffrey 48:09

Where's my login? Scott's asleep. Yeah. Scott went to a movie. It'll happen. Actually, you're always very nice that whenever anybody sends an email like that, you almost always turn it around within a matter of minutes.

Other voice 48:23

Whenever the complaints come up, I get to it.

Jeffrey 48:27

Because we want to get you in there. I mean, we are truly excited about that. And, you know, it's still a little bit in the gestation period, but I think we're heading towards some t-shirts.

Other voice 48:35

Oh, yeah, very soon. That's now, like, after weeks of experimenting with storefronts and whatnot, that's now happening. So that's almost done.

Jeffrey 48:45

Yeah, I think we found our company.

Other voice 48:48

We also have other ways to support the podcast. We now have an Amazon Associates account, which you can reach through armscontrolwonk.com slash podcast. There's a link there. If you do your shopping through that, we get a small cut, which is a way to just send a couple dollars here without doing Patreon if you don't want. And if you are interested in advertising on the podcast or sponsoring an episode, send us an email over at armscontrolwonkpodcast at gmail.com.

Yeah, we're thinking about it.

Jeffrey 49:17

Yeah. I'm not sure it's a good idea. We'll think about it. If you're Raytheon, you know, we're happy to, you know, the SM3. Fine, fine missile. Everybody should own one. I'd love an SM3. You know, the best part is when you do sponsors like that, you know, for, like, the mattress company, they send you the mattress. You think Raytheon would sponsor us and, like, send me an SM3?

Aaron 49:35

I just want the missile model for my desk.

Jeffrey 49:38

Yeah, me too. That'd be pretty awesome. Yeah. All right. Well, Aaron, thanks for joining me. Always a pleasure. And thank you for listening.