Restricted Data

No single nuclear weapons test did more to establish the grim realities of the thermonuclear age than Castle BRAVO. On March 1, 1954, it was the highest yield test in the United States’ highest-yield nuclear test series, exploding with a force of 15 million tons of TNT. It was also the greatest single radiological disaster in American history. 

Among BRAVO’s salient points:

• It was the first “dry fuel” hydrogen bomb test by the United States, validating that lithium-deuteride would work fine as a fusion fuel and making thermonuclear weapons relatively easy to deploy.
• It had a maximum predicted yield of only 6 megatons — so it was 250% as explosive than was expected.
• And, of course, it became famous for raining nuclear fallout down on inhabited islands over a hundred miles downwind, and exposing a crew of Japanese fishermen to fatal levels of radiation.

It was this latter event that made BRAVO famous — so famous that the United States had to admit publicly it had a hydrogen bomb. And accidentally exposing the Japanese fishing supply to radiation, less than a decade after Hiroshima and Nagasaki, has a way of making the Japanese people understandably upset. So the shot led to some almost frank discussion about what fallout meant — that being out of the direct line of fire wasn’t actually good enough.

I say “almost frank” because there was some distinct lack of frankness about it. Lewis Strauss, the secrecy-prone AEC Chairman at the time and an all-around awful guy, gave some rather misleading statements about the reasons for the accident and its probable effects on the exposed native populations. His goal was reassurance, not truth. But, as with so many things in the nuclear age, the narrative got out of his control pretty quickly, and the fear of fallout was intensified whether he wanted it to be or not.

We now know that the Marshallese suffered quite a lot of long-term harm from the exposures, and that the contaminated areas were contaminated for a lot longer than the AEC guessed they would be. Some of this discrepancy comes from honest ignorance — the AEC didn’t know what they didn’t know about fallout. But a lot of it also came from a willingness to appear on top of the situation, when the AEC was anything but.

I’ve been interested in BRAVO lately because I’ve been interested in fallout. It’s no secret that I’ve been working on a big new NUKEMAP update (I expect it to go live in a month or so) and that fallout is but one of the new amazing features that I’m adding. It’s been a long-time coming, since I had originally wanted to add a fallout model a year ago, but it turned out to be a non-trivial thing to implement. It’s not hard to throw up a few scaled curves, but coming up with a model that satisfies the aesthetic needs of the general NUKEMAP user base (that is, the people who want it to look impressive but aren’t interested in the details) and also has enough technical chops so that the informed don’t just immediately dismiss it (because I care about you, too!) involved digging up some rather ancient fallout models from the Cold War (even going out to the National Library of Medicine to get one rare one in its original paper format) and converting them all to Javascript so they can run in modern web browsers. But I’m happy to say that as of yesterday, I’ve finally come up with something that I’m pleased with, and so I can now clean up my Beautiful Mind-style filing system from my office and living room.

Recently I was sent a PDF of a recent report (January 2013) by the Defense Threat Reduction Information Analysis Center (DTRIAC) that looked back on the history of BRAVO. It doesn’t seem to be easily available online (though it is unclassified), so I’ve posted it here: “Castle Bravo: Fifty Years of Legend and Lore (DTRIAC SR-12-001).” I haven’t had time to read the whole thing, but skipping around has been rewarding — it takes a close look at the questions of fallout prediction, contamination, and several “myths” that have circulated since 1954. It notes that the above fallout contour plot, for example, was originally created by the USAF Air Research and Development Command (ARDC), and that “it is unfortunate that this illustration has been so widely distributed, since it is incorrect.” The plume, they explain, actually under-represents the extent of the fallout — the worst of the fallout went further and wider than in the above diagram.

You can get a sense of the variation by looking at some of the other plots created of the BRAVO plume:

The AFSWP diagram on the left is relatively long and narrow; the NRDL one in the middle is fat and horrible. The RAND one at the right is something of a compromise. All three, though, show the fallout going further than the ADRC model — some 50-100 miles further. On the open ocean that doesn’t matter so much, but apply that to a densely populated part of the world and that’s pretty significant!

DTRIAC SR-12-001 is also kind of amazing in that it has a lot of photographs of BRAVO and the Castle series that I’d never seen before, some of which you’ll see around this post. One of my favorites is this one, of Don Ehlher (from Los Alamos) and Herbert York (from Livermore) in General Clarkson’s briefing room on March 17, 1954, with little mockups of the devices that were tested in Operation Castle:

There’s nothing classified there — the shapes of the various devices have long been declassified — but it’s still kind of amazing to see of their bombs on the table, as it were. They look like thermoses full of coffee. (The thing at far left might be a cup of coffee, for all that I can tell —  unfortunately the image quality is not great.)

It also has quite a lot of discussion of several persistent issues regarding the exposure of the Japanese crew and the Marshallese natives. I didn’t see anything especially new here, other than the suggestion that the fatality from the Fortunate Dragon fishing boat might have been at least partially because of the very aggressive-but-ineffective treatment regime prescribed by the Japanese physicians, which apparently included the very dubious procedure of repeatedly drawing his blood and then re-injecting it into muscle tissue. I don’t know enough of the details to know what to think of that, but at least they do a fairly good job of debunking the notion that BRAVO’s contamination of the Marshallese was deliberate. I’ve seen that floating around, even in some fairly serious forums and publications, and it’s just not supported by real evidence.

One thing that I hadn’t appreciated as well before is that BRAVO is pretty much a worst-case scenario from a radiological point of view. It was a very high-yield weapon that was very “dirty” right out of the box: 10 of its 15 megatons (67%) were from fission. 1

It was detonated as a surface burst, which automatically means quite a significant fallout problem. Nuclear weapons that detonate so that their fireball does not come into contact with the ground release “militarily insignificant” amounts of fallout, even if their yields are very high. (They are not necessarily “humanly insignificant” amounts, but they are far, far, far less than surface bursts — it is not a subtle difference. 2 )

But even worse, it was a surface burst in a coral reef, which is just a really, really bad idea. Detonating nuclear weapons on a desert floor, like in Nevada, still presents significant fallout issues. But a coral reef is really an awful place to set them off, and not just because coral reefs are awesome and shouldn’t be blown up. They are an ideal medium for creating and spreading contamination: they break apart with no resistance, but do so in big enough chunks that they rapidly fall back to Earth. Particle size is a big deal when it comes to fallout; small particles go up with the fireball and stay aloft long enough to lose most of their radioactive energy and diffuse into the atmosphere, while heavy particles fall right back down again pretty quickly, en masse. So blowing up and irradiating something like coral is just the worst possible thing. 3

Note that the famous 50 Mt “Tsar Bomba” lacked a final fission stage and so only 3% of its total yield — 1.5 Mt — was from fission. So despite the fact that the Tsar Bomba was 3.3 times as explosive than Castle Bravo, it had almost 7 times fewer fission products. And its fireball never touched the ground (in fact, it was reflected upwards by its own shock wave, which is kind of amazing to watch), so it was a very “clean” shot radiologically. The “full-sized,” 100 Mt Tsar Bomba would have been 52% fission — a very dirty bomb indeed.

In the end, what I’ve come to take away from BRAVO is that it actually was a mistake even more colossal than one might have originally thought. It was a tremendously bad idea from a human health standpoint, and turned into a public relations disaster that the Atomic Energy Commission never really could kick. 

In retrospect the entire “event” seems to have been utterly avoidable as a radiological disaster, even with all of the uncertainties about yield and weather. It’s cliché to talk about nuclear weapons in terms of playing with “forces of nature beyond our comprehension,” but I’ve come to feel that BRAVO is a cautionary tale about hubris and incompetence in the nuclear age — scientists setting off a weapon whose size they did not know, whose effects they did not correctly forecast, whose legacy will not soon be outlived.

1. Chuck Hansen, Swords of Armageddon, IV-299.[↩]
2. The count difference is about three orders of magnitude or so less, judging by shots like Redwing CHEROKEE. That’s still a few rads, but the difference between 1,000 and 1 rad/hr is pretty significant.[↩]
3. Couldn’t they have foreseen this? In theory, yes — they had already blown up a high-yield, “dirty” fission hydrogen bomb on a coral reef in 1952, the MIKE test. But somewhere a number of AEC planners seem to have gotten their wires crossed, because a lot of them thought that MIKE had very little fallout, when in fact it also produced a lot of very similar contamination. Unlike BRAVO, however, MIKE’s fallout blew out over open sea. The only radiation monitoring seems to have been done on the islands, and so they don’t seem to have ever drawn up one of those cigar-shaped plumes for it. See e.g. the discussion here on page 51.[↩]

Tags: 1950s, Atomic Energy Commission, Bad ideas, Bomb design, H-bomb, Nuclear testing, NUKEMAP

This entry was posted on Friday, June 21st, 2013 at 12:52 pm and is filed under Redactions, Visions. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.

Citation: Alex Wellerstein, "Castle Bravo revisited," Restricted Data: The Nuclear Secrecy Blog, June 21, 2013, accessed April 25, 2024, https://blog.nuclearsecrecy.com/2013/06/21/castle-bravo-revisited/.