How to make an atomic thunderstorm

by Alex Wellerstein, published November 15th, 2012

The rapid temperature and pressure changes produced by atomic explosions can, in fact, alter the local weather. This isn’t conspiracy theory kookery — it’s actually occurred numerous times in the course of nuclear testing. Here, for example, is an excerpt from a report about Redwing CHEROKEE, a 1956 test of a 3.8 megaton hydrogen bomb:

As the cloud rose and cooled, a very  intense purple with Szchrinkoff [sic — Cherenkov] radiation. Rain started in the area at about H+3 minutes after the burst, and at about H+10 minutes, a thunderstorm developed within the stem. Mr. Tanner and I counted 21 flashes appearing exactly like lightning flashes within a cloud.1

Lightning accompanied many hydrogen bomb detonations. Ivy MIKE, the first H-bomb, produced quite a lot of lightning, later analysis of the Rapatronic footage found:

Great, another thing to worry about. Image from Colvin, et al., “An empirical study of the nuclear explosion-induced lightning seen on IVY-MIKE,” (1987). You can see film of this footage here. Note that these sorts of things should not be confused with the smoke trails that were sometimes used to visualize the moving blast wave on tests.

That local weather changes would follow nuclear explosions isn’t too surprising when you think about it. What is weather if not pressure, temperature, and electrostatic charge? All three of those things are present in quantity when you detonate a nuclear weapon. I wouldn’t necessarily have guessed, a priori, that rain, lightning, and thunderheads could be created in the immediate aftermath of a nuclear bomb, but after hearing about it, it makes sense.

What’s more surprising, though, is that this was actually investigated as a way to enhance nuclear weapons as early as 1945.

In late April 1945, two theorists working at Los Alamos working on the possible health hazards of the “Trinity” test stumbled upon the fact that rising hot air (such as that produced by a nuclear weapon)  might produce rain. They sent a memo to J. Robert Oppenheimer raising the possibility:

After the ball of fire and hot air produced by the gadget explosion start to rise, conditions could easily exist favorable for the formation of a thunderhead. The initial velocity of rise of the hot air should be about 25 meters per second. Hubbard believes that a velocity of only 15 meters per second would be sufficient to produce a thunderhead provided that the atmospheric conditions were just right. He believes that the time when the proper conditions of humidity and instability would prevail over Japanese targets can be predicted accurately. In general they would be quite likely to occur in the summer months,. We are going to make a careful study of this question and its consequences.2

Two days later, they sent him another memo, this time discussing raising the possibility of making this happen deliberately. (As a side note, I always love it when the defensive swings around to become the offensive — this might be a problem becomes this might be a cool weapon with amazing rapidity.)

The memo, written by physicist Joseph Hirschfelder, had an ominous title: “Strategic Possibilities Arising if a Thunderstorm is Induced by Gadget Explosion.”3 Hirschfelder, his 1990 obituary explains, was a leader of a theoretical group at Los Alamos, and later went on to do physics work on the bombs dropped at Bikini in 1946.

Hirschfeder’s 1945 memo explained that “it would be feasible, if desired, to choose the proper weather conditions for delivery [of the bomb] so that the gadget explosion would induce a thunderstorm.” The physics seems fairly clear:

Joseph Hirschfelder’s Los Alamos badge photograph.

Because of the high potential temperature of the hot air, the active material and fission products would surely rise to heights of the order of 10,000 feet (in a time of three minutes) before the thunderhead would develop. With even a light wind the major portion of the active materials would be carried away from the area of blast damage (for a 15 mile an hour wind, one mile in four minutes) and the products would rain down on an area which has not been severely damaged by the blast (a radius of A damage for blast is considerably under one mile).

A simple calculation shows that the radiation from the active material and fission products would be sufficient to to render an area of from one to one-hundred square kilometers uninhabitable. Calculations which I have made on the smoke column would indicate that the radius of our smoke column would be of the order of 500 to 1000 meters therefore we could not expect to poison an area of more than a few square kilometers. …

I do not believe that there would be any lessening of the blast damage if we deliver the gadget in weather conditions favorable for the formation of the thunderstorm (conditional instability, humidity above 60%) and therefore the radiation effects might cause considerable damage in addition to the blast damage ordinarily considered.

In plainer language, Hirschfelder is saying, “hey Oppy, I found a way to make the bomb even more radioactive than we had previously contemplated. We’ll set it off in a way that will create a thunderstorm, which will spread radiation all over the place, even to places that weren’t hit by the actual blast itself.” Clever? Undoubtedly. Horrible? I find it so — it’s an attempt to make the bomb even more unpleasant than it already was. But in a sense, that’s part of the job description, isn’t it?

Hirschfelder closed the memo by offering that, “if you are interested in this possibility, we should try to work out more explicit details: how long it would take before the rain started, how predictable would be the area on which the active material was dumped, etc.” It doesn’t appear that Oppenheimer followed up on the issue, but he didn’t condemn it either. My total speculation is that he never followed up on it because it sounds a little complicated to pull off under wartime conditions — and waiting around for ideal weather conditions was tricky enough as it was without trying to create atomic thunderstorms. 

(A small, weather-related meditation: As you probably know, bad weather saved the city of Kokura, Japan, from being the target of the Fat Man bomb. Nagasaki was the runner-up, and even its mission was almost scrapped because of cloud cover. There was probably somebody who lived in Kokura who complained about it being so cloudy on that day, August 9, 1945, without realizing how lucky he or she was. When clouds get you down, cheer up! You might be living in Kokura.)

I came across this memo for the first time while going through the footnotes of Sean Malloy’s excellent article on what was and wasn’t known about radiation effects prior to dropping the bomb on Hiroshima. What struck me about it, aside from the gee-whiz aspect of ATOMIC THUNDERSTORMS, was how bloodthirsty these physicists appeared. Another document from Sean’s article, Bill Penney’s calculations on the ideal height to detonate the atomic bomb (with the special goal of trying to kill as many Japanese firefighters as possible), similarly affected me.

A common depiction is of the Los Alamos scientists as a bunch of giddy geeks whose “technically sweet” lab experiments get appropriated by the military for awful ends. But it’s a far darker story than that. These were some of the smartest people around at the time, and they applied all of their mental energies to the making of war — to the production of deaths. It’s not incomprehensible, of course: they knew they were doing wartime work, and there was, of course, a particularly vicious war on. But the flip side of all of those cute films and photographs of them drinking at lab parties is that when they weren’t there, they were plotting, in meticulous fashion, for killing as many people as were possible.

I think we’ve lost some of that in our collective memory. It’s present in some of the earlier depictions of the scientists and their work, but we seem to have compartmentalized our “weapons scientists” into the “good guys” (Oppenheimer, Bethe, Feynman) and the “bad guys” (Teller, Von Neumann) in terms of who we think are more dovish or hawkish. And yet, they all made weapons of mass destruction — some with more ambivalence than others, but they made them nonetheless. I’m not a total dove about these things, but I still think it’s worth keeping that at the forefront of one’s mind when talking about these guys. What I think is easy to forget when we read about Feynman’s hijinks and Oppenheimer’s highballs is that these geniuses were applying the entirety of their brains to a very grim job, one they did quite well. It is impossible to imagine the military men thinking up atomic bombs — much less atomic thunderstorms — on their own.

  1. Cherokee Field Report Bikini Operations, page 10, quoted in Chuck Hansen, The swords of Armageddon: U.S. nuclear weapons development since 1945 (Sunnyvale, CA : Chukelea Publications, 1995), 1307. []
  2. Joseph O. Hirschfelder and J.M. Hubbard to J. Robert Oppenheimer (23 April 1945), Nuclear Testing Archive, Las Vegas, Nevada, document NV0123756. []
  3. Joseph O. Hirschfelder to J. Robert Oppenheimer, “Strategic Possibilities Arising if a Thunderstorm is Induced by Gadget Explosion,” (25 April 1945), Nuclear Testing Archive, Las Vegas, Nevada, document NV0124031. []

18 Responses to “How to make an atomic thunderstorm”

  1. Nige Cook says:

    “Rain started in the area at about H+3 minutes after the burst, and at about H+10 minutes, a thunderstorm developed within the stem. Mr. Tanner and I counted 21 flashes appearing exactly like lightning flashes within a cloud.”

    It is worth pointing out that the Cherokee 3.8 Mt, 50% fission air burst you mention, despite being associated with thunderstorm conditions, was heavily documented for fallout by Terry Triffet’s Redwing Project 2.63 and produced no significant local fallout. See WT-1317, “Characterization of Fallout”, Figures 4.1 and 4.2. The Liberty ship YAG 40, 80 miles directly downwind, measures a peak dose rate of only 0.25 mR/hour around 9 hours after burst, which decayed rapidly. No significant fallout was measured on the numerous rafts and barges like YFNB 13 and 29, within Bikini lagoon and those immediately downwind!

    • Hi Nige: I’m glad you’re enthusiastic and passionate but I can’t have you using my comments section to promote your own agenda so thoroughly, especially when you are relying on calling other groups of folks liars and calling their work “drivel.” If you want to bring the other post of yours to a more scholarly level I’ll be happy to approve it but this is my blog, my rules, and I don’t find that sort of thing interesting.

  2. Jason says:

    From what we’ve learned, is 3:00 on that video on wrong in any way? I think that is probably extremely close. What do you say?

    • From what I can tell there are still people who debate exactly the best way to explain the role of radiation transport — that is, the exact way that the radiation of the fission primary is used to compress the fusion secondary (does the “plastic” matter much or not?). I really don’t care too much personally about that level of detail. Rhodes is pretty confident about the amount and positioning of the plastic in his diagram of the MIKE device in Dark Sun (which is what the PBS design is based on), which is somewhat divergent from the more Morland-style “exploding foam”/”foam-as-radiation-channel” version.

      Anyway, short story is, I think everyone who cares about such things agrees that it is basically correct, but there are what seem to me to be somewhat slight disagreements about how exactly the energy is transferred from primary to secondary. If one were a weapon designer I suppose the disagreements would feel more than “slight.”

      • Jason says:

        I wonder if they used FOGBANK back then or some similar agent for the ‘plastic foam’, which is apparently an aerogel. It seems that it came later on as a part of shrinking the designs. I’m sure by the 80’s they tested with and without it. Always wonder about that interstage part though.

      • Wotan says:

        My understanding is that the “plastic” is used to re-radiate in a more uniform way the X-rays from the primary to ablate the secondary, so it seems to be an important piece in the Teller-Ulam’s scheme.

  3. Nige Cook says:

    I called them “evil”, Alex! If you want to act as a censor to facts, I won’t comment here again, so good luck. Cheers, Nigel

    • I have your comment in front of me as I write; you said exactly as I said you did. I’m happy for a diversity of opinions but you have to keep it civil if it’s going to be on my blog. If you can’t handle that, feel free to say what you will on your own blog. I don’t think that’s an unreasonable request.

      • Nige Cook says:

        Hi Alex, I’ve called them factual, brief names on my blog – using the dictionary definitions for those who use “repeated distortions and complete misrepresentations of science” by the same groups for political propaganda for decades since 1918 with the resulting millions killed in wars – and that is I fear, very “civil” in comparison to the language which could be used to describe these guys. I fear you are insulting me as “uncivil” for sticking to concise facts. Cheers, Nige

  4. Wotan says:

    Alex, do you think Von Neumann is in the same league as Teller as a “bad guy”?
    He died too early (before, partial ban treaty, abm and arms control) to have a proper comparison with Teller, but most of his ideas seems less crazy/irrational than Teller’s.

    • I actually find von Neumann much more problematic than Teller. He didn’t have Teller’s ridiculous audacity — he was far more self-assured than Teller, and probably for good reason — but he was in every other way even more Strangelovian. Cold as ice, a complete hawk, an advocate of preventative nuclear war. At some point I will write a post about him and these particular points of view…

      • Wotan says:

        Maybe my judgment on Von Neumann is a bit biased by his scientific accomplishments, he was clearly hawkish, but it seems to me he behaved much more rational than Teller (even in the infamous Oppenheimer trial). Even disregarding Teller’s late years and the Oppenheimer controversy, his unwillingness to work on implosion during project Y and his obsession with the Super has always made him look quite crazy in my mind.
        But i guess Strangelovian doesn’t mean nutty, so maybe i’m off topic.
        Anyway, looking forward to your post on Von Neumann.

        • Teller was obsessive and he was, I think, fairly insecure. He wanted to be an “idea man” — he thought if he focused his attention on only big, interesting questions, then he’d find the most success. His track record on this is mixed; a few of the ideas eventually panned out, but a lot of them didn’t, and even those that did pan out (like the Super) did so at the expense of a lot of other work (e.g. his shirking of his implosion duties, which were then passed along to… Klaus Fuchs). He was emotional, perhaps to a fault: his issues with Oppenheimer and the Super in general became very wrapped up in something personal for him, and I think it caused him to act disproportionately.

          Still, I find those traits at least relatable. They make him human, albeit a very flawed human. (Among other flawed humans, as it is: Oppenheimer had his own fatal flaws.) And I think Teller suffered for his flaws. His actions in the Oppenheimer hearing, and his other political activities, led to him being ostracized for much of his life, and it’s clear that he was the sort of person who really felt wounded by that.

          Von Neumann, on the other hand, I find utterly unrelatable. Even his contemporaries thought of him as a human computer. He was clearly gifted, probably more gifted cognitively than anyone else at Los Alamos with perhaps the exception of Feynman. And he applied those resources methodically to the question of “megadeaths,” especially at the end of his career. If he’d had his way, he’d have nuked the USSR as soon as it was feasible to do so — a few million lives on the ledger didn’t seem to bother him as long as things worked out in the long run. I find this tendency more disturbing than Teller’s, personally. I don’t think Teller actually wanted to kill anybody; I think he genuinely saw the bomb as “insurance” and nothing more.

          Anyway, more on von Neumann another day.

  5. […] intriguing, and full of excellent, historical imagery. His posts range for those interested in the “science” aspects of nuclear history, to more of the social aspects (Civil Defense), and logistical/political […]

  6. A commenter here asked why the document linked-to has “LANL” written at the top of it, when this is not what they would have called Los Alamos at the time of its creation. My answer:

    The “LANL” at the top of the document linked-to here, along with the “RG 326 ATOMIC ENERGY COMMISSION,” are stamps that were added later. (RG 326 is an explicit reference to the NARA reference group.) They are indicators of where the physical records are stored (or were stored at one point in their existence), and were not applied contemporaneously with its creation. They are just indicators of where that particular paper copy was located, prior to it being photocopied or microfilmed or whatever, long before it was ever scanned as a grainy PDF. (It comes from a government document repository, where many such documents are hoarded, often with nothing more than such stamps to indicate where they at one point originally lived.) Incidentally, the same archive has another copy of the same document (identical except for redactions) as well. The exact path of this document, from its creation to its being in my hands, is not entirely clear to me, but I have no doubts as to its authenticity.

  7. Kenny says:

    I’m not what any of you would call an educated man. But I’ve always been fascinated by the power of nuclear weapons. I bought the video “Trinity” years ago, and was even more fascinated. And I often wondered what was creating what I called “the devil’s claw marks” . But now I know that it’s ligtning created by the weapon itself. I have been curious for years what mad the white streaks in almost every detonation I watched. And now I know. I have read nearly every blog you have online, and I feel somewhat smarter now. Thanks for publishing this information!


  8. GeBeWubya says:

    Dr. Wellerstein ,

    You ask rhetorically
    What is weather if not pressure, temperature, and electrostatic charge?
    and I answer pedantically:

    “Water is the key factor in the dynamic nature of weather. The release of heat when water vapor condenses adds an element of feedback into the processes of adiabatic cooling and depressurization of rising air. Moreover, water droplets or ice crystals are involved in the electrostatic charge buildup that results in lightning discharges.”

    The preceding trivial note gives me the opportunity to thank you for this excellent blog. I happened upon a post recommended on, and after following a couple of intriguing links, I decided to start from the beginning and read all of your posts.

    I particularly like your scholarly treatment of the historical evidence and your presentation and annotation of original source documents. You have given me a greater understanding and appreciation of the work of an historian, as well as of the history of the nuclear energy.

    Until I retired, I had worked in another area of R&D requiring security clearance and compartmentalization, which gives me a taste of the insider’s view of secrecy issues you explore. Please continue your scholarship and exposition; it is an important contribution.