News and Notes | Visions

NUKEMAP at One Two Three Four Five Six Seven Eight Nine Million “Detonations”

by Alex Wellerstein, published February 23rd, 2012

I woke up this morning to find that NUKEMAP had hit well over one million “detonations.”

Remember when I was very impressed that I had 1,500 detonations? Yeesh.

These have been spread over about 190,000 unique visitors. The average of ~5 “detonations” per visitor has held pretty solidly over the last week.

Here are some visualizations I threw together showing where those million-and-change detonations fell. Each of the dots has an opacity of only 25%, so when they look bright red, that means they’re being stacked on top of each other. I’ve also thrown out any perfectly redundant data, so nuking the exact same spot repeatedly doesn’t change how it is rendered.

Click the image to zoom in. For details of various regions, click here: the USA, Europe, Asia and the Middle East, South America and Africa, and Oceania.

I will be soon writing up a somewhat formal analysis of this data, and other feedback I’ve gotten as to how NUKEMAP was used, talked about, spread around, and so on. I’ll let you know when it’s up.

Until then, Tom Lehrer will serve as analysis-by-proxy:

After the jump is a brief NUKEMAP FAQ of sorts, based on various blog comments, forum posts, news stories, and so on that I’ve seen on this.

Update: It’s only 2/26 (three days after I wrote the above) and we’re already at the second million. If I had a nickel for every Tsar Bomba dropped… well, I’d have about $25,000. Update: NUKEMAP hit three million around 3/6. Update: NUKEMAP hit four million sometime around 4/12. Update: NUKEMAP hit five million sometime around 7/12. Update: NUKEMAP hit six million detonations on 8/27 (another round on Reddit). Update: NUKEMAP hit seven million detonations on 10/22 (Reddit, again). Update: NUKEMAP hit eight million around 10/29 (tail end of that Reddit traffic). Update: Nine million around 12/22/12 (Reddit again).

I’ve written up an analysis of 4.3 million “detonations” and their locations for the online journal WMD Junction. Check it out.

The brief NUKEMAP FAQ of sorts:

  • This doesn’t take into account terrain or fallout or height of detonation or …  Yeah, I know it. It isn’t meant to be a rigorous simulator with fifty variables. Those things are out there in the world if you’re looking for them. This isn’t that. Sorry. (Hand-wavy fallout plumes will be added in the next big update, but those aren’t going to satisfy experts, either.)
  • Is this meant to be a game? No. “Game” is entirely the wrong word. “Toy” is probably closer to what is meant when people say “game,” and it’s not quite what I thought I was designing either, but so be it. I like to believe that a little bit of education is leaking through, even if a lot of the usage is just “for kicks.”1
  • Is this meant to be pro-Iran, anti-Iran, pro-Israel, anti-Israel…? None of the above! Sorry, I didn’t make this with the intention of pushing any particular political position. Of course, people will read into it what they want, what they think it means with regards to the nuclear issues of the day. But I didn’t make it to push any particularly partisan point of view. Now, that might seem rather naive, but my general position is that the effects of nuclear weapons can be interpreted through a multitude of political lenses. The hawks and the doves come together on their interest in nuclear effects — they just read the meaning of them very differently.
  • This isn’t new, there are others on the Internet who have done similar things… I tried to give credit in my original NUKEMAP post to the other simulators out there. Frankly, even those are just the latest iterations in a pretty old genre. I’m not claiming to have re-invented the wheel here, and I’ve been as surprised as you are to see this sent around to so many people. I think we can conclude from this that a significant number of people out there hadn’t seen the other simulators. Anyway, it’s not a contest. (As of this writing, I’ve not made a single cent off of NUKEMAP, so there’s no need to feel that there’s too much at stake in the “priority” department.)
  • Are you really as macabre/insensitive sounding as you come off? In the earliest publicity about it, I hadn’t really considered that it would end up being such a big deal. So I probably would have chosen my words a little more carefully. My enthusiasm for the tool should not be confused with an enthusiasm for the bomb. One is a little script I banged together over the course of a few days, the other is a massive historical and present issue. That being said, there’s a long history of using dark humor to talk about nuclear weapons. I’m not trying to trivialize their effects — much to the contrary, I’m trying to communicate them and make them feel “real.” Again, I hope a little education leaks through.
  • Why’d you list bomb X or city Y…? I tried to pick a number of “major targets” and a number of historically and presently interesting yields. There is no rigorous methodology behind it. I put Washington, D.C., first on the list not because I dislike the city, but because I live there. (I live within 4 kt of the U.S. Capitol building, so I’m sensitive to these things.)
  • It doesn’t work on my computer.  It seems a very small minority of folks can’t get it to work. Leave me a note with what browser and operating system you’re using and I’ll see if there’s anything I can do about it. The odds are it’s not my code but Google’s that is incompatible. But I’ll do what I can.
  • Wait, who are you, again? Some of the news accounts have garbled up my personal information, but I’m an historian of science who has a Ph.D. from Harvard and who currently works at the American Institute of Physics (a professional organization for American physicists) near Washington, D.C. I’m not a physicist, though I sometimes am labeled one on the Internet. NUKEMAP emerged out of teaching exercises I developed while teaching at Harvard and MIT, but I don’t teach anywhere at the moment. I study the history of nuclear weapons, and am finishing a book on the history of nuclear secrecy policies in the United States. This is my blog.
  1. If you do want to play a “nuke your neighbor” game, check out DEFCON from Introversion Software. Note that it’s fairly unrealistic, because it doesn’t just label everyone who plays a big loser. []

31 Responses to “NUKEMAP at One Two Three Four Five Six Seven Eight Nine Million “Detonations””

  1. Allen Thomson says:

    A very nice tool, perhaps educational in both directions (for us the users and you the proprietor[*]).

    A suggestion for future enhancements: Allow for multiple aimpoints in a single image, so as to visualize attacks with MIRVed missiles and all-out attacks against major target areas.

    [*] I’d bet you’re going to find hits on places that have no obvious targets but, on study, might turn out to be interesting.

    • Thanks, Allen!

      You can currently add more than one detonation if you click the “Detach marker” button. (I probably should come up with a better label than that — it’s not obvious what it means.) That removes the marker from any current detonations and allows you to plot new detonations as well. As long as you don’t click “Clear” the old ones will remain.

      I’ve been playing with the idea of making MIRVing somewhat more straightforward (e.g. you could select a maximum number of warheads for certain historical/current weapons and it would try to tesselate them perfectly based on the yield) but haven’t tried to implement it yet.

  2. I feel for you on the pro-/anti-/etc. part. HYDEsim has been used by hawks and doves alike to make their points over the years, and I’d occasionally get hate mail about wanting the terrorists to win. Good luck.

  3. Andrew Tubbiolo says:

    I nuked Tucson. I did it per the old Cold War targeting. I nuked Davis Monthan AFB, the Hughes (Now Raytheon) missile plant at KTUS, I hit the Titan Missile Museum, and then pretended an R-36/SS-18 made a first strike against the < 1983 Titan Missile force that surrounded Tucson, then for good measure I hit the air strips up to Phoenix. Next we need something to render the landscapes with the mushroom clouds as they develop and fall out.

    It's a good first stab at a tool for the public. Given the era, what is the likely targeting for your town? Given known traffic patterns how would you die on your way to or from work, shopping, school? How far would you be from your loved ones when the big one falls? Would your home be swept off the ground, burn, or have your windows break? Given your typical life patterns and you survived the initial hit, how soon could you expect fallout to affect you? What gas, water, electrical lines would be cut? Likely firestorm?

    A tool like this while dramatic would be useful in the public debate sure to come as the world goes multipolar, and the bankrupt major powers come to rely on nuclear weapons to maintain a balance of power.

    • NUKEMAP actually grew out of work on another map project (at the moment dubbed TARGETMAP) which seeks to overlay various historical targeting scenarios over Google Maps. (So you could see what the US thought they would do to the USSR in 1950, for example, and compare it to what the US thought the USSR was going to do to them in the 1980s.) That’s gotten a bit sidetracked at the moment (I’m still compiling the data for it), but the original intention was to help people figure out whether their cities really were on anybody’s “lists.” NUKEMAP was an easy side-project because it doesn’t require too much additional data; just a few equations. (Now that I’ve made NUKEMAP, I think it has more popular potential than TARGETMAP, because you can actually “do” something with it, rather than just look up information. But I didn’t expect either of them to take off to quite such a degree.)

      It would be fun to get together some genuinely technical people to translate some of the really more complicated simulators (e.g. KDFOC3 for fallout) into Javascript for Google Maps. But it would take a real team of folks to gin up that kind of effort, and would require people more gifted in math and physics than I am. I think the appeal would be more limited — I don’t think your average user wants to trade-off accuracy for usability and speed — but it would be a fun project.

  4. Bruce Roth says:

    There is no substitute for the best, and this is the best simulator I’ve seen! The incredible number of hits confirms that!

  5. Dr. Kenneth Noisewater says:

    BTW, you may wish to include Shot Grable (US nuclear artillery) in your list..

  6. Norris Krueger says:

    Just found this, Alex. Not to make your life even more complicated, but it seems there could be a mashup with population and other data – so what are the estimate casualties (fatal & otherwise)? The property damage? etc.

    If you’ve got an open API, who knows what others might come up with.

    p.s. when I was an undergrad at Caltech, one of my profs had been a SALT negotiator & you’ve brought back some find, if weird, memories. Thanks!

    • Hi Norris, thanks for writing! I address the “population question” here. Short version: it’s possible with current technology, if you marry the Landscan population density database with Google Maps, which has been done. But it’s not easy and would require significant work and probably money (for the aforementioned database and the programs needed to manipulate the data). So for the moment it’s out of the cards, though someday it may either a. get easier, or b. someone will provide me the money/technical expertise to do it myself. I would not be surprised if Google made it easier to do this sort of thing in, say, the next 5 years or so. But at the moment it is just on the edge of feasibility.

  7. Jason says:


    Kudos on a great sim. I’ve had great fun destroying the state of Nebraska over the past couple days. (Though all those damned silos way out in the Panhandle are a PITA to locate & target.)

    I know you alluded to it, but I just wanted to explicitly suggest that if feasible, it would be cool if a future iteration(s) could take into account air vs. ground-burst weapons.

    Incidentally, what’s the assumed height for the blasts in the current model?

    • Hi Jason: I think the height is just assumed to be “ideal,” which means it would vary based on the size of the yield. I might look into what kind of modifier would change it for a ground-burst as part of a future iteration.

  8. Alex says:

    Wow! What is missing is the animation

  9. kme says:

    I notice a conspicious block of red on the Falklands…

  10. Fan says:

    Is it possible to include a somewhat “up to date” feature of global detonation visualization like you had in the beginning of this article?

    I found that the overall detonation distribution has become an interesting talking point more than individual detonations. I know most people would blow up their own neighborhood first just to see what it would be like, but the overall chart can still show which areas people are most interested in.

    • Hi Fan: Sure, I can try to generate another one today with the updated info. (I tried a little while ago, when I hit 3 million, and my script somewhat overwhelmed Google Maps’ API, basically ending up crashing my browser; but with a little tweaking it should be doable.)

  11. J B says:

    Thermal radiation radius: 1.434975864189056e+21 km / 891652663562952200000 mi

    Nice. That is around 151 680 308 light years. Give or take a few angstroms.

  12. J B says:

    I set my sights too low.

    Thermal radiation radius: 4.24124808 × 10^112 Parsecs
    Third-degree burns to all exposed skin; starts fires in flammable materials, contributes to firestorm if large enough.

    That is around 1.514285 x 10^103 times the size of the observable universe. [ depending on when the Inflationary Epoch ended ]

  13. […] NUKEMAP news, I’ve written up an analysis of who-bombed-where for the online journal WMD Junction. […]

  14. Jim says:

    A lot of people seem to have a grudge against the ocean!

  15. […] million “detonations” since it was first made public. I’m both agog and aghast. Click here for a FAQ of […]

  16. Nate says:

    I love this tool, website, or whatever you may wish to call it. Nuclear weapons are such a hotly debated topic that facts like the ones this website provides are rarely revealed. There are people that wish to say that nukes WILL end the world as we know it, and those that say nukes will DEFEND it. But however your political stance ( mine would be that of a peace loving libertarian ) this website relays the hard facts, even though certain relevant conditions are ignored, the large majority of the information revealed rings true.

    My father works as an engineer at a nuclear power plant outside of DC (Calvert County Maryland) and while this hardly qualifies me as an expert, I do know a bit more than average on the topic. I love this site and so does my father! It is in the truest sense of the word, an educational tool.

  17. Nate says:

    I would love to see an air burst vs. ground burst function in the website though.

  18. […] noter : l’auteur indique dans cet article que plus de 6 millions d’explosions ont été testées. Il fournit même une carte avec la […]

  19. Mark says:

    Great app! However, I have a problem: using a link I had successfully used yesterday (Saturday, Sept 22), I got the target variables sidebar but no map to place my locations on; I typed my targets and other parameters but nothing appeared in the way of a map. Is it broken?

    • The map end of things is handled by Google, so if that part breaks down it is likely something wrong with them. It is working for me now just fine. There are a few small changes I have been meaning to make which might improve its compatibility with various browsers. But I can’t tell much if you don’t tell me what kind of operating system and browser you’re using.

  20. Mephisto says:

    Nice tool, I use it very often actually, to try to visualize my surroundings with weapons effects. Food for thought for many…
    I don’t know if this is possible, but there is a few more options I would like to see, if it is possbile to implement that is…or if You have the time for that matter 😉

    Having a MIRV-option would be nice yeah, although I predict for it to be accurate…ish it would require some serious coding…? I seem to recall reading somewhere that a D-5 MIRV footprint is an ellipse that is 300 km wide and 500 m long in the direction of fire, using a minimum energy trajectory. I guess you would need a point of origin as well, not sure if that is possible to implement.
    One other feature, that might be easier to implement, is a function for neutron warheads, and their enhanced radiation aspects. Would like to see that if You have the time 🙂

  21. Mephisto says:

    And the airburst-option already mentioned, that would be very nice as well 🙂

  22. Will says:

    Just wondering if you have taken into account the fact the the google maps API basically uses a Mercator projection and these euclidean distance buffers created for detonation radii are going to be horribly inaccurate. Given the fact that this buffers show up as perfect circles on your map I’m going to assume you haven’t taken that into account and I would advise trying to use geodesic buffering for a more accurate picture. ESRI has a article on this problem…

  23. […] of the traffic I received was driven not by my sterling content, but by that funny creature, the NUKEMAP. It was not my expectation that it would get so crazy and drive so much traffic to the blog itself, […]

  24. […] mapping of where people bombed doesn’t look significantly different than did the first million, so I won’t post another one of those images. But here’s some fun-with-data for you: […]

  25. Alan says:

    If you can somehow pull the code for

    and translate from fortran to javascript (completely capable of running the fairly simple math)

    then you will have better than hand-wavy fall out plumes. It would be even cooler if you could integrate wind direction from wunderground.

    just make sure you get the post-1999 version of the code, there was a sign error around line 6000 of the older versions.