Radiation Chart Update

Ellen and I made our radiation chart in the early days of the Fukushima disaster. I intended it to provide context for radiation exposure levels reported in the media.  I included a few example doses from monitoring sites around Fukushima (the only ones I could find at the time). But our main goal was to give people a better understanding of what different radiation levels meant.  It wasn’t a guide to what was happening at Fukushima because neither of us had hard data on that.

I’ve recently corrected a few things on the chart (the old version is available here). In particular, I’ve changed the mammogram dose from 3 mSv to 0.4 mSv, based on figures from this paper.  The other figures seem to hold up, and I’ve made only small corrections elsewhere.  I’ve added a few more Fukushima-related doses where I could find data, but they’re examples only—not full coverage of the effects.  Specifically, I added total exposure figures over the weeks following the accident for Tokyo, a typical spot in the Exclusion Zone, and a station place on the northwest edge of the zone that got a particularly heavy dose. Those data came from here (Google cache of now-dead MEXT page) and here.

Unfortunately, the disaster has progressed beyond simple radiation releases—there’s some amount of contaminated water, and radioactive material potentially getting in food. When radioactive material is ingested, the effects get a lot more complicated, and depend on what isotopes are there and how they’re processed by the body. Ellen’s page has a bit more information about that.

For reliable information on what’s happening in Japan, including discussions of the contamination levels, there are two sites Ellen and I recommend. One is the MIT Nuclear Science and Engineering hub, which posts periodic articles explaining aspects of the disaster, and the other is the International Atomic Energy Agency’s Fukushima Accident Update Log, which has detailed measurements from a variety of sources.

Note: Some people questioned the side-by-side comparison of short- and long-term doses.  It’s true that they’re not always the same, and I mentioned this in the intro note on the chart. Combining the two sacrificed precision for simplicity, but I don’t think it was a huge stretch—most regulatory dose limits are specified in terms of a total yearly (or quarterly) dose, which is a combination of all types of exposures.  And for those low doses, the comparison is pretty good; the place duration becomes important is up in the red and orange zones on the chart.

117 thoughts on “Radiation Chart Update

  1. “Extra dose in Tokyo” is reported as 40 mSv when it should clearly be 40 µSv.


  2. The link to the old version of the page has the same image on it as the most recent version – I suppose we were always at war with Eurasia?


  3. Confirmed. Both images have the updated 0.4 mSv for mammogram figure. We’ve definitely always been at war with Eastasia.


  4. Actually it seems both pages point to the same image, the old page isn’t old at all


  5. So, how does this compare to, say, standing next to a Super-Colliding Super-Button for about fifteen minutes?


  6. Using your chart, it is appearent that electrical engineers developed the sievert which like the farad or henry are rarely if ever observed without a pico or micro. In addition 1Sv has no tangible equivalent other than it will make you sick. I hereby suggest we change the standard unit of measure to the Banana. 1Banana (Bn)=.1uSv per your chart. This gives us an easy and tangible way to think about radiation. New York to LA=400 Banana trip. Break your arm? 10 Bananas. Radiation workers can have 500,000 Bananas per year.


  7. Thank you for this, I appreciate the education and visualization.

    I might be reading this wrong, but in the red quadrant of your chart: “Approximate total dose at the north-west edge of the Fukushima exclusion zone” says “40mSv” but only shows 2 red cubes. Compared to the 1:10 scale of the adjacent red examples (50, 75, 100 mSv), I would assume there should be 4 red cubes.

    Thanks again


  8. Pingback: Radiation Charts | Sphaerula

  9. Incidentally, the potassium from bananas have the same potassium mix as other sources of potassium. Hence, as long as you have a diet that gives you enough potassium, the extra dose from a banana is (to a good approximation, at least) exactly 0. Unfortunately, this renders the banana unit rather useless. Oh, and both Sievert, Farad and Henry are derived SI units, so it’s Napoleon’s fault, not electrical engineers.


  10. The figure I’d like to see added to the chart is the effect of standing in a TSA scanner for the duration of a scan. It’s immediately relevant to many people, and I think it would be a very good addition.

    Also, as others have pointed out, the Extra Dose in Tokyo says mSv rather than uSv, and the approximate total dose in the exclusion zone should have 4 boxes rather than the two it currently has.


  11. I was wondering what source you used for the chest and arm x-ray? I’m an x-ray tech and my textbook says the skin dose for a PA chest (shot back to front) is 17 mrem or 170 usv and a midline dose of 5 mrem or 50 uSv. A forearm x-ray comes out with a skin dose of 24 mrem or 240 uSv and a midline dose 13 mrem or 130uSv.

    Although the number used can vary dramatically based on the equipment and person the fact that your table shows an arm x-ray has less dose (20 times to be exact) flies in the face of everything I have ever learned, been told, or read.


  12. It’s a wonderful chart!
    I have just one comment: about the cellphone, I agree that it doesn’t produce ionizing radiation but, AFAIK, there is no evidence that it can’t cause cancer (i.e. there are many things that can lead to cancer without producing any ionizing radiation).


  13. Luis: Ionizing radiation is the only kind of radiation that can cause cancer. There’s just no physical mechanism by which non-ionizing radiation could do such a thing. I suppose you are technically correct, though, in that if you *eat* a cell phone, it might give you cancer.


  14. Randall, despite your statements about deferring to the experts in lieu of your chart you may be interested to hear that there is a blown up copy stuck up on the wall in the criticality, dose and shielding assessment department at the main office of Sellafield Ltd.


  15. I have completed experiments proving non ionizing radiation can ionize a gas with a chemical stimuli. contact me if you’re interested randall


  16. Randall, I am a nuclear instrumentation and controls technician at a nuclear plant in the eastern U.S and we also have a blown up copy stuck on the wall of the department. Plus I’ve handed out a few copies to friends and family who figured I was going to become a radioactive-zombie-mutant because of my job (Holla’ Marie). I LOVE your comics! thankyou.


  17. You could put smoking on the chart. Wikipedia says it’s a lot. So much so in fact that I doubt the accuracy.


  18. @JP
    > Ionizing radiation is the only kind of radiation that can cause cancer.

    Sorry, microwaves (non-ionizing) have the ability to rearrange biological molecules. For instance a microwave oven does this. Several (though not all) studies have found a carcinogenic effect from long term exposure. See the references in footnote 9 of the Wikipedia article on Microwaves.


  19. I’m hoping it’s just a unit error since it’s in the wrong box, with wrong colors next to it, but did tokyo get a 40mSv dose after the incident?


  20. @Verisimilidude
    Your reference is to an article where the data used is mostly Industrial and Military RF use not cell phones.

    Your average microwave is running at about 1000 watts. That is over 300 times more powerful than the highest cell phone wattage allowed on an old analog phone. That is equivalent to comparing the damage done to your body by an ice cube(32 F, 0 C,273 K ) to the damage done by attempting to walk on the surface of the sun (9930 F, 5498 C, 5777 K). Digital phones use even less wattage.


  21. I think another valuable addition to the chart would be to mention what exactly is the cancer risk. According to the Health Physics Society there is a 0.5% chance of dying from radiation-induced cancer per each 100 mSv of exposure.
    Taking this into account, I have no idea why the authorities forced everyone to stay out of the zone for the rest of the year. I don’t think I would seriously consider leaving my home for an entire year to lower my chance of cancer by 0.5%. For me the main takeaway from the accident is that the dose limits for the general population are paranoidally low.


  22. Listening to NPR, I heard the tail end of a report on radiation with X-Rays. The doctor being interviewed said there are studies being conducted concerning the ‘cumulative’ effects of x-rays. Does anyone have any more information on a cumulative effect or does the body absorb and the radiation eventually/naturally wear off?


  23. ../…../ˉn……..(‘(…′…′….ˉ~ /’)

    Google in the input: = tntn.us ==you can find many brand names, even more surprising is that he will sell you the unexpected o(∩_∩)o


  24. I would really love to see an addition for the amount of radiation exposure from the new-style airport scanners. I have a friend who refuses to be scanned for fear of radiation exposure – I would like to be able to show her how many bananas’ worth of exposure it is, and how it compares to the act of flying (up in the sky!) at all.


  25. Peter: Re Sellafield (neé Windscale), that powerplant doesn’t exactly have a good reputation over here – it’s sort of a byword for nuclear contamination in some places. Having to use that chart to reassure friends/family/visitors/new co-workers in an informal, friendly manner might be quite useful instead of trying to get them to digest a few pages of Times Roman text from an official RRPPS manual.

    Kryzysztof: Given that the official annual additional exposure limit to the general public outside of security screening or medical treatment is 1/100th your stated level (airline flights excepted), and 1/5th even for radiation workers, then the evac is just to make sure everything is as safe as it can be and no accusations of widespread increased risk can be levelled at the event. Same as all other situations where radiation is used. 0.5% is still a 1/200 fraction, or if you have a local population of 1 million people, it’s an estimated additional 5,000 people developing some form of cancer (which could be anything from a benign skin mole to some kind of aggressive, highly metastatic and rapidly fatal nervous system tumour). It’s a greater health benefit – or more accurately, a greater reduction in health risk – than they could gain from banning all private motor traffic and lowering the speed limit for commercial vehicles to 20mph, thereby almost completely eliminating fatal road accidents. On that basis, why WOULDN’T they evacuate until the area is cleaned up (to at least the 20mSv level…)?

    Graham: Pretty much all ionising radiation exposure is cumulative, that’s why you have annual dose limits for the public and workers (which, ultimately, form rough working-life and entire-expected-lifespan limits). You may instead be thinking about the difference between deterministic (a certain radiation strength is required for any effect) and stochastic (ANY exposure can potentially have an effect, but greater exposure = greater probability) effects?

    Verisimilidude: Why not just give us the link rather than spouting some babble and – to play devil’s advocate here – hoping no-one follows up a nonsense citation? I’ve worked in radiology and done the first couple years of a 4-year course, and at no point did we hear anything about that. They were FAR more bothered even about the potentially deleterious effects of standing next to a radio mast (the sheer amount of non-ionising energy causes resonant heating, that cooks you from inside), Lasers (very bad for your eyes, particularly the invisible IR/near-UV types, and at high intensities not very good for your skin either), and most of all far-UV X-Ray, Gamma, and A/B particles, all of which are so much more potentially damaging than Laser or RF (which is all that microwaves are) that the former barely even warrant thinking about. Hell, even in comparison to sunlight – something everyone enjoys standing in the way of! – you need a fairly strong radio or coherent light source to cause greater damage.

    Microwaves, with sufficient intensity and focus, CAN cause you damage …. because of the resonant heating effect which is the exact same reason they’re used for cooking. Put a piece of meat in your nuke-o-oven, set it to high for 30 minutes and come back. Looks pretty ruined, doesn’t it? That’s directly because of radiation, ionising or otherwise though. It’s because the water molecules in the dead muscle tissue were resonantly excited by the focussed, high intensity RF passing through them – same as they would by any other heat source, for example a resistive coil with a couple hundred watts going through it, releasing plentiful Infrared in the process – and the material underwent phase, oxidative, and other chemical changes as a result. As it is also effectively dead, rather than part of a living organism (actively respiring cells, in a semi fluid matrix, with excellent circulation connecting it remotely to gas exchange, fuel supply, waste removal and – crucially – temperature regulation mechanisms), it has none of the usual defenses against excessive heating that it would have done whilst still part of the unfortunate, slaughtered animal.

    Stand in front of a radar dish or sit on the aerials of a cellphone tower, and yes, you’ll probably come to some harm. Or if you stick your head in a microwave (# and get yourself a tan #) – assuming you’ve managed to defeat the deliberately explosive interlocks of course.

    Stand at one remove from said tower, or a household wifi device (limited to something silly like 10mW), or hold a cellphone to your head… there’ll be little measurable effect. Possibly a very small local increase in temperature, from holding a slightly warm electronic device operating in its highest power mode, in your hot and sweaty hand, against your face and therefore blocking off the pores and evaporative cooling that are your skin’s main method of temperature regulation.

    My phone’s alleged Specific Absorption Rate (I presume the amount of energy dumped into the tissues surrounding the ear whilst in use) is 0.35W/kg at maximum transmitting power, compared to a limit of 2W/kg. I should like to think that, given that the heated grips on my motorcycle can only just keep my hands at a normal temperature at full power (something like 36W) in deep winter, don’t become uncomfortable at minimum power (~9W?) until the outside air is above 20’C, and even my meaty paws probably don’t count for a kilo of flesh altogether, that even 2W per kilo would be barely noticeable after extended use, let alone one-seventh of that output. And that’s just resonant (and conductive, maybe?) heating of the live tissue, with pretty good circulation behind it. The manual does note that do not allow a damaged antenna to contact the skin, as a minor burn may result – one presumes this is an electrical or direct-conduction burn rather than anything radiological, because the phone is surely designed to allow as much of the antenna signal to escape as possible anyway.

    Unfortunately I have not found any information on the exact transmitting power, so I will have to improvise. The battery pack’s entire capacity is something like 5 watt-hours. I know that I can speak continuously on it for at least an hour, probably two, without fully draining it. So, let’s say that’s a peak output of 2.5 watts. Not a great deal. You’d have trouble burning yourself on a lightbulb of that power, unless it was a very small one. Now, that light bulb is also putting out a lot more infrared, which can directly heat your skin a lot more strongly (the microwaves from a typical phone aren’t in the cooking band, though cordless home phones, bluetooth and wifi ARE, which is why their power levels are incredibly low), but it does produce its output in much the same manner – unidirectionally. Or in other words, it (theoretically at least) spreads out evenly in all directions. It is unfocussed, so at any one time the portion of the signal that may be passing through you depends on how you hold it and where, taking into account slices through a theoretical sphere and all those kind of geometric things that make my head hurt – but is on average almost certainly less than half of that, if not a third. Let’s say you have 1.5W of weakly resonant, non-ionising microwave energy passing through you whilst making a call, then, shared between all parts of your body with which it may intersect, because, as before, it’s not focussed on any one point (and if your skull does actually act as a resonant/focussing cavity, it’d have to bring them down to a very fine point to create any significant heating, the phone would have to be surgically attached to your face to stop you moving it around and changing the target point of the focussing, and something would have to be done to slow down the not inconsiderable amount of cranial circulation…) …. and of course a good half or more of THAT gets through completely unscathed, as you can still successfully make calls even if you’re lying on the ground with your phone ear facing downwards.

    It’s not much is it. If you’re making a call outdoors, I’d be far more worried about the increased risk of skin cancer from sunlight exposure, or the chemical compounds in the sunblock you use to protect against the UV… (it can’t help against all of the higher powered rays, though).


    Luis: I’d call pretty much every study that’s ever been done into this either a/ being fundamentally flawed in ways that any reasonably bright high school student could point out, or b/ being inconclusive (meaning they failed to either disprove OR PROVE a link, because the stats are so muddy) AT THE VERY WORST, and as-good-as-disproving a relationship in most cases, a reasonable argument against it causing cancer in some way without needing non-ionising radiation.
    (it’s hard to fully disprove something in studies – it’s much better to design your experiment that discarding the null hypothesis = proving there is a link, and your success or otherwise depends on doing that)

    What, like there’s a little gremlin comes out whilst you’re making a voice call with a magic paintbrush, and he draws some cancer on you? You fancy telling us what these ways might be?
    Normally, non-radiative cancer causers are quite unfriendly chemicals… given that my own phone’s outer surface is glass on one side, and a combination of rubber and brushed metal on the other, perfectly safe and largely natural stuff, I’d like to hear what you suggest it could be.

    It’s Wifi, MMR, Spanish Cucumbers, Toyota Accelerators all over again.
    (yeah, that last one? turns out in all but maybe two cases, it was PEBSWAC rather than a mechanical fault. the old american propensity for being unable to decide between which of two spatially and topologically separate pedals to push rearing its head again. but the public imagination got hold of the story and ran with it… its not even like it’s unprecedented that drivers will go doo-lally and take off on a 130mph thrill ride (or 70+ in a truck) and then claim it on their vehicle, never mind that they had several minutes to investigate shifting out of gear, turning off the ignition, shoving both feet on the brake and/or clutch (though generally any servo boosted disc brake system is more than strong enough to lock the wheels, causing *clutch* slip (or torque converter stalling & overheating) should the engine actually be powerful enough to keep turning in the face of it. Not that these drivers would have bothered to figure that one out, what with one of their feet planted firmly into the carpet atop the throttle pedal)

    BTW when it comes to the towers? Inverse square (inverse cube?) law. If you’re outside the fence, you’re probably already far enough away that it won’t affect you even if you stand there all day. Cellphone receiving antennas actually have to be exceptionally sensitive and very carefully amplified in order to pick up the mast signal with any kind of reliability.

    Matthew Keen: NO U.
    If you’re gonna make wild-eyed claims like that, you can publish a link to your peer reviewed research paper right here in the thread, rather than trolling the good Mr Munroe for his personal email address.

    Was, e.g. strong acid involved, by any chance?

    Tarynn McKeage: How old’s your textbook? More modern X-ray sets are significantly more sensitive than the old ones. A similar quiet revolution was on the cards for Gamma Cameras at the point I left Nuc Med. (Reduction in dose along with an improvement in imaging quality AND less couch time for the patient? Gimme!)

    Technological improvement continues apace. The dose for a single X-ray 50-60 years ago would leave more greenhorn radiographers agog these days.

    Girts: What PART of Chernobyl are these people working in? You’re not referring to the bit in the partial concrete sarcophagus, are you? The one where there’s planned works to encase it properly in a better engineered and longer lasting concrete-and-steel oversized aircraft hangar affair to isolate it from the elements and trespass/interference whilst robots are sent in to chip away the old concrete, gain access to the reactor building, and start properly dismantling and decommissioning the ruined equipment, do you?
    There’s going to be some human-tended works during that building process, but they’re going to be on a piece of prepared ground (existing topsoil stripped, replaced by gravel grading and thick non-radioactive ashphalt etc) some distance from The Object itself.

    There are other parts of it still operating, but I would hazard that these days they’re a little more mindful of proper time-distance-shielding (and don’t let your reactor go into runaway with all the safety interlocks defeated during a fire drill, stupid!) and have cleaned down/replaced any contaminated surface or equipment so a typical worker’s annual exposure might not actually be too excessive.

    General reply to Randall et al:
    Yeah, I was following the TepCo press releases, particularly the supposedly raw monitoring data, more or less live – or at least, as they bothered to release it. I could sort of smell something wasn’t right, and it looked for all the world like figures were either being fudged, and/or monitoring procedures were changing to try and make things look better than they were. But was it a deliberate cover up, merely panic and mismanagement, or actually misleading readings despite all the proper procedures being followed?
    Even fooled me for a while at first. The readings looked high, but not terrible, and so I couldn’t understand quite what the panic was, or why greater measures weren’t being taken to handle the reactor overheating (even made a dumb youtube video suggesting/demonstrating how fire-fighting tanker planes could be used to quickly, repeatedly and safely drench the site with seawater; the crews suffering a high instantaneous exposure at some points, but only for mere seconds at a time, unlike the poor sods in the helicopters). Now it seems the reason may well be they were screwed, they knew they were screwed and that there wasn’t much that could help, and the exposure was so high that the helicopter pilots were either actually robots, or may have got dangerously high doses. This sort of behaviour – and the siting of reactors on fault lines in tsunami-prone areas without marine-grade generators to back them up for heaven’s sake – doesn’t do the industry or the idea of nuclear power any favours.

    Not that I’m a massive fan, as it does itself require a good amount of fossil fuel in the mining, production and transport of enriched uranium (unless there’s been a silent revolution and all of the mining, refining and transport equipment is powered from the reactors that they’re making uranium for), but until (not unless – *until*) renewables become a properly economically viable replacement option, it’s our best stopgap. Germany announced the shutdown of all of their nuke plants pretty much in the same week as (much more quietly) cutting much of their misguided solar feed-in subsidies, because the scheme wasn’t working all so well for making useful amounts of electricity, but it WAS quite a nice earner for those who had invested in panels, being paid back far, far more than the market value of the power they produced. But, without those two alternatives, what are they going to do now? Probably use coal. Yeah, that thing that, overall, produces quite a lot more radioactive contamination than actual nuclear plants (and more mercury than CFL bulbs…), spits out an insane amount of CO2 even compared to oil and methane, makes particles galore, and is still a finite resource in the end.


    Captcha: Χορων gntiont … seriously now, greek? WTF? It’ll be mandarin next… we’re doomed.
    (“ERROR: That reCAPTCHA response was incorrect”. I guess too many people just wrote “Xopwv” for the first one then rather than breaking out charmap.exe? The new one is “knotoo character”… not a character? too right…)


  26. that was a number of replies in roughtly reverse order, btw … and as often happens on boards like these, originally with quite a bit more clarifying linespacing than has actually survived the submission process. good luck.


  27. Hmm. No edit button. Corrections and errata, then, for technical issues and my own intermittent, subtle dyslexia.

    All backslashes = quotes. Thanks, autoconverting board software.

    In response to Verisimlfksjfoish….: Looks pretty ruined, doesn’t it? That’s directly because of radiation, ionising or otherwise though.
    should’ve course been That’s NOT directly because of radiation.

    In response to Graham: on further reflection… the deterministic effects may actually NOT be cumulative. But a cumulative build up could reach a deterministic level. Arrgh. It’s complicated and involves time constants and stuff like that. Even a megadose of 1Sv/sec still involves a certain number of radioactive decays to occur, the rays or particles to shoot out, intersect and be absorbed by your body over a certain portion of our journey through the 4th dimension. And your cells/tissues can repair some minor damage pretty much seamlessly and with no measurable lasting effects… so you have the battle of a short sharp exposure that can do you more harm than a longer term one where the instantaneous rate is a lot lower, but the overall dose is much higher. The effects of the latter may be a lot more subtle. And again, they’re both sort of cumulative, though the radiation sickness suffered with the “short, sharp” dose is deterministic in that it doesn’t happen in the latter case.

    As I said, it’s complex and unless you’re well grounded in it, can make your brain hurt. We hadn’t quite hit that part of the course when I was forced to quit.

    And to clarify my point vs Luis: We have plenty of studies effectively “disproving” any link between cellphones and cancer, and quite a few more that remain undecided (in neither case is there any properly definable correlation, but the latter type usually have some kind of experimental deficiency, e.g. very small sample sizes, that mean the statistical results simply aren’t precise enough for us to draw a definitive conclusion in either direction), but very few if any that “prove” a correlation between use of cellphones, or their transmissions, and increased cancer risk. Certainly, not any that aren’t one of the great many funded by anti-cellphone lobby groups (possibly backed by whoever’s still got a stake in coin-op payphones, landlines and the postal service?), or that don’t have significant deficiencies in their experimental quality control.

    I maintain my line, as I have done for the better part of a decade now, that if someone can bring to the table an independently funded, well designed and run, highly statistically valid, peer reviewed and hard-to-question study to the table which features, e.g —
    a study of two groups of 1000 mice/people/whatevers, randomly selected (but all with similar family histories of population-average cancer occurrence), who are then kept under similar-as-possible conditions but for one group being kept in a regular modern wifi- and cellphone signal rich area with a phone transmitter strapped to a set part of their body such that it’s unlikely to cause significant contact or insulative warming, left running at a typical “high power” level for several hours a day, and the others isolated from 21st century RF noise as much as possible and not even allowed a sniff of a phone.
    And then at the end of a couple years or so, shows that the transmitter-equipped group has a notable increase in cancer rate (in this case I think it would be 50 more, out of the 1000, vs the “control” group?*)….

    ….THEN we can talk. Until then, I’m putting my trust in the well-researched and field-proven theory, rather than goddamn snake oil.

    And perhaps we’ll see radiotherapists strapping their patients into exoskeletons made entirely of iPhones… whilst, outside the window, a flock of Vietnamese Pot-Bellies arcs gracefully through the summer sky.

    (* the question is, however… in this situation, which group is actually the control? Might it be a third one where the subjects are exposed to a population-average amount of cell signal and case-to-skin contact?)

    Captcha: Dratafer Daitō … romaji, now. Where will it end?


  28. As I said, it’s complex and unless you’re well grounded in it, can make your brain hurt. We hadn’t quite hit that part of the course when I was forced to quit. I agree! what’s your idea?


  29. Thank you for this graph – it’s been very helpful in explaining to friends and family about the kind of treatment my husband is getting for his cancer treatments (and they should be reading your comic anyway, because it’s excellent).


  30. Pingback: a long distance relationship

  31. Thanks for a really good infographic Randall. One of my colleagues and I were wondering if you’d consider doing a poster version of this (and no, we don’t work in nuclear safety…)


  32. I’ve heard on the news and radio lately that the authorities in Japan undersold the magnitude of the nuclear disaster (big surprise.) With the recent data coming out about the continuing disaster, I was wondering if you’re going to update the chart? It’s an awesome infographic regardless, so thanks for sharing!


Comments are closed.