TMI: Fear, Fukushima and Facts

Thirty-two years ago today, I noticed that things were odd at our pre-school. The teachers had drawn the blinds closed in a few rooms and at least one of the other kids had had his mom pick him up in a big wood-sided station wagon before the day was actually over. I didn't really know much more about what was going on, or why the day was strange, until years later I read much more about the accident at Three Mile Island and realized that since my pre-school and my home were just ten miles from the reactor meltdown, it's surprising that the parents of our classroom full of three year olds weren't more disturbed.

The accident's been on my mind a lot lately since the horrible earthquake and tsunami in Japan has put the Fukushima plant at risk. Beyond the obvious parallels, one thing that's stuck with me is that, aside from being a very young kid at the time, I hadn't originally known very much about what the hell actually happened at Three Mile Island, and much of that was because I didn't really know how nuclear power worked, or what the real risks were, or what had actually happened at TMI. Fortunately, even in that pre-Wikipedia era, the information was pretty easy to get hold of, because local papers back then were very interested in accurately covering the story.

When I was a kid, "TMI" was synonymous with "Three Mile Island". It's only since I've been an adult that it's ironically come to mean "Too Much Information". Because that's obviously not the case with news coverage of the Fukushima reactors here in the U.S. Fortunately, I did have access to an expert who could truly explain the context and severity of the threat at hand.

TMI Radiation XKCD

An Expert's Voice

Most of the folks I know online have seen Randall Munroe's Radiation Dose Chart. Randall's an acquaintance, and I'd been following his work since the beginning of XKCD, so I knew that he had a physics background and is a diligent researcher, so his work would be fact-driven and as accurate as possible. He talks about his sourcing and his caution in more detail on his blog, and on the whole I was very pleased with how meticulous he was.

But in my case, I had an unfair advantage: My father-in-law Stephen Browne is actually a health physicist, an expert scientist trained in measuring the amount of radiation that people are exposed to, and tracking it against the amount of exposure that's permitted. After he'd done a really insightful radio interview about his perspective on the dangers at hand, I'd sent him Randall's illustration and asked for any feedback or comments he had on it. He's kindly agreed to let me share them here, and I'll try to filter back any questions raised if clarifications are needed. His response to the XKCD Dose Chart follows. (I've added emphasis for clarity, but not changed any of the text.)

Well, the individual dose numbers are about right, but you cannot sum numbers which represent dissimilar things, such as one-time hypothetical exposure events, annual average exposures, thresholds for biological effects, and regulatory limits. Also, partial body doses (e.g., x-ray of hand, chest, arm, and mammogram) cannot be compared to whole body doses without an additional weighting factor. It would be better to just list the numbers in ascending order to show their relative magnitudes.

I think it would be more helpful for people to look at the two charts discussed below which were produced by the National Councial on Radiation Protection & Measurements (NCRP), a scientific body chartered by Congress.

  • The first chart breaks down the total dose to the U.S. population by major category. The magnitude of the average dose in the U.S. is about 620 millirem1 per year, of which 50% comes from natural background and 48% from medical exposure.
  • The second chart breaks down the collective occupational dose in the U.S and shows the groups receiving the greatest collective dose are medical and aviation -- not what most people would guess.

The public also ought to know that the fatalities attributable to radiation exposure from the two most serious previous commercial nuclear power accidents are very low.

  • TMI = 0 deaths or injuries
  • Chernobyl = 28 short term deaths among plant and emergency response personnel from acute radiation syndrome
    and 15 excess deaths from thyroid cancer. There were many more cases of thryroid cancer but, being that it is one of the most treatable forms of cancer, few deaths. These numbers are based on the recently published report of the United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) which looked at peer-reviewed studies of the radiation effects on populations over of a period of 20 years after the accident. You will hear claims that tens of thousands to a million deaths were caused by Chernobyl, but that is false. Those numbers are projected deaths based on cancer risk per unit dose x collective population dose, which is an invalid method because low doses carry no risk. Actual cancer death estimates can only be determined from post-accident epidemiological studies, which is why it has taken 20 years.

1 100 millirem = 1 millisievert (mSv)

I was really delighted at the amount of clear detail expressed there, and that a lot of facts that I think are underreported in most media sources seemed very comprehensible. There was also some great insight into how the media is often miscommunicating about the very measures of exposure:

It is not easy to make all of this stuff understandable to a layman and still get it technically correct.

One of the common problems I see in the media is the failure to distinguish between dose and dose rate. That's like mixing up miles and miles per hour. It makes a lot of what is reported confusing and hard to interpret.

Moreover, risk of harm is a function of both dose and dose rate. The same total dose spread relatively evenly over weeks, months, or years (chronic exposure) carries much lower risk of harm than the same total dose received over minutes, hours or days (acute exposure). This has to do with the body's ability to repair damage at the cellular level. So you can't really estimate risk accurately without knowing something about both dose and dose rate.

You are now starting to see a lot of reporting of radioactivity levels in food and water relative to government established concentration limits. Internal exposure is much more complex than external exposure. However, it is important to understand that accidental consumption of something that is above the limits, perhaps even many times greater than the limits, does not automatically mean the person is at great risk of harm. In general, it would take repeated or prolonged consumption at levels above the limits to have a significant risk as all limits are set very conservatively. Essentially, the risk is a function of the accumulated internal dose. The radioactivity concentration limits for food and water are just derived values, meaning you start with a dose limit and calculate backwards to find the concentration that, based on a complex set of assumptions and models, would deliver a certain internal dose to a hypothetical person in a given period of time.

Spread The Word

While I was very pleased to get this level of detail about the actual measures and dangers of radiation exposure, I have been a bit disappointed in the reality that more people don't know about these basic facts. This may be doubly true since I'm part of one of the world's preeminent organizations that's dedicated to sharing scientific knowledge.

So, I'm gonna make an unusual request: Please share this link with your social network. I've never asked people to tweet about or blog about what I write, but I am still optimistic that sharing actual facts through these networks can counteract the disinformation or misinformation that gets sensationalized in other media. And maybe we can do a bit to counteract that. And of course, if you've got links or comments with more details, please feel free to add them.