What a wonderful statistic

Marcellus Shale is 8 to 32 times background. This compares to an average radium-226 in
surface soil in New York State of 0.81 picoCuries per gram (pCi/g)5
Using this range of radium concentrations and a simple Fortran program that simulates
the production of radon in the well bore, and transit to the wellhead, we calculate a range
of radon concentrations at the wellhead between 36.9 picoCuries per liter (pCi/L) to 2576
pCi/L.
These wellhead concentrations in Marcellus shale are up to 70 times the average in
natural gas wells throughout the U.S. The average was calculated by R.H.Johnson of the
US Environmental Protection Agency in 1973 (pre-fracking) to be 37 pCi/L6 to a
maximum of 1450 pCi/L.

Somehow the range, from the same as other gas to under twice the amount of other gas, becomes up to 70 times the concentration. By comparing the lowest part of one range with the highest part of the other.

That\’s a really rather remarkable use of statistics. Entirely and absolutely true in detail and entirely and completely misleading overall.

Far more accurate would be to sway that fracked gas from the Marcellus has up to twice the radon content of other natural gas.

That may or may not be a problem what with radon exposure most certainly being capable of producing lung cancer. But that is what the problem is.

9 thoughts on “What a wonderful statistic”

  1. A salesman looking for a contract to investigate the fictitious problem.
    And he is assuming there is no ventilation in the kitchen

  2. a range between 37 and 2576 is so wide as to be basically meaningless – especially since it has such a considerable overlap with the range quoted for natural gas wells. The normal response would be to assume that it is not scarily different but they should supply more descriptive statistics.

  3. This is the paper in full. It’s not very well presented, but Tim’s criticism is not entirely to the point.

    The data reported by Johnson come from measuring actual radiation levels in natural gas supplies from various wells. You can download a pdf of that report here. The average (not the minimum) radon level was 37 pCi/l, and the maximum was 1450. Johnson’s analysis suggested that the radon would cause 15 excess lung cancer deaths per year nationally, and that it wasn’t worth doing anything about it.

    The range of radon concentrations in the current paper come from different assumptions in the simulation. It’s likely that the actual measured range would be much narrower, and at least plausible that the levels are many times higher than 37 pCi/l. It would be prudent to make some actual measurements. At some concentration, which handwavingly I estimate to be well within the plausible range for Marcellus shale gas, it would be worth doing something about the problem.

    The thing to do would be simply (but not very cheaply) to store the gas above ground for a few radon half-lives (the half-life is 3.8 days).

  4. Anecdata alert.

    A colleague (many jobs ago) took his radiation dosimeter home (Devon) for a long weekend. >500µSv. Once the dosimeters went in at the end of the month, there was a significant panic until they’d worked out why somebody who never went near anything nuclear had got the biggest dose of all of us that month.

    The local council installed forced air ventilation into his basement.

    As radon is trapped in basement areas (and I’ve no idea what size his house was), I’m not sure what that relates to in gas concentration etc. And it’s not relevant to fuels in any way …

  5. What’s a Curie, if a pico curie is 1 X 10 to the minus 12 of it?
    Enough to kill a Curie, perhaps. Well he did die of it.
    But 1,000 piCu X 1,000 litres (only in the kitchen) suggests that the danger might be significant over approximately 1,000 years.
    I should live so long.

  6. So Much For Subtlety

    blokeinfrance – “What’s a Curie, if a pico curie is 1 X 10 to the minus 12 of it?”

    A curie is an obsolete measurement of the rate of decay of a radioactive substance:

    The Curie (symbol Ci) is a non-SI unit of radioactivity, named after Marie and Pierre Curie.[1][2] It is defined as
    1 Ci = 3.7 × 1010 decays per second.
    Its continued use is discouraged.[3]
    One Curie is roughly the activity of 1 gram of the radium isotope 226Ra, a substance studied by the Curies.

    Or by extention, the amount of any given substance that would produce 1 Ci.

    Enough to kill a Curie, perhaps. Well he did die of it.

    Hang on, no he didn’t. He got hit by a horse drawn waggon. She did. Probably. But is the Curie named after him or her? Wikipedia is terribly politically correct so it says both but what is the betting it is just him?

    But 1,000 piCu X 1,000 litres (only in the kitchen) suggests that the danger might be significant over approximately 1,000 years.
    I should live so long.

    Doses this low are probably good for you. If they aren’t – and it is not wildly accepted they are – the other things in your life are probably much more of a threat. Coffee for instance.

  7. So Much For Subtlety

    PaulB – “Johnson’s analysis suggested that the radon would cause 15 excess lung cancer deaths per year nationally, and that it wasn’t worth doing anything about it.”

    And to think they wouldn’t let Project Orion go forward because it might have killed one extra person from cancer across the entire planet. What a travesty of a decision!!

    Orion must rise!

  8. Doses this low are probably good for you. If they aren’t – and it is not wildly accepted they are – the other things in your life are probably much more of a threat.

    Picocuries aren’t a measure of radiation dose, they’re a measure of decay rate (a picocurie is about one decay every 27 seconds). The problem with radon is that its decay creates radioactive dust, which, if lodged in the lungs, is certainly a risk factor for lung cancer.

  9. @ Paul B
    I did read the first few pages before I commented.
    You have missed the point that Radon has a density (in the gaseous form) more than six times that of Oxygen. So there is probably quite a high risk to mice running around the kitchen floor of inhaling it but a very much lower one for humans. His data is Bullshit, he is producing a scare story to pressurise the New York authorities in election year into giving him a contract to investigate something which is not a significant risk.

Leave a Reply

Your email address will not be published. Required fields are marked *