Jump to content

Talk:Banana equivalent dose/Archive 2

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 1Archive 2

Apparent Original research vs. WP:V and WP:DUE

This comment was sparked by a talk page discussion at Talk:Sievert#Possibly Incorrectly Quoted Value for Banana Equivalent Dose in Sv (discrepancy of one hundred times). An editor in that article apparently relied on info in this article (this article, of course, not being a reliable source). The Calculation section of this article appears to contain in original research unsupported by cited supporting sources in calculating and presenting a value for the BED of approximately 0.0001 μSv per hour. This in spite of a number of reliable sources which assert that one sievert is equal to 100 rem (or thereabouts — I have seen sources which present different figures, apparently growing out of differences in the sizes of bananas on which calculations are based. The values I've seen in other sources, however, were well within one order of magnitude of 100 rem.)(One such source is http://www.ehs.unr.edu/Documents/RadSafety.pdf, which was cited in some earlier versions of the article. That source asserts that The dose equivalent of a banana, if eaten, is about 0.01 mrem. Other sources (e.g., http://www.epa.gov/rpdweb00/docs/402-k-07-006.pdf, which is currently cited in this article, would support an assertion that one sievert is equal to 100 rem. One BED, therefore, would be by simple calculation (a calculation which I flubbed in an earlier version of this article) 0.1 microsieverts (not "0.0001 μSv per hour" — one wonders about the "per hour" there — it seems to me that a person eating 1 banana per hour would absorb an equivalent dose of 100 rem (0.1 microsievert) per hour from each hourly-eaten banana.).

Original research in the calculation section should be eliminated. The value of a BSD asserted by this article should be supported by cited verifiable reliable sources. If reliable sources disagree, due weight considerations apply.

I would have boldly corrected this myself by rewriting or eliminating the Calculation section, but at the moment I do not have the time to engage in the discussion which would ensue from that. Someone, please correct this. Wtmitchell (talk) (earlier Boracay Bill) 09:10, 28 March 2011 (UTC)

I agree the calc section is poor quality. Although I think the values are "roughy" OK and more reasonable than the old value, the derivation is flawed and does need solid refs or deletion.
The RadSafety pdf that you cite does not derive the value, it just produces it from nowhere without any justification, explanation or citation. I am pretty sure that they just copied from the RadSafe ML I cite in this article. This has , sadly , become something of an urban myth amongst may people with a pro-nuclear bias and seems to be repeated without question or verification.
I have contacted U. Nevada about the accuracy of this value since they still give prominence to that document. As yet they have not replied. Whether that is through embarrassment or the need to verify or simple lack of time remains to be seen. I will chase them up if they don't reply.
In any case it states the same value as used on the ML which I have expert comment indicating it is a) not legit use of K-40 dose and b) ignores homeostasis and is wrong, big time.
In the absence of a correct value I think this article cannot do better than cover the idea, inaccuracies etc without proposing a value. Climatedragon (talk) 10:08, 28 March 2011 (UTC)
Re "I think the values are 'roughly' OK ..." above, please note the lead paragraph of Wikipedia's verifiability policy:

The threshold for inclusion in Wikipedia is verifiability, not truth: whether readers can check that material in Wikipedia has already been published by a reliable source, not whether editors think it is true.

I've provided links above to reliable sources which appear to support 0.1 μSv as a value for BED. Other sources I've seen support values in the same order of magnitude. The unsupported figure of 0.001 μSv is out of whack with figures I've seen published in citeable sources. An unsupported figure should not be presented -- particularly an unsupported figure which would conflict with figures from reliable source if those figures were presented. If figures published by apparently reliable sources differ significantly, due weight considerations apply. Unsupported figures calculated by WP editors have zero weight.
http://www.nv.doe.gov/nssab/Documents/Handouts/FY%202007/TW/07FY07%20and%2011FY07.pdf supports a 10 mrem (0.1 μSv) figure, though only vaguely.
I see that chart 23 of http://kcc.ks.gov/nuclear/2008_KS_nuclear_waste.ppt says that a 4 mrem of exposure is equivalent to eating a banana a day for a year (4 mrem = .04 msV = 40 μSv; 40 μSv/365 = 0.1095 μSv per day as the daily dose from eating one banana per day).
Those sources are in the .gov domain. Other sources probably exist. Wtmitchell (talk) (earlier Boracay Bill) 14:31, 28 March 2011 (UTC)
I contacted the author of that pdf. His reply was 'I performed this exposure estimate of K-40 in a banana about 15 years ago. The method of K-40 dose estimate followed a simplified methodology and was made without potassium metabolism. The only purpose of this calculation was to use it internally for RS training of radiation workers at this university.'
Further question asking if this meant it was an error that this was publicly available and whether a value based on rapid elimination of K would be more reasonable remains unanswered.
Reading between the lines I believe he did not do the calc himself but lifted it from the ML post cited here from "about 15 years ago". ;) Climatedragon (talk) 14:29, 2 April 2011 (UTC)



This isn't a simple conflict between truth and verifiability, rather this is a case where competing sources present a distribution of values and the truth is probably somewhere in the middle. (Some sources gave the high value of 0.1uSv, while the UK Authority source gives the low value of absolutely zero Sv. The emphasised middle result of 1nSv falls within this range of cited values.)

Much of the calculation is not original research, rather it is to explain nuclear physics calculations that are too trivial to normally publish. And much of it is indeed already comparatively very well backed up with external references.

The part of the calculation for radiation power density should not be too difficult to find direct citations for, because it is such a fundamental and widely used result. For example, potassium's "decay heat" is the main cause for the current warmth of the planet's mantle. (That's probably notable enough to warrant an inline mention in this article.)

I think it would be interesting to cite something about ingestion of pure K-40, in a sentence to explain why differing isotopic ratios are negligible here.

For now perhaps it would be better if example calculations were continued for a few different exposure periods, namely: zero time, 8 hours, and 50 years. Hopefully before too long we (or Climatedragon?) can shame expert to present unambiguously superseding sources.

By the way, the UK Authority quote elsewhere in the article needs attention, since it seems to actually be a paraphrasing. Cesiumfrog (talk) 02:11, 29 March 2011 (UTC)

Re the assertion above, "This isn't a simple conflict between truth and verifiability, rather this is a case where competing sources present a distribution of values and the truth is probably somewhere in the middle", please see WP:DUE, which says in part, "Neutrality requires that each article or other page in the mainspace fairly represents all significant viewpoints that have been published by reliable sources, in proportion to the prominence of each viewpoint." It is not the place of WP editors to try to identify where they perceive truth among the different published viewpoints (note the lead paragraph of WP:V), and attempting to do so flouts WP:NPOV. Wtmitchell (talk) (earlier Boracay Bill) 10:46, 29 March 2011 (UTC)
There is very little about the topic of this article in reliable sources, and probably nothing that does not require careful interpretation. So far the climate on this talk page is halfway between cooperation and a battle field. If it gets much worse there is probably no alternative to turning the article into a redirect to another article that mentions the idea without giving any numbers.
I am very much surprised by just how opaque all our sources are as to where they get their initial numbers from and/or what model they used for their calculations. Most surprising is that the printed technical literature that I have seen so far is no better in this respect than the blog posts. Altogether I am beginning to have serious doubts about the notability of the topic outsided the blogosphere. Hans Adler 14:46, 2 April 2011 (UTC)

Cleanup

I have revised the whole article, removing the quadruplication of text noted above and some overly pedagogical (non-encyclopedic) details. (I suppose that the the BED concept has been redefined and used by MANY people in MANY contexts over the last 15 years. So it does not seem necessary to mention explicitly Fukushima, xkcd, and slashdot; it suffices to give refs to the last two as mere examples of those many.)

For the criticism section, there seems to be uncertainty as to what is the reequilibrium time for K in the body. That is the time the body takes to restore the "normal" K budget after one ingests food that is abnormally rich in K. Some comments above say 30 days; but that may be the turnover time, i.e. the time needed for half of the K atoms in the body to be replaced by fresh ones, given a normal diet. These two parameters are quite different. Consider for example a water tank with an overflow drain. If one drops a bucket of water into the tank, another bucket of water will come out from the drain almost instantly, so the reequilibrium time is on the order of seconds. On the other hand the turnover time depends on the size of the tank, how much it is stirred, and on the rate at which one pours water into it, and can be arbitrarily long - even infinite.

One way to put the BED concept in perspective (perhaps not for the article, but just for this talk page) is to note that the human body contains a certain concentration of potassium that is pretty constant over the lifetime. As long as one ingests the minimum amount of K needed to replace the inevitable losses, the average amount of K in the body is little affected by eating more or fewer bananas. As long as the isotope composition of the diet is the natural one, the amount of 40K in the body will be constant too. In adults, that 40K generates about 5000 Bq of activity, giving a total equivalent absorbed dose of 5000 x 5.02 = 25000 nanosieverts = 25 microsieverts over 50 years, or 0.05 microsieverts per year, or 0.00005 microsieverts per hour. Again, this lifetime dose is largely independent of how many bananas or brazilnuts one eats. (Potassium-40 that is still in the digestive system or in the bladder should also have little impact since only the amount that is within a few mm of the organ's walls will be able to irradiate them.) Jorge Stolfi (talk) 04:27, 8 June 2011 (UTC)

  • I have read a bit more and it seems that the problems with this article are worse than I thought. For starters, it must be noted that the dose in sievert is radiation absorbed per kilogram of tissue, not total per person; so the conversion between BEDs and sieverts depends on body weight. Also the value "50 years" assumed in EPA tables is relevant only for isotopes that remain in the body for deades; it does not apply for elements with short half-lifes in the body, such as potassium. Jorge Stolfi (talk) 16:39, 8 June 2011 (UTC)

Jorge, I appreciate your efforts, but to anyone with some radiation protection background there are a large numbers of technical problems in this article. For example "This potassium will naturally generate 175 g × 31 Bq/g ≈ 5400 Bq of radiation". Now Bq are NOT units of "radiation" they are units of the amount of radioactive material present (technically "activity"). Radiation is what is given off by radioactive materialand is measured in units of grays or sieverts. Then there is a good deal of confusion over dose - absorbed, equivalent and effective. The BED is (if it is anything) a unit of "effective dose" (roughly a unit that gives an indication of the effect of that dose of radiation in terms of the probablity that it will produce cancer) but in various places it is described as "a radiation dose equivalent" unit (effective dose), and "the absorbed radiation dose" (absorbed dose). In the sentence "The BED is supposed to be a radiation dose equivalent unit; that is, a unit for measuring potentially damaging radiation absorbed by body tissue, rather than the total radiation (of any kind) emitted by a source or absorbed by matter" there are at least three errors - equivalent dose as above, it is assumed that ALL (ionising) radiation absorbed by the body is "potentially damaging" and there is no unit (or even concenpt) for "the total radiation ... emitted by a source".

In response to your last comment, yes "equivalent dose" (in Sv) is "per kg", but "effective dose" (also in Sv) is not. Equivalent dose is usually measured only in one organ or tissue, but effective dose is integrated over the whole body. Very confusing I know, I've had 30 yrs in the business and I still get tangled up sometimes.

Bottom line (in my opinion) There is very little benefit in going into much technical detail in this article. I think it starts out very well: BED is a whimsical unit of dose. Then go on to say that conventionally it is taken as about 0.1microSv, but this is very debatable, and is probably nearer to zero (EPA ref). Full stop. If people want to know the details of dose calculations, this is not the place. Baska436 (talk) 02:30, 19 August 2011 (UTC)

Confusing

As it stands, this article is extremely confusing. The lede explains units for radioactivity; should this really be in the lede??? The lede explicitly states that the BED is equivalent to units of microsieverts. The article actually does list a value in microsieverts, but then quickly says that it's wrong. When the article finally wanders around to listing what a BED is, lists it in Bq-- units which are not explained anywhere in the article. Mostly, the article seems to be a (somewhat confusing) essay attempting to make the point that the "BED" is a poorly thought-out unit and shouldn't be used-- and goes to great length to avoid ever saying what it actually is, using the units that were defined in the beginning of the article.. Because of the poor citations, much of the article reads as if it were OR-- the main citations for the criticisms seems to be discussions on an internet mailing list and blog posts. It would be good to turn some of these into WP:RS references. Geoffrey.landis (talk) 19:52, 20 April 2011 (UTC)

OK, I've been trying to clear this up, and I think I've now made it less confusing. I moved the discussion of SI units out of the lede, since they are out of place there, and made the Bq to Sv conversion explicit. I've now just finished pulling out the criticism sections and putting them together into a separate subheading, so that discussion of "What is the BED" now comes before the criticism, "Why the BED is not an accurate way of thinking about radiation dose."
The criticism section now seems to say the same thing four different times, but I was trying to not delete text. Somebody might want to clean it up a bit more, but I think it's now at least clear what the ideas are. Geoffrey.landis (talk) 19:56, 22 April 2011 (UTC)

So from the figures mentioned, a banana adds around 15 Bq verses 5400 Bq for human body tissue overall (based on Potassium 40 decay alone). So this is only 0.28% does this get it into better perspective and is it worth mentioning ? --Andromedean (talk) 23:05, 1 January 2013 (UTC)