Talk:Radiocarbon dating

Reference 53
Proper bibliographic information is missing for Ref.53 of the current version, "Pasquier-Cardina et al. (1999), pp. 200–201." My guess is, it was meant to cite a paper by Pasquier-Cardin et al., not Pasquier-Cardina; However, unfortunately, I'm unable to access the paper beyond its abstract. Can anybody confirm the content and revise the citation? Thanks. --Deer hunter (talk) 15:17, 14 January 2023 (UTC)
 * Thanks for catching that; that is indeed the right paper. I've fixed the citation; let me know if that looks OK. Mike Christie (talk - contribs -  library) 15:31, 14 January 2023 (UTC)
 * Thanks for your prompt improvement! --Deer hunter (talk) 15:37, 14 January 2023 (UTC)

The origin of the neutron?
The origin of 14C from the nuclear reaction with a neutron is undisputed today. But where do the neutrons come from? Since neutrons decay with a half-life comparable to the transit time of light from the Sun, objects that are much farther away cannot be the explanation. Franz Scheerer aus Wiesbaden (talk) 11:36, 25 April 2023 (UTC)
 * See this section of the article; they are created in the atmosphere by cosmic rays. Mike Christie (talk - contribs - library) 12:17, 25 April 2023 (UTC)
 * I found there:
 * "They originate from the Sun, from outside of the Solar System in our own galaxy, and from distant galaxies."
 * So, I think the neutrons originate from the Sun, because the distance to objects outside of the Solar System or from distant galaxies are much too far away. Franz Scheerer aus Wiesbaden (talk) 14:34, 25 April 2023 (UTC)
 * The cosmic rays originate from those places. They strike the atmosphere (typically Nitrogen-14) and knock off a neutron. MrOllie (talk) 14:42, 25 April 2023 (UTC)
 * If you knock off a neutron from Nitrogen-14 you create Nitrogen-13, which decays to Carbon-13. I don't think, that this was observed on Earth. Franz Scheerer aus Wiesbaden (talk) 10:15, 26 April 2023 (UTC)
 * See Neutron for more details; there are sources cited there which give more information. There are multiple pathways that generate neutrons, including spallation of nitrogen nuclei. Mike Christie (talk - contribs -  library) 10:26, 26 April 2023 (UTC)

Decay graph for C14
I've removed the graph for C14 decay, which was just added to the lead, for a couple of reasons. I think a graph like this is a good idea but I have a couple of questions. (There's a spelling error in the caption, FYI, for when it's re-added.) One problem is I tried to check the source and am getting a security error, so there's apparently a problem (probably temporary) with the website. , if you have access, can you explain what makes that website reliable? A second issue is that I don't think it belongs in the lead -- I think the "Principles" section would be a better fit, since that's where decay is covered. I also wonder if it's necessary to be clear what half-life is used for the graph, since the Libby half-life is used in C14 calculations. Presumably this graph uses 5730 years, which is appropriate for the physical details. And I think the caption needs to be more explicit about what it shows rather than just saying there's a correspondence. Mike Christie (talk - contribs - library) 10:15, 22 May 2023 (UTC)


 * I couldn't access the website either, so I probably withhold from adding the image for now. CactiStaccingCrane (talk) 10:22, 22 May 2023 (UTC)

Lead image
The first half of the article is quite text-heavy. It would be nice to find an engaging image for the lead. There aren't a lot of great options on Commons, though. Maybe File:Mummy tooth sampling.jpg, but I'm not sure having an image of human remains right up front is wise? –&#8239;Joe (talk) 18:12, 22 June 2023 (UTC)
 * I agree it would be nice to have something there. I think a picture of Libby would be ideal but I've been unable to find a free one.  This might be dramatic; having something that has been radiocarbon dated and which is clearly old seems a reasonable option. Mike Christie (talk - contribs -  library) 18:23, 22 June 2023 (UTC)
 * The reflection on the glass is a bit of a shame with that one. Also it's a bit odd using something so amenable to dendro dating... but maybe that's just me. Would the Dead Sea Scrolls be too obvious? –&#8239;Joe (talk) 15:42, 23 June 2023 (UTC)
 * That could work. There's already an image of one of the scrolls in the article; we could move that to the top or perhaps use this one instead? Mike Christie (talk - contribs -  library) 16:01, 23 June 2023 (UTC)
 * I think that's a good idea. Nice to have two. –&#8239;Joe (talk) 15:14, 25 June 2023 (UTC)

Impact
The first sentence of the last paragraph in the article (under the section heading of "Impact" states: "Researchers have studied other radioactive isotopes created by cosmic rays to determine if they could also be used to assist in dating objects of archaeological interest; such isotopes include 3He, 10Be, 21Ne, 26Al, and 36Cl.  Thus, this sentence asserts that Helium-3 and Neon-21 are radioactive.  They are not; they are stable isotopes. 47.205.67.147 (talk) 17:09, 8 July 2023 (UTC)
 * You're right, of course; I've removed "radioactive" from that sentence. I don't think it's the place to go into how a non-radioactive cosmogenic isotope could be used for dating, so I don't think more is needed.  Thanks for pointing out the error. Mike Christie (talk - contribs -  library) 17:59, 8 July 2023 (UTC)

Tree rings
I am reading Jennifer French's 2021 Palaeolithic Europe, which covers the period down to 15,000 years BP, and on page 215 she comments that radiocarbon dates for her period are of inadequate precision to trace some demographic processes because they are before "the more reliable dendrochronology-based part of the 14C calibration curve", which only goes back to ~13,910 cal BP as of 2020 (see ). The article does discuss tree ring calibration and in brackets that it goes back 13,900 years, but unless I have missed the relevant passage I do not think it spells out clearly that radiocarbon dating is more accurate for this period than for earlier ones for which exact proxies are not available. Do you agree and do you think this is worth spelling out? Dudley Miles (talk) 19:36, 24 October 2023 (UTC)
 * You're right that the article doesn't go into this. It would certainly be worth putting something into the radiocarbon calibration article, which is currently just the leftovers from this article.  I would like to add something here too, but I think it's going to be tricky to get it right.  If we're going to discuss how accurate the calibration curve is we shouldn't restrict ourselves to the statement that the dendochronology part is the most accurate; as I understand it the error bars get larger as you go back in time because the smaller amounts of 14C automatically decrease the accuracy.  I would imagine there is also material to be found on the relative accuracy of the varve data and so on.  The articles on the various IntCal curves probably cover this but I'm not sure I have the technical savvy to distill the relevant bits out accurately -- those articles are technical discussions, not survey works. Mike Christie (talk - contribs -  library) 02:01, 25 October 2023 (UTC)
 * I agree that the diminishing accuracy as dates get older should be covered and I have not found a source discussing this, but I do not think this is a reason for not briefly covering dendro dates. There is a good source for this at, which also mentions Miyake events, which are not currently mentioned in the Wiki article. How about "The continuous sequence of tree ring dates goes back to 13,910 BP as of 2020, and this provides close to annual dating for much of the period, reduced where there are calibration plateaus, and increased when short term $14$C spikes due to Miyake events provide additional correlation. Radiocarbon dating earlier than the continuous tree ring sequence relies on correlation with less accurate proxies and has wider error bars." Dudley Miles (talk) 10:03, 25 October 2023 (UTC)

Just looking at the abstract I don't think we can say "close to annual dating for much of the period"; the source only has "for certain time ranges", which could mean less than half of the range. Perhaps just "for part of the period" would work. Or does the body of the cited article support "much of"? Again just looking at the abstract I think we'd also need more support for "wider error bars", but perhaps French's book would provide that if the body of the article doesn't. If you agree, go ahead and make the change -- I'm off to work in a few minutes. Mike Christie (talk - contribs - library) 11:34, 25 October 2023 (UTC)


 * The first link was to the abstract, but the reference was to the full article. Can you see whether you think this supports my wording. Dudley Miles (talk) 12:01, 25 October 2023 (UTC)
 * Will look later today. Mike Christie (talk - contribs - library) 12:07, 25 October 2023 (UTC)
 * , sorry about the delay getting back to you on this. I agree that the article covers this.  For the last phrase of your proposed sentence, "less accurate proxies and has wider error bars", are you relying on just the fact that the tree-ring data is said to have become more accurate, or was there a specific place in the article that supports this?  I think it's OK either way but the latter would be ideal.  Mike Christie (talk - contribs -  library) 10:04, 29 October 2023 (UTC)
 * . I have looked again at the source article. It frequently refers to using dendro dates for the period where they are available and other records (not proxies) for earlier periods. I think this covers your point. There is no reference to error bars. My amended suggestions is "The continuous sequence of tree ring dates for the northern hemisphere goes back to 13,910 BP as of 2020, and this provides close to annual dating for IntCal20 for much of the period, reduced where there are calibration plateaus, and increased when short term $14$C spikes due to Miyake events provide additional correlation. Radiocarbon dating earlier than the continuous tree ring sequence relies on correlation with less accurate records." I think "more approximate" would be better than "less accurate", but I am not sure whether it is good English. What do you think? Dudley Miles (talk) 11:26, 29 October 2023 (UTC)
 * That seems good to me, and I think "more approximate" and "less accurate" are both acceptable; use whichever you prefer. Mike Christie (talk - contribs - library) 11:45, 29 October 2023 (UTC)
 * Looking at the article I think there is an error. It says "The IntCal20 data includes separate curves for the northern and southern hemispheres". The source says "For calibrating atmospheric samples from the Northern Hemisphere, the new curve is called IntCal20. This is accompanied by associated curves SHCal20 for the Southern Hemisphere, and Marine20 for marine samples." So IntCal20 is just for the northern hemisphere, not covering both as the Wiki article says. Do you agree? Dudley Miles (talk) 14:29, 29 October 2023 (UTC)
 * Yes -- I think when I wrote that I thought "IntCal20" was used as a shorthand for the whole 2020 recalibration as well as the NH curve, but even if that's true I think it would be better to rephrase to avoid the ambiguity. Mike Christie (talk - contribs - library) 14:56, 29 October 2023 (UTC)

Recent edits to the lead
One reason I reverted is that the changes led to repetition -- the mention of the Nobel Prize and date range of the dating method are already in the lead. If they're to be moved up, they shouldn't also be mentioned lower down in the lead. Another reason is that the existing lead structure mentions them where they come naturally. Yes, they're important points, but I don't think it's necessary to mention them right at the top -- particularly the Nobel Prize.

Can we please follow the WP:BRD recommendations? Once an edit is reverted, it should be discussed on the talk page rather than redone. This page has other watchers who may have an opinion; let's see what they think of the proposed changes first. Mike Christie (talk - contribs - library) 02:51, 15 November 2023 (UTC)


 * May I suggest, that we keep the year range and the Nobel Prize in the first paragrpah, and remove them from the rest of the Intro.  I strongly feel, that the useful range of the method must be in the very first paragraph ,because the general public  thinks, that early hominid (i.e. Australopithecus) remains were carbon-dated. Walter Tau (talk) 03:04, 15 November 2023 (UTC)


 * Obviously the two repetitions have to be removed somehow. The second long paragraph is a bit convoluted and putting the two mentions of Libby (invention + Nobel) into a single sentence would help. I also suggest removing some material from the second paragraph (the 98% and the "special preparation methods") as I think such fine detail doesn't need to be in the lead. I don't find Walter's reason for mentioning the time period in the first paragraph convincing; the second paragraph is soon enough though it could be written better. Zerotalk 03:18, 15 November 2023 (UTC)
 * I don't think these two details are so important that they need to be mentioned in the first paragraph. Also, Walter Tau's edit introduced a lower bound to the range (ca. in the range between 1,000 and 50,000 years) which is unsourced and incorrect: with bomb curve dating, you can date the cells in your own body, if you want. –&#8239;Joe (talk) 08:40, 15 November 2023 (UTC)