Radiocarbon Calibration Is Stretchy

The latest calibration curve for radiocarbon dating is raising eyebrows. Will it upset what is “known” about the past?

Sometimes “new and improved” is welcome news. For example, the Hubble Space Telescope improved the resolution of faint objects compared to ground-based telescopes tremendously. Without controversy, it has dazzled the world with the beauty of astronomical objects. Other times, though, “new and improved” suggests that what came before was bad. And sometimes one doesn’t want a “new and improved” version of something, like the standard kilogram or meter, because it throws off trusted measurements made prior to the change. Well, now we have a new calibration curve for radiocarbon dating, also called carbon-14 dating. What might be the reaction of scientists and observers to this change?

Notes: Radiocarbon dating is only used for recent samples, not the “millions of years” dates assigned to dinosaurs and other moyboy targets of interest. Because of its relatively short half-life of 5730 years, carbon-14 would all be gone in 100,000 years; practically speaking, 60,000 years is a maximum date a sample could theoretically yield. For this reason, it is mainly used to date archaeological sites, early man sites and fossils that could not be older than a few tens of thousands of years.

Radiocarbon dating is also more complicated than a plug-and-play method. Multiple sources of error can creep in. Scientists have to assume, for instance, that the atmosphere was stable, that incoming radiation did not fluctuate wildly, and that all points on the globe got the same input of radiocarbon. If C-14 concentrates due to ocean currents or other processes, those differences could influence dates in certain environments. Then there are the “unknown unknowns” that bedevil any dating method; do all organisms take in ¹⁴C in identical ratios? Are there seasonal variations, or climate variations? What if, as creationists say, atmospheric conditions changed drastically during a worldwide flood? Error bars tend to get larger backward in time, especially when measuring dates before human records were available to provide corroboration.

Researchers unlock secrets of the past with new international carbon dating standard (University of Sheffield). These UK scientists sound optimistic. “Radiocarbon dating is key for archaeology and environmental science to date everything from the oldest modern human bones to ancient massive volcanic eruptions,” they note. They believe that the new calibration curves, the first in 7 years, “will help scientists build up a more accurate picture of the past.” Of course, the University of Sheffield was glad to be included in the project that involved looking at some 15,000 samples in order to come up with the curves called the “IntCal20” set; they don’t want to sound less than optimistic. But will the new technology used for IntCal20 alter any conclusions based on previously-trusted curves? Why does there even need to be a new calibration?

If the level of atmospheric 14C were constant, this would be easy. However, it has fluctuated significantly throughout history. In order to date organisms precisely scientists need a reliable historical record of its variation to accurately transform 14C measurements into calendar ages. The new IntCal curves provide this link.

The curves are created based on collecting a huge number of archives which store past radiocarbon but can also be dated using another method. Such archives include tree-rings from up to 14,000 years ago, stalagmites found in caves, corals from the sea and cores drilled from lake and ocean sediments. In total, the new curves were based upon almost 15,000 measurements of radiocarbon taken from objects as old as 60,000 years.

The calibration curve does not just pop up out of the ground for scientists to behold. In any dating method, there is inevitably some human judgment involved in deciding which samples are valid. In addition, the other methods used for comparison are not always reliable themselves (see 19 Aug 2016). Scientists have to make judgment calls about which methods and samples are the most trustworthy. And this article from the University of Sheffield does not state explicitly what the team decided to do with anomalous measurements, such as when two dating methods disagree, or an outlier measurement falls off the expected curve (see 19 Jan 2017, “Extraordinary radiocarbon anomaly found in tree rings”). To some degree of error, therefore, the calibration curve is an interpretation of data by a consensus of selected experts who choose to agree on methods and samples. It is not an unbiased conclusion from nature. Let us see what impacts it is having so far.

Manning et al., “Radiocarbon offsets and old world chronology as relevant to Mesopotamia, Egypt, Anatolia and Thera (Santorini).” Nature Scientific Reports 10, 13785 (2020). https://doi.org/10.1038/s41598-020-69287-2.
We learn from this paper that small changes in the calibration can result in big changes in historical interpretations. “While small, the impact of these ¹⁴C offsets can be substantial for Mediterranean and Near Eastern archaeology because of the intricate and densely integrated timeframes involved and the small margins of tolerance,” the eight Cornell authors warn. They also state some of the wiggle room that enters the determination of the curve:

Moreover, where present, apparent seasonal ¹⁴C offsets fluctuate over time, and appear associated with changes in ¹⁴C production and thus likely with variations in solar activity and climate (and ocean systems), and potentially also, therefore, changes in percentage contributions of early and late wood to given tree-rings. These circumstances complicate the elegant hypothesis of a single NH [northern hemisphere] calibration curve, with any variation assumed as effectively comparable with (or incorporated within) error terms.

Take a look at the human judgment calls required to come up with a date:

Even small changes in  ¹⁴C ages can make large calendar differences during reversals and plateaus in the calibration record. There is a narrow distinction between a late seventeenth and earlier-mid sixteenth century BCE date range with IntCal20. Yet this determines the much-debated date of the Thera/Santorini volcanic eruption (Supplementary Discussion 3). Analysis with IntCal20 using (1) weighted average ¹⁴C ages2, (2) a published dataset and alternative appropriate method, or (3) the series of ¹⁴C dates on an olive branch found buried by the Santorini/Thera eruption, all indicate a most likely late seventeenth century BCE date, but include varying probability in the earlier-mid sixteenth century BCE (Supplementary Figs. S10a,b, S11a,b, Fig. 5a). However, if the eruption was coeval with a small positive offset—for example of up to ~ 8 ¹⁴C years (1‰) (see above, Supplementary Discussions 1, 3, Supplementary Fig. S6)—this moves substantial or majority probability from the later 17th to the earlier-mid sixteenth centuries BCE in (1) and (3) (Supplementary Figs. S10c,d, S11c, Fig. 5b).

So even with a “new and improved” calibration curve, there are still bound to be disagreements on dates for things. The more a preferred chronology depends on the outcome, the more the possibility for dispute. One gets the impression that coming up with a firm consensus for a given date is more art than science. “Our examples highlight the need to determine a high-resolution Mediterranean-Near Eastern ¹⁴C record in order to clarify the question of fluctuating small offsets as relevant to regional ¹⁴C levels over time.”

Bard, Heaton et al., “Extended dilation of the radiocarbon time scale between 40,000 and 48,000 y BP and the overlap between Neanderthals and Homo sapiens.” PNAS first published August 17, 2020 https://doi.org/10.1073/pnas.2012307117.
These authors don’t sound happy with the new calibration curve. It affects their favorite hypothesis for the arrival of modern humans in Europe. In the process of discussing it, they mention questionable parts of the curve that yield a “time dilation” difference between radiocarbon age and what they believe was the real age.

The new radiocarbon calibration curve (IntCal20) allows us to calculate the gradient of the relationship between ¹⁴C age and calendar age over the past 55 millennia before the present (55 ka BP). The new gradient curve exhibits a prolonged and prominent maximum between 48 and 40 ka BP during which the radiocarbon clock runs almost twice as fast as it should. This radiocarbon time dilation is due to the increase in the atmospheric ¹⁴C/¹²C ratio caused by the ¹⁴C production rise linked to the transition into the Laschamp geomagnetic excursion centered around 41 ka BP. The major maximum in the gradient from 48 to 40 ka BP is a new feature of the IntCal20 calibration curve, with far-reaching impacts for scientific communities, such as prehistory and paleoclimatology, relying on accurate ages in this time range. To illustrate, we consider the duration of the overlap between Neanderthals and Homo sapiens in Eurasia.

Here we see that the “radiocarbon age” and the “calendar age” are not necessarily the same thing. Without getting into a debate about whose chronology is correct, the take-home lesson is that radiocarbon dating calibration does not absolve scientists from critical thinking.

Additional reading: Browse through recent articles about radiocarbon dating at CEH.

The AAAS has a motto you can get on their T-shirts that states, “Facts Are Facts.” Certainly that must be the case to have any hope of objective truth, but the problem is that people are biased and will resist facts they don’t like. Look how Democrats and Republicans can have the same facts in front of their faces and yet arrive at opposite conclusions: depending on your party affiliation, you probably have heard that Trump’s response to the coronavirus pandemic was either awful or exemplary. One would hope that the further removed from politics a scientific question like radiocarbon dating gets, the less bias there is. These examples show that (1) there’s ample room for human judgment in arriving at calibration, and (2) scientists are not eager to give up their preferred chronologies.

Keep these papers in mind when someone pretends that “facts are facts” about radiocarbon dates.