Exploring the wonders of geology in response to young-Earth claims...

Never been here? Please read my guidelines and background posts before proceeding!

Saturday, November 13, 2010

Radiocarbon evidence for the antiquity of the Earth

While a number of Answers in Genesis (AiG) articles related to radiometric dating have focused on discordant ages obtained from igneous suites (such as K/Ar dates obtained from volcanic flows, see last post), I have found that the most intriguing claims deal with the radiocarbon, or 14-C, dating method. The reason is that AiG authors do not simply try and persuade their readers to discount this method as wholly unreliable (even when the ages obtained exceed 10,000 years) but actually present the results as positive evidence for a young Earth. Here, my goal is to take a more careful look at the conclusions put forth by various AiG articles with regard to both anomalous ages obtained by the 14-C method (such as from coal, diamonds, and permineralized wood) and typical ages obtained from latest-Pleistocene (~80,000-12,000 years B.P.) organic samples. In doing so, I will consider their use of sources (from scientific literature), their understanding of the method itself, and the assumptions that go into their reasoning for why these ages (commonly more than 10,000 years B.P.) are positive evidence for a young Earth (less than 10,000 years old).

Why the focus on 14-C?

When the average person hears the term "radiometric dating", he/she most often recalls the radiocarbon method, even though this method is relatively unimportant to most research in geology. However, this is not out of ignorance on anyone's part. The radiocarbon method is often used as a starting point for understanding radiometric dating techniques, especially in classes unrelated to geology, because: 1) most are familiar with Carbon, as opposed to elements like Osmium, Neodymium, Rubidium, Thorium, etc.; 2) the technique is relatively easy to understand (try explaining a U-Pb concordia over dinner, if you don't believe me); and 3) the method is used in historical studies, such as the dating of artifacts or trees, which can be confirmed by more 'tangible' witnesses like tree ring counts (Sakurai et al., 2004). The third point is most relevant to our discussion, since it results in 'both sides' affirming the accuracy of radiocarbon dating for any 'recent' samples (as opposed to nearly any other method, which must be discounted in all cases by anyone that believes in a young Earth). Thus even from a 'young-Earth' standpoint, all radiocarbon dates (assuming that care is taken to eliminate contamination) are taken to be meaningful indicators of a given sample's age.

Hasn't the issue already been settled?

Anyone familiar with typical studies employing the radiocarbon method knows that model ages obtained often exceed 10,000 years (e.g. Hogg et al., 2006). So doesn't the method already affirm that the Earth (or at least it's now deceased inhabitants) must be at least this old? AiG author Mike Riddle addresses this very question in an article entitled Doesn't Carbon-14 Dating Disprove the Bible? The article begins with a simplified explanation of the radiocarbon method. While his synopsis includes a number of minor factual errors (see below), I would recommend it to anyone not entirely familiar with the method at this point. At the end of his explanation, he states:

"Since no one was there to measure the amount of 14C when a creature died, scientists need to find a method to determine how much 14C has decayed."

This is a valid point: if we don't know how much 14-C was present in the sample to begin with, our age estimate based on the remaining 14-C will simply be wrong. Thus a large portion of research in radiocarbon dating has dealt with that very issue. Riddle then notes:

"To do this, scientists use the main isotope of carbon, called carbon-12 (12C). Because 12C is a stable isotope of carbon, it will remain constant; however, the amount of 14C will decrease after a creature dies. All living things take in carbon (14C and 12C) from eating and breathing. Therefore, the ratio of 14C to 12C in living creatures will be the same as in the atmosphere. This ratio turns out to be about one 14C atom for every 1 trillion 12C atoms. Scientists can use this ratio to help determine the starting amount of 14C."

A couple of factual errors are found in this quote (it is apparent from this point on that much of this article is taken from second-hand knowledge of the issue). First, carbon is taken in from the atmosphere primarily through photosynthesis (not breathing), which is passed down through the food chain (eating). Thus the 14/12C ratio found in anything other than primary producers is inherited from the constituents of its diet. Secondly, although photosynthetic organisms (trees, grass, algae, etc.) take in all isotopes of carbon, it is not at the same rate. There is a simple kinetic discrimination, due to the minor difference in mass between each isotope, that causes 12-C to be taken in 'preferentially' (i.e. 12-C does not remain constant). Thus the ratio of 14-Carbon to 12-Carbon in any given plant material (and, consequently, anything that has eaten that plant material) will always be less than that of the atmosphere. Of course, researchers have always known about this phenomenon, which is why δ13C values are reported along with Δ14C — the original isotopic fractionation must be normalized before an actual age is calculated.

Getting lost in the technical jargon — a minor detour

I would like to make it clear that my purpose here is not to cloud the issue by introducing meaningless complexities to the discussion. I fully understand that the author is writing to a general audience, and bound to make simplifications. I have no problem with this and I believe he does a great job illustrating the basics of the radiocarbon method. However, it is one thing for an expert in his/her respective field to give a "dumbed down" illustration to an audience unfamiliar with the subject, for the sake of helping them to understand. It is quite another thing to make an argument against that simplified explanation without fully understanding or appreciating the true complexity of the issue. This attack is called a strawman argument, and abounds anytime a complex issue is debated by those only vaguely familiar (one need only watch 5 minutes of any politically driven show — liberal or conservative — to see my point. Personally, I think the best example can be seen on The Colbert Report, where strawman accusations are regularly used, although in this case it is intentional and to his own humorous advantage!).

So how do we know the original Δ14C again?

One can start to see the basis of the young Earth argument — it lies with that "critical assumption" that we can know the original ratio of 14/12-Carbon in the atmosphere when the organic material was still alive. Riddle gives his own version here of how that is calculated:

"Dr. Willard Libby, the founder of the carbon-14 dating method, assumed this ratio [14/12-Carbon] to be constant. His reasoning was based on a belief in evolution, which assumes the earth must be billions of years old...In [his] original work, he noted that the atmosphere did not appear to be in equilibrium. This was a troubling idea for Dr. Libby since he believed the world was billions of years old and enough time had passed to achieve equilibrium. Dr. Libby’s calculations showed that if the earth started with no 14C in the atmosphere, it would take up to 30,000 years to build up to a steady state (equilibrium)...What does this mean? If it takes about 30,000 years to reach equilibrium and 14C is still out of equilibrium, then maybe the earth is not very old."

Fortunately, Dr. Libby's book (Libby, 1955) is available in any university library and can be read by any of you interested (I only read it because it was referenced in Riddle's article, but now will say the work is nothing short of genius). One thing is very clear: nowhere did Dr. Libby simply assume the 14/12-Carbon ratio to be constant. On the contrary, he devoted ample discussion to why that ratio would always be in flux (e.g. it depends on the ever-changing strength of the geomagnetic field). He did make the point that one might expect this ratio to be constant when averaged over the last 10-20,000 years, but it is a far stretch to accuse him of making a blind assumption based on a "belief in evolution", which actually has nothing to do with his work (such a reference is an ad hominem argument buried within a Red Herring, as it calls on the audience to reject Dr. Libby's conclusions based on unrelated beliefs). Since the strength of the geomagnetic field, atmospheric composition, etc. could be calculated for the last 8,000 years (or just consider the last 4,000 years if you're not comfortable with that number), Dr. Libby was able to test his predictions rigorously against samples of known age (like tree rings, or a 2nd century copy of a Biblical text). Since that time, a detailed record of 14/12-C ratio has been constructed for the past 50,000 years (Hughen et al., 2004) and is used by any radiocarbon laboratory.

In the quote above, Riddle also makes the argument that perhaps the Earth is less than 30,000 years old, since the 14/12-C ratio is 'not yet' in equilibrium. As pointed out, it would never be expected to reach true equilibrium. (In fact, part of the reason this was true in Dr. Libby's time is that atomic-bomb testing introduced massive quantities of 14-C into the atmosphere, throwing off any equilibrium that would have been reached.) Nonetheless, I am not sure that Riddle realizes the full implications of his argument. According to the young Earth model, how much 14-C was in the original, pre-Flood atmosphere? How much at the time of Creation, and how much needed to accumulate up to and after the Flood? If the accumulation rate were not substantially higher in the past (I would contend that the young Earth model predicts it to be lower), has enough time passed to produce the amount of 14-C in the modern atmosphere? In order to address these questions, we need to know something about the young Earth position.

How does Answers in Genesis interpret radiocarbon dates?

Riddle devotes the rest of the article to a discussion on the young Earth interpretation of the carbon cycle, accumulation of 14-C in the atmosphere, and how these issues effect calculated radiocarbon ages. He first notes that a much stronger magnetic field in the past would have resulted in a much lower production rate of 14-C in the atmosphere, which is true. Secondly, he notes that if all organisms now buried in geological strata were alive at the onset of the flood, the carbon reservoir of the biosphere would have been significantly diluted, resulting in much lower 14/12-C ratios. This is also true. Thus he concludes:

"When the Flood is taken into account along with the decay of the magnetic field, it is reasonable to believe that the assumption of equilibrium is a false assumption. Because of this false assumption, any age estimates using 14C prior to the Flood will give much older dates than the true age. Pre-Flood material would be dated at perhaps ten times the true age."

John Woodmorappe complements the paradigm here, suggesting that volcanism associated with the flood may have added CO2 gas devoid of any 14C to the atmosphere, resulting in highly inflated ages of any organic samples from directly after the flood. This effect can also be seen, for example, in the radiocarbon dating of artifacts made from shell material (e.g. jewelry). The age can be inflated for two reasons: 1) the shell may have already been old when utilized by the person; 2) shells are produced from the dissolved inorganic carbon (DIC) in the ocean, lake, or river in which the organism lived, which is a mixture of atmospheric CO2 and dissolved carbonate from ancient rocks that contain no 14C.

I would summarize the young Earth model like this:

1) The original atmosphere would have contained little to no 14C; furthermore, the production rate of 14C in the atmosphere would have been significantly lower than today, due to a much higher strength geomagnetic field.

2) During the Flood, and shortly after, the strength of the magnetic field rapidly decayed, leading to a sharp increase in the production of 14C in the atmosphere. Therefore, 14C has been accumulating to this day, and the production rate is increasing.

3) Since the vegetation and other organisms now preserved as coal, kerogen, fossils, etc. must have been living near the onset of the flood, the pre-Flood biosphere must have been much larger (up to 500 times) than today. Any model of the 14/12-C ratio must account for this massive burial of carbon, which would have greatly enriched the previously diluted biosphere/atmosphere with 14C.

4) Any plant matter formed within a few hundred hundred years after the Flood would be depleted in 14C, due to the large amount of volcanic gas (which contains no 14C) released to the atmosphere from the Flood. The degree of depletion would decrease over time as the system equilibrated toward modern values.

5) Therefore, any organic material buried in the Flood (i.e. fossil wood, coal) should give an "age" of 50,000 years or more; organic material produced in the years that followed the flood should also give anomalously old dates; the more recent the sample, however, the more accurate the calculated age, which explains the practical use of the method in archaeology.

Stressing the Carbon Cycle

While nobody can accuse Answers in Genesis of ignoring the issue of radiocarbon dating, this is hardly a case of "you have your model, we have ours; since we start with different assumptions, however, we will have differing but equally valid interpretations of the same data" as is often purported by AiG. At first, the AiG model appears to offer a consistent explanation that would obviously be ignored or discredited by secular science. However, there are critical assumptions made with regard to the carbon cycle that are completely out of touch with reality. Since the carbon cycle is by itself a complex issue, I will only comment briefly here before discussing the positive evidence for a young Earth from radiocarbon dating.

The idea that the pre-Flood biosphere could have been 500 times larger than today dose not come without consequence. Currently, the amount of carbon dioxide and oxygen in the atmosphere is roughly balanced by a number of fluxes. For example, dissolved CO2 in the surface water of the ocean represents a mixture of CO2 from the atmosphere, and the oxidation of organic matter (i.e. decay). This is balanced by the escape of gaseous CO2 to the atmosphere, and photosynthetic production. However, even if these fluxes were balanced before the Flood, significantly different reservoir sizes, such as a 500-fold increase in the marine biota, would have noticeable effects on the isotopic composition of the ocean (in this case, it would raise the δ13C value of marine carbonates substantially, which is not observed in the geologic record). If Woodmorappe were correct concerning the inflation of radiocarbon ages due to a large flux of volcanic gases during the flood, we could also make predictions concerning the global change in both δ13C and δ34S (stable isotopes in carbon and sulfur) immediately following the flood. In other words, the model is testable by methods other than radiocarbon, but can not hold up to consistent scrutiny between fields. Furthermore, calcite precipitated as cement in sediments should retain 14C in equilibrium with the atmosphere. Thus one could predict that in the young Earth model, meteoric calcite cements would give reasonable estimates of the 14/12-C ratio in the pre-Flood atmosphere, and the radiocarbon "ages" of calcite cements should be broadly consistent between each other. However, no studies have shown this to be the case (nor could they).

Before anyone accepts AiG's reasoning behind their interpretation of radiocarbon ages, they should consider whether AiG could offer an internally consistent model of the carbon cycle before, during, and after the Flood, which can explain a wide range of phenomena (and not simply levels of 14C in the atmosphere).

Back to the evidence

Riddle cites evidence produced by the RATE team to confirm the young Earth model concerning radiocarbon age estimates, all of which has been discussed in more detail by various authors in other AiG articles. A closer look, however, shows that these studies rather provide more evidence for the antiquity of the Earth.

In an article entitled Radiocarbon in Diamonds Confirmed, Dr. Snelling reviews a study by Taylor and Southon (2007) that reports radiocarbon ages of Paleozoic diamonds around 70,000 years B.P. Referencing the study, he states: "Confirmation that there is in situ carbon-14 in diamonds has now been reported in the conventional literature." Such a find would be surprising to the field of radiocarbon dating, given that these diamonds are hundreds of millions of years old and should not contain any radioactive carbon. In fact, it could rather prove that the Paleozoic era was not so distant after all, thereby substantiating the young Earth model. However, Dr. Snelling's reference is misleading, to say the least. The purpose of the diamond study was not to demonstrate the moot point that 500-million year old samples should not contain any 14C, but rather to measure the background values of the AMS (Accelerated Mass Spectrometer) instrumentation.

But what is a background value? In short, it represents anything that would cause the machine to 'think' that intrinsic 14C is being measured. Taylor and Southon (2007) discuss the potential sources of false 14C thoroughly in their article. The most obvious would be contamination. If the sample were handled by human hands that transferred modern carbon, it would appear younger. If CO2 from the lab atmosphere leaked into the sampling tube, it would also produce a falsely young age (anyone that has worked with instruments require a vacuum to take measurements knows that this problem is the rule, not the exception). Other sources are not so obvious. Atmospheric CO2 (or mineralized carbon during washing) can adhere on to the sampling tube or even the sample itself, only to be released during analysis. But my personal favorite is the ionized molecular carbon. Many samples contain hydrogen in addition to carbon. Mass spectrometers work by ionizing the carbon and running it through a magnetic field. However, the machine will occasionally ionize a carbon that is attached to a hydrogen. Therefore, if a 13C atom is attached to a hydrogen and they are ionized as a molecule, the machine will think it is measuring 14C, when in fact it is not.

Of course, all of these effects (I've only mentioned a few) are rather negligible, producing only minor amounts of 'false' 14C. Nonetheless, a radiocarbon age of ~70,000 years represents only 1 14C atom for every 4 quadrillion atoms of 12C (can you appreciate modern technology at this point?). I understand that Dr. Snelling is eager to conclude these background values actually represent intrinsic 14C, but there is simply no reason to believe this, and good reason to reject it. First of all, Taylor and Southon (2007) do not blame the readings on background values arbitrarily — there is a solid, physical basis for their findings that does not require them to believe these diamonds naturally contained 14C in their mineral structure. Secondly, I mentioned earlier that δ13C values were always reported with radiocarbon ages to correct for the kinetic fractionation of carbon isotopes. Did Taylor and Southon report these values? Yes they did, and their results are profound. Reported δ13C values in diamond faces varied from -23.1 to 4.2. This means that diamonds are not isotopically homogenous. In other words, the amount of 14C in the diamond structure should also vary substantially from sample to sample, but it doesn't. The radiocarbon "age" is relatively constant between diamond faces, which means the age actually does represent background values, and not intrinsic 14C. Unfortunately, Dr. Snelling ignores this fact, which may indicate that he is not entirely familiar with the lab procedure. To elucidate:

"Yet this begs the question as to why then did the Precambrian graphite contain on average more carbon-14 to yield younger ages than the diamonds? And why did the diamonds have such different carbon-14 contents to yield different apparent radiocarbon 'ages'? Because the same instrument was used to analyze all the diamonds and the graphite, the results should surely have all been affected by the same 'machine background.'”

The ages between diamond samples varied slightly, but cut faces of any given diamond were remarkably consistent. Nonetheless, background values arise from an interaction between the sample and the machine itself. Thus it will vary between different samples, and especially between different types of samples. This is why researchers use substances that are chemically similar to the sample they actually want to date when measuring background values.

In case one is still convinced that these diamonds did contain intrinsic 14C, however, I must pose this question: why should natural diamonds contain any 14C, ever? Diamonds are formed deep in the mantle, far removed from the atmosphere where 14C is actually produced. To suggest that radioactive diamonds are evidence for a young Earth requires an intentional ignorance, or downright dishonesty on the part of AiG.

Playing games with the evidence

A number of other articles are available at AiG concerning radiocarbon ages obtained from petrified wood and coal. I will not make an effort to discuss them all in detail, but rather challenge you to seriously study any claim made within. This requires checking the sources (were they used properly?), and testing the consistency of the argumentation (are the assumptions of the argument valid in all fields of science?). For example, when someone reports that 20-million year old fossil wood, which has spent at least a few thousand years near the surface of the Earth (where it was subject to groundwater, microbial activity, and more) was sent to a lab that gave it an age of 33,000 years, what is more likely? That the plethora of scientific evidence regarding the antiquity of the Earth has all been misunderstood? Or that a tiny, negligible fraction of the sample is actually comprised of a recent carbonate mineral (or even humic acid, which has the same δ13C value as wood and bonds strongly to metal cations) that has precipitated in the pore spaces during weathering processes and was not removed during the washing process? Yes, labs take all precautions to account for contamination, but when the lab is intentionally not told that the sample is from a coal bed or volcanic flow, they can not properly rule out all forms of contamination or machine background values.

Concluding remarks

Vast amounts of research in radiocarbon dating are available in the scientific literature. While the method is admittedly susceptible to a range of analytical errors, it has proven accurate under rigorous testing and in some very important studies (such as Hughen et al., 2004, which dated ice cores layer by layer and obtained consistent ages for 50,000 annual layers) that actually help prove the antiquity of the Earth. While AiG has made a noble effort to reinterpret the evidence out of a desire to be consistent (and I understand that desire), their models simply can not account for the evidence from the radiocarbon method at this time. The radiocarbon method is an elegant tool that has provided much useful information over the past 60 years, which should not be discredited without a definitive argument. So the challenge remains.

References cited

Hogg, A.G., Fifield, L.K., Turney, C.S.M., Palmer, J.G., Galbraith, R., Baillie, M.G.K., 2006, Dating ancient wood by high-sensitivity liquid scintillation counting and accelerator mass spectrometry—Pushing the boundaries: Quaternary Geochronology, v. 1, p. 241-248.

Hughen, K., Lehman, S., Southon, J., Overpeck, J., Marchal, O., Herring, C., Turnbull, J., 2004, 14C Activity and Global Carbon Cycle Changes over the Past 50,000 Years: Science, v. 303, p. 202-207.

Libby, W. F., 1955, Radiocarbon dating: University of Chicago Press, 2nd ed.

Sakurai, H., Gandou, T., Kato, W., Sawaki, Y., Matsumoto, T., Aoki, T., Matsuzaki, H., Gunji, S., Tokanai, F., 2004, AMS measurement of C-14 concentration in a single-year ring of a 2500-yr-old tree: Nuclear Instruments and Methods in Physics Research B, v. 223-224, p. 371-375.

Taylor, R.E., and Southon, J., 2007, Use of natural diamonds to monitor 14C AMS instrument backgrounds: Nuclear Instruments and Methods in Physics Research B, v. 259, p. 282-287.


  1. I think you missed the point in Taylor/Southon. If the diamonds have measurable differences between them, but each face *doesn't* have measurable differences between them, then the problem is *not* machine background. Machine background would have the same statistical pattern on both the individual cuts and between diamonds. The background does not know that the different cuts came from the same diamond. The only reason that they would be equivalent only on faces from the same diamond is if they are really measuring C14.

  2. Is there a particular section of their article to which you are referring?

    Having worked with mass spectrometers, I can say this statement is simply not true: "Machine background would have the same statistical pattern on both the individual cuts and between diamonds." Else why did they measure individual faces of the same diamond? In any case, I admit I was not entirely clear, since I included analytical noise in this category of 'background' (atmospheric contamination, variable vacuum levels, interaction between beam/sample due to variable angle, etc.). But these factors are bound to vary slightly between samples, and AMS radiocarbon equipment is already amplified to the point that significant errors are to be expected when model ages approach and exceed 50 kyrs. Hence no geochronologist accepts the idea that actual 14C is being measured in these diamonds.

    In case you think I am wrong here, can you explain the more pertinent problem—namely that there is no correlation between δ13C and 14C activities? If 14C levels were measuring actual 14C, they should covary with 13C. But they don't. Therefore, the machine is not measuring intrinsic 14C.

  3. Just found the blog, appreciating it so far. Just one note, as a lay person, delta notation for isotope ratios is not familiar to me, and tracking it down in wikipedia took some searching (first to recognize a lower case delta, and then to figure out what it was referring to). You might want to provide an explanation for the notation to make your intent easier to understand.

    1. Thank you for the heads up! This article was one of the first that I wrote, so admittedly I less weary of how to mix technical jargon into popular articles. ;)

  4. Love it. Well done and very gracious. Nice to see a balanced view where someone sites the evidence against YEC claims. What about your views on Genesis and Science?

    1. Thank you! You can search more easily through my articles at the new site. If you're interested in Genesis and Science, maybe start with this category:


  5. With RentalCars you can discover the cheapest car rental from over 49,000 locations worldwide.