Fossil fuel emissions threaten to reduce radiocarbon dating reliability

Radiocarbon dating is one of the great tools of science that has allowed archeologists to shed new light on everything from the building of Stonehenge to the beginnings of international trade. However, a new study from the Imperial College London suggests that fossil fuel carbon emissions may be so diluting radioactive carbon isotopes that within decades it will difficult to differentiate between modern artifacts and those over a thousand years old.

It may conjure up a very odd mental picture, but every living thing on Earth has its own internal clock that's ready to start ticking the moment it dies. And that's what the American physical chemist Willard Libby won the Nobel Prize for when he discovered radiocarbon dating in the late 1940s.

It's based on the very simple principle that radioactive isotopes decay at a steady, predictable rate. Radium, for example, has a half-life of about 1,600 years. That is, if you had a solid block of radium, half of it would decay into other elements in 1,600 years. In another 1,600 years, half of that would decay, and so on until it was all gone. If you know how pure your block of radium originally was, it's relatively simple to calculate how old it is by measuring how much radium is left. This is a very useful tool if you have solid blocks of radium that need dating, but if you don't know how much radium was there in the first place, the job is a lot harder.

For living things, we do have a radioactive isotope that we can measure. It's called carbon-14 and is created by cosmic ray neutrons striking nitrogen atoms in the atmosphere. This naturally radioactive form of carbon has a half life of 5,568 years ± 30 years, and is absorbed by living things in the form of food or respiration. When an organism dies, it ceases to absorb new carbon-14 and the clock starts ticking.

What allows scientists to measure how much carbon-14 was in an organism when it died is the assumption that the isotope is produced at a more or less steady rate that can be calibrated by comparing radiocarbon dates against more reliable tree ring dating. Since the ratio of carbon-14 to other carbon isotopes is fixed, and no new carbon-14 is absorbed after an organism dies, it's possible to estimate the time since the organism died going back tens or even hundreds of thousands of years by measuring the ratio of carbon-14 against stable carbon isotopes in ancient organic compounds.

It's served archeology and other fields well, but now it's in a spot of bother. A new study by Heather D Graven from the Department of Physics and Grantham Institute at Imperial College London indicates that fossil fuel emissions are confusing things by drastically altering the ratio of carbon-14 to other isotopes. Her study uses a simple carbon cycle model with a one-dimensional box diffusion model of the ocean, with the atmosphere and the biosphere represented as one-box carbon reservoirs.

According to this model, the carbon being pumped into the atmosphere by vehicles and industry is diluting the carbon-14 by introducing large amounts of "fossil" carbon. Though fossil fuel sources are largely derived from dead organisms, they are many millions of years old and all of the carbon-14 they once contained has long decayed away, leaving only stable isotopes. As carbon compounds from these fuels are emitted into the atmosphere, they change the naturally fixed ratio of carbon-14 to other isotopes, so the atmosphere appears much older. The result is that as modern organisms ingest this carbon and eventually die, they, too, seem much older than they actually are.

Graven says that the amount of carbon being pumped into the atmosphere is so great that by 2050, a newly made cotton t-shirt will have the same radiocarbon date as a robe worn by William the Conqueror in the 11th century. She also says that at the present rate, the effect will be noticeable by 2020.

The study indicates that if this occurs, it will not only have implications for archeology, but also for detecting fraudulent art works, detecting illegal ivory, and other important fields. However, if carbon emissions could be drastically reduced, the effects could be reduced or even eliminated over time.

One point to keep in mind is that the important thing about this study isn't that carbon emissions are affecting radiocarbon dating, but the scale at which they are doing so according to Graven's modeling. In fact, as she points out, artificial carbon's effect has always been well known since the early days of the technique. It's called the Industrial or Suess effect, named after Hans Suess, who first studied it. Industrial carbon has been throwing off dating since about 1880 and it hasn't been restricted to atmospheric emissions. Many fertilizers, pesticides, and other agricultural chemicals use petrochemicals, which are routinely sprayed on crops and have to be taken into account as a dating contaminant.

If that isn't enough, there's also the Bomb effect, which Graven also notes. This gets its name from the open air atomic testing that began with the Allies in 1945 and ended with the Chinese in 1980. This 45 years of setting off nuclear weapons above ground generated a massive spike in carbon-14 that doubled the normal amount, but this has been dropping as time goes on and the atmospheric isotope is absorbed into the biosphere.

This has made the atmosphere appear much younger just as the fossil carbon makes it appear much older. The upshot is that radiocarbon dating has always had a large margin of error. Part of this is an inherent part of the technique, but part is due to the confusing mixtures in the modern atmosphere. The result is that anything within 300 years of today ("today" is defined as 1950) is modern. If Graven's model holds up, it means that this 300 year margin will extend several centuries into the past until it exceeds the 1,000 year mark.

But what practical effect will this have on radiocarbon dating? Star Trek tricorders and CSI episodes notwithstanding, it is not a matter of waving a device over an artifact and reading a dial. It's a complex process of intersecting lines of logic designed to eliminate errors and isolate the only possible conclusions. In addition, a lot of the famous cases involving murder and forgery don't rest on how old something is, but how young, which is much easier to prove. Put new wine in an old bottle and you might get away with palming it off as being from the Last Supper. Seal it with a plastic cork, and you're busted.

The other thing to bear in mind is that, though the study has very important implications, it does not invalidate carbon dating. The fossil fuel carbon will not affect the age of any ancient materials. It simply makes more recent ones more tricky to handle. Murderers aren't going to walk free and people aren't going to think that William the Conqueror wore "Eric Bloodaxe Rules Okay" t-shirts.

What is does mean is that radiocarbon dating is going to need much more corroboration for more recent periods. Scientists will need to keep an eye out for very modern isotopes, such as strontium 90, which spiked in the atmosphere after the atomic tests, as well as being careful about which carbon compounds they use for dating materials. Generally collecting carbon that could come from anywhere, for example, may need to give way to looking specifically for lignin or collagen.

Also the provenance of finds will be very important. Archeologists are very keen on what are called sealed contexts. That is, artifacts discovered buried in stratigraphy that provides verifiable bookends for possible dates, such as a layer of soil in the ancient Romano-British city of Colchester with a layer below containing coins of a certain date and the layer above made of ash from the year the city was famously burned to the ground in a rebellion. Anything in the layer between these two must have been deposited between those two dates, which provides a cross reference.

However, one problematic aspect of the study is Graven's assertion that by drastically controlling emissions, it could return carbon-14 ratios to preindustrial levels. In a sense, it could correct the problem, but only by introducing another, which would be to firmly establish that carbon-14 levels can be artificially manipulated by means of adding radioactive or nonradioactive isotopes as desired. The premise of radiocarbon dating is that carbon-14 is produced naturally at a constant rate. If it is understood that the carbon-14 ratio is now a product of regulation, then radiocarbon dating from the mid-20th century onward will always be qualified with a footnote as to the reliability of the regulator.

The study was published in the Proceedings of the National Academy of Sciences (PNAS).

Source: Imperial College London