Maintained by:
Evan James Gowan
evangowan@gmail.com
Since different organisms and materials fractionate carbon differently, this can lead to errors when determining the true age of an sample. It is now standard to normalize material to a δ13C value -25‰ using the PeeDee Belemnite standard. This is important, because not doing this correction can lead to errors in the age of hundreds of years. The error amounts to about 16 years per 1‰. As an example, marine shells often have a δ13C value of 0‰, which would lead to an error of about 400 years.
In the early years of radiocarbon dating, the fractionation of carbon isotopes was often not taken into account. Some labs corrected with measured δ13C values, while others use an assumed value based on the material that was dated. One of the issues I have run into is that there is no comprehensive list of what labs did this correction, and which ones did not. With modern AMS labs, you can almost be assured that they measure the δ13C and normalize the date, but with older conventional labs this was often not true. Some labs corrected some material, while not doing it for others. As an example, the Groningen lab only corrected terrestrial materials and did not correct freshwater and marine shells. These kind of details are often only available in the lab lists that used to be published in the journal Radiocarbon.
There is also the large difference between conventional and AMS methods in terms of how much material was needed for a sample. Conventional dating requires a lot larger sample to get a precise age. If a large block of wood was sampled, it is likely that the conventional and AMS dates are comparable. For samples like marine shells, conventional ages are likely derived from a mixture of dozens of shells, sometimes with different species. This can be problematic, since there is a big difference in how carbon is taken up by suspension feeders (like Hiatella arctica) versus deposit feeders (like Portlandia arctica). If a sample contains a lot of suspension feeder organisms, the apparent age might be significantly older than reality. It may be desirable for some to add an additional correction for these kind of samples.
Some labs also can report ages in non-standard ways. The most prominent example is the Geological Survey of Canada (GSC), which reported ages with 2-sigma uncertainties. Notes on these are also necessary to correctly calibrate dates.
I also thought it would be a good idea to document the laboratory name, the lab code and the country. This information might be useful. I am open to having additional columns. I also include a reference to where you can find the details on the fractionation correction.
The purpose of this list is to make these details available, and done in a way that it is publicly available. If you have any additions or corrections, please let me know via the Issues tab.
This repository has a spreadsheet file with the lab information that I use to enter everything. I also included a markdown file which can be viewed in a web browser.