Time flies, which means busy days and definitely no boredom! Last week we had a successful workshop, which focused on Asian monsoon research at the Bolin Centre, and brought together scientists interested in different aspects of the Asian monsoon: paleo-data, climate modeling and modern climatology. Our invited guests, Kevin Anchukaites and Jessica Tierney from Woodshole Oceanographic Institute and Hans Linderholm from Gothenburg University gave very interesting and inspiring talks.
A few days ago, Moo, Francesco, Nut and I visited the 14CHRONO Centre at Queen’s University in Belfast, and also this venue had an Asian monsoon theme, since Paula Reimer, Maarten Blaauw and Chris Hunt are part of the Asian monsoon project. The students could for the first time see where all our radiocarbon samples are dated and could discuss age modeling, its potentials and limitations with Maarten Blaauw.
Chronology forms such an important part of our research, because without a valid time scale, we have no means of discussing when changes occurred and how these compare to regional or even global events. But establishing chronologies and age models is not a simple thing. Radiocarbon dates may not always turn out the way one had hoped for, but may be older or younger than expected. Often these ‘wrong’ dates are not even published, and elegantly eliminated from the record. In my view radiocarbon dates are rarely wrong, but rather tell a story, and it is important to understand the story they tell. It may be that the ‘wrong’ type of material has been dated or that the sample is contaminated by younger carbon. The ‘wrong’ type of material may actually tell a story of erosion and reworking, or may indicate lake water reservoir effects; and contamination will tell you to be more careful next time you prepare your sample.
Without an understanding of the underlying reasons for strange radiocarbon dates, step number two, which is age modeling, becomes even more difficult. The choice of model is one issue, but an even more crucial issue is to decide which dates may be outliers, and which ones may be correct. Of course all this is no problem if the dates appear in a nice sequence, i.e. if they become older with depth. But if they don’t, even the most adequate statistical approach cannot help to produce a reasonable age model.
Quite often, the explanation for ‘wrong’ radiocarbon dates can be found in the type of sediments from which the organic material has been selected, or in the type of material selected for dating. This brings us back to the most basic part: a good knowledge of the sediments and of the sediment stratigraphy, a fact that is just too often overlooked.