The study of isotopes – chemical signatures preserved in our bones and teeth that shed light on diet and movements during life – are increasingly becoming a major part of archaeology, frequently redefining how we look at different periods and featuring in most post-excavation analyses. But we still have a long way to go in terms of being able to use them to confidently pinpoint a person’s specific origins. At the moment, most isotopic maps are still fairly crude and the science is better at identifying local vs non-local rather than confidently determining exact locations. A new study, recently published in Science Advances, has highlighted the need to make sure these maps are more accurate, bringing up the potential impact that agricultural practices might have in certain regions.
Erik Thomsen and Rasmus Andreasen from Aarhus University in Denmark assessed the effect that adding lime to fields (in order to change the soil pH) has on strontium concentration and isotopic composition. After analysing 84 samples of surface water from both agricultural areas (24 samples) as well as pristine locations (60 samples) in different areas of Jutland, they found that lime can indeed have an impact in certain regions. In West and Central Jutland there was a significant difference in strontium isotope values between the two samples: while the pristine samples had strontium values that were high and variable (0.7104-0.7150), farmland ones were low and much narrower (0.7091-0.7115). They also found a difference in strontium concentrations, which were low and variable in the pristine areas but higher and less variable in the farmlands. By contrast, in East Jutland there were no significant differences between the pristine and farmland samples.
The reason for this disparity appears to be due to the soil composition in these different regions. West and Central Jutland are characterised by non-calcareous, or non-chalky, soil, while the East Jutland soil is chalky. The reason why chalky soil is not affected is that it is already naturally high in lime. As Thomsen and Andreasen describe, ‘A hypothetical soil, 20cm thick containing 10% CaCO3, would naturally hold about 320 metric tons/ha of calcium carbonate. The addition of 500kg/ha of agricultural lime would increase this amount by about 0.16%, which would have minimal influence on both the strontium concentration and isotopic composition.’ They also assessed the impact of other fertilisers, manure, animal feed, and pesticides and found their effect to be negligible.
These results were seen as being significant enough to affect some previous isotopic studies in Denmark. Previous analysis of a Bronze Age burial (c.1370 BC) from Central Jutland, known as the Egtved Girl, had shown that she was probably born in southern Germany, moved to Denmark 13 months before her death but only stayed there for nine months before moving elsewhere, and then returned shortly before her death. But, taking into account that the map for the area where she is buried is largely based on samples from non-calcareous farmland soils, after correcting this with pristine samples Thomsen and Andreasen suggest that instead she was most likely local to the area and that her varying isotope values may instead reflect a seasonal migration pattern between where she was buried and the Vejle Tunnel Valley just to the north.
So what does this mean for isotopic studies in the UK? Well, it does not mean that we have to redo all previous research (although these findings should certainly be taken into account in refining British isotopic maps). For example, many of the larger, more ‘headline-grabbing’ isotopic studies have been done in the Stonehenge landscape, which is located on Wessex chalklands (see CA 334 and 344). As the study found that chalky soil is not significantly affected by agricultural lime practices, the maps for these areas are likely to still be quite accurate. And that means that any samples from this area that have been found to be ‘non-local’ are likely to be correct. Moreover, in the UK it is not only the Wessex chalklands that are calcareous but also large areas across the east coast.
Most studies have been quite cautious in determining potential origins for ‘non-local’ individuals. Only studies that have identified origins in areas solely defined by samples taken from agricultural land in non-calcareous areas will potentially be affected. These studies can be easily rectified by taking new samples from pristine landscapes. While this is easier said than done in localities where farming is ubiquitous, Thomsen and Andreasen found that these pristine samples only need to be about 150m from farmland, although a buffer of at least 300m might be wise.