Sie sind hier: Die Projekte / Verbesserte Zeitskalen aus Knochendatierungen / Projektbeschreibung
Verbesserte Zeitskalen aus Knochendatierungen: Isotopenuntersuchungen nach Reservoiralter und Diät


An increasing number of bone radiocarbon dates brought to light inconsistencies with ages obtained on associated charcoal, the latter being younger, for example at Lepenski Vir in the Iron Gates Section of the Danube River (Cook et al., 2001, Bonsall et al., 2004). The puzzle could be solved at another Iron Gates site, Schela Cladovei, where arrowheads made out of ungulate bones, that were found embedded in, or directly associated with, human skeletons dated several hundred years younger than the skeletons. Although δ13C values were only marginally heavy, associated δ15N values were clearly enriched and pointed to a riverine fish diet and a reservoir effect of the River Danube as source of the discrepancies.

Fischer and Heinemeier (2003) observed in Lake Tissø in the Åmose Valley on Sjaelland, Denmark, apparent reservoir ages of more than 1000 years in fish and freshwater mussels that showed light δ13C values, and derived reservoir ages of up to several hundred years for food residues on the inside of stone age potshards from a comparison with associated ox and deer bone. They argue that remarkably early appearances of various cultural phenomena in northern Germany, based on 14C dates on food residues in pottery, may not be real and reflect unrecognized reservoir effects. This fits with the age inconsistencies observed in the Ostorf grave field near Schwerin, which may be attributable to diet-related reservoir ages of 250 to 710 years (possibly up to 1050 years). Philippsen et al. (2008) discuss variable reservoir ages in river water and samples of fluvial origin from hundreds up to 3000 years that partly transfer into food crusts in Northern Germany, and very high fresh water reservoir ages - up to 4430 years – in the Netherlands, and their influence on ages measured for 11th-13th century Counts of Holland - were reported by Lanting and van der Plicht (1998). This suggest that in the northern part of Middle Europe and southern Scandinavia, we have to reckon with potential reservoir ages in dating human bones, even at sites far from the sea.

What makes the fresh water reservoir effect particular problematic are its magnitude, high variability and its asymptomatic δ13C values, all related to its causal processes. In the oceans the local reservoir age is generally determined by 14C decay in the ocean and large scale mixing processes. The fresh water reservoir age, by contrast, starts with the dissolution of old – often 14C-free – carbonate by soil CO2 of atmospheric composition. The reservoir age is then determined by the fraction of soil CO2 used for dissolution, by mixing with runoff and groundwaters of different origin and age in streams and lakes, and by exchange with the atmosphere. As a result local fresh water reservoir ages are highly variable in space and time, depending on many parameters such as rainfall, hydrology, topography, vegetation, carbonate concentrations in the underground etc., and δ13C values do not give clear indications of their magnitude, which makes the construction of reliable precise chronologies from bones and food crusts problematic.

Isotopic indications for diet

pre-treatment of bones

The δ13C and δ15N data, are often collected to quantify diet-related reservoir ages in bone collagen. Yet reservoir ages, determined directly by 14C analysis of bone and associated materials, can also be applied to back calculate food sources. As the difference in δ13C between trophic levels (ca. 1 ‰) (Schoeninger and DeNiro., 1984) is small compared with the natural variability, the δ15N value with a 3‰ change per trophic level (Ambrose, 1991) is often used alone to indicate an aquatic component of the diet. As also δ15N has its, often poorly understood, variability and the bone collagen, due to its slow turnover (Bayliss et al., 2004), records long term averages in mostly protein consumption, the diet reconstruction based on δ15N is necessarily imprecise (Hedges and Reynard, 2007). Dating human bones together with associated plant material and/or terrestrial herbivores provides a direct human dietary reservoir age, while the dating of associated fish bones or mussels indicates the aquatic endmember. The three dates together make it possible to quantify the aquatic contribution to the (protein) diet, independent of trophic level assumptions. Thus results of the 14C reservoir age study via direct AMS 14C dating may be used to verify the biochemical isotope models used in the stable isotope studies of diet.

Yet, carbohydrates and fat, important sources of energy in the diet, are in this calculation poorly represented, as their carbon is largely oxidized to CO2 and used not to build proteins. The CO2 is transported in the blood and exhaled, but some of it is fixed as carbonate in the bone apatite (Kelner and Schoeninger, 2007). Turnover for this carbonate is considerably faster than for collagen, so it records the 14C and δ13C of recent carbohydrate and fat intake. Thus apatite carbonate might provide insights in the more recent total diet including the non-protein part (Kelner and Schoeninger, 2007). However, the fairly easy exchange that allows bone carbonate to record recent dietary details also can take place while the bone is buried in the soil, which leads to loss of information over time. In-soil exchange via groundwater also easily affects the amount and isotopic composition of strontium, present in trace amounts in bones, which may be monitored to indicate the significance of information obtained from apatite carbonate (Nelson et al., 1986).


The goals of the research are (a) to provide better age control for the skeletal material used in the aDNA project in support of the interpretation of genetic lineages and their implications for the social and cultural structures of Neolithic communities, (b) and to improve regional chronologies, including bone dates, for better comparison of contemporaneous social and cultural developments in different regions.

To achieve these aims, fresh water reservoir age influences in bone dating, and the stable isotopic composition – δ13C and δ15N – as potential indicator for diet, and reservoir age of human bones have to be quantified.