Kanstrup, Marie5; Holst, M.K.6; Jensen, Peter Mose7; Thomsen, I.K.8; Christensen, Bent Tolstrup8
1 School of Culture and Society - Prehistoric Archaeology, subject, School of Culture and Society, Arts, Aarhus University2 Moesgaard Museum3 Department of Agroecology - Soil Fertility, Department of Agroecology, Science and Technology, Aarhus University4 Department of Physics and Astronomy, Science and Technology, Aarhus University5 Department of Physics and Astronomy, Science and Technology, Aarhus University6 School of Culture and Society - Prehistoric Archaeology, subject, School of Culture and Society, Arts, Aarhus University7 Moesgaard Museum8 Department of Agroecology - Soil Fertility, Department of Agroecology, Science and Technology, Aarhus University
Current concepts of prehistoric manuring are founded on limited and mainly circumstantial evidence, giving rise to much ambiguity with respect to the onset of systematic use of manure to enhance cereal production. This paper reports carbon (C) and nitrogen (N) contents and isotopic compositions (δ13C, δ15N) of charred grains of naked barley, emmer and spelt dating to the first four millennia of early agriculture in Denmark. The δ15N values ranged from c. 0.5‰–5.5‰, 0.5‰–6.0‰ and 1.5‰–8‰ for spelt, emmer and naked barley, respectively. This study represents the until now most comprehensive investigation of long term trends in δ15N values of charred cereal grains, which previous research have proposed as an indicator for prehistoric manuring practice. Our study suggests a long-term (3900–500 BC) decrease of manuring intensity in emmer cropping. Conversely the long-term (2300 BC – AD 1) trend for naked barley cropping displays a more distinct and significant increase (+2‰) in grain δ15N values, reflecting an increased manuring intensity with an average δ15N value of as high as 6‰. We interpret this trend as indicating the initiation of a more intensive and systematic manuring practice associated with cultivation of barley in the Early Iron Age (500 BC- 0). Although the isotopic signal ascribed to manuring was (somewhat) variable, the relative manuring effect was detected throughout the chronological continuum being investigated. Further, we observed that the conventional sample pre-treatment (acid-base-acid) induced an average δ15N offset of 0.7‰ (pre-treated sample > non pre-treated sample). This has not previously been reported. Methodological advancements are needed to remedy this issue and provide consensus about appropriate pre-treatment of grain samples from archaeological sites. We conclude that N-isotope analysis of charred cereal grains constitutes a new and direct source of information about prehistoric manuring practice.
Journal of Archaeological Science, 2014, Vol 51, p. 115-125