Calculating a statistically robust δ13C and δ15N offset for charred cereal and pulse seeds

Abstract

This study measured δ13C and δ15N of fresh and charred seeds from six different taxa of cereals and pulses. For each taxon, 12 different batches of seeds, all from the same growing context or field, were charred for 4, 8, or 24 hours at 215, 230, 245, or 260°C, with the thirteenth batch left uncharred. The least charred samples showed no average offset compared to their uncharred counterparts, but among the charred seeds, every 15°C increase resulted in an increase in δ15N by 0.12‰, while increasing the duration of charring by 4 hours resulted in an average increase of 0.04‰. Differences in δ13C among the charred samples were only minimally dependent on time (+0.016‰/4 hours duration), but not temperature. The average offset for all batches and all taxa was minimal for δ13C, at +0.11‰ (95% confidence interval (CI) 0.003–0.22‰). The average offset for δ15N of 0.31‰ (95% CI 0.05–0.57‰) is smaller than the conservative estimate of 1‰ used in previous studies. We demonstrate that using this new, smaller charring offset for δ15N makes a significant difference in paleodietary models, one that further emphasizes the potentially high δ15N of manured crops relative to animal forage. Statement of significance Archaeologists and scientists measure carbon and nitrogen isotope ratios in ancient crop remains that have been preserved by charring to understand how crops were grown in the past. The charring process preserves the seed's physical form but results in a slight alteration to the carbon and nitrogen isotope ratios. Our study systematically measured this offset within a window of charring temperatures and times: a window in which the seeds and grains have been completely chemically converted into a stable form, but where this process has been gentle enough that the physical form of the seed has not been distorted. From this study, we understand more precisely the size of the small adjustment we need to make to our measurements of ancient charred crop remains so that we can work back to the original values that reflect the conditions in which these crops were once grown. Data availability The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are contained within the paper.