The design of the new radiocarbon intercomparison, GIRI

Full text: Given the complexity of the radiocarbon dating process, the diversity of materials being dated, the continued technical developments, GIRI (the Glasgow international radiocarbon intercomparison) is the next development of the series of inter-comparisons to support continuing quality assurance. As a result of the needs to deliver accurate and precise measurements but also as part of general, good laboratory practice, including laboratory benchmarking and quality assurance, the 14C community has previously undertaken a wide-scale, far reaching and evolving programme of global inter-comparisons, to the benefit of laboratories and users alike (Scott et al, 2018). GIRI has been designed to continue this programme and to meet a number of objectives, including the most fundamental one, to provide an independent assessment of the analytical quality of the laboratory/measurement and an opportunity for a laboratory to participate and improve (if needed). The principles that we followed in the creation of GIRI are to provide:- A series of unrelated individual samples, spanning the dating age range; Some linked samples to earlier inter-comparisons to allow traceability; Some known age samples, to allow independent accuracy checks; A small number of duplicates, to allow independent estimation of laboratory uncertainty; Two categories of samples, bulk and individual to support laboratory investigation of variability. All of the GIRI samples are natural (wood, peat and grain), some are known age, and overall their age spans approx. >40,000BP to modern. Sample materials include: humic acid, whalebone, grain, some single ring dendro-dated samples, some wood samples spanning a number of rings (10 rings), and a background wood sample. We have designed the study with two groups of samples. The first group is typical of the samples provided in previous intercomparisons, where the volume of material being provided is sufficient to make, at most, a very small number of repeat measurements. The second group of samples will provide a quantity of material, sufficient to allow AMS labs to run multiple measurements in different wheels/batches. The purpose of including the first group of samples is to allow each laboratory to quality check (once consensus values and uncertainties have been defined) their laboratory operation at the time of analyses (so a classical round robin trial). The second group of samples provides laboratories with well-characterised materials which can function as secondary standards, to be run routinely and thus allow assessment of within-laboratory variability. Analysis of the GIRI results will have several strands. Ultimately, we wish to define consensus values for all the samples and a quantified uncertainty supporting a more in-depth evaluation of laboratory performance and variability. The radiocarbon dating community inter-comparison programme has been running for more than 30 years and has evolved over time in line with the changing technology. This paper presents the most recent intercomparison (GIRI) and its design.