A community effort to improve inter-laboratory standardization of the carbonate clumped isotope thermometer using carbonate anchors

A Unified Clumped Isotope Thermometer Calibration (0.5–1,100°C) Using Carbonate‐Based Standardization

“…The regression from this study is nearly identical to the regression from Jautzy et al. (2020) when all Δ 47 values are calculated with “InterCarb” (Bernasconi et al., n.d.) anchor values. (b) T‐Δ 47 relationship for samples 0°C–100°C including regressions from studies with material reanalyzed for this study (Bonifacie et al.…”

Full Propagation of Analytical Uncertainties in Δ47 Measurements

“…What is more, as reported by Bernasconi, Daëron et al. (2021), using this framework yields a very reasonable ( p = 0.19) prediction for the distribution of inter‐laboratory scatter in Δ 47 values within the InterCarb data set.…”

“…Accurate comparisons of clumped‐isotope data produced by different laboratories have long remained a challenge (Petersen et al., 2019, and references therein). A striking result of the InterCarb comparison exercise (Bernasconi, Daëron, et al., 2021) is that despite data sets from different labs having extremely diverse analytical errors, the overall scatter between all laboratories is accurately predicted (i.e., neither too large nor too small) by the error propagation models described here, implying that carbonate‐standardized Δ 47 measurements are free of unrecognized systematic inter‐laboratory discrepancies. It is thus reasonable to expect that we are now capable of quantitative comparisons between results from different laboratories, but this requires that future studies report full analytical uncertainties.…”

“…Because the relative contributions of the autogenic and allogenic error components defined above (Equations ) are sensitive to the distribution of analyses among anchor and unknown samples and on the isotopic composition of unknowns relative to the anchor polygon, they are expected to vary greatly between laboratories and/or sessions. The InterCarb data set, comprising over five thousand analyses from 22 different laboratories (Bernasconi, Daëron, et al., 2021), offers an excellent opportunity to quantify these two components in a wide range of realistic settings.…”

“…(2019) tested the first two of these hypotheses and found that using unified methods for 17 O correction and subsequent data processing reduced inter‐laboratory discrepancies without eliminating them. Testing the third hypothesis is one of the goals of the recently completed inter‐laboratory comparison exercise “InterCarb”, whose results are reported in a companion study (Bernasconi, Daëron, et al., 2021).…”