Perspective: On the discontinuity between the scale realizations VPDB2020 and VPDB2006 – The underestimated role of measurement uncertainties?

Abstract

Several recent publications have discussed the discontinuity between the δ13C scale realization VPDB2020 (new IAEA carbonate reference materials (RMs) IAEA‐610/611/612) and the LSVEC‐traceable realization named as VPDB2006 (organic RMs normalized to LSVEC in 2006, and USGS44 with its δ13C based on those RMs and LSVEC). This looks like an unsolvable puzzle. The question from practical users is as follows: do we want to be correct, that is, use VPDB2020, or be compatible with the results obtained in previous years, that is, use VPDB2006? In metrology, the significance of any discrepancy is considered in terms of the associated combined uncertainty. Uncertainties estimated for VPDB2020 according to the metrological approaches are much lower than the discontinuity. However, the situation for VPDB2006 is unclear. By reviewing publications related to LSVEC and LSVEC‐normalized organic RMs of 2006, several missing and/or underestimated uncertainty components can be identified. When these missing or underestimated uncertainties are taken together, the discontinuity at negative δ13C appears to be the same as or similar to the combined underestimated uncertainties for VPDB2006.