On estimating the precision of stable isotope ratios in processed tree-rings

Abstract

Stable isotope dendrochronology is a well-developed field of research, but improvements to methodologies are on-going. We propose an improved method for estimating the precision of stable isotope ratios (δ) of tree-ring samples that are processed from whole wood to various end products such as cellulose-nitrate, α-cellulose, or cellulose intermediates. The status quo method for estimating the δ precision of organic solids is to characterise the long-term 2-sigma range of δ values for a ready-made Quality Assurance (QA) standard that is included in each analysis run of samples. While the status quo method is appropriate for characterising analytical uncertainties associated with the mass spectrometer, combustion or pyrolysis system, and analyte specifics, it does not reflect uncertainties associated with sample processing from inadvertent and unrealised operator error (e.g., contamination by airborne particles, incomplete chemical processing, sample storage issues, and other unforeseen errors), although such errors would probably be rare with an experienced operator. The proposed method improves upon the status quo method as it respects the Identical Treatment principle by subjecting QA standards to the same processing steps that samples undergo. As such, analytical uncertainties associated with sample processing would be integrated into the QA standard's δ value and precision estimate. In effect, the proposed method is a system to monitor inter-batch reproducibility and, by the same token, can be used to identify batches that were potentially compromised during processing. A pilot study example is used to demonstrate the proposed method for δ18O analysis of α-cellulose samples.