Analysis of a global hydrographic data product showed a clear pH-dependent discrepancy between pH on the total scale measured spectrophotometrically (pHspec) using purified meta-Cresol Purple (mCP) and pH calculated from total alkalinity and dissolved inorganic carbon. However, this was based mainly on US cruises, and three recent Japanese cruises do not show this pH-dependent discrepancy. One potential explanation is that purified mCP batches obtained from different institutions lead to significantly different pHspec. Here, we tested this hypothesis by comparing the performance of purified mCP obtained from four different institutions. We demonstrate that consistent pH of ±0.0012 (95% C.I.) can be achieved regardless of the institution when impurities are properly removed. However, there was at least one batch from three of the four institutions that had significant pH-dependent errors that were as large as −0.008. The presence of impurities that led to pH-dependent errors was identified using HPLC and, for 8 out of the 9 cases, by spectrophotometry (although issues still remain for the latter). We conclude that pH-dependent errors due to impurities that remain after the purification process are, by themselves, too small to account for the differences observed between the recent set of cruises. Identifying the source of this difference should be a top priority. This study also highlights the importance of establishing robust quality assurance and quality control protocols to ensure consistent behavior with previously published equations to compute pH. We recommend a centralized system where one or a handful of institutions distribute purified mCP for the community, as this distribution approach will lead to lower prices and simplify quality assurance.
- Meta-cresol purple