Internal consistency of marine carbonate system measurements and assessments of aragonite saturation state: Insights from two U.S. coastal cruises

Abstract This research assesses the thermodynamic consistency of recent marine CO 2 system measurements in United States coastal waters. As one means of assessment, we compared aragonite saturation states calculated using various combinations of measured parameters. We also compared directly measured and calculated values of total alkalinity and CO 2 fugacity. The primary data set consists of state-of-the-art measurements of the keystone parameters of the marine CO 2 system: dissolved inorganic carbon (DIC), total alkalinity (TA), CO 2 fugacity ( f CO 2 ), and pH. This study is the first thermodynamic CO 2 system intercomparison based on measurements obtained using purified meta cresol purple as a pH indicator. The data are from 1890 water samples collected during NOAA's West Coast Ocean Acidification Cruise of 2011 (WCOA2011) and NOAA's Gulf of Mexico and East Coast Carbon Cruise of 2012 (GOMECC-2). Calculations of in situ aragonite saturation states (Ω A ) near the saturation horizon exhibited differences on the order of ± 10% between predictions based on the (DIC, TA) pair of measurements vs. the (pH, DIC), ( f CO 2 , DIC), or ( f CO 2 , pH) pairs. Differences of this magnitude, which are largely attributable to the imprecision of Ω A calculated from the (DIC, TA) pair, are roughly equivalent to the magnitude of Ω A change projected to occur over the next several decades due to ocean acidification. These observations highlight the importance of including either pH or f CO 2 in saturation state calculations. Calculations of TA from (pH, DIC) and ( f CO 2 , DIC) showed that internal consistency could be achieved if minor subtractions of TA (≤ 4 μmol kg − 1 ) were applied to samples of salinity f CO 2 ): ~ 3 μmol kg − 1 , which is similar to the accuracy of the TA measurements. Systematic trends can be detected in the offsets between measured and calculated parameters, but for this high-quality data set the magnitude of methodological improvements required to achieve exact thermodynamic consistency is quite small.