LAUSR

A simple protocol was developed to measure seawater calcium carbonate saturation states (Ωspec) spectrophotometrically. Saturation states are typically derived from the separate measurement of two other carbon system parameters, with each requiring unique instrumentation and often complex measurement protocols. Using the new protocol, the only required equipment is a thermostatted laboratory spectrophotometer. For each seawater sample, spectrophotometric measurements of pH (visible absorbance) are made in paired optical cells, one with and one without added nitric acid. Ultraviolet absorbance is measured to determine the amount of added acid based on the direct proportionality between nitrate concentration and UV absorbance. Coupled measurements of pH and the alkalinity change that accompanies the nitric acid addition allow calculation of a seawater sample's original carbonate ion concentration and saturation state. These paired absorbance measurements yield Ωspec (and other carbonate system parameters), with each sample requiring about 12 min processing time. Initially, an instrument-specific nitrate molar absorptivity coefficient must be determined (due to small but significant discrepancies in instrumental wavelength calibrations), but thereafter no further calibration is needed. In this work, the 1σ precision of replicate measurements of aragonite saturation state was found to be 0.020, and the average difference between Ωspec and Ω calculated conventionally from measured total alkalinity and pH (Ωcalc) was -0.11% ± 0.96% (a level of accuracy comparable to that obtained from spectrophotometric measurements of carbonate ion concentration). Over the entire range of experimental conditions, 0.97 < Ω < 3.17 (n = 125), all measurements attained the Global Ocean Acidification Observing Network's "weather level" goal for accuracy and 90% attained the more stringent "climate level" goal. When Ωspec was calculated from averages of duplicate samples (n = 56), the precision improved to 0.014 and the average difference between Ωspec and Ωcalc improved to -0.11% ± 0.73%. Additionally, 97% of the duplicate-based Ωspec measurements attained the "climate level" accuracy goal. These results indicate that the simple measurement protocol developed in this work should be widely applicable for monitoring fundamental seawater changes associated with ocean acidification.