Assessment of Autonomous pH Measurements for Determining Surface Seawater Partial Pressure of CO2

Abstract

The Southern Ocean Carbon and Climate Observations and Modeling (SOCCOM) program currently operates >80 profiling floats equipped with pH sensors in the Southern Ocean. Theoretically, these floats have the potential to provide unique year‐around estimates of pCO2 derived from pH measurements. Here, we evaluate this approach in the field by comparing pCO2 estimates from pH sensors to directly measured pCO2. We first discuss data from a ship's underway system which covered a large range in temperature (2–30°C) and salinity (33.6–36.5) over 43 days. This pH sensor utilizes the same sensing technology but with different packaging than those on SOCCOM floats. The mean residual varied between −4.6 ± 4.1 and 8.6 ± 4.0 (1σ) μatm, depending on how the sensor was calibrated. However, the standard deviation of the residual, interpreted as the ability to track spatiotemporal variability, was consistently < 5 μatm and was independent of the calibration method. Second, we assessed the temporal stability of this approach by comparing pCO2 estimated from four floats over 3 years to the Hawaii Ocean Time‐series. Good agreement of −2.1 ± 10.4 (1σ) µatm was observed, with coherent seasonal cycles. These results demonstrate that pCO2 estimates derived from profiling float pH measurements appear capable of reproducing spatiotemporal variations in surface pCO2 measurements and should provide a powerful observational tool to complement current efforts to understand the seasonal to interannual variability of surface pCO2 in underobserved regions of the open ocean.