Quantification of seawater total alkalinity measurement uncertainty to support the evaluation of ocean alkalinity enhancement

Abstract

Ocean alkalinity enhancement is one approach being considered to contribute to marine Carbon Dioxide Removal techniques. It relies on the addition of alkalinity to the marine environment, either under the form of crushed-rock feedstocks or under a dissolved form. This conducts to the decrease of seawater partial pressure of CO2 (pCO2) allowing seawater, by equilibrium with air, to absorb more CO2 from the atmosphere. In order to determine the amount of CO2 removed from the atmosphere, a monitoring, reporting, and verification (MRV) system of marine carbon dioxide removal needs to be designed. The evaluation of ocean alkalinity enhancement will depend on the monitoring of measurable variables of the marine carbonate system, as well as on numerical simulations. In this context, acquiring accurate and robust data of seawater total alkalinity is highly relevant in order to quantify the background state alkalinity and to check that alkalinity has efficiently been added. In that goal, the application of the three pillars of metrology: metrological traceability, measurement procedure harmonization and validation, and uncertainty estimation, are required. However, up to date, the total alkalinity measurement method lacks of a rigorously established uncertainty budget. The establishment of measurement results uncertainty can be realised by the mean of two approaches. The “bottom-up” approach is the most rigorous way to thoroughly establish an uncertainty budget. It relies on the identification and quantification of each source of uncertainty involved at every step of the measurement process, as described by the Guide to the expression of Uncertainty in Measurements. The “top-down” approach relies on an experimental assessment of the uncertainty, from repeatability, reproducibility and trueness estimates. The presentation will focus on the evaluation of the uncertainty of seawater total alkalinity measurement results using the two approaches aforementioned. The sources of uncertainty originating from the potentiometric titration measurement method, and the mathematical model used for data treatment, will be presented and quantified. This study will also allow identifying which sources have the major contribution to the overall uncertainty budget, and thus the ones we should focus on to lower the uncertainty. The estimation of the uncertainty with the “top-down” approach is determined from an inter-laboratory comparison involving five laboratories, conducted on reference solutions. The developed artificial and natural seawater reference solutions, as well as the results of the inter-laboratory comparison, will be presented. Comparison, advantages and limitations of the two uncertainty estimation methods will be discussed. Finally, the level of uncertainty estimated will be discussed in the frame of MRV system in supporting the evaluation of ocean alkalinity enhancement.