Award: OCE-2049117

The ocean absorbs approximately 25% of human caused CO₂ emissions every year, curbing the impacts of climate change on society. Currently, the magnitude and variability of this ocean carbon sink is quantified through data products produced from highly accurate shipboard measurements. However, the number of shipboard measurements has been declining over the past decade, potentially leading to larger uncertainties. Furthermore, these assessments are typically delayed by several years, due to the time required to curate and quality control shipboard measurements.

Robotic platforms, such as profiling floats, equipped with pH sensors have the potential to vastly increase the number of observations in near-real time, and enable more timely assessment of the global ocean carbon sink. Currently there are over 300 pH profiling floats operating worldwide, and they can be used to estimate surface pCO₂ and dissolved inorganic carbon, which can then be used to quantify the ocean carbon sink. However, there are several key sources of uncertainty that need to be resolved to achieve the necessary accuracy from float measurements. This project addressed 4 of these sources of uncertainties.  

The first three are related to the thermodynamic model used for the marine CO₂ system. First, we developed a method to detect low levels of organic alkalinity in natural seawater, a component that is typically ignored in the thermodynamic model. We demonstrated that organic alkalinity is prevalent at low, but relatively constant concentrations from surface to deep ocean waters. Incorporating organic alkalinity into the thermodynamic model improves its performance. Second, we quantified the effects of pressure on equilibrium constants utilized in the thermodynamic model. These measurements were originally conducted in the 1960’s and have not been replicated. We used modern metrology and repeated these measurements, and got good agreement with historical values, suggesting that this is not a significant source of uncertainty. Finally, we propose an adjustment to the equilibrium constants K₁ and K₂ that significantly improves the thermodynamic model. Using these adjustments, we believe that the ocean carbon sink can be quantified through profiling float pH measurements with comparable uncertainty with other methods.

Finally, we improved methods to enable more accurate pH measurements from profiling floats. Specifically, two open-source algorithms (ESPER and TRACE) were developed that are used to predict reference deep pH values as part of the protocol to correct for pH sensor drift once floats have been deployed. These algorithms better incorporate the impacts of ocean acidification, leading to more accurate reference pH fields. Additionally, the algorithms have wide-spread applications outside of profiling floats, and is already being widely utilized by the oceanographic community.

 

Last Modified: 12/16/2025
Modified by: Yuichiro Takeshita