Reduced air–sea CO2 exchange in the Atlantic Ocean due to biological surfactants

This is the author accepted manuscript. The final version is available from Springer Nature via the DOI in this recordOcean CO2 uptake accounts for 20–40% of the post-industrial sink for anthropogenic CO2. The uptake rate is the product of the CO2 interfacial concentration gradient and its transfer velocity, which is controlled by spatial and temporal variability in near-surface turbulence. This variability complicates CO2 flux estimates and in large part reflects variable sea surface microlayer enrichments in biologically derived surfactants that cause turbulence suppression. Here we present a direct estimate of this surfactant effect on CO2 exchange at the ocean basin scale, with derived relationships between its transfer velocity determined experimentally and total surfactant activity for Atlantic Ocean surface seawaters. We found up to 32% reduction in CO2 exchange relative to surfactant-free water. Applying a relationship between sea surface temperature and total surfactant activity to our results gives monthly estimates of spatially resolved ‘surfactant suppression’ of CO2 exchange. Large areas of reduced CO2 uptake resulted, notably around 20° N, and the magnitude of the Atlantic Ocean CO2 sink for 2014 was decreased by 9%. This direct quantification of the surfactant effect on CO2 uptake at the ocean basin scale offers a framework for further refining estimates of air–sea gas exchange up to the global scale.This work was supported by grants from the Leverhulme Trust to R.C.U.G. (RPG-303) and the UK Natural Environment Research Council (NERC) to R.C.U.G. (NE/K00252X/1) and J.D.S. (NE/K002511/1). Both NERC grants are components of RAGNARoCC (Radiatively Active Gases from the North Atlantic Region and Climate Change), which contributes to NERC's Greenhouse Gas Emissions and Feedbacks programme (www.nerc.ac.uk/research/funded/programmes/greenhouse). J.D.S. and I.A. acknowledge additional support from the European Space Agency (grant 4000112091/14/I-LG). R.P. acknowledges support from T. Wagner. This study is a contribution to the international IMBeR project and was supported by UK NERC National Capability funding to Plymouth Marine Laboratory and the National Oceanography Centre, Southampton. This is contribution no. 324 of the AMT programme