Improvements on the back-calculation technique for estimating anthropogenic CO2

• Published in: Deep Sea Research Part I: Oceanographic Research Papers
• Main Authors: Pérez, Fiz F., Álvarez, Marta, Ríos, Aida F.
• Format: Article in Journal/Newspaper
• Language: English
• Published: Elsevier 2002
• Subjects: Anthropogenic carbon; Performed alkalinity; North Atlantic ocean; CO2 disequilibrium; Alvarez; North Atlantic
• Online Access: http://hdl.handle.net/10261/52746; https://doi.org/10.1016/S0967-0637(02)00002-X

17 páginas. 3 tablas, 5 figuras The back-calculation technique for evaluating the anthropogenic CO2 (CANT) input corrects total inorganic carbon for the remineralisation of organic matter and the dissolution of calcium carbonate, along with subtracting the pre-industrial equilibrium total inorganic carbon. This technique is mainly dependent on the estimation of the preformed values of total alkalinity, oxygen and partial pressure of CO2 (pCO2). A new approach for estimating the preformed alkalinity (TA0) in the North Atlantic Ocean is suggested. TA0 is parameterised as a function of the inorganic to organic carbon decomposition ratio in the water column in agreement with sediment trap data and a recent review by Milliman et al. (Deep-Sea Res. I 46 (1999) 1653), which gives evidence for a considerable calcium carbonate dissolution in the upper 500–1000 m of the ocean, above the lysocline. We recalculated the CANT content in a section along the Eastern North Atlantic (OacesNAtlIIB-93) using this new TA0, the Mehrbach et al. (Limnol. Oceanogr. 8 (1973) 897) CO2 dissociation constants, and including the effect of water vapour pressure on estimation of the pre-industrial pCO2. Our findings lead us to conclude that the CO2 disequilibrium when water masses are formed is small and not significant. Direct and indirect evidence based on available high-quality CO2 data taken in formation areas in late winter conditions support this argument. M. Alvarez was financed by the ‘‘Xunta de Galicia’’ with a predoctoral grant. This work was supported by grant number CICYT-MAR97-0660. Peer reviewed