Special Section: Global Analysis, Interpretation, and Modeling
An improved method for detecting anthropogenic CO2 in the oceans
Article first published online: 14 JUN 2010
Copyright 1996 by the American Geophysical Union.
Global Biogeochemical Cycles
Volume 10, Issue 4, pages 809–837, December 1996
How to Cite
1996), An improved method for detecting anthropogenic CO2 in the oceans, Global Biogeochem. Cycles, 10(4), 809–837, doi:10.1029/96GB01608., , and (
- Issue published online: 14 JUN 2010
- Article first published online: 14 JUN 2010
- Manuscript Accepted: 20 MAY 1996
- Manuscript Received: 19 JAN 1996
An improved method has been developed for the separation of the anthropogenic CO2 from the large natural background variability of dissolved inorganic carbon (C) in the ocean. This technique employs a new quasi-conservative carbon tracer ΔC*, which reflects the uptake of anthropogenic CO2 and the air-sea disequilibrium when a water parcel loses contact with the atmosphere. The air-sea disequilibrium component can be discriminated from the anthropogenic signal using either information about the water age or the distribution of ΔC* in regions not affected by the anthropogenic transient. This technique has been applied to data from the North Atlantic sampled during the Transient Tracers in the Ocean North Atlantic (TTO NAS) and Tropical Atlantic study (TTO TAS) cruises in 1981–1983. The highest anthropogenic CO2 concentrations and specific inventories (inventory per square meter) are found in the subtropical convergence zone. In the North Atlantic, anthropogenic CO2 has already invaded deeply into the interior of the ocean, north of 50°N it has even reached the bottom. Only waters below 3000 m and south of 30°N are not yet affected. We estimate an anthropogenic CO2 inventory of 20 ± 4 Gt C in the North Atlantic between 10°N and 80°N. The 2.5-dimensional ocean circulation model of Stocker et al.  and the three-dimensional ocean general circulation biogeochemistry model of Sarmiento et al.  predict anthropogenic CO2 inventories of 18.7 Gt C and 18.4 Gt C, respectively, in good agreement with the observed inventory. Important differences exist on a more regional scale, associated with known deficiencies of the models.