Anthropogenic CO2 in the oceans estimated using transit time distributions

The distribution of anthropogenic carbon (C ant ) in the oceans is estimated using the transit time distribution (TTD) method applied to global measurements of chlorofluorocarbon-12 (CFC12). Unlike most other inference methods, the TTD method does not assume a single ventilation time and avoids the...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Tellus. Series B, Chemical and physical meteorology Chemical and physical meteorology, 2006-11, Vol.58 (5), p.376-389
Hauptverfasser: Waugh, D. W., Hall, T. M., Mcneil, B. I., Key, R., Matear, R. J.
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The distribution of anthropogenic carbon (C ant ) in the oceans is estimated using the transit time distribution (TTD) method applied to global measurements of chlorofluorocarbon-12 (CFC12). Unlike most other inference methods, the TTD method does not assume a single ventilation time and avoids the large uncertainty incurred by attempts to correct for the large natural carbon background in dissolved inorganic carbon measurements. The highest concentrations and deepest penetration of anthropogenic carbon are found in the North Atlantic and Southern Oceans. The estimated total inventory in 1994 is 134 Pg-C. To evaluate uncertainties the TTD method is applied to output from an ocean general circulation model (OGCM) and compared the results to the directly simulated C ant . Outside of the Southern Ocean the predicted C ant closely matches the directly simulated distribution, but in the Southern Ocean the TTD concentrations are biased high due to the assumption of 'constant disequilibrium'. The net result is a TTD overestimate of the global inventory by about 20%. Accounting for this bias and other centred uncertainties, an inventory range of 94-121 Pg-C is obtained. This agrees with the inventory of Sabine et al., who applied the ΔC * method to the same data. There are, however, significant differences in the spatial distributions: The TTD estimates are smaller than ΔC * in the upper ocean and larger at depth, consistent with biases expected in ΔC * given its assumption of a single parcel ventilation time.
ISSN:0280-6509
1600-0889
DOI:10.1111/j.1600-0889.2006.00222.x