National Oceanic and
Atmospheric Administration
United States Department of Commerce


FY 1995

Measurement of fugacity of carbon dioxide in seawater: An evaluation of a method based on infrared analysis

Chen, H., R. Wanninkhof, R.A. Feely, and D. Greeley

NOAA Data Report ERL AOML-85, 49 pp (1995)

An analysis system is described to measure the fugacity of CO (fCO) in 500 mL of seawater (discrete samples) by infrared analysis. (The fugacity of CO is the partial pressure corrected for non-ideality of interaction between CO and N/O. Fugacity is 0.7 to 1.2 µatm lower than the corresponding partial pressure and depends primarily on temperature.) The unit is used on oceanographic research cruises sponsored by the Ocean-Atmosphere Carbon Exchange Study (OACES) of NOAA since 1991 for a total of approximately 10,000 measurements. The precision of the analyses based on replicate samples during routine analysis is 0.2%. Precision is primarily limited by sample storage, and the difference between headspace gas and water CO concentrations prior to equilibration. The precision and accuracy of the instrument is estimated in several different ways. A preliminary side by side test of two different discrete fCO systems using an infrared (IR) analyzer and a gas chromatograph (GC) show a variability of 0.8% without significant bias. For surface waters the discrete measurements are compared with measurements from a continuous flowing underway system which was deployed on all cruises. Select comparisons with fCO calculated from spectrophotometric pH and DIC (Total Dissolved Inorganic Carbon) measurements are performed. These comparisons are limited by the uncertainty in the magnitude and the temperature dependence of the carbon and borate dissociation constants. Indirect comparisons are made with pCO measurements of Drs. Takahashi and Chipman of LDEO in the South Atlantic and Equatorial Pacific (EqPac). Although the measurements were performed months (for the EqPac study in 1992) to years (for the South Atlantic study in 1989 and 1991) apart, some inferences can be made by normalizing the fCO values and by performing the comparison in property space. The comparison indicates that the precision of our system is comparable to that of the LDEO system. The surface values obtained from the LDEO system during EqPac are 1.5 to 3% lower. There is inconclusive evidence that some of the subsurface LDEO pCO values are lower as well. However, the comparison of subsurface values for the North Atlantic Deep Water in the South Atlantic study shows good agreement. Because of the current uncertainty in dissociation constants there is no a priori way to determine which value is correct.

Feature Publications | Outstanding Scientific Publications

Contact Sandra Bigley |