Discrete pCO₂ Measurements

The discrete measurements of pCO₂ were performed by the LDEO group on three of four sections of the North Atlantic survey. During the WOCE sections A24, A20, and A22, a total of 2,465 samples were analyzed onboard the R/V Knorr (1,103, 595, and 767 samples respectively). On the earlier WOCE section AR24, discrete pCO₂ was not measured.

An automated equilibrator-IR gas analyzer system was used during the expedition for the determination of partial pressure of CO₂ in the seawater samples. Its design is similar to that described by Chipman, Marra, and Takahashi (1993) with the exception that the gas chromatograph was replaced with an IR gas analyzer. The equilibrator-IR system is shown schematically in Fig. 4.

The system consists of a circulation pump plumbed to recirculate air in a closed system through porous plastic gas dispersers immersed in a 250-mL seawater sample. The seawater sample is contained in a 250-mL Pyrex reagent bottle with a standard taper-ground glass stopper that serves as an equilibration vessel. A Pyrex extension tube (~20 mL), which has a standard taper-ground glass male-joint to form an airtight seal with the reagent bottle, is connected to the mouth of the reagent bottle to provide an extra headspace to prevent seawater from entering the gas circulation line. Four sets of flasks and circulation pumps are used so that four water samples can be processed concurrently. Because the partial pressure of CO₂ is sensitive to temperature, the equilibration flasks are kept immersed in a water bath maintained at 20C. The temperature at which the water sample is equilibrated with circulating gas is measured with a precision of 0.01C and is recorded.

An electrically driven Valco 10-port valve (the equilibrator selection valve in Fig. 4) is used to isolate each of the equilibrators during the initial equilibration. Manually operated 2-way and 3-way Whitey valves allow the headspace in each equilibrator to be filled with a calibration gas of known CO₂ concentration, creating a known initial condition for the headspace (about 40 mL) before equilibration. The equilibrator is open to the laboratory air through isolation coils attached to the low-pressure side of the equilibrator, keeping the total pressure of equilibration the same as the ambient atmospheric pressure. The atmospheric pressure is measured with a high-precision electronic barometer with an accuracy of better than 0.05% and is recorded. It takes about 20 minutes for each water sample to be thermally equilibrated with the constant-temperature water bath, and the headspace gas is recirculated through the water sample throughout the period to ensure CO₂ equilibration.

An electrically driven Valco 6-port valve (the sample selection valve in Fig. 4) is connected to the equilibrator selection valve and to the calibration gas selection valve. This allows selection of the gas sample to be analyzed for CO₂: the equilibrated sample gas or one of the four calibration gases. A 2-way normally-closed Skinner solenoid valve on the output of the calibration gas selection valve controls the flow of the calibration gases to the sample selection valve. It also provides a necessary second means of stopping the flow of the calibration gases to prevent their accidental loss in case of a control malfunction. The concentration of CO₂ in the gas equilibrated with the seawater sample is determined using an IR gas analyzer (LICOR Model 6125) in a flow-through mode. A 0.5-mL aliquot of equilibrated headspace gas, representing less than 1% of the circulating gas, is isolated using a gas pipette (attached to the sampling valve in Fig. 4) and swept with CO₂-free air (or pure nitrogen gas) flowing at a constant rate of about 50 mL/min. For low-pCO₂ samples, a 1-mL gas pipette (attached to the sampling valve) is used. The sample gas is passed through a permeation drying tube for the removal of water vapor and injected into the IR gas analyzer cell (about 7 mL in volume) filled previously with CO₂-free air. The displaced CO₂-free air is discharged out of the cell into the laboratory. The small volume of the gas sample ensures that all of the CO₂ from the gas pipette is found in the analyzer cell at the same time, so that the peak height is proportional to the amount of CO₂ present in the gas pipette. Drying of the sample gas avoids the effects of pressure-broadening of the CO₂ absorption spectra and of dilution caused by water vapor. The amount of CO₂ in the sampling pipette is a function of the loop volume, temperature, and pressure. The temperature is held constant and measured, and the pressure of the sample gas is same as the barometric pressure, which is measured with an accuracy of better than 0.05%. The peak height, which represents the number of moles of CO₂ in the sample gas, is calibrated every 1.5 hours using a quadratic equation fitted to three calibration gas mixtures (366.52, 788.8 and 1211.4 ppm mole fraction in dry air).

The analytical procedure begins with water samples being drawn from the 10-L Niskin bottles off a rosette directly into 250-mL Pyrex reagent bottles. These served as both sample containers and equilibration vessels. The samples were immediately inoculated with 100 µL of 50% saturated mercuric chloride solution, sealed airtight with ground glass stoppers to prevent biological modification of the pCO₂, and stored in the dark until analysis. Measurements were normally performed within 24 hours of sampling. A headspace of 3 to 5 mL was left above the water to allow for thermal expansion during storage. Prior to analysis, the sample flasks were brought to the water bath temperature of 20°C in the constant-temperature bath. The equilibrator headspace, including the extension tube and the gas circulation tubings, was filled with a calibration gas of known CO₂ concentration. The gas in the equilibrators, and in the tubing that connects them to the gas pipette loop, was recirculated continuously for about 20 minutes through a gas disperser immersed in the water. This provided a large surface area for gas exchange between the sample water and circulating gas, and equilibrium for CO₂ was attained in 15 min. The temperature of the bath water was assumed to be that of the sample water and was measured at the time of equilibration with a precision of 0.01°C using a thermometer calibrated against a NIST-certified thermometer. This temperature is reported in the data tables as TEMP_PCO2 and showed no variation at a limit of 0.01°C.

The equilibrated air samples were saturated with water vapor at the temperature of equilibration and had the same pCO₂ as the water. By injecting the air aliquot into the IR analyzer after the water vapor was removed, the concentration of CO₂ was measured. Therefore, the effect of water vapor must be taken into consideration for computing pCO₂ as follows:

pCO₂ (atm) = [C_meas (ppm)] x [total press. of equilibration (atm) - water vapor press. (atm)]

where C_meas is the mole fraction concentration of CO₂ in dried equilibrated air. The total pressure of equilibrated air is measured by having the headspace in the equilibrator flask always at atmospheric pressure. The latter was measured with an electronic barometer at the time each equilibrated air sample was injected into the IR analyzer for CO₂ determination. The water vapor pressure was computed at the equilibration temperature, and salinity of the seawater. C_meas was determined by using a quadratic equation fit to three of the calibration gas mixtures.

The concentrations for standard gases used are traceable to the WMO reference scale through analysis in the laboratories of C. D. Keeling of SIO (La Jolla, California) and of Pieter P. Tans of NOAA/CMDL (Boulder, Colorado). The values of the standard gas mixtures used during this cruise were 366.52 ppm CO₂, 788.0 ppm CO₂, and 1211.4 ppm CO_2.

Corrections were made to account for the change in pCO₂ of the sample water due to the transfer of CO₂ between the water and circulating air during equilibration. We know the pCO₂ in equilibrated, perturbed water and the TCO₂ by coulometry before the equilibration. We can also calculate the change in TCO₂ in the water based on the change in pCO₂ between the post-equilibrium value and the known concentration in the pre-equilibrium condition. With the pre-equilibrium TCO₂ plus the perturbation in TCO₂ during equilibration, the post-equilibrium TCO₂ value was obtained. Using the post-equilibrium TCO₂ and measured pCO₂ values, TALK at the end of the equilibration was calculated, using the temperature, salinity, phosphate, and silicate data. Since the perturbation does NOT change the TALK, the pre-equilibrium pCO₂ from the pre-equilibrium TCO₂, the calculated TALK, and the temperature, salinity, etc., were calculated. This is the value that was reported as pCO₂, the pre-equilibrium calculated value. The magnitude of this correction is generally less than 2 atm. Details of the computational scheme are presented in a DOE technical report by Takahashi, et al. (1998).

The pCO₂ values reported in this data set are expressed as micro-atmospheres at the temperature of equilibration. The precision of the pCO₂ measurement for a single hydrographic station was estimated to be about 0.15% based on the reproducibility of replicate equilibrations. The station-to-station reproducibility was estimated to be about 0.5%.