Total Carbon Dioxide Measurements

The TCO₂ was determined using two automated dynamic headspace SOMMA sample processors with coulometric detection of the CO₂ extracted from acidified samples. A description of the SOMMA-coulometry system and its calibration can be found in Johnson et al. (1987); Johnson and Wallace (1992); and Johnson et al. (1993). A schematic diagram of the SOMMA analytical system and sequence may be found in Johnson et al. (1993), and further details concerning the coulometric titration can be found in Huffman (1977) and Johnson et al. (1985). Samples were collected in 300-mL precombusted (450°C for 24 h) glass standard biological oxygen demand (BOD) bottles, poisoned with 200 µL of a 50% saturated solution of mercury chloride (HgCl₂), and analyzed for TCO₂ within 24 hours of collection (Handbook of Methods, DOE 1994). The samples were stored in a refrigerator in darkness at approximately 15°C until analyzed. Analyses of duplicate samples separated in time by up to 8 hours showed no evidence of any significant biological consumption or production of CO₂ during storage under these conditions. CRMs were routinely analyzed according to the DOE handbook (1994). The CRMs were supplied by Dr. Andrew Dickson of the Scripps Institution of Oceanography (SIO), and for this cruise batch 21 was used with a salinity of 34.54 and a certified TCO₂ of 1991.94 ± 0.79 µmol/kg. The CRM TCO₂ concentration was determined by vacuum extraction and manometry in the laboratory of C. D. Keeling at SIO. Some CRM were lost in transit probably as a result of overheating which compromised the bottle seals. However, most of the damaged seals could be seen with the naked eye, therefore effort was made to carefully inspect and analyze only uncompromised CRM.

Seawater was introduced into an acidified stripping chamber from an automated "to-deliver" (TD) pipette. The resultant CO₂ from continuous gas extraction was dried and coulometrically titrated on a model 5011 UIC coulometer. The coulometer was adjusted to give a maximum titration current of 50 mA, and the samples were run in counts mode. In counts mode, the number of pulses or counts generated during the titration by the coulometer's voltage to frequency converter (VFC) are recorded and displayed on a personal computer (PC). In each coulometer cell, the acid, hydroxyethylcarbamic acid [HO(CH₂)2NHCOOH], formed from the reaction of CO₂ and ethanolamine (C₂H₇NO), is titrated coulometrically to electrolytically generate hydrogen ions (OH^-) with photometric endpoint detection. The product of the time and the current passed through the cell during this titration (charge in coulombs) is related by Faraday's constant to the number of moles of OH^- generated and thus to the moles of CO₂ that reacted with ethanolamine to form the acid. The age of each titration cell is logged from its birth, which is the time that electrical current is applied to the cell, until its death, which is the time when the current is turned off. The age is measured as chronological age in minutes from birth and in mg of carbon (mgC) titrated since birth (i. e., carbon age).

Each system was controlled with an IBM-compatible PC equipped with two RS232 serial ports for the coulometer and barometer, a 24-line digital input/output (I/O) card for the solid state relays and valves, and an analog to digital (A/D) card for the temperature, conductivity, and pressure sensors. The cards were manufactured by Real Time Devices (State College, Penn.). The temperature sensors (model LM34CH, National Semiconductor, Santa Clara, Calif.) with a voltage output of 10 mV/°F built into the SOMMA were calibrated against thermistors certified to 0.01°C (PN CSP60BT103M, Thermometrics, Edison, N.J.) using a certified mercury thermometer as a secondary standard. These sensors monitored the temperature of SOMMA components, including the pipette, gas sample loops, and the coulometer cell. The SOMMA software was written in GWBASIC Version 3.20 (Microsoft Corp., Redmond, Wash.), and the instruments were driven from an options menu appearing on the PC monitor. With the coulometers operated in the counts mode, conversions and calculations were made using the SOMMA software rather than the programs and the constants hardwired into the coulometer circuitry.

The SOMMA-coulometry systems were calibrated with pure CO₂ using an eight-port gas sampling valve (GSV). The GSV had two sample loops of known volume determined gravimetrically by the method of Wilke et al. (1993). These two loops were connected to a source of pure CO₂ through an isolation valve with the vent side of the GSV plumbed to a barometer. When a gas loop was filled with CO₂, the mass (moles) of CO₂ contained therein was calculated by dividing the loop volume (V) by the molar volume of CO₂ at the ambient temperature (T) and pressure (P). The molar volume of CO₂ [V(CO₂)] was calculated from the gas constant (R), loop temperature (T), the instantaneous barometric pressure (P), and the first virial coefficient B(T) for pure CO₂:

V(CO₂) = RT / P x [1+ B(T) / V(CO₂)] .

The ratio of the calculated mass to that determined coulometrically, known as the gas calibration factor (CALFAC), was used to correct the subsequent titrations for small departures from 100% recoveries (DOE, 1994). Pressure was measured with a barometer, model 216B-101 Digiquartz Transducer (Paroscientific, Inc., Redmond, Wash.) which was factory-calibrated for pressures between 11.5 and 16.0 psi. The standard operating procedure was to make three sequential gas calibrations for each newly born titration cell (normally, one cell per day) at a carbon age of between 3 and 6 mgC titrated.

The "to-deliver" volume (V_cal) of the sample pipettes was determined and calibrated gravimetrically prior to the cruise and periodically during the cruise by collecting aliquots of deionized water dispensed from the pipette into preweighed serum bottles. The serum bottles were crimp sealed and returned to shore, where they were reweighed on a model R300S balance (Sartorius, Goettingen, Germany). The apparent weight (g) of water collected (W_air) was corrected to the mass in vacuo (M_vac) from:

M_vac = W_air + W_air (0.0012 / p - 0.0012 / 8.0),

where 0.0012 is the sea level density of air at 1 atm, p is the density of the calibration fluid at the pipette temperature and sample salinity, and 8.0 is the density of the stainless steel weights. The "to-deliver" volume (V_cal) was calculated by dividing the mass in vacuo (M_vac) by the density of the calibration fluid at the pipette temperature and sample salinity (p), as illustrated:

V_cal = M_vac / p .

The V_cal of the pipette for the BNL discrete system (004) calculated from the pipette aliquots taken during the cruise and weighed post cruise was 20.8386 ± 0.0044 mL (n = 12, Rel. Std. Dev. = 0.02%) at a calibration temperature (t_cal) of 22.24°C. The precruise volume could not be used because the original pipette installed on the BNL system was broken during transit and had to be replaced with a new pipette that had not been calibrated prior to the cruise. The precruise volume for the IfMK discrete system (014) was 21.4371 mL at 14.66°C. These pipette volumes were used for the calculation of the sample TCO₂. During the A8 section, the mean temperature of the sample pipettes was 21.5 ± 1.26°C (n = 1588). The sample "to-deliver" volume (V_t) at the measured pipette temperature was calculated from the expression:

V_t = V_cal [1 + a_v (t - t_cal)] ,

where a_v is the coefficient of volumetric expansion for Pyrex-type glass (1 x 10^-5 / °C), and t is the temperature of the pipette sample at the time of measurement.

The coulometers used to detect CO₂ were electronically calibrated as described in Johnson et al. (1993) and DOE (1994). Briefly, at least two levels of current (usually 50 and 2 mA) were passed through an independent and very precisely known resistance (R) for a fixed time. The voltage (V) across the resistance was continuously measured, and the instantaneous current (I) across the resistance was calculated from Ohm's law and integrated over the calibration time. Then, the number of pulses (counts) accumulated by the VFC during this time was compared with the theoretical number computed from the factory-calibration of the VFC [frequency = 10⁵ pulses (counts) generated per sec at 200 mA] and the measured current. If the VFC was perfectly calibrated at the factory, then the electronic calibration procedure would yield a straight line passing through the origin (intercept = 0 = INT_ec) with a slope (SLOPE_ec) of 1. The results for the precruise in-house electronic calibration and mean gas CALFAC for the coulometers on the Section A8 Cruise are given in Table 3. The BNL laboratory's electronic calibrations showed that the factory-calibration for both coulometers was almost perfect. The CALFAC remained very stable over the duration of the cruise (Rel. St. Dev. = 0.05% corresponds to a 1 µmol/kg change in sample TCO₂ concentration).

Table 3. The electronic calibration and the mean gas calibration coefficients for the coulometers used on WOCE Section A8

System	SLOPEec	INTec (µmol/min)	CALFAC (n)	S. D.	Rel. S. D. (%)
004	1.000490	0.000740	1.004000 (19)	0.000520	0.05
014	1.000052	0.0000584	1.002816 (38)	0.000408	0.04

The factory-calibration of the VFC and the value of the Faraday (96489 Coulomb/mol) yielded a scaling factor of 4.82445 x 10³ counts/µmol. The theoretical mass (M) of carbon titrated (M, in µmol) from water samples or the gas loops was calculated by (1) dividing the number of pulses or counts by the scaling factor, (2) subtracting the difference between the products of (a) the slope intercept (INT_ec) and the time in minutes during the titration where current flow was continuous (T_i) and (b) the system blank in mol per minute and the length of the titration in minutes; and then (3) dividing the resultant by the slope (SLOPE_ec). More simply stated:

M = [Counts / 4824.45 - (Blank x T_t) - (INT_ec x T_i)] / SLOPE_ec.

Note that the slope obtained from the electronic calibration procedure applied for the entire length of the titration, but the intercept correction applied only for the period of continuous current flow (usually 3 to 4 min) because the electronic calibration procedure can only be carried out for periods of continuous current flow. For water samples, the TCO₂ concentration in µmol/kg was calculated from:

TCO₂ = M x CALFAC x [ 1 / (V_t x p)] x d_Hg

where p is the density of seawater in g/mL at the measurement temperature and sample salinity calculated from the equation of state given by Millero and Poisson (1981), and d_Hg is the correction for sample dilution with bichloride solution (for Section A8 d_Hg = 1.00066).

System 004 was equipped with a conductance cell (Model SBE-4, Sea-Bird Electronics, Inc., Bellevue, Wash.) for the determination of salinity as described by Johnson et al. (1993). SOMMA and CTD salinity were compared to ensure that the salinity of the analyzed samples matched the assigned salinity. The agreement between CTD and SOMMA salinity was 0.04 or better. However, all calculations of TCO₂ in the last expression are based on the WOCE sample salinity furnished by the chief scientist.

The first phase of the three-phase quality control-quality assurance (QC-QA) process was assessed through the accuracy of the 77 CRM analyzed on board the ship on WOCE Section A8. These data are summarized in Table 4 and their temporal distribution during Section A8 is plotted in Fig. 3.

Table 4. Comparison of the at-sea mean analytical difference ( TCO₂ = measured certified) and the standard deviation of the differences between analyzed and certified TCO₂ on WOCE Section A8

System	No.* (n)	Mean (µmol/kg)	S. D. (µmol/kg)	δTCO2 (µmol/kg)	Outliers**
004	27	1992.23	0.92	0.29	0
014	50	1991.21	0.85	-0.73	1
All	77	1991.57	0.99	-0.37	1

* Batch 21 CRM with a certified TCO₂ of 1991.94 ± 0.76 µmol/kg (n = 10) at S = 34.54.
**See text for description.

Only one CRM analysis was considered an outlier and was dropped from the data set. This occurred on April 4 on System 014 for CRM No. 42 at a carbon age of 29 mgC titrated. The analytical difference was 8 µmol/kg. This might be explained as one of the compromised CRM (damaged seal) described earlier. After this result, CALFAC was redetermined, but it was not different than the factor originally determined at a carbon age of 6 mgC titrated. Hence there is no evidence for a change in system response as the cell prepared on 4 April aged. Although no visual damage was noted to the stopper seal of this bottle, this CRM exhibited an unusual amount of grease in and floating on the surface of the liquid phase, and it was suspected that it may have been compromised. However, no additional seawater samples were analyzed with this cell. Table 4 and Fig. 3 show that both systems gave very high accuracy throughout the A8 section with results virtually identical to the Certified TCO₂ with an overall precision of ± 0.99 µmol/kg (n = 77), which also compares favorably with the precision (± 0.76 µmol/kg; n = 10) of the bench mark vacuum extraction/manometric method. The response of both systems (Table 4) remained constant during the cruise.

The second phase of the QC-QA procedure was an assessment of precision that is presented in Table 5. The single-system precision was determined from samples with duplicates analyzed on the same system (either 004 or 014). The sample precision was calculated using duplicates that were analyzed on both systems (004 and 014).

Table 5. Precision of the discrete TCO₂ analyses on WOCE Section A8

Mean absolute difference						Pooled standard deviation
Sigma(bs) µmol/kg	S. D. µmol/kg	K	Sigma(bn) µmol/kg	S. D. µmol/kg	K	Sp2 µmol/kg	K	n	d.f.
Single-system precision
0.92	1.01	198	0.77	1.07	49	1.03	208	488	280
Sample precision
1.45	1.26	46	1.20	0.78	19	1.17	46	155	109

Single-system and sample precision have been separately assessed in Table 5 as:

• "between-sample" precision [Sigma(bs)], which is the mean absolute difference between duplicates (n = 2) drawn from the same Niskin bottle;
• "between-Niskin" precision [Sigma(bn)], which was the mean absolute difference between duplicates (n = 2) drawn from two different Niskin Bottles closed at the same depth;
• the pooled standard deviation (S_p²), calculated according to Youden (1951) where K was the number of samples with duplicates analyzed, n was the total number of replicates analyzed from K samples, and n - K was the degrees of freedom (d.f.).

Single-system precision provided a measure of drift in system response during a sequence of sample analyses. This is because the time elapsed between duplicate analyses on the same system using the same coulometer cell was deliberately kept at from 3 to 12 hours on the assumption that drift or change in response would be reflected in the single-system precision by an increase in the imprecision of the duplicate analyses. Sample precision, on the other hand, was measured because TCO₂ measurements were made on two separate systems and an estimate of overall sample precision for the section(s), independent of which analytical system was used, was required. Sample precision is the most conservative estimate of precision, incorporating several sources of random or systematic (bias) error.

As on other Sections in the North Atlantic (e.g. A10, A24, A20, A22) where SOMMA-coulometer systems have been run in parallel, the sample precision was slightly less precise than the single-system precision. Following established precedent for systems run in parallel the precision and accuracy of the TCO₂ determination on Section A8 was taken to be the pooled sample standard deviation (S_p²) of ± 1.17 µmol/kg given in Table 5. These data, showing an equivalent high precision between systems 004 and 014 and good agreement between single-system and sample precision, indicate that changes in system response during the coulometer cell lifetime nor system bias (see also Table 4) did not occur in either system during the cruise. The agreement between "between-sample" and "between-Niskin" precision indicates that there were no significant analytical effects due to gas exchange with the overlying headspace of the Niskin bottles during sampling. These findings were was consistent with data from other cruises (Johnson et al., 1996; 1998). The precision and accuracy of the TCO₂ determination on Section A8 was ± 1.17 µmol/kg. These data probably represent the maximum performance to be expected for these systems under field conditions.

The final step in the QC-QA procedure was the ship-to-shore comparison. Here sample duplicates were analyzed in real time at sea by continuous gas extraction/coulometry and later, after storage, on shore by vacuum extraction/manometry. The manometric analyses were completed by February 1995 in the laboratory of C. D. Keeling at SIO for 16 samples collected at 10 Section A8 stations. The results of the comparison are given in Table 6. Tables 4, 5, and 6 and Fig. 3 show an internally consistent TCO₂ data set for the A8 section with excellent accuracy, consistently good precision, no significant analytical bias between the systems, and excellent agreement between the at-sea and shore analyses. Hence no correction for instrumental bias or CRM differences has been applied to the data, and the TCO₂ data clearly meet the survey criterion for accuracy and precision. Additionally, the data submitted have not been normalized to a salinity of 35. Figure 4 summarizes the analytical results as a contour-section plot of the TCO₂ data from the WOCE Section A8 in Atlantic Ocean along 11.3° S.

The mean ship-to-shore analytical difference [δTCO₂(SIO), n = 16) was -1.62 ± 1.50 µmol/kg, and the mean absolute difference was 1.92 µmol/kg. The lower ship-based TCO₂ for Section A8 was consistent with previously reported results for A9, A1E, and A10 (Johnson et al. 1995, 1996, 1998) and for the program in general (Wallace 2002). The reason for ship-based TCO₂ being lower than the shore-based results is not known as of this time. While only three analyses shown in Table 6 were made on system 004, (δTCO₂(SIO) = - 0.66 µmol/kg) and 13 were made on the IfMK system 014 (δTCO₂ (SIO) = - 1.96 µmol/kg), the difference between the two systems was consistent with the sample precision (± 1.17 µmol/kg).

Table 6. Comparison of the at-sea analyses of TCO₂ by coulometry and the on-shore analyses of TCO₂ by manometry on aliquots of the same sample

Station	Date	Niskin (No.)	Depth (m)	TCO2 (at-sea) (µmol/kg)	TCO2 (SIO) (µmol/kg)	TCO2(sea-SIO) (µmol/kg)
232	21.04.94	324	12.9	2034.13(2)*	2035.26	-1.13
232**	21.04.94	311	2500.0	2186.15	2186.05	+0.10
234**	22.04.94	324	10.8	2058.46	2056.96	+1.50
238	23.04.94	311	2849.1	2192.00(2)	2195.11	-3.11
244	25.04.94	324	9.2	2050.01(2)	2051.39	-1.38
44	25.04.94	311	3501.9	2205.30(2)	2208.57	-3.27
248	26.04.94	324	12.9	2058.02(2)	2059.47	-1.45
248	26.04.94	311	3758.5	2208.84(2)	2213.14	-4.30
250**	27.04.94	311	3852.5	2209.01(2)	2212.60	-3.59
256	29.04.94	324	13.0	2056.62(2)	2058.27	-1.65
260	01.05.94	324	11.5	2012.12(2)	2013.60	-1.48
260	01.05.94	311	3001.1	2199.53(2)	2200.24	-0.71
264	02.05.94	324	12.3	2039.73	2039.99	-0.26
264	02.05.94	311	996.9	2235.96(2)	2238.27	-2.31
268	03.05.94	324	1.0	2020.53(2)	2021.10	-0.57
268	03.05.94	311	1597.7	2191.11(2)	2193.38	-2.27
Mean						-1.62
S. D.						±1.50***
n						16

*The number 2 in parentheses means the TCO₂ is the mean of two analyses. The SIO results are always the mean of 2 analyses
**Analyzed on System 004.
***The precision of the method was ± 1.17 µmol/kg.