The P6 Section really began in mid-April 1992 in Jacksonville, Florida, when BNL personnel (K. M. Johnson, R. Ramirez, and R. Wilke) placed two Single-Operator Multiparameter Metabolic Analyzers (SOMMA, S/Ns 004 and 006) and a gas chromatograph (GC) system for measuring discrete pCO2 aboard the R/V Knorr. The ship was berthed in Jacksonville for the final adjustments following its first shake- down cruise after refitting. The BNL CO2 group trucked its scientific gear to Jacksonville to save the cost and trouble of shipping it to Chile. Preliminary operational tests of the measurement systems were made, customs declarations filled out, and the instruments were secured for transit to Valparaiso before the BNL group departed ship. The equipment placed on board the ship in Jacksonville was used for the entire P6 Section so that subsequent staging was minimal.
The BNL CO2 group, consisting of K. M. Johnson, C. Neill, and K. Wills, departed New York on April 28, 1992, and arrived in Santiago, Chile, on the 29th. Upon arrival to the pier, the R/V Knorr was docking, and the group boarded the ship almost immediately and began the testing of the SOMMA and pCO2 GC systems placed aboard in Jacksonville. A number of problems developed during the testing phase. The SOMMAs were plagued with contamination and noise from unknown sources. Trouble-shooting was interrupted temporarily while the CO2 group moved aboard the ship on May 1, 1992, and the P6E Leg began with the departure of the ship from Valparaiso at 0900 on May 2 under the command of Captain Carl Swanson with Harry Bryden as Chief Scientist.
Trouble-shooting and repair continued, and two serious problems with the SOMMA systems were determined. The first finding was that very brief electrical surges were spontaneously causing the 8-port gas sample valve used in the gas calibration procedure to attempt to switch from the "load" to the "inject" position. The surges were not powerful enough to cause the valve to actually switch or cause the operator to hear the valve trying to switch, but they were strong enough to enable "cross talk" between the valve ports. Cross-talk momentarily connects the calibration gas, in this case pure CO2, to the carrier gas line, and the unwanted sporadic introduction of CO2 into the analytical stream interfered with sample titrations. This was only discovered after three very frustrating days by resting the palm of the hand on the gas sample valve chassis. In this way the silent and spontaneous surges could be felt through the metal valve cover. This problem was corrected by placing an uninterruptable power supply (UPS) in-line and by unplugging the gas sample valve from the mains when it was not needed. This left a second problem a much smaller, steadier, but still unsatisfactory source of noise manifesting itself as an unacceptably high blank. This noise resulted because the carrier gas cylinders of N2 were stored in a hangar external to the laboratory containing the SOMMA systems. Furthermore, thermoplastic lines (in some cases as long as 50 feet) were used to connect the cylinders to the SOMMA systems. These lines apparently allowed enough air (containing CO2) to diffuse through the tubing walls, thus interfering with the sample titrations. This problem was corrected by relocating the cylinders within the laboratory to within 6 to 10 feet of the SOMMAs. This relocation along with the installation of the normal CO2 scrubbers in the carrier gas lines reduced the blank to satisfactory levels. The pCO2 system also got off to a rocky start resulting from some initial software problems that required some code to be rewritten.
The first WOCE CTD station on P6E was station No. 4 with a depth of only 100 m at 32° 30' S and 71° 30' W. Stations 1 through 3 were not included in Section P6E so that there were 69 WOCE CTD stations on this leg. A CO2 group collected samples at 34 of the 69 stations (~ 50%). (see Fig. 2).
Some 1,043 individual water samples along with 98 duplicates and 92 samples for a certified reference material (CRM) were analyzed for TCO2. The pCO2 was determined on 323 of these samples. Because of initial set-up problems, the pCO2 sampling could not begin until Station 23.
As the analytical work proceeded during P6E, the members of the CO2 group became aware of an additional problem with one of the SOMMA systems (S/N 004). In this case, the recovery of the CO2 calibration gas during the gas calibration procedure was approximately 0.2% lower than expected based on prior experience with these systems (99.4 vs 99.6%). This lower recovery was associated with an analytical difference between the measured and CRM TCO2 of +3 to -4 µmol/kg (measured amount minus certified amount). This suggested that the gas-calibration was in error, however, the problem was identified and corrected. The relief measurement group was asked to bring a spare gas sample valve with them to Easter Island for installation at the conclusion of the P6E Leg.
The ship's operations following the retrofit were not without incident. In addition to the power surges mentioned, at least 2 complete power black-outs were experienced, the ship vibrated strongly, and there was noticeable buckling of specific deck plates. The vibration was of particular concern because of the possibility of damage to the computer hard disks. With the exception of the power outages, these problems did not cause any instrumental downtime. A lack of fuel availability on Easter Island required conservation measures such as lower ship speeds for the P6 Legs.
The P6E Leg ended with the anchoring of the R/V Knorr at 0800 on May 26, 1992, off the village of Hanga Roa on Easter Island. The first measurement group went ashore after lunch and obtained accommodations. The second measurement group consisting of R. Wilke, D. Hunter, and M. Anderson came aboard as the first group left. They carried with them some badly needed replacement parts including a gas sample valve, a refrigerated bath, silver electrodes, and fittings. Change-over between the groups took place on May 28 and 29, and the equipment was ready when the ship pulled anchor and departed on Leg P6C on May 30, 1992, with Mike McCartney as the Chief Scientist.
The second leg, P6C, which crossed the Southwest Pacific Basins and occupied stations from the East Pacific Rise ridge crest to the Kermadec Islands, began with Station 73 which was a reoccupation of the last station of the P6E Leg (72). Both Stations 73 and 74 are co- located with Station 72. The leg continued along 32µ 30' S from 112µ 40' W until 178µ 55' E, the location of Station 184. Thereafter P6C angled northward and ended with Station 188 at 31µ 05' S and 177µ 32' E. During this leg, additional duplicate stations were made such that Station 141 was co-located with 142 and 186 with 188. The CTD results from station 112 are unreliable because of the failure of the CTD. Of the 113 CTD stations occupied during P6C, 109 are unique WOCE stations. The CO2 group collected samples at 51 of the 109 stations (Fig. 3) or 47%. 1739 samples along with 157 duplicates and 163 CRM were analyzed for TCO2. Discrete measurements of pCO2 were made on 314 of these samples.
With respect to the TCO2 systems, the gas sample valve for system 004 was replaced before the start of section P6C. However, the recovery of the CO2calibration gas did not improve, and it remained constant and slightly low (99.4%) as on the first Leg P6E. No additional repairs were attempted. Otherwise sampling and analysis for the carbonate system parameters remained routine. Unfortunately, of the three legs comprising the P6 Section the P6C measurement group experienced the worst weather, and it was the longest of the three legs. The pCO2 system experienced numerous problems indicated by a loss of precision as this leg went on. Modifications were made in coordination with the shore-based BNL measurement group members, but precision continued to decline such that the last station run on P6C was Station 125 on June 19, 1992. In addition to poor precision, other problems reported included component temperature changes as a result of lack of sufficient insulation and traffic through the area where GC was located, exhaustion of the H2 generator's silica gel scrubber, unprogrammed gas sample valve switching in the GC (further evidence of electrical problems), dirty valve rotors, very rough weather that made trouble-shooting very difficult, and finally a thorough cleaning of the GC hardware on June 26 that resulted in the loss of all response upon restart. Daily efforts were made to bring the system back and response was gradually restored. It was not until the Knorr docked in Auckland and replacement parts brought by the relief crew were installed that the system was restored to function for the last Leg (P6W). Hence, pCO2 was not determined between Stations 126 and 190. The P6C station work concluded on July 4, and the R/V Knorr arrived at Auckland on July 7.
The relief crew from BNL, consisting of K. M. Johnson and V. Coles, were joined in Auckland by Bronte Tilbrook from the CSIRO Division of Oceanography located in Tasmania, Australia. Change-over and coordination between the relieved and relief groups occurred between July 7 and July 9. Complaints from several sources about the electric power on board ship on both Legs P6E and P6C lead to the acquisition and installation of a harmonic filter for the ship's "clean power" supply. Repairs were also made to one of the CTD winches. The ship departed from Auckland to take on fuel for the third and final Leg (P6W) at 0900 on July 13, 1992, with John Toole as Chief Scientist and Carl Swanson as Master. At 1600, following fueling, the R/V Knorr headed north and picked up where P6C left off with Station 189 which is co-located with Stations 188, 187, and 190. It continued northward to 30µ 05' S and 176µ 30' E. From 176µ 30' E the cruise continued along 30µ 05' S to the Australian coast finishing with Station 246 at 153µ 29' E. This station was followed by several additional CTD calibration stations (247-267) which are not considered or included in the WOCE data sets.
During P6W, additional duplicate casts were made such that Station 233 is co-located with 234, Station 246 with 247, and Stations 237 through 246 are co-located with Stations 248 through 257 and 258 through 267, so that of the 78 CTD stations occupied during P6W, only 55 represent unique WOCE stations. The COa2 group collected samples at 22 of these 55 stations (40%)(Fig. 4), and 502 water samples along with 47 duplicates and 41 CRM were analyzed for TCO2. Discrete measurements of pCO2 were made on 171 of these samples beginning with station 191. On this leg only one SOMMA (S/N 006) was used. The station work concluded on July 29, and the R/V Knorr arrived in Sydney on July 30, 1992 after uniformly good working conditions.
As on previous cruises, not all stations could be sampled for TCO2 and pCO2 parameters because of the lack of the time required for analysis. With 46% of the P6 stations sampled and profiled for TCO2, the goal of 50% coverage was essentially achieved. The coverage for pCO2 was considerably less (25%). The standard WOCE parameters (temperature, oxygen, nutrients, and salinity) were analyzed on all samples, and the carbonate system samples were normally drawn in conjunction with the tracer samples which included He, 14C, and the CFCs. Sampling frequency was at least every 30 nautical miles, but shorter intervals as a function of bathmetry shallow waters or changing depths required more stations were common especially on P6W. A minimum vertical depth resolution was maintained at 200 m throughout the P6 section. Table 3 presents a summary of the carbonate system measurements made on the WOCE section P6.