RB-07-11 CLIVAR P18 Leg 1 Readme File [RB-07-11 included the TAO-110W operations] Ship: NOAA research vessel Ronald H. Brown Cruise Start: San Diego, CA, Dec. 15, 2007 Cruise End: Easter Island, Jan. 18, 2008 Chief Scientist: John Bullister System Operator: Robert Castle Expocode: 33RO20071215 Method: Infrared absorption of dried gas. For details of the system see: Measurement of fugacity of Carbon Dioxide in surface water and air using continuous sampling methods. Wanninkhof and Thoning, 1993 in Marine Chemistry 44, 189-205, And: Feely, R.A., R. Wanninkhof, H.B. Milburn, C.E. Cosca, M. Stapp, and P.P. Murphy, A new automated underway system for making high precision pCO2 measurements onboard research ships, Analytica Chim. Acta, 377, 185-191, 1998. The three standard gases come from CMDL in Boulder and are directly traceable to the WMO scale. Sampling Cycle: The system runs on an hourly cycle during which 3 standard gases, 3 air samples from the bow tower and 8 surface water samples (from the equilibrator head space) are analyzed on the following schedule: Mins. after hour Sample 4 Low Standard 8 Mid Standard 12 High Standard 16.5 Water 21 Water 25.5 Water 30 Water 34 Air 38 Air 42 Air 46.5 Water 51 Water 55.5 Water 60 Water Units: All xCO2 values are reported in parts per million (ppm) and fCO2 values are reported in microatmospheres (uatm) assuming 100 % humidity at the equilibrator temperature. Notes: 1. Any values outside the range of the standards (289.06, 370.90, & 514.29 ppm) should be considered approximate (within 5 ppm). While individual data points above 411 or below 289 may not be accurate, the general trends should be indicative of the seawater chemistry. 2. The salinity readings from the ship's TSG were inaccurate and were replaced with the salinity readings from the Seabird Micro TSG in the Hydro Lab sink. 3. The millivolt response of the calibration standards was incorrect due to low gas flow which caused incomplete flushing of the Licor sample cell. The problem was corrected on Jan. 9 at 1600 and was apparently caused by a gas flow meter that was reading ~40 ml/min too high. During the period from Jan. 2 to Jan. 8, reference gases of concentrations 411.42 and 287.95 were passed through the system as samples for several hours. The millivolt response of these gases has been used to correct the millivolt response of the standards during the period of low gas flow. All air and equilibrator headspace samples prior to Jan. 9 at 1600 have been flagged as 3 or questionable for this reason. Because of the incomplete flushing, gas from the preceding measurement phase was present in the sample cell. For the mid and high standards this gas was lower in concentration and caused the voltage response to be too low. For the low standard, the preceding gas was equilibrator head space gas and was always higher in concentration, although it varied from 345 to 575 ppm over the course of the cruise. The low standard voltage response was set to 3.75 mV for the entire section of incomplete flushing. The mid and high responses were calculated using the equation adjusted voltage = original voltage + 16 mV + 0.6mV * (80 - f) where f is the recorded gas flow in ml/min. The last term was added to account for the fact that the higher the gas flow (which ranged from 45 to 55 ml/min), the more complete the flushing and the less depression in the response. The corrections were arrived at using a process of trial and error and were based on responses of recent cruises using the same set of standards. The xCO2 values of the standards that were run as samples and the air values were then examined to determine the equation that best matched the expected values. 4. The following calibration gases were run through the system as samples: Jan. 2 2334 - 2355 411.42 ppm Jan. 3 1825 - 2100 411.42 ppm Jan. 6 1111 - 1600 411.42 ppm Jan. 6 1611 - 2200 287.95 ppm Jan. 8-9 2211 - 0530 411.42 ppm 5. From Jan. 5 at 1840 to Jan 6 at 1045, gas flow in the water phase was low due to an apparent blockage in the tubing from the pump. The first 3 of each set of 4 headspace gas samples has been removed during this period. The data indicates that there was enough gas flow to flush the Licor sample cell by the time the fourth sample was measured. 6. Atmospheric air samples were higher than expected (386-405 ppm) for about a week after the ship left port and did not fall below 386 ppm until around Dec. 23. As no indication of stack gas contamination can be found they are assumed to be caused by shoreside sources and have been left in. As with all samples in this time period they have been flagged as 3 (questionable). 7. On Dec. 23 at 1750 the air input line to the bow mast was changed to a second air line that had previously been run to see if contamination in the line was responsible for the high air values. 8. On Dec. 30 the glass wool in the filter at the air line intake on the bow mast was replaced. 9. On Jan. 3 at 2123 the system was shut down and a bad solenoid valve on the Licor sample input was replaced. The system was restarted at 2201. 10. On Jan 4 at 2314 the water was shut off and the equilibrator showerhead was cleaned. Water was turned back on at 2323. 11. On Jan. 7 at 1528 the system was shut down and the Licor was replaced with the spare after first setting its zero and span. The system was turned back on at 2245. 12. On Jan. 8 the system was shut down from 0210 to 2125 to clean the rotor of the Valco valve. 13. On Jan. 9 at 1600 the gas flow of the calibration gases was increased to ~ 100 ml/min. This resulted in air values in the expected range (~383 ppm). 14. On Jan. 9 from 2045 to 2200 the system was shut down to reinstall the original Licor and to replace the nylon tubing to its inputs and outputs. 15. During February, 2017, the data was reexaminded and rearranged slightly for easy resubmission to international databases. The numerical values were not changed; the data file structure including column arrangement was reformatted. The mole fraction of CO2 in air is listed close to the time of its measurement, and the fugacity of CO2 in air based on this measurement is listed along side of the following water analyses. The fugacity of CO2 in water is listed alongside of the closest, preceeding good measurement of CO2 in air. The only QC flag listed is the WOCE_QC_FLAG, which is for the entire, particular data record. For questions or comments contact: Kevin Sullivan 4301 Rickenbacker Causeway Miami, FL 33149 305-361-4382 kevin.sullivan@noaa.gov