This data was collected in February 2013 by the IMOS Ship of Opportunity Underway CO2 Measurement research group on MV L'Astrolabe (IMOS platform code: FHZI) voyage AL1213R4.
Departed: Hobart, Australia, February 19, 2013
Arrived: Dumont d'Urville station, Antarctica, February 25, 2013
CO2 System Overview:
The fugacity of carbon dioxide (fCO2) in surface seawater was measured using a General Oceanics Inc. automated system (Model 8050; Pierrot et al 2009). Seawater is sprayed into an equilibration chamber and CO2 in the headspace gas equilibrates with the seawater. The headspace gas is pumped through a thermoelectric condenser followed by a nafion drying tube, before flowing through a Licor 7000 non-dispersive infrared gas analyser used to measure the CO2 mole fraction (XCO2) of the dried air. The gas flow is stopped temporarily for the CO2 measurements, which are made at atmospheric pressure. A set of four CO2 standards (Table 1) that cover the range of CO2 values expected in the ocean are analysed about every three hours to calibrate the gas analyser. Atmospheric XCO2 (dry) is measured after the standards by pumping clean outside air from an intake on the forward mast of the ship.
The seawater intake is at about 3m depth in the bow of the ship. A remote temperature sensor (Seabird Electronics SBE38) located at the intake is used to measure sea surface temperature (SST). Sea surface salinity is measured using a thermosalinograph (Seabird Electronics SBE21) mounted in the bow thruster space near the intake. The travel time between the intake and CO2 system is typically about 50 seconds with warming usually less than 0.4ºC. The thermosalinograph water is from the same intake and supply line. Meteorological data, salinity, SST, and ships position and time are taken from the ship’s logging system. The salinity values measured by the SBE21 were checked against discrete bottle samples collected on the voyage.
Meteorological data, salinity, SST, and ships position and time are taken from the ships logging system.
Statement: Parameters logged by the fCO2 system and ship sensors are quality controlled after each voyage.
1. Data with missing parameters or obvious outliers for the ship or fCO2 system parameters are marked as missing and removed from the calculations. Parameter values are flagged as good (flag=2), questionable (flag=3), or bad (flag=4), depending on the range of values expected. Many of the ship and CO2 system parameters are not reported in the final dataset, but are used to establish that the system is functioning correctly. For example, water flow rates to the equilibrator below 2 LPM are flagged as questionable and the cause investigated with the flag value changed to 4 if the flow has been interrupted or is insufficient. Similar checks are made to ensure the gas flow through the infrared gas analyser is in a suitable range (50 to 120ml/min). The list of parameters checked are:
CO2 system data quality controlled:
GPS date and Time
Latitude and Longitude
Water flow rate
Gas flow rates through licor analyser
Equilibrator water temperature
Mole fraction CO2
Water vapour in gas stream
Licor NDIR temperature
Ship’s data quality controlled:
GPS date and time
Latitude and Longitude
Sea surface temperature
Sea surface salinity
Relative wind speed and direction
True wind speed and direction
2. The data sets are next evaluated for excessive warming of the seawater flowing to the equilibrator, and for contamination of the atmospheric measurements by ship stack gas.
The fCO2 value in the water is sensitive to warming between the ship intake and equilibrator. The travel time between the ship intake and equilibrator is first checked by comparing the timing of rapid changes in surface water temperature for the intake (SST) and the equilibrator temperatures. The travel time or lag time is normally about 50 seconds. The warming in the system used on MV l’Astrolabe is typically about 0.3 °C, increasing to about 0.4°C in cooler regions.
Atmospheric CO2 values can be influenced by contamination from ship stack gas. There are two atmospheric air intakes that are interchangeable with a manual valve. One intake is located above the wheelhouse to collect air in the front part of the ship and the other intake is located below the helicopter deck in the stern of the ship. The intake above the wheelhouse of the ship is within about 20m of the ship stacks. The relative wind speed and direction recorded by the ship meteorological sensors are used to select the best intake to use and later to evaluate if anomalous atmospheric measurements could be due to stack gas contamination. High XCO2ATM_PPM values due to stack gas is often observed at relative wind speeds below about 5 knots and relative wind direction perpendicular to the ship (in both directions 90 degrees or 270 degrees). Data with likely stack gas contamination are flagged as bad (flag = 4) and not included in the calculations outlined below.
3. After completion of the quality control checks, the measured mole fractions are corrected to final values using measurements of the four CO2-in-air standards (Table 1). The standards are run about every four hours to bracket the air and equilibrator measurements. The offsets between the measured and certified values of each standard are linearly interpolated to the times of measurement of the air and equilibrator samples. At each measurement time, a linear regression of offset values versus certified standard values is used to calculate the offset to apply to the measured air and equilibrator values. The corrections are typically small (about 1 to 2 ppm) and account for drift of the gas analyser response over time. The corrected mole fractions (dry) for the equilibrator and air samples flagged as good are then used to calculate the fugacity of CO2. Only data flagged as good or suspect are reported in the final data set.
A time lag of 320 seconds was calculated between the intake temperature and the equilibrator temperature (located in the pCO2 wet box). Intake temperature and salinity values were replaced with corresponding lagged values (thermosalinograph is located in the bow thruster’s space close to the seawater intake). A calibration to the TSG salinity values was also applied based on discrete bottle salinity samples.
There is insufficient seawater flow to the pCO2 system between the 22 Feb 2013 21:32 and 23 Feb 2013 21:13. These measurements were flagged ‘4’ and excluded from the calculations.
Please see the voyage dataset report for additional processing details and fugacity of carbon dioxide calculations (fCO2SW and fCO2ATM)..
Australia’s Integrated Marine Observing System (IMOS) is enabled by the National Collaborative Research Infrastructure Strategy (NCRIS). It is operated by a consortium of institutions as an unincorporated joint venture, with the University of Tasmania as Lead Agent.
Australian Marine National Facility (MNF)
CSIRO Marine and Atmospheric Research (CMAR)
Craig Neill and Alan Poole from CSIRO and the ship-based personnel are thanked for their effort and expertise in making the data collection possible.