PREPARATION AND ANALYSIS OF AN EXTENSIVE HISTORIC DATA SET OF OCEAN CARBON DIOXIDE PARTIAL PRESSURE AND RELATED MEASUREMENTS

Ralph Keeling (SIO)

Link to NOAA Strategic Plan: NOAA’s Mission; Goal 2: Understand Climate Variability and Change to Enhance Society’s Ability to Plan and Respond; Goal 3: Serve Society’s Needs for Weather and Water Information

RESEARCH OBJECTIVES AND SPECIFIC PLANS TO ACHIEVE THEM

Our ultimate goal is to produce a data set of the partial pressure of CO2 gas (pCO2) in the ambient air and the surface ocean from measurements made on five expeditions in all of the major oceans during the period between October 1957 and September 1967 (Table 1). These measurements were made on ships operated by Scripps Institution of Oceanography (SIO) by personnel of the Carbon Dioxide Research Group (CDRG) under the direction of Charles D. Keeling. We plan to make final computations and produce a fully calibrated data set of the true pCO2 in the air and surface ocean as hourly averages and as a function of geographic location.

Scale differences on the CO2 analyzer between reference gas traces and those of ambient and equilibrated air, on an hourly frequency, will be entered into a database. Calibration data of the sensitivity of the instrument from scale differences of CO2 reference gases will also be entered as well as shipboard auxiliary data, consisting of ambient and equilibrator temperatures and barometric pressures along with geographic locations and salinities where available. Scale differences will be converted to CO2 index values using the gas calibration data and then to mole fractions and pCO2s in the water and the atmosphere using the manometric calibrations made at SIO over the years and auxiliary data. Ultimately a database of final results will be constructed and deposited with the National Oceanographic Data Center (NODC), operated by NOAA, and with the Carbon Dioxide Information and Analysis Center (CDIAC) of the U. S. Dept. of Energy.

RESEARCH ACCOMPLISHMENTS

Progress has been made on three of the five shipboard data sets as to location and assembly of original data, construction of the databases, and calculation of the CO2 mole fractions of ambient air and of air equilibrated with surface seawater. Table 1 lists the five expeditions and their spatial and temporal coverage of the surface ocean.

Most of the original shipboard data, and also preliminary workups, digitization of recorder strip chart records, and early reports, have been located in the files of the CO2 Research Program at SIO and in the SIO Library’s archives of research materials of the late Professor Charles Keeling. Shipboard CO2 reference gas data as well as early workups of the equilibrator data were found in an early report (“Shipboard Carbon Dioxide Project Report No. 3,” dated January 1, 1968).

In the 1980's, Lee Waterman, then located in Boulder, Colorado at the NOAA Climate Monitoring and Diagnostics Laboratory (now the Global Monitoring Division of Earth System Research Laboratory), began workup of the data encompassed by this project. By 1988 he had obtained from SIO all the data necessary to create a computerized database and calculation scheme for the first three expeditions, DOWNWIND, MONSOON, and LUSIAD. In 1996, with the assistance of Pieter Tans at NOAA, he produced a draft report on his workups, entitled “Quasi-simultaneous CO2 Measurements in the Atmosphere and Surface Ocean Waters from Scripps Institution of Oceanography DOWNWIND, MONSOON, and LUSIAD Expeditions 1957-1963.” Inquiries to Tans confirmed that the data sets from these three cruises had been worked up to hourly averages of CO2 in the ambient and equilibrated air. In January 2007 Tans sent these data sets to us in electronic form. The original proposal called for producing data sets with ½-hourly averages. In order to take advantage of Waterman's prior work, it seems more straightforward to proceed using hourly averages. Hourly resolution will suffice for the important scientific applications.

The calculation procedure given in Waterman’s draft report was confirmed for a few lines of hourly averaged data. The approach seems in general to be correct, although the details of algorithms and corrections need to be reviewed. The data workup had been completed as far as the mole fraction of CO2 in dried ambient and equilibrated air. Waterman discussed but did not complete the calculation of fugacities. Additional features of Waterman’s workup, such as corrections for temperature gains in the ships’ sea water lines, also will be reviewed and made to conform to modern schemes.

Daily infrared analyzer sensitivities were obtained from reference gas calibrations run at sea on a less-than-daily frequency. For LUSIAD Expedition, by far the longest of the five expeditions, estimation of these daily “Recorder scale factors” (RSFs) was complicated by substantial drifts with time. In the 1996 workup, Waterman divided the RSFs on the 15-month LUSIAD expedition into four periods represented by functions: a cubic function of time for the 1962 RSFs, a linear function for early 1963, a single determination for a short period, and finally a quadratic function for March to August 1963. The 1996 Waterman functions have been plotted in Figure 1 along with the individual data and seem to be reasonable, although subject to further review.

Table 1 List of SIO expeditions between 1957 and 1967 with concurrent oceanic and atmospheric pCO2 data.

Fig. 1 Nondispersive infrared analyzer CO2 sensitivities (“Recorder Scale Factors”) during LUSIAD expedition in 1962-1963. Fits to the data are shown for the several time periods.