Measurements, Instrumentation, and Calibrations

The Princeton underway system was designed and constructed by OTL personnel for automated, high-resolution surface water and atmospheric boundary layer CO₂ concentration measurements. The system is controlled by a personal computer that is programmed to perform periodic calibrations, determine detector stability, and alternately measure the seawater and marine air CO₂ concentrations. A dual-beam infrared spectrometer (Li-Cor 6251) is used to measure the CO₂ concentration in the gas stream. The input gas to the detector (either one of four calibrated standards, air-equilibrated with surface seawater or marine air) is selected with an electronic 6-port valve. Prior to entering the detector, the gas is passed through a hygroscopic ion-exchange membrane (Nafion) and a small magnesium perchlorate/Aquasorb^TM column to remove water vapor. Marine air is pumped from the bow or stern of the research vessel to avoid contamination from the ship's exhaust. The surface water CO₂ concentration is determined by continuously pumping seawater from the ship's intake (depth ~7m) through the counterflow disk equilibrator. The equilibrator design is a modification of a disk stripper that has been shown to be very efficient at extracting soluble gases from seawater. Water flows through the bottom half of the chamber at a rate of approximately 18 L/min and is then dumped overboard. A fixed volume of air is recirculated through the top half of the chamber in the opposite direction as the water flow. Sixty disks are mounted on a stainless steel shaft that runs along the axis of the chamber. The disks are rotated at 135 rpm so they can pick up a thin film of water on either side, greatly increasing the surface area of the water. Thus, the chamber equilibrates a very large volume of water with a small fixed volume (~6 L) of air. With the rotating disks, the equilibrator's response to an instantaneous change in the CO₂ of the water is an exponential mixing function. Laboratory and "at sea" tests indicate that the response time for this system was approximately 1 minute. The precision of the measurements, estimated from times when the ship was not moving and multiple measurements were made at the same location, was estimated to be approximately 0.4 ppm. This is comparable to the precision obtained from standard gas and marine air measurements. The average water and air sampling frequencies for the Indian Ocean legs were ~2.5 and 9.5 min, respectively. Comparison of these measurements with those from an independent underway system operated by R. Weiss of Scripps Institution of Oceanography (SIO) on the same vessel agreed to better than 1 ppm (R. Weiss, personal communication, 1995). Further details of the underway system design and operation can be found in Sabine and Key (1996), which is reprinted in Appendix A of this documentation.

The infrared detector used during the Indian Ocean survey cruises had an instrumental drift that could be significant on the timescale of a day. The primary calibration method for this system, therefore, was the periodic analysis of gas standards having known CO₂ concentrations. A detailed description of the philosophies and mechanics of how the detector readings were calibrated is given in Appendix A. In addition to the accuracy of the CO₂ standard gases, the accuracy of the final results at in situ conditions depends on supporting measurements of temperature, pressure, and salinity. This section discusses the calibration of relevant parameters given in this report.