A comparison of methods for the determination of dissolved oxygen in seawater

An intercalibration of dissolved oxygen methods was conducted at 2 stationsin the Sargasso Sea between April 28 and May 3, 1990. The experiment comparedthree techniques using automated endpoint detection with the manual Winkler method using a starch endpoint. Institutions participating in theintercomparison were the Bedford Institute of Oceanography (automated photometrictitration), the University of Delaware (automated amperometric titration), theScripps Institution of Oceanography (manual titration), and the Woods HoleOceanographic Institution (automated amperometric titration).Differences in measured oxygen concentrations between institutions wereencouragingly small. However, small, systematic differences in dissolved oxygenbetween institutions did exist. The range between the highest and lowest oxygenvalues reported by the 4 institutions never exceeded 0.6% over the entireconcentration range studied (3.4 to 6.2 mlj1). The good agreement is probablydue to the use of the essentials of Carpenter's (1965) modification of theWinkler method by all institutions.The intercalibration revealed several aspects of dissolved oxygenmeasurements that require further research: (1) the intercalibration should beextended to very low oxygen concentrations; (2) procedures for measur ing andapplying corrections for the seawater blank need to be formalized; (3) a simpleprocedure to measure the temperature of seawater at the time of sampling needsto be developed; and (4) the solubility of atmospheric oxygen in the Winklerreagents must be measured as a function of temperature.The intercalibration also revealed that analytical techniques required forprecise and accurate volumetric measurements were often not applied, even byexperienced analysts. It was found that uncalibrated pipets were used todispense standards, that the volumes of oxygen flasks were not corrected forbuoyancy, and that corrections for the thermal expansion of aqueous solutions were often not applied.

This research was supported by National Science Foundation Grants OCE 88-22542 and OCE 88-21977 and OCE 89-07815. Preparation and distribution of thisreport by the WHP Office, Woods Hole Oceanographic Institution, Woods Hole, MA.02543 USA, was supported by NSF Grant OCE 89-07815.