The oxygen content of air-saturated fresh waters over ranges of temperature and atmospheric pressure of limnological interest

SIL Mitteilung No. 6 (MORTIMER 1956) and its aids for calculation of per cent oxygen saturation, widely distributed and still in use, were later shown to be based on erroneous experimental data. Four subsequent independent determinations (1963–1969) of the oxygen content of air-saturated pure water yielded consistent results, with a scatter in experimental points of not more than ±0.4 % over the temperature range 0 to 40°C. Centrally located within that scatter (Fig. 2) are the recent, much more accurate results of BENSON & KRAUSE (1980) based on determinations of the HENRY coefficient for oxygen (BENSON, KRAUSE & PETERSON 1979). When those results are translated into equilibrium solubility of atmospheric oxygen, a small additional uncertainty arises concerning the correction for molecular interactions in the air. Nevertheless, that uncertainty (± 0.04 % near 40 °C and ±0.07% at lower temperatures) is less than the experimental scatter in all previous estimates of equilibrium concentrations and much less than the uncertainties involved in field measurements of dissolved oxygen using the WINKLER titrimetric procedure. The BENSON & KRAUSE solubility tables (reproduced in Appendix 1) are therefore recommended as the standard for use in limnological applications, for which this review provides a commentary. Also presented are graphical and other aids for calculating equilibrium concentration (using, for example, equations 6 and 7 and a pocket calculator) or per cent saturation (nomogram. Fig. 5) over appropriate ranges of temperature (0 to 40°C) and atmospheric pressure (0.6 to 1.1 atm) or lake altitude (–0.8 to +4.0 km relative to sea level).