Abstract
Abstract. The determination of salinity by means of electrical conductivity relies on stable salt proportions in the North Atlantic Ocean, because standard seawater, which is required for salinometer calibration, is produced from water of the North Atlantic. To verify the long-term stability of the standard seawater composition, it was proposed to perform measurements of the standard seawater density. Since the density is sensitive to all salt components, a density measurement can detect any change in the composition. A conversion of the density values to salinity can be performed by means of a density–salinity relation. To use such a relation with a target uncertainty in salinity comparable to that in salinity obtained from conductivity measurements, a density measurement with an uncertainty of 2 g m−3 is mandatory. We present a new density–salinity relation based on such accurate density measurements. The substitution measurement method used is described and density corrections for uniform isotopic and chemical compositions are reported. The comparison of densities calculated using the new relation with those calculated using the present reference equations of state TEOS-10 suggests that the density accuracy of TEOS-10 (as well as that of EOS-80) has been overestimated, as the accuracy of some of its underlying density measurements had been overestimated. The new density–salinity relation may be used to verify the stable composition of standard seawater by means of routine density measurements.
Highlights
For almost 40 years, the salinity1 of seawater has been indirectly determined by means of electrical conductivity
Since the absolute conductivity cannot be measured as accurately as required for precise salinity measurements (Seitz et al, 2010), the conductivity has been measured relative to that of standard seawater2; the conversion to salinity is carried out by means of the conductivity–salinity relation PSS-78 (JPOTS, 1981a, b). This is achieved by calibrating salinometers and conductivity–temperature– depth devices using standard seawater, which is diluted to obtain the conductivity of the potassium chloride standard (Culkin, 1986; Bacon et al, 2007) used as a conductivity reference
The substitution method and the apparatus used for the density measurement are briefly outlined, as they have already been described in detail by Schmidt et al The uncertainty in density is discussed with regard to the uncertainty in salinity obtained from a density measurement and the subsequent calculation by means of the density– salinity relation
Summary
For almost 40 years, the salinity of seawater has been indirectly determined by means of electrical conductivity. The long-term comparability of salinity measurement results was discussed, with two main deficiencies being elaborated (Pawlowicz et al, 2016): a lack of traceability to a long-term stable and ubiquitous reference like the International System of Units (SI) and chemical composition variabilities in standard seawater These variabilities are likely to increase in the coming decades, due especially to the absorption of carbon dioxide into the ocean resulting from accumulation in the atmosphere (Millero, 2007). To remedy the deficiencies, Seitz et al (2011) proposed to perform routine measurements of the standard seawater density This would be achieved by determining the salinity of a standard seawater batch by conductivity measurement, and by density measurement; the conversion to salinity is carried out in this second approach by means of a density–salinity relation. The determination of salinity by means of conductivity is retroactively ensured in case of consistency; on the other hand, in case of inconsistency, a need for action is demonstrated
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
