59
Composition of the Major Components of Seawater
equation is quite general for a given estuarine system formed by mixing river water
with seawater. Both the intercept and slope vary from location to location and from
time to time.
With the development of precise conductivity bridges in the 1950s, it became possible to
determine conductivity salinities to ±0.003‰. In 1961, nonthermostated commercial conductivity bridges became available. All these bridges gave conductivity ratios between the
sample and standard seawater (R = C sample / C std ) and used standard seawater to calibrate
the bridges. Although standard seawater was calibrated for chlorinity, it was not meant
to be a standard for conductivity. At about the same time, the old definition of salinity
came under question because of the uncertain accuracy and small number of samples
used. The Joint Panel for Oceanographic Tables and Standards (JPOTS), sponsored by
UNESCO (United Nations Educational, Scientific, and Cultural Organization), ICES, IAPSO
(International Association of Physical Sciences of Ocean), and SCOR (Scientific Committee
on Oceanic Research), was appointed to develop a conductivity standard for salinity.
Samples were collected from around the world. These samples were analyzed for their
chemical composition, chlorinity, and conductivity ratios. The JPOTS panel decided to
revise the old relationship between salinity and chlorinity using Equation 2.4, which is
equivalent to the original Knudsen definition at S = 35 or Cl = 19.374. Near S = 35, the
two equations are identical; however, at S = 32 or 38, the difference is 0.003‰.
S(‰) = 1.80655 Cl(‰)
(2.4)
Cox, Culkin, and Riley (1967) developed a relationship between the conductivity ratio at
15°C [R 15 = C 15 (sample)/C 15 (std seawater)] and the chlorinity of samples collected throughout the world. Samples from deeper than 200 m were deleted (due to the Ca 2+ affecting
the R 15 ). Since most of the deep samples had S of 34.8, a discontinuity occurred when the
results were deleted. The polynomial was adjusted by adding 0.0018 to make S = 35 and
R = 1.0. The Cl(‰) as a function of R 15 was converted to salinity using Cl = S/1.80655. The
resultant equation was
S(‰) = –0.08996 + 28.2970R 15 + 12.80832R 15
2 – 10.67869R 15
3 + 5.98624R 15
4 – 1.32311R 15
5 (2.5)
Unfortunately, this relationship was referred to as the new definition of salinity, while
in fact this polynomial merely expresses Cl in terms of R 15 and is strictly valid at S of 35,
which is equivalent to the old definition.
About the same time (1969) as the various organizations recommended the acceptance of
the so- called redefinition of salinity, in situ salinometers became commercially available.
Since the new definition only went to 10°C, it became necessary to use either extrapolated
values or equations based on the dilution of seawater with pure water. Many workers
attempted to use equations that joined these two methods.
Our laboratory examined many samples of standard seawater bottled from 1962 to 1975
for conductivity, salinity, and density. All the conductance measurements were made
relative to the same standard seawater (P 64 – 1973). Our results were in good agreement
(±0.0012 average deviation) with measurements made by Poisson (University of Paris).
The differences ranged from –0.002 to 0.008. The maximum spread of 0.0098 in salinity
represents the maximum error one would obtain in using the various standard seawater samples. No systematic correlations of the differences were found with the age of the
standard. These results pointed out the need to characterize the conductivity of seawater
relative to a KCl standard. To examine the causes of these differences, we measured the
Composition of the Major Components of Seawater
equation is quite general for a given estuarine system formed by mixing river water
with seawater. Both the intercept and slope vary from location to location and from
time to time.
With the development of precise conductivity bridges in the 1950s, it became possible to
determine conductivity salinities to ±0.003‰. In 1961, nonthermostated commercial conductivity bridges became available. All these bridges gave conductivity ratios between the
sample and standard seawater (R = C sample / C std ) and used standard seawater to calibrate
the bridges. Although standard seawater was calibrated for chlorinity, it was not meant
to be a standard for conductivity. At about the same time, the old definition of salinity
came under question because of the uncertain accuracy and small number of samples
used. The Joint Panel for Oceanographic Tables and Standards (JPOTS), sponsored by
UNESCO (United Nations Educational, Scientific, and Cultural Organization), ICES, IAPSO
(International Association of Physical Sciences of Ocean), and SCOR (Scientific Committee
on Oceanic Research), was appointed to develop a conductivity standard for salinity.
Samples were collected from around the world. These samples were analyzed for their
chemical composition, chlorinity, and conductivity ratios. The JPOTS panel decided to
revise the old relationship between salinity and chlorinity using Equation 2.4, which is
equivalent to the original Knudsen definition at S = 35 or Cl = 19.374. Near S = 35, the
two equations are identical; however, at S = 32 or 38, the difference is 0.003‰.
S(‰) = 1.80655 Cl(‰)
(2.4)
Cox, Culkin, and Riley (1967) developed a relationship between the conductivity ratio at
15°C [R 15 = C 15 (sample)/C 15 (std seawater)] and the chlorinity of samples collected throughout the world. Samples from deeper than 200 m were deleted (due to the Ca 2+ affecting
the R 15 ). Since most of the deep samples had S of 34.8, a discontinuity occurred when the
results were deleted. The polynomial was adjusted by adding 0.0018 to make S = 35 and
R = 1.0. The Cl(‰) as a function of R 15 was converted to salinity using Cl = S/1.80655. The
resultant equation was
S(‰) = –0.08996 + 28.2970R 15 + 12.80832R 15
2 – 10.67869R 15
3 + 5.98624R 15
4 – 1.32311R 15
5 (2.5)
Unfortunately, this relationship was referred to as the new definition of salinity, while
in fact this polynomial merely expresses Cl in terms of R 15 and is strictly valid at S of 35,
which is equivalent to the old definition.
About the same time (1969) as the various organizations recommended the acceptance of
the so- called redefinition of salinity, in situ salinometers became commercially available.
Since the new definition only went to 10°C, it became necessary to use either extrapolated
values or equations based on the dilution of seawater with pure water. Many workers
attempted to use equations that joined these two methods.
Our laboratory examined many samples of standard seawater bottled from 1962 to 1975
for conductivity, salinity, and density. All the conductance measurements were made
relative to the same standard seawater (P 64 – 1973). Our results were in good agreement
(±0.0012 average deviation) with measurements made by Poisson (University of Paris).
The differences ranged from –0.002 to 0.008. The maximum spread of 0.0098 in salinity
represents the maximum error one would obtain in using the various standard seawater samples. No systematic correlations of the differences were found with the age of the
standard. These results pointed out the need to characterize the conductivity of seawater
relative to a KCl standard. To examine the causes of these differences, we measured the