Evaporites and the Salinity of the Ocean During the Phanerozoic: Implications for Climate, Ocean Circulation and Life

Abstract

A compilation of data on volumes and masses of evaporite deposits is used as the basis for reconstruction of the salinity of the ocean in the past. Chloride is tracked as the only ion essentially restricted to the ocean, and past salinities are calculated from reconstructed chlorine content of the ocean. Models for ocean salinity through the Phanerozoic are developed using maximal and minimal estimates of the volumes of existing evaporite deposits, and using constant and declining volumes of ocean water through the Phanerozoic. We conclude that there have been significant changes in the mean salinity of the ocean accompanying a general decline throughout the Phanerozoic. The greatest changes are related to major extractions of salt into the young ocean basins which developed during the Mesozoic as Pangaea broke apart. Unfortunately, the sizes of these salt deposits are also the least well known. The last major extractions of salt from the ocean occurred during the Miocene, shortly after the large scale extraction of water from the ocean to form the ice cap of Antarctica. However, these two modifications of the masses of H2O and salt in the ocean followed in sequence and did not cancel each other out. Accordingly, salinities during the Early Miocene were between 37‰ and 39‰. The Mesozoic was a time of generally declining salinity associated with the deep sea salt extractions of the North Atlantic and Gulf of Mexico (Middle to Late Jurassic) and South Atlantic (Early Cretaceous). The earliest of the major extractions of the Phanerozoic occurred during the Permian. There were few large extractions of salt during the earlier Palaeozoic. The models suggest that this was a time of relatively stable but slowly increasing salinities ranging through the upper 40‰'s into the lower 50‰'s.

Full text

Evaporites and the salinity of the ocean during the Phanerozoic:
Implications for climate, ocean circulation and life
William W. Hay
a,
⁎, Areg Migdisov
b,✠
, Alexander N. Balukhovsky
b
,
Christopher N. Wold
c
, Sascha Flögel
d
, Emanuel Söding
e
a
2045 Windcliff Dr., Estes Park, CO 80517, USA
b
VI Vernadski Institute of Geochemistry and Analytical Chemistry, Russian Academy of Sciences, Kosygin 19, Moscow 119991, Russia
c
Platte River Associates, 2790 Valmont Road, Boulder, CO 80304, USA
d
Leibniz-Institute of Marine Sciences (IFM-GEOMAR), Wischhofstrasse 1-3, D-24148 Kiel, Germany
e
Integrated Ocean Drilling Program Management International, Inc., Sapporo Office, Creative Research Initiative “Sousei”(CRIS),
Hokkaido University, N21W10 Kitaku, Sapporo 001-0021, Japan
Received 23 March 2005; accepted 24 March 2006
Abstract
A compilation of data on volumes and masses of evaporite deposits is used as the basis for reconstruction of the salinity of the
ocean in the past. Chloride is tracked as the only ion essentially restricted to the ocean, and past salinities are calculated from
reconstructed chlorine content of the ocean. Models for ocean salinity through the Phanerozoic are developed using maximal and
minimal estimates of the volumes of existing evaporite deposits, and using constant and declining volumes of ocean water through
the Phanerozoic. We conclude that there have been significant changes in the mean salinity of the ocean accompanying a general
decline throughout the Phanerozoic. The greatest changes are related to major extractions of salt into the young ocean basins which
developed during the Mesozoic as Pangaea broke apart. Unfortunately, the sizes of these salt deposits are also the least well known.
The last major extractions of salt from the ocean occurred during the Miocene, shortly after the large scale extraction of water from
the ocean to form the ice cap of Antarctica. However, these two modifications of the masses of H
2
O and salt in the ocean followed
in sequence and did not cancel each other out. Accordingly, salinities during the Early Miocene were between 37‰and 39‰. The
Mesozoic was a time of generally declining salinity associated with the deep sea salt extractions of the North Atlantic and Gulf of
Mexico (Middle to Late Jurassic) and South Atlantic (Early Cretaceous). The earliest of the major extractions of the Phanerozoic
occurred during the Permian. There were few large extractions of salt during the earlier Palaeozoic. The models suggest that this
was a time of relatively stable but slowly increasing salinities ranging through the upper 40‰'s into the lower 50‰'s.
Higher salinities for the world ocean have profound consequences for the thermohaline circulation of the ocean in the past. In
the modern ocean, with an average salinity of about 34.7‰, the density of water is only very slightly affected by cooling as it
approaches the freezing point. Consequently, salinization through sea-ice formation or evaporation is usually required to make
water dense enough to sink into the ocean interior. At salinities above about 40‰water continues to become more dense as it
approaches the freezing point, and salinization is not required. The energy-consuming phase changes involved in sea-ice formation
and evaporation would not be required for vertical circulation in the ocean.
The hypothesized major declines in salinity correspond closely to the evolution of both planktonic foraminifera and calcareous
nannoplankton. Both groups were restricted to shelf regions in the Jurassic and early Cretaceous, but spread into the open ocean in
Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46
www.elsevier.com/locate/palaeo
⁎Corresponding author.
E-mail address: whay@gmx.de (W.W. Hay).
✠
April 3, 2003.
0031-0182/$ - see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.palaeo.2006.03.044

the mid-Cretaceous. Their availability to inhabit the open ocean may be directly related to the decline in salinity. The Permian
extraction may have created stress for marine organisms and may have been a factor contributing to the end-Permian extinction.
The modeling also suggests that there was a major salinity decline from the Late Precambrian to the Cambrian, and it is tempting to
speculate that this may have been a factor in the Cambrian explosion of life.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Salinity; Salt; Palaeoceanography; Phanerozoic; Sedimentary cycling
1. Introduction
The history of the salinity of the ocean has been a
matter of inquiry since the end of the 19th century. Irish
Physicist John Joly (1899) used the present salinity of
the ocean and the present rate of supply of salt by rivers
to estimate the age of the Earth. Using the methodology
shown in Fig. 1, he determined that these ions would
accumulate in the ocean to their present level in about
90 million years. He reasoned that this must therefore be
the age of the ocean, and that the planet could not be
much older. His estimate for the age of the earth, being
in close agreement with that of Lord Kelvin (1864) was
widely accepted by the non-geological community. He
held to this argument, rejecting even his own radiomet-
ric evidence for a much greater age for the planet,
through at least the first quarter of the 20th century (Joly,
1925).
The amount of sodium chloride in the ocean is only
about 10% that of saturation, so it might be expected that
the steady supply of salts to the ocean would cause its
salinity to increase continuously. However, it has not
been known how the supply of salts might have changed
over time. Further, it has been recognized that there were
numerous salt deposits in the geologic record and that
these must represent extractions of salt from the ocean.
With the lack of quantification of supply and
extraction of salts from the ocean, many geologists
and palaeontologists have assumed that the salinity of
the ocean has remained essentially constant near its
Fig. 1. Schematic version of Joly's (1899) method for determining the age of the Earth from ocean salinity and rate of salt supply.
4W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

present value of 34.7‰through the Phanerozoic
(Railsback et al., 1989). Others have made a correction
for the present mass of fresh water in glacial ice on
Antarctica and Greenland (Shackleton and Kennett,
1975; Shackleton, 1987; Duplessy et al., 1991, 1993),
assuming a global mean salinity of 34.03‰during ice-
free times. Others have constructed more complex
schemes taking into account the gradual buildup of
Antarctic ice (Zachos et al., 1994). However, the
extractions of salt from the ocean to form evaporite
deposits have been generally disregarded, although it
has been recognized that they must have a significant
effect on ocean salinity (Southam and Hay, 1981).
In this paper we attempt a quantitative reconstruction
of the salinity of the ocean through the Phanerozoic,
based solely on deposition and sedimentary cycling of
evaporite deposits. However, this may be only part of
the salinity history of the ocean because there is another
potentially large reservoir of salt stored in brines in the
pore space of deeply buried sediments on the continental
blocks.
2. Evaporite deposits
2.1. Early estimates of the size of evaporite deposits
The first detailed data on the volumes of evaporite
deposits was Zharkov's (1974, 1981) compilation of
Palaeozoic salt deposits. It fueled speculation that there
might have been changes in salinity of the ocean
through time, although it was thought that these could
not have been more than a few per parts per thousand
(Holland, 1974, 1978). The then known inventory of
Palaeozoic salt deposits did not allow for a change in
salinity of more than about 10%.
When Sigsbee Knoll was drilled on Leg 1 of the
Deep Sea Drilling Project in 1967, a very major
discovery was made: much of the Gulf of Mexico was
underlain by a vast layer of salt (Ewing et al., 1969).
This was soon realized to be the offshore extension of
the Jurassic Louann salt known from petroleum drilling
in Louisiana and east Texas. The idea that salt could be
deposited in basins of oceanic depth was new, and the
projected volume of the salt deposit was far larger than
any known until then on land. This discovery was soon
followed by the discovery that large areas of the
Mediterranean and Red Seas were also underlain by salt.
Holland (1974) reported on a series of calculations to
determine whether the relative proportions of major ions
of seawater might have changed sufficiently to be
reflected in an alteration of the sequence and mineralogy
of salts deposited as seawater evaporates. He reckoned
that the mass of evaporites deposited during the
Phanerozoic might be equal to the amount in seawater
and so concluded that any salinity changes could not
possibly have been greater than a factor of two. He also
found it unlikely that the relative proportions of major
ions in seawater could have varied by more than a factor
of two.
From other compilations of evaporite deposits it
became evident that so much salt has been extracted
from seawater during the Phanerozoic that the salinity of
the ocean during the Palaeozoic must have been higher
than it is today (Ronov, 1968; Holser et al., 1980), but a
quantitative estimate remained elusive. Holland (1984)
discussed the problem of salinity and relative ratios of
major ions. He cited a personal communication from
Holser in 1981 to the effect that the inventory of halite in
sedimentary rocks today amounts to about 30% of the
NaCl in the ocean, so that the maximum salinity that
might be expected would be about 45‰.
Since Zharkhov's compilation of Palaeozoic salt
deposits, a number of other Phanerozoic salt deposits
have been described. These include not only new
discoveries on land, but very large deposits of
Mesozoic salt in the Gulf of Mexico and in the
North, Central and South Atlantic, and extensive
Miocene salt underlying broad areas of the Mediterra-
nean and Red Seas and Persian Gulf. The present total
inventory amounts to a minimum of 9.1 × 10
6
km
3
of
halite, equal to about 19.6 × 10
18
kg or more than 50%
of the total of halite dissolved in the ocean today
(36.8× 10
18
kg); the maximum estimate is about
16.3× 10
6
km
3
of halite, equal to 35.2 × 10
18
kg, or
95% of the present total in the ocean. If all this salt
were in the ocean at one time, the salinity would have
been 57‰to 73‰, but as Holland (1984) noted there
is no evidence that the total inventory of halite was
dissolved in the sea at any one time.
The first comprehensive compilation of evaporite
volumes was published by Ronov (1980: Table 13) who
extracted the data from the compilations of sediment
volumes and masses for the major geologic intervals of
the Phanerozoic that had been published in a series of
papers by Ronov, Khain, Balukhovsky and Seslavins-
kiy. The 25 stratigraphic intervals they recognized
included subunits of most of the 9 geologic systems of
the Palaeozoic and Mesozoic, and the 5 units of the
Tertiary. The compilations were made in terms of three
tectonic regimes which had distinct erosion–sedimen-
tation histories —(1) platform (cratonic), (2) geosyn-
clinal, and (3) orogenic regions. These tectonic regimes
are recognized in the lithologic–palaeogeographic maps
for the same time intervals published by Ronov et al.,
5W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

1984, 1989). It was recognized that some areas have a
transition in time from one tectonic regime to another.
The evaporite data were presented as total volumes of
halite, gypsum and anhydrite lumped together. In
modern seawater the components of halite (Na
+
,Cl
−
)
and anhydrite/gypsum (Ca
2+
,SO
4
2−
) together make up
slightly over 80% of the total salts and are at present in
the ratio of 1:0.04. However, in evaporite deposits their
ratio is more often in the order of 1:0.25. Thus the total
evaporite volumes which include all three phases are of
limited value in reconstructing ocean salinity in the past.
Holser (1985, Fig. 1) presented an inventory of
Phanerozoic evaporite deposits showing volumes of
both NaCl and CaSO
4
. This compilation was based on
Zharkov (1981) for the Palaeozoic, and Holser et al.
(1980) for the Mesozoic and Cenozoic. He did not
speculate on the specifics of changes in ocean salinity
through time, but noted that the relative changes in
CaSO
4
induced by evaporite deposition were much
greater than those for NaCl, and are reflected in the
temporal distribution of sulfur isotopes.
Continental margin and ocean floor sediments were
included in the compilation of Budyko et al. (1987).
However, as the purpose of their study was to establish
the history of the atmosphere, only the data for total
sediment, carbonates, and organic carbon were pub-
lished. However, it was this landmark compilation that
showed clearly the exponential decay of total sediment
volumes and masses on a global scale.
Berner and Berner (1987) noted that extractions of
salt into evaporite deposits would have resulted in rapid
lowerings of salinity, and gradual return of the salt from
dissolution of the deposits would have produced longer-
term gentle rises in salinity. They suggested that
although salinity must have varied through time, the
variations were probably about a mean not greatly
different from that observed today. Their reason for
assuming that salinity must have remained close to its
present level throughout the Phanerozoic is based on the
fossil record and its diversity since the Early Cambrian.
A more complete compilation was presented by
Ronov (1993: Tables 20 and 21). Table 20 presented
data for pre-Quaternary Phanerozoic except for Antarc-
tica, recognizing 28 stratigraphic units in the Phanero-
zoic and 3 in the Late Precambrian. Sedimentary
materials are broken down into 18 different deposit
types, 12 sediments and 6 volcanics. The amounts of
evaporites, broken down into halite and gypsum–
anhydrite for the Early Palaeozoic and combined for
younger deposits, are expressed as a percent of the total
volume of sedimentary and volcanic deposits on the
continents. Table 21 presented the information for the
Late Jurassic through Pliocene intervals for the
continents, offshore regions, and ocean floor in terms
of volumes and masses of 32 different kinds of deposit,
26 sediments and 6 volcanics. Ronov made very
conservative estimates of the volumes of salt in the
deep sea, and also assumed that the majority of the
offshore deposits are gypsum/anhydrite. It should be
noted that the volumes of the deep sea deposits were
estimated at a time when data were still rather
incomplete. According to Ronov (1993), the Phanero-
zoic–Late Precambrian total for all evaporites is
23.77× 10
18
kg. The halite mass was estimated to be
18.39× 10
18
kg.
Land (1995) explored the potential role of saline pore
waters in the history of ocean salinity. Using the
maximum values of the Holser (1985) evaporite
inventory, and adding his own estimate of the volume
of halite in the Jurassic Louann Formation (3500 ×10
3
km
3
), he estimated the known inventory of Phanerozoic
halite to be approximately 9600 × 10
3
km
3
or 12.6×
10
18
kg Cl
−
. This is almost half the amount of Cl
−
dissolved in the ocean today. He noted that the present
river flux of Cl
−
, given by Drever et al. (1988) as
115 × 10
9
kg/yr, implies a residence time of Cl
−
in the
ocean of 220× 10
6
years. Further, he noted that if this
rate of delivery had been sustained over the entire
Phanerozoic (550 my) it would imply dissolution of five
times the existing inventory of halite, which seemed
unreasonable. He suggested that the deficit could be
made up by Cl
−
in the pore waters of sedimentary rocks.
Using Garrels and Mackenzie's (1971) estimate of the
volume of pore water (330 × 10
18
kg), and assuming a
high pore water salinity with 100,000 mg Cl
−
/l, the Cl
−
in pore water would contain 30 × 10
18
kg of Cl
−
, almost
three times the amount in halite deposits. This implies a
crustal residence time for Cl
−
of 420× 10
6
years.
However, he argued that this still does not solve the
problem that the present riverine flux is apparently too
large, and he presented evidence that over the long term
it may have been about 1/3 of the modern value. This
would increase the crustal residence time for Cl to about
1.3× 10
9
years, and its oceanic residence time to
680× 10
6
years. These calculations were made in a
context of assuming that ocean salinity has remained
essentially constant as Cl
−
is cycled between oceanic
and crustal reservoirs.
Knauth (1998) used the same data to speculate on the
possibility that more salt has been removed from the
ocean into the crustal reservoirs over the course of
geologic time. He concluded that the ocean may have
had very high salinities (>70‰) in the Precambrian.
These projected high salinities have been used by
6W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Tab le 1
Volumes and masses of halite compiled by Areg Migdisov and Alexander Balukhovsky (Pliocene–Late Devonian, Early Devonian) and from Ronov (1993) (Middle Devonian, older Palaeozoic and Neoproterozoic
Stratigraphic units Age of
top
after
Ronov
(1993)
Volumes of halite Masses of halite —calculated as 2.16 volume
Platforms Geosynclines Orogenic
regions
Shelf and
slope of
platforms
Shelf and
slope of
geosynclines
Shelf
and
slope of
orogenic
regions
Ocean
floor
Global
total
Platforms Geosynclines Orogenic
regions
Shelf and
slope of
platforms
Shelf and
slope of
geosynclines
Shelf
and
slope of
orogenic
regions
Ocean
floor
Global
total
Ma 10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
15
kg
Pliocene 1.8 0.00 0.00 4.00 0.00 0.00 0.00 0.00 4.00 0.00 0.00 8.64 0.00 0.00 0.00 0.00 8.64
Miocene 5.3 1.00 0.00 241.00 435.00 0.00 183.00 0.00 860.00 2.16 0.00 520.56 939.60 0.00 395.28 0.00 1857.60
Oligocene 23.7 4.00 0.00 9.00 13.00 1.00 0.00 2.00 29.00 8.64 0.00 19.44 28.08 2.16 0.00 4.32 62.64
Eocene 36.6 0.00 2.00 6.00 8.00 0.00 0.00 0.00 16.00 0.00 4.32 12.96 17.28 0.00 0.00 0.00 34.56
Paleocene 57.8 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Late Cretaceous 66.4 41.00 0.00 0.00 0.00 0.00 0.00 0.00 41.00 88.56 0.00 0.00 0.00 0.00 0.00 0.00 88.56
Early Cretaceous 97.5 182.00 46.00 2.00 231.00 0.00 0.00 300.00 761.00 393.12 99.36 4.32 498.96 0.00 0.00 648.00 1643.76
Late Jurassic 144.0 428.00 49.00 0.00 354.00 12.00 0.00 0.00 843.00 924.48 105.84 0.00 764.64 25.92 0.00 0.00 1820.88
Middle Jurassic 169.0 138.00 0.00 1.00 0.00 0.00 0.00 0.00 139.00 298.08 0.00 2.16 0.00 0.00 0.00 0.00 300.24
Early Jurassic 187.0 42.00 0.00 0.00 0.00 0.00 0.00 0.00 42.00 90.72 0.00 0.00 0.00 0.00 0.00 0.00 90.72
Late Triassic 208.0 384.00 8.00 0.00 0.00 0.00 0.00 0.00 392.00 829.44 17.28 0.00 0.00 0.00 0.00 0.00 846.72
Middle Triassic 230.0 23.50 0.00 0.00 0.00 0.00 0.00 0.00 23.50 50.76 0.00 0.00 0.00 0.00 0.00 0.00 50.76
Early Triassic 240.0 42.50 0.00 0.00 0.00 0.00 0.00 0.00 42.50 91.80 0.00 0.00 0.00 0.00 0.00 0.00 91.80
Late Permian 245.0 180.00 0.00 1.00 0.00 0.00 0.00 0.00 181.00 388.80 0.00 2.16 0.00 0.00 0.00 0.00 390.96
Early Permian 258.0 1046.00 0.00 55.00 0.00 0.00 0.00 0.00 1101.00 2259.36 0.00 118.80 0.00 0.00 0.00 0.00 2378.16
MandL
Carboniferous
286.0 129.00 0.00 0.00 0.00 0.00 0.00 0.00 129.00 278.64 0.00 0.00 0.00 0.00 0.00 0.00 278.64
Early Carboniferous 320.0 37.00 0.00 6.30 0.00 0.00 0.00 0.00 43.30 79.92 0.00 13.61 0.00 0.00 0.00 0.00 93.53
Late Devonian 360.0 121.60 2.74 0.00 0.00 0.00 0.00 0.00 124.34 262.66 5.92 0.00 0.00 0.00 0.00 0.00 268.58
Middle Devonian 374.0 168.33 14.58 2.55 0.00 0.00 0.00 0.00 185.46 363.59 31.49 5.52 0.00 0.00 0.00 0.00 400.60
Early Devonian 387.0 1.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00 2.16 0.00 0.00 0.00 0.00 0.00 0.00 2.16
Late Silurian 408.0 27.44 0.00 0.00 0.00 0.00 0.00 0.00 27.44 59.27 0.00 0.00 0.00 0.00 0.00 0.00 59.27
Early Silurian 421.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Late Ordovician 438.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Middle Ordovician 458.0 17.67 9.41 0.00 0.00 0.00 0.00 0.00 27.08 38.16 20.33 0.00 0.00 0.00 0.00 0.00 58.49
Early Ordovician 478.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Late Cambrian 505.0 6.32 0.00 0.00 0.00 0.00 0.00 0.00 6.32 13.65 0.00 0.00 0.00 0.00 0.00 0.00 13.65
Middle Cambrian 523.0 130.59 0.00 0.00 0.00 0.00 0.00 0.00 130.59 282.07 0.00 0.00 0.00 0.00 0.00 0.00 282.07
Early Cambrian 540.0 1155.75 0.00 0.00 0.00 0.00 0.00 0.00 1155.75 2496.42 0.00 0.00 0.00 0.00 0.00 0.00 2496.42
Vendian 570.0 423.09 0.00 0.00 0.00 0.00 0.00 0.00 423.09 913.88 0.00 0.00 0.00 0.00 0.00 0.00 913.88
Late Riphean 680.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
M and E Riphean 1050.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
7W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Table 2
Volumes and masses of halite compiled by William Holser and Christopher Wold
Stratiraphic
units
Age of
top —
time
scale of
Gradstein
and Ogg
(1996)
Volumes of halite Masses of halite —calculated as 2.16* volume
Platforms Geosynclines Orogenic
regions
Shelf and
slope of
platforms
Shelf and slope
of geosynclines
Shelf and slope
of orogenic
regions
Ocean
floor
Global
total
Platforms Geosynclines Orogenic
regions
Shelf and
slope of
platforms
Shelf and slope
of geosynclines
Shelf and slope
of orogenic
regions
Ocean
floor
Global
total
Ma 10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
15
kg
Pliocene 1.8 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Miocene 5.3 0.00 0.00 1000.00 0.00 0.00 0.00 1100.00 2100.00 0.00 0.00 2160.00 0.00 0.00 0.00 2376.00 4536.00
Oligocene 23.8 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Eocene 33.7 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Paleocene 54.8 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Late Cretaceous 65.0 54.50 0.00 0.00 0.00 0.00 0.00 0.00 54.50 117.72 0.00 0.00 0.00 0.00 0.00 0.00 117.72
Early
Cretaceous
98.9 0.00 0.00 0.00 0.00 0.00 0.00 2500.00 2500.00 0.00 0.00 0.00 0.00 0.00 0.00 5400.00 5400.00
Late Jurassic 142.0 0.00 0.00 0.00 0.00 0.00 0.00 2100.00 2100.00 0.00 0.00 0.00 0.00 0.00 0.00 4536.00 4536.00
Middle Jurassic 159.4 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Early Jurassic 180.1 0.00 0.00 0.00 0.00 0.00 0.00 1800.00 1800.00 0.00 0.00 0.00 0.00 0.00 0.00 3888.00 3888.00
Late Triassic 205.7 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Middle Triassic 227.4 115.00 0.00 0.00 0.00 0.00 0.00 0.00 115.00 248.40 0.00 0.00 0.00 0.00 0.00 0.00 248.40
Early Triassic 241.7 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Late Permian 248.2 485.00 50.00 0.00 0.00 0.00 560.00 0.00 1095.00 1047.60 108.00 0.00 0.00 0.00 1209.60 0.00 2365.20
Early Permian 256.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Pennsylvanian 290.0 69.00 0.00 0.00 0.00 0.00 0.00 0.00 69.00 149.04 0.00 0.00 0.00 0.00 0.00 0.00 149.04
Mississippian 323.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Late Devonian 354.0 98.80 62.00 0.00 0.00 0.00 0.00 0.00 160.80 213.41 133.92 0.00 0.00 0.00 0.00 0.00 347.33
Middle
Devonian
370.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Early Devonian 391.0 2.32 0.00 0.00 0.00 0.00 0.00 0.00 2.32 5.02 0.00 0.00 0.00 0.00 0.00 0.00 5.02
Late Silurian 417.0 26.00 0.00 0.00 0.00 0.00 0.00 0.00 26.00 56.16 0.00 0.00 0.00 0.00 0.00 0.00 56.16
Early Silurian 423.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Late Ordovician 443.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Middle
Ordovician
458.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Early
Ordovician
880.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Late Cambrian 495.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Middle
Cambrian
505.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Early Cambrian 518.0 675.00 0.00 0.00 0.00 0.00 0.00 0.00 675.00 1458.00 0.00 0.00 0.00 0.00 0.00 0.00 1458.00
Vendian 545.0 600.00 0.00 0.00 0.00 0.00 0.00 0.00 600.00 1296.00 0.00 0.00 0.00 0.00 0.00 0.00 1296.00
Late Riphean 680.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
MandE
Riphean
1050.0 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
8W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Table 3
Maximum estimates of volumes and masses of halite (compiled from data of Migdisov, Balukhovsky, Ronov, Holser, and Wold)
Stratiraphic units Age of
top —
time
scale of
Gradstein
et al.
(2004)
Volumes of halite Masses of halite —calculated as 2.16⁎volume
Platforms Geosynclines Orogenic
regions
Shelf and
slope of
platforms
Shelf and
slope of
geosynclines
Shelf and slope
of orogenic
regions
Ocean
floor
Global
total
Platforms Geosynclines Orogenic
regions
Shelf and
slope of
platforms
Shelf and
slope of
geosynclines
Shelf and slope
of orogenic
regions
Ocean
floor
Global
total
Ma 10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
15
kg
Pliocene 1.81 0.05 0.00 4.00 0.00 0.00 0.00 0.00 4.05 0.11 0.00 8.64 0.00 0.00 0.00 0.00 8.75
Miocene 5.33 1.20 0.00 1000.00 435.00 0.00 183.00 1100.00 2719.20 2.59 0.00 2160.00 939.60 0.00 395.28 2376.00 5873.47
Oligocene 23.03 3.60 0.00 9.20 13.00 1.00 0.00 2.00 28.80 7.78 0.00 19.87 28.08 2.16 0.00 4.32 62.21
Eocene 33.90 0.00 1.30 5.40 8.00 0.00 0.00 0.00 14.70 0.00 2.81 11.66 17.28 0.00 0.00 0.00 31.75
Paleocene 55.80 0.10 0.00 4.20 0.00 0.00 0.00 0.00 4.30 0.22 0.00 9.07 0.00 0.00 0.00 0.00 9.29
Late Cretaceous 65.50 54.50 0.00 0.00 0.00 0.00 0.00 0.00 54.50 117.72 0.00 0.00 0.00 0.00 0.00 0.00 117.72
Early Cretaceous 99.60 177.50 46.20 1.50 231.00 0.00 0.00 2800.00 3256.20 383.40 99.79 3.24 498.96 0.00 0.00 6048.00 7033.39
Late Jurassic 145.50 489.00 11.00 21.00 354.00 12.00 0.00 2100.00 2987.00 1056.24 23.76 45.36 764.64 25.92 0.00 4536.00 6451.92
Middle Jurassic 161.20 138.00 0.00 1.00 0.00 0.00 0.00 0.00 139.00 298.08 0.00 2.16 0.00 0.00 0.00 0.00 300.24
Early Jurassic 175.60 42.00 0.00 0.00 0.00 0.00 0.00 1800.00 1842.00 90.72 0.00 0.00 0.00 0.00 0.00 3888.00 3978.72
Late Triassic 199.60 384.00 8.00 0.00 0.00 0.00 0.00 0.00 392.00 829.44 17.28 0.00 0.00 0.00 0.00 0.00 846.72
Middle Triassic 228.00 115.00 0.00 0.00 0.00 0.00 0.00 0.00 115.00 248.40 0.00 0.00 0.00 0.00 0.00 0.00 248.40
Early Triassic 245.00 42.50 0.00 0.00 0.00 0.00 0.00 0.00 42.50 91.80 0.00 0.00 0.00 0.00 0.00 0.00 91.80
Late Permian 251.00 485.00 50.00 1.00 0.00 0.00 560.00 0.00 1096.00 1047.60 108.00 2.16 0.00 0.00 1209.60 0.00 2367.36
Early Permian 270.60 1046.00 0.00 55.00 0.00 0.00 0.00 0.00 1101.00 2259.36 0.00 118.80 0.00 0.00 0.00 0.00 2378.16
M and L Carboniferous 299.00 129.00 0.00 0.00 0.00 0.00 0.00 0.00 129.00 278.64 0.00 0.00 0.00 0.00 0.00 0.00 278.64
Early Carboniferous 318.10 37.00 0.00 6.30 0.00 0.00 0.00 0.00 43.30 79.92 0.00 13.61 0.00 0.00 0.00 0.00 93.53
Late Devonian 359.20 121.60 62.00 0.00 0.00 0.00 0.00 0.00 183.60 262.66 133.92 0.00 0.00 0.00 0.00 0.00 396.58
Middle Devonian 385.30 168.33 14.58 2.55 0.00 0.00 0.00 0.00 185.46 363.59 31.49 5.52 0.00 0.00 0.00 0.00 400.60
Early Devonian 397.50 2.32 0.00 0.00 0.00 0.00 0.00 0.00 2.32 5.01 0.00 0.00 0.00 0.00 0.00 0.00 5.01
Late Silurian 416.00 27.44 0.00 0.00 0.00 0.00 0.00 0.00 27.44 59.27 0.00 0.00 0.00 0.00 0.00 0.00 59.27
Early Silurian 428.20 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Late Ordovician 443.70 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Middle Ordovician 460.90 17.67 9.41 0.00 0.00 0.00 0.00 0.00 27.08 38.16 20.33 0.00 0.00 0.00 0.00 0.00 58.49
Early Ordovician 471.80 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Late Cambrian 488.30 6.32 0.00 0.00 0.00 0.00 0.00 0.00 6.32 13.65 0.00 0.00 0.00 0.00 0.00 0.00 13.65
Middle Cambrian 501.00 130.59 0.00 0.00 0.00 0.00 0.00 0.00 130.59 282.07 0.00 0.00 0.00 0.00 0.00 0.00 282.07
Early Cambrian 513.00 1155.75 0.00 0.00 0.00 0.00 0.00 0.00 1155.75 2496.42 0.00 0.00 0.00 0.00 0.00 0.00 2496.42
Ediacaran 542.00 600.00 0.00 0.00 0.00 0.00 0.00 0.00 600.00 1296.00 0.00 0.00 0.00 0.00 0.00 0.00 1296.00
Cryogenian–Tonian 630.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Middle Proterozoic 1000.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
9W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Table 4
Minimum estimates of volumes and masses of halite (compiled from data of Migdisov, Balukhovsky, Ronov, Holser, and Wold)
Stratiraphic units Age of
top —
time
scale of
Gradstein
et al.
(2004)
Volumes of halite Masses of halite —calculated as 2.16 ⁎volume
Platforms Geosynclines Orogenic
regions
Shelf and
slope of
platforms
Shelf and
slope of
geosynclines
Shelf and slope
of orogenic
regions
Ocean
floor
Global
total
Platforms Geosynclines Orogenic
regions
Shelf and
slope of
platforms
Shelf and
slope of
geosynclines
Shelf and slope
of orogenic
regions
Ocean
floor
Global
total
Ma 10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
3
km
3
10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
15
kg
Pliocene 1.81 0.05 0.00 4.00 0.00 0.00 0.00 0.00 4.05 0.11 0.00 8.64 0.00 0.00 0.00 0.00 8.75
Miocene 5.33 1.20 0.00 240.80 435.00 0.00 183.00 1100.00 1960.00 2.59 0.00 520.13 939.60 0.00 395.28 2376.00 4233.60
Oligocene 23.03 3.60 0.00 9.20 13.00 1.00 0.00 2.00 28.80 7.78 0.00 19.87 28.08 2.16 0.00 4.32 62.21
Eocene 33.90 0.00 1.30 5.40 8.00 0.00 0.00 0.00 14.70 0.00 2.81 11.66 17.28 0.00 0.00 0.00 31.75
Paleocene 55.80 0.10 0.00 4.20 0.00 0.00 0.00 0.00 4.30 0.22 0.00 9.07 0.00 0.00 0.00 0.00 9.29
Late Cretaceous 65.50 44.40 0.00 0.00 0.00 0.00 0.00 0.00 44.40 95.90 0.00 0.00 0.00 0.00 0.00 0.00 95.90
Early Cretaceous 99.60 177.50 46.20 1.50 231.00 0.00 0.00 300.00 756.20 383.40 99.79 3.24 498.96 0.00 0.00 648.00 1633.39
Late Jurassic 145.50 489.00 11.00 21.00 354.00 12.00 0.00 2100.00 2987.00 1056.24 23.76 45.36 764.64 25.92 0.00 4536.00 6451.92
Middle Jurassic 161.20 138.00 0.00 1.00 0.00 0.00 0.00 0.00 139.00 298.08 0.00 2.16 0.00 0.00 0.00 0.00 300.24
Early Jurassic 175.60 42.00 0.00 0.00 0.00 0.00 0.00 1800.00 1842.00 90.72 0.00 0.00 0.00 0.00 0.00 3888.00 3978.72
Late Triassic 199.60 384.00 8.00 0.00 0.00 0.00 0.00 0.00 392.00 829.44 17.28 0.00 0.00 0.00 0.00 0.00 846.72
Middle Triassic 228.00 23.50 0.00 0.00 0.00 0.00 0.00 0.00 23.50 50.76 0.00 0.00 0.00 0.00 0.00 0.00 50.76
Early Triassic 245.00 42.50 0.00 0.00 0.00 0.00 0.00 0.00 42.50 91.80 0.00 0.00 0.00 0.00 0.00 0.00 91.80
Late Permian 251.00 180.00 50.00 1.00 0.00 0.00 560.00 0.00 791.00 388.80 108.00 2.16 0.00 0.00 1209.60 0.00 1708.56
Early Permian 270.60 1046.00 0.00 55.00 0.00 0.00 0.00 0.00 1101.00 2259.36 0.00 118.80 0.00 0.00 0.00 0.00 2378.16
M and L Carboniferous 299.00 129.00 0.00 0.00 0.00 0.00 0.00 0.00 129.00 278.64 0.00 0.00 0.00 0.00 0.00 0.00 278.64
Early Carboniferous 318.10 37.00 0.00 6.30 0.00 0.00 0.00 0.00 43.30 79.92 0.00 13.61 0.00 0.00 0.00 0.00 93.53
Late Devonian 359.20 121.60 2.74 0.00 0.00 0.00 0.00 0.00 124.34 262.66 5.92 0.00 0.00 0.00 0.00 0.00 268.57
Middle Devonian 385.30 168.33 14.58 2.55 0.00 0.00 0.00 0.00 185.46 363.59 31.49 5.52 0.00 0.00 0.00 0.00 400.60
Early Devonian 397.50 1.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00 2.16 0.00 0.00 0.00 0.00 0.00 0.00 2.16
Late Silurian 416.00 27.44 0.00 0.00 0.00 0.00 0.00 0.00 27.44 59.27 0.00 0.00 0.00 0.00 0.00 0.00 59.27
Early Silurian 428.20 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Late Ordovician 443.70 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Middle Ordovician 460.90 17.67 9.41 0.00 0.00 0.00 0.00 0.00 27.08 38.16 20.33 0.00 0.00 0.00 0.00 0.00 58.49
Early Ordovician 471.80 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Late Cambrian 488.30 6.32 0.00 0.00 0.00 0.00 0.00 0.00 6.32 13.65 0.00 0.00 0.00 0.00 0.00 0.00 13.65
Middle Cambrian 501.00 130.59 0.00 0.00 0.00 0.00 0.00 0.00 130.59 282.07 0.00 0.00 0.00 0.00 0.00 0.00 282.07
Early Cambrian 513.00 1155.75 0.00 0.00 0.00 0.00 0.00 0.00 1155.75 2496.42 0.00 0.00 0.00 0.00 0.00 0.00 2496.42
Ediacaran 542.00 600.00 0.00 0.00 0.00 0.00 0.00 0.00 600.00 1296.00 0.00 0.00 0.00 0.00 0.00 0.00 1296.00
Cryogenian–Tonian 630.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Middle Proterozoic 1000.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
10 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Knauth (2005) in discussing the possible early history of
life on Earth.
All of these studies, whether suggesting that the
general trend is toward higher ocean salinities with age,
or more simply, variations about a mean recognize that
ocean salinity must have changed with time. However,
quantification of the salinity of the ocean in the past has
been elusive because there was no agreement on how to
reconstruct the original extractions and the subsequent
delivery of dissolved evaporite and pore water salt to the
ocean in the past.
2.2. Recent compilations of evaporite deposits
More recently, William Holser and Christopher
Wold compiled a list of major evaporite deposits that
included several large accumulations that had escaped
earlier attention. Estimates of the size of some of the
deposits in the deep sea vary significantly. For the
South Atlantic, volume estimates range from 4000 km
3
(Southam and Hay, 1981) to 1000 km
3
(Burke, 1975;
Burke and Sengor, 1988), and Holser and Wold chose
an intermediate value. They were also able to include
some recently discovered deposits, such as those in
Thailand (Japakasetr and Workman, 1981). This list
has been published in terms of masses of NaCl and
CaSO
4
as Table 1 in Floegel et al. (2000). They found
the total mass of evaporite deposits existing today to
be 32.02× 10
18
kg and the mass of halite to be
24.40× 10
18
kg.
Areg Migdisov and Alexander Balukhovsky, have
gone back to the original data on areas, thicknesses, and
volumes of sediment which they worked on as part of
the group led by Alexander Ronov and Victor Khain at
the Vernadski Institute in the 1960s and ’70s. They
recompiled the information for the Pliocene though
Carboniferous and for the Early Devonian in terms of
the original geographical areas defined for the global
compilation. The evaporites have been divided into
halite and gypsum–anhydrite. These data have been
placed in the GERM (Geochemical Earth Reference
Model) data bank. The Ronov data for evaporites differ
from those of Zharkov (1974, 1981) because they used
different maps and also included the additional amounts
of halite in diapirs and other salt tectonic features. For
this reason estimates from the Ronov database are
generally larger than those of Zharkov.
The results of these recent compilations are presented
here as four data sets, Tables 1–4.Table 1 shows the
Ronov data for halite as revised by Migdisov and
Balukhovsky; it still lacks some of the deep sea deposits;
Table 2 shows the Holser and Wold data on major
evaporite deposits recast into the same format. It
includes the deep sea deposits (previously published in
Floegel et al., 2000). Table 3 is a combination of Tables
1 and 2 using the smaller number when there are
differences between the Ronov/Migdisov/Balukhovsky
and Holser/Wold data sets. Table 4 is a combination of
Tables 1 and 2, using the larger number whenever there
are differences between the data sets, and Table 3 can be
regarded as a minimum estimate and Table 4 as a
maximum estimate of the size of known halite deposits.
It should be noted that these are not the maximum
estimates of evaporite volumes proposed; Southam and
Hay (1981) gave the volume of the Early Cretaceous
South Atlantic halite as 4000 km
3
, and Land (1995)
cited the volume of the Jurassic Louann salt of the Gulf
of Mexico to be 3500 km
3
.
We do not know whether these data represent a
complete inventory of the world's evaporite deposits.
There are no data from Antarctica, and many areas of the
ocean basins and marginal seas have not been
sufficiently explored to be sure that there are not more
deposits to be found. Furthermore, many of the
estimates on volumes and masses of halite in the deep
sea are based on seismic data, some of which show only
diapirs or the top of the salt layer, and none of which
have been drilled though except near their edges.
However, future discoveries and corrections are not
likely to be so large as to invalidate the results presented
here.
3. Factors affecting ocean salinity
Two factors are important in determining the salinity
of the ocean in the past: 1) the amount of salt, and 2) the
amount of water. The salinity of seawater is a measure of
the amount of dissolved solids in terms of weight. Today
the ocean contains about 1322.746 × 10
18
kg H
2
O, and
about 47.578× 10
18
kg of salts, for a mean salinity of
34.72.The four major anions, Cl
−
,SO
4
2−
, HCO
3
?−
,Br
−
,
and four major cations Na
+
,Mg
2+
,Ca
2+
,K
+
, make up
99.8% by weight of the dissolved solids. The two most
abundant ions, Cl
−
and Na
+
, comprise 85.1% by weight
of the salt in present day seawater. Chloride alone makes
up 55% by weight and, as shown in Table 5, there are
17% more moles of Cl
−
than Na
+
. Of the major ions in
seawater, Cl
−
is unique in that it is incompatible with
almost all minerals and resides almost entirely in
seawater, in the pore space of sediments, and in
evaporite deposits derived from seawater. Because all
of the other major ions can enter into and be exchanged
with counterparts in minerals, the only element that can
be used as a proxy for salinity in the past is Cl
−
.
11W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Salinity has a complicated history of definitions. It
was intended to be an expression of the weight of salt
in a given volume of water. Unfortunately this cannot
be determined simply by drying. As the water is
evaporated chemical reactions occur, some of the solid
salts deposited are hydrated, and gasses other than
H
2
O vapor are lost. Accordingly, salinity was
originally defined to be the total amount of dissolved
solids in seawater, when carbonate is converted to
oxide; bromide and iodide are converted to chloride;
organic matter is oxide; and the remainder dried to
480 °C. It is obviously impossible to determine
salinity in the past using this definition. Subsequently,
the definition of salinity was changed to accommodate
a simpler analytical technique using titration with
silver nitrate to determine the amount of halides (Cl
−
,
Br
−
,F
−
, and I
−
) in the water. This measure is termed
the “Chlorinity, Cl”and salinity was then redefined as
1.80655 Cl. The abundance ratios of these halides in
modern seawater are 325,000:1117:22:1. Chlorine is
so dominant that for practical purposes the other
halides need not be considered. The chloride content
of the ocean could be used to make precise estimates
of ocean salinity in the past were it not for the fact
that we know that the relative proportions and total
amounts of the major cations have changed with time
(Sandberg, 1983; Hardie, 1996; Stanley and Hardie,
1998, 1999; Lowenstein et al., 2001; Hardie, 2003).
Because salinity is expressed in terms of weight, not
moles, changing the proportions of cations having
different atomic weights could change the salinity
without changing the chlorinity.
It should be noted that since the 1970s salinity, or
“practical salinity”to be exact, has been defined in terms
of electrical conductivity of the seawater sample relative
to a KCl solution. Again, this definition cannot be
applied to ancient seawater.
The sulfate minerals in evaporite deposits, gypsum
and anhydrite, can not be used to reconstruct past
ocean salinity because sulfur can also leave the ocean
as pyrite. For simplicity we do not report their
abundances here, but note that extractions of sulfate
from the ocean during the Miocene, Early Cretaceous
and Late Jurassic and included in the data sets of
Holser (1985),Ronov (1993) and Floegel et al. (2000)
appear to have exceeded the entire amount dissolved
in the ocean today. In contrast halite extractions rarely
exceeded 10% of the amount in solution. The Late
Palaeozoic extractions of sulfate, which probably took
place in short periods of time, produced significant
upward jumps in the value of δ
34
S, as discussed by
Holser (1977), suggesting a strong impact on the
dissolved inventory. Berner (2004) reviewed earlier
modeling attempts to determine the relative propor-
tions of major ions in seawater, including sulfate. He
developed a new model to reconstruct the variation in
calcium, magnesium and sulfate content of the ocean
throughout the Phanerozoic.
In the discussion of Phanerozoic palaeosalinities
below we have assumed that the amount of salt in the
Table 5
Composition of modern seawater: major ions after Gill (1989) with modifications to make chlorinity (Cl) of 19.2 equal to a salinity of 34.72‰
Species Atomic/
Molecular
weight
Concentration o/oo
by weight (=g/kg)
Proportion of total
salts by weight, as %
Molar
concentration
Mass in the
ocean (10
15
kg)
Moles in the
ocean (10
15
)
H
2
O 18.02 965.28 17,389.8474 1,371,346 76,121,000
Na
+
22.99 10.59 30.51 0.4608 15,049 654,602
Mg
2+
24.31 1.27 3.67 0.0524 1810 74,432
Ca
2+
40.08 0.41 1.17 0.0102 578 14,422
K
+
39.10 0.38 1.10 0.0097 541 13,836
Sr
2+
87.62 0.01 0.04 0.0001 19 211
Cl
−
35.45 19.12 55.08 0.5394 27,168 766,311
SO
4
2−
96.06 2.67 7.69 0.0278 3793 39,484
HCO
3
−
61.02 0.12 0.35 0.0020 172 2823
CO
3
2−
60.01 0.02 0.05 0.0003 26 427
Br
−
79.91 0.07 0.19 0.0008 94 1176
F
−
19.00 0.03 0.08 0.0015 41 2173
Other 0.00 0.02 0.07 35
Sum of halides 19.22 55.35 27,303
Sum of salts 34.72 49,326 1,569,899
Sum of salts+water 1000.00 1,420,671
Note: The mass of seawater assumes an ocean volume of 1,370,323,000 km
3
with an average temperature of 3.5 °C, average salinity of 34.72‰, and
average pressure of 2000 dbar, giving an average density of 1035.8 kg/m
3
(calculated after Millero and Poisson, 1981).
12 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

ocean retains its present proportionality of 1.81558
times the amount of the chloride ion. The periodic
removal of salts other than NaCl, most importantly
CaCO
3
,CaMg(CO
3
)
2
,CaSO
4
,MgSO
4
,KCl,etc.,
implies that chloride is not an exact but only an
approximate proxy for the salinity of seawater.
4. Delivery of salts to the ocean today
Salts can be brought to the ocean by rivers,
groundwater, glaciers, and through the atmosphere. Of
these potential transport mechanisms, the dissolved load
of rivers and groundwater dominate. The dissolved load
of rivers is relatively well known (Meybeck, 1979).
River water is essentially a bicarbonate solution, with
more than 80% of the solutes consisting of HCO
3
−
,SO
4
2−
,
Ca
2+
and H
4
SiO
4
(Allen, 1997), but it also contains the
other anions and cations that are common is seawater,
including Cl
−
.
Chlorine is the classic “excess volatile”and is very
rare in silicate minerals. Chlorine occurs regularly only
in chlorapatite, cerargyrite, lazurite, sodalite, vanadinite,
and in the evaporite salts halite, sylvite, kainite, and
boracite (Emiliani, 1992). It is not a regular constituent
of chlorite or chloritoid although in these and some other
minerals it may substitute for OH
−
.Cl
−
is released in the
weathering of shales (Meybeck, 1987), it is presumably
bound with the interlayer water. NaCl must be contained
in magmas because NaCl brine is present in fluid
inclusions in silicate minerals, but it must be a relatively
minor constituent because there is no record of NaCl as
a primary magmatic precipitate (Mueller and Saxena,
1977).
Determining the global average concentration of Cl
−
in river waters, [Cl
−
], is difficult because concentrations
vary through three (Meybeck, 1979) and possibly even
five orders of magnitude (Feth, 1981). Essentially there
have only been two attempts to determine an average
value, that of Livingston (1963): 7.8 mg/l =0.220 mmol;
subsequently cited by Garrels and Mackenzie (1971),
Lisitzin (1974),Holland (1978), and Brown et al. (1989)
among others, and that of Meybeck (1979): 8.3 mg/
l= 0.233 mmol, subsequently cited by Drever et al.
(1988),Berner and Berner (1987) and others. The
estimates of average values were determined by
averaging the [Cl
−
] in major rivers. Meybeck (1979)
averaged data from 61 rivers or regional groups of
rivers, having a total discharge of 23,413 km
3
/yr. This
value for the concentration is then multiplied by
estimates of the total water discharge from land to
extrapolate the total amount of Cl
−
delivered from land
to the sea each year.
Estimates of the total discharge of rivers to the ocean
range between 32,500 km
3
/yr (Garrels and Mackenzie,
1971) and 37,500 km
3
/yr (Marcinek and Rosenkranz,
1989); Berner and Berner (1987) use a value of
37,400 km
3
/yr. From the values for concentration and
discharge, the total flux of Cl
−
from land to sea via rivers
may range between 254 × 10
9
kg and 311×10
9
kg;
Berner and Berner (1987) give the flux as 308 × 10
9
kg.
Meybeck's (1979) global average for [Cl
−
] in rivers is
dominated by those rivers which drain areas with large
evaporite deposits in the continental interior (see Fig. 1
in Feth, 1981). The total discharge of the rivers for
which he had reliable data is only about 60% of the
global discharge from land to sea, and the assumption
that Cl
−
has the same average concentration in the other
40% of the world's rivers, which drain more coastal
areas mostly lacking evaporite deposits, may not be
justified.
The origin of the Cl
−
in river and groundwater is
controversial. There are four potential sources of
riverine Cl
−
: 1) recycled directly from the ocean via
the atmosphere, 2) from volcanic emissions, including
juvenile chloride outgassed from the mantle, 3) from
human activities, 4) from introduction of saline ground-
waters, and 5) from dissolution of evaporites.
4.1. The atmospheric recycling correction for chloride
in rivers
The bursting of bubbles generated by breaking waves
injects droplets of seawater from the sea surface into the
air. Evaporation of the water leaves the salt as an
aerosol. Sea salt aerosol is concentrated below 1 to 2 km
above the ocean surface and decreases exponentially
with height (Ryan and Mukherjee, 1975). The salt is
hygroscopic, and serves as a nucleus for raindrops.
Hence rainwater has a small content of sea salt, the
amount decreasing away from its source. The [Cl
−
]
ranges from >8 mg/l over the ocean to 1 mg/l in coastal
regions, to 0.1 mg/l or less in the continental interiors
(Junge and Werby, 1958; Stallard and Edmond, 1981;
Drever, 1982; Berner and Berner, 1987). Because
concentrations of Cl
−
in rainwater vary through two or
more orders of magnitude (Feth, 1981), a global average
is difficult to determine. Garrels and Mackenzie (1971)
give a [Cl
−
] of 3.8 mg/l in rainwater as a global average;
this seems to be generally accepted (Brown et al., 1989).
Estimates of the proportion of riverine Cl
−
due to
atmospheric recycling vary greatly. Sverdrup et al.
(1942) made a first attempt to separate salts atmospher-
ically recycled from the ocean from those derived from
weathering on land. As a simplification they assumed
13W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

that all of the Cl
−
in rivers is atmospherically recycled.
Using this assumption, the amount of non-atmospher-
ically recycled solutes in river water can be determined
by setting [Cl
−
] equal to 0 and subtracting the amounts
of the other ions proportional to their abundance in
seawater. Feth (1981) noted that subsequent authors
have sometimes dropped the modifying statements and
stated assumptions of Sverdrup et al. (1942), with the
result that the idea that all Cl
−
in river water is recycled
from the sea has appeared in some textbooks (e.g.
Rankama and Sahama, 1950; Brown et al., 1989).
Garrels and Mackenzie (1971) cited the amount of
atmospherically recycled Cl
−
in river water to be 55%.
Holland (1978) presented a more detailed argument. He
concluded that if the Cl
−
in North American rivers were
derived solely from atmospheric precipitation, [Cl
−
]
would be 1–4 ppm depending on the distance from the
coast rather than the average of 8 ppm observed. He
interpreted this to mean that at present about 27% of the
salt in rivers comes from the sea through atmospheric
cycling. Meybeck (1983) estimated the proportion of
Cl
−
dissolved in rivers that is atmospherically recycled
from the ocean is 72%. However Berner and Berner
(1987) argue that most of the rivers on which Meybeck's
estimate is based are short and strongly affected by
precipitation coming directly from the ocean. In
contrast, Berner and Berner (1987) use a value of 13%
for atmospherically recycled salt, based on studies of the
Amazon by Stallard and Edmond (1981). It can also be
argued that this is too small because very little of the
Amazon Basin is close to the coast. It is important to
recall that the 100 largest rivers deliver about 60% of the
water entering the ocean. The remaining 40% comes
from smaller rivers that are shorter and closer to the
coast.
Using maximum and minimum estimates for the
atmospherically recycled component of river water and
the total Cl
−
flux of rivers, the short-term atmospher-
ically recycled flux ranges between 183 × 10
9
kg/yr and
40× 10
9
kg/yr. The flux from other sources would then
be between 71× 10
9
kg/yr and 271× 10
9
kg/yr, a very
large uncertainty.
4.2. Chloride from volcanic emissions
The chloride emitted from volcanoes comes from
two possible sources: subducted seawater, and out-
gassing from the Earth's mantle. The total amount of
Cl
−
in the oceans, pore waters in sediments, and salt
deposits is probably between 55 and 65 ×10
18
kg or 1.25
and 1.60× 10
21
mol. It is not known whether this has
been at the surface of the Earth since early in its history,
or has been gradually added through time. If added
steadily since the accretion of the Earth, this would
require a flux of 12 to 14 × 10
9
kg Cl
−
/yr.
HCl is emitted by volcanoes, but many of these are
situated along subduction zones and it can be argued
that the emissions are simply returning the salt
dissolved in subducted seawater. Present volcanic
emissions of Cl
−
into the troposphere and stratosphere
are thought to be about 500 times smaller than the input
of Cl
−
into the ocean from rivers. Prior to the advent of
plate tectonics, Correns (1956) evaluated the role of
magmatic sources, and concluded that they were
inadequate to maintain the balance observed today.
Subsequently, Bartels (1972) estimated annual volcanic
emissions of 7.6 × 10
9
kg Cl
−
/yr from chlorine mea-
surements for the Greenland icecap. This is half the rate
expected if the rate of outgassing were constant over
geologic time. Over the Phanerozoic this amounts to
4142× 10
15
kg, or almost 16% of the present amount in
the ocean. Anderson (1974) arrived at a much lower
estimate of 1.7 × 10
9
kg Cl
−
/yr in total volcanic
emissions, which over the course of the Phanerozoic
would amount to only 3.5% of the oceans' Cl
−
.
Unfortunately, the rate of addition of juvenile Cl
−
to
the total inventory remains an open question.
4.3. Chloride from human activities
The amount of Cl
−
introduced into rivers and
groundwater is also difficult to estimate. Meybeck
(1979) proposed that the amount of Cl
−
“pollution”of
rivers is about 30%. This estimate was based on a
comparison of Cl
−
concentrations in river in the early
and late 20th century. Applying this value to the
estimate of non-atmospherically recycled Cl
−
,the
residual “natural”Cl
−
flux might range between
50× 10
9
kg/yr and 190× 10
9
kg/yr. However, salt has
been an important commodity since prehistoric times
(Kurlansky, 2002). Originally salt was used in food
preparation and storage (e.g. salt cod, in which the salt
used for preservation may outweigh the fish!), and more
recently for the chlorination of water supplies and
deicing of roads. According to the Salt Institute (www.
saltinstitute.org) global salt production increased during
the 20th century from 10 × 10
9
kg/yr early in the century
to 183× 10
9
kg/yr in 1990 and reached 225 × 10
9
kg/yr
in 2002. These correspond to Cl
−
masses of 6× 10
9
kg/
yr, 110× 10
9
kg/yr, and 135× 10
9
kg/yr. Only a fraction
of this salt is extracted from evaporite deposits. The
major portion comes from coastal salt evaporation pans.
Wold et al. (1999) give the locations of 690 solar salt
production facilities located in low and mid-latitude
14 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

coastal regions of all continents except Antarctica.
Human recycling of salt from the ocean is probably
greater than atmospheric recycling. Because of its
solubility and ways in which it is used, the human
recycling begins to approach the residual “natural”
fluxes cited above. Wilkinson (2005, p. 161) stated that
“Humans are now an order of magnitude more
important at moving sediment than the sum of all
other natural processes operating on the surface of the
planet.”It appears that the human effect on salt may be
just as great as or greater than it is on sedimentary
materials generally.
4.4. Chloride from of saline groundwaters
It is well known that the salinity of pore waters
tends to increase with depth (Feth, 1981; Land, 1995).
On the continental blocks, deeper groundwaters may
have [Cl
−
] as high as 100 g/l, 5 times the concentration
in seawater. The source of this Cl
−
is uncertain, part of
it may be from dissolution of evaporite deposits, but
such saline waters also occur in basins that are not
known to contain halite as a solid phase. Speculation is
that clay minerals may act as osmotic membranes
allowing escape of water but retaining Cl
−
as
compaction occurs (Feth, 1981). As noted above,
Land (1995) suggested that the amount of Cl
−
in brines
could equal that in evaporite deposits.
Although the brines may approach the land surface in
arid regions, they are ordinarily overlain by fresh,
circulating groundwater. Because of the density differ-
ences the two water masses do not readily mix, but it is
thought that upward diffusion gradually introduces salt
into the fresher layer from which it eventually enters
rivers. The rates of these processes are unknown.
4.5. Chloride from evaporite deposits
Holland (1978) was the first geochemist to recognize
the importance of evaporites in river water chemistry.
He noted that there is an obvious relation between
elevated chloride and sulfate in rivers and the presence
of evaporites in the drainage basin. He estimated the
average chloride [Cl
−
] in North American rivers to be
8 ppm. He estimated that 75% of this comes from the
erosion of evaporites. On a global scale this implies an
annual delivery of about 225 × 10
9
kg/yr from the
erosion of evaporites. Berner and Berner (1987)
concluded that, after correcting for atmospherically
recycled salt and pollution (using the value of 30% from
Meybeck, 1979), the delivery from evaporites would be
about 188× 10
9
kg/yr.
Allen (1997), working from the Oxford Global
Sediment Flux Database, concluded that evaporites
contribute 18% of the total solute load of rivers,
although they are only 1% of the outcrop area.
5. Water
Critical to evaluation ocean salinity in the past is
information on the amounts of water in the ocean and
other reservoirs.
5.1. The total amount of free water on Earth
Free water is H
2
O that is not incorporated into the
solid phases of minerals. It occurs on the surface of the
Earth as a solid: ice; as a liquid: fresh water in rivers and
lakes, seawater, pore water in sediments on land and in
the ocean, water in fissure and cracks in igneous and
metamorphic rocks, water in the biosphere; and as a gas,
vapor (Table 6). Three of these reservoirs, ice, seawater,
and pore water, are large; all of the others are small and
can be neglected. The possible exception is that the
Arctic Ocean basin may have been isolated from the
world ocean and filled with fresh or brackish water at
times during the Late Cretaceous and Palaeogene
forming by far the largest “lake”on Earth. Its present
volume is about 16.7 × 10
6
km
3
, and its volume at the
beginning of the Late Cretaceous would have been
about 15.6× 10
6
km
3
. Both values are smaller than the
present volume of ice on Antarctica (23.56 ×10
6
km
3
=
21.6× 10
6
km
3
water).
Some water is bound in sedimentary minerals, such
as gypsum (−600× 10
15
kg), and in hydrated clays. The
amount in clay is unknown but probably small.
Potentially very large amounts of water, possibly several
ocean equivalents, are incorporated into the mineral
wadsleyite which is stable in the Earth's mantle.
Table 6
Water at or near the surface of the Earth (from various sources)
Mass (10
15
kg) Moles (10
15
)
Free water
Ice 27,820.0 1,544,269
Rivers and lakes 225.0 12,490
Oceans 1,371,345.7 76,122,437
Pore water 58,613.0 3,253,566
Fissures in crystalline rocks 75.0 4163
Atmosphere 13.0 722
Biosphere 0.6 33
Bound water (speculative values)
in Gypsum 600 33,306
in clays 7500 416,320
15W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

There are two ultimate sources for water being
supplied to the surface of the earth: outgassing from the
earth's interior, “juvenile water”, and water carried by
incoming extraterrestrial objects, particularly comets.
Based on consideration of ‘excess volatiles,’Rubey
(1951) estimated the rate of outgassing of water from the
earth's interior at 0.370× 10
12
kg/yr, making the
assumption that it had occurred at the same rate
throughout geologic time and that the oceans had been
generated through this process. However, in the context
of plate tectonics most of the “outgassed”H
2
O,
especially that of volcanic arc systems, is now thought
to be the result of recycling of seawater through
subduction zones.
Turekian (1968) proposed that the oceans originated
as saline water condensed in the last phases of Earth
accretion and hence have always had roughly their
present volume. Others have argued that bombardment
of the early Earth by asteroids would have vaporized any
early ocean, and that the water accumulated later from
comets colliding with the Earth.
Wallmann (2001) has estimated the outgassing rate of
juvenile water from the mantle to be 0.11 × 10
12
kg/yr
into ocean crustal rocks, and about 0.04 ×10
12
kg/yr into
the atmosphere, mostly through intraplate volcanoes.
Geologists had generally assumed that cometary
delivery of water would occur as rare events largely
restricted to the early history of the Earth, but Frank et
al. (1986) suggested that the Earth might be
continuously bombarded by cosmic snowballs. On
the basis of observations from satellites they estimated
that small comets could bring in water at a rate of
about 1× 10
12
kg; this would fill the oceans in about
1.3 billion years. At first their idea was ridiculed as
being a misinterpretation of the instrumental data, but
more recently there seems to be additional evidence
from ground-based telescopes that this might be the
case (Frank and Sigwarth, 2001). However, even if
there is an influx of water from space, the rate of
supply remains highly controversial. Where it has
been possible to make measurements, the water in
comets has about twice as much Deuterium as ocean
water, suggesting that the cometary contribution is
small.
The other side of the equation is losses of free water;
there are three possibilities: subduction into the mantle,
loss into space, and incorporation into minerals.
Southam and Hay (1981) estimated the water lost to
subduction to be 0.26 × 10
12
kg/yr but suggested that
most, if not all, of this is returned as volcanic water
emissions. Von Huene and Scholl (1991) estimated the
global long-term rate of subduction of sediment to be
1km
3
solid per year. If the pore space is 30%, the mass
of subducted water is 0.30 × 10
12
kg/yr. These numbers
for subducted water so closely balance estimates of
outgassing that it is evident that the contribution of truly
juvenile water must be small or negligible. Thus until
recently it seemed likely that the volume of the oceans
was essentially in a steady state.
Another method of estimating the amount of pore
water subducted is to assume that the amount of
sediment on the ocean floor is in a steady state, i.e.
what is delivered from the continents is ultimately
subducted. The age–area distribution of the ocean floor
is not an exponential decay, as might be expected, but is
more nearly linear. It can be fit by a second order
polynomial equation
A¼0:077t2−2:98tþ281:54 ð1Þ
where Ais the area of ocean floor older than age t(my)
still in existence today. Using the data of Hay et al.
(1988),onthemodernmass–age distribution of
sediment on the ocean floor, and making the correction
for the loss of ocean floor with age, it works out that the
long-term average mass of sediment delivered to the
ocean floor is 3.08 × 10
12
kg/yr. For the sediment mass
to be in steady state, this amount must be subducted
each year. Assuming an average density of the solid
phase to be 2700 kg/m
3
, the volume subducted would be
1.14 km
3
solid/yr, very close to the estimate of Von
Huene and Scholl (1991).
Wallmann (2001) has reevaluated the data on
subduction, taking into account not only pore water,
which he estimates to be between 1.08 and 1.80 ×
10
12
kg/yr, but also the structurally bound water in the
sediments, 0.09 × 10
12
kg/yr, pore water in the upper-
most 0.5 km of ocean crust, 0.07 ×10
12
kg/yr, water in
the upper ocean crustal rocks, 0.11×10
12
kg/yr, water in
the deeper ocean crust and peridotites from 0.5 to 7 km,
between 0.36 and 1.26× 10
12
kg/yr, and water from the
mantle trapped in the ocean crust 0.11× 10
12
kg/yr.
These total between 1.82 and 3.33 × 10
12
kg/yr as an
estimate of water subducted. The difference between his
estimate of pore water subducted and those given above
is largely related to differences in the assumed porosity
of the sediment being subducted.
Wallmann (2001) estimates the amounts returned at
subduction zones: cool submarine water reflux, 1.08 to
1.80× 10
12
kg/yr, and recycling through arc volcanoes,
0.14 to 1.21× 10
12
kg/yr, for a total between 1.22 and
2.90× 10
12
kg/yr. He estimates that the net amount of
water subducted into the deeper mantle (> 250 km) is
between 0.16 and 0.41× 10
12
kg/yr.
16 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

In addition to the water subducted, there is flow
through the mid-ocean ridge basalts and older ocean
crust, estimated to be of the order of 130 × 10
12
kg/yr
(Humphris and McCollom, 1998). Although this water
is altered chemically, it is returned to the ocean and Cl
−
is not affected.
The loss of water to space comes from dissociation of
H
2
O into its components by radiation at the top of the
atmosphere, and subsequent loss of the hydrogen into
space. This may have been an important process in the
Earth's early history, but today the presence of ozone in
the atmosphere makes an effective cold trap, the
stratosphere, through which H
2
O vapor from the surface
cannot pass. The presence of significant quantities of
free oxygen in the atmosphere, a prerequisite to making
ozone, was essential to ending the loss of water to space,
and in this respect the present oceans may owe their
existence to the development of life on this planet.
It is possible that water brought by small comets
today is dissociated and the hydrogen escapes back into
space.
It is obvious that the uncertainties concerning supply
and loss of free water from the surface of the Earth are so
great that it is impossible to make a balance or even
know the sign of the change. However, there is another
possible source of information on the volume of the
ocean basins —palaeogeography. It has long been
recognized that there are long-term episodes of
continental flooding and emergence on the scale of
hundreds of millions of years. A long-term trend was
already evident in the data on continental flooding of
Budyko et al. (1987), shown in Fig. 2.
The greatest flooding occurred in the Early Ordovi-
cian and Late Cretaceous and the greatest emergence in
the Late Triassic and at present. These long-term trends
are thought to reflect changes in the rate of sea-floor
spreading. However, Tardy et al. (1989), in compiling
areas of land on the palaeogeographic maps (redrawn
from Scotese et al., 1979; Parrish et al., 1982), noted that
there appeared to be a secular decline in sea level on
which these episodes of flooding and emergence were
superimposed. They considered the possibility of a
long-term loss of water from the oceans, but concluded
that the observations were more likely to reflect a loss of
information through erosion of older deposits. They
assumed that the sedimentary cycling known from the
Fig. 2. Percent of continental area flooded during the Phanerozoic, based on data in Budyko et al. (1987). Diagonal line is linear regression through
the Phanerozoic, showing an apparent decline in the volume of seawater with time.
17W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

mass–age distribution of sediments also applied to the
apparent areas of the continents on palaeogeographic
maps. They calculated the palaeoland area A(in
10
3
km
2
) to decrease exponentially as a function of
time t(in my) following the formula
A¼EXPð−0:00122tþ11:941Þð2Þ
Hallam (1992) assumed that the observed secular
decline is not an artifact of the loss of information with
age. He estimated that the oceans had lost about 10% of
their water over the course of the Phanerozoic. This
amounts to 0.25 × 10
12
kg seawater/yr, in the middle of
the range of Wallmann's (2001) estimates of the net
water subducted into the deep mantle.
There is a problem in the quantitative evaluation of
the long-term trend, noted by Ronov (1994). From the
palaeogeographic maps one can measure the area of
continent flooded, but any assumption about the
absolute magnitude of the sea level change requires
knowledge of the hypsography of the continents at the
time. The assumption usually made is that the
hypsography does not change with time. However,
Ronov (1994) showed that the sea level trends are
similar but not exactly the same for the Gondwanan
and Laurasian continents. More importantly, the
degree of flooding of the Gondwanan continents is
only about half that of the Laurasian continents. This
implies that the two regions have had different
hypsographies (or hypsographic histories) throughout
the Phanerozoic. Because this problem remains
unsolved, and because we know nothing about the
possible variations in the rate of loss of water to the
mantle, we have used a constant rate of decline of
256× 10
15
kg/my throughout the Phanerozoic, based
on Hallam (1992) and Wallmann (2001), for our
“long-term trend”model.
In our calculations of palaeosalinity we use two
models for the total amount of free water on Earth, a
steady-state model assuming that the amount of free
water has remained constant over the Phanerozoic, and a
“long-term trend”model another assuming that it has
declined at the rate of 0.256 × 10
12
kg/yr.
Table 7
Late Cenozoic ice volumes —after Flint (1971) and Denton and Hughes (1981a,b)
Region Age Area Average thickness Ice volume Water volume Sea level
10
12
m
2
km 10
6
km
3
10
6
km
3
m
Antarctic
Present 12.53 1.88 23.56 21.60 59.84
Glacial 13.81 1.88 25.96 23.81 65.95
Greenland
Present 1.73 1.52 2.63 2.41 6.68
Glacial 2.30 1.52 3.50 3.21 8.88
Arctic ocean
Present 15.00 0.00 0.05 0.04 0.00
Glacial 15.00 1.36 20.40 18.71 0.00
Laurentide
Present 0.00 0.00 0.00 0.00 0.00
Glacial 13.39 2.20 29.46 27.01 74.83
Cordilleran
Present 0.30 0.30 0.09 0.08 0.23
Glacial 2.37 1.50 3.56 3.26 9.03
British–Scandinavian–Barents–Kara
Present 0.30 0.30 0.09 0.08 0.23
Glacial 8.37 2.00 16.74 15.35 42.52
Other
Present 0.64 0.30 0.19 0.18 0.49
Glacial 5.20 0.30 1.56 1.43 3.96
Totals with thick Arctic ice (Denton and Hughes' “Outrageous hypothesis”)
Present 30.50 26.60 24.39 67.46
Glacial 60.44 101.17 92.77 205.17
Glacial–Present 29.94 74.57 68.38 137.71
Totals without thick Arctic ice
Present 30.50 26.60 24.39 67.46
Glacial 60.44 80.82 74.11 205.17
Glacial–Present 29.94 54.21 49.71 137.71
18 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

5.2. The amount of water in the ocean
There are two reservoirs of free water outside the
ocean large enough to significantly affect the ocean
volume: ice and groundwater. The volumes of ice at
present and during the last glacial maximum are shown
in Table 7. The estimates are from Flint (1971), partially
revised by Denton and Hughes (1981a). The amount of
water currently present as ice is about 24.4 × 10
6
km
3
.It
is mostly in the Antarctic and Greenland ice sheets, but
there are also lesser amounts in the Cordillera of North
America, the mountains of Scandinavia, and elsewhere.
During the Quaternary as much as 2% of the total
ocean volume may have been incorporated into ice
sheets. During the Last Glacial Maximum the mean
salinity of the ocean was probably about 36.0‰, but if
the Denton and Hughes (1981b) “outrageous hypothe-
sis”of a thick sheet of ice floating on the Arctic Ocean is
true, the ocean salinity could have been as high as
37.6‰.
The amount of water stored in the pore space of
sediments is larger that the amount in ice sheets, but
varies on a much longer time scale (10
6
years) and only
as the total mass of sedimentary rocks increases or
decreases. If the total sedimentary mass remains
constant, the pore space, and hence the amount of
water stored in it, also remain constant. This is because
the rate at which older sediments are being eroded and
releasing pore water is the same as the rate at which new
sediments are being formed and enclosing pore water.
Of course, young sediments contain more pore space
than older, more deeply buried sediments, but over the
long term the dewatering of sediments through com-
paction has almost no effect on the fluxes. However,
there are both marine and non-marine sediments. The
pore space in the former is filled with seawater which
may be altered by reactions with the minerals. Pore
space in the latter is filled with rainwater and salts
derived from weathering. Although the long-term net
pore water flux into and out of sediments changes only
with the total sedimentary mass, the associated fluxes of
fresh water from non-marine sediments and of salt water
flux from marine sediments may change on the time
scale of 10
6
years.
On the shorter term (10
4
–10
3
years) the portion of
the pore space that is filled with water may vary.
However, the effects of variations in groundwater
storage in response to climate change are much less
Fig. 3. Mass of water in the oceans through the Phanerozoic taking glaciations into account. Upper sloping line assumes that there has been loss of
water to the mantle at a constant rate, lower quasi-horizontal line assumes that the mass of free water at the Earth's surface has remained constant.
19W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

well known. It is probably about an order of magnitude
less than the volume changes due to buildup and melting
of ice sheets. The possibility of relatively short-term
changes in the volume of groundwater has been
explored by Hay and Leslie (1990), who concluded
that changes in storage equivalent to 30 m of sea level
(10.8× 10
18
kg H
2
O) were possible and changes
equaling 10 m of sea level (3.6 × 10
18
kg H
2
O) were
probable. The maximum possible salinity change due to
variations in groundwater storage would be in the order
of 0.3‰.
Fig. 3 shows two models for the mass of H
2
O in the
ocean through the Phanerozoic, taking into account the
fluxes into and out of ice sheets and the possible long-
term flux of water from the ocean into the mantle via
subduction. The calculations are given in Table 8. The
net fluxes of water into and out of the pore space in
sediments are so small that they cannot be shown in this
figure. The upper, sloping line includes the long-term
loss of water to the mantle at a steady rate, and includes
the fluctuations due to glacial buildup and decay. The
lower, quasi-horizontal line assumes that there has been
a constant mass of free water on the surface of the earth
and shows only the effect of fluxes into and out of ice.
For the Palaeozoic glaciations, the amounts of ice are
speculative; we assume that the Late Palaeozoic
Gondwanan glaciation involved ice masses comparable
to today's Antarctic ice sheets, and that the Ordovician
glaciation was much smaller. The steep declines in all
models approaching the present are related to both the
buildup of ice sheets, first on Antarctica and then in the
northern hemisphere.
Table 8
Masses of water (H
2
O only) in the ocean during the Phanerozoic according to two different models
Stratigraphic unit Time scale of Gradstein et al. (2004) Masses of water
Age of
top
Age of
base
Length Age of
mid-point
Mass of
water in
ice
Ocean water —
taking only glacial
changes into
account —Model A
Water in ocean
and ice with
long-term trend
Ocean water —
with both long-term
trend and glacial
variations —Model B
Ma Ma my Ma 10
15
kg 10
15
kg 10
15
kg 10
15
kg
Holocene 0 0.01 0.01 0.01 24,390 1,371,346 1,395,736 1,371,346
Pleistocene 0.01 1.81 1.80 0.91 74,110 1,321,626 1,395,969 1,321,859
Pliocene 1.81 5.33 3.52 3.57 22,000 1,373,736 1,396,649 1,374,649
Miocene 5.33 23.03 17.70 14.18 22,000 1,373,736 1,399,365 1,377,365
Oligocene 23.03 33.90 10.87 28.47 10,000 1,385,736 1,403,020 1,393,020
Eocene 33.90 55.80 21.90 44.85 0 1,395,736 1,407,213 1,407,213
Paleocene 55.80 65.50 9.70 60.65 0 1,395,736 1,411,257 1,411,257
Late Cretaceous 65.50 99.60 34.10 82.55 0 1,395,736 1,416,861 1,416,861
Early Cretaceous 99.60 145.50 45.90 122.55 0 1,395,736 1,427,097 1,427,097
Late Jurassic 145.50 161.20 15.70 153.35 0 1,395,736 1,434,979 1,434,979
Middle Jurassic 161.20 175.60 14.40 168.40 0 1,395,736 1,438,831 1,438,831
Early Jurassic 175.60 199.60 24.00 187.60 0 1,395,736 1,443,744 1,443,744
Late Triassic 199.60 228.00 28.40 213.80 0 1,395,736 1,450,449 1,450,449
Middle Triassic 228.00 245.00 17.00 236.50 0 1,395,736 1,456,258 1,456,258
Early Triassic 245.00 251.00 6.00 248.00 0 1,395,736 1,459,201 1,459,201
Late Permian 251.00 270.60 19.60 260.80 0 1,395,736 1,462,477 1,462,477
Early Permian 270.60 299.00 28.40 284.80 20,000 1,375,736 1,468,619 1,448,619
M and L Carboniferous 299.00 318.10 19.10 308.55 25,000 1,370,736 1,474,697 1,449,697
Early Carboniferous 318.10 359.20 41.10 338.65 5000 1,390,736 1,482,400 1,477,400
Late Devonian 359.20 385.30 26.10 372.25 0 1,395,736 1,490,998 1,490,998
Middle Devonian 385.30 397.50 12.20 391.40 0 1,395,736 1,495,899 1,495,899
Early Devonian 397.50 416.00 18.50 406.75 0 1,395,736 1,499,827 1,499,827
Late Silurian 416.00 428.20 12.20 422.10 0 1,395,736 1,503,755 1,503,755
Early Silurian 428.20 443.70 15.50 435.95 0 1,395,736 1,507,300 1,507,300
Late Ordovician 443.70 460.90 17.20 452.30 0 1,395,736 1,511,484 1,511,484
Middle Ordovician 460.90 471.80 10.90 466.35 0 1,395,736 1,515,079 1,515,079
Early Ordovician 471.80 488.30 16.50 480.05 5000 1,390,736 1,518,585 1,513,585
Late Cambrian 488.30 501.00 12.70 494.65 0 1,395,736 1,522,322 1,522,322
Middle Cambrian 501.00 513.00 12.00 507.00 0 1,395,736 1,525,482 1,525,482
Early Cambrian 513.00 542.00 29.00 527.50 0 1,395,736 1,530,728 1,530,728
20 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

6. Reconstructing sediment masses and fluxes in the
past
6.1. Reconstructing the total sediment masses
originally deposited
The reconstruction of the masses of sediment that
existed in the past and of ancient sediment fluxes rests
on the resemblance of the mass–age distribution to an
exponential decay. Gilluly (1969) was the first geologist
to realize (on the basis of area–age distribution of
sediments on geologic maps of North and South
America) that younger sediments are formed mostly
from the erosion of older sediments, and that the
distribution has the form of a decay curve. Veizer and
Jansen (1979, 1985) have shown that the exponential
decay with age holds for: 1) the age/area distribution of
continental basement; 2) the thicknesses of both
sedimentary and volcanogenic units; 3) the thickness,
area and volume of sedimentary rocks; and 4) the
cumulative reserves of most mineral commodities. They
concluded that “the described exponential relationship is
a fundamental law of the present day age distribution of
geologic entities”(Veizer and Jansen, 1979, p. 342).
Wold and Hay (1990) described how to use a simple
exponential decay having the form
y¼Ae−bt ð3Þ
fit through the data to represent the long-term average
sedimentary cycle, as shown in Fig. 4. Here yis the
remnant of the original sediment deposited at time t,
after tmy of cycling at a constant rate of erosion b
(decay constant, or “average recycling proportionality
parameter”of Veizer and Jansen, 1985), and Ais the
long-term average rate at which sediment was deposited.
For this study we used the Ronov (1993) data set on
global sediment masses for both continents and ocean
basins supplemented with data for the Quaternary from
Hay (1994), and adjusted to include a projection of
Antarctic values. Table 9 shows these data in terms of
Fig. 4. Existing masses of sediment on the continental blocks and in the ocean basins (shaded areas), exponential decay curve fit through the data
(solid curve), and reconstructed original masses of sediment deposited (open areas). The exponential is fit through the total existing sedimentary mass,
so the reconstruction of original masses includes sediment that would have been deposited on the ocean floor and has been lost to subduction.
Because, except for evaporites, sedimentary material cannot be stored in the ocean, the dashed line can also be interpreted as the detrital and dissolved
flux from erosion of older sediments. Note the high sediment volumes reconstructed for the early Palaeozoic, reflecting erosion and sedimentation
before the spread of land plants, and the unusually low sediment masses preserved from the Late Precambrian (Ediacarian) possibly reflecting the
weathering system after the “snowball Earth.”
21W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Table 9
Reconstruction of original masses of sediment deposited during given geologic intervals
Stratigraphic unit Time scale of Gradstein et al. (2004) Masses of sedimentary material (after Ronov, 1993; Hay, 1994) Calculations for reconstruction
Age of
top
Age of
base
Length Age of
mid-
point
Platforms Geosynclines Orogenic
regions
Offshore
and
ocean
floor
Estimated
Antarctic
Estimated
Global
total
Total sediment
in interval
normalized to
time
Exponential
curve fit to
data
a
Ratio
observed/
exponential
Reconstructed
original rate of
deposition
Reconstructed
original sediment
mass deposited
during interval
Ma Ma my Ma 10
18
kg 10
18
kg 10
18
kg 10
18
kg 10
18
kg 10
18
kg 10
18
kg/my 10
18
kg/my 10
18
kg/my 10
18
kg
Holocene 0 0.01 0.01 0.01 9.549
Pleistocene 0.01 1.81 1.80 0.91 2.996 5.060 9.434 25.220 1.836 44.546 24.748 9.523 2.599 24.818 44.673
Pliocene 1.81 5.33 3.52 3.57 5.076 2.253 16.014 44.808 2.451 70.502 20.029 9.444 2.121 20.253 71.289
Miocene 5.33 23.03 17.70 14.18 9.193 6.794 29.696 117.234 4.797 167.714 9.475 9.138 1.037 9.903 175.275
Oligocene 23.03 33.90 10.87 28.47 6.993 6.543 9.999 62.297 2.471 88.303 8.124 8.741 0.929 8.876 96.477
Eocene 33.90 55.80 21.90 44.85 17.720 20.564 11.878 82.048 5.267 138.477 6.323 8.306 0.761 7.270 159.204
Paleocene 55.80 65.50 9.70 60.65 7.111 3.768 5.134 23.663 1.681 41.357 4.264 7.908 0.539 5.149 49.942
Late Cretaceous 65.50 99.60 34.10 82.55 41.873 57.247 20.817 118.203 12.593 250.733 7.353 7.387 0.995 9.505 324.122
Early Cretaceous 99.60 145.50 45.90 122.55 45.160 70.941 11.194 124.581 13.366 265.242 5.779 6.523 0.886 8.460 388.298
Late Jurassic 145.50 161.20 15.70 153.35 18.180 29.202 6.580 48.305 5.666 107.933 6.875 5.927 1.160 11.076 173.891
Middle Jurassic 161.20 175.60 14.40 168.40 13.502 27.463 12.214 5.584 58.763 4.081 5.656 0.721 6.890 99.209
Early Jurassic 175.60 199.60 24.00 187.60 14.866 23.970 9.460 5.071 53.367 2.224 5.328 0.417 3.985 95.644
Late Triassic 199.60 228.00 28.40 213.80 16.864 34.109 6.390 6.023 63.386 2.232 4.912 0.454 4.340 123.244
Middle Triassic 228.00 245.00 17.00 236.50 10.738 19.138 2.433 3.392 35.701 2.100 4.577 0.459 4.382 74.493
Early Triassic 245.00 251.00 6.00 248.00 17.966 13.220 1.006 3.380 35.572 5.929 4.416 1.343 12.821 76.926
Late Permian 251.00 270.60 19.60 260.80 12.767 23.391 6.009 4.428 46.595 2.377 4.244 0.560 5.350 104.854
Early Permian 270.60 299.00 28.40 284.80 18.201 30.879 13.338 6.554 68.972 2.429 3.938 0.617 5.889 167.240
M and L Carboniferous 299.00 318.10 19.10 308.55 24.130 28.056 19.301 7.506 78.993 4.136 3.658 1.131 10.797 206.222
Early Carboniferous 318.10 359.20 41.10 338.65 16.158 46.505 7.074 7.322 77.059 1.875 3.331 0.563 5.375 220.915
Late Devonian 359.20 385.30 26.10 372.25 18.670 34.032 8.890 6.467 68.065 2.608 3.001 0.869 8.300 216.624
Middle Devonian 385.30 397.50 12.20 391.40 13.060 36.362 6.386 5.860 61.914 5.075 2.827 1.795 17.143 209.139
Early Devonian 397.50 416.00 18.50 406.75 9.631 34.248 9.559 5.611 59.049 3.192 2.695 1.184 11.309 209.214
Late Silurian 416.00 428.20 12.20 422.10 8.381 22.952 3.285 3.635 38.253 6.375 2.570 2.481 23.693 142.160
Early Silurian 428.20 443.70 15.50 435.95 13.446 30.209 5.172 5.127 53.948 2.697 2.461 1.096 10.466 209.311
Late Ordovician 443.70 460.90 17.20 452.30 9.039 36.156 0.548 4.803 50.546 3.370 2.339 1.440 13.756 206.342
Middle Ordovician 460.90 471.80 10.90 466.35 14.440 59.523 0.606 7.830 82.399 6.867 2.239 3.066 29.283 351.398
Early Ordovician 471.80 488.30 16.50 480.05 14.340 39.771 0.962 5.783 60.856 2.434 2.146 1.134 10.833 270.822
Late Cambrian 488.30 501.00 12.70 494.65 15.629 22.536 1.109 4.124 43.398 4.340 2.051 2.116 20.210 202.101
Middle Cambrian 501.00 513.00 12.00 507.00 17.903 30.936 2.717 5.413 56.969 4.382 1.973 2.221 21.207 275.692
Early Cambrian 513.00 542.00 29.00 527.50 21.063 27.620 1.872 5.308 55.863 2.069 1.851 1.117 10.672 288.136
Ediacaran 542.00 630.00 88.00 586.00 27.247 41.820 6.125 7.895 83.087 0.615 1.543 0.399 3.808 514.065
Cryogenian–Tonian 630.00 1000.00 370.00 815.00 25.530 107.112 0.337 13.963 146.942 0.397 0.757 0.524 5.009 1853.232
Mesoproterozoic 1000.00 1600.00 600.00 1300.00 39.271 36.854 0.546 8.050 84.721 0.141 0.168 0.843 8.048 4828.875
a
y=9.5496e
(−0.0031t)
.
22 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

existing masses of sediment representing time intervals
of differing length. It also shows step by step the
calculations involved in reconstructing the original
masses of sediment deposited during each interval of
time. In order to use Eq. (3), the volumes/masses must
be normalized by dividing by the length of each time
interval, so that they are expressed in terms of mass per
unit time. We have taken tto be expressed as millions of
years. The exponential curve is then fit to the time-
normalized mass age data using the age mid-point of
each stratigraphic interval.
Assuming a constant total sedimentary mass, and
plotted against the Gradstein et al. (2004) time scale,
A=9549.63 × 10
15
kg/my and b=−0.0031/my (Fig. 4).
However, the original data show strong temporal
deviations from the exponential decay curve. Wold
and Hay (1990) proposed that these temporal variations
in the rates of erosion and deposition shown by the data
can be described in terms of proportional deviations
from the decay curve, and that the original fluxes of both
erosion and deposition can be reconstructed by
multiplying these proportions by the long-term average
rate of sediment deposition, A, as given by:
Mt¼Mo=Mpð4Þ
where M
t
is the mass/my of sediment originally
deposited at time t,M
o
is the mass/my of sediment of
that age existing today, and M
p
is the mass/my predicted
by the exponential decay for that age. The mass in each
stratigraphic interval canthenbedeterminedby
multiplying the mass/my by the length of the interval,
expressed in my.
There still remain possible complications. Wold and
Hay (1993) noted that the mass represented by the area
under the curve is not equal to the mass of sediment
observed, and successive reconstructions based on this
method yield a varying total mass of sediment. They
suggested that variations from the average rate of
erosion and deposition involve variations in both the
rate of deposition A, and the decay constant b, with time.
They showed how correction for these changes can be
made assuming either a constant mass of sediment
through time, or growth or decline of the sedimentary
mass due to imbalances between the formation of new
sediment from weathering of igneous and metamorphic
rocks and losses due to subduction and metamorphism.
However, they found that the corrections are small and
do not affect general trends. To keep the calculations
here as simple as possible we have not made corrections
for the assumption of a constant or steadily increasing or
decreasing sediment mass.
Implicit in the reconstructions of ancient sediment
masses is the assumption that the Earth as a whole can
be considered a quasi-closed system with regard to
sediment. In reality, the Earth itself is not a closed
system because new sediments are generated from the
weathering of igneous rocks and sediments deposited on
the sea floor may be subducted. However for the Earth
as a whole these processes probably nearly balance, so
that assumption of the Earth as a closed system used for
the calculations is a reasonable approximation. Hay
(1999) showed that the assumption of constant growth
of sedimentary mass since the Early Precambrian versus
a constant mass throughout the Proterozoic and
Phanerozoic makes very little difference in terms of
masses reconstructed for the Phanerozoic.
Evidence that the reconstructions of past sediment
fluxes are realistic has come from an unexpected source.
McArthur et al. (2001) noted that there is a strong
similarity between the reconstructions of original
sediment masses and the strontium isotope curve for
the Phanerozoic. Because most sediment cannot be
stored, original sediment masses are a direct reflection
of erosion rates. Wold and Hay (1990) had suggested
that times of high deposition rates were times of uplift
and erosion of the continents. This fits with the
hypothesis that changes in the Sr-isotope ratio also
reflect uplift and erosion. A more precise correlation
between the two phenomena is presented in Hay et al.
(2001).
6.2. Reconstructing the original masses of evaporite
deposits
Ancient masses of detrital sediments, such as sands
and shales, can be reconstructed in the same way as the
total sedimentary mass. Being a particulate material,
they are eroded from one site and deposited at another in
a brief period of time. For detrital matter, the rate of
erosion must equal the rate of deposition on geological
time scales. The same applies to carbonates because
carbonate is stored only briefly in solution in the ocean.
Since they cannot be stored, the proportions of most
sediment types within the total sediment mass change
only slowly with time as they evolve (Ronov, 1972,
1982).
Evaporites, which constitute about 1.33% of the
total existing sedimentary mass, present a special case
because they can be stored in solution for long periods
of time in the ocean. As shown in Fig. 5, their
deposition is episodic and depends on the existence of
restricted passages between basins and the open ocean,
and the requirement that the basin must be located in a
23W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

region where evaporation exceeds precipitation and
runoff. Because of their episodic deposition the
proportion of evaporites in the total existing sediment
of a particular age varies significantly with strati-
graphic interval. For the reconstruction of original
halite masses we assume that all halite deposits on or
within the continental blocks are recycled at the same
rate as the total sediment and that their original mass
had the same proportion to the total sediment
deposited at that time as the existing mass has to
the total sediment of that age existing today. We also
assume that evaporites deposited in the continental
margins and in the deep sea (Mediterranean, Red Sea,
Gulf of Mexico and the North and South Atlantic)
have not been involved in recycling and retain their
original size. This latter assumption is not strictly true:
in some places the Mediterranean evaporites are near
the sediment surface and are being dissolved. The
Gulf of Mexico evaporites have been mobilized by the
weight of sediments deposited above them, particular-
ly during the Late Neogene and Quaternary, forming
diapirs that approach the sediment surface and
undergo dissolution. In both cases the amounts of
evaporite lost to dissolution are unknown and difficult
to estimate. Tables 10a and b show the reconstruction
of minimum and maximum estimates of original halite
masses based on the assumptions as to whether they
are recyclable or not. For those involved in recycling,
their original mass is calculated assuming their
proportion of the sediment of each stratigraphic
interval has remained constant.
This latter assumption might be questioned because
evaporites are much more soluble than other rock types.
Garrels and Mackenzie (1971) postulated that because
of their solubility evaporites are preferentially eroded.
Meybeck (1979), on the basis of data on rivers draining
terrains having different geology, concluded that
evaporites were 80 times more soluble than silicic
rocks and 50 times more soluble than crystalline
(igneous and metamorphic) rocks. Einsele (1992) stated
that halite has a chemical denudation rate three orders of
magnitude greater than most igneous and metamorphic
rocks. It is also evident that halite, once deposited, may
subsequently be dissolved in the subsurface and
Fig. 5. Distribution of existing halite deposits through the Phanerozoic showing the sporadic nature of deposition. Masses are maximum values, taken
from Table 5. Total mass is proportional to the area of the bar; mass for an interval is equal to the value shown on the abscissa times the length of the
interval in my. Solid bars are existing masses. Open bars are halite masses subsequently eroded. It is evident that more halite has been deposited than
has been eroded.
24 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

contribute to the saline brines found in most sedimen-
tary basins. Land (1995) cited abundance evidence for
dissolution of halite at depth, including collapse
breccias.
However, Wold and Hay (1993) argued that the
amount of evaporites dissolved and eroded must be
closely related to the general rate of sedimentary
cycling. For them to be selectively eroded they must
be close enough to the surface so that the groundwaters
circulate. We assume that on the time scale of the major
stratigraphic units in the Ronov database (average
length 18.7 my) little selective erosion of older deposits
takes place. Interestingly, the half-life of all sediment,
calculated from the equation and values given above, is
Table 10a
Reconstruction of original masses of halite and amounts eroded since deposition —minimum estimates
Age Time scale of Gradstein et al. (2004) Minimum halite masses
Age of
top
Length Age of
mid-
point
Length
between
mid-
points
Total
existing
halite
mass
Halite on
ocean floor
—not
recyclable
Total
recyclable
halite
mass
Recyclable halite
mass as proportion
of total existing
sediment
Reconstructed
original
recyclable
halite mass
Mass of
haite
eroded
since
deposition
Ma my Ma my 10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
18
kg
Holocene 0 0.01 0.005
Pleistocene 0.01 1.80 0.91 0.905 0.000 0.000 0.000 0.000000 0.000 0.000
Pliocene 1.81 3.52 3.57 2.66 8.748 0.000 8.748 0.000124 8.846 0.098
Miocene 5.33 17.70 14.18 10.61 4233.600 2376.000 1857.600 0.011076 1941.353 83.753
Oligocene 23.03 10.87 28.47 14.29 62.208 4.320 57.888 0.000656 63.246 5.358
Eocene 33.90 21.90 44.85 16.39 31.752 0.000 31.752 0.000229 36.505 4.753
Paleocene 55.80 9.70 60.65 15.80 9.288 0.000 9.288 0.000225 11.216 1.928
Late Cretaceous 65.50 34.10 82.55 21.90 95.904 0.000 95.904 0.000382 123.975 28.071
Early
Cretaceous
99.60 45.90 122.55 40.00 1633.392 648.000 985.392 0.003715 1442.553 457.161
Late Jurassic 145.50 15.70 153.35 30.80 6451.920 4536.000 1915.920 0.017751 3086.743 1170.823
Middle Jurassic 161.20 14.40 168.40 15.05 300.240 0.000 300.240 0.005109 506.896 206.656
Early Jurassic 175.60 24.00 187.60 19.20 3978.720 3888.000 90.720 0.001700 162.587 71.867
Late Triassic 199.60 28.40 213.80 26.20 846.720 0.000 846.720 0.013358 1646.304 799.584
Middle Triassic 228.00 17.00 236.50 22.70 50.760 0.000 50.760 0.001422 105.914 55.154
Early Triassic 245.00 6.00 248.00 11.50 91.800 0.000 91.800 0.002581 198.520 106.720
Late Permian 251.00 19.60 260.80 12.80 1708.560 0.000 1708.560 0.036669 3844.867 2136.307
Early Permian 270.60 28.40 284.80 24.00 2378.160 0.000 2378.160 0.034480 5766.441 3388.281
M and L
Carboniferous
299.00 19.10 308.55 23.75 278.640 0.000 278.640 0.003527 727.425 448.785
Early
Carboniferous
318.10 41.10 338.65 30.10 93.528 0.000 93.528 0.001214 268.127 174.599
Late Devonian 359.20 26.10 372.25 33.60 268.574 0.000 268.574 0.003946 854.764 586.190
Middle
Devonian
385.30 12.20 391.40 19.15 400.602 0.000 400.602 0.006470 1353.194 952.592
Early Devonian 397.50 18.50 406.75 15.35 2.160 0.000 2.160 0.000037 7.653 5.493
Late Silurian 416.00 12.20 422.10 15.35 59.270 0.000 59.270 0.001549 220.267 160.997
Early Silurian 428.20 15.50 435.95 13.85 0.000 0.000 0.000 0.000000 0.000 0.000
Late Ordovician 443.70 17.20 452.30 16.35 0.000 0.000 0.000 0.000000 0.000 0.000
Middle
Ordovician
460.90 10.90 466.35 14.05 58.492 0.000 58.492 0.000710 249.447 190.954
Early
Ordovician
471.80 16.50 480.05 13.70 0.000 0.000 0.000 0.000000 0.000 0.000
Late Cambrian 488.30 12.70 494.65 14.60 13.647 0.000 13.647 0.000314 63.553 49.906
Middle
Cambrian
501.00 12.00 507.00 12.35 282.074 0.000 282.074 0.004951 1365.043 1082.968
Early Cambrian 513.00 29.00 527.50 20.50 2496.420 0.000 2496.420 0.044688 12,876.227 10,379.807
Ediacaran 542.00 88.00 586.00 58.50 1296.000 0.000 1296.000 0.015598 8018.430 6722.430
Cryogenian–
Tonian
630.00 370.00 815.00 229.00 0.000 0.000 0.000 0.000000 0.000 0.000
Mesoproterozoic 1000.00 600.00 1300.00 485.00 0.0000 0.000 0.000 0.000000 0.000 0.000
25W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

224 my, very close to the 220 my half-life of evaporites
determined by Garrels and Mackenzie (1971).
A remaining problem is that our calculations are
based solely on still existing evaporite deposits. Global
sedimentary recycling implies that only half of the
sediment deposited in the Late Palaeozoic is still in
existence, and that 2/3 of the sediments originally
present in the Early Palaeozoic have been destroyed, and
some of them would surely have been evaporites. It is
likely that many older evaporite deposits have been
completely destroyed by erosion. Correspondingly,
estimates for variations in salinity of the ocean in the
past, particularly in the Palaeozoic, may well be
minimal.
Table 10b
Reconstruction of original masses of halite and amounts eroded since deposition —maximum estimates
Age Time scale of Gradstein et al. (2004) Maximum halite masses
Age of
top
Length Age of
mid-
point
Length
between
mid-
points
Total
existing
halite
mass
Halite on
ocean floor
—not
recyclable
Total
recyclable
halite
mass
Recyclable halite
mass as proportion
of total existing
sediment
Reconstructed
original
recyclable
halite mass
Mass of
haite
eroded
since
deposition
Ma my Ma my 10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
18
kg
Holocene 0 0.01 0.005
Pleistocene 0.01 1.80 0.91 0.905 0.000 0.000 0.000 0.000000 0.000 0.000
Pliocene 1.81 3.52 3.57 2.66 8.748 0.000 8.748 0.000124 8.846 0.098
Miocene 5.33 17.70 14.18 10.61 5873.472 2376.000 3497.472 0.020854 3655.161 157.689
Oligocene 23.03 10.87 28.47 14.29 62.208 4.320 57.888 0.000656 63.246 5.358
Eocene 33.90 21.90 44.85 16.39 31.752 0.000 31.752 0.000229 36.505 4.753
Paleocene 55.80 9.70 60.65 15.80 9.288 0.000 9.288 0.000225 11.216 1.928
Late Cretaceous 65.50 34.10 82.55 21.90 117.720 0.000 117.720 0.000470 152.176 34.456
Early
Cretaceous
99.60 45.90 122.55 40.00 7033.392 6048.000 985.392 0.003715 1442.553 457.161
Late Jurassic 145.50 15.70 153.35 30.80 6451.920 4536.000 1915.920 0.017751 3086.743 1170.823
Middle Jurassic 161.20 14.40 168.40 15.05 300.240 0.000 300.240 0.005109 506.896 206.656
Early Jurassic 175.60 24.00 187.60 19.20 3978.720 3888.000 90.720 0.001700 162.587 71.867
Late Triassic 199.60 28.40 213.80 26.20 846.720 0.000 846.720 0.013358 1646.304 799.584
Middle Triassic 228.00 17.00 236.50 22.70 248.400 0.000 248.400 0.006958 518.301 269.901
Early Triassic 245.00 6.00 248.00 11.50 91.800 0.000 91.800 0.002581 198.520 106.720
Late Permian 251.00 19.60 260.80 12.80 2367.360 0.000 2367.360 0.050808 5327.401 2960.041
Early Permian 270.60 28.40 284.80 24.00 2378.160 0.000 2378.160 0.034480 5766.441 3388.281
M and L
Carboniferous
299.00 19.10 308.55 23.75 278.640 0.000 278.640 0.003527 727.425 448.785
Early
Carboniferous
318.10 41.10 338.65 30.10 93.528 0.000 93.528 0.001214 268.127 174.599
Late Devonian 359.20 26.10 372.25 33.60 396.576 0.000 396.576 0.005826 1262.142 865.566
Middle
Devonian
385.30 12.20 391.40 19.15 400.602 0.000 400.602 0.006470 1353.194 952.592
Early Devonian 397.50 18.50 406.75 15.35 5.011 0.000 5.011 0.000085 17.755 12.744
Late Silurian 416.00 12.20 422.10 15.35 59.270 0.000 59.270 0.001549 220.267 160.997
Early Silurian 428.20 15.50 435.95 13.85 0.000 0.000 0.000 0.000000 0.000 0.000
Late Ordovician 443.70 17.20 452.30 16.35 0.000 0.000 0.000 0.000000 0.000 0.000
Middle
Ordovician
460.90 10.90 466.35 14.05 58.492 0.000 58.492 0.000710 249.447 190.954
Early
Ordovician
471.80 16.50 480.05 13.70 0.000 0.000 0.000 0.000000 0.000 0.000
Late Cambrian 488.30 12.70 494.65 14.60 13.647 0.000 13.647 0.000314 63.553 49.906
Middle
Cambrian
501.00 12.00 507.00 12.35 282.074 0.000 282.074 0.004951 1365.043 1082.968
Early Cambrian 513.00 29.00 527.50 20.50 2496.420 0.000 2496.420 0.044688 12,876.227 10,379.807
Ediacaran 542.00 88.00 586.00 58.50 1296.000 0.000 1296.000 0.015598 8018.430 6722.430
Cryogenian–
Tonian
630.00 370.00 815.00 229.00 0.000 0.000 0.000 0.000000 0.000 0.000
Mesoproterozoic 1000.00 600.00 1300.00 485.00 0.0000 0.000 0.000 0.000000 0.000 0.000
26 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Table 11a
Reconstruction of original masses of sediment and halite and amounts eroded since deposition —minimum estimates
Stratigraphic unit Time to
mid-
Holocene
since unit's
mid-point
age
Amount of
unit's
sediment
remaining
Amount of
unit's
sediment
eroded
during each
interval
Halite
fluxes
from unit
during
each
interval
Time to
mid-
Holocene
since unit's
mid-point
age
Amount of
unit's
sediment
remaining
Amount of
unit's
sediment
eroded
during each
interval
Halite
fluxes from
unit during
each
interval
Time to
mid-
Holocene
since unit's
mid-point
age
Amount of
unit's
sediment
remaining
Amount of
unit's
sediment
eroded
during each
interval
Halite
fluxes
from unit
during
each
interval
my 10
15
kg 10
15
kg 10
15
kg my 10
15
kg 10
15
kg 10
15
kg my 10
15
kg 10
15
kg 10
15
kg
Ediacaran Early Cambrian Middle Cambrian
Holocene 586.00 83,088.45 234.19 3.65 527.50 55,864.14 157.45 7.04 507.00 56,970.27 160.57 0.80
Pleistocene 585.09 83,322.64 692.15 10.80 526.59 56,021.60 465.37 20.80 506.09 57,130.84 474.58 2.35
Pliocene 582.43 84,014.79 2818.49 43.96 523.93 56,486.96 1895.00 84.68 503.43 57,605.42 1932.52 9.57
Miocene 571.82 86,833.28 3944.66 61.53 513.32 58,381.96 2652.17 118.52 492.82 59,537.94 2704.69 13.39
Oligocene 557.54 90,777.94 4745.69 74.02 499.04 61,034.14 3190.74 142.59 478.54 62,242.63 3253.92 16.11
Eocene 541.15 95,523.63 4811.08 75.04 482.65 64,224.88 3234.70 144.55 462.15 65,496.55 3298.75 16.33
Paleocene 525.35 100,334.71 7071.79 110.31 466.85 67,459.59 4754.69 212.48 446.35 68,795.30 4848.83 24.01
Late Cretaceous 503.45 107,406.50 14,228.01 221.93 444.95 72,214.27 9566.14 427.49 424.45 73,644.13 9755.55 48.30
Early Cretaceous 463.45 121,634.50 12,227.40 190.72 404.95 81,780.41 8221.03 367.38 384.45 83,399.68 8383.81 41.51
Late Jurassic 432.65 133,861.90 6414.42 100.05 374.15 90,001.44 4312.70 192.73 353.65 91,783.49 4398.10 21.78
Middle Jurassic 417.60 140,276.32 8631.31 134.63 359.10 94,314.15 5803.22 259.33 338.60 96,181.59 5918.13 29.30
Early Jurassic 398.40 148,907.63 12,641.32 197.18 339.90 100,117.37 8499.33 379.82 319.40 102,099.71 8667.62 42.92
Late Triassic 372.20 161,548.95 11,817.09 184.32 313.70 108,616.70 7945.17 355.05 293.20 110,767.34 8102.48 40.12
Middle Triassic 349.50 173,366.04 6312.65 98.47 291.00 116,561.87 4244.28 189.67 270.50 118,869.82 4328.32 21.43
Early Triassic 338.00 179,678.69 7296.92 113.82 279.50 120,806.15 4906.05 219.24 259.00 123,198.14 5003.19 24.77
Late Permian 325.20 186,975.61 14,489.90 226.01 266.70 125,712.20 9742.22 435.36 246.20 128,201.33 9935.11 49.19
Early Permian 301.20 201,465.51 15,444.09 240.90 242.70 135,454.42 10,383.77 464.03 222.20 138,136.45 10,589.37 52.43
M and L Carboniferous 277.45 216,909.60 21,285.88 332.02 218.95 145,838.19 14,311.46 639.55 198.45 148,725.81 14,594.83 72.26
Early Carboniferous 247.35 238,195.48 26,237.46 409.25 188.85 160,149.65 17,640.64 788.33 168.35 163,320.65 17,989.92 89.07
Late Devonian 213.75 264,432.94 16,227.12 253.11 155.25 177,790.29 10,910.23 487.56 134.75 181,310.57 11,126.25 55.09
Middle Devonian 194.60 280,660.06 13,723.25 214.06 136.10 188,700.52 9226.76 412.33 115.60 192,436.83 9409.46 46.59
Early Devonian 179.25 294,383.31 14,394.26 224.52 120.75 197,927.28 9677.92 432.49 100.25 201,846.28 9869.54 48.87
Late Silurian 163.90 308,777.57 13,590.73 211.99 105.40 207,605.20 9137.67 408.34 84.90 211,715.83 9318.59 46.14
Early Silurian 150.05 322,368.30 16,815.85 262.29 91.55 216,742.86 11,306.06 505.25 71.05 221,034.42 11,529.92 57.09
Late Ordovician 133.70 339,184.15 15,149.39 236.30 75.20 228,048.92 10,185.62 455.18 54.70 232,564.34 10,387.30 51.43
Middle Ordovician 119.65 354,333.54 15,423.33 240.57 61.15 238,234.55 10,369.80 463.41 40.65 242,951.64 10,575.13 52.36
Early Ordovician 105.95 369,756.87 17,176.18 267.92 47.45 248,604.35 11,548.33 516.07 26.95 253,526.77 11,776.99 58.31
Late Cambrian 91.35 386,933.05 15,150.61 236.32 32.85 260,152.68 10,186.44 455.21 12.35 265,303.76 10,388.14 51.44
Middle Cambrian 79.00 402,083.66 26,469.67 412.88 20.50 270,339.12 17,796.76 795.30 0.00 275,691.90
Early Cambrian 58.50 428,553.33 85,511.91 1333.82 0.00 288,135.88
Ediacaran 0.00 514,065.24
Total flux from unit
through mid-Holocene
6722.41 10,379.77 1082.96
27W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

6.3. Reconstructing chlorine fluxes to the ocean in the
past
As discussed above, the long-term chlorine fluxes to
the ocean in the past have four possible sources: 1) from
volcanic emissions, 2) from the weathering of crystal-
line rocks, 3) from release of saline pore waters through
erosion of sediments and 4) from erosion of evaporite
deposits. As noted above, the Cl
−
from volcanic
emissions is mostly, if not entirely, recycled from
subducted ocean water, and any contribution of juvenile
Cl
−
from the mantle is probably negligible. Further, we
neglect the contribution of the weathering of crystalline
rocks which contain little Cl, most of which is probably
also derived from subduction processes. The effect of
saline waters both as a sink and source for Cl
−
is more
speculative, but as a first approximation we will assume
that the total sedimentary mass and its contained pore
space have remained almost constant during the
Proterozoic and Phanerozoic (cf. Hay, 1999). We also
assume that the proportions of saline brine and fresh
water in the pore space have remained approximately
constant. Hence the fluxes into and out of this reservoir
would remain equal and have little effect on oceanic
salinity, so we assume here that they can also be
neglected. Consequently we base our calculations solely
on the removal of halite into evaporite deposits and its
subsequent recycling.
6.4. Flux of chlorine from the erosion of halite deposits
To determine the amount of halite eroded during each
interval of time a series of exponential decays are
recalculated starting with the mid-point age of the
interval t
a
−t
b
during which deposition occurred and
calculating the flux rate at the mid-points of each
subsequent stratigraphic interval. As discussed above,
the total amount of sediment originally deposited during
the interval t
a
−t
b
is calculated using Eq. (3) (Table 9).
The original amount of halite deposited is assumed to
have the same proportional relation to the total sediment
as exists today (Tables 10a–b). For each successive
interval after deposition, the amount of sediment of
interval remaining is determined with Eq. (3) using the
time since deposition as t. The detailed calculations for
the interval by interval erosion of sediment originally
deposited during t
a
−t
b
are shown in Tables 11a–g. The
difference between the amounts remaining in successive
stratigraphic intervals is the amount of sediment eroded
from t
a
−t
b
during that time. Analogous to the
reconstruction of the amount of halite originally
deposited, the amount of halite eroded during each of
the successive intervals is determined by multiplying the
amount of t
a
−t
b
sediment eroded by the proportion of
halite originally deposited in interval t
a
−t
b
.Tables 11a–
gare based on the minimum estimates of recyclable
halite. Table 12 shows the fluxes for those intervals for
which the maximum estimates of recyclable halite differ
from the minimum estimates. The total fluxes from each
stratigraphic unit shown at the bottom of Tables 11a–g
and 12 are slightly smaller than the masses of halite
shown to be eroded since deposition in Tables 10a–b
because they are calculated for the mid-Holocene
(0.005 Ma) rather than for present as in Tables 10a–b.
As is evident from Tables 9 and 10a–b, at any time
the youngest sediments are most likely to be eroded. The
greatest amount of erosion will occur immediately after
deposition, before the deposits are protected by burial,
and in each successive time interval less and less will be
eroded. We assume this to be true not only for detrital
sediment, but for evaporites as well. The erosional
fluxes from halite deposits in each stratigraphic interval
are then summed to obtain the flux of halite and Cl
−
(0.607× NaCl) returned to the ocean, and are shown in
Table 13 and Fig. 6. While the removal of salt from the
ocean into evaporites is sporadic, the return is
continuous but varies by a factor of three with time.
The possible rates of Cl
−
delivery by rivers during the
Phanerozoic are hindcast by our calculations to vary
between 13 and 42 × 10
9
kg/yr. For the Quaternary/
Holocene the minimum and maximum flux estimates
are 29.6× 10
9
kg/yr and 34.6× 10
9
kg/yr. Although only
about 10% of the present Cl
−
flux of rivers, these values
are appropriate for the range of uncertainties for the
“natural”Cl
−
flux in rivers discussed above.
7. Reconstructing ocean salinity in the past
As discussed above, the salinity of the ocean depends
on two variables, the amount of water in the ocean, and
the amount of salts dissolved in the water. Although
changes in any of the reservoirs of water shown in Table
6would affect the mass of water in the ocean, only three
of the reservoirs are potentially large enough to affect
ocean salinity: ice, pore water, and fresh water lakes.
Buildup and decay of the mostly northern hemisphere
ice sheets during the Quaternary can cause an oscillation
of ocean salinity from 34.7‰to 36‰, almost a 4%
increase. In contrast, although the pore water reservoir is
large, it cannot change rapidly and its effect on ocean
salinity has been neglected in our calculations. The
amount of water in lakes is trivial throughout most of
geologic time, but if the Arctic ocean were a fresh water
lake in the Eocene, with a mass of 16.5 × 10
18
kg it
28 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Table 11b
Reconstruction of original masses of sediment and halite and amounts eroded since deposition —minimum estimates
Stratigraphic unit Time to
mid-
Holocene
since unit's
mid-point
age
Amount of
unit's
sediment
remaining
Amount of
unit's
sediment
eroded
during each
interval
Halite
fluxes
from unit
during
each
interval
Time to
mid-
Holocene
since unit's
mid-point
age
Amount of
unit's
sediment
remaining
Amount of
unit's
sediment
eroded
during each
interval
Halite
fluxes
from unit
during
each
interval
Time to
mid-
Holocene
since unit's
mid-point
age
Amount of
unit's
sediment
remaining
Amount of
unit's
sediment
eroded
during each
interval
Halite
fluxes
from unit
during
each
interval
my 10
15
kg 10
15
kg 10
15
kg my 10
15
kg 10
15
kg 10
15
kg my 10
15
kg 10
15
kg 10
15
kg
Late Cambrian Middle Ordovician Late Silurian
Holocene 494.65 43,398.44 122.32 0.04 466.35 82,400.03 232.25 0.16 422.10 38,253.48 107.82 0.17
Pleistocene 493.74 43,520.76 361.52 0.11 465.44 82,632.27 686.42 0.49 421.19 38,361.30 318.66 0.49
Pliocene 491.08 43,882.29 1472.14 0.46 462.78 83,318.69 2795.14 1.98 418.53 38,679.97 1297.62 2.01
Miocene 480.47 45,354.43 2060.36 0.65 452.17 86,113.83 3911.98 2.78 407.92 39,977.58 1816.10 2.81
Oligocene 466.19 47,414.79 2478.75 0.78 437.89 90,025.81 4706.37 3.34 393.64 41,793.69 2184.89 3.39
Eocene 449.80 49,893.54 2512.90 0.79 421.50 94,732.18 4771.21 3.39 377.25 43,978.58 2214.99 3.43
Paleocene 434.00 52,406.44 3693.71 1.16 405.70 99,503.39 7013.20 4.98 361.45 46,193.57 3255.81 5.04
Late Cretaceous 412.10 56,100.15 7431.52 2.34 383.80 106,516.59 14,110.12 10.02 339.55 49,449.38 6550.50 10.15
Early Cretaceous 372.10 63,531.67 6386.57 2.01 343.80 120,626.71 12,126.09 8.61 299.55 55,999.88 5629.43 8.72
Late Jurassic 341.30 69,918.24 3350.35 1.05 313.00 132,752.79 6361.27 4.52 268.75 61,629.31 2953.16 4.58
Middle Jurassic 326.25 73,268.59 4508.27 1.42 297.95 139,114.06 8559.80 6.08 253.70 64,582.48 3973.81 6.16
Early Jurassic 307.05 77,776.87 6602.77 2.08 278.75 147,673.86 12,536.58 8.90 234.50 68,556.29 5820.00 9.02
Late Triassic 280.85 84,379.63 6172.26 1.94 252.55 160,210.44 11,719.18 8.32 208.30 74,376.28 5440.53 8.43
Middle Triassic 258.15 90,551.89 3297.20 1.04 229.85 171,929.62 6260.35 4.44 185.60 79,816.81 2906.31 4.50
Early Triassic 246.65 93,849.09 3811.30 1.20 218.35 178,189.97 7236.47 5.14 174.10 82,723.12 3359.47 5.21
Late Permian 233.85 97,660.39 7568.31 2.38 205.55 185,426.43 14,369.84 10.20 161.30 86,082.58 6671.07 10.34
Early Permian 209.85 105,228.70 8066.70 2.54 181.55 199,796.27 15,316.13 10.87 137.30 92,753.66 7110.38 11.02
M and L Carboniferous 186.10 113,295.40 11,117.96 3.50 157.80 215,112.41 21,109.52 14.99 113.55 99,864.04 9799.91 15.18
Early Carboniferous 156.00 124,413.36 13,704.25 4.31 127.70 236,221.92 26,020.07 18.47 83.45 109,663.94 12,079.59 18.72
Late Devonian 122.40 138,117.61 8475.69 2.67 94.10 262,241.99 16,092.67 11.42 49.85 121,743.53 7470.88 11.58
Middle Devonian 103.25 146,593.30 7167.87 2.25 74.95 278,334.66 13,609.54 9.66 30.70 129,214.41 6318.11 9.79
Early Devonian 87.90 153,761.17 7518.36 2.36 59.60 291,944.21 14,275.00 10.13 15.35 135,532.52 6627.04 10.27
Late Silurian 72.55 161,279.53 7098.66 2.23 44.25 306,219.21 13,478.12 9.57 0.00 142,159.56
Early Silurian 58.70 168,378.19 8783.19 2.76 30.40 319,697.33 16,676.52 11.84
Late Ordovician 42.35 177,161.38 7912.77 2.49 14.05 336,373.86 15,023.87 10.66
Middle Ordovician 28.30 185,074.15 8055.85 2.53 0.00 351,397.73
Early Ordovician 14.60 193,130.00 8971.40 2.82
Late Cambrian 0.00 202,101.40
Total flux from unit
through mid-Holocene
49.91 190.95 161.00
29W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Table 11c
Reconstruction of original masses of sediment and halite and amounts eroded since deposition —minimum estimates
Stratigraphic unit Time to
mid-
Holocene
since unit's
mid-point
age
Amount of
unit's
sediment
remaining
Amount of
unit's
sediment
eroded
during each
interval
Halite
fluxes
from unit
during
each
interval
Time to
mid-
Holocene
since unit's
mid-point
age
Amount of
unit's
sediment
remaining
Amount of
unit's
sediment
eroded
during each
interval
Halite
fluxes
from unit
during
each
interval
Time to
mid-
Holocene
since unit's
mid-point
age
Amount of
unit's
sediment
remaining
Amount of
unit's
sediment
eroded
during each
interval
Halite
fluxes
from unit
during
each
interval
my 10
15
kg 10
15
kg 10
15
kg my 10
15
kg 10
15
kg 10
15
kg my 10
15
kg 10
15
kg 10
15
kg
Early Devonian Middle Devonian Late Devonian
Holocene 406.75 59,049.91 166.43 0.01 391.395 61,914.80 174.51 1.13 372.25 68,066.22 191.85 0.76
Pleistocene 405.84 59,216.34 491.90 0.02 390.490 62,089.31 515.77 3.34 371.34 68,258.06 567.01 2.24
Pliocene 403.18 59,708.25 2003.06 0.07 387.830 62,605.08 2100.25 13.59 368.68 68,825.08 2308.91 9.11
Miocene 392.57 61,711.31 2803.42 0.10 377.220 64,705.33 2939.43 19.02 358.07 71,133.99 3231.47 12.75
Oligocene 378.29 64,514.73 3372.70 0.12 362.935 67,644.76 3536.33 22.88 343.79 74,365.46 3887.68 15.34
Eocene 361.90 67,887.43 3419.17 0.13 346.550 71,181.09 3585.06 23.20 327.40 78,253.14 3941.24 15.55
Paleocene 346.10 71,306.60 5025.83 0.18 330.750 74,766.15 5269.67 34.10 311.60 82,194.38 5793.22 22.86
Late Cretaceous 324.20 76,332.43 10,111.66 0.37 308.850 80,035.82 10,602.25 68.60 289.70 87,987.61 11,655.61 45.99
Early Cretaceous 284.20 86,444.10 8689.86 0.32 268.850 90,638.06 9111.46 58.95 249.70 99,643.22 10,016.71 39.52
Late Jurassic 253.40 95,133.95 4558.64 0.17 238.050 99,749.52 4779.81 30.93 218.90 109,659.92 5254.70 20.73
Middle Jurassic 238.35 99,692.60 6134.16 0.22 223.000 104,529.33 6431.77 41.62 203.85 114,914.63 7070.79 27.90
Early Jurassic 219.15 105,826.76 8984.02 0.33 203.800 110,961.11 9419.90 60.95 184.65 121,985.41 10,355.79 40.86
Late Triassic 192.95 114,810.79 8398.26 0.31 177.600 120,381.00 8805.71 56.98 158.45 132,341.21 9680.58 38.20
Middle Triassic 170.25 123,209.04 4486.32 0.16 154.900 129,186.71 4703.98 30.44 135.75 142,021.79 5171.33 20.41
Early Triassic 158.75 127,695.36 5185.83 0.19 143.400 133,890.69 5437.43 35.18 124.25 147,193.12 5977.65 23.59
Late Permian 145.95 132,881.19 10,297.79 0.38 130.600 139,328.12 10,797.40 69.86 111.45 153,170.78 11,870.15 46.84
Early Permian 121.95 143,178.98 10,975.92 0.40 106.600 150,125.52 11,508.44 74.46 87.45 165,040.93 12,651.83 49.92
M and L Carboniferous 98.20 154,154.90 15,127.60 0.55 82.850 161,633.95 15,861.54 102.63 63.70 177,692.76 17,437.43 68.81
Early Carboniferous 68.10 169,282.50 18,646.63 0.68 52.750 177,495.50 19,551.30 126.50 33.60 195,130.19 21,493.78 84.81
Late Devonian 34.50 187,929.13 11,532.41 0.42 19.150 197,046.80 12,091.92 78.24 0.00 216,623.97
Middle Devonian 15.35 199,461.54 9752.94 0.36 0.000 209,138.72
Early Devonian 0.00 209,214.48
Total flux from unit
through mid-Holocene
5.49 952.59 586.19
30 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Table 11d
Reconstruction of original masses of sediment and halite and amounts eroded since deposition —minimum estimates
Stratigraphic unit Time to
mid-
Holocene
since unit's
mid-point
age
Amount of
unit's
sediment
remaining
Amount of
unit's
sediment
eroded
during each
interval
Halite
fluxes
from unit
during
each
interval
Time to
mid-
Holocene
since unit's
mid-point
age
Amount of
unit's
sediment
remaining
Amount of
unit's
sediment
eroded
during each
interval
Halite
fluxes
from unit
during
each
interval
Time to
mid-
Holocene
since unit's
mid-point
age
Amount of
unit's
sediment
remaining
Amount of
unit's
sediment
eroded
during each
interval
Halite
fluxes
from unit
during
each
interval
my 10
15
kg 10
15
kg 10
15
kg my 10
15
kg 10
15
kg 10
15
kg my 10
15
kg 10
15
kg 10
15
kg
Mississippian Pennsylvanian Early Permian
Holocene 338.65 77,060.58 217.20 0.26 308.545 78,994.36 222.65 0.79 284.80 68,972.96 194.40 6.70
Pleistocene 337.74 77,277.78 641.94 0.78 307.640 79,217.01 658.05 2.32 283.89 69,167.36 574.57 19.81
Pliocene 335.08 77,919.72 2614.02 3.17 304.980 79,875.06 2679.61 9.45 281.23 69,741.93 2339.67 80.67
Miocene 324.47 80,533.73 3658.48 4.44 294.370 82,554.67 3750.29 13.23 270.62 72,081.60 3274.52 112.91
Oligocene 310.19 84,192.22 4401.40 5.34 280.085 86,304.96 4511.85 15.92 256.34 75,356.12 3939.47 135.83
Eocene 293.80 88,593.62 4462.04 5.42 263.700 90,816.81 4574.02 16.13 239.95 79,295.59 3993.75 137.70
Paleocene 278.00 93,055.66 6558.75 7.96 247.900 95,390.83 6723.34 23.72 224.15 83,289.34 5870.40 202.41
Late Cretaceous 256.10 99,614.41 13,195.80 16.02 226.000 102,114.16 13,526.94 47.71 202.25 89,159.73 11,810.88 407.24
Early Cretaceous 216.10 112,810.21 11,340.33 13.76 186.000 115,641.10 11,624.91 41.01 162.25 100,970.61 10,150.14 349.98
Late Jurassic 185.30 124,150.54 5949.07 7.22 155.200 127,266.01 6098.35 21.51 131.45 111,120.76 5324.70 183.60
Middle Jurassic 170.25 130,099.61 8005.13 9.72 140.150 133,364.36 8206.01 28.95 116.40 116,445.46 7164.98 247.05
Early Jurassic 151.05 138,104.74 11,724.22 14.23 120.950 141,570.38 12,018.43 42.39 97.20 123,610.44 10,493.75 361.83
Late Triassic 124.85 149,828.96 10,959.79 13.30 94.750 153,588.81 11,234.82 39.63 71.00 134,104.19 9809.54 338.23
Middle Triassic 102.15 160,788.74 5854.68 7.11 72.050 164,823.62 6001.60 21.17 48.30 143,913.73 5240.22 180.68
Early Triassic 90.65 166,643.43 6767.55 8.21 60.550 170,825.22 6937.38 24.47 36.80 149,153.96 6057.28 208.86
Late Permian 77.85 173,410.98 13,438.69 16.31 47.750 177,762.60 13,775.92 48.59 24.00 155,211.24 12,028.28 414.74
Early Permian 53.85 186,849.66 14,323.66 17.38 23.750 191,538.52 14,683.10 51.79 0.00 167,239.52
M and L Carboniferous 30.10 201,173.33 19,741.64 23.96 0.000 206,221.63
Early Carboniferous 0.00 220,914.97
Total flux from unit
through mid-Holocene
174.60 448.78 3388.24
31W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Table 11e
Reconstruction of original masses of sediment and halite and amounts eroded since deposition —minimum estimates
Stratigraphic unit Time to
mid-
Holocene
since unit's
mid-point
age
Amount of
unit's
sediment
remaining
Amount of
unit's
sediment
eroded during
each interval
Halite
fluxes
from unit
during
each
interval
Time to
mid-
Holocene
since unit's
mid-point
age
Amount
of unit's
sediment
remaining
Amount of
unit's
sediment
eroded during
each interval
Halite
fluxes
from unit
during
each
interval
Time to
mid-
Holocene
since unit's
mid-point
age
Amount
of unit's
sediment
remaining
Amount of
unit's
sediment
eroded during
each interval
Halite
fluxes
from unit
during
each
interval
my 10
15
kg 10
15
kg 10
15
kg my 10
15
kg 10
15
kg 10
15
kg my 10
15
kg 10
15
kg 10
15
kg
Late Permian Early Triassic Middle Triassic
Holocene 260.80 46,595.26 131.33 4.82 247.995 35,572.71 100.26 0.26 236.50 35,702.00 100.63 0.14
Pleistocene 259.89 46,726.59 388.15 14.23 247.090 35,672.98 296.33 0.76 235.59 35,802.63 297.41 0.42
Pliocene 257.23 47,114.74 1580.58 57.96 244.430 35,969.31 1206.68 3.11 232.93 36,100.04 1211.07 1.72
Miocene 246.62 48,695.33 2212.13 81.12 233.820 37,175.99 1688.83 4.36 222.32 37,311.10 1694.97 2.41
Oligocene 232.34 50,907.46 2661.34 97.59 219.535 38,864.82 2031.78 5.24 208.04 39,006.07 2039.16 2.90
Eocene 215.95 53,568.80 2698.01 98.93 203.150 40,896.60 2059.77 5.32 191.65 41,045.23 2067.26 2.94
Paleocene 200.15 56,266.81 3965.80 145.42 187.350 42,956.36 3027.65 7.81 175.85 43,112.49 3038.65 4.32
Late Cretaceous 178.25 60,232.60 7978.94 292.58 165.450 45,984.01 6091.45 15.72 153.95 46,151.14 6113.58 8.69
Early Cretaceous 138.25 68,211.54 6857.01 251.44 125.450 52,075.46 5234.92 13.51 113.95 52,264.72 5253.95 7.47
Late Jurassic 107.45 75,068.55 3597.15 131.90 94.650 57,310.38 2746.21 7.09 83.15 57,518.67 2756.19 3.92
Middle Jurassic 92.40 78,665.70 4840.36 177.49 79.600 60,056.59 3695.33 9.54 68.10 60,274.86 3708.76 5.27
Early Jurassic 73.20 83,506.06 7089.14 259.95 60.400 63,751.92 5412.14 13.97 48.90 63,983.62 5431.81 7.72
Late Triassic 47.00 90,595.20 6626.92 243.00 34.200 69,164.06 5059.26 13.06 22.70 69,415.43 5077.65 7.22
Middle Triassic 24.30 97,222.12 3540.08 129.81 11.500 74,223.31 2702.64 6.97 0.00 74,493.07
Early Triassic 12.80 100,762.20 4092.05 150.05 0.000 76,925.95
Late Permian 0.00 104,854.25
Total flux from unit
through mid-Holocene
2136.28 106.72 55.15
32 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Table 11f
Reconstruction of original masses of sediment and halite and amounts eroded since deposition —minimum estimates
Stratigraphic unit Time to mid-
Holocene since
unit's mid-
point age
Amount of
unit's
sediment
remaining
Amount of unit's
sediment eroded
during each
interval
Halite fluxes
from unit
during each
interval
Time to mid-
Holocene since
unit's mid-
point age
Amount of
unit's
sediment
remaining
Amount of unit's
sediment eroded
during each
interval
Halite fluxes
from unit
during each
interval
Time to mid-
Holocene since
unit's mid-
point age
Amount of
unit's
sediment
remaining
Amount of unit's
sediment eroded
during each
interval
Halite fluxes
from unit
during each
interval
my 10
15
kg 10
15
kg 10
15
kg my 10
15
kg 10
15
kg 10
15
kg my 10
15
kg 10
15
kg 10
15
kg
Late Triassic Early Jurassic Middle Jurassic
Holocene 213.80 63,387.10 178.66 2.39 187.595 53,367.91 150.42 0.26 168.40 58,763.71 165.63 0.85
Pleistocene 212.89 63,565.76 528.04 7.05 186.690 53,518.33 444.57 0.76 167.49 58,929.34 489.52 2.50
Pliocene 210.23 64,093.79 2150.19 28.72 184.030 53,962.90 1810.32 3.08 164.83 59,418.86 1993.36 10.18
Miocene 199.62 66,243.98 3009.33 40.20 173.420 55,773.22 2533.66 4.31 154.22 61,412.21 2789.83 14.25
Oligocene 185.34 69,253.31 3620.43 48.36 159.135 58,306.89 3048.17 5.18 139.94 64,202.05 3356.36 17.15
Eocene 168.95 72,873.74 3670.31 49.03 142.750 61,355.06 3090.17 5.25 123.55 67,558.40 3402.60 17.39
Paleocene 153.15 76,544.05 5394.98 72.07 126.950 64,445.22 4542.23 7.72 107.75 70,961.00 5001.47 25.55
Late Cretaceous 131.25 81,939.02 10,854.36 144.99 105.050 68,987.45 9138.68 15.54 85.85 75,962.47 10,062.65 51.41
Early Cretaceous 91.25 92,793.38 9328.12 124.61 65.050 78,126.13 7853.69 13.35 45.85 86,025.13 8647.74 44.18
Late Jurassic 60.45 102,121.51 4893.48 65.37 34.250 85,979.82 4120.00 7.00 15.05 94,672.86 4536.55 23.18
Middle Jurassic 45.40 107,014.98 6584.72 87.96 19.200 90,099.81 5543.91 9.42 0.00 99,209.41
Early Jurassic 26.20 113,599.70 9643.90 128.82 0.000 95,643.72
Late Triassic 0.00 123,243.59
Total flux from
unit through
mid-Holocene
799.57 71.87 206.65
Stratigraphic unit Time to mid-
Holocene since
unit's mid-
point age
Amount of
unit's
sediment
remaining
Amount of unit's
sediment eroded
during each
interval
Halite fluxes
from unit
during each
interval
Time to mid-
Holocene since
unit's mid-
point age
Amount of
unit's
sediment
remaining
Amount of unit's
sediment eroded
during each
interval
Halite fluxes
from unit
during each
interval
Time to mid-
Holocene since
unit's mid-
point age
Amount of
unit's
sediment
remaining
Amount of unit's
sediment eroded
during each
interval
Halite fluxes
from unit
during each
interval
my 10
15
kg 10
15
kg 10
15
kg my 10
15
kg 10
15
kg 10
15
kg my 10
15
kg 10
15
kg 10
15
kg
Late Jurassic Early Cretaceous Late Cretaceous
Holocene 153.35 107,934.69 304.22 5.40 122.545 265,246.10 747.60 2.78 82.55 250,737.28 706.71 0.27
Pleistocene 152.44 108,238.90 899.13 15.96 121.640 265,993.70 2209.59 8.21 81.64 251,443.99 2088.72 0.80
Pliocene 149.78 109,138.03 3661.31 64.99 118.980 268,203.29 8997.56 33.43 78.98 253,532.71 8505.40 3.25
Miocene 139.17 112,799.35 5124.25 90.96 108.370 277,200.85 12,592.67 46.78 68.37 262,038.11 11,903.86 4.55
Oligocene 124.89 117,923.59 6164.81 109.43 94.085 289,793.52 15,149.83 56.28 54.09 273,941.97 14,321.14 5.48
Eocene 108.50 124,088.41 6249.75 110.94 77.700 304,943.35 15,358.56 57.06 37.70 288,263.12 14,518.46 5.55
Paleocene 92.70 130,338.15 9186.49 163.07 61.900 320,301.91 22,575.51 83.87 21.90 302,781.57 21,340.65 8.16
Late Cretaceous 70.80 139,524.65 18,482.66 328.09 40.000 342,877.43 45,420.55 168.74 0.00 324,122.22
Early Cretaceous 30.80 158,007.30 15,883.80 281.95 0.000 388,297.98
Late Jurassic 0.00 173,891.10
Total flux from
unit through
mid-Holocene
1170.79 457.15 28.07
33W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Table 11g
Reconstruction of original masses of sediment and halite and amounts eroded since deposition —minimum estimates
Stratigraphic
unit
Time to mid-
Holocene
since unit's
mid-point age
Amount of
unit's
sediment
remaining
Amount of
unit's sediment
eroded during
each interval
Halite
fluxes from
unit during
each
interval
Time to mid-
Holocene
since unit's
mid-point age
Amount of
unit's
sediment
remaining
Amount of
unit's sediment
eroded during
each interval
Halite
fluxes from
unit during
each
interval
Time to mid-
Holocene
since unit's
mid-point age
Amount of
unit's
sediment
remaining
Amount of
unit's sediment
eroded during
each interval
Halite
fluxes from
unit during
each
interval
my 10
15
kg 10
15
kg 10
15
kg my 10
15
kg 10
15
kg 10
15
kg my 10
15
kg 10
15
kg 10
15
kg
Paleocene Eocene Oligocene
Holocene 60.65 41,358.01 116.57 0.03 44.845 138,479.16 390.31 0.09 28.46 88,304.55 248.89 0.16
Pleistocene 59.74 41,474.58 344.53 0.08 43.940 138,869.47 1153.58 0.26 27.56 88,553.44 735.61 0.48
Pliocene 57.08 41,819.10 1402.93 0.32 41.280 140,023.05 4697.43 1.08 24.90 89,289.04 2995.43 1.96
Miocene 46.47 43,222.03 1963.49 0.44 30.670 144,720.47 6574.36 1.51 14.29 92,284.47 4192.30 2.75
Oligocene 32.19 45,185.52 2362.21 0.53 16.385 151,294.83 7909.39 1.81 0.00 96,476.77
Eocene 15.80 47,547.73 2394.76 0.54 0.000 159,204.23
Paleocene 0.00 49,942.48
Total flux
from unit
through
mid-
Holocene
1.90 4.66 5.19
Stratigraphic
unit
Time to mid-
Holocene since
unit's mid-point
age
Amount of unit's
sediment
remaining
Amount of unit's
sediment eroded
during each interval
Halite fluxes
from unit during
each interval
Time to mid-
Holocene since
unit's mid-point
age
Amount of
unit's sediment
remaining
Amount of unit's
sediment eroded
during each interval
Halite fluxes
from unit during
each interval
my 10
15
kg 10
15
kg 10
15
kg my 10
15
kg 10
15
kg 10
15
kg
Miocene Pliocene
Holocene 14.18 167,716.32 472.71 5.24 3.565 70,503.11 198.71 0.02
Pleistocene 13.27 168,189.03 1397.13 15.47 2.660 70,701.83 587.31 0.07
Pliocene 10.61 169,586.17 5689.20 63.01 0.000 71,289.14
Miocene 0.00 175,275.36
Total flux from
unit through
mid-Holocene
83.72 0.10
34 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Table 12
Reconstruction of amounts of halite eroded since deposition —only for those stratigraphic units in which maximum estimates for recyclable halite differ from minimum estimates
Stratigraphic unit Time to
mid-
Holocene
since unit's
mid-point
age
Halite
fluxes
from unit
during
each
interval
Time to
mid-
Holocene
since unit's
mid-point
age
Halite
fluxes
from unit
during
each
interval
Time to
mid-
Holocene
since unit's
mid-point
age
Halite
fluxes
from unit
during
each
interval
Time to
mid-
Holocene
since unit's
mid-point
age
Halite
fluxes
from unit
during
each
interval
Time to
mid-
Holocene
since unit's
mid-point
age
Halite
fluxes
from unit
during
each
interval
Time to
mid-
Holocene
since unit's
mid-point
age
Halite
fluxes
from unit
during
each
interval
my 10
15
kg my 10
15
kg my 10
15
kg my 10
15
kg my 10
15
kg my 10
15
kg
Early Devonian Late Devonian Late Permian Middle Triassic Late Cretaceous Miocene
Holocene 406.75 0.01 372.25 1.12 260.795 6.67 236.50 0.70 82.55 0.33 14.18 9.86
Pleistocene 405.84 0.04 371.34 3.30 259.890 19.72 235.59 2.07 81.64 0.98 13.27 29.14
Pliocene 403.18 0.17 368.68 13.45 257.230 80.31 232.93 8.43 78.98 3.99 10.61 118.64
Miocene 392.57 0.24 358.07 18.83 246.620 112.39 222.32 11.79 68.37 5.59 0.00
Oligocene 378.29 0.29 343.79 22.65 232.335 135.22 208.04 14.19 54.09 6.72
Eocene 361.90 0.29 327.40 22.96 215.950 137.08 191.65 14.38 37.70 6.82
Paleocene 346.10 0.43 311.60 33.75 200.150 201.49 175.85 21.14 21.90 10.02
Late Cretaceous 324.20 0.86 289.70 67.91 178.250 405.39 153.95 42.54 0.00
Early Cretaceous 284.20 0.74 249.70 58.36 138.250 348.39 113.95 36.56
Late Jurassic 253.40 0.39 218.90 30.62 107.450 182.76 83.15 19.18
Middle Jurassic 238.35 0.52 203.85 41.20 92.400 245.93 68.10 25.80
Early Jurassic 219.15 0.76 184.65 60.34 73.200 360.18 48.90 37.79
Late Triassic 192.95 0.71 158.45 56.40 47.000 336.70 22.70 35.33
Middle Triassic 170.25 0.38 135.75 30.13 24.300 179.86 0.00
Early Triassic 158.75 0.44 124.25 34.83 12.800 207.91
Late Permian 145.95 0.87 111.45 69.16 0.000
Early Permian 121.95 0.93 87.45 73.71
M and L Carboniferous 98.20 1.28 63.70 101.60
Early Carboniferous 68.10 1.58 33.60 125.23
Late Devonian 34.50 0.98 0.00
Middle Devonian 15.35 0.83
Early Devonian 0.00
Total flux from unit
through mid-Holocene
12.74 865.56 2960.00 269.90 34.45 157.63
35W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Table 13
Reconstructions of fluxes of halite and chloride from evaporite deposits to ocean
Stratigraphic unit Age of
mid-
point
Halite and chloride fluxes to ocean-minimum estimates Halite and chloride fluxes to ocean-maximum estimates
Total return flux of
halite to ocean in
each time interval
Return flux
rate of halite
Total return flux of
chloride to ocean in
each time interval
Return flux
rate of chloride
Total return flux
of halite to ocean in
each time interval
Return flux
rate of halite
Total return flux of
chloride to ocean in
each time interval
Return flux
rate of chloride
Ma 10
15
kg 10
15
kg/my 10
15
kg 10
15
kg/my 10
15
kg 10
15
kg/my 10
15
kg 10
15
kg/my
Holocene 0.01 44.192 48.831 26.824 29.640 51.658 57.081 31.356 34.648
Pleistocene 0.91 130.612 49.102 79.282 29.805 152.679 57.398 92.676 34.841
Pliocene 3.57 531.563 50.100 322.659 30.411 621.422 58.569 377.203 35.552
Miocene 14.18 655.765 45.906 398.050 27.865 703.674 49.260 427.130 29.901
Oligocene 28.47 785.623 47.948 476.873 29.104 843.260 51.465 511.859 31.239
Eocene 44.85 794.609 50.292 482.327 30.527 853.040 53.990 517.795 32.772
Paleocene 60.65 1167.203 53.297 708.492 32.351 1253.091 57.219 760.626 34.732
Late Cretaceous 82.55 2331.918 58.298 1415.474 35.387 2500.984 62.525 1518.097 37.952
Early Cretaceous 122.55 1859.012 60.358 1128.420 36.637 2004.305 65.075 1216.613 39.500
Late Jurassic 153.35 827.315 54.971 502.180 33.367 903.535 60.036 548.446 36.442
Middle Jurassic 168.40 1082.055 56.357 656.807 34.209 1184.617 61.699 719.063 37.451
Early Jurassic 187.60 1570.962 59.960 953.574 36.396 1721.174 65.694 1044.753 39.876
Late Triassic 213.80 1348.109 59.388 818.302 36.049 1488.526 65.574 903.536 39.803
Middle Triassic 236.50 716.298 62.287 434.793 37.808 776.293 67.504 471.210 40.975
Early Triassic 248.00 819.922 64.056 497.693 38.882 889.272 69.474 539.788 42.171
Late Permian 260.80 1330.202 55.425 807.432 33.643 1353.022 56.376 821.284 34.220
Early Permian 284.80 975.751 41.084 592.281 24.938 1000.074 42.108 607.045 25.560
M and L Carboniferous 308.55 1273.448 42.307 772.983 25.680 1306.971 43.421 793.331 26.357
Early Carboniferous 338.65 1540.147 45.838 934.869 27.823 1581.468 47.067 959.951 28.570
Late Devonian 372.25 900.083 47.002 546.351 28.530 900.640 47.031 546.689 28.548
Middle Devonian 391.40 695.033 45.279 421.885 27.484 695.504 45.310 422.171 27.503
Early Devonian 406.75 728.643 47.469 442.286 28.813 728.643 47.469 442.286 28.813
Late Silurian 422.10 678.273 48.973 411.712 29.726 678.273 48.973 411.712 29.726
Early Silurian 435.95 839.229 51.329 509.412 31.157 839.229 51.329 509.412 31.157
Late Ordovician 452.30 756.061 53.812 458.929 32.664 756.061 53.812 458.929 32.664
Middle Ordovician 466.35 758.875 55.392 460.637 33.623 758.875 55.392 460.637 33.623
Early Ordovician 480.05 845.120 57.885 512.988 35.136 845.120 57.885 512.988 35.136
Late Cambrian 494.65 742.968 60.159 450.981 36.517 742.968 60.159 450.981 36.517
Middle Cambrian 507.00 1208.178 58.936 733.364 35.774 1208.178 58.936 733.364 35.774
Early Cambrian 527.50 1333.821 22.800 809.630 13.840 1333.821 22.800 809.630 13.840
36 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

would have increased the salinity of the world ocean by
only slightly more than 1‰.
Table 8 summarizes two models of the amount of
water in the ocean through the Phanerozoic and Late
Precambrian. The first, termed model A, assumes a
constant mass of free water on the surface of the Earth
divided between ice sheets and the ocean. The second,
model B, assumes that in addition there has been a
continuous steady loss of water from the surface to the
mantle through subduction at a rate of 256 ×10
15
kg/my.
It is important to note that Model B assumes that only
H
2
O and none of the salts in seawater are subducted
over the long term. Tables 14a–binclude these two
models as part of our calculations of the past salinity of
the ocean based on removal and return of Cl
−
.
Table 14a is based on the existing amounts of halite
in evaporite deposits using the minimum estimates of
Ronov/Migdisov/Balukhovsky/Holser/Wold (Table 3).
The total mass of Na
+
and Cl
−
in the ocean today is
42,217× 10
15
kg, but as shown in Table 5, there are
more moles of Cl
−
than there are of Na
+
. Today the total
chloride in the ocean is 27,168× 10
15
kg. Including
bromine, fluorine and iodine, the total halides in the
ocean are 27,303× 10
15
kg. The modern concentration
of halides by weight (“chlorinity, Cl”) is 19.22. The
salinity of the ocean today is 1.80655× Cl. Unfortu-
nately, this formula cannot be used to determine the
salinity of the ocean in the past because it is the ratio
between the total halides and the mass of seawater,
which includes both H
2
O and salts (Table 5). The
removal of NaCl as halite into evaporite deposits and its
return though dissolution with time means that the
relative molar proportions of Na
+
and Cl
−
in the ocean
will have changed during the Phanerozoic and the
relative proportions of the major anions and cations will
undoubtedly have changed as well (Hardie, 1996;
Stanley and Hardie, 1998; Lowenstein et al., 2001;
Hardie, 2003). As a close approximation, we assume
that the salinity (mass of salts:mass of seawater) of the
ocean varies directly with the proportional relation
between Cl
−
and H
2
O today in terms of mass
(1:50.476).
Table 14a shows the masses of water in the ocean for
the two models, A and B, through the Phanerozoic, the
minimum and maximum fluxes of chlorine into and out
of evaporite deposits, assuming Cl
−
to be 60.7% of the
Fig. 6. Fluxes of Cl to the ocean during the Phanerozoic resulting from the erosion of evaporite deposits. Calculations based on both minimum (Tables
3, 14a) and maximum (Tables 4, 14a) estimates are shown, the upper line being based on the maximum estimates. The peaks in flux reflect general
increases in erosion rates. Fluxes are maximal if high erosion rates occur in the time interval immediately after halite has been deposited on a
continental block and is vulnerable to erosion before it becomes deeply buried.
37W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

weight of the halite, and minimum and maximum
estimates of the mass of chloride in the ocean. Table 14b
shows the minimum and maximum estimates of the total
mass of salts in the ocean assuming the total mass of
salts to be 1.81558 times the mass of chloride, the
minimum and maximum estimates of the mass of
seawater, and minimum and maximum estimates of
salinity for water models A and B.
Fig. 7 summarizes these results for both the
minimum and maximum models. The short-dash lines
are the minimum and maximum estimates of salinity
assuming water model A, with no loss of water to the
mantle. The upper solid line is the average of these
values. The long-dash lines are the minimum and
maximum estimates of salinity assuming water model B,
with loss only of water to the mantle. The lines slope
sharply downward toward the present as extractions
occurred, and slope gently upward toward present as
evaporite deposits are eroded and salts returned to the
ocean. Glaciations have little effect except for the Late
Palaeozoic where they cause a broad depression in the
lines, and for the Neogene. There were significantly
Table 14a
Masses of ocean water —models A and B
a
, minimum and maximum chloride fluxes from and to the ocean, and minimum and maximum estimates of
mass of chloride in the ocean
Stratigraphic unit Age of
mid-
point
Length Mass of
ocean water
—taking
only glacial
changes into
account —
Model A
Mass of
ocean water
—with both
long-term
trend and
glacial
variations —
Model B
Chloride
flux from
ocean into
evaporite
deposits
—
minimum
rate
Chloride
flux from
ocean into
evaporite
deposits
—
maximum
rate
Chloride
flux from
evaporite
deposits
to ocean
—
minimum
rate
Chloride
flux from
evaporite
deposits
to ocean
—
maximum
rate
Mass of
chloride
in the
ocean —
minimum
estimate
Mass of
chloride
in the
ocean —
maximum
estimate
Ma my 10
15
kg 10
15
kg 10
15
kg/
my
10
15
kg/
my
10
15
kg/
my
10
15
kg/
my
10
15
kg 10
15
kg
Recent 0.000 0.00 1,371,346 1,371,346 0.00 0.00 29.64 34.65 27,168 27,168
Holocene 0.005 0.01 1,371,346 1,371,346 0.00 0.00 29.64 34.65 27,168 27,168
Pleistocene 0.910 1.80 1,321,626 1,321,859 0.00 0.00 29.81 34.84 27,114 27,105
Pliocene 3.570 3.52 1,373,736 1,374,649 1.53 1.53 30.41 35.55 27,012 26,985
Miocene 14.180 17.70 1,373,736 1,377,365 148.06 206.83 27.86 29.90 29,140 30,117
Oligocene 28.465 10.87 1,385,736 1,393,020 3.77 3.77 29.10 31.24 28,864 29,818
Eocene 44.850 21.90 1,395,736 1,407,213 1.01 1.01 30.53 32.77 28,218 29,123
Paleocene 60.650 9.70 1,395,736 1,411,257 0.70 0.70 32.35 34.73 27,911 28,793
Late Cretaceous 82.550 34.10 1,395,736 1,416,861 2.21 2.71 35.39 37.95 26,780 27,591
Early Cretaceous 122.550 45.90 1,395,736 1,427,097 27.65 99.06 36.64 39.50 26,367 30,325
Late Jurassic 153.350 15.70 1,395,736 1,434,979 294.71 294.71 33.37 36.44 30,470 34,379
Middle Jurassic 168.400 14.40 1,395,736 1,438,831 21.37 21.37 34.21 37.45 30,285 34,148
Early Jurassic 187.600 24.00 1,395,736 1,443,744 102.45 102.45 36.40 39.88 31,870 35,649
Late Triassic 213.800 28.40 1,395,736 1,450,449 35.19 35.19 36.05 39.80 31,846 35,518
Middle Triassic 236.500 17.00 1,395,736 1,456,258 3.78 18.51 37.81 40.97 31,267 35,136
Early Triassic 248.000 6.00 1,395,736 1,459,201 20.08 20.08 38.88 42.17 31,155 35,004
Late Permian 260.800 19.60 1,395,736 1,462,477 119.07 164.99 33.64 34.22 32,829 37,567
Early Permian 284.800 28.40 1,375,736 1,448,619 123.25 123.25 24.94 25.56 35,621 40,341
M and L Carboniferous 308.550 19.10 1,370,736 1,449,697 23.12 23.12 25.68 26.36 35,572 40,279
Early Carboniferous 338.650 41.10 1,390,736 1,477,400 3.96 3.96 27.82 28.57 34,591 39,268
Late Devonian 372.250 26.10 1,395,736 1,490,998 19.88 29.35 28.53 28.55 34,366 39,289
Middle Devonian 391.400 12.20 1,395,736 1,495,899 67.33 67.33 27.48 27.50 34,852 39,775
Early Devonian 406.750 18.50 1,395,736 1,499,827 0.25 0.58 28.81 28.81 34,323 39,253
Late Silurian 422.100 12.20 1,395,736 1,503,755 10.96 10.96 29.73 29.73 34,094 39,024
Early Silurian 435.950 15.50 1,395,736 1,507,300 0.00 0.00 31.16 31.16 33,611 38,541
Late Ordovician 452.300 17.20 1,395,736 1,511,484 0.00 0.00 32.66 32.66 33,050 37,979
Middle Ordovician 466.350 10.90 1,395,736 1,515,079 13.89 13.89 33.62 33.62 32,834 37,764
Early Ordovician 480.050 16.50 1,390,736 1,513,585 0.00 0.00 35.14 35.14 32,255 37,184
Late Cambrian 494.650 12.70 1,395,736 1,522,322 3.04 3.04 36.52 36.52 31,830 36,759
Middle Cambrian 507.000 12.00 1,395,736 1,525,482 69.05 69.05 35.77 35.77 32,229 37,158
Early Cambrian 527.500 29.00 1,395,736 1,530,728 269.51 269.51 13.84 13.84 39,643 44,573
a
From Table 8.
38 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Table 14b
Mass of salts in the ocean, masses of seawater for ocean water models A and B, and minimum and maximum salinities calculated for ocean water models A and B
Stratigraphic unit Age of
mid-
point
Length Mass of salts
in ocean —
minimum
estimate
Mass of salts
in ocean —
maximum
estimate
Mass of
seawater —
Model A —
minimum
estimate
Mass of
seawater —
Model A —
maximum
estimate
Salinity —
Model A
minimum
estimate
Salinity —
Model A
maximum
estimate
Mass of
seawater —
Model B —
minimum
estimate
Mass of
seawater —
Model B —
maximum
estimate
Salinity —
Model B
minimum
estimate
Salinity —
Model B
maximum
estimate
Ma my 10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
15
kg 10
15
kg
Recent 0.000 0.00 49,326 49,326 1,420,672 1,420,672 34.72 34.72 1,420,672 1,420,672 34.72 34.72
Holocene 0.005 0.01 49,325 49,325 1,420,671 1,420,671 34.72 34.72 1,420,671 1,420,671 34.72 34.72
Pleistocene 0.910 1.80 49,228 49,212 1,370,854 1,370,837 35.91 35.90 1,371,087 1,371,070 35.90 35.89
Pliocene 3.570 3.52 49,043 48,994 1,422,779 1,422,730 34.47 34.44 1,423,693 1,423,643 34.45 34.41
Miocene 14.180 17.70 52,906 54,680 1,426,642 1,428,416 37.08 38.28 1,430,270 1,432,044 36.99 38.18
Oligocene 28.465 10.87 52,406 54,138 1,438,1,42 1,439,874 36.44 37.60 1,445,426 1,447,158 36.26 37.41
Eocene 44.850 21.90 51,232 52,875 1,446,968 1,448,611 35.41 36.50 1,458,446 1,460,088 35.13 36.21
Paleocene 60.650 9.70 50,675 52,276 1,446,411 1,448,011 35.04 36.10 1,461,932 1,463,532 34.66 35.72
Late Cretaceous 82.550 34.10 48,621 50,094 1,444,357 1,445,829 33.66 34.65 1,465,482 1,466,955 33.18 34.15
Early Cretaceous 122.550 45.90 47,872 55,057 1,443,607 1,450,793 33.16 37.95 1,474,969 1,482,154 32.46 37.15
Late Jurassic 153.350 15.70 55,321 62,419 1,451,057 1,458,155 38.12 42.81 1,490,301 1,497,398 37.12 41.68
Middle Jurassic 168.400 14.40 54,986 61,998 1,450,721 1,457,734 37.90 42.53 1,493,816 1,500,829 36.81 41.31
Early Jurassic 187.600 24.00 57,864 64,725 1,453,599 1,460,461 39.81 44.32 1,501,608 1,508,469 38.53 42.91
Late Triassic 213.800 28.40 57,819 64,487 1,453,555 1,460,223 39.78 44.16 1,508,268 1,514,936 38.33 42.57
Middle Triassic 236.500 17.00 56,769 63,793 1,452,505 1,459,529 39.08 43.71 1,513,027 1,520,052 37.52 41.97
Early Triassic 248.000 6.00 56,564 63,553 1,452,300 1,459,288 38.95 43.55 1,515,766 1,522,754 37.32 41.74
Late Permian 260.800 19.60 59,604 68,206 1,455,340 1,463,942 40.96 46.59 1,522,081 1,530,683 39.16 44.56
Early Permian 284.800 28.40 64,673 73,243 1,440,409 1,448,979 44.90 50.55 1,513,292 1,521,862 42.74 48.13
M and L Carboniferous 308.550 19.10 64,584 73,131 1,435,320 1,443,867 45.00 50.65 1,514,281 1,522,828 42.65 48.02
Early Carboniferous 338.650 41.10 62,804 71,294 1,453,539 1,462,030 43.21 48.76 1,540,203 1,548,694 40.78 46.04
Late Devonian 372.250 26.10 62,394 71,333 1,458,130 1,467,068 42.79 48.62 1,553,392 1,562,331 40.17 45.66
Middle Devonian 391.400 12.20 63,276 72,215 1,459,012 1,467,950 43.37 49.19 1,559,175 1,568,114 40.58 46.05
Early Devonian 406.750 18.50 62,317 71,267 1,458,053 1,467,002 42.74 48.58 1,562,144 1,571,094 39.89 45.36
Late Silurian 422.100 12.20 61,901 70,851 1,457,637 1,466,587 42.47 48.31 1,565,657 1,574,606 39.54 45.00
Early Silurian 435.950 15.50 61,024 69,974 1,456,760 1,465,710 41.89 47.74 1,568,324 1,577,274 38.91 44.36
Late Ordovician 452.300 17.20 60,004 68,954 1,455,740 1,464,690 41.22 47.08 1,571,488 1,580,438 38.18 43.63
Middle Ordovician 466.350 10.90 59,614 68,563 1,455,350 1,464,299 40.96 46.82 1,574,693 1,583,643 37.86 43.29
Early Ordovician 480.050 16.50 58,561 67,511 1,449,297 1,458,247 40.41 46.30 1,572,147 1,581,096 37.25 42.70
Late Cambrian 494.650 12.70 57,789 66,739 1,453,525 1,462,475 39.76 45.63 1,580,111 1,589,061 36.57 42.00
Middle Cambrian 507.000 12.00 58,514 67,464 1,454,250 1,463,200 40.24 46.11 1,583,996 1,592,946 36.94 42.35
Early Cambrian 527.500 29.00 71,976 80,926 1,467,712 1,476,661 49.04 54.80 1,602,704 1,611,654 44.91 50.21
39W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

higher salinities in the early Neogene after initiation of
the glaciation of Antarctica and before the extractions
into the Persian Gulf, Red Sea, Mediterranean and
Carpathian regions. The maximum and minimum
evaporite models diverge in the Mesozoic, primarily
because of the uncertainties involved in estimating the
size of the salt deposits in young ocean basins. The large
decline in ocean salinity through the Mesozoic is related
to the major salt extractions into the Gulf of Mexico and
Atlantic, and indicate that in the early Mesozoic
salinities would have been in the high 30's to low
40's (‰) depending on the minimum or maximum
evaporite model. Palaeozoic salinities were in the low to
high 40's and reached the 50's (‰) in the Early
Cambrian, depending on the evaporite data. Water mass
model B reduces the salinities by about 5‰for both
maximum and minimum evaporite models. Assuming
that salts were subducted along with the water would
bring the salinities back up almost to the level of the
constant water mass model A, implying only a change in
the mass of seawater. There would be a small difference
because the seawater subducted early in the Phanerozoic
would have been somewhat more saline than that in the
later Phanerozoic.
8. Implications
Salinity and temperature affect a number of the
physical properties of seawater, including its density,
specific heat, saturation vapor pressure, and osmotic
pressure. The largest effects are related to the density
where there is a complex interplay between temperature
and salinity. These relationships are described by the
equation of state for seawater (Millero et al., 1980;
Millero and Poisson, 1981), which is highly non-linear.
The Millero et al. (1980) and Millero and Poisson
(1981) equation of state is a polynomial with 17 terms,
one of which is the bulk modulus. The bulk modulus is
described by another polynomial with 25 terms. This
equation of state is thought to be valid for the range T=
−7°C–50 °C and S=0–60‰(Millero, personal
communication). Salinity causes only a minor effect
on specific heat which decreases by about 9% from a
salinity of 0‰to 60‰. This is unlikely to have any
Fig. 7. Four models reconstructing the mean salinity of the ocean during the Phanerozoic. The short-dashed lines are the maximum and minimum
estimates based on water Model A (no loss of water through subduction); the upper solid line is the average for Model A. The long-dashed lines are
the maximum and minimum estimates based on water Model B (water steadily lost through subduction); the lower solid line is the average for Model
B. In the Palaeozoic, the average for Model A nearly coincides with the maximum for Model B, and the average for Model B nearly coincides with the
minimum for Model A.
40 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

detectable influence on climate. Similarly, saturation
vapor pressure is overwhelmingly dominated by the
effect of temperature. Increased salinity lowers the
saturation vapor pressure but only very slightly, so that it
is unlikely to have any effect on climate. Osmotic
pressure varies directly with salinity and might be
expected to have a major effect on organisms.
8.1. The effect of ocean salinity differences on
circulation
Rooth (1982) called attention to the fact that the
mean salinity of the ocean has profound implications for
the behavior of the ocean. To understand the implica-
tions of differences in mean ocean salinity for the
thermohaline circulation, one needs to only consider
what would happen if the salinity of the ocean were
significantly lower than it is today. Fig. 8 is a graphical
representation of the equation of state for seawater
(Millero and Poisson, 1981). Below a salinity of 27.4‰,
the maximum density of seawater lies above the freezing
point, and it behaves like fresh water in that the coldest
water will float. This in fact occurs today along the
Arctic shelf off Siberia, so that region has become the
“ice factory”from the Arctic Ocean. In geologic
perspective polar regions with lower than average
salinities could become excluded as major sites of
deep water formation even if they are very cold. This
situation has existed periodically in the Arctic since the
Arctic Ocean Basin became isolated from the Pacific in
the Cretaceous (see palaeogeographic maps of Kazmin
and Napatov, 1998). By contrast, for seawater with
salinities greater than 27.4‰, the maximum density lies
below the freezing point. It might be expected that ice
formation could not occur unless the entire ocean was
chilled to the freezing point, but near the freezing point
salinity plays a critical role. It is thought that sea-ice
formation in the open ocean requires freshening of a
minute surface layer, probably through precipitation in
the form of snowflakes. The snowflakes themselves can
serve as nuclei for ice formation.
With present day ocean salinities, density changes
due to temperature become very small as the freezing
point is approached. At salinities of 30–35‰the
maximum density of seawater is just below the freezing
point. The temperature change as the seawater cools
from 0EC to the freezing point (about −2EC) has
almost no effect on its density. As a result of this
peculiar situation the formation of sea ice becomes an
important factor in increasing the density of sea water.
Sea ice initially has a salinity of about 7‰, so that as it
freezes salt is expelled into the surrounding water,
increasing its density. For this reason, most of the high
latitude sites where deep water is generated involve sea-
ice formation.
The temperature–salinity–density relation becomes
significantly different when ocean salinities are signif-
icantly higher than at present. When the salinity reaches
Fig. 8. Graphical representation of the equation of state of seawater showing the relation between temperature, salinity, pressure, and density. Solid
lines are density at the surface beneath one atmosphere of pressure. Long-dashed lines are density at a pressure of 2000 dbar (=depth of approximately
2 km). Short-dashes lines are density at a pressure of 4000 dbar (=depth of approximately 4 km). After Millero et al. (1980), and Millero and Poisson
(1981). See text for discussion.
41W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

40‰, the density curve never approaches the vertical,
but slopes down to the freezing point. This means that
the densest water in the ocean will be the coldest water.
As the salinity increases, temperature becomes more and
more dominant in controlling the density of seawater.
This has profound implications for formation of high-
latitude bottom waters in the past.
Another factor that plays an important role in deep
water formation is the way in which its volume changes
in response to pressure changes, sometimes termed
compressivity. This is described by its bulk modulus,
which is a strongly non-linear function (Millero and
Poisson, 1981) that is required for the equation of state
to determine the density of seawater below the sea
surface. Like the density, the bulk modulus changes with
both temperature and salinity. Fig. 8 shows the
isopycnals (density lines) for the surface, 2000 dbar
and 4000 dbar, corresponding approximately to depths
of 2 and 4 km. The isopycnals are almost parallel at
temperatures above 20EC. However, they converge at
cooler temperatures. The convergence is the expression
of the fact that colder water is more compressible than
warmer water even if the temperature differences are
very small. This phenomenon is very important in deep
water formation in the ocean today. As two surface
waters having the same density but slightly different
temperatures begin to sink, the cooler water will be
compressed more rapidly. Hence its density will become
greater and it will “win the race”to the bottom, and will
become the bottom water. The warmer water will stop
sinking at an intermediate level. The compressibility
also changes with salinity; as shown in Fig. 8 the salinity
increases, the surface, 2000 dbar and 4000 dbar lines
become more parallel as they approach the freezing
point. At higher salinities the difference in compress-
ibility of colder water becomes less important in deep
water formation.
Today the thermohaline circulation is driven by
small density differences (Hay, 1993) and this has
undoubtedly been true in the past. Today, formation of
the deep waters that drive thermohaline circulation
requires a density increase through salinization, and
this involves phase changes. At high latitudes this
salinization results from the phase change from
seawater to sea ice, and involves about 0.34 × 10
6
J/kg.
Dense water can also form at low latitudes, as in the
Mediterranean today. There it involves salinization
through evaporation; the phase change is from water
to vapor and involves about 2.5 × 10
6
J/kg. At higher
salinities deep water formation can be achieved
simply by cooling the water, which requires only
0.0042× 10
6
J/kg.
Changing mean ocean salinity implies significant
changes in the thermohaline circulation. If salinization is
not required to produce deep water, the energy required
by the phase changes at the atmosphere–ocean surface is
eliminated. The phase changes required to increase
seawater density today act as a flywheel on the ocean–
atmosphere system, consuming energy and thereby
slowing down the rate at which deep water formation
can take place. At higher ocean salinities these phase
changes are not required, and much less energy is
needed to drive the thermohaline circulation. A more
saline ocean could convect much more readily than it
does today.
8.2. The effect of salinity on marine life
Many palaeontologists assume that conditions in the
ocean have remained much as they are today, but some
obvious changes have occurred. Stanley and Hardie
(1998, 1999) have described the alternations between
calcite and aragonite secreting organisms through the
Phanerozoic, attributed to changing relative proportions
of Ca
2+
and Mg
2+
in seawater. However, it has generally
been assumed that most marine organisms cannot
tolerate salinities much higher that those of the open
ocean today. Interestingly, coral reefs in the northern
part of the Red Sea are among the most diverse on Earth
in terms of both corals and fish (Schuhmacher, 1991).
Today they thrive in salinities ranging between 41‰and
43‰, and their ancestors must have survived even
higher salinities during the Last Glacial Maximum. It
may be that at present many marine animals and plants
are living nearer their low salinity tolerance limit rather
than their high limit.
Palaeozoic fossils present a special opportunity to
explore the hypothesis that ocean salinities then were
significantly higher than today. There are very few
Palaeozoic fossils, such as inarticulate brachiopod
Lingula, that have modern counterparts, and those are
almost always found in peculiar environments. It must
be recalled that the fossil record prior to the Jurassic is
almost exclusively from epeiric and shelf seas. These
areas are fundamentally different from open ocean in
that their salinity depends on local fresh water balance.
The fresh water balance is an expression of the supply of
fresh water through precipitation and runoff and loss of
water to evaporation. Fresh water balance is negative in
mid-latitudes where it is reflected in deserts on land and
higher salinity marginal seas such as the Mediterranean.
Fresh water balance is positive at high latitudes where
marginal seas such as the Baltic and Hudson Bay have
low salinities. The epeiric seas of the geologic past,
42 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

some of which have very diverse assemblages of fossils
probably reflect abnormal conditions, having salinities
either higher or lower than the mean of the world ocean
at the time.
The best record of oceanic salinity would be in
pelagic deposits, but these are almost unknown for the
early Mesozoic and Palaeozoic. As far as is known,
acritarchs were a major constituent of the Palaeozoic
ocean plankton, but are rare in younger deposits
(Mendelson, 1993). The evolution of the calcareous
nannoplankton (Brown et al., 2004) fits well with the
hypothesized decline in ocean salinity during the
Mesozoic. Jurassic calcareous nannoplankton are most-
ly found in shelf deposits, and the invasion of the ocean
proper seems to have occurred in mid- and Late
Cretaceous. The planktonic foraminifera have a similar
history (Culver, 1993). One peculiar group is the
Radiolaria, which are known from the Cambrian on.
Curiously Palaeozoic radiolaria are known from shallow
water deposits (Casey, 1993), whereas today they are
strictly inhabitants of the open ocean.
There is a question why the organic carbon-rich
deposits of the Late Jurassic and Early Cretaceous are
such prolific producers of petroleum. The coincidence
of petroleum-prone organic carbon-rich deposits and
sharply declining ocean salinity may not be unrelated.
In many instances Cyanobacteria are important
contributors of organic carbon in these deposits. It
may be that the rapidly declining salinities of these
times, when very large salt extractions were occurring
in the Gulf of Mexico and South Atlantic, were
instrumental in favoring an abundance of these
organisms.
9. Summary and conclusions
The reconstructions presented here closely resemble
those that have been published previously based on less
complete data (Holser et al., 1980; Hay and Wold, 1997;
Hay et al., 1998; Floegel et al., 2000; Hay et al., 2001).
The data suggest that there have been significant
changes in the mean salinity of the ocean during the
Phanerozoic. The biggest changes are related to major
extractions of salt into the young ocean basins which
developed as Pangaea broke apart. Unfortunately, these
salt deposits are the least well known. The last big
extractions were those of the Miocene, and they
occurred just after there had been a large scale extraction
of water from the ocean to form the ice cap of
Antarctica. However, these two modifications of the
masses of H
2
O and salt in the ocean followed in
sequence and did not cancel each other out. Accord-
ingly, salinities during the Early Miocene were between
37‰and 39‰.
The Mesozoic was a time of generally declining
salinity associated with the deep sea salt extractions of
the North Atlantic and Gulf of Mexico (Middle to Late
Jurassic) and South Atlantic (Early Cretaceous). Al-
though none of these salt layers have been penetrated by
drilling, the extent and thickness of salt in the Gulf of
Mexico is relatively well known. The thin edge of that in
the South Atlantic has been drilled along the Brazilian
margin, but is known to thicken offshore. The North
Atlantic salt deposits are known only though seismic
interpretation. The decline in salinity corresponds
closely to the evolution of both planktonic foraminifera
and calcareous nannoplankton. Both groups were
restricted to shelf regions and in the Jurassic and early
Cretaceous, but spread into the open ocean in the mid-
Cretaceous. Their availability to inhabit the open ocean
may be directly related to the decline in salinity.
The earliest of the major extractions of the
Phanerozoic occurred during the Permian and involved
both halite and unusually large amounts of gypsum/
anhydrite. These may have created stress for marine
organisms and may have been a factor contributing to
the end-Permian extinction. There were few major
extractions of salt during the Palaeozoic. The models
suggest that this was a time of relatively stable but
slowly increasing salinities ranging through the upper
30 and low 40‰range into the lower 50‰concentra-
tions in the Early Cambrian.
The modeling suggests that there was a major salinity
decline from the Late Precambrian to the Cambrian, and
it is tempting to speculate that this may have been a
factor in the Cambrian explosion of life. However, the
Late Precambrian salt deposits of the Hormuz region
cannot be precisely dated, and the apparent sharp
decline before the Cambrian may be an artifact of the
lack of better information.
The largest uncertainties in the reconstruction of past
ocean salinities lie first in the knowledge of the history
of water on Earth. Has the mass of free water remained
nearly constant or has it grown or decreased with time?
Is the recycling of saline basin waters significant? How
much of the salt in subducted seawater is returned to the
ocean? Is there a steady flux of juvenile chlorine from
the mantle, or a steady loss to the mantle? Are the deep
sea deposits of salt as large as we think, or are there
possible other deep sea deposits that remain undiscov-
ered? Is there an unknown mechanism for regulating the
salinity of the ocean? The answers to these questions
must lie in innovative new approaches to understanding
the history of seawater.
43W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Acknowledgements
This work was carried out with the support from the
Deutsche Forschungsgemeinschaft through grant HA
2891/1-2. We thank Robert Berner and Klaus Wallmann
for thoughtful reviews and helpful suggestions.
References
Allen, P.A., 1997. Earth Surface Processes. Blackwell Science,
Oxford.
Anderson, A.T., 1974. Chlorine, sulfur, and water in magmas and
oceans. Geological Society of America Bulletin 85, 1485–1492.
Bartels, O.G., 1972. An estimate of volcanic contributions to the
atmosphere and volcanic gases and sublimates as the source of
radioisotopes
10
Be,
35
S,
32
P, and
23
Na. Health Physics 22,
387–392.
Berner, R.A., 2004. A model for calcium, magnesium and sulfate in
seawater over Phanerozoic time. American Journal of Science 304,
438–453.
Berner, E.K., Berner, R.A., 1987. The Global Water Cycle:
Geochemistry and Environment. Prentice-Hall, Englewood Cliffs,
N.J.
Brown, J., Colling, A., Park, D., Phillips, J., Rothery, D., Wright, J.,
1989. Seawater: Its Composition, Properties and Behaviour. Open
University, Pergamon Press, Oxford.
Brown, P.R., Lees, J.A., Young, J.R., 2004. Calcareous nannoplankton
evolution and diversity through time. In: Thierstein, H.R., Young,
J.R. (Eds.), Coccolithophores —from Molecular Processes to
Global Impact. Springer Verlag, Berlin, pp. 481–508.
Budyko, M.I., Ronov, A.B., Yanshin, A.L., 1987. History of the
Earth's Atmosphere. Springer-Verlag, New York.
Burke, K., 1975. Atlantic evaporites formed by evaporation of water
spilled from Pacific, Tethyan, and Southern Oceans. Geology 3,
613–616.
Burke, K., Sengor, A.M., 1988. Ten metre sea-level change associated
with South Atlantic Aptian salt deposition. Marine Geology 83,
309–312.
Casey, R.E., 1993. Radiolaria. In: Lipps, J.H. (Ed.), Fossil
Prokaryotes and Protists. Blackwell Scientific Publications,
Oxford, pp. 249–284.
Correns, C.W., 1956. The geochemistry of the halogens. In:
Ahrens, L.H., Rankama, K., Runcorn, S.K. (Eds.), Physics
and Chemistry of the Earth, vol. 1. Pergamon Press, London,
pp. 181–233.
Culver, S.J., 1993. Foraminifera. In: Lipps, J.H. (Ed.), Fossil
Prokaryotes and Protists. Blackwell Scientific Publications,
Oxford, UK, pp. 203–247.
Denton, G.H., Hughes, T.J., 1981a. The Last Great Ice Sheets. Wiley
Interscience, New York.
Denton, G.H., Hughes, T.J., 1981b. The Arctic ice sheet: an outrageous
hypothesis. In: Denton, G.H., Hughes, T.J. (Eds.), The Last Great
Ice Sheets. Wiley Interscience, New York, pp. 437–467.
Drever, J.I., 1982. The Geochemistry of Natural Waters. Prentice-Hall,
Englewood Cliffs, NJ.
Drever, J.I., Li, Y.-H., Maynard, J.B., 1988. Geochemical cycles:
the continental crust and the oceans. In: Gregor, C.B., Garrels,
R.M., Mackenzie, F.T., Maynard, J.B. (Eds.), Chemical Cycles
in the Evolution of the Earth. John Wiley & Sons, New York,
pp. 17–53.
Duplessy, J.C., Labeyrie, L., Juillet-Leclerc, A., Maitre, F., Duprat, J.,
Sarnthein, M., 1991. Surface salinity reconstruction of the North
Atlantic Ocean during the last glacial maximum. Oceanologica
Acta 14, 311–324.
Duplessy, J.C., Bard, E., Labeyrie, L., Duprat, J., Moyes, J., 1993.
Oxygen isotope records and salinity changes in the northeastern
Atlantic Ocean during the last 18,000 years. Paleoceanography 8,
341–350.
Einsele, G., 1992. Sedimentary Basins: Evolution, Facies, and
Sediment Budget. Springer-Verlag, New York.
Emiliani, C., 1992. Planet Earth. Cambridge University Press,
Cambridge, UK.
Ewing, W.M., Worzel, J.L., Burk, C.A., 1969. Regional aspects of
deep water drilling in the Gulf of Mexico, east of the Bahama
Platform, and on the Bermuda Rise. In: Ewing, W.M., Worzel, J.L.,
Beall, A.O., Berggren, W.A., Bukry, J.D., Burk, C.A., Fischer, A.
G., Pessagno Jr., E.A. (Eds.), Initial Reports of the Deep Sea
Drilling Project, vol. 1, pp. 624–640.
Feth, J.H., 1981. Chloride in natural continental water —a review. U.
S. Geological Survey Water-Supply Paper 2176, 1–30.
Flint, R.F., 1971. Glacial and Quaternary Geology. John Wiley &
Sons, New York.
Floegel, S., Wold, C.N., Hay, W.W., 2000. Evolution of sediments and
ocean salinity. Abstracts Volume, 31st International Geological
Congress, Rio de Janeiro —Brazil, August 6–17, 2000. 4 pp., CD-
ROM.
Frank, L.A., Sigwarth, J.B., 2001. Detection of small comets with a ground-
based telescope. Journal of Geophysical Research 106, 3665–3683.
Frank, L.A., Sigwarth, J.B., Craven, J.D., 1986. On the influx of small
comets into the Earth's upper atmosphere: I. Observations.
Geophysical Research Letters 13, 303–306.
Garrels, R.M., Mackenzie, F., 1971. Evolution of Sedimentary Rocks.
W. W. Norton, New York.
Gill, R., 1989. Chemical Fundamentals of Geology. Unwin Hyman,
London.
Gilluly, J., 1969. Geological perspective and the completeness of the
geologic record. Geological Society of America Bulletin 80,
2303–2312.
Gradstein, F.M., Ogg, J., 1996. A Phanerozoic time scale. Episodes 19,
3–5.
Gradstein, F.M., Ogg, J.G., Smith, A.G., Agterberg, F.P., Bleeker, W.,
Cooper, R.A., Davydov, V., Gibbard, P., Hinnov, L., House, M.R.,
Lourens, L., Luterbacher, H.-P., McArthur, J., Melchin, M.J.,
Robb, L.J., Shergold, J., Villeneuve, M., Wardlaw, B.R., Ali, J.,
Brinkhuis, H., Hilgen, F.J., Hooker, J., Howarth, R.J., Knoll, A.H.,
Laskar, J., Monechi, S., Powell, J., Plumb, K.A., Raffi, I., Röhl, U.,
Sanfilippo, A., Schmitz, B., Shackleton, N.J., Shields, G.A.,
Strauss, H., Van Dam, J., Veizer, J., van Kolfschoten, Th., Wilson,
D., 2004. A Geologic Time Scale 2004. Cambridge University
Press, Cambridge, UK.
Hallam, A., 1992. Phanerozoic Sea-level Changes. Columbia
University Press, New York.
Hardie, L.A., 1996. Secular variation in seawater chemistry: an
explanation for the coupled secular variation in the mineralogies of
marine limestones and potash evaporites over the past 600 m.y.
Geology 24, 279–283.
Hardie, L.A., 2003. Secular variations in Precambrian seawater
chemistry and the timing of Precambrian aragonite seas and
calcite seas. Geology 31, 785–788.
Hay, W.W., 1993. The role of polar deep water formation in global
climate change. Annual Review of Earth and Planetary Sciences
21, 227–254.
44 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Hay, W.W., 1994. Pleistocene–Holocene fluxes are not the Earth's
norm. In: Hay, W.W. (Ed.), Material Fluxes on the Surface of the
Earth. National Academy Press, Washington, D.C., pp. 15–27.
Hay, W.W., 1999. Carbonate sedimentation through the late Precam-
brian and Phanerozoic. Zentralblatt für Geologie und Paläontolo-
gie, Teil I, 1998, Heft 5–6, 435–445.
Hay, W.W., Leslie, M.A., 1990. Could possible changes in global
groundwater reservoir cause eustatic sea-level fluctuations. In:
Revelle, R. (Ed.), Sea-Level Change. National Academy Press,
Washington, D.C., pp. 161–170.
Hay, W.W., Wold, C.N., 1997. Preliminary reconstruction of the
salinity of the ocean in the Cenozoic and Mesozoic. Freiberger
Forschungshefte, C 468 Geowissenschaften-Paläontologie,
Paläontologie, Stratigraphie, Fazies, Heft 5, Karl-Armin-Tröger-
Festschrift, 119–127.
Hay, W.W., Sloan II, J.L., Wold, C.N., 1988. The mass/age distribution
of sediments on the ocean floor and the global rate of loss of
sediment. Journal of Geophysical Research 93, 14933–14940.
Hay, W.W., Wold, C.N., DeConto, R.M., 1998. The role of salinity in
circulation of the Cretaceous ocean. Zentralblatt für Geologie und
Paläontologie 1996 (11/12), 1445–1454.
Hay, W.W., Wold, C.N., Söding, E., Flögel, S., 2001. Evolution of
sediment fluxes and ocean salinity. In: Merriam, D.F., Davis, J.C.
(Eds.), Geologic Modeling and Simulation: Sedimentary Systems.
Kluwer Academic/Plenum Publishers, Dordrecht, The Nether-
lands, pp. 153–167.
Holland, H.D., 1974. Marine evaporites and the composition of
sea water during the Phanerozoic. In: Hay, W.W. (Ed.), Studies
in Paleo-Oceanography. Society of Economic Paleontologists
and Mineralogists Special Publication, vol. 20. Society of
Economic Paleontologists and Mineralogists, Tulsa, Oklahoma,
pp. 187–192.
Holland, H.D., 1978. The Chemistry of the Atmosphere and Oceans.
Wiley, New York.
Holland, H.D., 1984. The Chemical Evolution of the Atmosphere and
Oceans. Princeton University Press, Princeton, NJ.
Holser, W.T., 1977. Catastrophic chemical events in the history of the
ocean. Nature 267, 403–498.
Holser, W.T., 1985. Gradual and abrupt shift in ocean chemistry during
Phanerozoic time. In: Holland, H.D., Trendall, A.F. (Eds.), Patterns
of Change in Earth Evolution. Dahlem Konferenzen 1984.
Springer Verlag, Berlin, Germany, pp. 123–143.
Holser, W.T., Hay, W.W., Jory, D.E., O'Connell, W.J., 1980. A census
of evaporites and its implications for oceanic geochemistry.
Abstracts with Programs - Geological Society of America 12/7,
449.
Humphris, S.E., McCollom, T., 1998. The cauldron beneath the sea
floor. Oceanus 18/2, 18–21.
Japakasetr, T., Workman, D.R., 1981. Evaporite deposits of Thailand.
In: Halbouty, M.T. (Ed.), Energy Resources of the Pacific Region.
American Association of Petroleum Geologists, Tulsa, Oklahoma,
pp. 179–187.
Joly, J., 1899. An estimate of the geological age of the Earth.
Transactions of the Royal Society of Dublin 2 (7), 23–66.
Joly, J., 1925. The Surface History of the Earth. Clarendon Press,
Oxford, UK.
Junge, C.E., Werby, R.T., 1958. The concentration of chloride, sodium,
potassium, calcium, and sulfate in rain water over the United
States. Journal of Meteorology 15, 417–425.
Kazmin, V.G., Napatov, L.M. (Eds.), 1998. The Paleogeographic Atlas
of Northern Eurasia. Institute of Tectonics of the Lithospheric
Plates, Russian Academy of Natural Sciences, Moscow, Russia.
Knauth, L.P., 1998. Salinity history of the Earth's early ocean. Nature
395, 554–555.
Knauth, L.P., 2005. Temperature and salinity history of the
Precambrian ocean: implications for the course of microbial
evolution. Palaeogeography, Palaeoclimatology, Palaeoecology
219, 53–69.
Kurlansky, M., 2002. Salt —A World History. Walker and Company,
New York.
Land, L.S., 1995. The role of saline formation water in crustal cycling.
Aquatic Geochemistry 1, 137–145.
Lisitzin, A.P., 1974. Osadkovrazovanie v Okeanach. Izdatelstvo
“Nauka”, USSR, Moscow.
Livingston, D.A., 1963. Chemical composition of rivers and
lakes, Chapter G. In: Fleischer, M. (Ed.), Data of Geochem-
istry. U.S. Geological Survey Professional Paper, vol. 440-G.
64 pp.
Lord Kelvin, W.T., 1864. On the secular cooling of the earth.
Transactions of the Royal Society of Edinburgh 23,
157–170.
Lowenstein, T.K., Timofeeff, M.N., Brennan, S.T., Hardie, L.A.,
Demicco, R.V., 2001. Oscillations in Phanerozoic seawater
chemistry: evidence from fluid inclusions. Science 294,
1086–1088.
Marcinek, J., Rosenkranz, E., 1989. Das Wasser der Erde: Lehrbuch
der geographischen Meeres- und Gewässerkunde. Verlag Harri
Deutsch, Thun, Switzerland.
McArthur, J.M., Howarth, R.J., Bailey, T.R., 2001. Strontium isotope
stratigraphy: LOWESS Version 3: best fit to the marine Sr-isotope
curve for 0–509 Ma and accompanying look-up table for deriving
numerical age. Journal of Geology 109, 155–170.
Mendelson, C.V., 1993. Acritarchs and prasinophytes. In: Lipps, J.H.
(Ed.), Fossil Prokaryotes and Protists. Blackwell Scientific
Publications, Oxford, UK, pp. 77–104.
Meybeck, M., 1979. Concentrations des eaux fluviales en éléments
majeurs et apports en solution aux océans. Revue de Géologie
Dynamique et de Géographie Physique 21, 215–246.
Meybeck, M., 1983. Atmospheric inputs and river transport of
dissolved substances. International Association of Hydrological
Sciences Special Publication 141, 173–192.
Meybeck, M., 1987. Global chemical weathering from surficial rocks
estimated from river dissolved loads. American Journal of Science
287, 401–428.
Millero, F.J., Poisson, A., 1981. International one-atmosphere equation
of state for sea water. Deep-Sea Research 28A, 625–629.
Millero, F.J., Chen, C.T., Bradshaw, A., Schleicher, K., 1980. A new
high pressure equation of state for seawater. Deep-Sea Research
27A, 255–264.
Mueller, R.F., Saxena, S.K., 1977. Chemical Petrology: with
Applications to Terrestrial Planets and Meteorites. Springer-
Verlag, New York.
Parrish, T.J., Ziegler, A.M., Scotese, C.R., 1982. Rainfall patterns and
the distribution of coals and evaporites in the Mesozoic and
Cenozoic. Palaeogeography, Palaeoclimatology, Palaeoecology
40, 67–101.
Railsback, L.B., Anderson, T.F., Ackerly, S.C., Cisne, J.L., 1989.
Paleoceanographic modeling of temperature–salinity profiles from
stable isotopic data. Paleoceanography 4, 585–591.
Rankama, K., Sahama, Th.G., 1950. Geochemistry. Chicago Univer-
sity Press, Chicago, Illinois, USA.
Ronov, A.B., 1968. Probable changes in the composition of
seawater during the course of geologic time. Sedimentology 10,
25–43.
45W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46

Ronov, A.B., 1972. Evolution of rock composition and geochemical
processes in the sedimentary shell of the earth. Sedimentology 19,
157–172.
Ronov, A.B., 1980. Osadochnaya Obolochka Zemli (Kolichestvennoe
Zakonomernosti Stroeniya, Sostava y Evoliutsii. Izdatelstvo
“Nauka”, USSR, Moskva. 78 pp.
Ronov, A.B., 1982. The earth's sedimentary shell (Quantitative
patterns of its structure, compositions, and evolution). International
Geology Review 24, 1365–1388.
Ronov, A.B., 1993. In: Yaroshevskii, A.A. (Ed.), Stratisfera —Ili
Osadochnaya Obolochka Zemli (Kolichestvennoe Issledovanie).
Nauka, Moskva. 144 pp.
Ronov, A.B., 1994. Phanerozoic transgressions and regressions on the
continents: a quantitative approach based on areas flooded by the
sea and areas of marine and continental deposition. American
Journal of Science 294, 777–801.
Ronov,A.B.,Khain,V.Y.,Seslavinsky, K.B., 1984. Atlas of
Lithological–Paleogeographical Maps of the World: Late Precam-
brian and Paleozoic of the Continents. U.S.S.R. Academy of
Science Press, Leningrad.
Ronov, A.B., Khain, V.E., Balukhovsky, A.N., 1989. In: Barsukhov, V.
L., Laviorov, N.P. (Eds.), Atlas of Lithological–Paleogeographical
Maps of the World: Mesozoic and Cenozoic of Continents and
Oceans. Editorial Publishing Group, Moscow.
Rooth, C., 1982. Hydrology and ocean circulation. Progress in
Oceanography 11, 131–149.
Rubey, W.W., 1951. Geologic history of sea water. Geological Society
of America Bulletin 62, 1111–1148.
Ryan, J.A., Mukherjee, N.R., 1975. Sources of atmospheric gaseous
chlorine. Reviews of Geophysics and Space Physics 11, 650–658.
Sandberg, P.A., 1983. An oscillating trend in Phanerozoic nonskeletal
carbonate mineralogy. Nature 305, 19–22.
Scotese, C.R., Bambach, R.K., Barton, C., Van der Voo, R., Ziegler,
A., 1979. Paleozoic base maps. Journal of Geology 87, 217–277.
Shackleton, N.J., 1987. Oxygen isotopes, ice volume and sea level.
Quaternary Science Reviews 6, 183–190.
Shackleton, N.J., Kennett, J.P., 1975. Paleotemperature history of the
Cenozoic and the initiation of Antarctic glaciation: oxygen and
carbon isotope analyses on DSDP sites 277, 279, and 281. In:
Kennett, J.P., Houtz, R.E., et al. (Eds.), Initial Reports of the Deep
Sea Drilling Project, vol. 29. U.S. Government Printing Office,
Washington, D.C., pp. 743–755.
Schuhmacher, H., 1991. Korallenriffe —Verbreitung, Tierwelt,
Ökologie. BLV Verlagsgesellschaft mbH, München, Germany.
Southam, J.R., Hay, W.W., 1981. Global sedimentary mass
balance and sea level changes. In: Emiliani, C. (Ed.), The
Sea, v. 7, the Oceanic Lithosphere. Wiley-Interscience, New
York, pp. 1617–1684.
Stallard, R.F., Edmond, J.M., 1981. Chemistry of the Amazon,
precipitation chemistry and the marine contribution to the
dissolved load at the time of peak discharge. Journal of
Geophysical Research 86, 9844–9858.
Stanley, S.M., Hardie, L.A., 1998. Secular oscillations in the carbonate
mineralogy of reef-building and sediment-producing organisms
driven by tectonically forced shifts in seawater chemistry.
Palaeogeography, Palaeoclimatology, Palaeoecology 144, 3–19.
Stanley, S.M., Hardie, L.A., 1999. Hypercalcification: paleontology
links plate tectonics and geochemistry to sedimentology. GSA
Today 9 (2), 1–7.
Sverdrup, H.U., Johnson, M.W., Fleming, R.H., 1942. The Oceans,
Their Physics, Chemistry, and General Biology. Prentice-Hall,
New York.
Tardy, Y., N'kounkou, R., Probst, J.-L., 1989. The global water cycle
and continental erosion during Phanerozoic time (570 my).
American Journal of Science 289, 455–485.
Turekian, K.K., 1968. Oceans. Prentice-Hall, Englewood Cliffs, NJ.
Veizer, J., Jansen, S.L., 1979. Basement and sedimentary recycling and
continental evolution. Journal of Geology 87, 341–370.
Veizer, J., Jansen, S.L., 1985. Basement and sedimentary cycling —2:
Time dimension to global tectonics. Journal of Geology 93,
625–664.
Von Huene, R., Scholl, D.W., 1991. Observations at convergent
margins concerning sediment subduction, subduction erosion, and
the growth of continental crust. Reviews of Geophysics 29,
279–316.
Wallmann, K., 2001. The geological water cycle and the evolution of
marine δ
18
O values. Geochimica et Cosmochimica Acta 65,
2469–2485.
Wilkinson, B., 2005. Humans as geologic agents: a deep-time
perspective. Geology 33, 161–164.
Wold, C.N., Hay, W.W., 1990. Estimating ancient sediment fluxes.
American Journal of Science 290, 1069–1089.
Wold, C.N., Hay, W.W., 1993. Reconstructing the age and lithology of
eroded sediment. Geoinformatics 4, 137–144.
Wold, C.N., Schwartz, G.J., Morrill, C., 1999. Digital database of
modern evaporites and their predicted distribution based on results
from an atmospheric general circulation model simulation. In:
Harff, J., Lemke, W., Stattegger, K. (Eds.), Computerized
Modeling of Sedimentary Systems. Springer, Berlin, pp. 149–167.
Zachos, J.C., Stott, L.D., Lohmann, K.C., 1994. Evolution of early
Cenozoic marine temperatures. Paleoceanography 9, 353–387.
Zharkov, M.A., 1974. Paleozoickie Solenosn'ie Formatsii Mira.
Nedra, Moscow, USSR.
Zharkov, M.A., 1981. History of Paleozoic salt Accumulation.
Springer Verlag, New York.
46 W.W. Hay et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 240 (2006) 3–46