As for carbonates, the d
18 O of silica is a function of the
temperature and of the d
18 O of the water in which the silica
precipitated:
1000 ln a ¼ 3:09 Â
10
6
T K
ð Þ
2
!
ð26:1Þ
where the fractionation factor
a ¼
1000 þ d
18 O chert SMOW
ð
Þ
1000 þ d
18 O water SMOW
ð
Þ
ð26:2Þ
ln denotes its natural logarithm, and d
18 O water is the d
18 O of
the water from which the cherts precipitated.
Applied to the cherts, assuming that the isotopic composition of seawater was quite similar to that of today, this
paleothermometer predicts the temperatures T of the water
from which the cherts precipitated to be close to 85 °C 3 Ga
ago, and 50 °C at the Precambrian-Cambrian boundary.
These temperatures are either the sign of very hot oceans
or the consequence of an alteration of the cherts after
deposition by meteoric or hydrothermal fluids, in which case
they provide no information on climate. The question
remains open because the d
18 O isotopic signal is particularly
sensitive to diagenesis. A third hypothesis has been formulated more recently: it assumes that d
18
O of seawater was
much lower than its present value, due to tectonic processes.
Under these conditions, the temperature of the sea water
could have been similar to the current one. Unfortunately,
there is no constraint on the d
18 O of seawater during the
Archean. Authors generally assume that this ratio remained
constant over time at −1‰ in comparison with the SMOW,
calculated from the level of the current ocean which would
have received the melt water from all the ice caps on land.
This value is the result of the equilibrium that is supposed to
exist between the
18 O depletion of sea water resulting from
the interactions between water and lithosphere at low temperatures and its
18 O enrichment during water/lithosphere
interactions at high temperature in the hydrothermal systems.
Recent work (Kasting et al. 2006) suggests that the d
18
O
ratio may have been significantly lower (−9‰ compared to
SMOW) during the Archean and Proterozoic periods. This
argument is based on the fact that the Archean oceans were
probably shallower than the oceans are currently. This
results in a shallower water column over the oceanic ridges
before 800 Ma, implying a reduction in hydrostatic pressure
in the hydrothermal systems. This decrease in pressure
limited the penetration of seawater into the ridges in the
depths, thus reducing the gain in
18 O by sea water through
alteration at high-temperature of the oceanic crust
(T > 350 °C). This reduction in flux at high temperature
results in an imbalance in the
18 O cycle in the
ocean-atmosphere system and its stabilization at lower
values than is currently the case. If this scenario proves to be
correct, the temperature of the water as inferred from the
d
18 O data on cherts, could be significantly lower. Kasting
and Howard (2006) argue in favor of ‘moderate’ climates at
the end of the Archean and during the Proterozoic.
An additional element was added to this debate by Robert
and Chaussidon (2006), who measured the isotopic composition of silicon (d
30 Si) in Precambrian cherts which is
distinctly less sensitive to diagenesis than d
18 O. These data,
when translated into temperatures, (requiring the use of a
silicon cycle model and therefore additional assumptions),
suggest temperatures of around 70 °C, 3 Ga ago, and 20 °C,
800 million years ago, thus confirming very high temperatures in the distant past.
There remains a potentially major problem: the formation
conditions of Precambrian cherts are unknown. Nevertheless, these results showing a gradual cooling of sea water
from very high values seem to have been confirmed recently
by a totally independent method, based on the resurrection of
proteins of unicellular Archean organisms using phylogenetic and statistical methods of analysis (Gaucher et al.
2008).
Finally, irrespective of any debate on the terrestrial temperature during the Archean/Proterozoic, the d
18 O of the
cherts show a significant increase at the end of the Archean,
between 2.7 and 2.5 Ga (of about 10‰), suggesting a rapid
cooling of the oceans of about 20 °C (Fig. 26.2).
The Theory of the Paleothermostat
In 1981, Walker, Hays and Kasting published a
ground-breaking article explaining why the climate
remained relatively stable (within a temperature range
allowing water to remain in the liquid state) for between one
million to one billion years (Walker et al. 1981). This study
was carried out in order to solve the faint young sun paradox. The models of stellar evolution predict the evolution of
the solar constant during the history of the Earth and make it
possible to calculate that during the Archean, it would have
been 20–30% weaker than today. Under these conditions,
the Earth should have totally frozen over although this
contradicts the isotopic data which provide an estimate of the
temperature of the fluid envelopes of the Earth.
The residence time of the exosphere carbon content (i.e.
all the carbon contained in the ocean, the biosphere and the
atmosphere) is around 200,000 years which is very short
compared to the geological processes of sedimentary carbon
burial and continental weathering (François and Goddéris
1998). This measure gives an indication of the average time
spent by a carbon atom entering the ocean-atmosphere system via volcanic degassing, for example, before exiting via
sedimentary deposits. This response time is very short in the
26 The Precambrian Climate
345
18 O of silica is a function of the
temperature and of the d
18 O of the water in which the silica
precipitated:
1000 ln a ¼ 3:09 Â
10
6
T K
ð Þ
2
!
ð26:1Þ
where the fractionation factor
a ¼
1000 þ d
18 O chert SMOW
ð
Þ
1000 þ d
18 O water SMOW
ð
Þ
ð26:2Þ
ln denotes its natural logarithm, and d
18 O water is the d
18 O of
the water from which the cherts precipitated.
Applied to the cherts, assuming that the isotopic composition of seawater was quite similar to that of today, this
paleothermometer predicts the temperatures T of the water
from which the cherts precipitated to be close to 85 °C 3 Ga
ago, and 50 °C at the Precambrian-Cambrian boundary.
These temperatures are either the sign of very hot oceans
or the consequence of an alteration of the cherts after
deposition by meteoric or hydrothermal fluids, in which case
they provide no information on climate. The question
remains open because the d
18 O isotopic signal is particularly
sensitive to diagenesis. A third hypothesis has been formulated more recently: it assumes that d
18
O of seawater was
much lower than its present value, due to tectonic processes.
Under these conditions, the temperature of the sea water
could have been similar to the current one. Unfortunately,
there is no constraint on the d
18 O of seawater during the
Archean. Authors generally assume that this ratio remained
constant over time at −1‰ in comparison with the SMOW,
calculated from the level of the current ocean which would
have received the melt water from all the ice caps on land.
This value is the result of the equilibrium that is supposed to
exist between the
18 O depletion of sea water resulting from
the interactions between water and lithosphere at low temperatures and its
18 O enrichment during water/lithosphere
interactions at high temperature in the hydrothermal systems.
Recent work (Kasting et al. 2006) suggests that the d
18
O
ratio may have been significantly lower (−9‰ compared to
SMOW) during the Archean and Proterozoic periods. This
argument is based on the fact that the Archean oceans were
probably shallower than the oceans are currently. This
results in a shallower water column over the oceanic ridges
before 800 Ma, implying a reduction in hydrostatic pressure
in the hydrothermal systems. This decrease in pressure
limited the penetration of seawater into the ridges in the
depths, thus reducing the gain in
18 O by sea water through
alteration at high-temperature of the oceanic crust
(T > 350 °C). This reduction in flux at high temperature
results in an imbalance in the
18 O cycle in the
ocean-atmosphere system and its stabilization at lower
values than is currently the case. If this scenario proves to be
correct, the temperature of the water as inferred from the
d
18 O data on cherts, could be significantly lower. Kasting
and Howard (2006) argue in favor of ‘moderate’ climates at
the end of the Archean and during the Proterozoic.
An additional element was added to this debate by Robert
and Chaussidon (2006), who measured the isotopic composition of silicon (d
30 Si) in Precambrian cherts which is
distinctly less sensitive to diagenesis than d
18 O. These data,
when translated into temperatures, (requiring the use of a
silicon cycle model and therefore additional assumptions),
suggest temperatures of around 70 °C, 3 Ga ago, and 20 °C,
800 million years ago, thus confirming very high temperatures in the distant past.
There remains a potentially major problem: the formation
conditions of Precambrian cherts are unknown. Nevertheless, these results showing a gradual cooling of sea water
from very high values seem to have been confirmed recently
by a totally independent method, based on the resurrection of
proteins of unicellular Archean organisms using phylogenetic and statistical methods of analysis (Gaucher et al.
2008).
Finally, irrespective of any debate on the terrestrial temperature during the Archean/Proterozoic, the d
18 O of the
cherts show a significant increase at the end of the Archean,
between 2.7 and 2.5 Ga (of about 10‰), suggesting a rapid
cooling of the oceans of about 20 °C (Fig. 26.2).
The Theory of the Paleothermostat
In 1981, Walker, Hays and Kasting published a
ground-breaking article explaining why the climate
remained relatively stable (within a temperature range
allowing water to remain in the liquid state) for between one
million to one billion years (Walker et al. 1981). This study
was carried out in order to solve the faint young sun paradox. The models of stellar evolution predict the evolution of
the solar constant during the history of the Earth and make it
possible to calculate that during the Archean, it would have
been 20–30% weaker than today. Under these conditions,
the Earth should have totally frozen over although this
contradicts the isotopic data which provide an estimate of the
temperature of the fluid envelopes of the Earth.
The residence time of the exosphere carbon content (i.e.
all the carbon contained in the ocean, the biosphere and the
atmosphere) is around 200,000 years which is very short
compared to the geological processes of sedimentary carbon
burial and continental weathering (François and Goddéris
1998). This measure gives an indication of the average time
spent by a carbon atom entering the ocean-atmosphere system via volcanic degassing, for example, before exiting via
sedimentary deposits. This response time is very short in the
26 The Precambrian Climate
345