sequences, respectively. On the ocean floor, they are separated by the normal, calm Cretaceous period, between 118
and 84 Ma, during which no geomagnetic reversal occurred
and therefore no magnetic anomaly is observed.
Some important features of the geomagnetic field appear
evident when looking at the scale of magnetic polarities. On
the one hand, the total time for normal and reverse polarity
intervals is essentially the same, with no tendency for the
field to remain in one or the other of the polarities. On the
other hand, during the Cenozoic, the rate of reversal
increased: there was a reversal every million years or so at
the beginning of this period, but this reached four reversals
per million years during the last 5 million years. The current
normal period which has lasted for about 780,000 years
therefore seems abnormally long.
Before the Upper Jurassic, the entire ocean floor corresponding to the current tectonic phase was absorbed in ocean
trenches. As a result, the polarity scale can only be extended
in time through the study of continental rocks and is much
less continuous and precise than for the Cenozoic and
Mesozoic eras. The most obvious and probably most documented characteristic is the long period of reverse polarity
during the Permo-Carboniferous which lasted about 70
million years, and which is called the Kiaman interval. This
interval was preceded and followed by periods where the
field reversed frequently.
Astronomical Calibration of the Polarity Scale
In a famous article published in 1976, Jim Hays, John Imbrie
and Nicholas Shackleton were the first to show that some
indicators (proxies) of paleoclimate, such as isotopic records
of oxygen, evolved over time depending on the orbital
cycles of the obliquity, eccentricity and precession. These
cycles, initially calculated by Milutin Milankovitch, were
established much more precisely by André Berger, and gave
paleoclimatologists precise solutions for the last three million years. Hays et al. have “adjusted” their initial age model
to “match” the obliquity cycle in the record to that given by
astronomical calculation. In doing so, they established the
first ‘cyclostratigraphy’. This method was widely used in the
1980s to constrain the ages of isotopic records of oxygen
during the Brunhes period with a precision of a few thousand
years.
Subsequently, the development of a hydraulic corer in the
DSDP program resulted in the acquisition of previously
inaccessible, very deep sediments, allowing the study of
continuous sequences in even older sediments. Applied to
the Matuyama period, astronomical calibration based on
obliquity cycles did not initially show significant age differences between the limits of “astronomical” polarity and
those of the Mankinen and Dalrymple scale. But when an
astro-timeline, based on precession, was obtained at
Site ODP 677, it became clear that the polarity intervals of
the Mankinen and Dalrymple scale were not accurately
dated.
This study paved the way for a complete revision of the
ages of the polarity intervals, especially during the Gauss and
Gilbert periods (Hilgen 1991a, b). The realization of the
“youth” of the ages obtained by the K/Ar method for the
Plio-Pleistocene era compared to those obtained by
cyclostratigraphy has led to the extensive use of the
40 Ar/
39 Ar
method (Chap. 5), which had recently been developed to test
the validity of cyclostratigraphic ages of this period. The age
of the Brunhes/Matuyama reversal (initially set at 0.73 Ma
by K/Ar) has been re-evaluated to 0.78 Ma due to a large
number of independent
40 Ar/
39 Ar measurements. The latter
value is consistent with the cyclostratigraphic age. Good
agreement is also observed for the Plio-Pleistocene era,
where the new
40 Ar/
39 Ar estimations coincide with the ages
given by Shackleton et al. (1990), and Hilgen (1991a, b).
Although these new estimations, in general, confirmed
the astronomical ages, the ages of geomagnetic reversals
given by Cande and Kent (1992), based on magnetic
anomalies at sea, did not seem to be in agreement with
astronomical dating. The authors had in fact adopted durations for the Plio-Pleistocene polarity intervals that have
proven inaccurate. As mentioned above, it is precisely the
astronomical ages for the Plio-Pleistocene that Cande and
Kent (1995) adopted in their new version of the scale. The
consistency they then obtained between
40 Ar/
39 Ar estimates,
magnetic anomalies at sea and astronomical calculations
over the entire time period covered by the new scale can be
considered as a validation of this method. With this method,
the age of a geomagnetic reversal can be estimated within
the duration of one precession cycle, in other words, less
than two or three times the length of a polarity reversal. This
very high resolution led Renne et al. (1994) to calibrate the
age of one of the standards of the
40 Ar/
39 Ar method (Fish
Canyon Tuff, FCTs) with the astronomical method, which
reduced the uncertainty to 0.6% for the calibrated ages
compared with the standard.
Principle and Practice of Magnetostratigraphy
The magnetic polarities scale shows that polarity reversals
are largely random in time, which means that sequences of
four or five successive reversals do not repeat themselves
identically in time; therefore, they constitute a kind of
“fingerprint” of specific geological periods. Magnetostratigraphy is based on this characteristic: if a sequence
characteristic of the magnetostratigraphic scale can be
identified within a particular series, then a specific age can
be assigned to this section.
106
C. Laj et al.
and 84 Ma, during which no geomagnetic reversal occurred
and therefore no magnetic anomaly is observed.
Some important features of the geomagnetic field appear
evident when looking at the scale of magnetic polarities. On
the one hand, the total time for normal and reverse polarity
intervals is essentially the same, with no tendency for the
field to remain in one or the other of the polarities. On the
other hand, during the Cenozoic, the rate of reversal
increased: there was a reversal every million years or so at
the beginning of this period, but this reached four reversals
per million years during the last 5 million years. The current
normal period which has lasted for about 780,000 years
therefore seems abnormally long.
Before the Upper Jurassic, the entire ocean floor corresponding to the current tectonic phase was absorbed in ocean
trenches. As a result, the polarity scale can only be extended
in time through the study of continental rocks and is much
less continuous and precise than for the Cenozoic and
Mesozoic eras. The most obvious and probably most documented characteristic is the long period of reverse polarity
during the Permo-Carboniferous which lasted about 70
million years, and which is called the Kiaman interval. This
interval was preceded and followed by periods where the
field reversed frequently.
Astronomical Calibration of the Polarity Scale
In a famous article published in 1976, Jim Hays, John Imbrie
and Nicholas Shackleton were the first to show that some
indicators (proxies) of paleoclimate, such as isotopic records
of oxygen, evolved over time depending on the orbital
cycles of the obliquity, eccentricity and precession. These
cycles, initially calculated by Milutin Milankovitch, were
established much more precisely by André Berger, and gave
paleoclimatologists precise solutions for the last three million years. Hays et al. have “adjusted” their initial age model
to “match” the obliquity cycle in the record to that given by
astronomical calculation. In doing so, they established the
first ‘cyclostratigraphy’. This method was widely used in the
1980s to constrain the ages of isotopic records of oxygen
during the Brunhes period with a precision of a few thousand
years.
Subsequently, the development of a hydraulic corer in the
DSDP program resulted in the acquisition of previously
inaccessible, very deep sediments, allowing the study of
continuous sequences in even older sediments. Applied to
the Matuyama period, astronomical calibration based on
obliquity cycles did not initially show significant age differences between the limits of “astronomical” polarity and
those of the Mankinen and Dalrymple scale. But when an
astro-timeline, based on precession, was obtained at
Site ODP 677, it became clear that the polarity intervals of
the Mankinen and Dalrymple scale were not accurately
dated.
This study paved the way for a complete revision of the
ages of the polarity intervals, especially during the Gauss and
Gilbert periods (Hilgen 1991a, b). The realization of the
“youth” of the ages obtained by the K/Ar method for the
Plio-Pleistocene era compared to those obtained by
cyclostratigraphy has led to the extensive use of the
40 Ar/
39 Ar
method (Chap. 5), which had recently been developed to test
the validity of cyclostratigraphic ages of this period. The age
of the Brunhes/Matuyama reversal (initially set at 0.73 Ma
by K/Ar) has been re-evaluated to 0.78 Ma due to a large
number of independent
40 Ar/
39 Ar measurements. The latter
value is consistent with the cyclostratigraphic age. Good
agreement is also observed for the Plio-Pleistocene era,
where the new
40 Ar/
39 Ar estimations coincide with the ages
given by Shackleton et al. (1990), and Hilgen (1991a, b).
Although these new estimations, in general, confirmed
the astronomical ages, the ages of geomagnetic reversals
given by Cande and Kent (1992), based on magnetic
anomalies at sea, did not seem to be in agreement with
astronomical dating. The authors had in fact adopted durations for the Plio-Pleistocene polarity intervals that have
proven inaccurate. As mentioned above, it is precisely the
astronomical ages for the Plio-Pleistocene that Cande and
Kent (1995) adopted in their new version of the scale. The
consistency they then obtained between
40 Ar/
39 Ar estimates,
magnetic anomalies at sea and astronomical calculations
over the entire time period covered by the new scale can be
considered as a validation of this method. With this method,
the age of a geomagnetic reversal can be estimated within
the duration of one precession cycle, in other words, less
than two or three times the length of a polarity reversal. This
very high resolution led Renne et al. (1994) to calibrate the
age of one of the standards of the
40 Ar/
39 Ar method (Fish
Canyon Tuff, FCTs) with the astronomical method, which
reduced the uncertainty to 0.6% for the calibrated ages
compared with the standard.
Principle and Practice of Magnetostratigraphy
The magnetic polarities scale shows that polarity reversals
are largely random in time, which means that sequences of
four or five successive reversals do not repeat themselves
identically in time; therefore, they constitute a kind of
“fingerprint” of specific geological periods. Magnetostratigraphy is based on this characteristic: if a sequence
characteristic of the magnetostratigraphic scale can be
identified within a particular series, then a specific age can
be assigned to this section.
106
C. Laj et al.