axial dipole. Thus reconstructed, the latitudinal distribution
of the paleoclimate indicators is statistically identical to their
present distribution, thus supporting the hypothesis of the
axial centered dipole. Other studies carried out in the 1960s
made it possible to validate this hypothesis, first for the
Plio-Pleistocene and then gradually as far back as the Precambrian. In 1968, Le Pichon proposed a model of movement of rigid land masses relative to each other, thereby
reconstructing seafloor spreading for all the Cenozoic (Le
Pichon 1968).
The trajectories of the APWP make it possible to place
the continents in their original position with respect to the
axis of rotation of the Earth, in other words, in terms of
paleolatitude and orientation. However, paleomagnetism is
not enough, since it does not specify the paleolongitude of
the continents because of the spherical symmetry of the
magnetic field (assuming an axial and centered dipole
magnetic field). Kinematic parameters of the oceans are used
to pin down the position of a continent relative to another.
However, this method can only be used when the ocean
separating two continents is bordered by passive margins
and especially when these kinematic parameters are known
(available for the last 170 million years only). If one of these
oceanic margins is active, the positioning of one plate relative to another is nevertheless possible by traveling across
one or more other intermediate continents separated by
passive margins. Nevertheless, there are cases where the
deformation of the continental land masses by tectonics (in a
collision zone, for example), the presence of active margins,
or the age of reconstruction prevent the use of this method
combining paleomagnetism and ocean kinematics. In these
cases, paleogeographic reconstructions depend on paleomagnetism and the APWP.
These trajectories of the apparent polar wander are not all
of equivalent quality. To overcome this disadvantage, Besse
and Courtillot (1991, 2002) and Torsvik et al. (2012) proposed calculating artificial trajectories of APWP where the
poles of all the plates of a certain age are integrated into a
single referential (i.e. transferred from one lithospheric plate
to another using the kinematic parameters of the oceans).
This method is applicable to the last 320 million years
(despite the absence of kinematic data for the oceans beyond
170 Ma since all the continents were then combined into a
supercontinent, the Pangea). Before 320 million years, for
all continents and regardless of the period for those land
masses surrounded by active margins, reconstructions rely
solely on the trajectories of APWP, although geological
and/or paleontology arguments may provide constraints on
the relative position of the masses. Of course, there is an
overall increase in uncertainties with age.
The now well-known movements of large lithospheric
masses are part of the classical theory of plate tectonics.
However, the whole Earth can also tilt relative to the axis of
rotation in response to the heterogeneities of masses in the
mantle, modifying the tensor inertia of our planet (the
maximum axis of inertia is aligned with the axis of rotation
of the Earth). In the paleomagnetic reference system, the
APWP is thus caused by the movement of the plate due to
plate tectonics and to the overall movement of the
continents.
In 1972, Morgan proposed using hot spots associated
with convective plumes from the D″ transition zone as a
definitive frame of reference. Deep mantle convective
plumes are assumed to be fixed because of the sufficiently
slow movements in the lower mantle. The movements of
some lithospheric plates can thus be positioned within the
“hot spots” reference frame. Unlike magnetic poles, paleolatitude and paleolongitude are constrained (by a frame of
reference associated with the mantle), but this method can
only be used for the last 130 million years. Before this time,
the traces of hot spots on the ocean floor cannot be traced,
erased by the subduction zones, while the few remaining
traces are not constraining enough.
The differences in latitude and in rotation of the continents
between the paleomagnetic and hotspot reference frames
make it possible to isolate the movement of the global drift of
the crust-mantle couple with respect to the axis of rotation or
the true polar wander from the perspective of a continent. The
amplitude of the global drift generally does not exceed 1° per
million years over the last 130 million years (Besse and
Courtillot 2002). Nevertheless, there are some rapid events
(occurring over a few million years) during which a global
drift of about 10° of all the continental masses has been
observed, for example in the Paleocene (Moreau et al. 2007).
Even with a lack of data in the hot spot reference, it is possible to isolate the true polar wander by determining common
drifts in APWP of different continents. Events of great
amplitude have also been suggested, for example a 90°
movement between the Lower and Middle Cambrian
(Kirschvink et al. 1997) or during the Ediacaran (Robert et al.
2017). These events result in a continental drift rate of the
order of 10° per Ma, therefore much higher than the maximum continental drift rate caused by plate tectonics of about
2° per Ma (Seton et al. 2012) but perfectly compatible with
the theoretical maximum speed of the true polar wander of
about 10° per Ma, taking into account a viscoelastic Earth
(Greff-Lefftz and Besse 2014; Robert et al. 2017).
The Paleomagnetic Tool, Tests
and Uncertainties
To understand the difficulty of obtaining robust paleogeographic reconstructions, it is important to look at the paleomagnetic tool used (Chap. 7). The direction of the magnetic
field is measured in a laboratory using a magnetometer, an
2 The Changing Face of the Earth Throughout the Ages
27
of the paleoclimate indicators is statistically identical to their
present distribution, thus supporting the hypothesis of the
axial centered dipole. Other studies carried out in the 1960s
made it possible to validate this hypothesis, first for the
Plio-Pleistocene and then gradually as far back as the Precambrian. In 1968, Le Pichon proposed a model of movement of rigid land masses relative to each other, thereby
reconstructing seafloor spreading for all the Cenozoic (Le
Pichon 1968).
The trajectories of the APWP make it possible to place
the continents in their original position with respect to the
axis of rotation of the Earth, in other words, in terms of
paleolatitude and orientation. However, paleomagnetism is
not enough, since it does not specify the paleolongitude of
the continents because of the spherical symmetry of the
magnetic field (assuming an axial and centered dipole
magnetic field). Kinematic parameters of the oceans are used
to pin down the position of a continent relative to another.
However, this method can only be used when the ocean
separating two continents is bordered by passive margins
and especially when these kinematic parameters are known
(available for the last 170 million years only). If one of these
oceanic margins is active, the positioning of one plate relative to another is nevertheless possible by traveling across
one or more other intermediate continents separated by
passive margins. Nevertheless, there are cases where the
deformation of the continental land masses by tectonics (in a
collision zone, for example), the presence of active margins,
or the age of reconstruction prevent the use of this method
combining paleomagnetism and ocean kinematics. In these
cases, paleogeographic reconstructions depend on paleomagnetism and the APWP.
These trajectories of the apparent polar wander are not all
of equivalent quality. To overcome this disadvantage, Besse
and Courtillot (1991, 2002) and Torsvik et al. (2012) proposed calculating artificial trajectories of APWP where the
poles of all the plates of a certain age are integrated into a
single referential (i.e. transferred from one lithospheric plate
to another using the kinematic parameters of the oceans).
This method is applicable to the last 320 million years
(despite the absence of kinematic data for the oceans beyond
170 Ma since all the continents were then combined into a
supercontinent, the Pangea). Before 320 million years, for
all continents and regardless of the period for those land
masses surrounded by active margins, reconstructions rely
solely on the trajectories of APWP, although geological
and/or paleontology arguments may provide constraints on
the relative position of the masses. Of course, there is an
overall increase in uncertainties with age.
The now well-known movements of large lithospheric
masses are part of the classical theory of plate tectonics.
However, the whole Earth can also tilt relative to the axis of
rotation in response to the heterogeneities of masses in the
mantle, modifying the tensor inertia of our planet (the
maximum axis of inertia is aligned with the axis of rotation
of the Earth). In the paleomagnetic reference system, the
APWP is thus caused by the movement of the plate due to
plate tectonics and to the overall movement of the
continents.
In 1972, Morgan proposed using hot spots associated
with convective plumes from the D″ transition zone as a
definitive frame of reference. Deep mantle convective
plumes are assumed to be fixed because of the sufficiently
slow movements in the lower mantle. The movements of
some lithospheric plates can thus be positioned within the
“hot spots” reference frame. Unlike magnetic poles, paleolatitude and paleolongitude are constrained (by a frame of
reference associated with the mantle), but this method can
only be used for the last 130 million years. Before this time,
the traces of hot spots on the ocean floor cannot be traced,
erased by the subduction zones, while the few remaining
traces are not constraining enough.
The differences in latitude and in rotation of the continents
between the paleomagnetic and hotspot reference frames
make it possible to isolate the movement of the global drift of
the crust-mantle couple with respect to the axis of rotation or
the true polar wander from the perspective of a continent. The
amplitude of the global drift generally does not exceed 1° per
million years over the last 130 million years (Besse and
Courtillot 2002). Nevertheless, there are some rapid events
(occurring over a few million years) during which a global
drift of about 10° of all the continental masses has been
observed, for example in the Paleocene (Moreau et al. 2007).
Even with a lack of data in the hot spot reference, it is possible to isolate the true polar wander by determining common
drifts in APWP of different continents. Events of great
amplitude have also been suggested, for example a 90°
movement between the Lower and Middle Cambrian
(Kirschvink et al. 1997) or during the Ediacaran (Robert et al.
2017). These events result in a continental drift rate of the
order of 10° per Ma, therefore much higher than the maximum continental drift rate caused by plate tectonics of about
2° per Ma (Seton et al. 2012) but perfectly compatible with
the theoretical maximum speed of the true polar wander of
about 10° per Ma, taking into account a viscoelastic Earth
(Greff-Lefftz and Besse 2014; Robert et al. 2017).
The Paleomagnetic Tool, Tests
and Uncertainties
To understand the difficulty of obtaining robust paleogeographic reconstructions, it is important to look at the paleomagnetic tool used (Chap. 7). The direction of the magnetic
field is measured in a laboratory using a magnetometer, an
2 The Changing Face of the Earth Throughout the Ages
27