Geology Reference
In-Depth Information
The paleomagnetism of igneous rocks is much
stronger than that of sedimentary rocks, so it is more
robust and withstands the effects of remagnetization
more easily than that of sedimentary rock paleomag-
netism. However, just averaging the magnetizations of
a pile of lava fl ows is no guarantee that enough time
has passed to adequately average the effects of geo-
magnetic secular variation. For instance, data from the
Hawaiian Volcano Observatory (Kauahikaua et al .
1998) show that the number of fl ows erupted in
Hawaii per thousand years over the past 12,000 years
varies from 1 to 11 in any given 1000 year period.
Based on the sequence of fl ows erupted over the past
12,000 years and assuming that about 3000 years is
needed to adequately average secular variation and
obtain the GAD fi eld, anywhere from 6 to 17 fl ows
should be measured for a paleomagnetic study that can
be reliably used to reconstruct paleolatitude or for
other tectonic applications. Since the volcanic history
in any particular region isn't known in detail, most
workers use the amplitude of the circular standard
deviation of virtual geomagnetic poles (VGPs) derived
from lava fl ows to estimate whether secular variation
has been adequately averaged. The behavior of the geo-
magnetic fi eld over the past 5 million years is the only
guide to the amount of secular variation, i.e. the ampli-
tude of the circular standard deviation, expected if a
sequence of igneous rocks has faithfully recorded
secular variation.
Although there can be unrecognized unconformities
and hiatuses in any sequence of sedimentary rocks,
collecting samples from a thick stratigraphic section
gives confi dence that enough time has been sampled to
average paleosecular variation. Knowing the average
sediment accumulation rate from magnetostratigra-
phy, rock magnetic cyclostratigraphy, fossils, or from
radiometric control can give assurance that secular
variation has been averaged; even without this infor-
mation, knowing the typical sedimentation rate for dif-
ferent lithologies can however guide sampling strategy
and data interpretation (Table 1.1).
There is another reason why the continuity of sedi-
mentary paleomagnetic records is important to paleo-
magnetists. Recent sediments (marine and lacustrine)
and high-fi delity records from ancient sedimentary
rocks allow the detailed observation of geomagnetic
fi eld behavior. A continuous record of Earth's mag-
netic fi eld behavior is critical for understanding the
generation of the geomagnetic fi eld and for providing
constraints on models of the geodynamo. The best con-
straints on geodynamo models come from records of
transitional fi eld behavior during polarity transitions
(Merrill et al . 1996). There is a rich array of data from
marine sediments showing the behavior of the fi eld
during the most recent polarity transition, the Brunhes-
Matayama, some 780,000 years ago. Clement (1991)
was the fi rst to show preferred longitudinal bands of
virtual geomagnetic pole paths during the Brunhes-
Matuyama polarity transition using the paleomagnet-
ism of marine sediments. The accuracy of this result
was questioned and then modifi ed by observations
from igneous rocks, but it was the continuity of the
sedimentary paleomagnetic record that was critical to
the Clement's initial observation. Recent work on sedi-
mentary records of secular variation of the geomag-
netic fi eld at high latitudes (Jovane et al . 2008 ) shows
fi eld behavior close to the so-called tangent cylinder
(latitudes >79.1°), a cylinder parallel to the Earth's
rotation axis that includes the inner core of the Earth.
These workers collected 682 samples from a 16 m long
marine core at 69.03°S in the Antarctic showing that
the dispersion of secular variation was high (about
30°) during the past 2 myr suggesting vigorous fl uid
motion in the outer core. This kind of record would be
nearly impossible to obtain from igneous rocks.
Finally, continuous records of geomagnetic fi eld
paleointensity variations from marine and lacustrine
sediments not only allow another constraint on geody-
namo models, but they can also provide a way to cor-
relate and date marine sediments globally. The best
example of this is Valet's work (Guyodo & Valet 1996,
1999; Valet et al . 2005) on constructing stacked rela-
Table 1.1 Typical sediment accumulation rates for sedimentary rocks
Sedimentary environment
of deposition
Average sediment
accumulation rate
Sampling thickness to
average secular variation
Deep marine
1 cm/1000 years
3cm
Near-shore marine
0.1-1 m/1000 years
0.3-3 m
Continental lacustrine
1 m/1000 years
3m
Search WWH ::




Custom Search