The Magnetization of Sedimentary Rocks: Processes and their Interpretation - Paleomagnetism of Sedimentary Rocks: Process and Interpretation - page 125

tively quiet depositional environments and relatively
continuous sedimentation, so that a complete record
of the geomagnetic fi eld is probable. Gray-colored
mudrocks would usually suggest enough magnetite for
a strong paleomagnetic signal; their fi ne grain size
would increase the probability that it is well aligned
with the ancient geomagnetic fi eld.
Red sedimentary rocks, red shales, mudstones, silt-
stones or sandstones are always good targets for paleo-
magnetic studies because the red color indicates that
hematite, a very stable magnetic mineral because of its
high coercivity, is more than likely present in the rock.
Of course, as indicated earlier in the topic, the age of a
red bed's magnetization may not be the same as its
depositional age, but this can be investigated using dif-
ferent laboratory and fi eld tests. The magnetization will
most likely be stable and ancient.
Pure quartz sandstones (white in color) will proba-
bly have weak paleomagnetism and be more prone to
secondary magnetic overprinting. This is also true for
pure limestones, mainly because of the dearth of depo-
sitional magnetite in these rocks. Organic-rich lake and
marine sediments and sedimentary rocks are likely
affected by reduction diagenesis. The original deposi-
tional magnetite in these rocks will have been (at least
partially and maybe completely) dissolved and replaced
by secondary iron sulfi des. However, if the sediment
accumulation rate was high enough, the depositional
magnetite may have passed through the reduction dia-
genesis zone fast enough such that little alteration
occurred. These types of rocks should therefore not be
avoided, but simply tested for the predominance of iron
sulfi de magnetizations.
Orange-brown weathering stains on the outcrop
indicate present-day chemical weathering of the rock's
iron minerals that has probably either altered the depo-
sitional Fe oxides or produced secondary magnetic
minerals from the oxidation of iron silicates. Similarly,
a sickly brown-yellow color for a fi ne - grained sand-
stone does not bode well for good paleomagnetic
results. The color of the rock hints that secondary iron
oxide hydroxides (e.g. limonite, goethite) have over-
printed a weak primary magnetization.
Intensely deformed rocks, for instance sedimentary
rocks with a well-developed spaced cleavage, will prob-
ably not yield pristine primary paleomagnetic results.
They are likely to have secondary magnetizations formed
during deformation or primary magnetizations that
have been defl ected by grain-scale strain. Metamorphic
rocks of greenschist grade and higher were almost cer-

tainly reset paleomagnetically, either thermally or by
the growth of secondary magnetic minerals from
metamorphic fl uids. Even lower-grade rocks such as
slates probably had their primary magnetizations
affected by deformation.
The best sampling strategy for a study depends on
the objective of the study. Sampling schemes are as
varied as the types of observations being made. For
good paleomagnetic poles, either for global or regional
tectonic studies, paleomagnetic sites should be distrib-
uted throughout a geologic formation to make sure
changes in the geomagnetic fi eld due to paleosecular
variation are adequately averaged. Paleosecular varia-
tion is averaged over periods of 10 3 - 10 4 years, so even
most sites in a sedimentary rock will average secular
variation adequately. Recent plate motions suggest
that sampling a unit that was deposited over periods of
longer than 5 million years could include long-term
directional changes due to plate motion. The unit
should therefore be sampled over thicknesses that
record shorter time periods than 5 or more million
years. Although the defi nition of a paleomagnetic site
is usually given as sampling the geomagnetic fi eld at
a geologic instant of time (e.g. Butler 1992), for
paleomagnetic poles paleomagnetists typically collect
about 8-10 individually oriented samples over a strati-
graphic thickness of about a meter or less. This is done
to ensure the best results as any one stratigraphic layer
may not be a perfect paleomagnetic recorder; sampling
over different strata at a site increases the chances of
getting good results.
For magnetostratigraphic studies designed to catch
the reversals of the geomagnetic fi eld, sites are taken
from as close to one stratum as possible so a geologic
instant of time is captured. Many sites are however
collected throughout the formation to see how the geo-
magnetic fi eld is changing through time. Obviously,
continuous sampling of the strata would be perfect but
is not logistically possible. In order to properly identify
a polarity interval, it is optimal to have at least three
sequential sites of the same polarity. The stratigraphic
sampling interval therefore depends on the expected
reversal rate of the geomagnetic fi eld and an estimate
of the sediment accumulation rate for the rocks. The
geomagnetic fi eld's reversal rate is not stationary but
has changed through geologic time, so a rough esti-
mate of the time period sampled is very helpful to
design a good sampling strategy. For instance, for the
Mid - Tertiary the fi eld reversed on average 4-5 times
per million years. To get three sequential sites during a