Magnetization of Sediments and the Environment - Paleomagnetism of Sedimentary Rocks: Process and Interpretation - page 115

Given all these problems, it might be tempting to
simply throw up hands in despair and avoid the whole
topic. The best approach is to assume that patterns in
Nature probably mean something, and should be
studied. Since most systems in Nature are only incom-
pletely understood, however, always remember that
the observer could sometimes be fooled.
If we however accept that the evidence is reasonably
strong that sedimentary facies respond in some way to
changes in Earth's climate and that Milankovitch
forcing of global climate occurs, it is not a great leap
to think that environmental magnetic parameters
will somehow record orbitally forced climate cycles.
Because the timescale of the sedimentary record is best
suited to study millennial-scale cyclicity, rock magnetic
cyclostratigraphy should be perfect for the detection of
Milankovitch - scale cyclicity.
In our treatment of inclination shallowing using
magnetic anisotropy, we avoided the use of magnetic
susceptibility because it has many sources (diamag-
netic, paramagnetic and ferromagnetic minerals) and
is therefore a complicated signal to untangle. We follow
the same philosophy with rock magnetic cyclostratig-
raphy, although susceptibility records of Milankovitch
cycles have been reported in the literature - particu-
larly for deep-sea marine sediments - which have a
fairly straightforward depositional history (Hays et al.
1976 ; Bloemendal & de Menocal 1989 ; Mayer & Appel
1999). Remanence, particularly the remanence of a
specifi c magnetic mineral and a specifi c magnetic grain
size, can be used to construct a simpler measure of
magnetic particle concentration, the primary param-
eter used in rock magnetic cyclostratigraphic studies.
ARM data series for low-coercivity depositional mag-
netite or IRM data series for high-coercivity minerals
such as hematite can be used to detect orbitally forced
climate variations.
Other environmental magnetic parameters can be
used to detect more complicated signals. For instance,
the S -ratio can be used to quantify the changes in rela-
tive amounts of magnetite and hematite, or the ARM/
SIRM or ARM/χ ratios can detect changes in magnetic
grain size. More complicated parameters can be con-
structed on a case-by-case basis to measure other
climate-sensitive proxies, such as the goethite to hema-
tite ratio (see Fig. 8.1). While the response of sedimen-
tary facies to cyclic climate changes can be diffi cult to
measure or interpret, for instance subtle changes in
sediment grain size due to either runoff variations or

small changes in relative sea level, magnetic measure-
ments can be very sensitive to these subtle changes.
Moreover, magnetic measurements are relatively quick
to make when compared to non-magnetic measure-
ments (e.g. chemical, isotopic) and non-destructive.
One magnetic measurement is in effect counting thou-
sands of magnetic particles in a sample, which is one
of the reasons for the sensitivity of a magnetic meas-
urement. The short measurement time for one sample
( c. 1-2 minutes) allows the collection of hundreds-
thousands of measurements from one stratigraphic
section.
Often the cyclic signals are seen in the data series in
rock magnetic measurements, but time series analysis
is needed to quantify the periodicities and sophisticated
noise analysis is used to determine the statistical sig-
nifi cance of the frequency peaks in the power spec-
trum. In a typical rock magnetic cyclostratigraphic
study, a relatively thick stratigraphic section (tens to
hundreds of meters) is sampled at very close intervals
(sub-meter scale) and a rock magnetic parameter
(ARM, SIRM, IRM) is measured for the hundreds-
thousands of samples collected. Ideally, some inde-
pendent measure of time needs to be established;
standard magnetostratigraphy is one of the best tech-
niques to do this. An independent time measure is criti-
cal so that Milankovitch periodicities (20 kyr, 40 kyr,
100 kyr, 400 kyr) can be unequivocally identifi ed.
Ratios of periodicities (1 : 2 : 5 : 20, particularly the 5 : 1
eccentricity to precession ratio) can be used to tenta-
tively identify Milankovitch cycles, but this is not
always foolproof (see the section on 'Latemar contro-
versy ' below).
One of the diffi culties of rock magnetic cyclostratig-
raphy is the determination of the rock magnetic sam-
pling interval. Time series analysis theory dictates that
the highest frequency detectable in a data (time) series
has a period of twice the sampling interval. This is
called the Nyquist frequency and means that, if you
want to detect precession in the stratigraphic section,
you need to sample at least once every 10,000 years.
For good resolution of precession it would be better to
sample every 5000 years. For this, of course, you need
a rough idea of the sediment accumulation rate for the
section being studied. This could come from a previous
magnetostratigraphic study or from a detailed bio-
chronology for the section. At the worst sediment
accumulation rates can be estimated from depositional
environments and lithologies, but a preliminary pilot