Geomagnetic Events and Relative Paleointensity Records
639
boundary was identified at 2.7 m sub-bottom, where
pore waters contain minute iron concentrations
(Enneking et al. 1998). Magnetic grain-size sensitive parameters derived from high-resolution
hysteresis measurements (2 cm intervals, Fig. 2)
revealed on average slightly coarser magnetic
minerals below around 3.5 m, possibly reflecting
some minor diagenetic corrosion. The mean ratio
of specific saturation remanence (G rs ) to specific
saturation magnetization (G s ) decreases from 0.25
above to 0.21 below this horizon (Day et al. 1977;
see also Frederichs et al. this volume), the median
field of remanent hysteretic magnetization (Brh)
drops from 48 to 42.10- 3 T (von Dobeneck 1996).
Both the magnetite concentration and contribution
of the sedimentary matrix to the susceptibility
signal remain close to constant overall as documented by the specific ferrimagnetic (Xfe) and
non-ferrimagnetic (Xnf) susceptibilities. In clear
contrast to the magnetic grain-size indices, their
variations consistently show a distinct correlation
to the climatic 8 18 0 record. The same is true for
the remanence carrying magnetic mineral component represented by the saturation remanence.
There is thus no definite evidence for a chemically
or biologically mediated neoformation of magnetic
minerals at depth. A limited influx of reactive
organic matter obviously sustained only moderate
diagenesis in the late Quaternary Ceara Rise sediments which did not severely affect the magnetic
mineral inventory.
For the two shallowest sites of both transects
(GeoB 1515 and 1523) stable oxygen isotope stratigraphies were determined on planktic foraminiferal
species Globigerinoides sacculifer (Mulitza
1994; Rilhlemann et al. 1997). Following Prell et
al. (1986), the authors developed an age scheme
by graphic correlation of characteristic isotopic
events in their 8 18 0 records to the SPECMAP
stack ofImbrie et al. (1984). Based on this initial
model more detailed chronostratigraphies, used as
time framework in this study, were defined for all
cores by orbital tuning and sub-Milankovitch correlations of high-resolution susceptibility logs (von
Dobeneck and Schmieder this volume). Fluctuations in the amount of terrigenous material (-30 to
> 90%) supplied by the Amazon River and also in
magnetic susceptibility very clearly correspond to
the precession frequency band. They are superimposed on predominantly obliquity controlled variable bulk sedimentation rates ranging from about
I to 4 cmlkyr (Fig. 6).
Paleomagnetic Analyses
For a paleomagnetic study the sediments have been
sub-sampled at on average 5 cm intervals. Natural remanent magnetizations were measured on a
three-axis SQUID magnetometer (Cryogenics
Consultants, GM 400). To isolate the stable characteristic remanent magnetization (ChRM), each
individual sample was subjected to a systematic
stepwise alternating field (AF) demagnetization
using a standard procedure of 10,20,35,50 and
70 mT. Different graphic representations of the
data sets were employed to evaluate the demagnetization results, including NRM normalized intensity variations, stereographic projections of the
resultant and difference remanent directions (Hoffman and Day 1978) and orthogonal component
diagrams (Zijderveld 1967). After an essentially
single-component remanence had been identified
on these grounds, several successive demagnetization steps were averaged to give the stable ChRM
directions. Relative ChRM declinations refer to the
common split surface of the working core halves
as no absolute orientation in azimuth is available.
Because of the equatorial latitude of the coring
sites (geocentric axial dipole inclinations < ± 10°)
the polarities were exclusively derived from absolute ChRM declinations under the following
assumptions:
- Only short geomagnetic events have interrupted
the predominantly normal polarity ofthe Brunhes Chron. The large majority of samples from
each core thus carries a normal polarity.
- The average relative declination of these samples, i.e., excluding all declinations deviating by
more than 90° from the mean, coincides with
true north (0°) and is equivalent to the absolute
azimuth of the core.
An anticlockwise rotation of the cores around
their mean relative declination consistently results
in absolute ChRM declinations of about 0° for the
uppermost (youngest) sediments in agreement with
the fact that no Holocene geomagnetic event has
639
boundary was identified at 2.7 m sub-bottom, where
pore waters contain minute iron concentrations
(Enneking et al. 1998). Magnetic grain-size sensitive parameters derived from high-resolution
hysteresis measurements (2 cm intervals, Fig. 2)
revealed on average slightly coarser magnetic
minerals below around 3.5 m, possibly reflecting
some minor diagenetic corrosion. The mean ratio
of specific saturation remanence (G rs ) to specific
saturation magnetization (G s ) decreases from 0.25
above to 0.21 below this horizon (Day et al. 1977;
see also Frederichs et al. this volume), the median
field of remanent hysteretic magnetization (Brh)
drops from 48 to 42.10- 3 T (von Dobeneck 1996).
Both the magnetite concentration and contribution
of the sedimentary matrix to the susceptibility
signal remain close to constant overall as documented by the specific ferrimagnetic (Xfe) and
non-ferrimagnetic (Xnf) susceptibilities. In clear
contrast to the magnetic grain-size indices, their
variations consistently show a distinct correlation
to the climatic 8 18 0 record. The same is true for
the remanence carrying magnetic mineral component represented by the saturation remanence.
There is thus no definite evidence for a chemically
or biologically mediated neoformation of magnetic
minerals at depth. A limited influx of reactive
organic matter obviously sustained only moderate
diagenesis in the late Quaternary Ceara Rise sediments which did not severely affect the magnetic
mineral inventory.
For the two shallowest sites of both transects
(GeoB 1515 and 1523) stable oxygen isotope stratigraphies were determined on planktic foraminiferal
species Globigerinoides sacculifer (Mulitza
1994; Rilhlemann et al. 1997). Following Prell et
al. (1986), the authors developed an age scheme
by graphic correlation of characteristic isotopic
events in their 8 18 0 records to the SPECMAP
stack ofImbrie et al. (1984). Based on this initial
model more detailed chronostratigraphies, used as
time framework in this study, were defined for all
cores by orbital tuning and sub-Milankovitch correlations of high-resolution susceptibility logs (von
Dobeneck and Schmieder this volume). Fluctuations in the amount of terrigenous material (-30 to
> 90%) supplied by the Amazon River and also in
magnetic susceptibility very clearly correspond to
the precession frequency band. They are superimposed on predominantly obliquity controlled variable bulk sedimentation rates ranging from about
I to 4 cmlkyr (Fig. 6).
Paleomagnetic Analyses
For a paleomagnetic study the sediments have been
sub-sampled at on average 5 cm intervals. Natural remanent magnetizations were measured on a
three-axis SQUID magnetometer (Cryogenics
Consultants, GM 400). To isolate the stable characteristic remanent magnetization (ChRM), each
individual sample was subjected to a systematic
stepwise alternating field (AF) demagnetization
using a standard procedure of 10,20,35,50 and
70 mT. Different graphic representations of the
data sets were employed to evaluate the demagnetization results, including NRM normalized intensity variations, stereographic projections of the
resultant and difference remanent directions (Hoffman and Day 1978) and orthogonal component
diagrams (Zijderveld 1967). After an essentially
single-component remanence had been identified
on these grounds, several successive demagnetization steps were averaged to give the stable ChRM
directions. Relative ChRM declinations refer to the
common split surface of the working core halves
as no absolute orientation in azimuth is available.
Because of the equatorial latitude of the coring
sites (geocentric axial dipole inclinations < ± 10°)
the polarities were exclusively derived from absolute ChRM declinations under the following
assumptions:
- Only short geomagnetic events have interrupted
the predominantly normal polarity ofthe Brunhes Chron. The large majority of samples from
each core thus carries a normal polarity.
- The average relative declination of these samples, i.e., excluding all declinations deviating by
more than 90° from the mean, coincides with
true north (0°) and is equivalent to the absolute
azimuth of the core.
An anticlockwise rotation of the cores around
their mean relative declination consistently results
in absolute ChRM declinations of about 0° for the
uppermost (youngest) sediments in agreement with
the fact that no Holocene geomagnetic event has