tmp82AF TMP
tmp82AF TMP
tmp82AF TMP
Terra Antartica
2008-2009, 15(1), 193-210
Abstract We conducted initial palaeomagnetic studies on cores from site AND-2A (7745.488S,
16516.605E, ~383.57 metres water depth). A total of 813 samples were collected that span from the
top of the section down to the base at 1138.54 metres below sea floor (mbsf). Samples were collected
every one or two metres down the core, with paired (pilot) samples being collected about every ten to
twenty metres to allow us to assess the demagnetisation behaviour of the samples using either alternating
field (AF) or thermal demagnetisation. With the exception of only a few intervals, AF demagnetisation
was observed to resolve a characteristic remanent magnetisation (ChRM) as well or better than thermal
demagnetisation. Thermal demagnetisation was particularly ineffective in many intervals owing to thermal
alteration that was common above 500C and was evident in some samples even at low temperatures.
Above Lithostratigraphic Unit (LSU) 8 (436.18 mbsf), where lithologies are generally more coarse grained
than lower in the section, resolving a ChRM is difficult and recent overprints or a drilling overprint are a
concern. Within LSU 8 and below, most samples have a ChRM that can be resolved. The ChRM is most
likely an original depositional magnetisation throughout most of this lower section, although orthogonal
demagnetisation diagrams contain evidence that normal polarity overprinting affects some intervals. Based on
40
Ar/39Ar dates and diatom datums, the magnetozones identified from the base of the hole up to ~266mbsf
are consistent with spanning from either Chron C6n (18.748-19.772 Ma) or C6An.1n (20.040-20.213 Ma)
up through Chron C5Br (15.160-15.974 Ma). Above this, intervals of constant polarity are isolated within
longer stratigraphic intervals of uncertain polarity, making their correlation with the geomagnetic polarity
timescale (GPTS) speculative and highly dependent on ages obtained from other dating methods. One
exception is a reversed-to-normal polarity transition that occurs at ~31 mbsf and is interpreted to most
likely be the Brunhes/Matuyama boundary. The spacing of polarity reversals below 266 mbsf and their
correlation with the GPTS indicates that this part of the stratigraphic section was deposited between 15
to 20 Ma at a mean sedimentation rate of about 18 centimetres (cm)/ thousand year (k.y.).
Introduction
The Southern McMurdo Sound (SMS) Project of the
ANDRILL Programme cored site AND-2A (7745.488S,
16516.605E, ~383.57 m water depth) to a total
depth of 1138.54 mbsf (Fig. 1). The thick sequence
of rocks recovered contains a geologic history of the
region, including records of climate change, tectonics,
and much more.
In this study, we examine the palaeomagnetic
record of the core with a primary focus on determining
a preliminary magnetostratigraphy, which can be
used to assist in dating the stratigraphic section.
We follow a sampling and measurement strategy
similar to that applied in the palaeomagnetic study of
cores from Site AND-1B from the McMurdo Ice Shelf
(MIS) Project of ANDRILL (Wilson et al., 2007), with
a goal of building a long continuous palaeomagnetic
record that spans from the early Miocene to present.
To accomplish this, we collected oriented mini-
G. Acton et al.
195
G. Acton et al.
197
Analysis
The characteristic remanent magnetisation
direction (ChRM) was estimated from the data using
principal component analysis (PCA) (Kirschvink,
1980) and Fisher statistics (Fisher, 1953). For the
PCA direction, we find the best-fit line that passes
through the vector demagnetisation data in two
ways. First, we use the FREE option of PCA in which
the line is fit through the data free of the constraint
that the line passes through the origin of orthogonal
demagnetisation diagrams. Second, we use the
ANCHOR option of PCA in which the line is fit through
the data but is anchored to the origin of orthogonal
demagnetisation diagrams (Fig. 4).
In both cases, we use an iterative search program
to find and delete data from demagnetisation steps
that are outliers, where an outlier is defined as a
datum that degrades the fit of the line relative to all
other demagnetisation data used. We require that
data from at least five steps are used to find the best
estimates of the FREE and ANCHORED PCA direction.
To avoid contamination by drilling overprints, we do
not use NRM data from demagnetisation steps <25mT
or <250C in the PCA. We also never use data from
demagnetization steps >80mT or >650C because
the NRM directions typically become erratic owing
to spurious anhysteretic remanent magnetisations
(ARM) imparted during AF demagnetisation or to
Rock Magnetism
Currently, rock magnetic observations are limited
to susceptibility measurements and NRM. Because
drilling imparts a strong magnetic overprint to the
core, the NRM prior to demagnetisation provides
information somewhat similar to a low-field isothermal
remanent magnetisation (IRM). Hence, both the
susceptibility and the NRM prior to demagnetisation
provide measures of the magnetic concentration in
the core, with their variability being similar along the
G. Acton et al.
Fig. 5 Variations in the volume susceptibility, the magnetisation intensity, the inclination of the NRM (prior to demagnetisation), and
the inclination of the characteristic remanent magnetisation (ChRM) for the oriented palaeomagnetic samples. The susceptibility and
intensity data are smoothed with a 5-point moving average. The Lithostratigraphic Units (LSU) are given on the far right with dashed
horizontal lines at LSU boundaries.
199
Fig. 6 (A) Variation of volume susceptibility from the palaeomagnetic samples compared with that from whole-core measurements.
(B) Comparison of the low-frequency susceptibility with the high-frequency susceptibility for the palaeomagnetic samples. The lowfrequency values are generally slightly higher than the high-frequency. This difference is used in computing the frequency dependence
(Xfd) of magnetic susceptibility (see text). (C) The Xfd of the palaeomagnetic samples compared with clasts taken from the core. (D)
Lithostratigraphic Units (LSU) with the LSU boundaries given by the dashed lines.
G. Acton et al.
from low (<5 mT) to very high (>100 mT) (Figs. 7-9).
It also appears to have a relatively low unblocking
temperature, but this is difficult to fully assess because
the thermal results give mostly erratic directions,
which may result from thermal alteration as nearly
all samples in this interval have a notably increase
201
G. Acton et al.
Fig. 10 Magnetic susceptibility variation with temperature for palaeomagnetic samples collected in different lithostratigraphic units
of AND-2A drillcore. The increase in susceptibility with temperature is indicative of alteration of preexisting magnetic minerals and/or
the formation of new magnetic minerals.
203
G. Acton et al.
205
Fig. 15 Representative orthogonal demagnetisation diagrams for samples from LSUs 11-14. Dark squares give the vertical component
(inclination).
206
207
Fig. 18 Characteristic remanent magnetisation (ChRM) inclinations, magnetozones, and magnetostratigraphic interpretation for Core
AND-2A. The different quality of ChRM inclinations are plotted with different symbols, where the highest quality (quality 1) are circles,
quality 2 are grey squares, quality 3 are triangles, and quality 4 are Xs (see Supplementary SMS 10 Table A3 for explanation of quality
factors). The magnetozones are black=normal polarity, white=reversed polarity, and grey=uncertain polarity. The magnetostratigraphic
interpretation uses the chron nomenclature of Cande and Kent (1995), where for example a geomagnetic excursion within Chron C5Dn
would be named C5Dn-1r.
208
Bottom
Depth
(mbsf)
Middle
Depth
(mbsf)
Depth
Range
(m)
Reversal
Type
Age (Ma)
22.28
22.28
22.28
0.00
U/N
<1.945
25.34
36.86
31.10
11.52
N/R
0.781-1.945
83.20
83.92
83.56
0.72
R/U
>0.781
Base of magnetozone R1.1. Boundary between the bottom of a reversed polarity magnetozone (part or all of
which is probably within the Matuyama) and the top of an
uncertain polarity zone.
96.78
96.81
96.80
0.03
U/R
114.37
117.05
115.71
2.68
R/U
263.47
267.62
265.55
4.15
U/R
<15.160
277.07
279.88
278.48
2.81
R/U
>15.974
328.17
328.87
328.52
0.70
U/N
<16.268
351.97
352.32
352.14
0.35
N/U
<16.268
388.50
388.50
388.50
0.00
U/N
<16.268
412.00
414.52
413.26
2.52
N/R
16.268
427.33
431.78
429.55
4.45
R/U
16.268-16.543
438.50
439.96
439.23
1.46
U/R
16.268-16.543
446.95
451.37
449.16
4.42
R/N
16.543
460.07
461.16
460.62
1.09
N/U
16.543-16.721
467.42
475.63
471.52
8.21
U/N
<16.721
486.68
489.74
488.21
3.06
N/R
16.721
502.07
506.07
504.07
4.00
R/U
16.721-17.235
209
512.42
517.39
514.90
4.97
U/R
16.721-17.235
532.70
537.72
535.21
5.02
R/U
16.721-17.235
547.69
552.01
549.85
4.32
U/R
16.721-17.235
579.05
583.63
581.34
4.58
R/N
17.235
640.46
645.57
643.02
5.11
N/R
17.235-17.533
646.08
646.68
646.38
0.60
R/N
17.235-17.533
706.80
708.16
707.48
1.36
N/R
17.235-17.533
Base of magnetozone N5.1 that overlies a possible excursion in Chron C5Dn, referred to as C5Dn-2r(o).
708.16
708.90
708.53
0.74
R/N
17.235-17.533
Top of magnetozone N5.2 that underlies a possible excursion in Chron C5Dn, referred to as C5Dn-2r(o).
723.36
724.10
723.73
0.74
N/R
17.533
748.25
750.55
749.40
2.30
R/N
17.717
757.08
761.42
759.25
4.34
N/R
17.740
783.16
784.22
783.69
1.06
R/N
18.056
957.93
959.25
958.59
1.32
N/R
18.524
978.92
992.36
985.64
13.44
R/N
18.748
<19.772
645.57
643.02
5.11
N/R
17.533
646.08
646.68
646.38
0.60
R/N
18.056
706.80
708.16
707.48
1.36
N/U
18.056-18.524
Base of magnetozone N5.1 that overlies a possible excursion in Chron C5En, referred to as C5En-1r(y)
708.16
708.90
708.53
0.74
U/N
18.056-18.524
Top of magnetozone N5.2 that underlies a possible excursion in Chron C5En, referred to as C5En-1r(o
723.36
724.10
723.73
0.74
N/R
18.524
748.25
750.55
749.40
2.30
R/N
18.524-18.748
757.08
761.42
759.25
4.34
N/R
18.524-18.748
783.16
784.22
783.69
1.06
R/N
18.748
957.93
959.25
958.59
1.32
N/R
19.772
978.92
992.36
985.64
13.44
R/N
20.040
<20.213
Reversal types are N=Normal Polarity; R=Reversed Polarity; U=Uncertain Polarity. N/R = normal polarity interval on top of a reversed
polarity interval; R/N = reversed polarity interval on top of a normal polarity interval; the magnetozones are labelled and plotted
versus depth in figure 18 of this paper.
G. Acton et al.
Supplementary Information
The following supplementary information in tables for this contribution is available on-line at the Terra Antartica
website www.mna.it/english/Publications/TAP/terranta.html and at the ANDRILL data site www.andrill.org/data.
Appendix 1
Supplementary SMS 10 Table A1 - Palaeomagnetic and rock magnetic results for samples collected from the
AND-2A core.
Supplementary SMS 10 Table A2 - Rock magnetic measurements made on selected volcanic, basement, and
sedimentary clasts from AND-2A core.
Supplementary SMS 10 Table A3 - Principal component analysis and preferred inclination for palaeomagnetic
samples from AND-2A core.