-
Tamaki, K., Suyehiro, K., Allan, J., McWilliams, M., et al.,
1992Proceedings of the Ocean Drilling Program, Scientific Results,
Vol. 127/128, Pt. 2
59. MAGNETIC PROPERTIES AND PALEOMAGNETISM OF VOLCANIC ROCKS
ANDINTERLAYERED SEDIMENTS FROM THE JAPAN SEA (ODP LEG 127)1
L. Vigliotti2
ABSTRACT
Measurements of natural remanent magnetization (NRM), initial
susceptibility (K), anisotropy of magnetic susceptibility,frequency
dependent susceptibility (Xfd), and viscous remanent magnetization
(VRM) are reported from volcanic rocks recoveredduring ODP Leg 127
in the Japan Sea. The results indicate a significant difference
between the basalts drilled in the Yamato Basin(Site 794 and 797)
and in the Japan Basin (Site 795). The Koenigsberger ratios (Q)
show very low values in the Yamato Basinattesting that the
remanence is not dominant over the induced magnetization. This
evidence could explain why no magneticanomaly pattern has been
recognized in this basin.
Experiments of VRM acquisition and decay show that both the
processes are multistage with the acquisition processproceeding
more rapidly and deviates more from a log (t) law than the
corresponding decay. The sediments interlayered with thebasalts in
the acoustic basement of the Yamato Basin show processes of
remagnetization related to the emplacement of the
dikes.Temperatures of heating between 200° and 250° C were
estimated from the different unblocking temperatures of the
twocomponents of magnetization.
INTRODUCTION
One of the aims of Ocean Drilling Program (ODP) Leg 127 was
todrill the basement of the Japan Sea. The acoustic basement of the
basinwas successfully recovered from three sites: 794, 795, 797.
This studyreports the magnetic properties and the paleomagnetic
measurements of100 basalt samples recovered at Hole 794C (Yamato
basin, 24 samples),Hole 795B (Japan basin, 19 samples), and Hole
797C (Yamato basin,57 samples). At Sites 794 and 797 the basalts
were interlayered withsediments, and paleomagnetic measurements
were also carried out on asmall collection of sedimentary
samples.
The magnetic properties of rocks are useful indicators of the
com-position, concentration, and alteration state of magnetic
minerals; suchstudies examine the sources of the remanence and its
petrogeneticsignificance. In this paper the magnetic properties of
the rock are used toinvestigate its relationships with modes and
times of emplacement, andits implications for the magnetic
anomalies in the Japan Sea. The mag-netic parameters included in
the measurements are: intensity and directionof natural remanent
magnetization (NRM), acquisition and decay ofviscous remanent
magnetization (VRM), initial susceptibility (K),frequency dependent
susceptibility (Xfd), and anisotropy of magneticsusceptibility
(AMS).
The basement drilled at Site 794 consists of a total of 33.5 m
ofdolerite of tholeiitic composition and 1.5 m of welded tuffs. The
32 mof igneous rocks are divided into six units, each representing
separateintrusive sills (Tamaki, Pisciotto, Allan, et al., 1990);
the tuffs divideUnit V from Unit VI. Most of the igneous rocks are
highly altered,and their ages are unknown.
At Site 795, about 80 m of basalt, basaltic andesites, and
basalticbreccias were drilled in the acoustic basement. The
recovered 17.6 m ofvolcanic rocks are moderately to highly altered
and vesicular; onthe basis of texture and mineralogy, they are
divided into three units ofunknown ages.
Basalt, alkali basalt, hawaiite, and dolerite interlayered with
sedi-ments represent the acoustic basement cored in Hole 797C. A
total of
1 Tamaki, K., Suyehiro, K., Allan, J., McWilliams, M., et al.,
1992. Proc. ODP, Sci.Results, 127/128, Pt. 2: College Station, TX
(Ocean Drilling Program).
2 Istituto di Geologia Marina-CNR, 40127 Bologna, Italy.
347 m of this basement was cored and 140 m of igneous rocks
andsediments was recovered. The lithology of the interlayered
sedimentsvaries with increasing depth, from sandstone to siltstone,
and claystone.The 80 m of the recovered igneous rocks basically
includes two broadand overlapping groups containing 21 separate
units on the basis ofcontact relations, intervening sedimentary
rocks, composition, and tex-ture (Tamaki, Pisciotto, Allan, et al.,
1990). The upper group consists ofUnits 1, 2, and 4 and lacks
chilled margins which mark the lower groupof units consisting of
Units 3 and 5-21. The upper group is considered tobe representative
of lava flows and is predominantly massive while group2 is highly
fractured and brecciated.
SAMPLES AND LABORATORY PROCEDURES
Sedimentary samples were collected from the working half of
thecore sections, cutting cubic samples (7 cm3) with a spatula.
More con-solidated sediments were drilled with a double-blade press
diamond saw.Cylindrical igneous samples (2.54 cm diameter) were
drilled with adiamond drill bit.
Paleomagnetic measurements were carried out in the
Paleomag-netic Lab of the Istituto di Geologia Marina of CNR of
Bologna(Italy). The anisotropy of magnetic susceptibility (AMS) was
meas-ured with a Jelinek Kappa bridge in the Paleomagnetic Lab of
theUniversità di Torino (Italy). The natural remanent magnetization
(NRM)of the sediments was measured with a Jelinek JR-4 spinner
magne-tometer, while for igneous rocks a Schonstedt spinner
magnetometerwas used. Few samples were studied on board the JOIDES
Resolutionusing a Molspin spinner magnetometer. The frequency
dependencemagnetic susceptibility (Xfd (%) = 100 × (KLf - KHf)/KLf)
wasmeasured with a Bartington Instrument magnetic
susceptibilitymeter (model M.S.2) which allows measurements in two
frequencies:0.47 Hz (Lf) and 4.7 Hz (Hf). The Koenigsberger ratio
(Q) wascalculated assuming a local field value of 45000 nT.
Acquisition and decay of viscous remanent magnetization
(VRM)were carried out on the original remanence of the rock. The
wholecollection of igneous rocks was arbitrarily oriented and in
the laboratoryunder the Earth Magnetic Field (about 45000 nT), and
the NRM meas-ured at intervals, spaced approximately
logarithmically, for 1376 hr. Onthe basis of the results of the VRM
acquisition, 20 samples wereselected and stored in a magnetic
shield with a low field of about 70 nT.The NRM was measured on an
interval of time lasting 7176 hr.
933
-
L. VIGLIOTTI
RESULTS
The magnetic properties for each specimen of igneous rock
arelisted in Tables 1, 2, and 3, for Site 794,795 and 797,
respectively, inorder of increasing depth in each hole. Figure 1
shows typical or-thogonal plots of samples demagnetized by thermal
and alternatingfield methods. Magnetic properties of some sediments
from Site 797are discussed in the paper of Torii et al. (this
volume). In this chapterthe distribution of the data in the three
sites will be presented.
Table 1. Magnetic properties of igneous rocks recovered at Site
794.
Table 3. Magnetic properties of igneous rocks recovered at Site
797.Symbols as in Table 1.
Depth(mbsf) K K,/K3 NRM
VRMNRM (%) Xfd
Depth(mbsf) K NRM
VRM/NRM% VRM K,/K,
542.72560.54565.67573.03582.29586.17587.04588.10592.16601.95602.51605.71606.98615.50623.44624.63627.37628.79634.98635.62635.87636.80640.43645.93Mean
35981417203721049320348204303032720388503996026650258033097032800324634982043360341073831135866505605484941560289103975838303
5470645589347030610408898981190228
166073680130508821970996325024901070
64863501756
8.609.6081.719.3
20660.639.570.415.1
194
7.2056.339.030.061.04.6031.511.44.4016.8
60.55.70
46.9
47062
481670630630351632180444
12041431293053890314370110180
392360395
1.331.272.192.23
1.012.930.861.181.141.162.121.100.931.050.671.330.48
0.650.771.320.801.26
1.0051.0121.0241.0311.0401.0451.0361.0221.0111.0181.0051.0121.0081.0311.0091.0151.0041.0171.0041.0061.0071.0081.0121.017
1.0051.0201.0381.0471.0451.0571.0531.0361.0131.0271.0071.0221.0121.0351.0121.0191.0061.0201.0061.0101.0101.0091.0161.0201.21
4.100.420.431.910.240.660.740.630.810.23
1.460.610.671.660.551.570.712.461.340.53
0.614.34
Symbols: K, magnetic susceptibility (× 10 S.I. Units). NRM,
intensity of natural remanentmagnetization (mA/m). VRM/NRM %, ratio
of the viscous remanent magnetization acquiredafter 1376 hr with
respect to the initial NRM. VRM, difference between VRM acquired
after1376 hr and the initial NRM; values are expressed in mA/m.
Xfd, frequency dependencesusceptibility (%). F, parameter of
emplacement mode (Ellwood 1975). K,/K3, ratio betweenthe maximun
and the minimum susceptibility. Q, Koenigsberger ratio.
Table 2. Magnetic properties of igneous rocks recovered at Site
795.Symbols as in Table 1.
Depth(mbsf)
684.33693.71702.94705.08712.88722.74724.08725.57726.90727.59732.05733.50741.82741.98742.32742.69751.78752.23753.07Mean
NRM
188129154
1710015900176008990
2040056405650210003720018900
9418
15000827
16700924011091
VRM
1701727230060070048011004103908001600900142
600200500650508
VRM/NRM%
90.4133.346.81.83.84.05.35.47.36.93.84.34.815.311.54.0
41.33.07.020
F
1.0071.0101.008
1.0121.0151.0251.0121.0141.0201.0121.0151.009
1.0011.0161.0011.0121.0141.012
K
23383231112344928870290933150428943321593419935909199572506324116930391
14576280
190703136022440
Q
0.20.50.216.114.915.28.417.24.54.328.640.321.32.71.3
28.063.623.88.015.7
Xfd
0.571.020.811.71.972.321.91
1.231.424.242.572.482.15
3.57
1.99
553.66561.08561.70562.37570.39572.49573.47590.39592.35594.95599.17609.33618.16620.02621.37627.77629.10637.53646.70648.74656.54657.67660.77669.61676.23678.73682.50682.99685.30704.46708.60713.07724.37725.70733.35733.68734.80743.30743.85745.07754.44758.69765.72769.20773.33774.37781.78785.37806.99852.92881.63883.70884.84885.45891.21892.76894.96Mean
1569023443248772801220817294702833436330194102620329746107792914931658338102535021749409803158923616272102421020420255221747810814174381606515225343903173028250
15411061076763009361075460101930805909608093740701737723087500730908123076260562604550340735432075017452480
83938190357502829043677
1.0261.0591.0311.0331.0091.0331.0081.0201.0281.0301.0051.0401.0311.0291.0321.0821.0071.0541.0321.0221.0611.1241.1941.0231.0671.0271.1171.0421.0071.0761.2351.1201.0171.0251.0031.0211.0191.0351.0261.0191.0101.0141.0461.0101.0201.0111.0241.0131.0361.0491.0231.0261.0361.0031.0251.0491.046
475
5284941070146042233614008627381280445885
89736603570779724553729524075320011016701180710
611274029801060686737103011201700146014401570210011001510018305840
15101270190573545
13556722760
1546
72.0
108.391.924.215.889.8
192.620.740.447.7
3.180.752.5
48.38.2
11.836.119.580.344.020.230.383.593.69.0
20.415.8
88.225.229.543.4
125.990.061.283.022.935.650.731.815.794.512.632.83.8
4.618.1
43160.658.2
5.450.049.323.754.9
0.661.410.880.610.591.350.930.642.191.221.160.571.000.682.021.81
1.630.760.932.360.822.261.051.220.982.861.160.54
3.08
2.120.670.640.750.700.861.871.210.860.641.340.99
0.930.550.590.740.740.460.500.420.420.501.280.771.09
0.82
0.580.481.401.350.400.251.960.890.673.230.410.76
0.965.012.370.670.830.55
6.970.800.310.282.601.661.27
0.522.640.530.470.240.210.370.300.570.410.420.610.740.345.610.612.08
0.900.850.120.310.28
0.400.550.731.14
1.0221.0441.0201.0261.0051.0251.0061.0161.0231.0231.0041.0261.0171.0271.0231.0731.0061.0431.0311.0191.0451.1001.1471.0191.0451.0161.0871.0391.0061.0641.1861.0971.0111.0801.0021.0161.0171.0291.0191.0121.0101.0091.0361.0081.0121.0101.0161.0101.0241.0301.0211.0261.0301.0021.0181.0431.036
Site 794
The NRM intensities (Jr), plotted downhole together with
themagnetic susceptibility and the Koenigsberger ratio (Q) (Fig.
2)range from 228 to 6350 mA/m without any trend, but with the
twohigher values at the top and bottom of the section. The mean
valueof 1756 mA/m is within the range of the oceanic basalts.
Themagnetic susceptibility (K) exhibited constant values around
amean value of 38303 × 10~6 S.I. units. With these results for Jr
and
934
-
127-794B-26R-CCi(15-17cm) 127-794C -4R-4, 28-30cm
127-794C-10R-1i64-66cm 127-794O12R-1,132 -134cm
NRM N
Ui W
3080 mA/m
500°
300°
NRM
127-795B-41R-1,93-95cm 127-795B-38R-3,77"79cm 127-797C-1OR-2
,116-118cm 127-797C-24R-1, 46-48cm
Figure 1. Examples of vector demagnetization diagrams from
igneous rocks. Square: horizontal plane. Dots: vertical plane.
Horizontal orientation is arbitrary.
-
L. VIGLIOTTI
K, the distribution of the Koenigsberger ratio resembles that of
theNRM and it is noteworthy that, although the mean value for Q is
1.2,most of the samples had a value < 1. All the samples showed
to bevery susceptible to acquisition of a VRM. The ratio
VRM/NRM,which is an expression of the potential seriousness of the
effect ofVRM on the sample, varies from 4.4 % to 206% with a mean
valueof 46.9% (Table 1). The values of the frequency dependent
suscepti-bility (Xfd) in most of the samples is more than 1% with a
meanvalue of 1.26%. The anisotropy of magnetic susceptibility (AMS)
isweak, with the ratio K,/K3, commonly a measure of the degree
ofanisotropy, variable from 1.005 to 1.057. In order to distinguish
theemplacement mode of basalts, Elwood (1975) defined a parameterF
(F2 = K!2/(K2 × K3)) which separates intrusives (F > 1.040)
fromextrusives (F < 1.040) with 80% confidence. Although all the
recov-ered igneous rocks are considered sills, only two dolerite
samplesfrom Unit 2 exhibited values of F > 1.040. A plot of the
parameter Fas function of depth is shown in Figure 3 A.
1,04 -
1,03 -
1,02-
1,01 -
1,00-
Å
/IB SITE 794
\ \
540 560 580 600 620
Depth (mbsf)
640 660
10000 •j
1000 -;
100 -.
10 i
1 -
,1 -
K
X/SITE
D
794
. . T
540 560 580 600 620 640 660
Depth (mbsf)
Figure 2. NRM intensity, magnetic susceptibility (K), and
Koenigsberger ratio(Q) plotted vs. depth. Jr is expressed in mA/m;
K in 1CT6 SI Units.
Alternating field (AF) and thermal demagnetization were car-ried
out on the basalts recovered from the basement of Site 794.The
median destructive field of the rock is below 10 mT as canbe
expected from their VRM. The paleomagnetic record showsdifferences
not only between the lithologic units, but even fromcore to core,
suggesting different units from a paleomagnetic pointof view. One
sample (127-794B-26R-CC, 15-17 cm) from the firstunit yielded a
clear direction of normal polarity (Fig. 1), con-trasting with the
negative inclinations evidenced by the over-lying black mudstone.
The basalts recovered within the upper11 cores from Hole 794C
(560.0-629.2 mbsf) mostly exhibiteda reversed polarity. This
polarity was completely overprinted by asecondary component of
normal polarity in the first six cores. Thisoverprint, very
probably the result of the present field on the VRMof the rock, was
successfully removed by AF demagnetization withpeak-field of 10-15
mT, or thermal treatment with temperaturesabove 300° (Fig. 1). From
Core 127-794C-7R through Core 127-794C-11R, only the primary
magnetization was observed. Twosamples deviated from this picture:
Sample 127-794C-3R-1,103-105 cm, and Sample 127-794C-8R-2, 38-40
cm. Both thesesamples exhibited a normal polarity, but while the
latter belongsto a small piece that could have been rotated in the
core, the polarityof the former is confirmed by the result obtained
from an adjacentsample demagnetized on the ship. The stronger NRM
intensityobserved in the samples could give credit to a possible
different
1,05
1,04 -
1,03 -
1,02 -
1,01 -
1,00680 700 720
Depth (mbsf)
740 760
1,25
1,00500 600 700
Depth (mbsf)
800 900
Figure 3. Parameter F as function of depth in the basalts
drilled in the JapanSea. Dotted line separates intrusives (F >
1.040) from extrusives (F < 1.040)emplacement (Ellwood
1975).
age for this sill. The inclination appears systematically
shallowerthan the expected value of 56° and in a few cases exceeds
the valueof 30°-35°. The last two cores drilled at Hole 794C had a
stablemagnetization vector of normal polarity with consistent
inclina-tions of about 60°. The same polarity was observed in the
tuffdividing the last two units.
936
-
PALEOMAGNETISM OF VOLCANIC ROCKS
Site 795
Site 795 is the only site drilled in the Japan Basin which
reachedthe acoustic basement. Basaltic samples were studied only
from thelithological units 1 and 3 while no samples were studied
from the verysmall Unit 2 (703.3-704 mbsf) which probably is only
an highlyaltered part of Unit 1 (Tamaki, Pisciotto, Allan, et al.,
1990). Magneticproperties show clearly differences between the two
units as can beseen in Figure 4 which summarize the results for the
basalts drilledfrom Hole 795B. The magnetic susceptibility (K) is
mostly constantalong the section (mean value: 22,440 × 10"6 S.I.
units) except fewvery altered specimens from the lower part of Unit
3 which isbrecciated and is considered a subunit. The intensity of
the NRM, theKoenigsberger ratio (Q), and the ratio between VRM and
NRM arevery different between the two units. Unit 1 is
characterized by anintensity of remanence (mean: 157 mA/m) which is
about two ordersof magnitude less than the NRM of Unit 3 (mean:
13,141 mA/m). Theinduced magnetization appears predominant over the
remanence(Q mean: 0.3) in the upper unit, while in the Unit 3 the
remanence ispredominant (Q mean: 18.5). Moreover, the VRM of the
upper unitis significant with a mean value of 90% for the ratio
VRM/NRM,while the lower unit had only a few percent of viscosity
(mean valueof VRM/NRM% = 5.6). The frequency dependence
susceptibility(Xfd) tends to increase from Unit 1 (mean: 0.8) to
Unit 3 (mean: 2.32).Optical observations suggest that the grain
size of the magneticminerals is different in the two units with
finer grains in the lowerunit. The anisotropy of magnetic
susceptibility (AMS) is weak withthe parameter F well below the
value of 1.040 (Fig. 3B).
Demagnetizations of samples from this site (Fig. 1) confirmed
the
SITE 795
680 700 720
Depth (mbsf)
740 760
Figure 4. Magnetic susceptibility (K), NRM intensity, VRM/NRM%,
andKoenigsberger ratio (Q) of igneous rocks recovered at Site 795
plotted vs. depth.Values as in Table 1.
normal polarities obtained on board, except two samples
whichexhibited negative inclinations. One belong to Section
127-795B-36R-1 and another one to Section 127-795B-41R-1. Probably
the twosamples were misoriented during the sampling, however, it
must benoted that the former exhibited a normal polarity at the
measure ofthe NRM and the direction changed drastically with AF
peak field ofonly 5 mT. However, it is preferred to interpret all
the basalts asnormally magnetized.
Site 797
NRM intensities (Jr) and magnetic susceptibility (K) divide
thesequence into three groups as showed by the downhole plots
ofFigure 5 A and 5B. The first group includes the first 16 cores
which
1öOUU -
12000 -
| 8000 -E
4000 -
0 -
NRM
SITE 797
1• i •
j
A
500 600 700 800Depth (mbsf)
900
150000 -
100000 -
50000 -
0 -
SITE
K -
797
i W
B
l
fa500 600 700 800 900
Depth (mbsf)
Figure 5. NRM intensity (A) and magnetic susceptibility (B)
plotted vs. depth.
recovered igneous rocks (from Core 127-797C-8R through
-24R;550.9-713.6 mbsf; representing nine lithologic units)
characterizedby an intensity of magnetization that in most of the
cases is below103 mA/m (mean 1190 mA/m) and a magnetic
susceptibility in therange 10-41 × 10~3 S.I. units (mean 24,993 ×
lO^S.I. units). The secondgroup is represented by the basalts
recovered from Cores 127-797C-26R through -36R (lithologic units
10-18; from 723.2 to 814.4 mbsf)with higher Jr (mean: 2579 mA/m)
and K (mean: 82,361 S.I. units).The last three cores drilled into
the basement (852.4-895.6 mbsf)which represents three lithologic
units are characterized by lowerNRM intensities (mean: 579 mA/m),
by a mean value of susceptibilityof 36,208 × 10~6 S.I. units) and
by a different polarity with respect tothe upper units. The
Koenigsberger ratio (Q), with the exception offew samples, is
systematically below 1 as can be seen in the plot ofFigure 6,
indicating that the induced magnetization play an importantrole on
the magnetic anomalies. Similar to Site 794, the basalts fromSite
797 exhibited a high VRM with a mean value for the ratioVRM/NRM% of
about 55%. The frequency dependent susceptibility(Xfd) had a mean
value of 1.09 %. The anisotropy of magnetic suscep-tibility (AMS)
is variable, and the parameter F is above the valuewhich defines
intrusive emplacement mode (F > 1.040) only in the25% of the
samples (Fig. 3C). Nevertheless, the majority of the
937
-
L. VIGLIOTTI
18 -
1 6 -
s 14~ε 12-
2 i o -o^ ft —
ε 6 -i
2 -
0 -
11•u
I1i11 1
•
1
SITE 797
•-2 -1,5 -1 -0,5 0 0,5 1 1,5 2
LogQ
Figure 6. Histogram showing a log-normal distribution of the
Koenigsbergerratio (Q) values.
samples are clearly representative of sill, or dikes, as
testified by thechilled margins. Most of the samples with a
significant AMS belongto the basalts collected from Units 5-10.
Twenty-six specimens, that is at least one for each
lithologicalunit, were demagnetized using both AF (22 samples) and
thermal(4 samples) treatment. The basalts from the upper 18 units
(from Core127-797C-8R through Core -36R) exhibited consistently
normalpolarity while negative inclinations, mostly overprinted by a
normalpolarity, mark the samples belonging to the last three units.
Twoexamples of vector demagnetization diagrams are shown in Figure
1.
The intrusion of the basaltic sills appears to have strongly
affectedthe interlayered sediments that in many case exhibited
remagnetiza-tion clearly due to the emplacement of the sills. Whole
core measure-ments, on the archive halves of the cores, carried out
on-board, exhibiteda long normal polarity for the sediments
interbedded with the basaltsof the upper 12 units until Section
127-797C-30R-2 at 755.5 mbsf.These sediments represent
lithostratigraphic Units V and VI. Ther-mal demagnetization of
discrete samples collected from Unit VIchanged dramatically this
configuration. The samples showed a strongoverprint that in several
cases could be only partially removed, but thechange of directions
during the cleaning follow a great circle clearlyindicating a
reverse polarity. An example is reported in Figure 7. Thesediments
of lithostratigraphic Unit VI below 755.5 mbsf, fromCore
127-797C-30R to Core 127-797C-37R still yield an overprintof normal
polarity, but it could be easily removed (Fig. 8). Boththermal and
AF cleaning were effective in isolating the primarydirection of the
sediment. Sedimentological evidence suggests thatthe sediments
recovered with these cores are finer grained siltyclaystone. It is
noteworthy that the two vectors of primary andsecondary component
were exactly opposite each other, so duringthermal cleaning the
intensity of the remanence increased, reachingin one sample
(127-797C-37R-2, 116-118 cm), at 200°C, a value of376% with respect
to the NRM intensity (Fig. 9). The darker coloredclaystone
recovered with Core 127-797C-41R again exhibited acomplex
paleomagnetic record with a normal polarity that probablyis of
secondary origin. Even thermal demagnetization was ineffectivein
resolving the direction of a sample collected from this
claystone.The direction changed during progressive demagnetization,
but anegative polarity was obtained only at 480° C when the sample
isviscous and the susceptibility increases. The reason of this
overprintcannot be related to the emplacement of the overlying sill
which alsoshows negative inclinations with an overprint of normal
polarity.
127-797C-23R-5,79-81cm
90
180"Figure 7. Change of magnetic directions of sedimentary rocks
interlayered with the basalts during thermal and AF
demagnetization. A stable end-point is not
reached. Open symbols: negative inclinations; solid symbols:
positive inclinations.
938
-
PALEOMAGNETISM OF VOLCANIC ROCKS
270
127-797C-22R-4/l24-126cm280°
/ 127-797C-30R-3J46-48cm
180Figure 8. Examples of magnetic directions reaching a stable
end-point during progressive thermal demagnetization from sediments
interlayered with the basalts.Open symbols: negative inclinations;
solid symbols: positive inclinations.
VISCOUS REMANENT MAGNETIZATIONEXPERIMENT
It is well known that oceanic basalts are sometimes very
suscep-tible to acquiring a viscous remanent magnetization (VRM),
and forthis reason Lowrie (1973) proposed that this kind of
magnetizationmight be an important magnetization component of the
oceanic crust.Many experiments have been carried out to estimate
the seriousnessof VRM on the oceanic basalt magnetization. However,
most of theseexperiments were carried out after the NRM was
completely destroyedby AF-demagnetization and on a selected number
of specimens whichshowed instability during measurement of NRM
properties. More-
over, the VRM was given to the samples in a field of 0.1 mT.
Theseexperiments therefore tried to establish the behavior of the
VRM, butin artificial conditions. Lowrie and Kent (1976,1978) and
Tivey andJohnson (1981) found that VRM acquired in the presence of
NRM ismore serious than after AF demagnetization.
In order to apply the VRM acquisition to the basement of the
JapanSea, the present experiment was carried out on the whole
collectionof samples and under conditions similar to those existing
in nature.The results, relative to Sites 794,795, and 797 are
reported in Tables 1,2, and 3, respectively. The VRM was calculated
by subtracting theoriginal NRM values from the remanence measured
after 1376 hr.This VRM was used to calculate the ratio VRM/NRM%
which is an
939
-
L. VIGLIOTTI
200°
N
D
127-797C-30R-3,46-48cm
103 mA/m
i i i
s
127-797O37R-2,
U
NRM /
116-118cm
D
W/̂ -200°
/^-250°
150° /7SθO°330?//- /y-360°100/
< < - ΛOO°- J^Λ80°
"510°i i i i
-
E
Figure 9. Examples of vectors demagnetization diagrams showing
two opposite components with different
unblocking temperatures. 1 division: 15 mA/m for the upper
sample, and 10.3 mA/m for the lower sample.
940
-
PALEOMAGNETISM OF VOLCANIC ROCKS
expression of the seriousness of the VRM. After the VRM
acquisitionexperiment (lasting 1376 hr), a select number of samples
were storedin a magnetic shield for 7176 hr to observe the decay of
the VRM.
The VRM acquisition of the basalts recovered from the sills of
theacoustic basement of the Yamato Basin (Sites 794 and 797) was
higher(mean value for the ratio VRM/NRM% respectively of 46.9%
and54.9%) than that exhibited by the basalts from Site 795 in the
JapanBasin, which had only 20% of viscosity. Lowrie and Kent
(1978)reported a mean value for the VRM/NRM% measured after
AFdemagnetization of 111 samples from 30 DSDP sites of only 20%.
Acomparison of these results confirm that the VRM acquired in
thepresence of the NRM is much more serious than the VRM
acquiredfrom a demagnetized state.
Whether the VRM is acquired by multidomain or by very fine,near
superparamagnetics grains, according to theory, the growth mustbe
logarithmic with time (Dunlop 1973). However, as observed inseveral
oceanic basalts (Lowrie, 1974; Lowrie and Kent, 1978), theVRM was
acquired by a multistage process. Two or three stage wereobserved
during our experiments. About 50% of the samples exhib-ited a
three-stage process (Fig. 10), and some results seem to suggestthat
in the samples showing two stages, the third was not
observedbecause the length of the experiment was not sufficient to
reveal it.The basalts which exhibited a three-stage process had,
after only1376 hr, a VRM of the same magnitude of the NRM; two
samples(127-794C-4R-1,79-81 cm; and 127-797C-44R-1, 33-35 cm)
exceedthis value by 3 and 5 times, respectively (Fig. 11). Samples
from thelower unit of Site 795 exhibited low viscosity with a slow,
but constant,increase of VRM with time (Fig. 12). Few specimens
decreased the NRMprobably because the VRM grew in the opposite
direction of themagnetization vector.
1200
1000 -
800 -
600 -
4 0 0 -
200 -
127-794C-8R-2,38-40cm
ET
127-795B-35R-l,51-53cm
127-797C-8R-2,145-147cm
10 100
Time (hr)1000 10000
Figure 10. Examples of three-stage acquisition of viscous
remanent magneti-zation from basaltic rocks.
Experimental observations about acquisition and decay of
VRMindicate contrasting results. Several authors found that the
initial rateof decay of VRM is the same as the initial rate of
acquisition (seeDunlop, 1973, for a complete list of references),
however, more recentstudies suggest that VRM decay proceeds more
slowly than thecorresponding VRM acquisition (Dunlop, 1983; Tivey
and Johnson1984) with the viscous acquisition coefficient (Sa)
about twice theviscous decay coefficient (Sd) (Sa = 2Sd). The
experiments carriedout on the basalts from the Japan Sea for this
study indicate that theVRM decay curves do not follow the VRM
acquisition curves, andany correlation between the two curves
appears impossible. VRM
1200
1000-
800-
600 -
400 -
200 -
127-794C4R-l,79-81cm
O 127-797C44R-l,33-35cm
1 10 100Time (hr)
1000 10000
Figure 11. Examples of very viscous samples which acquired VRM
exceedingthe NRM after 1370 hr under the present magnetic
field.
decay appears more constant with time with respect to the
VRMacquisition (Figs. 13 and 14). After 7176 hr of storage in a
magneticshield (more than 1 yr) VRM decayed below 50% of the NRM in
veryfew samples. The behavior was variable from sample to sample,
andit is difficult to establish a common trend. Some samples
increased inintensity, and this could be related to the erasing of
a VRM with adirection opposite to the magnetization vector. An
example of thisprocesses is reported in Figure 15 showing the
change of the NRMdirection during VRM acquisition and decay. Some
samples, after aninitial decay of the remanence, changed, acquiring
VRM. The basaltsfrom the lower unit of Site 795 were stable, and
the VRM changedonly of few percent with very consistent
paleomagnetic directions.
DISCUSSION AND SUMMARY
The results of the magnetic properties measured from the
igneousrocks recovered in the Japan Sea indicate differences
between thebasalts of the acoustic basement of the Yamato Basin
(Sites 794 and797) and the Japan Basin (Site 795). The former are
characterized by
25000
20000 -
15000 -
10000
127-795B-40R-l,22-24cm
o
127-795B-37R-l,28-3Ocm
10 100
Time (hr)
1000 10000
Figure 12. Examples of VRM acquisition from two samples
collected atSite 795.
941
-
L. VIGLIOTTI
1200
200-
127-794C-7R-l,105-107cm
1200
10 100 1000 10000Time (hr)
Figure 13. Comparison between the curves of VRM acquisition
(solid symbols)and VRM decay (open symbols) from basalts recovered
from Site 794.
higher values of magnetic susceptibility while the NRM intensity
isabout one order of magnitude less with respect to the samples
meas-ured at Site 795. The Koenigsberger (Q) ratios are probably
one ofthe most significant differences between the two basins. In
the YamatoBasin low Q values testify that the remanence does not
dominant overthe induced magnetization, so one of the requirements
of the Vine andMatthews (1963) hypothesis to produce symmetrical
oceanic mag-netic anomalies over ridges is not present in the
Yamato Basin. Thisis the reason for the unclear magnetic anomaly
pattern observed inthis basin. On the contrast, in the Japan Basin,
remanence dominatesover the induced magnetization and a marine
magnetic anomalypattern can be recognized (Tamaki and Kobayashi
1988). Q valuesaround unity as observed at Sites 794 and 797 and in
the upper unitat Site 795, together with the high VRM of these
rocks, testify to largegrain size in the multidomain range for the
carrier of the remanence.Optical observations confirm this
statement and support Irving's(1970) assumption that conditions at
initial rifting would favor devel-opment of coarse grain sizes in
basalts.
The frequency dependence susceptibility Xfd should exhibit
adecrease with frequency according to the Neel's (1955) theory
ofmagnetic viscosity. Before the measurements reported in this
paper,Xfd measurements on oceanic basalts were carried out only by
a recentstudy of Trigui andTabbagh (1990). The mean decrease in
suscepti-bility observed in the basalts drilled from the Japan Sea
is in the range1.09%-1.99% which is lower than the theoretical
decrease of 4%calculated by Vincenz (1965) for multidomain grains
and a tenfoldincrease in frequency.
Ellwood's emplacement parameter (F) was unsuccessfully appliedto
the basalts in this study. At Site 794 where all the units have
beenrecognized as sills, only two samples exhibited an AMS
representa-tive of an intrusive emplacement (F > 1.040). Also at
Site 797,parameter F does not show a correlation with the
recognized emplace-ment mode of the basalts. One of the reasons for
the inapplicabilityof this criterion is that it has been applied to
relatively younger (andless altered) basalts (Ellwood, 1978).
However, some debate couldexist about the emplacement mode of the
first basaltic unit at Site 797which has been considered a lava
flow. One sample from this unitexhibited a F value of 1.044 typical
for intrusive rock. As found byEllwood (1975), intrusive rocks show
a wide range of distribution ofF values, but extrusive rocks are
much more characterized by F valuesclose to one. On this basis it
is possible that Unit 1 at Site 797 also is a
1000-
800
600-
400
• VRM(a)
O VRM(d)
127-797C-9R-1,110-112cm
10 100 1000Time (hr)
10000
1200
1000-
800 -
• VRM(a)
O VRM(d)
-o-o
100
Time (hr)
1000 10000
800
700 -
600 -
500 -
400-
300
127-797C-45R-2,79-81cm
• VRM(a)
O VRM(d)
10 100Time (hr)
1000 10000
Figure 14. Comparison between the curves of VRM acquisition and
VRMdecay from basalts recovered from Site 797.
942
-
PALEOMAGNETISM OF VOLCANIC ROCKS
dike. Petrographic similarity with Unit 2 suggests a similar
emplace-ment mode for this unit, and probably only intrusive rocks
charac-terize the acoustic basement of the Yamato Basin. F values
from Site 795are well in the range of the extrusive emplacement
mode supporting theobservation that the basalts recovered at this
site represent lava flows.
Igneous rocks from Site 795 exhibit only normal polarities
whileboth polarities were observed in the dikes from Sites 794 and
797testifying to different ages for the emplacement of these
rocks.Especially at Site 794 the emplacement history of the dikes
appearscomplicated with normal polarity at the top and the bottom
of thesection, while in the center negative inclinations
predominate, butwith a normal polarity in the lower part of Core
127-794C-3R. Theinclinations of the basalts recovered with the
upper eleven cores atHole 794C are systematically shallower than
the expected value andonly the last two cores, which also had a
different polarity, exhibitedinclinations consistent with the
latitude of the site. Three possibilitiescan explain these
results:
1. The basement is disrupted only in the upper 90 m.2. The
basalts at the bottom of the hole have a different age,
possibly younger, and were emplaced after the disruption.3. The
basalts recovered with the upper eleven cores were emplaced
in a short period of time and their directions represent a
virtual geomag-netic pole (VGP) which was characterized by shallow
inclinations.
Against this latter hypothesis, play the differences observed in
themagnetic properties which could indicate that the emplacement
wasnot a single event as expected in the case of a not-averaged
VGP.
The intrusion of the dikes clearly affected the magnetization
ofthe interlayered sediments as testified by the process of
remagnetiza-tions observed at Site 797. The presence of two
opposite componentsof magnetization with different unblocking
temperatures, gives anopportunity to estimate the temperature at
which the sediments wereheated. The separation of these two
components show that unblockingtemperatures are restricted to
200°-220°C and in few case reached250° C (Fig. 9). These
temperatures can be considered representativeof the heating of the
sandstones and siltstones interlayered with theigneous rocks.
Viscous remanent magnetization (VRM) experiments showedthat the
VRM acquisition occurs in two or three different stages,each of
which following a logarithmic dependence respect to theacquisition
time (Fig. 10). VRM decay is a slower process, but stillinvolving
more than one stage (Figs. 13 and 14). The VRMobserved in the
studied samples was significantly higher than theVRM reported from
other study on demagnetized samples, sup-porting the observation of
Lowrie and Kent (1978) that VRM ismore serious in undemagnetized
samples.
ACKNOWLEDGMENTS
The author is grateful to Dr. L. Monti for the help in
measuringthe VRM of the rocks and to Dr. C. Ferretti for plotting
the data. This
paper was supported by Contribution No. 826 of the Istituto
diGeologia Marina, CNR, Bologna, Italy.
REFERENCES
Dunlop, D. J., 1973. Theory of the magnetic viscosity of lunar
and terrestrialrocks. Rev. Geophys. Space Phys., 11:855-901.
, 1983. Viscous magnetization of 0.04-100 µm magnetites.
Geo-phys. J. R. Astron. Soc, 74:667-687.
Ell wood, B. B., 1975. Analysis of emplacement mode in basalt
from Deep-SeaDrilling Project Holes 319A and 321 using anisotropy
of magnetic suscep-tibility. J. Geophys. Res., 80:4805-4808.
, 1978. Flow and emplacement direction determined for
selectedbasaltic bodies using magnetic susceptibility anisotropy
measurements.Earth Planet. Sci. Lett, 41:254-264.
Irving, E., 1970. The mid-Atlantic ridge at 45°N: XIV. Oxidation
and mag-netic properties of basalt: review and discussion. Can. J.
Earth. Sci.,7:1528-1538.
Lowrie, W., 1973. Viscous remanent magnetization in ocean
basalt. Nature,243:27-30.
, 1974. Ocean basalt magnetic properties and the Vine and
Matthewshypothesis. J. Geophys., 40:513-563.
Lowrie, W., and Kent, D. V, 1976. Viscous remanent magnetization
in basaltsamples. In Yeats, R. S., Hart, S. R., et al., Init.
Repts. DSDP, 34: Washington(U.S. Govt. Printing Office),
479-484.
, 1978. Characteristics of VRM in ocean basalts. J.
Geophys.,44:297-315.
Neel, L., 1955. Some theoretical aspects of rock magnetism.
Advan. Phys.(Phys. Mag. Suppl.), 4:191-243.
Tamaki, K., and Kobayashi, K., 1988. Geomagnetic anomaly
lineation in theJapan Sea. Mar. Sci. Mon., 20:705-710.
Tamaki, K., Pisciotto, K., Allan, J., et al., 1990. Proc. ODP,
Init. Repts., 127:College Station, TX (Ocean Drilling Program).
Tivey, M., and Johnson, H. P., 1981. Characterization of viscous
remanentmagnetization in single- and multi-domain magnetite grain.
Geophys. Res.Lett, 8:217-220.
, 1984. The characterization of viscous remanent magnetization
inlarge and small magnetite particles. J. Geophys. Res.,
89:543-552.
Trigui, M., and Tabbagh, A., 1990. Magnetic susceptibility of
ocean basalts inalternative fields. /. Geomagn. Geoelectr.,
42:621-636.
Vincenz, S. A., 1965. Frequency dependence of magnetic
susceptibility ofrocks in weak alternating fields. J. Geophys.
Res., 70:1371-1377.
Vine, F. J., and Matthews, D. H., 1963. Magnetic anomalies over
ocean ridges.Nature, 199:947-949.
Date of initial receipt: 23 April 1991Date of acceptance: 10
February 1992Ms 127/128B-212
943
-
L. VIGLIOTTI
\ 127-794C-4R-1J79-81cm
Figure 15. Stereonet plot of paleomagnetic directions changing
with time (numbers express hours) during acquisition and decay of
VRM.
944
-
PALEOMAGNETISM OF VOLCANIC ROCKS
fc N R M- h -4- •+•
127-797C-44R-2l90-92cm
Figure 15 (continued).
945
