7/29/2019 paleocurrent
1/15
Paleocurrent analysis for the Late PleistoceneHoloceneincised-valley ll of the Yangtze delta, China by using
anisotropy of magnetic susceptibility data
Baozhu Liua,*, Yoshiki Saitoa, Toshitsugu Yamazakia, Abdelaziz Abdeldayema,b,Hirokuni Odaa, Kazuaki Horic, Quanhong Zhaod
aMRE Geological Survey of Japan, AIST, Higashi 1-1-1, Tsukuba, Ibaraki 305 8567, Japan
bDepartment of Geology, Faculty of Science, Tanta University, Tanta 31527, Egypt
cDepartment of Geography, Graduate School of Science, University of Tokyo, Hongo 7-3-1,
Bunkyou-ku, Tokyo 113 0033, JapandMarine Geology Laboratory, Tongji University, 1239 Siping Road, Shanghai 200092, People's Republic of China
Received 9 August 2000; accepted 16 March 2001
Anisotropy of magnetic susceptibility (AMS) analysis has been conducted on samples from borehole core CM-97 from the
Yangtze River (Changjiang) incised-valley ll, China, to determine the paleocurrent directions to help in reconstructing
sedimentary paleoenvironments. Borehole CM-97 consists of uvial (Unit 1), estuarine (Units 26), and deltaic (Units 79)sediments after the Last Glacial Maximum in ascending order. The AMS results show that the paleocurrent directions for the
tide-dominated estuarine and deltaic sediments were westerly or northwesterly directed (ood-tide dominated), but give no
denite trend for the uvial sediments.
Comparison between the paleocurrent directions inferred from primary sedimentary structures and in situ AMS data shows
that they are in good agreement, conrming the applicability of AMS as a good paleocurrent indicator for sediments deposited
in coastal tide-dominated environments. Considering that these sediments were strongly tide-inuenced and the tidal pattern
since 12 kyr bp has not changed signicantly, we think that the westerly or northwesterly current direction most probably
resulted from ood-tidal currents, and the sedimentary paleoenvironment was a ood-tide dominated estuary or delta.
Additionally, it has been found that downhole changes of some AMS parameters, including the mean magnetic susceptibility
(K), the corrected anisotropy degree (Pj) and the magnetic foliation (F), clearly mark the dened stratigraphic boundaries in the
borehole. This further extends the validity of AMS as a good stratigraphic marker in addition to its long known credibility as a
sensible paleocurrent recorder.q
2001 Elsevier Science B.V. All rights reserved.Keywords: Anisotropy of magnetic susceptibility (AMS); Yangtze delta; Paleocurrent; Tide-dominated; Incised-valley ll
All materials acquire magnetization in a magnetic
eld and thus have a magnetic susceptibility. This
susceptibility is not always isotropic and varies with
the orientations of the rock (Ising, 1942). This
Marine Geology 176 (2001) 175189
0025-3227/01/$ - see front matter q 2001 Elsevier Science B.V. All rights reserved.
PII: S0025-3227(0 1)00151-7
www.elsevier.nl/locate/margeo
* Corresponding author. Tel.: 181-298-61-3719; fax: 181-298-
61-3747.
E-mail addresses: [email protected] (B. Liu).
7/29/2019 paleocurrent
2/15
spatial susceptibility variation is dened as the
anisotropy of magnetic susceptibility (AMS) and
reects the preferred orientation of magnetic
minerals in the rock or unconsolidated sediments,
i.e. its magnetic fabric (Hrouda, 1982; Tarling and
Hrouda, 1993). AMS has long been demonstrated as
a useful tool for paleocurrent determination, in parti-
cular for deep-sea sediments such as contourites
(Ellwood and Ledbetter, 1977, 1979; Ellwood et
al., 1979; Ellwood, 1980; Ledbetter and Ellwood,
1980; Abdeldayem et al., 1999), turbidites (Ellwood
and Ledbetter, 1977; Ledbetter and Ellwood, 1980),
submarine canyon and fan sediments (Rees et al.,
1968), as well as Mid-Proterozoic embayment
shales (Schieber and Ellwood, 1993) and Palaeo-
zoic ysch shales (Piper et al., 1996), modernbeach sand sediments (Rees, 1965; Taira and
Lienert, 1979), and laboratory deposited sediments
(Rees, 1965; Rees and Woodall, 1975). AMS of
experimental tidal at sediments has also been
reported (Ellwood, 1984). However, to the best of
our knowledge, no similar work has been done on
natural sediments deposited in coastal tide-domi-
nated estuary or delta environments. If AMS were
also applicable, it would be of great help for
detailed reconstruction of coastal sedimentary
paleoenvironments.AMS of a rock sample corresponds to a symmetri-
cal second-rank tensor (Hrouda, 1982) which can be
described by a triaxial ellipsoid with the principal
eigenvectors K1. K2. K3 representing the maxi-
mum, intermediate and minimum susceptibility
axes, respectively. Usually, current would be parallel
to the K1 axis and in favorable conditions its absolute
direction may be inferred from the tilting direction of
K3 axis (Rees, 1965; Tarling and Hrouda, 1993;
Tarling and Shi, 1995; Piper et al., 1996; Abdeldayem
et al., 1999). However, current could be perpendicular
to the axes ofK1 if the ow is strong enough and thegrains are very ne under traction sedimentation
(Ellwood and Ledbetter, 1977, 1979; Ledbetter and
Ellwood, 1980).
This study, therefore, attempts for the rst time to
use AMS, to determine paleocurrent direction for
sediments from the Late PleistoceneHolocene
Yangtze incised-valley ll, China, which are charac-
terized by typical tide-dominated estuarine and deltaic
sediments (Hori et al., 1999, 2001a,b).
The Yangtze River (Changjiang) of China, the
longest river in Asia, has a length of about 6300 km,
a total catchment area of about 1.8 106 km2, meanannual runoff of 893 109 m3, and mean annual sedi-
ment discharge of 481 106 t (Milliman and Meade,
1983; Huang et al., 1996; Li and Wang, 1998; Chen et
al., 2000). It originates in the Kunlun Mountains in the
southwestern part of Qinghai Province, north of the
Tibet Plateau, and ows towards the east to its mouth
into the East China Sea, about 23 km north of Shang-
hai. The present Yangtze delta has been forming since
the maximum transgression in the Holocene at about
7 kyr bp (Liu et al., 1992; Huang et al., 1996; Chen
and Chen, 1997; Li and Wang, 1998; Li et al., 2000).The Yangtze delta plain, with low relief of 35 m, is
located at its end and faces the East China Sea and the
South Yellow Sea (Chen, 1999; Li et al., 2000).
The Yangtze delta is one of the typical tide-domi-
nated deltas in the world. At present, it is in a meso-to
macrotidal environment with an average tidal range of
2.6 m, and the maximum of about 5.0 m (Li and Wang,
1998). The present tide in the Yangtze River estuary
area is irregular semidiurnal, the ood-tidal current
direction is towards the NW, while the ebb-tidal current
is obviously diverted towards the south at SSE (Chen etal., 1988; Huang et al., 1996; Chen, 1999; Chen et al.,
2000). However, tidal currents were inferred to be
much stronger during the transgression stage in the
Holocene than at present, during which the average
tidal range was greater than 4 m (Li et al., 2001).
During the Last Glacial Maximum, the shoreline was
located near the edge of the East China Sea continental
shelf (Zhu et al., 1979), and a huge incised-valley was
formed in the present Yangtze delta area (Li et al., 2000;
Fig. 1), whereas during the postglacial sea-level rise,
most of the incised-valley had been lled, and the
present Yangtze delta had downlapped onto the estuar-ine deposits (Li et al., 2000). The incised-valley
sequence after the Last Glacial Maximum consists of
a uvial unit, estuary units, and delta units in ascending
order (Li et al., 2000; Hori et al., 2001b).
A seventy-meter-long borehole, CM-97 (latitude
B. Liu et al. / Marine Geology 176 (2001) 175 189176
7/29/2019 paleocurrent
3/15
31837 029 00N; longitude 121823 038 00E, the elevation of
2.48 m), was taken by rotary drilling using drilling
mud in 1997 on Chongming Island (Fig. 1). The
CM-97 site is located inside the incised-valley formed
during the Last Glacial Period (Fig. 1). Core recoveryis about 90%. Detailed core description and radio-
carbon dating are reported by Hori et al. (1999,
2001a,b).
CM-97 was divided into nine stratigraphic units
from bottom to top (Hori et al., 1999; Fig. 2). Radio-
carbon ages show that CM-97 recorded the sedimen-
tary environments of the Yangtze estuary for
approximately 10 kyr, from 11.5 to 1.5 kyr bp (Hori
et al., 1999, 2001a,b; Fig. 3). In general, borehole
sediments consist of transgressive uvial sediments
(Unit 1), transgressive estuarine sediments (Units
26), and regressive deltaic sediments (Units 79).Unit 1, formed prior to about 11 kyr bp, consists of
ne to medium sand uvial sediments with clear high-
angle trough-cross bedding. Units 26, deposited
during about 116 kyr bp, are transgressive estuarine
sediments consisting of thinly alternating silt and clay
layers with a few foraminifera. In detail, Unit 2
consists of inshore subtidal deposits, Unit 3 of
muddy intertidal to subtidal deposits, Unit 4 of trans-
gressive lag deposits, Unit 5 of muddy intertidal to
subtidal at deposits, and Unit 6 of estuarine central
basin deposits (Hori et al., 1999, 2001b). Units 79
were formed during about 61.5 kyrbp, and are
regressive deltaic sediments consisting of clayey silt
to ne sand. Sedimentary environments of these units
are prodelta for Unit 7, delta front for Unit 8, subtidal
to intertidal at and surface soil for Unit 9 (Hori et al.,
1999, 2001a).
Altogether there were 39 subcores of samples from
borehole CM-97. All the subcores were split into two
halves. A total of 2543 sequentially numbered discrete
samples were taken by continuously pressing 7 cm3plastic boxes into the face of the working half. Initial
low eld magnetic susceptibility (K) and its aniso-
tropy were rst measured using a KappaBridge
KLY-3S susceptibility meter. The natural remanent
magnetization (NRM) was then measured and demag-
netized using a three-axis 2G Enterprises cryogenic
magnetometer with an in-line alternating eld (AF)
demagnetizer with a peak eld strength of 80 mT.
All odd-numbered samples were subjected to incre-
mental AF demagnetization at steps of 0, 5, 10, 15, 20,
25, 30, 35, 40, 50, 60 and 80 mT. Following statisticaland visual analysis of this detailed demagnetization
spectrum, we found that most samples exhibited RM
stability. Therefore, it was decided to treat the remain-
ing samples (the even-numbered samples) at steps of
0, 20, 30 and 40 mT. Fig. 4 shows typical demagne-
tization behavior for these examples. The majority of
samples showed a stable magnetization expressed as a
single component that heads toward the origin of ortho-
gonal plots (Zijderveld diagram, Zijderveld, 1976, Fig.
4Ba). Although they still exhibited a general steady
decay toward the origin of the plot, the remaining
samples either acquired some spurious magnetization
at high elds (Fig. 4b) or behaved in an erratic manner
because their magnetization was too weak and
they acquired spurious magnetizations at low elds
(Fig. 4c).
Combined visual (using stereographic and orthogo-
nal plots) and statistical (using the principal compo-
nent analysis of Kirschvink, 1980) inspections of
demagnetization data indicted that 20 mT AF demag-
netization was sufcient to remove most of the
B. Liu et al. / Marine Geology 176 (2001) 175189 177
Fig. 1. Location of borehole CM-97 and map of the Yangtze
(Changjiang) delta. The area between the two dashed lines is the
huge incised-valley of the Yangtze River, which was formed during
the Last Glacial Maximum (after Li et al., 2000). The shadowed
rectangle in the inset shows the location of the study area in China.
7/29/2019 paleocurrent
4/15
viscous remanence and isolate the stable magnetic
north direction for most samples. For the remaining
samples, magnetic north had to be computed using the
principal component analysis (Kirschvink, 1980) by
tting a line through a minimum of three consecutive
steps and toward the origin of the orthogonal plot.
Thus we could obtain the magnetic north of each
sample. Then the relative magnetic north direction
B. Liu et al. / Marine Geology 176 (2001) 175 189178
Fig. 2. Column section with sedimentological features of borehole CM-97 (after Hori et al., 1999, 2001a,b).
7/29/2019 paleocurrent
5/15
at each sample level was calculated using the equation
determined from linear tting of obtained magnetic
north versus depth in order to correct the affection
of paleosecular variation. This linear tting was
done on each subcore as it has the same cut direction.
And this relative magnetic north of each sample was
used for reorientation of AMS directions of the rela-
tive sample to their geographic coordinates so as to
obtain the absolute paleocurrent directions (Abdel-
dayem et al., 1999).
Following the recommendation of Jelinek (1981),
Ellwood et al. (1988) and Tarling and Hrouda(1993), the following set of AMS parameters that
dene the mean magnetic susceptibility (K), the
corrected anisotropy degree (Pj), the magnetic
lineation (L), the magnetic foliation (F) and the
ellipsoid shape (q) were calculated and used to
evaluate the magnetic fabric of borehole CM-97
sediments:
K K1 1 K2 1 K3=3
(Mean magnetic susceptibility, Nagata, 1961)
Pj exp
{2n1 2 nm
2 1 n2 2 nm2 1 n3 2 nm
2}
q;
(Corrected anisotropy degree, Jelinek, 1981)
where n1 ln K1; n2 ln K2; n3 ln K3; nm
n1 1 n2 1 n3=3
L K1=K2
(Magnetic lineation, Balsley and Buddington, 1960)
F K2=K3
(Magnetic foliation, Stacey et al., 1960)
q K1 2 K2=K1 1 K2=22 K3
(Shape factor, Granar, 1958)
Magnetic fabric of laboratory deposited materials
have shown that the magnetic grains are mostly
aligned within or close to the bedding plane, with
their longer axes in the direction of ow, with some
degree of imbrication (Rees, 1965; Rees and Woodall,
1975). More specically, the magnetic fabric para-
meters fall within specic ranges, such as
0.06, q, 0.7 and the imbrication angle, the angle
between horizontal and the plane of maximum-inter-mediate susceptibility, is less than 208 (Hamilton and
Rees, 1970). These ranges have been widely adopted
as being diagnostic of primary sedimentary fabrics
when they are found in natural sediments, while
values outside these ranges can generally be attributed
to coring disturbances, bioturbation and the like or
specically secondary fabrics. On the other hand,
the depositional magnetic fabrics of most deposited
sediments are characterized by clearly oblate suscept-
ibility ellipsoids (Hrouda, 1982). In the present study,
a foliated ellipsoid with q values ,0.7 and K3 direc-
tions lying within 258 of the vertical were considered
indicative of a primary fabric that is credible in
providing information on paleocurrent direction and
depositional conditions (Hamilton and Rees, 1970;
Hrouda, 1982; Tarling and Hrouda, 1993). Fig. 5
shows the features of F versus L of all the samples
with primary AMS, from which we can see that all the
samples have oblate ellipsoids, indicating primary
depositional nature of these samples (Hrouda,
1982). Thus, only those samples that have a primary
B. Liu et al. / Marine Geology 176 (2001) 175189 179
Fig. 3. Accumulation curve of borehole CM-97 and sea-level curve
in the East China Sea since 12 kyr bp. The accumulation curve is
after Hori et al. (1999), while the sea-level curve is after Saito
(1998).
7/29/2019 paleocurrent
6/15
magnetic fabric were used in paleocurrent determina-
tion for borehole CM-97 samples. There were a total
of 664 samples with secondary AMS, which were
mostly distributed on the top part of each subcore.
Description of the primary sedimentary structures
was done on the basis of detailed examination of
X-ray photographs of samples. Paleocurrent direc-
tions, relative to the subcore section, were inferred
from the dipping directions and angles of the foresets
of cross lamination. Then paleocurrent direction of the
sample located at the same horizon as the cross lami-
nation was determined by using the in situ (un-reor-iented) AMS data, so that we were able to compare the
results inferred from the sedimentary structures and
the in situ AMS data. In this case, we assumed that the
cut section of each subcore was along the `eastwest'
line, and the sample box was pressed into the subcore
section towards the `north'. Thus, paleocurrent direc-
tion from both the primary sedimentary structures and
the in situ AMS data were in the same coordinate
system, and were comparable.
5.1. Comparison between the paleocurrent directions
determined from the primary sedimentary structures
and the in situ AMS data
Sedimentary structures, especially the cross lami-
nations with clear foresets that have current direction
implications (Reineck and Singh, 1980; Allen, 1984;
Reading, 1996) were observed and described based on
detailed examination of X-ray photographs. In order
to demonstrate the applicability of AMS in paleocur-rent determination of CM-97 sediments, we selected
one subcore (Subcore B30) to conduct detailed
comparison between the paleocurrent directions
from the primary sedimentary structures and the in
situ AMS data, respectively (Fig. 6). Subcore B30
was located at the depth of 36.80 37.50 m within
stratigraphic Unit 6, and consisted of thinly inter-
bedded coarse silt and silty clay.
Firstly, we reconstructed the paleocurrents from
B. Liu et al. / Marine Geology 176 (2001) 175 189180
Fig. 4. Typical examples of directional and intensity changes during AF demagnetization of samples from borehole CM-97. (A) Normalized
NRM intensity versus AF peak amplitudes. Sample number and NRM of each sample are shown in the plots. (B) Orthogonal projections
(Zijderveld diagrams) of stepwise demagnetization of the same samples. Units of each sample are shown in the plots. The NRM measurement
for each sample is marked with a larger symbol. Horizontal projections are marked with solid squares and vertical projections are marked with
open squares.
7/29/2019 paleocurrent
7/15
primary sedimentary structures. We examined the X-
ray photos of Subcore B30 and found that there were
three horizons with clear cross lamination at depths of
(A) 36.86036.864 m, (B) 37.06837.071 m and (C)
37.07337.080 m, respectively (Fig. 6). LaminationA has a thickness of 4 mm, foreset laminae dipping
to the right with an apparent dip angle of 158, indicat-
ing a paleocurrent directed to the right; Lamination B
has a thickness of 3 mm, foreset laminae dipping to
the right with an apparent dip angle of 108, also indi-
cating a paleocurrent owing to the right; Lamination
C has a thickness of 7 mm, foreset laminae dipping to
the left with an apparent dip angle of about 308, indi-
cating a paleocurrent towards the left (Fig. 6). These
data clearly show that there were bi-directional ows
and these sediments were deposited under bi-direc-
tional currents (Fig. 6).
Secondly, we reconstructed the downhole paleocur-
rents of Subcore B30 based on the in situ AMS data,i.e. axes of the magnetic ellipsoid were in the subcore
coordinate system, to compare the paleocurrent direc-
tions between sedimentary structures and the in situ
AMS data. These downhole paleocurrent directions
were determined from the declination ofK1 (Fig. 6).
Results show that the paleocurrent of the sample at
Horizon A was to the right, and that of combined
sample B and C was to the left (Fig. 6), indicating
that paleocurrent directions from sedimentary structures
B. Liu et al. / Marine Geology 176 (2001) 175189 181
Fig. 5. Flinn-type (Fversus L) plot of the samples with primary AMS. It is very clear that all the samples have oblate ellipsoids, indicating their
primary depositional nature.
7/29/2019 paleocurrent
8/15
and AMS were in good accordance. Laminations B
and C were combined within one single AMS sample
(the thickness of one AMS sample is 20 mm), so only
one average paleocurrent direction could be inferred
from the AMS data. B was much thinner than C, thus
the paleocurrent direction from AMS showed the
predominated C direction.
Furthermore, there were a total of 41 horizons
where primary sedimentary cross laminations were
observed, among which 27 (about 65.9%) had similarpaleocurrent directions to those determined from the
in situ AMS data (Table 1). Among the 41 cross lami-
nations, 16 has a thickness of equal to or greater than
10 mm, of which 15 (about 93.8%) has similar paleo-
current directions to those determined from the in situ
AMS data (Table 1). Even 56% of those cross lamina-
tions with a thickness of less than 10 mm shows
similar paleocurrent directions to those determined
from the in situ AMS data (Table 1). These results
B. Liu et al. / Marine Geology 176 (2001) 175 189182
Fig. 6. Detailed comparison between the paleocurrent directions from primary sedimentary structures and the in situ AMS data for Subcore B30
samples from CM-97. Photos of the primary sedimentary structures were enlarged so as to make them clearer. Paleocurrent directions relative
to the core section from these sedimentary structures are shown with arrows: A, right; B, right; C, left. The downhole paleocurrent directions are
inferred from in situ AMS data (K1). A and C are in good agreement with the AMS results. In situ declinations and inclinations at 20 mT AF
eld demagnetization of the corresponding samples are also shown in the right part of the plot.
Table 1Statistics of primary cross laminations that have similar paleocur-
rent directions to those determined from the in situ AMS data
Cross
lamination
numbers
Numbers
with similar
paleocurrent
directions
Percentage
Totally 41 27 65.9
Thickness$ 10 mm 16 15 93.8
Thickness, 10 m 25 14 56.0
7/29/2019 paleocurrent
9/15
indicate that the absolute paleocurrent directions
inferred from the reoriented AMS data represent the
true current directions.
5.2. Paleocurrent reconstruction for each
stratigraphic unit
The paleocurrent directions for the nine strati-
graphic units were established from the reoriented
AMS data, based on paleomagnetically oriented K1and K3 for samples that satised the afore-mentioned
criteria for a primary fabric. K1 axes display a moder-
ate to complete girdle on the lower hemisphere equal-
area projections ofK1 and K3 (Fig. 7). Absolute paleo-
current direction was estimated from the imbrication
of K3 in the plot of lower hemisphere equal-areaprojections of reoriented K1 and K3 (open arrows in
Fig. 7). Rose diagram ofK1 was also done to show the
azimuthal distribution of K1 axes (Fig. 7). Preferred
imbrication that enables absolute current estimation
could also be observed for quite a few units (Fig. 7).
Note that there is a slight disagreement between the
absolute paleocurrent directions (open arrows in
Fig. 7) and the azimuthal K1 axes rose diagrams
(Fig. 7). This is because the absolute directions take
into account imbrications and are more realistic. The
following is a summary of AMS patterns and paleo-current evaluation for each unit in borehole CM-97
(Fig. 7).
Unit 1: K1 axes display a girdle. K1 axes are mostly
along the NS in the rose diagram, which also shows
a subordinate EW trend. The K3 axes, on the other
hand, are very steep to vertical with no clear imbrica-
tion. Such a pattern reects a predominant NS
running paleocurrent that seems to have oscillated
back and forth for short intervals, yet no absolute
current direction could be inferred from the AMS
data (Fig. 7).
Unit 2: K1 axes display a moderate girdle, while its
azimuthal distribution in the rose diagram shows a
good grouping along the SEENWW. K3 axes slightly
tilted toward the NW away from vertical, which
shows a weak imbrication indicating a probable abso-
lute NW paleocurrent direction (Fig. 7).
Unit 3: K1 axes are grouped along the NEE while K3axes are clearly tilted toward NWW indicating a
predominant NWW paleocurrent direction (Fig. 7).
Unit 4: K1 axes display a moderate girdle. K1 axes
are mostly to the east with a clear imbrication of K3axes that indicates a strong paleocurrent trending
generally westward. Rose diagram also shows a
subordinate NWWSEE running paleocurrent that
might have occurred for short periods (Fig. 7).
Unit 5: K1 axes display a moderate girdle. Although
extending over a wide range of azimuths, the K1 axes
show a higher concentration along the EW trend.
The K3 axes, on the other hand, show a slight tilt
toward the west marking a predominant paleocurrent
in this direction (Fig. 7).
Unit 6: Similar to Unit 5 but with a better
constrained ow.
Unit 7: K1 axes display a moderate girdle. K1 axes
extend over a wide range with a pronounced SWNE
concentration, accentuated by the rose diagram. Aslight imbrication that reects a NW owing paleo-
current can also be observed for the K3 axes (Fig. 7).
Unit 8: K1 axes are broadly grouped around an east-
erly direction with a general tendency for the K3 axes
to slightly imbricate toward the NWW, marking a
relatively strong paleocurrent in this direction (Fig. 7).
Unit 9: K1 axes are widely distributed around an
easterly trend. K3 axes display a pronounced imbrica-
tion that marks a strong paleocurrent owing toward
the west (Fig. 7).
In summary, the predominant paleocurrent direc-tions for the tide-dominated estuarine and deltaic sedi-
ments in the borehole were westerly or northwesterly,
and generally parallel or oblique to the azimuth ofK1axes shown in the rose diagrams. However, no abso-
lute paleocurrent direction could be obtained for the
uvial sediments from CM-97.
5.3. The mean magnetic susceptibility (MS), the
anisotropy degree, the magnetic lineation, the
magnetic foliation, the ellipsoid shape parameter and
their sedimentological implications
The mean magnetic susceptibility (K, or MS)
changes positively with grain size (Fig. 8). For the
whole borehole section, the maximum MS value is
3139 1026 SI, minimum is 228 1026 SI, and the
average is about 560 1026 SI. The lowest values of
MS occur in stratigraphic Unit 7 (Fig. 8), which is the
nest in whole borehole section (Fig. 2), with an aver-
age value of about 346 1026 SI. The highest values of
MS occur in stratigraphic Unit 1 (Fig. 8), which is the
B. Liu et al. / Marine Geology 176 (2001) 175189 183
7/29/2019 paleocurrent
10/15
B. Liu et al. / Marine Geology 176 (2001) 175 189184
Fig. 7. Paleocurrent directions from the reoriented AMS data for the nine stratigraphic units in borehole CM-97. Upper: Lower hemisphere
equal-area stereographic projections of K1 (solid square) and K3 (solid circle), showing absolute paleocurrent directions with open arrows.
Lower: Rose diagrams showing the azimuthal distribution of K1 axes.
7/29/2019 paleocurrent
11/15
B. Liu et al. / Marine Geology 176 (2001) 175189 185
Fig. 8. Downhole changes of the mean magnetic susceptibility (K), the corrected anisotropy degree (Pj), the magnetic lineation (L), the
magnetic foliation (F) and the magnetic ellipsoid shape parameter (q). These changes clearly mark the dened stratigraphic boundaries (after
Hori et al., 1999, 2001a,b), which are shown with dashed lines. The stratigraphic units are shown on the right hand side with numbers.
7/29/2019 paleocurrent
12/15
coarsest in the borehole (Fig. 2), with an average value
of about 955 1026 SI. In general, the mean MS exhi-
bits clear changes at horizon where lithology changes.
Both the corrected anisotropy degree (Pj) and the
magnetic foliation (F) also change with grain size, i.e.
the general trend increases with a decrease in grain
size (Fig. 8). The lowest, highest and average values
are 1.015, 1.208 and 1.089 for Pj, and 1.009, 1.173 and
1.067 for F, respectively.
Unlike the mean MS, Pj and F, there seem to be no
change in the magnetic lineation (L) and the shape
parameter (q) with grain-size, although there are
large downhole changes in both L and q (Fig. 8).
The lowest, highest and average values are 1.000,
1.067 and 1.015 for L, and 0.005, 0.700 and 0.226
for q, respectively.The eight stratigraphic boundaries from strati-
graphic Units 19 can be recognized clearly from
the characteristic downhole changes in K, Pj, L, F
and q, but particularly in K, Pj and F (Fig. 8). The K
value increases at the horizon where grain size
increases for each stratigraphic unit, even at the
erosion surface within Unit 4 (Figs. 2 and 8). Sharp
boundaries in K are identical to changes in lithology.
For example, the boundary between stratigraphic
Units 1 and 2, and the boundary between the Units 7
and 8 (Fig. 8) show marked changes in K. The chan-ging patterns of Pj and F also correspond to changes
between stratigraphic units although both parameters
have changes within individual units, e.g. within Unit
7 (Fig. 8). Although the downhole changes ofLand q
are not so clearly related to lithology, some bound-
aries can also be recognized from their curves, such
as, the boundaries between Units 1 and 2, and between
Units 8 and 9 (Fig. 8). All of these changes made it
possible for us to easily recognize the stratigraphic
boundaries throughout the CM-97 borehole (Figs. 2
and 8).
6.1. Paleocurrents and estimated sedimentary
paleoenvironments
It has been demonstrated that core samples could be
reoriented to their geographical coordinates by using
remanent magnetization (Hailwood and Ding, 1995;
Rolph et al., 1995). In this study, paleocurrent direc-
tions were determined from the reoriented AMS data
by using the relative magnetic north instead of the in
situ magnetic north. Relative magnetic north of each
sample was calculated from the equation determined
from the linear tting of obtained magnetic north
versus depth on each subcore. We used the relative
magnetic north in order to correct the affection of
paleosecular variation.
The AMS data indicates that most of the Yangtze
estuarine and deltaic sediments were deposited under
a relatively strong current that mostly owed toward
the west to northwest direction. Sedimentary facies
showed most of the CM-97 sediments were inuenced
by tidal uxes (Hori et al., 1999; Hori et al., 2001a,b).
Moreover, other borehole data from the Yangtze deltaarea also showed similar effects on sediment facies (Li
et al., 2000; Hori et al., 2001a,b). Present sedimentary
environments in the Yangtze estuary indicate that
strong tidal currents are dominant and therefore fora-
minifera living in coastal seas of the East China Sea
are transported into the estuary by ooding tidal
currents (Li and Wang, 1998). Moreover, paleo-tidal
patterns at 6 and 10 kyr bp for the Yangtze estuary
estimated through numerical simulation, also showed
tidal domination and similar pattern, but with a differ-
ent magnitude (Uehara et al., 2000). Thus it can beconcluded that the estimated west to northwest paleo-
current directions were induced from the ood-tidal
currents throughout the last 12 ka.
Based on the above-mentioned conclusion, the
sedimentary paleo-environments and their paleocur-
rent directions have been estimated for each of the
nine stratigraphic units, and are as follows.
Unit 1: The reason why no absolute ow direction
could be inferred from AMS for the uvial sediments
of this unit may be due to the complicated riverine
processes characterized with a strong transverse circu-
lation (Huang et al., 1996) or coarser grain size withhigh-angle trough cross bedding.
Units 2 6: The westerly or northwesterly paleocur-
rent directions inferred from the AMS data may have
resulted from a ood tidal current-induced fabric
structure. This is evident from the sedimentological
and paleontological characteristics of these units that
indicate a strong tidal inuence during the last trans-
gression at about 116 kyr bp (Hori et al., 2001b;
Figs. 2 and 3). Furthermore, tidal currents during the
B. Liu et al. / Marine Geology 176 (2001) 175 189186
7/29/2019 paleocurrent
13/15
Holocene transgression stage were inferred to be much
stronger than those at present, with an average tidal
range of more than 4 m at that time (Li et al., 2001).
Units 7 9: These units constitute a regressive
deltaic succession. An estimated west to northwest
paleocurrent reects an onshore ow direction,
consistent with ood-tidal current. Because the
modern environment in the Yangtze estuary and
delta is tide-dominated, ood tidal currents transport
marine particles like foraminifera into the estuary for
a distance of up to 200 km (Li et al., 1983). Flood-
tidal currents during spring tide were measured
230 km upstream from the Yangtze River mouth in
the March of 1982 (the dry season), while 200 km
upstream in the August of 1983 (the ood season)
(Li and Wang, 1998). As shown by sedimentary struc-tures of wave ripples in Unit 9 (Hori et al., 2001a),
other processes such as waves action and river
currents may have also inuenced these sediments.
6.2. MS and AMS parameters
In this study, we considered that the primary AMS
represented the primary deposition nature of sedi-
ments, thus could supply us with reliable paleocurrent
directions. The above-mentioned criteria for a
primary magnetic fabric have been widely acceptedas indicator of depositional nature of sediments
(Hamilton and Rees, 1970; Hrouda, 1982; Tarling
and Hrouda, 1993). CM-97 Samples with secondary
AMS were mostly distributed on the top part of each
subcore, which may most probably resulted from
coring disturbance, similar to deep sea sediments
reported elsewhere (Abdeldayem et al., 1999). Inter-
estingly, it has been found that there were better
`primary' AMS features with the development of
bioturbation in experimental tidal at sediments,
which was induced by the re-alignment of magnetic
minerals under the action of bioturbation and pore
water (Ellwood, 1984). However, there is not so
much bioturbation in the natural CM-97 sediments
(Hori et al., 2001a,b). Therefore, we thought that the
above-mentioned criteria for primary AMS were still
valid for this study, and the primary AMS results are
also valid and reliable.
Magnetic susceptibility has been successfully used
in stratigraphic division and correlation in the Chinese
and American loesspaleosol sequences (Heller and
Liu, 1982; Heller et al., 1991; Bloemendal et al.,
1995; Grimley et al., 1998), continental shelf sedi-
ments (Arai et al., 1997), deep-sea sediments (Barthes
et al., 1999), and in Paleozoic marine sequences
(Crick et al., 1997; Ellwood et al., 1999). In this
study, we found that not only magnetic susceptibility,
but also other AMS parameters could be used in
stratigraphic division. K and other AMS parameters
reect the grain size and perhaps mineral assemblage
as well. This was inferred from the fact that samples
where demagnetization behavior showed very low
intensities (Fig. 4c) mainly distributed in the strati-
graphic Unit 1 (Fig. 2), but quite few such samples
occurred in other stratigraphic units. This may extend
the use of AMS in stratigraphic division and maybe
also in stratigraphic correlation in addition to paleo-current determination.
Remarkable similarity in downhole changes of Pjand Findicates that the AMS of the sediments studied
here was induced by magnetic foliation that probably
formed due to compaction processes, a mechanism
that has been recognized in deep-sea sediments
(Ellwood, 1979; Abdeldayem et al., 1999).
The AMS analysis was applied to tidal-dominated
coastal sediments of late PleistoceneHolocene
incised-valley ll from the Yangtze delta, China
(borehole CM-97). Based on the comparison between
the paleocurrent directions from the primary sedimen-
tary structures and the in situ AMS data, we have
shown that AMS analysis is applicable in determining
paleocurrent directions for these sediments.
Throughout borehole CM-97 sediments, the domi-
nated paleocurrent directions were westerly to north-
westerly for transgressive estuarine and regressive
deltaic sediments. This is estimated to be mainly
due to ood tidal currents in the paleo- and modern
Yangtze (Changjiang) estuary. No absolute paleocur-
rent direction could be inferred from AMS for the
uvial coarse sediments. These results indicate that
the sedimentary paleoenvironment since about
11 kyr bp was a ood-tide dominated estuary or delta.
In addition, we have demonstrated that AMS can be
useful in stratigraphic division and may also be in
stratigraphic correlation. Our work indicates that in
B. Liu et al. / Marine Geology 176 (2001) 175189 187
7/29/2019 paleocurrent
14/15
addition to its long known credibility as a sensible
paleocurrent indicator, AMS may also be useful as a
stratigraphic marker.
The authors are grateful to staff of Marine Geology
Laboratory of Tongji University, Shanghai, China for
their great help during the coring and subsampling.
The authors are also grateful to Dr Brooks B. Ellwood
and an anonymous reviewer whose critical review has
greatly improved the manuscript. This research is
funded by the Global Environment Research Fund
of the Environment Agency of Japan. B. Liu would
like to express his special thanks to STA/JST/JISTECof Japan that made it possible for him to conduct this
study under the STA Fellowship. Demagnetization
data analysis was carried out by using Dr R. Enkin 0s
PC program (http://www.pgc.emr.ca/tectonic/
enkin.htm). Fig. 1 was created by using GMT
(http://www.soest.hawaii.edu/gmt/).
Abdeldayem, A.L., Yamazaki, T., Ikehara, K., 1999. Magnetic
susceptibility anisotropy and remanence of some deep-seasediments of the Tokai Basin. In: Yuasa, M. (Ed.), Marine
Geological Investigations of the Tokai Offshore Area.
Geological Survey of Japan, Japan, pp. 127146.
Allen, J.R.L., 1984. Sedimentary Structures Their Character and
Physical Basis. Elsevier, Amsterdam.
Arai, K., Sakai, H., Konishi, K., 1997. High-resolution rock-
magnetic variability in shallow marine sediment: a sensitive
paleoclimatic metronome. Sedim. Geol. 110, 723.
Balsley, J.R., Buddington, A.F., 1960. Magnetic susceptibility
anisotropy and fabric of some Adirondack granites and
orthogneisses. Am. J. Sci. 258, 6 20.
Barthes, V., Pozzi, J.P., Vibert-Charbonnel, P., Thibal, J., Melieres,
M.A., 1999. High-resolution chronostratigraphy from downhole
susceptibility logging tuned by palaeoclimatic orbital frequen-
cies. Earth Planet. Sci. Lett. 165, 97 116.
Bloemendal, J., Liu, X.M., Rolph, T.C., 1995. Correlation of the
magnetic susceptibility stratigraphy of Chinese loess and the
marine oxygen isotope record: chronological and palaeoclimatic
implications. Earth Planet. Sci. Lett. 131, 371380.
Chen, Z., 1999. Geomorphology and coastline change of the Lower
Yangtze delta plain, China. In: Miller, A.J., Gupta, A. (Eds.),
Varieties of Fluvial Form. Wiley, New York, pp. 427443.
Chen, Z., Chen, Z., 1997. Quaternary stratigraphy and trace-element
indices of the Yangtze delta, Eastern China, with special refer-
ence to marine transgressions. Quat. Res. 47, 181191.
Chen, J., Shen, H., Yu, C., 1988. Process of Dynamics and Geomor-
phology of the Changjiang Estuary. Shanghai Scientic and
Techical Publishers, Shanghai.
Chen, Z., Song, B., Wang, Z., Cai, Y., 2000. Late Quaternary evolu-
tion of the sub-aqueous Yangtze delta, China: sedimentation,stratigraphy, palynology, and deformation. Mar. Geol. 162,
423441.
Crick, R.E., Ellwood, B.B., El Hassani, A., Feist, R., Hladil, J.,
1997. Magnetosusceptibility event and cyclostratigraphy
(MSEC) of the EifelianGivetian GSSP and associated
boundary sequences in north Africa and Europe. Episodes 20,
167175.
Ellwood, B.B., 1979. Particle occulation: one possible control on
the magnetization of deep-sea sediments. Geophys. Res. Lett. 6,
237240.
Ellwood, B.B., 1980. Application of the anisotropy of magnetic
susceptibility method as a indicator of bottom-water ow
direction. Mar. Geol. 34, M83M90.
Ellwood, B.B., 1984. Bioturbation: minimal effects on the magneticfabric of some natural and experimental sediments. Earth
Planet. Sci. Lett. 67, 367376.
Ellwood, B.B., Ledbetter, M.T., 1977. Antarctic bottom water
uctuations in the Vema Channel: effects of velocity changes
on particle alignment and size. Earth Planet. Sci. Lett. 35,
189198.
Ellwood, B.B., Ledbetter, M.T., 1979. Paleocurrent indicators in
deep-sea sediment. Science 203, 13351337.
Ellwood, B.B., Ledbetter, M.T., Johnson, D.A., 1979. Sedimentary
fabric: a tool to delineate a high-velocity zone within a deep
Western Indian Ocean Bottom Current. Mar. Geol. 33,
M51M55.
Ellwood, B.B., Hrouda, F., Wagner, J., 1988. Symposia on magneticfabrics: introductory comments. Phys. Earth Planet. Inter. 51,
249252.
Ellwood, B.B., Crick, R.E., El Hassani, A., 1999. The magneto-
susceptibility event and cyclostratigraphy (MSEC) method
used in geological correlation of Devonian rocks from Anti-
Atlas Morocco. AAPG Bull. 83, 1119 1134.
Granar, L., 1958. Magnetic measurements on Swedish varved
sediments. Arkiv. f. Geofysik 3, 1 40.
Grimley, D.A., Follmer, L.R., Donald McKay, E., 1998. Magnetic
susceptibility and mineral zonations controlled by provenance
in loess along the Illinois and Central Mississippi River valleys.
Quat. Res. 49, 2436.
Hailwood, E.A., Ding, F., 1995. Palaeomagnetic reorientation of
cores and the magnetic fabric of hydrocarbon reservoir sands.In: Turner, P., Turner, A. (Eds.), Palaeomagnetic Applications
in Hydrocarbon Exploration and Production. Geological Society
Special Publication No. 98, pp. 245258.
Hamilton, N., Rees, A.I., 1970. The use of magnetic fabric in
palaeocurrent estimation. In: Runcorn, S.K. (Ed.), Palaeo-
geophysics. Academic, London, pp. 445464.
Heller, F., Liu, T.S., 1982. Magnetostratigraphical dating of loess
deposits in China. Nature 300, 431 433.
Heller, F., Liu, X.M., Liu, T.S., Xu, T.C., 1991. Magnetic suscept-
ibility of loess in China. Earth Planet. Sci. Lett. 103, 301310.
Hori, K., Saito, Y., Zhao, Q., Cheng, X., Wang, P., Li, C., 1999.
B. Liu et al. / Marine Geology 176 (2001) 175 189188
7/29/2019 paleocurrent
15/15
Sedimentary characteristics of post-glacial deposits beneath the
Changjiang River delta. In: Saito, Y., Ikehara, K., Katayama, H.,
(Eds.), Proceedings of an International Workshop on Sediment
Transport and Storage in Coastal Sea-ocean System, Tsukuba,
Japan, March 1519, 1999, pp. 3943.Hori, K., Saito, Y., Zhao, Q., Cheng, X., Wang, P., Sato, Y., Li, C.,
2001a. Sedimentary facies and Holocene progradation rates of
the Changjiang (Yangtze) delta, China. Geomorphology (in
press).
Hori, K., Saito, Y., Zhao, Q., Cheng, X., Wang, P., Sato, Y., Li, C.,
2001b. Sedimentary facies of the tide-dominated paleo-
Changjiang (Yangtze) estuary during the last transgression.
Mar. Geol. 177, in press.
Hrouda, F., 1982. Magnetic anisotropy of rocks and its application
in geology and geophysics. Geophys. Surv. 5, 37 82.
Huang, H., Tang, B., Yang, W., Yuan, Y., Zhou, C., Zan, Y., 1996.
Sedimentary Geology of the Yangtze Delta. China Geological
Publishing House, Beijing (in Chinese).
Ising, G., 1942. On the magnetic properties of varved clay. Ark.Mat. Astr. Phys. 29, 137.
Jelinek, V., 1981. Characterization of the magnetic fabric of rocks.
Tectonophysics 79, T63T67.
Kirschvink, J.L., 1980. The least-squares line and plane and the
analysis of palaeomagnetic data. Geophys. J. R. Astron. Soc.
62, 699718.
Ledbetter, M.T., Ellwood, B.B., 1980. Spatial and temporal changes
in bottom-water velocity and direction from analysis of particle
size and alignment in deep-sea sediment. Mar. Geol. 38,
245261.
Li, C., Wang, P., 1998. Late Quaternary Stratigraphy of the
Changjiang Delta. China Science Press, Beijing (in Chinese).
Li, C., Li, P., Cheng, X., 1983. The inuence of marine factors on
sedimentary characteristics of Yangtze River channel below
Zhenjiang. Acta Geographica Sinica 38, 128140 (in Chinese,
with English abstract).
Li, C., Chen, Q., Zhang, J., Yang, S., Fan, D., 2000. Stratigraphy
and paleoenvironmental changes in the Yangtze Delta during
the Late Quaternary. J. Asian Earth Sci. 18, 453 469.
Li, C., Fan, D., Hori, K., Zhao, Q., Saito, Y., Cheng, X., 2001. Some
problems on postglacial transgressionregression and environ-
mental change in the Changjiang delta. J. Nanjing Univ. (in
Chinese, with English abstract) (in press).
Liu, K.-B., Sun, S., Jiang, X., 1992. Environmental change in the
Yangtze River delta since 12,000 years bp. Quat. Res. 38,
3245.
Milliman, J.D., Meade, R.H., 1983. World-wide delivery of riversediments to the ocean. J. Geol. 91, 121.
Nagata, T., 1961. Rock Magnetism, 2nd ed. Maruzen, Tokyo.
Piper, J.D.A., Elliot, M.T., Kneller, B.C., 1996. Anisotropy of
magnetic susceptibility in a Paleozoic ysch basin: the
Windermere Supergroup, northern England. Sedim. Geol. 106,
235258.
Reading, H.G., 1996. Sedimentary Environments: Processes, Facies
and Stratigraphy, 3rd ed. Blackwell, Osney Mead, Oxford.
Rees, A.I., 1965. The use of anisotropy of magnetic susceptibility in
the estimation of sedimentary fabric. Sedimentology 4,
257271.Rees, A.I., Woodall, W.A., 1975. The magnetic fabric of some
laboratory-deposited sediments. Earth Planet. Sci. Lett. 25,
121130.
Rees, A.I., von Rad, U., Shepard, F.P., 1968. Magnetic fabric of
sediments from the La Jolla submarine canyon and fan,
California. Mar. Geol. 6, 145178.
Reineck, H.-E., Singh, I.B., 1980. Depositional Sedimentary
Environments (Second, revised and updated edition). Springer,
Berlin.
Rolph, T.C., Shaw, J., Harper, T.R., Hagan, J.T., 1995. Viscous
remanent magnetization: a tool for orientation of drill cores.
In: Turner, P., Turner, A. (Eds.), Palaeomagnetic Applications
in Hydrocarbon Exploration and Production. Geological Society
London Special Publ. 98, 239243.
Saito, Y., 1998. Sea levels of the Last Glacial in the East China Sea
continental shelf. Quat. Res. (Jpn) 37, 235242 (in Japanese,
with English abstract).
Schieber, J., Ellwood, B.B., 1993. Determination of basinwide
paleocurrent patterns in a shale succession from anisotropy of
magnetic susceptibility (AMS): a case study of the mid-
Proterozoic Newland formation, Montana. J. Sedim. Petrol.
63, 874880.
Stacey, F.D., Joplin, G., Lindsay, J., 1960. Magnetic anisotropy and
fabric of some foliated rocks from SE Australia. Geophys. Pura.
Appl. 47, 3040.
Taira, A., Lienert, B.R., 1979. The comparative reliability of
magnetic, photometric and microscopic methods of determiningthe orientations of sedimentary grains. J. Sedim. Petrol. 49,
759772.
Tarling, D.H., Hrouda, F., 1993. The Magnetic Anisotropy of
Rocks. Chapman & Hall, London.
Tarling, D.H., Shi, H., 1995. Magnetic anisotropy of borehole core
samples. In: Turner, P., Turner, A. (Eds.), Palaeomagnetic
Applications in Hydrocarbon Exploration and Production.
Geological Society London Special Publ. 98, 273280.
Uehara, K., Saito, Y., Hori, K., 2000. Numerical simulation of
paleotides around the Changjiang estuary at 10 and 6 kyr bp.
Abstracts of the International Workshop on Deltas: Their
Dynamics, Facies and Sequences, 1617 March, 2000,
Tsukuba, Japan, p. 12
Zhu, Y., Li, C., Zeng, C., 1979. On Late Pleistocene low stand of sea
level in East China Sea. Chin. Sci. Bull. 24, 317320 (in
Chinese).
Zijderveld, J., 1976. AC Demagnetization of rocks: analysis of
results. In: Collinson, D. (Ed.), Methods in Paleomagnetism.
Elsevier, New York, pp. 254286.
B. Liu et al. / Marine Geology 176 (2001) 175189 189