alaeoecology 247 (2007) 65–73www.elsevier.com/locate/palaeo

Palaeogeography, Palaeoclimatology, P

Diagenetic modification of magnetic properties observed in a pistoncore (MD01–2407) from the Oki Ridge, Japan Sea

Akira Hayashida a,⁎, So Hattori b,1, Hirokuni Oda c

a Department of Environmental Systems Science, Doshisha University, Kyotanabe, Kyoto 610–0321, Japanb Graduate School of Engineering, Doshisha University Kyotanabe, Kyoto 610–0321, Japan

c Geological Survey of Japan, AIST, Tsukuba 305–8567, Japan

Received 9 May 2005; accepted 9 November 2006

Abstract

We investigated the magnetic properties of u-channel samples from a giant piston core (IMAGES MD01–2407), in order toobtain high-resolution records of past geomagnetic field and paleoenvironmental changes in the Japan Sea. Measurements ofnatural remanent magnetization (NRM) show occurrence of anomalous magnetic directions at several horizons, which mayrepresent geomagnetic polarity excursions in the Brunhes Chron. However, most of them occur close to the lithological boundaryincluding tephra layers, suggesting that these features might have been caused by deformation of core sediments. Intensity of NRMnormalized by artificially induced remanence shows no significant correlation with a standard paleointensity record. This isprobably caused by a variation of lithology including volcanic ash layers and effect of the diagenetic changes of magnetic minerals.Down-core variations of magnetic properties in the topmost part depict in detail the process of dissolution and alteration ofmagnetic minerals during reductive diagenesis, which is to be expected for hemipelagic sediments characterized by a highsedimentation rate and a high organic matter content. Magnetic concentration parameters including low-field susceptibility,anhysteretic remanent magnetization and isothermal remanent magnetization (IRM) indicate that a loss of fine-grained magnetiteoccurs from 0.6 to 1.2 mbsf. While remanence intensities below the level of the early diagenesis are one or two orders of magnitudelower than those in the topmost part, most of the thinly-laminated (TL) dark layers showed rather enhanced magnetic intensities.Typically, TL2 shows a marked peak at the top with a gradual decrease towards the bottom. Variation of S-ratio, which gives theproportion of a low-coercivity component in saturation IRM, suggests that low-coercivity and fine-grained ferromagnetic mineralsare dominant. These data suggest new magnetic minerals formed during deposition of the TL layers, which may be analogous tothe diagenetic magnetic enhancement known in sapropels of the Mediterranean Sea.© 2007 Elsevier B.V. All rights reserved.

Keywords: Environmental magnetism; Magnetite diagenesis; Hemipelagic sediments; Japan Sea

⁎ Corresponding author.E-mail address: ahay@mail.doshisha.ac.jp (A. Hayashida).

1 Now at Obayashi Corporation, Tokyo.

0031-0182/$ - see front matter © 2007 Elsevier B.V. All rights reserved.doi:10.1016/j.palaeo.2006.11.022

1. Introduction

The Japan Sea is a marginal basin of the Asiancontinent, and is connected to the East China Sea, thePacific Ocean and the Okhotsk Sea. The paleoceano-graphy of the Japan Sea has been studied using Neogene

66 A. Hayashida et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 247 (2007) 65–73

strata exposed on the Japanese Islands (Kitamura et al.,2001) and the sub-bottom sediments recovered by pistoncoring (Oba et al., 1991) and by Legs 127 and 128 of theOcean Drilling Program (Tada, 1994; Tada et al., 1999).These studies revealed that the paleoceanographiccondition of the Japan Sea has responded sensitively toclimate changes through glacial eustatic sea-levelchanges and evolution of the Asian monsoon. Study ofremanent magnetization of sediments provides chrono-logical data on the history of the geomagnetic field, andthe magnetic properties of the sediments reflect theorigin, transportation, deposition and post-depositionalprocesses of the sediments which often correlate wellwith past climate changes (Evans and Heller, 2003). It istherefore expected that paleomagnetic and rock magnet-ic investigations can contribute to the delineation of theenvironmental history of the Japan Sea.

Since the pioneering work by Kobayashi andNomura (1972), several authors have studied thepaleomagnetic and rock magnetic properties of theJapan Sea sediments (Hamano and Krumsiek, 1992;Torii et al., 1992; Vigliotti, 1997; Vigliotti et al., 1999;Yamazaki et al., 2003). Among them, Hamano andKrumsiek (1992) demonstrated that magnetic polarity

Fig. 1. Map showing the location of the coring sites during the IMAGES VII cAkita). Also shown are locations of the core GH98–1232 (Yamazaki et al.,

records of the Pliocene–Pleistocene sediments at OceanDrilling Program (ODP) Sites 794, 797 and 797 arepreserved and even suggested that paleointensityrecords of the geomagnetic field could be obtained.Other investigations suggested that diagenetic modifi-cation of magnetic mineral assemblages might affectrecords of remanent magnetization. Yamazaki et al.(2003) showed that paleomagnetic secular variations ininclination for the last 30 kyr are preserved in a pistoncore (GH98–1232) from the northeastern Japan Sea. Tounderstand the preservation of paleomagnetic andpaleoclimatic signals in Japan Sea sediments, it isnecessary to investigate the magnetic properties atvarious locations and for longer time intervals.

During the IMAGES VII cruise in 2001, a giantpiston-core MD01–2407 was recovered from 932 mwater depth on the Oki Ridge in the Japan Sea(37°04.01′N, 134°42.18′E; Fig. 1). This core, 55.28 min length, is mainly composed of alternating layers ofhomogeneous light colored silty clay and laminated orbioturbated dark colored silty clay. Several volcanic ashlayers and characteristic thinly-laminated (TL) darkcolor layers, which are called TL layers (Tada et al.,1999), also exist. In this paper, we describe the magnetic

ruise in the Japan Sea (MD01–2407 at Oki Ridge and MD01–2408 off2003) and Sites 794 to 799 drilled in ODP Legs 127 and 128.

67A. Hayashida et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 247 (2007) 65–73

properties of u-channel samples taken from this core,and examine the diagenetic modification of magneticminerals in the surficial sediments and the TL layers.

2. Samples and methods

Core MD01–2407 was cut into thirty seven 1.5 m-long sections, which were split lengthwise into work-ing and archive halves on board the R/V Marion-Dufresne. For magnetic measurements, we collectedu-channel samples from the working halves onshore.The u-channels are 1.5 m long with a 2×2 cm2 cross-section and were pushed into the sediment of the splitface of each section, and removed from the core usinga fishing line. The u-channel was then sealed with aclip-on lid and a plastic film to prevent dehydration ofthe sediment.

Fig. 2. Magnetic intensity, inclination and relative declination of the NRM bmiddle panel shows the inclination of the geocentric axial dipole field expec

Magnetic measurements of the thirty seven u-channels were made at 1-cm intervals using a 2GEnterprises Model 755R cryogenic magnetometer. Themagnetometer has a narrow-access with high-resolutionpick-up coils that produce a response function that isabout 5 cm wide. As we have not deconvoluted the pass-through data, the results presented here should be re-garded as a moving average over about a 5 cm interval.We divided the outputs of the three pick-up coils (in emu)by the effective volume, which is the amount of thesediments sensed by each pick-up coil (Tauxe and Wu,1990), to calculate declination, inclination and volumet-ric intensity (in A/m). We also made pass-throughmeasurements of low-field (0.1 mT) magnetic suscep-tibility of the u-channels using a Bartington Instrumentssusceptibility meter with a core-logging sensor (MS2C)operated at a frequency of 0.47 kHz.

efore and after AF demagnetization at 10 mT. The dashed line in theted at the coring site.

68 A. Hayashida et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 247 (2007) 65–73

Our experimental procedure was as follows: (1)measurements of low-field magnetic susceptibility (k);(2) measurement of NRM with progressive alternatingfield (AF) demagnetization at levels of 0, 5, 10, 15, 20, 25,30 and 40 mT; (3) acquisition of anhysteretic remanentmagnetization (ARM) in a 100mTpeakAFwith a 0.1mTsteady field, and AF demagnetization at 0, 10, 15, 20, 30and 40 mT; (4) acquisition of isothermal remanentmagnetization (IRM) in a direct magnetic field at100 mT, and AF demagnetization at 0, 10, 15, 20, 30and 40 mT; and (5) acquisition of IRM at 1 T and back-field IRMs at 0.1 Tand 0.3 T. Assuming the IRM at 1 T isa saturation IRM (SIRM), we calculated the S-ratio(S−0.1 Tand S−0.3 T) using the definition given inKing andChannell (1991):

S ¼ −back−field IRM=IRM:

The S-ratios, ranging from −1 to 1, measure theproportion of low-coercivity magnetic minerals to highercoercivity grains. The IRMwas imparted by sliding the u-channels in a static field produced by an electric magnet.We also obtained ARM susceptibility (kARM) by dividingthe ARM intensity by the biased field (0.1 mT).

3. Results

Down-core plots of the NRM intensities, inclinationsand relative declinations measured on the u-channelsbefore and after AF demagnetization (at 10 mT) areshown in Fig. 2. In the topmost part above 5 mbsf

Fig. 3. Orthogonal projection of AF demagnetization data from the horizonssymbols indicate projection on the vertical and horizontal planes, respective

(meters below sea floor), the NRM intensity was about1×10−2 A/m, and the AF demagnetization at 10 mT didnot cause significant decay of the magnetic intensities.Below 5 mbsf, the NRM intensity was generally of theorder of 10−3 or 10−4 A/m and decreased to around1×10−4 A/m by the AF demagnetization at 40 mT.Examples of orthogonal projections of the AF demag-netization data are shown in Fig. 3.

The inclinations are mostly plotted between 40° and60°. This result is generally concordant with theinclination of the geocentric axial dipole field (56.5°)expected at the site latitude (37.1°). The declinationsshow significant stepwise changes at 24.0, 32.5 and43.5 mbsf as shown in Fig. 2. These horizonscorrespond to the intervals of core gaps or fluidizedsediments probably due to gas expansion. It is likely thatthe core segments were relatively rotated during thesampling or cutting onboard. The declinations are alsodisturbed in the topmost part of the core, suggesting thatsignificant rotation of the piston corer occurred duringcore penetration or recovery. Although this core seemsto be partly distorted, the magnetic directions in generalshow normal polarity reflecting deposition during theBrunhes Normal Chron.

Fig. 4 shows variations of magnetic susceptibility andintensities of ARM and IRM. While these data aregenerally regarded as magnetic concentration para-meters, the magnetic susceptibility record does notresemble the ARM and IRM in several intervals. Themagnetic susceptibility data of our samples is dominatedby paramagnetic minerals, while the ARM and IRM

at 3.5, 25.0 and 47.5 mbsf in the MD01–2407 core. Open and closedly.

Fig. 4. Magnetic susceptibility, ARM and IRM intensities and S-ratio. Horizontal lines show horizons of the tephra deposits described by Kido et al.(2007-this issue) and Chun and Ikehara (personal communication).

69A. Hayashida et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 247 (2007) 65–73

represent the concentration of ferromagnetic minerals,which are the carriers of the remanent magnetization. Inthe topmost part of the core, these parameters show asharp down-core decrease of about one order ofmagnitude, followed by significant swings down toabout 5 mbsf. The variation of magnetic mineralconcentration parameters in this interval is reflected inthe NRM intensity shown in Fig. 2 and also correlatedwith the variation of S-ratios (Fig. 4). The S−0.3 T ratio isaround 0.9, suggesting that magnetic grains in thesediments are dominated by low-coercivity minerals(e.g., magnetite or titanomagnetite) rather than antifer-romagnetic minerals (e.g., hematite). The more pro-nounced change in the S−0.1 T suggests that ferrimagneticminerals of different composition or grain-size aredominant in intervals of increased ARM and IRM.

Below 5 mbsf, the ARM and IRM are distributedmostly around decreased values accompanying numerousspiky peaks, which are also observed in the susceptibilityrecord. The abrupt increases of the magnetic concentra-tion parameters occur mostly associated with changes ofS-ratios, suggesting episodic inputs of magnetic particlesof distinct mineralogy or composition, such as derived byvolcaniclastic deposits. Indeed, most of the spikes ofARM and IRM occur at horizons corresponding to thetephra deposits noted in the core description (Kido et al.,2007-this issue; Chun and Ikehara, personal communica-tion), as shown in Fig. 4.

Fig. 5 shows variations of the magnetic concentrationparameters (k, ARM and IRM) and the grain-sizeindicators (kARM/k, SIRM/k and S-ratios) for the interval0–10 mbsf, corresponding to the last 112 ka. This

Fig. 5. Plots of magnetic susceptibility, ARM and IRM intensities, kARM/k, SIRM/k and S-ratios for the interval 0–10 mbsf. Horizontal lines showhorizons of TL layers (shaded) and tephra deposits (hatched). Also shown are lightness of the sediments (L⁎, shipboard data) and concentration of Sand Fe (Kido et al., 2007-this issue).

70 A. Hayashida et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 247 (2007) 65–73

diagram also shows variations of lightness of thesediments (L⁎, shipboard data) and concentrations ofsulfur and total iron (Kido et al., 2007-this issue).

4. Discussion

4.1. Paleomagnetic records

Paleomagnetic studies of Japan Sea sediments havebeen made using piston-core samples (Yamazaki et al.,2003) or discrete samples from ODP cores obtained by

advanced piston coring (Hamano and Krumsiek, 1992).The pass-through measurement of u-channel samplesfrom the IMAGES core was a unique attempt to assess acontinuous high-resolution record of Middle Pleistocenepaleomagnetic and rock magnetic signals in Japan Seasediments. However, the core sediments seeminglysuffered distortion during core recovery, which probablycaused disturbance of the magnetic directions (Fig. 2).We should be cautious, therefore, in interpretingfluctuations of the paleomagnetic records. It was alsonoted that the magnetic mineral concentrations show

71A. Hayashida et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 247 (2007) 65–73

significant fluctuations probably because of diageneticmodification of magnetic minerals and episodic input ofvolcanic ash. Such heterogeneity of the magneticproperties makes it impossible to obtain a reliableestimate of a paleointensity record by normalizing theNRM intensity with magnetic concentration parameters.

Kido et al. (2007-this issue) constructed an age modelfor MD01–2407, which covers the last 640 ka, bycomparing the δ18O curve of planktonic foraminifera withthe SPECMAP δ18O record. Previous paleomagneticstudies (Langereis et al., 1997; Nowaczyk et al., 1997;Lund et al., 2001; Oda et al., 2002; Kent et al., 2002)suggested that several geomagnetic polarity excursionsoccurred during this time interval, including theLaschamp (41 ka), Blake (120 ka), Lake Baikal/IcelandBasin (180–190 ka), and Jamaica/Pringle Falls (200–220 ka) excursions. It was also suggested that most of theexcursions occur in association with geomagnetic fieldintensity minima (Guyodo and Valet, 1999). Except for alarge fluctuation caused by core disturbance in thetopmost part, the NRM directions (both declination andinclination) show an abrupt fluctuation at 4.15mbsf in theMD01–2407 core. Since this horizon corresponds to atephra layer of about 10 cm thick (the AT volcanic ashlayer dated at 29 ka; Miyairi et al., 2004), the anomalousNRM direction was probably caused by mechanicaldeformation during sampling. Thus, clear evidence ofpolarity excursions were not recognized in this core,although the sedimentation rate ranges from 5.1 to15.0 cm/kyr. We suggest that diagenetic modification ofmagnetic minerals obscured detailed features of thepaleomagnetic record. There is also no apparent correla-tion between the normalized intensity data and thestandard paleointensity record (Sint800; Guyodo andVallet, 1999).

4.2. Alteration of magnetic minerals

The diagenetic change of the sediments is observedin down-core variations of the magnetic properties in thetopmost part of the core (Fig. 5). The depletion ofmagnetic minerals is indicated by a decrease insusceptibility, ARM and SIRM from 0.6 to 1.2 mbsf(about 1.1–2.8 ka). Most likely, this represents theeffects of reductive diagenesis, the process that has beendocumented in numerous studies (Leslie et al., 1990a,b;Karlin, 1990a,b; Bloemendal et al., 1992, 1993). Thegradual increase of sulfur content and the decrease oftotal iron content at the core top suggest the occurrenceof iron oxide reduction and sulfate reduction.

As shown in Fig. 5, the intensity drop accompanies asignificant decrease of the S−0.1 T ratio, suggesting the

loss of ferrimagnetic minerals. It is also noticeable that theloss of ARM at 0.6–1.0 mbsf proceeds from the decreaseof susceptibility and SIRM, as represented by a decreaseof kARM/k between 0.6 and 1.2 mbsf. These variations canbe explained by the diagenetic process that fine-grainedferrimagnetic components are selectively dissolved be-cause of its relatively large surface area to volume ratio(Karlin et al., 1990a). Such phenomena are commonlyobserved in anoxic interstitial micro-environments asso-ciated with high concentrations of organic matter in thesediment. Yamazaki et al. (2003) reported that in theGH98–1232 core from 838 m water depth in the north-eastern Japan Sea (Fig. 1), dissolution of magnetic min-erals occurs at a core depth between 1.2 and 1.6 m, whichcorresponds to about 5–8 ka. The organic carbon contentof the MD01–2407 core, estimated from lightness (L⁎) ofthe sediments (Kido et al., 2007-this issue) ranges from 1to 5% in the topmost 3 m and the variation is comparableto that of the GH98–1232 core. Thus, our data show that asimilar diagenetic change has occurred at a similar sub-bottom depth in the sediments of the Oki Ridge.

As shown in Fig. 5, the ARM and SIRM intensitiesshow remarkable peaks at several horizons below theearly diagenesis zone. While some of them representtephra layers, other peaks occur close to the TL layersdescribed by Watanabe et al. (2007-this issue).Typically, TL2 at 3.3–3.7 mbsf shows a marked peakat the top with a gradual decrease towards the bottom.Variations of the S-ratio suggest that low-coercivity andfine-grained ferromagnetic minerals are dominant inthese intervals. Kido et al. (2007-this issue) showed thatthe dark layers of the glacial periods, which wereformed under water column stratification and lowersurface productivity, are characterized by relativelylow organic carbon contents and better preservationof calcium carbonate, if compared with dark layers ofinterglacial stages. It is therefore suggested that thefine ferromagnetic minerals in the TL layers might bepreserved because of the decrease of organic matterdecomposition.

Alternatively, the high magnetic concentration in theTL layers may suggest formation of new magneticminerals during deposition, which may be analogous tothe diagenetic magnetic enhancement known in sapro-pels in the Mediterranean Sea (e.g., Roberts et al., 1999;Passier et al., 2001). The magnetic intensity enhance-ment could be explained by formation of a ferrimagneticiron sulfide in the reducing environment or byprecipitation of iron oxide because of reoxygenation ofthe bottom water after the deposition of the TL layer. Liuet al. (2004) reported the occurrence of fine-grainedauthigenic greigite in the Holocene continental shelf

72 A. Hayashida et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 247 (2007) 65–73

sediments from the Korean Strait. Fine-grained greigitemight have formed and have been preserved in the TLlayer with significantly high sulfur content, although thepresence of iron sulfide should be examined in moredetailed rock magnetic studies.

5. Summary

We described the magnetic properties of u-channelsamples of the giant piston-core MD01–2407 recoveredfrom the Oki Ridge in the Japan Sea. Although the NRMmeasurements were expected to provide a high-resolution record of geomagnetic polarity excursionsand paleointensity during the late Brunhes Chron, thepaleomagnetic record was affected by deformation ofthe core, significant lithological changes and diageneticchanges of magnetic minerals. Down-core variations ofmagnetic properties in the topmost part depicted theprocess of dissolution and alteration of magneticminerals during reductive diagenesis, which is to beexpected for hemipelagic sediments characterized by ahigh sedimentation rate and a high organic mattercontent. As a result, the magnetic intensity below thelevel of the early diagenesis was one or two orders ofmagnitude lower than in the topmost part of the core, butmost of the TL layers showed rather enhanced magneticintensities. This finding suggests new magnetic mineralsformed during deposition of the TL layers, which maybe analogous to the diagenetic magnetic enhancementknown from sapropels in the Mediterranean Sea.

Acknowledgements

We sincerely thank Hodaka Kawahata and TadamichiOba and the crew and staff of the IMAGES VII cruise forproviding the MD01–2407 core samples. The samplingof u-channelsweremadewith kind support byRyuji Tada,Ken Ikehara, Kazuho Fujine, Kana Nagashima, YoshikiKido, Ikue Minami, Satoko Watanabe, Takuya Itaki,Mitsuteru Suyehiro and Masaharu Fukuoka. YoshikiKido and Ken Ikehara also provided the geochemical andtephrostratigraphic data. We also thank Carl Richter,Masayuki Torii, Richard Jordan and Ryuji Tada for theirconstructive review, which substantially improved themanuscript.

References

Bloemendal, J., King, J.W., Hall, F.R., Doh, S.-J., 1992. Rockmagnetism of Late Neogene and Pleistocene deep-sea sediments:relationship to sediment source, diagenetic processes, andsediment lithology. J. Geophys. Res. 97, 4361–4375.

Bloemendal, J., King, J.W., Hunt, A., deMenocal, P.B., Hayashida, A.,1993. Origin of the sedimentary magnetic record at Ocean DrillingProgram sites on the Owen Ridge, western Arabian Sea. J. Geophys.Res. 98, 4199–4219.

Evans, M.E., Heller, F., 2003. Environmental magnetism. Principlesand Applications of Envirommagnetics. Academic Press, London.299 pp.

Guyodo, Y., Valet, J.P., 1999. Global changes in intensity of the Earth'smagnetic field during the past 800 kyr. Nature 399, 249–252.

Hamano, Y., Krumsiek, K.A.O., 1992. Magnetic intensity variations insediment cores from Leg 128. Proc. ODP Sci. Results 127/128,959–967.

Karlin, R., 1990a. Magnetite diagenesis in marine sediments from theOregon continental margin. J. Geophys. Res. 95, 4405–4419.

Karlin, R., 1990b. Magnetic mineral diagenesis in suboxic sedimentsat Bettis Site W-N, NE Pacific Ocean. J. Geophys. Res. 95,4421–4436.

Kent, D.V., Hemming, S.R., Turrin, B.D., 2002. Laschamp excursionat Mono Lake ? Earth Planet Sci. Lett. 197, 151–164.

Kido, Y., Minami, I., Tada, R., Fujine, K., Irino, T., Ikehara, K., Chun, J.H.,2007. Orbital-scale stratigraphy and high-resolution analysis ofbiogenic components and deep-water oxygenation conditions in theJapan Sea during the last 640 kyr. Palaeogeogr. Palaeoclimatol.Palaeoecol. 247, 32–49 (this issue). doi:10.1016/j.palaeo.2006.11.020.

King, J.W., Channell, J.E.T., 1991. Sedimentary magnetism, environ-mental magnetism, and magnetostratigraphy. Rev. Geophys. 29(Suppl.), 358–370.

Kitamura, A., Takano, O., Takata, H., Omote, H., 2001. Late Pliocene–early Pleistocene paleoceanographic evolution of the Sea of Japan.Palaeogeogr. Palaeoclimatol. Palaeoecol. 172, 81–98.

Kobayashi, K., Nomura, M., 1972. Iron sulfides in the sediment coresfrom the Sea of Japan and their geophysical implications. EarthPlanet. Sci. Lett. 16, 200–208.

Langereis, C.G., Dekkers, M.J., deLange, G.D., Paterne, M., vanSant-voort, P.J.M., 1997. Magnetostratigraphy and astronomical calibra-tion of the last 1.1Myr from an easternMediterranean piston core anddating of short events in the Brunhes. Geophys. J. Int. 129, 75–94.

Leslie, B.W., Hammond, D.E., Berelson, W.M., Lund, S.P., 1990a.Diagenesis in anoxic sediments from the California continentalborderland and its influence on iron, sulfur, and magnetitebehavior. J. Geophys. Res. 95, 4453–4470.

Leslie, B.W., Lund, S.P., Hammond, D.E., 1990b. Rock magneticevidence for the dissolution and authigenic growth of magneticminerals within anoxic marine sediments of the Californiacontinental borderland. J. Geophys. Res. 95, 4437–4452.

Liu, J., Rixiang, Z., Roberts, A.P., Shaoquan, L., Chang, J.H., 2004.High-resolution analysis of early diagenetic effects on magneticminerals in post–middle-Holocene continental shelf sedimentsfrom the Korea Strait. J. Geophys. Res. 109, B03103. doi:10.1029/2003JB002813.

Lund, S.P., Williams, T., Acton, G.D., Clement, B., Okada, M., 2001.Brunhes Chron magnetic field excursions recovered from Leg 172sediments. In: Keigwin, L.D., Rio, D., Acton, G.D., Arnold, E.(Eds.), Proc. ODP, Sci. Results, vol. 172. [Online]. Available fromWorld Wide Web: http://www-odp.tamu.edu/publications/172_SR/chap_10/chap_10.htm. [Cited 2006–01–17].

Miyairi, Y., Yoshida, K., Miyazaki, Y., Matsuzaki, H., Kaneoka, I.,2004. Improved 14C dating of a tephra layer (AT tephra, Japan)using AMS on selected organic fractions. Nucl. Instrum. Methodsin Phys. Res., B Beam Interact. Mater. Atoms 223–224, 555–559.

Nowaczyk, N.-R., Antonow, M., 1997. High-resolution magnetostrati-graphy of four sediment cores from the Greenland Sea − I.

73A. Hayashida et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 247 (2007) 65–73

Identifcation of the Mono Lake excursion, Laschamp and Biwa I/Jamaica geomagnetic polarity events. Geophys. J. Int. 131, 310–324.

Oba, T., Kato, M., Kitazato, H., Koizumi, I., Omura, A., Sakai, T.,Takayama, T., 1991. Paleoenvironmental changes in the Japan Seaduring the last 85,000 years. Paleoceanography 6, 499–518.

Oda, H., Nakamura, K., Ikehara, K., Nakano, T., Nishimura, M.,Khlystov, O., 2002. Paleomagnetic record from AcademicianRidge, Lake Baikal: a reversal excursion at the base of marineoxygen isotope stage 6. Earth Planet. Sci. Lett. 202, 117–132.

Passier, H.-F., de Lange, G.J., Dekkers, M.J., 2001. Magneticproperties and geochemistry of the active oxidation front and theyoungest sapropel in the eastern Mediterranean Sea. Geophys.J. Int. 145, 604–614.

Roberts, A.P., Stoner, J.S., Richter, C., 1999. Diagenetic magneticenhancement of sapropels from the eastern Mediterranean Sea.Mar. Geol. 153, 103–116.

Tada, R., 1994. Paleoceanographic evolution of the Japan Sea.Palaeogeogr. Palaeoclimatol. Palaeoecol. 108, 487–508.

Tada, R., Irino, T., Koizumi, I., 1999. Land–ocean linkages overorbital and millennial timescales recorded in late Quaternarysediments of the Japan Sea. Paleoceanography 14, 236–247.

Tauxe, L., Wu, G., 1990. Normalized remanence in sediments of thewestern equatorial Pacific: relative paleointensity of the geomag-netic field? J. Geophys. Res. 95 (12), 337–350.

Torii, M., Hayashida, A., Vigliotti, L., Wippern, J., 1992. Rockmagnetic properties of sediments from Site 797, Japan Sea. In:Tamaki, K., Suyehiro, K., Allan, J., McWilliams, M., et al. (Eds.),Proc. ODP, Sci. Results, vol. 127/128, pp. 947–957.

Vigliotti, L., 1997. Magnetic properties of light and dark sedimentlayers from the Japan sea: Diagenetic and paleoclimatic implica-tions. Quat. Sci. Rev. 16, 1093–1114.

Vigliotti, L., Capotondi, L., Torii, M., 1999. Magnetic properties ofsediments deposited in suboxic–anoxic environments: relation-ships with biological and geochemical proxies. In: Tarling, D.H.,Turner, P. (Eds.), Paleomagnetism and Diagenesis in Sediments,Geol. Soc. Lond. Spec. Publ, vol. 151, pp. 71–83.

Watanabe, S., Tada, R., Ikehara, K., Fujine, K., Kido, Y., 2007. Sedimentfabrics, oxygenation history, and circulation modes of Japan Seaduring the Late Quaternary. Palaeogeogr. Palaeoclimatol. Palaeoe-col. 247, 50–64 (this issue). doi:10.1016/j.palaeo.2006.11.021.

Yamazaki, T., Abdeldayem, A.L., Ikehara, K., 2003. Rock-magneticchanges with reduction diagenesis in Japan Sea sediments andpreservation of geomagnetic secular variation in inclinationduring the last 30,000 years. Earth Planets and Space 55,327–340.