Lithos 224–225 (2015) 324–341

Contents lists available at ScienceDirect

Lithos

j ourna l homepage: www.e lsev ie r .com/ locate / l i thos

Variations in the geochemical structure of themantle wedge beneath thenortheast Asian marginal region from pre- to post-opening of theJapan Sea

Shuang-shuang Chen a,b, Jia-qi Liu a,⁎, Sheng-sheng Chen b,c, Zheng-fu Guo a, Chun-qing Sun a,b

a Key Laboratory of Cenozoic Geology and Environment, Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing 100029, Chinab University of Chinese Academy of Sciences, Beijing 100049, Chinac Key Laboratory of Continental Collision and Plateau Uplift, Institute of Tibet Plateau Research, Chinese Academy of Sciences, Beijing 10085, China

⁎ Corresponding author. Tel.: +86 10 82998203; fax: +E-mail address: liujq@mail.iggcas.ac.cn (J. Liu).

http://dx.doi.org/10.1016/j.lithos.2015.03.0080024-4937/© 2015 Elsevier B.V. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 27 July 2014Accepted 8 March 2015Available online 19 March 2015

Keywords:BasaltDeep asthenospheric processMantle wedgeJapan Seanortheast Asia

Deep asthenospheric processes and the dynamic mechanism of magmatism in the northeast Asian marginalregion are of significant interest but have been difficult to study in detail. We completed comprehensive studieson Japan Sea basalts including petrography, whole-rock major and trace elements, Sr–Nd–Pb isotopic composi-tions, and K–Ar geochronology, then combined our results with previous research to study the tectonic evolutionof northeast Asia. The Japan Sea basalts, divided into Upper and Lower layers of Site 794 (US794, LS794), andUpper and Lower layers of Site 797 (US797, LS797) based on their stratigraphic level, belong to the calc-alkalicseries and are characterized by flat HREE with significantly positive anomalies of Ba, Sr, and Pb, and slightanomalies of Eu (δEu= 0.81–1.21). The US797 sample group has lower LREE, LILE and relatively depleted radio-active isotope ratios (87Sr/86Sr = 0.704051–0.704254; 143Nd/144Nd = 0.513035–0.513139; 206Pb/204Pb =17.758–18.223), whereas the sample groups LS797, US794, and LS794 have relatively higher incompatibleelements and slightly enriched isotopic values (87Sr/86Sr = 0.704248–0.705222; 143Nd/144Nd =0.512705–0.512917; 206Pb/204Pb = 18.077–18.377) due to the involvement of Pacific subducted fluid andsediments. K–Ar and 40Ar–39Ar geochronological data indicate age ranges of LS797, US794, and LS794 samplesof 17.7 ± 0.5–21.2± 0.8 Ma, significantly older than those of US797 (15.1 ± 0.9–17.2± 0.7 Ma). Our data com-piled with other data show sharply defined Sr–Nd isotopic variations of the Cenozoic basalts from Sikhote-Alin,the Japan Sea, the back-arc side of NE Japan and SWHokkaido, and north Hokkaido from a slightly enriched to adepleted isotopic signature at 23–24Ma, 17±2Ma (15–19Ma), 15Ma, and N12Ma, respectively, indicating thatthe upwelling asthenosphere beneath northeast Asia progressed eastward relative to the lithosphere. Weconclude that the temporal and spatial variations of basaltic magma sources in the northeast Asian marginalregion are closely associated with the extension of the Japan Sea.

© 2015 Elsevier B.V. All rights reserved.

1. Introduction

The Japan Sea, northeast Japan arc and Hokkaido make up anintegrated and typical back-arc basin and island arc system that, withthe related Sikhote-Alin region in far eastern Russia, comprise thenortheastern margin of the Eurasian continent (Fig. 1). At this margin,volcanism occurred continuously from the earlyMiocene to the Quater-nary, with obvious zonal distribution of volcanic features from thenortheast Japan arc to Hokkaido (Fukase and Shuto, 2000; Kondo et al.,2000; Nohda, 2009; Ohki et al., 1995; Okamura et al., 2005; Shutoet al., 1993, 1997, 2004, 2006). The early Miocene to middle Mioceneback-arc spreading of the Japan Sea, associated with drifting and thin-ning of subcontinental lithosphere and upwelling of the asthenosphere,

86 10 62010846.

began the process of change from a continental margin environment tothe present back-arc basin and island arc setting, and the formation ofthe northeast Asian integral tectonic framework (e.g., Garfunkel et al.,1986; Hager et al., 1983; Ishimoto et al., 2006; Otofuji et al., 1985;Shuto et al., 1993, 2004; Tatsumi et al., 1988). Concurrent and subse-quent to this back-arc spreading, themagma sources beneath the geotec-tonic unit consisting of the Sikhote-Alin region, the Japan Sea, northeastJapan, and Hokkaido changed regularly. Thus, rocks in this geotectonicunit (Sikhote-Alin, Japan Sea, northeast Japan, and Hokkaido) providean excellent opportunity to study the temporal changes in the magmasource that accompany the evolution of subduction- and extension-related volcanismduring the opening of a typical back-arc basin. Becauseof the considerable changes in the tectonic environment and magmasource from the pre- to the post-opening stage, we divided the tectonicevolution process into two stages, namely the pre-opening stage andthe syn- and post-opening stage.

Fig. 1. Indexmap showing the study area for the northeast Asianmargin (A), Sikhote-Alin, Japan Sea, NE Japan arc and Hokkaido (B) (modified from Takanashi et al., 2011). In Fig. 1B, thedotted line represents the boundary of different volcanismof NE Japan arc andHokkaido; Q.V.F. representQuaternary volcanic front in theNE Japan arc andHokkaido. EVZ, TVZ andWVZ inSW Hokkaido respectively represent an eastern volcanic zone, a transitional volcanic zone, and a western volcanic zone.

325S. Chen et al. / Lithos 224–225 (2015) 324–341

Previous studies of the magma sources of the Japan Sea, northeastJapan, Hokkaido and the Sikhote-Alin volcanics concluded that theapparently regular variations of the Sr–Nd isotopic compositions wereassociated with asthenospheric upwelling and extension of the JapanSea (e.g., Nohda, 2009; Nohda et al., 1988; Ohki et al., 1995; Okamuraet al., 2005; Pouclet et al., 1995; Sato et al., 2007; Shuto et al., 1993,2006; Tatsumi et al., 1988; Ujike and Tsuchiya, 1993). However, thesestudies focused only on basaltic magma sources in a limited area cover-ing a certain time span (Sato et al., 2007; Shuto et al., 2004, 2006), anddid not consider the Sikhote-Alin, the Japan Sea, northeast Japan, and

Hokkaido as an entire geotectonic unit to reasonably interpret theplate tectonic evolution and deep asthenospheric processes of theEurasian continent's northeast margin. Significantly, the point-in-timesequence of changes in the Sr–Nd isotopic composition of the volcanicrocks has a geographic regularity from the Sikhote-Alin region via theJapan Sea to the JapanArc, suggesting the presence of a certain directionand rhythm of asthenosphere flow beneath northeast Asia (Nohda,2009). However, given that the generation and formation of andesiteor rhyolite may be influenced by crustal contamination or magmamixing (e.g., Yamashita et al., 1999), geochemical characteristics cannot

326 S. Chen et al. / Lithos 224–225 (2015) 324–341

be the sole tool used to constrain the nature of the magma sources ofthese rocks. Thus, here we present a systemic investigation of basaltsfrom the Japan Sea in terms of detailed petrology, major and traceelements, Sr–Nd–Pb isotopic composition and geochronology.We com-pare these Japan Sea data with those of Cenozoic basalts from Sikhote-Alin, the back-arc side of northeast Japan, the trench side of northeastJapan, and north and southwest Hokkaido. The new data and the com-parisons allow us to interpret the variation of the lithosphere andasthenosphere over time in the marginal region of northeast Asia, andthe direction of asthenospheric flow, and finally to derive the temporaland spatial relationship between the magma source and the opening ofthe Japan Sea.

2. Geological background

The Japan Sea, northeast Japan arc, and Hokkaido are located inthe convergent plate boundary between the northeast Asian conti-nental margin region and the western Pacific plate, displaying thecharacteristics of a typical back-arc basin and island arc system.The Japan Sea, one of the marginal basins, is generally consideredto have been formed by extension in a typical back-arc basin(Celaya and McCabe, 1987; Hilde and Wageman, 1973; Karig, 1971,1974; Lallemand and Jolivet, 1986; Shimazu et al., 1990; Tamaki,1985); it comprises two sub-basins, the Yamato basin and theJapan basin, that are separated by the Yamato Rise (Fig. 1B). Ourtwo sample sites, Site 794 and Site 797, are located in the northernand central Yamato basin, respectively (Fig. 1B).

During theMiocene extension of the Japan Sea, the tectonic environ-ment of northeast Japan was transformed from Eurasian continentalmargin to island arc. Previous investigations divided the volcanic beltin northeast Japan, dated early Miocene to present, into three parts,namely the trench side, transition zone, and back-arc side (Fig. 1B;Nohda et al., 1988; Ohki et al., 1995; Sato et al., 2007; Shuto et al.,1993, 2004, 2006; Takanashi et al., 2011; Tatsumi et al., 1988; Yagiet al., 2001).

Hokkaido, located at the junctions of the Japan Sea basin andKurile basin, and of the northeast Japan arc and Kurile arc (Fig. 1B),is assumed to be associated with the opening of the Kurile back-arcand the concurrent sublithospheric thinning and asthenosphericupwelling (Ikeda, 1988; Ikeda et al., 2000; Yamashita et al., 1999).Tectonically, Hokkaido can generally be divided into north Hokkaidoand southwest Hokkaido, as these areas formed in response to some-what different tectonic processes (Ikeda et al., 2000; Takanashi et al.,2011; Fig. 1B).

The Sikhote-Alin area, comprising the Sikhote-Alin Mountains andsurrounding region of far eastern Russia, is located in the northeasternEurasian continental margin (Fig. 1B). Paleomagnetic studies indicatethat the islands of Japan moved eastward and separated from Sikhote-Alin as the Japan Sea opened (Otofuji and Matsuda, 1984; Otofujiet al., 1994). Previous age dating indicates that volcanism in theSikhote-Alin and Sakhalin Island areas occurred in three stages: thesubduction-related continental margin activity (Eocene to Oligocene,55–24 Ma), the opening of the Japan Sea (early to mid-Miocene,23–15 Ma) and the post-opening of the Japan Sea (mid-Miocene toPliocene, 14–5 Ma) (Okamura et al., 1998a, 1998b, 2005).

We analyzed in detail samples from two locations — Site 794 andSite 797 of ODP 127/128 located in the Yamato Basin (Fig. 1B). Site794 is on the northern margin of the Yamato Basin, and the coreincludes basalts up to 191m thick (Fig. 2). Pouclet and Bellon (1992) di-vided these basalts into three parts. The upper part includes units 1–3,consisting mainly of fine-grained tholeiitic basalts or phyric dolerites.All the samples are relatively fresh and showmassive structure and por-phyritic texture, with Ca-rich plagioclases, augites and minor olivinesand Cr-spinel phenocrysts (Fig. 2; Allan and Gorton, 1992; Chen et al.,submitted for publication). The lowest sequence contains units 5–7,comprising olivine-rich dolerite sills, massive structure and porphyritic

texture. The phenocrysts are mainly Ca-rich plagioclases and augiteswith relatively high contents of olivine (Fig. 2). The middle layer(unit 4) was not included in this study because the samples fromunit 4 were significantly influenced by seawater alteration and thepetrogenesis of unit 4 is controversial (Chen et al., submitted forpublication). Site 797, in the central Yamato Basin, consists of 21 igne-ous units (Fig. 2); units 1, 2, and 4 are mainly lava flows, whereasunits 3 and 5–21 are interpreted as representing sills that intrudedinto soft sandstones (Allan and Gorton, 1992). Tamaki et al. (1990) pro-posed that Site 797 could be divided into two distinct geochemicalgroups. The upper group contains units 1–9, which are dominated byaphyric basalts and composed of tholeiites having relatively depletedLILEs and LREEs; the lower group comprises units 10–21which containsa series of enriched tholeiites and mildly alkaline basalts with aphyrictexture (Fig. 2).

3. Analytical methods

Most analytical samples are petrographically fresh and show nosignificant alteration. To avoid the contamination of weathered sur-face, the collected samples were sawed into slabs and the centralparts were used for bulk-rock analysis. Samples were cleaned re-peatedly three times using deionized water, dried, and then crushedand powdered in an agate mortar for geochemical and geochronolo-gy analysis.

K–Ar dating was performed using the high sensitivity mass spec-trometer (MM1200B) at the K–Ar and 40Ar–39Ar Isotopic Laboratory ofInstitute of Geology, China Earthquake Administration. Detailed de-scriptions of the analytical line are available from Wang et al., 1983and Fan et al., 1999. The fresh samples without or minor phenocrystswere chosen and powdered into 40–60 mesh (grain diameter:0.45–0.3 mm), and finally are used for K–Ar dating. The K–Ar dating isdivided into potassic measure and argon measure. The former wereundertaken by HG-5 type flame photometer, and the latter are acquiredby isotope dilution analysis using anMM-1200mass spectrometer anda38Ar drilution agent of 99.98% purity. The background values of 40Armass analyzer is 10−15 mol, at the price of samples 100 mg and diluent3–5 × 10−13 mol. Given that the analytical samples are relatively young,the sampleswere conducted to low temperature baking (180 °C–200 °C)for 2–3 days and degas prior to the analysis, in order to wipe off the in-fluences of atmosphere Ar and increase the contents of radiogenic Ar.At this static state, the contents of 40Ar, 38Ar, 36Ar were recorded, andAr initial value can be figured out based on linear regression, finally theapparent age were attained according to K–Ar formula. The K–Ar agedata of Japan Sea basalts are listed in Table 1.

Whole-rock major-element analysis was determined using fusedglass discs with a Phillips PW1400 sequential X-ray fluorescence spec-trometer (XRF) at the Institute of Geology and Geophysics, ChineseAcademy of Sciences, Beijing, P. R. China (IGGCAS). The precision formajor elements was better than 2%. Details regarding the analyticaltechniques were discussed by Guo et al. (2005). Major element dataare presented in Table A1.

Whole-rock trace-element abundances were obtained by inductive-ly coupled plasma mass spectrometer (ICP-MS) using a FINNIGANMATII element system at IGGCAS. During the analytical runs, frequent stan-dard calibrationswere performed to correct for instrumental signal driftfollowing the procedure of Gao et al. (1999). Four replicates and twointernational standards (GSR-1 and GSR-3) were prepared using thesame procedure to monitor the analytical reproducibility. On the basisof repeated analysis of samples and international standards, the preci-sion for trace elements is better than 5% (Table A1). The detaileddescriptions of analytical procedures were given in Guo et al. (2005,2006, 2007).

Sr–Nd isotope analysis was performed on a Finnigan MAT262 massspectrometer at IGGCAS. Rb, Sr, Sm and Nd were separated from otherrare earth element fractions in solution using AG50W × 8 (H+) cationic

Fig. 2. Lithological column of igneous rocks at Sites 794 and 797; 40Ar–39Ar plateau ages, K–Ar ages for Japan Sea basalts from the Yamato Basin (Sites 794 and 797); microphotographs ofcross-polarized light of 127-794C-1R, 127-794C-10R, 128-794D-17R, 127-797C-9R, 127-797C-16R, 127-797C-27R, 127-797C-44R; the 40Ar–39Ar plateau ages are from references Kaneokaet al. (1990, 1992)and Nohda (2009); the K–Ar ages are from references Pouclet and Bellon (1992); the 40Ar–39Ar age of thick tuff deposits from Site 794 is from Barnes et al. (1992); andthe red ages are from this study.

327S. Chen et al. / Lithos 224–225 (2015) 324–341

ion-exchange resin columns. Collected Sr andNd fractionswere evaporat-ed and dissolved in 2% HNO3 for analysis by mass spectrometry. Massfractionation corrections were based on 86Sr/88Sr = 0.1194 and146Nd/144Nd = 0.7219. International standard NBS987 gave 87Sr/86Sr =0.710254 ± 16 (n = 8) and NBS607 yielded 87Sr/86Sr = 1.200320 ±30 (n = 12); The international La Jolla standard yielded 143Nd/144Nd =0.511862 ± 7 (n = 12) and BCR-1 yielded 143Nd/144Nd = 0.512626 ±9 (n = 12). The whole procedure blank is less than 2 × 10−10 g for Srand 5 × 10−11 g for Nd isotope analysis. Analytical results and errors(2σ) are reported in Table A2.

Pb isotopic ratios were measured with a VG354 mass spectrometer(UK) at IGGCAS. For Pb isotope measurements, 150 mg, 100 meshwhole-rock powder was weighed and dissolved in Teflon capsulesusing concentrated HF at 120 °C for 7 days. Pb was separated from thesilicate matrix and purified using AG1 × 8 anionic ion-exchange col-umns with dilute HBr as eluant. The whole procedure blank is lessthan 1 ng. Repeat analysis of the international standard NBS981 yielded

204Pb/206Pb = 0.059003 ± 0.000084 (n = 6), 207Pb/206Pb =0.914490 ± 0.00017 (n = 6), and 208Pb/206Pb = 2.166910 ±0.00097 (n = 6). Pb isotope fractionations were corrected usingcorrection factors from the certified values of the internationalstandard NBS981. The average 2σ uncertainties for measured ratiosof 206Pb/204Pb, 207Pb/204Pb and 208Pb/204Pb are respectively 0.6%,0.4% and 0.5%. The Pb isotope data are presented in Table A2.Detailed sample preparation and analytical procedures for theSr–Nd–Pb isotope follow those of Fan et al. (2003) and Zhanget al. (2002).

4. Results

The K–Ar age data of Site 794 and Site 797 basalts from Japan Sea arelisted in Table 1, and the major, trace elements and Sr–Nd–Pb isotopiccompositions are listed in Tables A1 and A2. Based on the previousinvestigations (Allan and Gorton, 1992; Chen et al., submitted for

Table 1K–Ar and 40Ar/39Ar dating results for Cenozoic basalts from Japan Sea.

Sample No. Weight (mg) 40K (%) 40K (10−9 mol/g) 40Ar* (%) 40Ar (10−12 mol/g) Age (Ma) Source

127-794C-9R 33.25 0.51 15.20 13.8 25.00 28.0 ± 1.7 This study127-794C-10R 23 0.57 17.00 17.25 28.80 28.9 ± 1.4 This study127-797C-9R 32.45 0.19 5.67 20.39 5.69 17.2 ± 0.7 This study127-797C-12R 35.1 0.17 5.07 13.73 4.48 15.1 ± 0.9 This study127-797C-16R 38.35 0.14 4.18 6.902 3.84 15.7 ± 1.9 This study128-794D-1R 20.6 ± 2.9 Pouclet and Bellon (1992)128-794D-3R 18.3 ± 1.3 Pouclet and Bellon (1992)128-794D-3R 18.1 ± 1.2 Pouclet and Bellon (1992)128-794D-8R 23.7 ± 5.0 Pouclet and Bellon (1992)

Sample No. 36Ar/40Ar (10−3) 36Ar/40Ar 36Ar/40Ar (10−2) 40Ar*/39Ar* Total age (Ma) Plateau age (Ma) Source

127-797C-27R 1.875 1.299 22.10 2.991 20.8 19.0 ± 1.1 Kaneoka et al. (1992)127-797C-34R 2.713 0.420 9.980 3.230 19.9 19.9 ± 1.1 Kaneoka et al. (1992)127-797C-41R 2.710 0.349 9.173 3.238 20.3 19.0 ± 0.3 Kaneoka et al. (1992)127-797C-45R 2.366 0.339 13.56 2.908 18.6 17.7 ± 0.5 Kaneoka et al. (1992)127-794C-3R 1.815 1.753 27.00 2.990 20.9 20.6 ± 0.6 Kaneoka et al. (1992)127-794C-8R 3.142 0.121 3.081 3.059 21.7 20.0 ± 2.0 Kaneoka et al. (1992)128-794D-15R 3.657 2.334 4.838 4.202 24.0 19.9 ± 0.7 Kaneoka et al. (1992)128-794D-17R 3.282 0.479 1.772 5.125 39.9 Kaneoka et al. (1992)128-794D-20R 3.441 0.850 2.103 4.998 31.2 21.2 ± 0.8 Kaneoka et al. (1992)

328 S. Chen et al. / Lithos 224–225 (2015) 324–341

publication; Nohda, 2009) and the following geochemical characteris-tics, 127-794C-1R, 127-794C-5R, 127-794C-6R, 127-794C-9R, 127-794C-10R, 127-794C-12R, 128-794D-1R, 128-794D-4R, 128-794D-8Rare defined as Upper layer of Site 794 (Fig. 2; US794); 128-794D-17R,128-794D-20R are labeled as as Lower layer of Site 794 (Fig. 2;LS794); 127-797C-9R, 127-797C-12R, 127-797C-16R are defined asUpper layer of Site 797 (Fig. 2; US797), 127-797C-27R, 127-797C-32R,127-797C-44R are labeled as Lower layer of Site 797 (Fig. 2; LS797).

4.1. K–Ar dating results for Site 794 and Site 797

The K–Ar age data of Japan Sea basalts are listed in Table 1, includingtwo samples fromUS794 (127-794C-9R, 127-794C-10R), three samplesfrom US797 (127-797C-9R, 127-797C-12R, 127-797C-16R). The K–Arage data of these two US794 samples and three US797 samples respec-tively are 28.0 ± 1.7 Ma, 28.9 ± 1.4 Ma, 17.2 ± 0.7 Ma, 15.1 ± 0.9 Ma,15.7± 1.9Ma. Combinedwith previous reliable ages, this K–Ar age dataand stratigraphic information (Allan and Gorton, 1992; Barnes et al.,1992; Chen et al., submitted for publication; Kaneoka et al., 1990,1992; Nohda, 2009; Pouclet and Bellon, 1992), we can get detailed

Fig. 3. Na2O + K2O (wt.%) versus SiO2 (wt.%) (A) (Le Bas et al., 1986) and AFM ternary (B) pcalcalkaline dividing curve are from Irvine and Baragar (1971).

stratigraphic columns and corresponding ages distribution for Site 794and Site 797 (Fig. 2).

4.2. Major- and trace-elements

Generally, the major elements of Japan Sea basalts display alarge range: SiO2 46.81%–52.55%, FeOT 7.05%–10.96%, MgO 5.77%–11.95%, CaO 3.75%–10.81%, K2O + Na2O 2.84%–6.05%, Mg# values(Mg/(Mg + Fe2+)) 0.34–0.54 (Table A1). It's worthy noting thatMgO values are relatively high, whereas the total alkali content(K2O + Na2O = 2.84%–6.05%) are relatively low, indicating theybelong to calc-alkaline basalt which is confirmed from the diagram ofNa2O + K2O vs. SiO2 and AFM ternary plot (Fig. 3).

On the Chondrite-normalized REE plot (Fig. 4A), all heavy rare earthelements (HREE) display relatively flat patterns. US797 is characterizedby the depletion of light rare earth elements (LREE), slight enrichmentof HREE, remarkable low ratios of (La/Yb)N (0.81–1.58) and (La/Sm)N(0.64–0.99), which is similar to the distribution pattern of N-MORB(Fig. 4A). Regarding LS797, US794 and LS794, they all share similardistribution patterns, with slight enrichment of LREE over flat HREE,

lots of Cenozoic basalts from Japan Sea; alkaline–subalkaline dividing line and tholeiitic–

Fig. 4. Chondrite normalized REE patterns (A) and Primitive mantle normalized trace element patterns (B) of Cenozoic basalts from Japan Sea; the data of Chondrite, Primitivemantle, E-MORB, N-MORB and OIB are from Sun and McDonough (1989).

329S. Chen et al. / Lithos 224–225 (2015) 324–341

resembling the pattern of E-MORB but higher total REE contents. Andtheir (La/Yb)N and (La/Sm)N ratios are respectively 1.15–3.50 and0.86–1.97, and slight Eu anomalies are observed, with the values

Fig. 5. Plots of the downhole variation in 143Nd/144Nd, 87S

of δEu 0.81–1.21. Similarly, on the trace elements spider diagram(Fig. 4B), US797 is distinguished by obvious LILE depletion, consistentwith those of N-MORB with exception of the anomalies of active

r/86Sr, 206Pb/204Pb isotopes and Zr/Nb, Nb/Ta ratios.

Fig. 6. Plots of 143Nd/144Nd vs. 87Sr/86Sr (A), 143Nd/144Nd vs. 206Pb/204Pb (B), 207Pb/204Pb vs. 206Pb/204Pb (C), 208Pb/204Pb vs. 206Pb/204Pb (D), 143Nd/144Nd vs. (La/Sm)N (E), 143Nd/144Nd vs.Th/Nd (F) of Cenozoic basalts from Japan Sea. Fig. 6A ismodified from Cousens and Allan (1992); Fig. 6B ismodified from Cousens et al. (1994); the data of alkaline basalts fromUIrueng &Dog(U–D) Island are from Tatsumoto and Nakamura (1991); the data of NE Japan Arc, Mariana Arc, Izu/Central Japan Arc are from Stern and Bibee (1984) and Nohda and Wasserburg(1986); the data of Pacific MORB, Indian MORB, Pacific Sediment, DM, EM , EMП are from Zindler and Hart (1986). In Fig. 6C and D, the data of Japan Arc, Mariana or Izu arc/backarcare from Stern and Bibee (1984), Ewart and Hawkesworth (1987), Othman et al. (1989), Stern et al. (1990), Volpe et al. (1990); the data of PacificMORB, Pacific Sediment are fromZindlerand Hart (1986). Fig. 6E is modified from Cousens and Allan (1992); Fig. 6F is modified from Cousens et al. (1994); the data of Pacific MORB, Pacific Sediment are from Zindler and Hart(1986).

330 S. Chen et al. / Lithos 224–225 (2015) 324–341

elements such as Rb, Ba, Pb. Whereas the total trace elements concen-trations of LS797, US794, LS794 are significantly higher comparedwith US797, relatively similar to the pattern of E-MORB (Fig. 4).

4.3. Sr–Nd–Pb isotopes

The Sr–Nd–Pb isotopic compositions of the Japan Sea basalts areshown in Figs. 5 and 6. The US797 is characterized by relatively high

143Nd/144Nd (0.513035–0.513139) value, slightly low 87Sr/86Sr(0.704051–0.704254) and Pb isotopic ratios (206Pb/204Pb =17.758–18.223; 207Pb/204Pb = 15.463–15.559; 208Pb/204Pb =37.636–38.304), whereas LS797, US794, LS794 have totally oppositeSr–Nd–Pb isotopic compositions, with low 143Nd/144Nd (0.512705–0.512917) ratio, high 87Sr/86Sr (0.704248–0.705222) and Pb isotopicratios (206Pb/204Pb = 18.077–18.377; 207Pb/204Pb = 15.531–15.610;208Pb/204Pb=38.152–38.557) (Figs. 5; Fig. 6A, B, C, D). The Sr–Nd isotopic

331S. Chen et al. / Lithos 224–225 (2015) 324–341

compositions are within the field of mantle evolution array, similarto the isotopic compositions of basalts from the Japan Arc (Fig. 6A).And the lead isotopic variation of all basalts show quite nicecorrelationship, locating above the North Hemisphere ReferenceLine (NHRL) of Hart (1984), similar to those of Japan Arc but muchhigher 207Pb/204Pb and 208Pb/204Pb and lower 143Nd/144Nd for a given206Pb/204Pb values than Pacific MORB and Mariana or Izu arc/backarcsystems (Fig. 6B, C, D).

5. Discussion

5.1. Implications from the ages of the volcanic rocks

K–Ar and 40Ar–39Ar dating of basalts from the Japan Sea basementprovides vital information for determining the timing of the extensionthat formed the Japan Sea. Although some investigations that includeconstraints and comparisons of different layers of the basalts were car-ried out in the past few years, comprehensive studies are still limited(Kaneoka et al., 1990, 1992; Nohda, 2009; Pouclet and Bellon, 1992).Here, we compile previously published reliable ages of Japan Sea basaltsand our own K–Ar dating results to restrict and compare the age rangeof rocks from Sites 794 and 797 (Table 1; Fig. 2). Previous studies foundthat K–Ar dates are inconsistent with 40Ar/39Ar results even for thesame sample for various reasons such as excess 40Ar and 39Ar loss(Baksi, 2007, 2014). Thus, it is important to evaluate the reliability ofthe dates before the age data are used to constrain the geologicalevent (for detail see Baksi, 2003, 2005, 2014).

With regard to the lower layer of Site 797 (LS797), the reliable40Ar–39Ar plateau ages of samples 797C-27R-1, 797C-34R-1, 797C-41R-1, and 797C-45R-2 are respectively 19.0 ± 1.1 Ma, 19.9 ± 1.1 Ma,19.0 ± 0.3 Ma, and 17.7 ± 0.5 Ma based on 40Ar–39Ar whole-rockstepwise-heating experiments (Fig. 2; Kaneoka et al., 1990, 1992;Nohda, 2009). However, for the upper layer of Site 797 (US797), the ba-salts are characterized by relatively depleted incompatible elements,with an especially low K content, making the K–Ar and 40Ar–39Ar datingdifficult. Because of this K deficiency, the age of the US797 samples re-mains poorly constrained (Kaneoka et al., 1990, 1992; Nohda, 2009;Tamaki et al., 1990). Fortunately, using the latest advanced analyticalmethods we obtained the following K–Ar dating results for samples797C-9R, 797C-12R, and 797C-16R: 17.2 ± 0.7 Ma, 15.1 ± 0.9 Ma, and15.7 ± 1.9 Ma, respectively (Fig. 2).

For the upper and lower layers of Site 794 (US794, LS794), the reli-able 40Ar–39Ar plateau ages of 794C-3R-1, 794C-8R-1, 794D-15R-1,and 794D-20R-1 are 20.6 ± 0.6 Ma, 20.0 ± 2.0 Ma, 19.9 ± 0.7 Ma,and 21.2 ± 0.8 Ma, respectively (Fig. 2; Kaneoka et al., 1990, 1992;Nohda, 2009); the K–Ar ages of 794D-1R, 794D-3R, and 794D-8R are20.6 ± 2.9 Ma, 18.3 ± 1.3 Ma, 23.7 ± 5.0 Ma, respectively (Fig. 2;Pouclet and Bellon, 1992). Note that the total 40Ar–39Ar age is olderthan the 40Ar–39Ar plateau age. For example, the total Ar ages of794D-17R-1and 794D-20R-1 are 31.9 ± 2.0 Ma and 31.2 ± 2.1 Ma, re-spectively, whereas the plateau ages are 19.9 ± 0.7 Ma and 21.2 ±0.8 Ma (Kaneoka et al., 1992; Nohda, 2009; Tamaki et al., 1990). Theresults are associated with excess 40Ar in the highest temperaturefraction and/or partial 39Ar loss in the lowest temperature fraction.The excess 40Ar can be trapped in phenocrysts that tend to be releasedat high temperatures during the stepwise heating of 40Ar/39Ar dating.Selective 39Ar loss may result from the recoil effect of 39Ar or the redis-tribution of radiogenic 39Ar trapped at a relatively loose site during neu-tron irradiation (Kaneoka et al., 1992). Given that the plateau rangecovers most of the integrated 39Ar, the calculated 40Ar–39Ar plateauage can be considered to represent the time of crystallization of thelavas.

Similarly, we found that the K–Ar ages of 794C-9R and 794C-10R are28.0 ± 1.7 Ma and 28.9 ± 1.4 Ma, respectively, in agreement with thepreviously obtained total Ar ages (e.g., 794D-17R-1 and 794D-20R-1were 31.9 ± 2.0 Ma and 31.2 ± 2.1 Ma, respectively) and significantly

older than the previous 40Ar–39Ar plateau age. Therefore, we arguethat the high K–Ar agewas also related to excess 40Ar trapped in pheno-crysts. This also agrees with the petrological characteristics of thesamples, because the crystallinity of the samples from 794C-9Rand 794C-10R is relatively good, as inferred from the appearance ofplagioclase and clinopyroxene and minor olivine phenocrysts, and thenumber of crystal modes in 50–60% (Fig. 2; Chen et al., submitted forpublication). These mineral phenocrysts may contain excess 40Ar asdiscussed by Kaneoka et al. (1992). However, the possibility of the oc-currence of excess 40Ar in the US797 samples is low, since the basaltsfrom the US797 are characterized by poor crystallization or absence ofsignificant phenocrysts (Fig. 2) and the number of crystal modes isonly 6–8%. Baksi (1995) also suggested that the K–Ar dates can haveminor (5–7%) loss of 40Ar* with regard to aphyric basalts due to incom-plete breakdown of the mineral lattice. We therefore argue that theexcess 40Ar trapped in phenocrysts can have a relatively large effecton the K–Ar age of 794C-9R and 794C-10R whereas the influence onthe K–Ar age of US797 would be minor.

The basalts from the basement of Site 797 and Site 794 are overlainby thick tuff deposits, with 40Ar–39Ar ages of 14.9 ± 0.2 Ma (Fig. 2;Barnes et al., 1992). Because the sites are not far apart and there are sim-ilar geochemical correlations between them, Nohda (2009) estimatedthe minimum age of the basalts from Site 797 and Site 794 as 14.9 ±0.2 Ma. Note that our analytical ages of the basalts from US797 are con-sistentwith this result. The age distribution of the samples fromSite 797and Site 794 is shown in Fig. 2. Clearly, the ages of LS797, US794, andLS794 are confined to 17.7 ± 0.5 Ma–21.2 ± 0.8 Ma, whereas the agerange of US797 is 15.1 ± 0.9 Ma–17.2 ± 0.7 Ma, younger than LS797,US794, and LS794.

5.2. Petrogenesis

5.2.1. Effect of seawater alteration and crustal contaminationThe seafloor basalts were inevitably influenced by seawater alterna-

tion, and the basaltic magma passed through some thickness of crustalmaterial as it rose before erupting. Thus, we need to evaluate the poten-tial effect of seawater alteration and crustal contamination beforediscussing the nature of the mantle sources.

Some basalts from the Japan Sea are variably altered, as indicated bythe relatively high LOI (1.88–4.85%) (Table A1; Pouclet and Bellon,1992; Allan and Gorton, 1992). Based on the relationship between thecompositional variation and loss on ignition (LOI), an obvious correla-tion between the major-element compositions (e.g., MgO, SiO2, K2O,CaO, FeO), the LILEs (Rb, K, Ba, Sr) and LOI indicates that these elementswere considerably affected by alteration (Allan and Gorton, 1992).Whereas REEs, most HFSEs (Zr, Nb, Hf, and Ta) and Th, Ni, Cr are rela-tively immobile during low-temperature alteration, and do not havesignificant correlation with LOI (Barnes et al., 1985; Beswick, 1982;Pearce and Cann, 1973; Saunders and Tarney, 1984; Su et al., 2012).Thus, in the following geochemical analysis and discussion, we deter-mine the characteristics of the REEs, Nb, Th, Y, and Zr to constrain thenature of the mantle source.

The Zr/Hf ratios of these studied basalts are 33.5–37.4, similar tothose of the primitive mantle (Zr/Hf = ~36.3) and much larger thanthe ratio of the continental crust (Zr/Hf = ~11.0). The relatively lowTh/Ce (b0.071) and Th/La ratios (b0.156) are also consistent withthose of mantle-derived magma (Th/Ce = ~0.02–0.05, Th/La = ~0.12)and lower than those of crustal material (Th/Ce = ~0.15, Th/La =~0.30; Plank, 2005; Sun and McDonough, 1989; Taylor and McLennan,1995; Weaver, 1991). Collectively, these ratios of incompatible ele-ments suggest that crustal contamination plays an insignificant role inthe petrogenesis of these basalts. As another line of evidence, lead isoto-pic compositions can also be effectively used to identify crustal materialcontamination. Because of the high contents of Zr, Hf, U, Th and Pb incrustal materials and the significant fractionation of U/Pb and Th/Pb ra-tios during the process of magmatic interaction with crustal materials

Fig. 7. Plots of Th/Y vs. Sm/Th (A), Th/Yb vs. Ta/Yb (B), Sm/Yb vs. Sm (C), (Yb/Sm)P vs. (Tb/Yb)P (D), Ba/La vs. Th/Nd (E), La/Ba vs. La/Nb (F) of Cenozoic basalts from Japan Sea; Fig. 7Ais modified from Su et al, (2012); the data of PM, OIB and N-MORB are from Sun and McDonough (1989); Fig. 7B is modified from Pearce and Peate (1995); Fig. 7C is modified fromSu et al. (2012); dashed and solid curves are the melting trends from DM and PM, melting curves for spinel lherzolite(Ol0.530 + Opx0.270 + Cpx0.170 + Sp0.030 andOl0.060 + Opx0.280 + Cpx0.670 + Sp0.110) and garnet lherzolite (Ol0.600+ Opx0.200 + Cpx0.100 + Gt0.100 and Ol0.030 + Opx0.160 + Cpx0.880 + Gt0.090) are from Aldanmaz et al. (2000) andZhao and Zhou (2007). Numbers along melting curves represent the degree of partial melting; Fig. 7D is modified from Zhang et al. (2006); black lines indicate the 1%, 5%, 10%,and 15% degree of partial melting; gray lines indicate the proportion of melt formed in the presence of garnet (Gar); Fig. 7E and Fig. 7F are modified from Su et al. (2012); the data ofOIB, MORB and HIMU are from Saunders et al. (1992).

332 S. Chen et al. / Lithos 224–225 (2015) 324–341

(e.g. Kramers and Tolstikhin, 1997), crustal contamination would resultin the enrichment of Zr, Hf, and Th and a large scatter in 206Pb/204Pb,207Pb/204Pb, and 208Pb/204Pb ratios (Cousens et al., 1994; Su et al.,2012; Zhao and Zhou, 2007; Zhao et al., 2010). However, the good linearrelationship of the lead isotopic compositions (Fig. 5C, D) and nopositive anomalies of Zr, Hf, and Th in Fig. 4 further indicate that crustalcontamination is limited. Finally, the lack of correlation between144Nd/143Nd and SiO2, and 144Nd/143Nd and Th confirm that the basalticmagma was little affected by crustal contamination (Storey et al., 1997;Xie et al., 2006).We therefore conclude that therewas no significant in-teraction between the basaltic magmas and crustal materials duringmagma ascent, and thus the geochemical characteristics of the basaltscan be reasonably used for constraining the nature of the magmamantle source.

5.2.2. Nature of the mantle sourcesAccording to the total alkali contents (Na2O+ K2O= 2.84%–6.05%),

plots of Na2O + K2O–SiO2, and AFM diagrams, most of the Japan Seabasalts belong to the calc-alkaline basaltic series (Fig. 3). The REEs pat-terns and spider diagrams show that the concentrations of LREEs andLILEs of US797 are notably lower than those of US794, LS794, andLS797 (Fig. 4). The US797 basalts are also characterized by significantdepletions resembling those of N-MORB (Fig. 4), whereas those fromUS794, LS794, and LS797 display relatively enriched geochemistrysimilar to that of E-MORB. Similarly, the plots of incompatible elementratios can be used to constrain the nature of the magma source(Figs. 5, 7A, B; Hofmann, 1988; Su et al., 2012; Yuan et al., 2010). The ra-tios of Sm/Th (24.52–50.87), Th/Y (0.002–0.006), Ta/Yb (0.02–0.04),and Nb/Ta (12.03–12.36) of the US797 samples are closer to those of

333S. Chen et al. / Lithos 224–225 (2015) 324–341

relatively depleted mantle sources (e.g., N-MORB) (Figs. 5, 7A, B).In contrast, US794, LS794, and LS797 samples display distinctmantle features, with Sm/Th (2.27–13.85), Th/Y (0.009–0.073), Ta/Yb(0.05–0.32), and Nb/Ta (14.62–18.79) values similar to those of rela-tively enrichedmantle sources (e.g., E-MORB). The same characteristicscan also be observed from other ratios of some incompatible elements(Figs. 5, 6E, F; Dungan et al., 1986; Sun and McDonough, 1989;Vorontsov et al., 2010). Additionally, the basalts from US797 have rela-tively high 143Nd/144Nd (0.513035–0.513139) values and slightly low87Sr/86Sr (0.704051–0.704254) ratios, which are close to those ofdepleted mantle (Figs. 5 and 6A). Whereas those from LS797, US794,and LS794 are characterized by relatively low 143Nd/144Nd (0.512705–0.512917) ratios and high 87Sr/86Sr (0.704248–0.705222) ratios,which are closer to those of enriched endmembers (Figs. 5 and 6A).Thus, based on all of the above observations, US797 samples displaydepleted features in contrast to US794, LS794, and LS797 samples(Figs. 5, 6 and 7).

Aldanmaz et al. (2000) pointed out that REE ratios can be used toconstrain the depth of the magma source, since REE display similargeochemical behavior during partial melting or fractional crystalli-zation (Zhao and Zhou, 2007). Based on the plot of Sm/Yb–Sm and(Yb/Sm)P−(Tb/Yb)P (Fig. 7C, D), the basalts from the Japan Sea haveslightly low Tb/Yb and Sm/Yb ratios, indicating that the magma sourceis relatively shallow and these Japan Sea basalts are products of partialmelting of themantle in the stability field of spinel lherzolitewithout gar-net (Fig. 7C, D; Saunders et al., 1992; Su et al., 2012). Additionally, theJapan Sea basalts generally display a nearly flat medium-to-heavychondrite-normalized REE (MREE, HREE) pattern (Fig. 4A), and have

Fig. 8. TiO2 vs. FeOT/MgO (A) and Zr/Y vs. Zr (B) of basaltic rocks from the Japan Sea and the baMiocene volcanic rocks of the Japan Sea, the back-arc side, the trench side ofNE Japan arc. In Fig.and Bellon (1992) and Cousens et al. (1994); the age of the back-arc side of NE Japan are respectet al. (2006); fields of ocean island tholeiite (OIT), island-arc tholeiite (IAT) and abyssal tholeiitebasalt (BABB) is by Miyashita et al. (1995). Fields of continental arc basalts and oceanic arc basafrom this study and Allan and Gorton (1992), Pouclet and Bellon (1992) and Cousens et al. (19

limited variations of (La/Yb)N ratios (0.81–3.50), low Ce/Y ratios(0.40–1.28), and (Tb/Yb)N values of 1.08–1.46 (Wang et al., 2002), alsosuggesting a shallow magma source. Yamashita and Fujii (1992) pro-posed, based on high-pressure experiments on the Japan Sea samples,that the basaltic magma was segregated from the garnet-free mantle ofspinel lherzolite at shallow depths of ~40–50 km, at pressures of~10 kbar (Frey et al., 1978; Green et al., 1979; Hanson, 1989; Schilling,1975; Takahashi and Kushiro, 1983). What's more, the analytical plagio-clase and clinopyroxene electronic probedata of the Japan Sea samples in-dicate that the depth of the basaltic magma source beneath the Japan Seais less than 50 km, at pressure conditions of 3.48–11.19 kbar (Chen et al.,submitted for publication).

Given the above geochemical differences and the good correlationbetween the isotopes and trace element ratios (Fig. 6A, B, E, F; Chenet al., submitted for publication), we argue that the mantle sources be-neath the Japan Sea may be heterogeneous, that is, the mantle sourcesmay be formed by the mixing of depleted asthenosphere and enrichedcompositions. The spider diagram (Fig. 4B) shows that the Japan Sea ba-salts have anomalies of mobile elements (e.g., Rb, Ba, Sr) and positiveanomalies of Pb, suggesting the geochemical features of subduction-related fluid (e.g. Davidson, 1987; Gill, 1981; Pearce, 1983; Perfit et al.,1980). Moreover, the Japan Sea basalts are characterized by relativelylow Th/Nd (b0.2) ratios, and large variation of Ba/Th, Ba/Nb, and Ba/La(Fig. 7E; Su et al., 2012), and the Sr–Nd–Pb isotope composition plotsin the field of the Japan island have mixture characteristics of depletedmantle and subduction-related material (Fig. 6A, C, D), all indicatingthat the enriched geochemistry may be associated with fluids ormelts released from the subducting slab. Additionally, the isotopic

ck-arc side of NE Japan; K2O vs. SiO2 (C) and Rb vs. SiO2 (D) variation diagrams for middle8A and B, the data of the Japan Sea are from this study andAllan andGorton (1992), Poucletively 20–22Ma and 11–14Ma, and the data of the back-arc side of NE Japan are from Shuto(MORB) in FeOT/MgO–TiO2 diagramare from Shuto (1988); and field of the back-arc basinlts in Zr–Zr/Y diagram are from Pearce (1983). In Fig. 8C, 8D, the data of the Japan Sea are94); the data of the back-arc side and trench side are from Shuto et al. (1993).

334 S. Chen et al. / Lithos 224–225 (2015) 324–341

compositions have the features of enrichedmantle II (EM II) (Fig. 6A, B),and the compositions of EM II may be closely related to subducted oce-anic sediments (Zindler and Hart, 1986). This can also be demonstratedby the plots of 143Nd/144Nd vs. (La/Sm)N and 143Nd/144Nd vs. Th/Nd(Fig. 6E, F; Cousens and Allan, 1992; Cousens et al., 1994; Zindler andHart, 1986), given that subduction-related fluid has almost no impacton Nd isotopic composition (Ellam and Hawkesworth, 1988). Thus,the significantly negative correlation between 143Nd/144Nd and the

Fig. 9. 87Sr/86Sr versus age and 143Nd/144Nd versus age diagrams of basaltic rocks from the JapanNorth Hokkaido and Southwest Hokkaido (B), the Sikhote-Alin (C). The data of the Japan Sea arJapan are from Nohda et al. (1988) and Shuto et al. (1993); the data of the back-arc side of NEet al. (1997, 2004, 2006); the data of theNorthHokkaido are from Shuto et al. (2004); the data ofrom Nohda (2009) and Okamura et al. (2005).

trace element ratiosmanifests that the enriched compositions are likelyderived from subducted sediments (Fig. 6E, F). On the other hand,these basic rocks have relatively high La/Nb (1.00–4.80) and low La/Ba(0.05–0.16) ratios (Fig. 7F), which are also features of a subduction-modified continental lithospheric mantle source (Saunders et al.,1992; Thompson and Morrison, 1988). In summary, we propose thatthese enriched compositions are likely derived from subducted fluidsand/or sediments from the western Pacific slab, with subsequent

Sea, the trench side and transitional zone ofNE Japan, the back-arc side of NE Japan (A), thee from this study andNohda (2009); the data of the trench side and transitional zone of NEJapan are from Fukase and Shuto (2000), Kondo et al. (2000), Ohki et al. (1995) and Shutof the Southwest Hokkaido are fromTakanashi et al. (2011); the data of the Sikhote-Alin are

Fig. 10. Sr–Nd isotopic profile of the trench side, the back-arc side NE Japan, and Japan Seaduring the pre-opening stage (A) and the syn- and post-opening stage (B). The data of theJapan Sea are from this study andNohda (2009); the data of the trench side and transition-al zone of NE Japan are from Shuto et al. (1993); the data of the back-arc side of NE Japanare from Fukase and Shuto (2000), Kondo et al. (2000), Ohki et al. (1995) and Shuto et al.(1997, 2004, 2006).

335S. Chen et al. / Lithos 224–225 (2015) 324–341

interaction with the overlying extending lithosphere (Chen et al.,submitted for publication; Cousens and Allan, 1992; Nohda, 2009;Ujike and Tsuchiya, 1993).

5.3. Tectonic evolution of the northeast Asian margin: Sikhote-Alin, JapanSea, northeast Japan, and Hokkaido

5.3.1. Tectonic setting of the Japan SeaSynthesizing the above chronological and geochemical characteris-

tics, we can summarize the following: the basalts from US797 are char-acterized by a relatively young eruption time (15.1 ± 0.9 Ma–17.2 ±0.7 Ma) and depleted geochemical composition, whereas the relativelyold basalts (17.7 ± 0.5 Ma–21.2 ± 0.8 Ma) from LS797, US794, andLS794 have slightly enriched geochemical characteristics and are relat-ed to the character of the subcontinental lithospheric mantle andsubduction-related fluids/sediments. That is to say, the nature of themagma source beneath the Japan Sea shifted considerably fromenriched to depleted over 17 ± 2 Ma (15–19 Ma). This conspicuousgeochemical trend in the volcanic rocks is closely associated with theextension of the Japan Sea (Nohda et al., 1988). Specifically, duringthe subduction of the western Pacific plate beneath the northeastAsia continental margin, the fluids and sediments released from thesubducting slabmigrated into the overlyingmantle wedge and loweredthe solidus temperature, subsequently triggering the partial melting ofthe surrounding asthenospheric mantle (e.g. Gill, 1981). The meltedmantle-derived magma continuously ascended, forming the convectiveasthenosphere mantle, which further led to the partial melting andthinning of the overlying subcontinental lithospheric mantle. Accompa-nied by the continuous extension and thinning of the subcontinentallithospheric mantle (SCLM), the subducting slab retreated eastward,and this process in turn increased the extensional forces acting on theJapan Sea. Repeating these processes (subduction and extension)ultimately lead to the opening of the Japan Sea (Liu et al., 2001;Nohda, 2009).

Plots of TiO2–FeOT/MgO and Zr/Y–Zr (Fig. 8A, B) also show thatthe Japan Sea and the back-arc side of northeast Japan had contrast-ing tectonic settings during the early–middle Miocene. Prior to theopening of the Japan Sea (20–22Ma), the partial melting and the ten-sion of the overlying thickened lithosphere would have resulted inthe upwelling and intrusion of asthenospheric mantle with a smallsubducting-slab-derived component, forming characteristic conti-nental rift zone-type and continental arc-type basaltic magma(Fig. 8A, B). After the opening of the Japan Sea (11–14 Ma), theoverlying lithosphere rapidly thinned, accompanied by upwellingasthenosphere and subduction-related fluids and slab-derived sedi-ments, and the partial melting produced oceanic arc-type basalticmagma and/or abundant back-arc basin basalt (Fig. 8A, B). As aresult, basaltic magmas of the Japan Sea between 20 Ma and 15 Madisplay geochemical features between continental arc and oceanicarc-type basalt (Fig. 8A, B).

5.3.2. Sr, Nd isotopic characteristics of the magma sources beneathnortheast Asia

Here we present the 87Sr/86Sr vs. age and 143Nd/144Nd vs. agediagrams of basaltic rocks from the Japan Sea, the trench side, and thetransitional zone, the back-arc side of northeast Japan, north Hokkaidoand southwest Hokkaido, and the Sikhote-Alin area, including ournew data and compiled previous data (Fig. 9; Fukase and Shuto, 2000;Kersting et al., 1996; Kondo et al., 2000; Nohda et al., 1988; Nohda,2009; Ohki et al., 1995; Okamura et al., 2005; Sato et al., 2007; Shibataand Nakamura, 1997; Shuto et al., 1993, 1997, 2004, 2006; Takanashiet al., 2011). The basalts from the trench side and the transitional zoneof northeast Japan show 87Sr/86Sr and 143Nd/144Nd in the range of0.70411–0.70560 and 0.51274–0.51284, respectively, with relativelyconstant Sr, Nd isotopic compositions from 30 Ma to the present(Fig. 9A; Nohda et al., 1988; Shuto et al., 1993). In strong contrast, the

Sr–Nd isotopic compositions of the basalts from the back-arc side ofNE Japan, the Japan Sea, Hokkaido, and Sikhote-Alin dramaticallychange at certain time periods (Fig. 9). Specifically, the Sikhote-Alin ba-salts have relatively high Sr isotopic ratios (0.70367–0.70506) and lowNd isotopic ratios (0.51261–0.51290) during 55–24 Ma, but duringthe 23–15 Ma period, they display relatively low Sr isotopic ratios(0.70335–0.70363) andhighNd isotopic ratios (0.51288–0.51292), sug-gesting a considerable change at 23–24 Ma (Fig. 9C; Nohda, 2009;Okamura et al., 2005). The Japan Sea basalts are characterized by rela-tively high Sr isotopic ratios (0.70405–0.70513) and low Nd isotopic ra-tios (0.51268–0.51299) before 17 Ma, whereas in the 17–15 Ma periodthey have relatively low Sr isotopic ratios (0.70370–0.70425) and highNd isotopic ratios (0.51304–0.51317), indicating a major variationaround 17 ± 2 Ma (15–19 Ma) (Fig. 9A). The basalts from the back-arc side of NE Japan and southwest Hokkaido have relatively high Srisotopic ratios (N0.7040) and low Nd isotopic ratios (b0.51284) priorto 15 Ma. In contrast, they have relatively low Sr isotopic ratios(0.7030–0.7040) and high Nd isotopic ratios (N0.51284) after the15 Ma. That is, around 15 Ma the Sr–Nd isotopic compositions show arapid change (Fig. 9A, B; Takanashi et al., 2011; Shuto et al., 2004,2006). Similarly, the north Hokkaido basalts display relatively highSr isotopic ratios (0.70466–0.70473) and low Nd isotopic ratios(0.51279–0.51283) at 12 Ma and relatively low Sr isotopic ratios(b0.70466) and high Nd isotopic ratios (N0.51284) after 12Ma, indicat-ing that the point-in-time of major change in north Hokkaido wasbefore 12 Ma (Fig. 9B; Shuto et al., 2004).

336 S. Chen et al. / Lithos 224–225 (2015) 324–341

These significant changes in the Sr–Nd isotopic compositions of ba-saltic magma from the back-arc side of NE Japan, Japan Sea, Hokkaido,and Sikhote-Alin, from a slightly enriched isotopic signature to a deplet-ed isotopic signature at a given time period, are considered to be associ-ated with constant upwelling of the asthenosphere and its subsequentintrusion into the overlying thinned SCLM during the extension of theJapan Sea (e.g., Nohda et al., 1988; Ohki et al., 1995; Shuto et al., 1993,2004; Tatsumi et al., 1988; Ujike and Tsuchiya, 1993). In contrast, theisotopic compositions of the basalts from the trench side and the transi-tional zone of NE Japan have remained unchanged from 30 Ma to thepresent and are distinguished by undepleted features, suggesting thatsince 30 Ma the subcontinental lithospheric mantle has played a veryimportant role.

5.3.3. Eastward asthenospheric flow in northeast AsiaUpwelling asthenosphere and thinning of the overlying SCLM has

previously been shown to result in the temporal variation of the Sr,Nd isotopic compositions from enriched to depleted in the studiedarea (Nohda, 2009; Nohda et al., 1988; Okamura et al., 2005; Shutoet al., 1993, 2004, 2006; Takanashi et al., 2011; Tatsumi et al., 1988).The significance of our study is that we found systematic and verymarked variations in the basaltic magma of the Sikhote-Alin area,

Fig. 11. Skeptical models illustrating the structural changes that occurred beneath the northeaopening stage (B) (modified from Jolivet and Tamaki, 1992; Jolivet et al., 1995).

Japan Sea, the back-arc side of NE Japan and SW Hokkaido, and northHokkaido, from a slightly enriched isotopic signature to a depletedisotopic signature at 23–24 Ma, 17 ± 2 Ma (15–19 Ma), 15 Ma, andN12Ma, respectively (Fig. 9). That is to say, the time period of the injec-tions of asthenosphere into subcontinental lithospheric mantleoccurred roughly at 23–24 Ma, 17 ± 2 Ma (15–19 Ma), 15 Ma, andN12 Ma, in a trend indicating a southeastward to northeastwardasthenosphere flow beneath the northeast Asia continent (Fig. 11).Additionally, the eastern boundary of marginal sea basin displays aneastward convex curve, implying a similar trend of eastward astheno-spheric flow beneath northeast Asia, and this character could be termedas a “mantle tongue”. This agreeswith previous studies that have shownthat in the eastern margin of the Asian continent, volcanism associatedwith thewestern Pacific subduction zone began in the Sikhote-Alin areaand progressed eastward via the Japan Sea andfinally to northeast Japan(Nohda, 2009).

5.3.4. The relationship between the temporal variation of Nd, Sr and theopening of the Japan Sea

During the tensile processes of the Japan Sea, the separation ofthe Japan arc from the NE Asian continent resulted in major chang-es in the tectonic environments and mantle wedge compositions.

st margin of the Asian continent during the pre-opening stage (A) and the syn- and post-

337S. Chen et al. / Lithos 224–225 (2015) 324–341

Based on the above Sr–Nd isotopic characteristics, the temporalvariation of Sr and Nd are closely associated with the tension andopening of the Japan Sea. To further examine the variations in Sr–Nd isotopic compositions over time, we separately considered com-positions from the pre-opening stage and the syn and post-openingstage. We also constructed Sr–Nd isotopic profiles (Fig. 10) and adiagram of the geochemical structure of the mantle wedge(Fig. 11) for the trench side, the back-arc side of NE Japan, andthe Japan Sea.

For the pre-opening stage, the Sr isotopes remain unchangedfrom the trench side to the Japan Sea, whereas the Nd isotopes dis-play a depleted tendency from the back-arc side to the Japan Sea(Fig. 10). This isotopic profile of the pre-opening stage suggests fea-tures of a continental margin volcanic arc. The mantle wedge in theNE Asian continental margin of the profile is predominantly com-posed of thick and enriched subcontinental lithospheric mantleand eastward flowing asthenosphere (Fig. 11; Kersting et al., 1996;Kimura et al., 2002; Nohda et al., 1988; Shibata and Nakamura,1997). Before the opening of the Japan Sea, the fluids or melts re-leased from the subducting slab lowered the solidus temperatureand triggered partial melting of the surrounding asthenosphericmaterials and subsequently generated the convective asthenospher-ic mantle. The flowing and invasion of asthenosphere led to a certaindegree of partial melting and tension of the overlying thick litho-sphere, ultimately resulting in a slight thinning of subcontinentallithospheric mantle in the Japan Sea. In contrast, the thickness ofthe SCLM in the trench side and the back-arc side of NE Japanremained unchanged (Fig. 11). Therefore, the depleted astheno-spheric mantle plays a certain role in the magmatic activities of theJapan Sea, whereas for the volcanism in the trench side and theback-arc side of NE Japan, the relatively enriched SCLM makes agreater contribution. Additionally, enriched subduction fluids andsediments also have different effects on the mantle source, havinga decreasing impact with increasing distance from the trench. Collec-tively, due to the effects of the subcontinental lithospheric mantleand subduction fluids, the mantle sources displayed relativelyenriched characteristics in the pre-opening stage. Moreover, fromthe back-arc side to the Japan Sea side, the isotopic compositionsare gradually depleted.

For the syn- and post-opening stage, the Sr–Nd isotopic profileclearly shows depleted features compared with the pre-openingstage (Fig. 10). From the trench side to the Japan Sea side, Nd iso-topes rapidly increase as Sr isotopes decrease, with a larger slopethan during the pre-opening stage (Fig. 10). The isotopic profileof the syn- and post-opening stage reflects a mantle wedge of oce-anic arc, distinguished from the mantle wedge of thick andenriched subcontinental lithospheric mantle of the pre-openingstage. The generating mechanism of the basaltic magma for thesyn- and post-opening stage is similar to that of the pre-openingstage, including the fluids/melts or sediments released from thesubducting slab, the convective asthenosphere mantle, and thetension and thinning of the subcontinental lithospheric mantle.However, the opening of the Japan Sea resulted in the sudden thin-ning of the SCLM, and this thinning decreased the chemical influ-ence of the SCLM and increased the chemical influence of theMORB during the syn- and post-opening stage (Fig. 11). Thus, ina trend from the trench side to the Japan Sea side, the Sr–Nd iso-topic compositions vary from undepleted to depleted. Significantacross-arc Sr–Nd isotopic variations are also closely associatedwith the fluids released from the subducting slab, having a de-creasing effect with increasing distance from the trench (Shibataand Nakamura, 1997).

During the middle Miocene, in addition to the variations of theSr–Nd isotopic compositions, another conspicuous characteristicacross the profile from the trench side to the back-arc side, then tothe Japan Sea, is the variation of the LILE, especially the K and Rb

components. With increasing distance from the trench, the contentsof K2O and Rb significantly decrease for a given SiO2 (Fig. 8C, D;Allan and Gorton, 1992; Cousens et al., 1994; Pouclet and Bellon,1992; Shuto et al., 1993), which may reflect variable degrees ofpartial melting. That is, in the Japan Sea and the back-arc side,the degree of partial melting is apparently higher than the partialmelting of the trench side; this relatively high partial melting onthe Japan Sea and the back-arc side may be associated with anexpected increase in the geothermal gradient resulting from the in-jection of hot asthenosphere into the subcontinental lithosphericmantle.

Synthesizing the above discussion, the structural transition oftectonic setting led to significant variations of Sr–Nd isotope compo-sitions in the pre-opening stage and the syn- and post-opening stage.This structural transition is shown by geochronological and geo-chemical data to include the eastward flowing asthenosphere, injec-tion of asthenospheric materials, thinned SCLM, and the invasion ofsubduction-related fluids and slab-derived sediments. However,with regard to the trench side and the transitional zone of NEJapan, the undepleted isotopic features remain constant, suggestingthat the tectonic environment of the volcanic front does not signifi-cantly change.

6. Conclusions

The Japan Sea basalts can be divided into the Upper and Lowerlayer of Site 794 (US794, LS794), and the Upper and Lower layer ofSite 797 (US797, LS797) based on their stratigraphic level and geo-chemical characteristics. Geochronological data indicate that theage ranges of samples from LS797, US794, and LS794 are between17.7 ± 0.5 Ma and 21.2 ± 0.8 Ma, which are significantly olderthan those of the US797 samples, which have a range of 15.1 ±0.9 Ma to 17.2 ± 0.7 Ma.

The incompatible elements and radioactive isotope ratios of theUS797 samples are markedly lower compared with those of US794,LS794, and LS797, and are characterized by significantly depletedfeatures resembling those of N-MORB, whereas US794, LS794, andLS797 samples display enriched geochemical characteristics similarto those of E-MORB. The basaltic magma source from the Japan Seais relatively shallow, and all the Japan Sea basalts are products ofthe partial melting of mantle in the garnet-free and spinel-bearingstability field.

When all the Sr–Nd isotopic compositions and age data of basal-tic rocks from the whole NE Asian marginal region are synthesized,patterns of Sr–Nd isotopic compositions of basaltic magma fromthe Sikhote-Alin, Japan Sea, back-arc side of NE Japan and SW Hok-kaido, and north Hokkaido display dramatic variations from aslightly enriched isotopic signature to a depleted isotopic signatureat 23–24 Ma, 17 ± 2 Ma (15–19 Ma), 15 Ma, and N12 Ma, respec-tively. These variations are considered to be associated with con-stant upwelling of the asthenosphere, extension of the overlyingthinned SCLM, and the progressive movement of the astheno-sphere beneath the northeast Asia eastward relative to thelithosphere.

Acknowledgments

The authors thank the anonymous reviewers and the editor fortheir helpful reviews and constructive suggestions which greatlyimproved the original manuscript. We also thank Drs. Z. Guo, andW. Guo for their help during the preparation of this manuscript. Thiswork was supported by the National Natural Science Foundation ofChina (Grant code: 41476034, 41272369, 40802038, 41320104006and 41302102).

Table A1Major- (wt.%), trace-element (ppm) for Cenozoic basalts from Japan Sea.

Sample 127-794C-1R

127-794C-5R

127-794C-6R

127-794C-9R

127-794C-10R

127-794C-12R

128-794D-1R

128-794D-4R

128-794D-8R

128-794D-17R

128-794D-20R

128-794D-20R-p

127-797C-9R

127-797C-12R

127-797C-16R

127-797C-27R

127-797C-27R-p

127-797C-32R

127-797C-44R

GSR1 GSR3

Interval (cm) 27–30 89–92 33–39 127–131 122–126 33–38 101–105 25–29 131–135 50–53 0–3 0–3 44–47 29–32 28–31 21–24 21–24 79–82 90–93SiO2 49.10 49.25 47.71 48.68 49.48 48.92 48.77 49.10 49.62 48.65 47.55 47.49 46.81 48.64 48.50 50.01 49.89 48.54 52.55TiO2 1.13 1.26 1.24 1.07 1.29 1.42 1.21 1.00 1.26 1.43 1.38 1.38 1.24 1.10 1.04 1.76 1.75 1.77 2.11Al2O3 15.13 19.93 17.70 15.97 16.80 17.52 18.40 16.46 16.20 16.83 16.37 16.40 19.43 19.24 17.83 15.42 15.49 16.63 13.62FeO 10.03 7.89 10.39 9.35 8.40 10.55 9.74 8.81 8.14 9.27 10.40 10.38 8.96 7.05 7.77 10.12 10.08 10.01 10.07MnO 0.12 0.17 0.16 0.15 0.12 0.16 0.12 0.16 0.11 0.23 0.17 0.17 0.24 0.41 0.33 0.18 0.18 0.30 0.19MgO 9.68 6.03 8.72 11.14 8.82 6.70 6.07 10.49 8.92 8.10 7.70 7.75 8.15 6.18 7.14 7.84 7.87 6.47 5.80CaO 7.63 7.38 4.65 5.09 7.96 6.97 8.57 4.94 8.49 8.47 10.50 10.47 6.51 10.26 10.78 6.48 6.45 8.62 3.75Na2O 2.78 3.48 3.02 3.00 2.84 3.05 3.20 2.90 2.73 3.16 2.84 2.83 3.51 3.41 3.15 3.58 3.61 3.59 4.66K2O 0.70 0.80 1.54 0.64 0.64 1.23 0.61 1.22 0.72 0.27 0.15 0.15 0.14 0.09 0.08 0.38 0.38 0.44 1.39P2O5 0.19 0.22 0.20 0.17 0.23 0.23 0.21 0.17 0.21 0.21 0.16 0.15 0.14 0.12 0.10 0.24 0.24 0.24 0.44LOI 2.98 3.22 4.69 4.60 3.00 2.74 2.62 4.82 2.88 3.18 1.90 1.88 4.66 2.90 2.28 3.16 3.34 2.78 4.85σ 1.99 2.93 4.41 2.33 1.87 3.09 2.52 2.78 1.80 2.08 1.96 1.98 3.50 2.17 1.90 2.24 2.31 2.93 3.83Mg# 49.11 43.32 45.63 54.37 51.22 38.84 38.39 54.35 52.29 46.63 42.54 42.75 47.63 46.71 47.89 43.65 43.84 39.26 36.55Li 5.73 5.65 17.6 14.2 5.90 6.48 4.09 8.47 6.27 9.47 8.16 8.17 7.43 3.48 5.24 2.23 2.30 2.98 4.90 131 9.50Be 1.29 0.80 1.28 0.63 0.83 0.75 0.73 0.53 0.62 0.64 0.43 0.41 0.76 0.43 0.39 0.70 0.68 0.87 1.44 12.4 2.50Sc 50.7 36.7 53.4 42.7 53.9 29.3 30.3 30.0 33.5 54.3 42.3 40.3 41.2 39.0 34.3 41.2 38.5 44.7 25.5 6.10 15.2V 358 256 358 260 303 268 232 224 248 284 225 231 208 194 189 300 284 316 240 24.0 167Cr 371 108 120 329 283 61 125 392 370 429 349 370 294 381 345 230 222 281 122 5.00 134Co 54.4 28.6 36.8 30.1 42.5 31.7 26.2 41.7 34.5 44.8 40.5 42.0 51.9 48.8 45.5 33.7 33.7 44.5 25.9 3.40 46.5Ni 91.1 26.9 63.3 118 111 52.7 39.5 202 187 113 130 198 125 162 185 80.6 57.8 108 47.5 2.30 140Cu 64.6 37.9 69.2 46.1 64.4 76.4 34.6 46.5 79.9 64.4 58.1 82.8 100 104 104 72.5 61.9 82.5 36.5 3.20 48.6Zn 89.8 72.0 100 76.1 87.0 317 72.9 71.3 69.7 98.4 81.1 83.3 124 72.4 67.0 114 108 116 103 28.0 150Ga 24.2 20.2 25.9 17.7 22.0 18.6 19.3 15.9 17.3 22.1 17.2 17.4 18.8 17.6 16.2 21.7 21.1 23.0 22.7 19.0 24.8Rb 10.0 8.54 15.2 6.92 4.15 12.1 7.02 6.55 7.58 1.71 1.24 1.26 0.12 0.02 0.64 2.08 1.72 2.71 25.8 466 37.0Sr 349 413 422 244 337 337 384 222 274 292 188 188 292 295 246 314 315 388 191 106 1100Y 42.6 16.4 20.9 17.0 23.8 18.4 18.7 16.3 19.1 34.3 27.2 27.1 19.5 18.3 21.2 40.7 39.8 39.6 54.5 62.0 22.0Zr 115 97.6 110 73.0 111 86.5 84.8 74.1 93.1 131.4 85.7 85.6 96.9 84.5 74.1 170 165 177 309 167 277Nb 10.2 8.20 9.51 4.61 11.5 7.59 7.35 4.44 9.54 5.54 2.91 3.00 1.09 0.81 0.55 7.56 7.35 7.70 13.2 40.0 68.0Cs 0.13 0.10 0.19 0.10 0.06 0.17 0.12 0.06 0.09 0.05 0.28 0.28 0.10 0.10 0.19 0.05 0.05 0.05 0.09 38.4 0.49Ba 202 134 209 152 161 177 154 141 152 99.0 44.0 45.0 34.0 21.0 23.0 147 143 178 277 343 526La 12.0 8.23 10.3 7.00 12.0 8.23 9.04 7.08 9.98 9.27 4.68 4.65 4.56 3.31 2.67 12.0 11.5 12.7 25.5 54.0 56.0Ce 25.5 21.1 23.9 14.8 25.1 19.2 19.4 15.5 20.7 22.4 12.2 12.4 12.2 9.52 8.43 27.9 27.0 30.0 55.6 108 105Pr 3.70 2.75 3.26 2.20 3.59 2.78 2.90 2.22 2.95 3.58 2.20 2.19 2.07 1.69 1.58 4.27 4.15 4.59 8.15 12.7 13.2Nd 16.4 12.1 14.3 9.93 15.6 13.1 13.2 9.99 12.9 16.3 11.7 11.7 10.5 8.81 8.60 19.6 19.7 22.0 35.7 47.0 54.0Sm 4.77 3.26 3.76 2.67 4.05 3.39 3.28 2.74 3.27 4.63 3.52 3.50 2.97 2.64 2.70 5.42 5.32 5.71 9.01 9.70 10.2Eu 1.72 1.32 1.53 1.04 1.39 1.28 1.32 1.02 1.24 1.63 1.32 1.31 1.16 1.04 1.06 1.81 1.78 1.94 2.55 0.85 3.20Gd 5.91 3.37 3.97 2.94 4.25 3.51 3.56 3.09 3.74 5.48 4.29 4.29 3.38 2.99 3.32 6.60 6.42 6.44 10.2 9.30 8.50Tb 1.10 0.54 0.67 0.51 0.73 0.60 0.61 0.54 0.65 0.98 0.77 0.78 0.62 0.55 0.60 1.17 1.18 1.13 1.76 1.65 1.20Dy 7.40 3.24 4.22 3.30 4.81 3.92 3.89 3.38 4.04 6.41 5.11 5.11 4.00 3.66 4.08 7.72 7.70 7.74 11.2 10.2 5.60Ho 1.65 0.64 0.89 0.70 0.99 0.80 0.80 0.70 0.83 1.40 1.08 1.09 0.84 0.77 0.85 1.64 1.67 1.64 2.41 2.05 0.88Er 4.73 1.78 2.34 1.98 2.77 2.23 2.21 1.95 2.24 3.89 3.01 3.03 2.27 2.16 2.41 4.66 4.59 4.58 6.63 6.50 2.00Tm 0.72 0.27 0.35 0.31 0.42 0.34 0.34 0.30 0.33 0.58 0.45 0.45 0.34 0.33 0.37 0.69 0.67 0.67 0.97 1.06 0.28Yb 4.67 1.69 2.20 2.00 2.64 2.19 2.18 1.89 2.15 3.78 2.85 2.89 2.07 2.11 2.36 4.38 4.32 4.23 6.21 7.40 1.50Lu 0.73 0.25 0.32 0.30 0.39 0.33 0.34 0.28 0.32 0.59 0.44 0.44 0.29 0.31 0.36 0.67 0.67 0.65 0.98 1.15 0.19Hf 3.32 2.61 3.03 2.11 3.24 2.34 2.28 2.19 2.72 3.74 2.53 2.55 2.68 2.36 2.06 4.66 4.65 4.90 8.57 6.30 6.50Ta 0.63 0.54 0.62 0.26 0.77 0.51 0.50 0.27 0.65 0.30 0.16 0.16 0.09 0.07 0.05 0.50 0.48 0.51 0.90 7.20 4.30Tl 0.07 0.05 0.10 0.05 0.05 0.10 0.13 0.08 0.07 0.04 0.05 0.05 0.05 0.04 0.05 0.04 0.04 0.04 0.17 1.93 0.12Pb 4.39 1.06 1.60 2.76 2.70 13.0 0.90 4.91 1.84 1.27 1.60 1.73 3.29 0.75 1.77 2.17 2.10 4.07 4.16 31.0 4.70Bi 0.01 0.01 0.01 0.01 0.02 0.02 0.00 0.02 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.04 0.04 0.02 0.02 0.53 0.05Th 1.26 0.63 0.88 0.69 1.28 0.61 0.67 0.94 1.04 0.40 0.25 0.26 0.12 0.09 0.05 1.55 1.58 1.66 3.97 54.0 6.00U 1.47 0.23 0.28 0.14 0.33 0.24 0.21 0.30 0.31 0.40 0.06 0.06 0.09 0.03 0.02 0.37 0.36 0.40 0.94 18.8 1.40

Appendix A.338

S.Chenetal./Lithos

224–225

(2015)324

–341

TableA2

Sr–Nd–

Pbisotop

eco

ncen

trations

forCe

nozo

icba

saltsfrom

Japa

nSe

a.

Sample

127-79

4C-1R

127-79

4C-5R

127-79

4C-6R

127-79

4C-9R

127-79

4C-10

R12

7-79

4C-12

R12

8-79

4D-1R

128-79

4D-4R

128-79

4D-8R

128-79

4D-17R

128-79

4D-20R

127-79

7C-9R

127-79

7C-12

R12

7-79

7C-16

R12

7-79

7C-27

R12

7-79

7C-32

R12

7-79

7C-44

R

Interval

(cm)

27–30

89–92

33–39

127–

131

122–

126

33–38

101–

105

25–29

131–

135

50–53

0–3

44–47

29–32

28–31

21–24

79–82

90–93

87Sr/8

6Sr

0.70

4855

0.70

4530

0.70

4510

0.70

5129

0.70

4528

0.70

4405

0.70

4493

0.70

5222

0.70

4727

0.70

4951

0.70

4248

0.70

4155

0.70

4051

0.70

4254

0.70

5125

0.70

5125

0.70

4648

2SE(1

0−6)

914

98

912

1512

1414

1110

1113

1212

13⁎(8

7Sr/8

6Sr) i

0.70

480.70

450.70

450.70

510.70

450.70

440.70

450.70

520.70

470.70

490.70

420.70

420.70

410.70

430.70

510.70

510.70

46143Nd/

144Nd

0.51

2758

0.51

2852

0.51

2855

0.51

2705

0.51

2803

0.51

2887

0.51

2881

0.51

2708

0.51

2822

0.51

2776

0.51

2917

0.51

3035

0.51

3139

0.51

3094

0.51

2862

0.51

2862

0.51

2804

2SE(1

0−6)

1314

1313

1313

1312

913

1315

1111

1111

12⁎ (

143 N

d/14

4 Nd)

i0.51

270.51

280.51

280.51

270.51

280.51

280.51

280.51

270.51

280.51

270.51

290.51

300.51

310.51

300.51

280.51

280.51

28206Pb

/204Pb

18.283

18.092

18.152

18.338

18.146

18.231

18.077

18.341

18.142

18.377

18.225

18.223

17.758

18.143

18.334

18.300

18.340

2SE(1

0−5)

201

362

454

238

200

201

271

220

363

294

255

201

142

236

238

201

220

207Pb

/204Pb

15.598

15.531

15.533

15.589

15.532

15.549

15.535

15.582

15.555

15.610

15.544

15.559

15.463

15.540

15.577

15.580

15.568

2SE(1

0−5)

187

326

388

203

171

171

233

234

249

250

249

218

139

155

218

218

218

208Pb

/204Pb

38.466

38.152

38.198

38.488

38.191

38.318

38.152

38.471

38.271

38.557

38.243

38.304

37.636

38.193

38.413

38.410

38.392

2SE(1

0−5)

423

801

993

500

458

422

610

769

651

578

688

613

376

458

615

576

653

⁎Sr–Ndisotop

ecorrection

form

ula:

(87Sr/8

6 Sr)

i=(8

7Sr/8

6Sr) s–87 R

b/86Sr

×(e

λt−

1),λ

=1.42

×10

−11a−

1 ,87 R

b/86Sr

=Rb

/Sr×

2.98

1;(1

43Nd/

144 N

d)i=

(143Nd/

144Nd)

s–14

7 Sm/1

44Nd×(e

λt−

1),1

47 Sm/1

44 N

d=

Sm/N

d×[0.531

497+

0.14

2521

×(1

43 N

d/144Nd)

s].

339S. Chen et al. / Lithos 224–225 (2015) 324–341

References

Aldanmaz, E., Pearce, J.A., Thirlwall, M.F., Mitchell, J.G., 2000. Petrogenetic evolution of lateCenozoic, post-collision volcanism in western Anatolia, Turkey. Journal of Volcanolo-gy and Geothermal Research 102, 67–95.

Allan, J.F., Gorton, M.P., 1992. Geochemistry of igneous rocks from Legs 127 and 128, Seaof Japan. In: Tamaki, K., Suyehiro, K., Allan, J., McWilliams, S. (Eds.), Proceedings of theOcean Drilling Program, Scientific Results: 127/128 Pt. 2. Ocean Drilling Program,pp. 905–929.

Baksi, A.K., 1995. Fine tuning the radiometrically derived geomagnetic polarity time scale(GPTS) for 0–10 Ma. Geophysical Research Letters 22, 457–460.

Baksi, A.K., 2003. Critical evaluation of 40Ar/39Ar ages for the Central Atlantic MagmaticProvince: timing, duration and possible migration of magmatic centers. The CentralAtlantic Magmatic Province: Insights from Fragments of Pangeapp. 77–90.

Baksi, A.K., 2005. Evaluation of radiometric ages pertaining to rocks hypothesized to havebeen derived by hotspot activity, in and around the Atlantic, Indian, and PacificOceans. Geological Society of America Special Papers 388, 55–70.

Baksi, A.K., 2007. A quantitative tool for detecting alteration in undisturbed rocks andminerals—I: Water, chemical weathering, and atmospheric argon. Geological Societyof America Special Papers 430, 285–303.

Baksi, A.K., 2014. 40Ar/39Ar ages of flood basalts provinces in Russia and China and theirpossible link to global faunal extinction events: a cautionary tale regarding alterationand loss of 40Ar*. Journal of Asian Earth Sciences 84, 118–130.

Barnes, S.J., Naldrett, A.J., Gorton, M.P., 1985. The origin of the fractionation of platinumgroup elements in terrestrial magmas. Chemical Geology 53, 303–323.

Barnes, D.A., Thy, P., Renne, P., 1992. Sedimentology, phenocryst chemistry, and age-Miocene “Blue Tuff”: Sites 794 and 796, Japan Sea. In: Tamaki, K., Suyehiro, K.,Allan, J., McWilliams, S. (Eds.), Proceedings of the Ocean Drilling Program, ScientificResults: 127/128 Pt. 1. Ocean Drilling Program, pp. 115–130.

Beswick, A.E., 1982. Some geochemical aspects of alteration and genetic relations inkomatiitic suites. In: Arndt, N.T., Nisbet, E.G. (Eds.), Komatiities. George Allen andUnwin, London, pp. 283–308.

Celaya, M., McCabe, R., 1987. Kinematic model for the opening of the Sea of Japan and thebending of the Japanese islands. Geology 15, 53–57.

Chen, S.S., Liu, J.Q., Guo, Z.F., Chen, S.S., Sun, C.Q., 2015. Geochemical characteristics ofvolcanic rocks from Site 794, Yamato Basin: implications for deep mantle processesof the Japan Sea. Acta Geologica Sinica (submitted for publication, English Edition).

Cousens, B.L., Allan, J.F., 1992. A Pb, Sr, and Nd isotopic study of basaltic rocks from the Seaof Japan, Legs 127/128. In: Tamaki, K., Suyehiro, K., Allan, J., McWilliams, S. (Eds.),Proceedings of the Ocean Drilling Program, Scientific Results: 127/128 Pt. 2. OceanDrilling Program, pp. 805–817.

Cousens, B.L., Allan, J.F., Gorton, M.P., 1994. Subduction-modified pelagic sediments as theenriched component in back-arc basalts from the Japan Sea: Ocean Drilling ProgramSites 797 and 794. Contribution to Mineralogy and Petrology 117, 421–434.

Davidson, J.P., 1987. Crustal contamination versus subduction zone enrichment: exam-ples from the Lesser Antilles and implications for mantle source compositions ofisland arc volcanic rocks. Geochimica et Cosmochimica Acta 51, 2185–2198.

Dungan, M.A., Lindstrom, M.M., MeMiland, N.J., 1986. Open systemmagmatic evolution ofthe Taos Plateau volcanic field, northern New Mexico, the petrology and geochemis-try of the Servillets basalt. Journal of Geophysical Research 91, 5999–6028.

Ellam, R.M., Hawkesworth, C.J., 1988. Elemental and isotopic variations in subductionrelated basalts: evidence for a three component model. Contributions to Mineralogyand Petrology 98, 72–80.

Ewart, A., Hawkesworth, C.J., 1987. The Pleistocene–Recent Tonga-Kermadec arc lavas:interpretation of new isotopic and rare earth data in terms of a depleted mantlesource model. Journal of Petrology 28, 495–530.

Fan, Q., Liu, R., Li, D., Li, Q., 1999. Significance of K–Ar age of bimodal volcanic rocks atWangtian'e volcano, Changbaishan area. Chinese Science Bulletin 44, 660–663.

Fan, W.M., Guo, F., Wang, Y.J., Ge, L., 2003. Late Mesozoic calcalkaline volcanism of post-orogenic extension in the northern Da Hinggan Mountains, northeastern China.Journal of Volcanology and Geothermal Research 121, 115–135.

Frey, F.A., Green, D.H., Roy, S.D., 1978. Integratedmodels of basalt petrogenesis: a study ofquartz tholeiites to olivine melilitites from south eastern Australia utilizing geochem-ical and experimental petrological data. Journal of Petrology 19, 463–513.

Fukase, M., Shuto, K., 2000. Petrology of Oligocene volcanic rocks from the middle part oftheMonzen Formation, Oga Peninsula, northeastern Japan. The Journal of the Geolog-ical Society of Japan 106, 280–298 (in Japanese with English abstract).

Gao, S., Ling, W., Qiu, Y., Lian, Z., Hartmann, G., Simon, K., 1999. Contrasting geochemicaland Sm–Nd isotopic compositions of Archean metasediments from the Konglinghigh-grade terrain of the Yangtze craton: evidence for cratonic evolution and redistri-bution of REE during crustal anatexis. Geochimica et Cosmochimica Acta 63,2071–2088.

Garfunkel, Z., Anderson, C.A., Schubert, G., 1986. Mantle circulation and the lateral migra-tion of subducted slabs. Journal of Geophysical Research: Solid Earth (1978–2012) 91,7205–7223.

Gill, J.B., 1981. Orogenic Andesites and Plate Tectonics vol. 390. Springer-Verlag, Berlin.Green, D.H., Hibberson, W.O., Jaques, A.L., 1979. Petrogenesis of mid-ocean ridge basalts.

The Earth: Its Origin, Structure and Evolution. Academic Press, London, pp. 265–290.Guo, Z., Hertogen, J., Liu, J., Pasteels, P., Boven, A., Punzalan, L., He, H., Luo, X., Zhang, W.,

2005. Potassic magmatism in western Sichun and Yunnan provinces, SE Tibet,China: petrological and geochemical constraints on petrogenesis. Journal of Petrology46, 33–78.

Guo, Z., Wilson, M., Liu, J., Mao, Q., 2006. Post-collisional, potassic and ultrapotassicmagmatism of the northern Tibetan Plateau: constraints on characteristics of themantle source, geodynamic setting and uplift mechanisms. Journal of Petrology 47,1177–1220.

340 S. Chen et al. / Lithos 224–225 (2015) 324–341

Guo, Z., Wilson,M., Liu, J., 2007. Post-collisional adakites in south Tibet: products of partialmelting of subduction-modified lower crust. Lithos 96, 205–224.

Hager, B.H., O'Connell, R.J., Raefsky, A., 1983. Subduction, back-arc spreading and globalmantle flow. Tectonophysics 99, 165–189.

Hanson, G.N., 1989. An approach to trace element modeling using a simple igneoussystem as an example. Reviews in Mineralogy and Geochemistry 21, 79–97.

Hart, S.R., 1984. A large-scale isotope anomaly in the Southern Hemisphere. Nature 309,753–756.

Hilde, T.W., Wageman, J.M., 1973. Structure and origin of the Japan Sea. In: Coleman, P.J.(Ed.), The Western Pacific-Island Arcs Marginal Seas Geochemistry. University ofWestern Australia Press, pp. 415–434.

Hofmann, A.W., 1988. Chemical differentiation of the Earth: the relationship betweenmantle, continental crust, and oceanic crust. Earth and Planetary Science Letters 90,297–314.

Ikeda, Y., 1988. Geochemistry of Miocene back-arc basin basalts from northeast Hokkaido,Japan. The Journal of the Geological Society of Japan 104, 99–106 (in Japanese withEnglish abstract).

Ikeda, Y., Stern, R.J., Kagami, H., Sun, C.H., 2000. Pb, Nd, and Sr isotopic constraints on theorigin of Miocene basaltic rocks from northeast Hokkaido, Japan: implications foropening of the Kurile back-arc basin. The Island Arc 9, 161–172.

Irvine, T., Baragar,W., 1971. A guide to the chemical classification of the common volcanicrocks. Canadian Journal of Earth Sciences 8, 523–548.

Ishimoto, H., Shuto, K., Goto, Y., 2006. Middle Miocene to Quaternary primary basalt andhigh magnesian andesite magmas of North Hokkaido, Japan: source mantle charac-teristics and degrees of partial melting. Island Arc 15, 251–268.

Jolivet, L., Tamaki, K., 1992. Neogene kinematics in the Japan Sea region and volcanicactivity of the northeast Japan arc. Proceedings of the Ocean Drilling Program,Scientific Results vol. 127, No. 128, pp. 1311–1331.

Jolivet, L., Shibuya, H., Fournier, M., 1995. Paleomagnetic rotations and the Japan Seaopening. Active margins and marginal basins of the western Pacific, pp. 355–369.

Kaneoka, I., Notsu, K., Takigami, Y., Fujioka, K., Sakai, H., 1990. Constraints on the evolutionof the Japan Sea based on 40Ar–39Ar ages and Sr isotopic ratios for volcanic rocks ofthe Yamato Seamount chain in the Japan Sea. Earth and Planetary Science Letters 97,211–225.

Kaneoka, I., Takigami, Y., Takaoka, N., Yamashita, S., Tamaki, K., 1992. AR–AR analysis ofvolcanic rocks recovered from the Japan Sea floor: constraints on the age of formationof the Japan. In: Tamaki, K., Suyehiro, K., Allan, J., McWilliams, S. (Eds.), Proceedings ofthe Ocean Drilling Program, Scientific Results: 127/128 Pt. 2. Ocean Drilling Program,pp. 819–836.

Karig, D.E., 1971. Origin and development of marginal basins in the western Pacific.Journal of Geophysical Research 76, 2542–2561.

Karig, D.E., 1974. Evolution of arc systems in the western Pacific. Annual Review of Earthand Planetary Sciences 2, 51.

Kersting, A.B., Arculus, R.J., Gust, D.A., 1996. Lithospheric contributions to arcmagmatism: Isotope variations along strike in volcanoes of Honshu, Japan. Sci-ence 272, 1464–1468.

Kimura, J.I., Manton, W.I., Sun, C.H., Iizumi, S., Yoshida, T., Stern, R.J., 2002. Chemicaldiversity of the Ueno basalts, central Japan: identification of mantle and crust contri-butions to arc basalts. Journal of Petrology 43, 1923–1946.

Kondo, H., Shuto, K., Fukase, M., 2000. An AFC (assimilation and fractional crystallization)as the petrogenesis of andesites from the Pliocene Myojin-iwa Formation, the back-arc side of the Northeast Japan: combined major-and trace-element and Sr–Ndisotope constraints. The Journal of the Geological Society of Japan 106, 426–441(in Japanese with English abstract).

Kramers, J.D., Tolstikhin, I.N., 1997. Two terrestrial lead isotope paradoxes, forward trans-port modelling, core formation and the history of the continental crust. ChemicalGeology 139, 75–110.

Lallemand, S., Jolivet, L., 1986. Japan Sea: a pull-apart basin? Earth and Planetary ScienceLetters 76, 375–389.

Le Bas, M.J., Le Maitre, R.W., Streckeisen, A., Zanettin, B., 1986. A chemical classification ofvolcanic rocks based on the total alkali-silica diagram. Journal of Petrology 27,745–750.

Liu, J.Q., Han, J., Fyfe, W.S., 2001. Cenozoic episodic volcanism and continental rifting innortheast China and possible link to Japan Sea development as revealed from K–Argeochronology. Tectonophysics 339 (3), 385–401.

Miyashita, S., Tsuchiya, N., Ikeda, Y., Sakamoto, I., 1995. Petrology of the Okushiri RidgeBasalts in the Japan Sea Basin: arc-and plume-type magma series. Memoirs of theGeological Society of Japan 44, 1–21.

Nohda, S., 2009. Formation of the Japan Sea basin: reassessment from Ar–Ar ages andNd–Sr isotopic data of basement basalts of the Japan Sea and adjacent regions.Journal of Asian Earth Sciences 34, 599–609.

Nohda, S., Wasserburg, G.J., 1986. Trends of Sr and Nd isotopes through time nearthe Japan Sea in northeastern Japan. Earth and Planetary Science Letters 78,157–167.

Nohda, S., Tatsumi, Y., Otofuji, Y., Matsuda, T., Ishizaka, K., 1988. Asthenospheric injectionand back-arc opening: isotopic evidence from Northeast Japan. Chemical Geology 68,317–327.

Ohki, J.I., Shuto, K., Kagami, H., 1995. Middle Miocene bimodal volcanism by astheno-spheric upwelling: Sr and Nd isotopic evidence from the back-arc region of theNortheast Japan arc. Oceanographic Literature Review 42 (8).

Okamura, S., Arculus, R.J., Martynov, Y.A., Kagami, H., Yoshida, T., Kawano, Y., 1998a.Multiple magma sources involved in marginal-sea formation: Pb, Sr, and Nd isotopicevidence from the Japan Sea region. Geology 26, 619–622.

Okamura, S., Martynov, Y.A., Furuyama, K., Nagao, K., 1998b. K–Ar ages of the basalticrocks from Far East Russia: constraints on the tectono-magmatism associated withthe Japan Sea opening. The Island Arc 7, 271–282.

Okamura, S., Arculus, R.J., Marthynov, Y.A., 2005. Cenozoic magmatism on the north-eastern Eurasia margin: the role of lithosphere versus asthenosphere. Journal ofPetrology 46, 221–253.

Othman, B.D., White, W.M., Patchett, J., 1989. The geochemistry of marine sediments,island arc magma genesis, and crust–mantle recycling. Earth and Planetary ScienceLetters 94, 1–21.

Otofuji, Y., Matsuda, T., 1984. Timing of rotational motion of Southwest Japan inferredfrom paleomagnetism. Earth and Planetary Science Letters 70, 373–382.

Otofuji, Y., Matsuda, T., Nohda, S., 1985. Paleomagnetic evidence of the Miocene counter-clockwise rotation of northeast Japan-rifting process of the Japan arc. Earth andPlanetary Science Letters 75, 265–277.

Otofuji, Y., Kambara, A., Matsuda, T., Nohda, S., 1994. Counterclockwise rotation ofNortheast Japan: paleomagnetic evidence for regional extent and timing of rotation.Earth and Planetary Science Letters 121, 503–518.

Pearce, J.A., 1983. In: Hawkesworth, C.J., Norry, M.J. (Eds.), Continental Basalts and MantleXenolith (Shaiva Geology Series). Shiva, pp. 230–249.

Pearce, J.A., Cann, J.B., 1973. Tectonic setting of basic volcanic rocks determined usingtrace element analyses. Earth and Planetary Science Letters 19, 290–300.

Pearce, J.A., Peate, D.W., 1995. Tectonic implications of the composition of volcanic arcmagmas. Annual Review of Earth and Planetary Sciences 23, 251–285.

Perfit, M.R., Gust, D.A., Bence, A.E., Arculus, R.J., Taylor, S.R., 1980. Chemical characteristicsof island-arc basalts: implications for mantle sources. Chemical Geology 30, 227–256.

Plank, T., 2005. Constraints from Thorium/Lanthanum on sediment recycling at subduc-tion zones and the Evolution of the Continents. Journal of Petrology 46, 921–944.

Pouclet, A., Bellon, H., 1992. Geochemistry and isotopic composition of volcanic rocksfrom the Yamato Basin: Hole 794D, Sea of Japan. In: Tamaki, K., Suyehiro, K., Allan,J., McWilliams, S. (Eds.), Proceedings of the Ocean Drilling Program, Scientific Results:127/128 Pt. 2. Ocean Drilling Program, pp. 779–789.

Pouclet, A., Lee, J.-S., Vidal, P., Cousens, B., Bellon, H., 1995. Cretaceous to Cenozoic volca-nism in South Korea and in the Sea of Japan: magmatic constraints on the opening ofthe back-arc basin. In: Smellie, J.L. (Ed.), Volcanism Associatedwith Extension at Con-suming Plate Margins. Geological Society Special Publication, Lomdon, pp. 169–191.

Sato, M., Shuto, K., Yagi, M., 2007. Mixing of asthenospheric and lithospheric mantle-derived basalt magmas as shown by along-arc variation in Sr and Nd isotopic compo-sitions of Early Miocene basalts from back-arc margin of the NE Japan arc. Lithos 96,453–474.

Saunders, A.D., Tarney, J., 1984. Geochemical characteristics of basaltic volcanism withinback-arc basins. Geological Society, London, Special Publications 16, 59–76.

Saunders, A.D., Storey, M., Kent, R.W., Norry, M.J., 1992. Consequences of plume-lithosphere interactions. Geological Society, London, Special Publications 68, 41–60.

Schilling, J.G., 1975. Rare-Earth variations across ‘normal segments’ of the ReykjanesRidge, 60°–53° N, Mid-Atlantic Ridge, 29°S, and East Pacific Rise, 2°–19°S, andevidence on the composition of the underlying low-velocity layer. Journal of Geo-physical Research 80, 1459–1473.

Shibata, T., Nakamura, E., 1997. Across-arc variations of isotope and trace element compo-sitions from Quaternary basaltic volcanic rocks in northeastern Japan: implicationsfor interaction between subducted oceanic slab and mantle wedge. Journal of Geo-physical Research: Solid Earth (1978–2012) 102, 8051–8064.

Shimazu, M., Yoon, M., Tateishi, M., 1990. Tectonics and volcanism in the Sado-PohangBelt from 20 to 14 Ma and opening of the Yamato Basin of the Japan Sea.Tectonophysics 181, 321–330.

Shuto, K., 1988. Miocene abyssal tholeiitic-type basalts from Tobishima Island, easternmargin of the Japan Sea. Journal of Mineralogy, Petrology and Economic Geology83, 257–272 (in Japanese with English abstract).

Shuto, K., Ohki, J., Kagami, H., Yamamoto, M.,Watanabe, N., Yamamoto, K., Anzai, N., Itaya,T., 1993. The relationships between drastic changes in Sr isotope ratios of magmasources beneath the NE Japan arc and the spreading of the Japan Sea back-arcbasin. Mineralogy and Petrology 49, 71–90.

Shuto, K., Kato, S., Ohki, J., Kagami, H., Arato, H., Rezanov, A.L., 1997. Miocene bimodalvolcanism in the Niigata oil and gas fields, northeast Japan—with special referenceto the spreading of the back-arc basin. Journal of the Japanese Association for Petro-leum Technology 62, 45–58 (in Japanese with English abstract).

Shuto, K., Hirahara, Y., Ishimoto, H., Aoki, A., Jinbo, A., Goto, Y., 2004. Sr and Nd isotopiccompositions of the magma source beneath north Hokkaido, Japan: comparisonwith the back-arc side in the NE Japan arc. Journal of Volcanology and GeothermalResearch 134, 57–75.

Shuto, K., Ishimoto, H., Hirahara, Y., Sato, M., Matsui, K., Fujibayashi, N., Takazawa, E.,Yabuki, K., Sekine, M., Kato, M., Rezanov, A.I., 2006. Geochemical secular variationof magma source during Early to Middle Miocene time in the Niigata area, NEJapan: Asthenospheric mantle upwelling during back-arc basin opening. Lithos 86,1–33.

Stern, R.J., Bibee, L.D., 1984. Esmeralda Bank: geochemistry of an active submarine volca-no in theMariana Island Arc. Contributions toMineralogy and Petrology 86, 159–169.

Stern, R.J., Lin, P.N., Morris, J.D., Jackson, M.C., Fryer, P., Bloomer, S.H., Ito, E., 1990. Enrichedback-arc basin basalts from the northern Mariana Trough: implications for the mag-matic evolution of back-arc basins. Earth and Planetary Science Letters 100, 210–225.

Storey, M., Mahoney, J.J., Saunders, A.D., 1997. Cretaceous basalts in Madagascar and thetransition between plume and continental lithosphere mantle sources. Large IgneousProvinces: Continental, Oceanic and Planetary Flood Volcanism. American Geophysi-cal Union, Geophysical Monograph vol. 100, pp. 95–122 (Washington, DC, USA).

Su, Y., Zheng, J., Griffin,W.L., Zhao, J., Tang, H., Ma, Q., Lin, X., 2012. Geochemistry and geo-chronology of Carboniferous volcanic rocks in the eastern Junggar terrane, NW China:Implication for a tectonic transition. Gondwana Research 22, 1009–1029.

Sun, S.S., McDonough, W.F., 1989. Chemical and isotopic systematics of oceanic basalts:implications for mantle composition and processes. Geological Society, London,Special Publications 42, 313–345.

341S. Chen et al. / Lithos 224–225 (2015) 324–341

Takahashi, E., Kushiro, I., 1983. Melting of a dry peridotite at high pressures and basaltmagma genesis. American Mineralogist 68, 859–879.

Takanashi, K., Shuto, K., Sato, M., 2011. Origin of Late Paleogene to Neogene basalts andassociated coeval felsic volcanic rocks in Southwest Hokkaido, northern NE Japanarc: constraints from Sr and Nd isotopes and major-and trace-element chemistry.Lithos 125, 368–392.

Tamaki, K., 1985. Two modes of back-arc spreading. Geology 13, 475–478.Tamaki, K., Pisciotto, K., Allan, J., et al. (Eds.), 1990. Proceeding of ODP Initial Reports.

Ocean Drilling Program 127.Tatsumi, Y., Nohda, S., Ishizaka, K., 1988. Secular variation of magma source compositions

beneath the Northeast Japan arc. Chemical Geology 68, 309–316.Tatsumoto, M., Nakamura, Y., 1991. DUPAL anomaly in the Sea of Japan: Pb, Nd, and Sr

isotopic variations at the eastern Eurasian continental margin. Geochimica etCosmochimica Acta 55, 3697–3708.

Taylor, S.R., McLennan, S., 1995. The geochemical evolution of the continental crust.Reviews of Geophysics 33, 241–265.

Thompson, R.N., Morrison, M.A., 1988. Asthenospheric and lower-lithospheric mantlecontributions to continental extension magmatism: an example from the BritishTertiary Province. Chemical Geology 68, 1–15.

Ujike, O., Tsuchiya, N., 1993. Geochemistry of Miocene basaltic rocks temporally strad-dling the rifting of lithosphere at the Akita-Yamagata area, northeast Japan. ChemicalGeology 104, 61–74.

Volpe, A.M., Macdougall, J.D., Lugmair, G.W., Hawkins, J.W., Lonsdale, P., 1990. Fine-scaleisotopic variation in Mariana Trough basalts: evidence for heterogeneity and arecycled component in backarc basin basalts. Earth and Planetary Science Letters100, 251–264.

Vorontsov, A.A., Yarmolyuk, V.V., Fedoseev, G.S., Nikiforov, A.V., Sandimirova, G.P., 2010. Iso-topic and geochemical zoning of Devonian magmatism in the Altai–Sayan rift system:composition and geodynamic nature of mantle sources. Petrology 18, 596–609.

Wang, S.S., Liu, J.Q., Zhu,M., Zhao, D.Z., 1983. K–Ar dating of the Cenozoic volcanic rocks inthe Changbaishan region. Scientia Geologica Sinica 3, 205–216.

Wang, K., Plank, T., Walker, J.D., Smith, E.I., 2002. Amantlemelting profile across the Basinand Range, SW USA. Journal of Geophysical Research 107 (ECV-5).

Weaver, B.L., 1991. The origin of ocean island basalt end-member compositions: traceelement and isotopic constraints. Earth and Planetary Science Letters 104, 381–397.

Xie, Z., Zheng, Y.F., Zhao, Z.F., Wu, Y.B., Wang, Z., Chen, J., Liu, X., Wu, F.Y., 2006. Mineralisotope evidence for the contemporaneous process of Mesozoic granite emplacementand gneiss metamorphism in the Dabie orogen. Chemical Geology 231, 214–235.

Yagi, M., Hasenaka, T., Ohguchi, T., Baba, K., Sato, H., Ishiyama, D., Mizuta, T., Yoshida,T., 2001. Transition of magmatic composition reflecting an evolution of riftingactivity—a case study of the Akita-Yamagata basin in Early to Middle Miocene,Northeast Honshu, Japan. Japanese Magazine of Mineralogical and PetrologicalSciences 30, 265–287 (in Japanese with English abstract).

Yamashita, S., Fujii, T., 1992. Experimental petrology of basement basaltic rocks from Sites794 and 797, Japan Sea. Proceeding of the Ocean Drilling Program, Scientific Results127/128, 891–898.

Yamashita, S., Shuto, K., Kakihara, Y., Kagami, H., 1999. Coeval volcanism due to interac-tion of back-arc basin basalt (BABB) magmawith the island-arc crust in the late Mio-cene Engaru volcanic field, northeastern Hokkaido, Japan: the evidence of Sr and Ndisotopic ratios combined with major- and trace-element compositions. The Journal ofthe Geological Society of Japan 105, 625–642 (in Japanese with English abstract).

Yuan, C., Sun, M., Wilde, S., Xiao, W.J., Xu, Y.G., Long, X.P., Zhao, G.C., 2010. Post-collisionalplutons in the Balikun area, East Chinese Tianshan: evolving magmatism in responseto extension and slab break-off. Lithos 119, 269–288.

Zhang, H.F., Sun, M., Zhou, X.H., Fan, W.M., Zhai, M.G., Yin, J.F., 2002. Mesozoic lithospheredestruction beneath the North China Craton: evidence from major-, trace- elementand Sr–Nd–Pb isotope studies of Fangcheng basalts. Contributions to Mineralogyand Petrology 144, 241–253.

Zhang, Z.C., Mahoney, J.J., Mao, J.W., Wang, F.S., 2006. Geochemistry of picritic and associ-ated basalt flows of the western Emeishan flood basalt province, China. Journal ofPetrology 47, 1997–2019.

Zhao, J.H., Zhou, M.F., 2007. Geochemistry of Neoproterozoic mafic intrusions in thePanzhihua district (Sichuan Province, SW China): implications for subduction-related metasomatism in the upper mantle. Precambrian Research 152, 27–47.

Zhao, J.H., Zhou, M.F., Zheng, J.P., 2010. Metasomatic mantle source and crustal contami-nation for the formation of the Neoproterozoic mafic dike swarm in the northernYangtze Block, South China. Lithos 115, 177–189.

Zindler, A., Hart, S.R., 1986. Chemical geodynamics. Annual Review of Earth and PlanetarySciences 14, 493–571.