Early Jurassic Subduction of the Paleo-Pacific Ocean in NE China: Petrologic and Geochemical Evidence from the Tumen Mafic Intru

Lithos 224–225 (2015) 46–60

Contents lists available at ScienceDirect

Lithos

journal homepage: www.elsevier.com/locate/lithos

Early Jurassic subduction of the Paleo-Paciﬁc Ocean in NE China: Petrologic and geochemical evidence from the Tumen maﬁc intrusive complex

Feng Guo a,⁎,HongxiaLia, Weiming Fan a,JingyanLia,b, Liang Zhao a, Miwei Huang a,b,WenliangXuc a State Key Laboratory of Isotope Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou 510640, China b University of Chinese Academy of Sciences, Beijing 100049, China c School of Earth Sciences, Jilin University, Changchun 130062, China article info abstract

Article history: Subduction of the Paleo-Pacific Oceanic Plate is widely considered to have caused extensive Mesozoic Received 3 December 2014 magmatism, lithospheric deformation and mineralization in East Asia. However, it is still unclear when this sub- Accepted 10 February 2015 duction began. Here we report an Early Jurassic (~187 Ma) mafic intrusive complex (including olivine norite, Available online 21 February 2015 gabbro, and diorite) from the Tumen area in NE China. The olivine norite contains a mineral assemblage of olivine, pyroxene, Ca-plagioclase, and hornblende that crystallized in a water-saturated parental magma. The rocks in the Keywords: complex show variable degrees of plagioclase and ferromagnesian mineral accumulation as reflected by positive Paleo-PacificOceanicsubduction Hydrous mineral assemblage Sr and Eu anomalies in primitive mantle-normalized incompatible element patterns. Mass-balance calculations Arc-type geochemistry indicate that the parental magma was calc-alkaline with arc-type trace element features (i.e., large ion incompat- Maficintrusion ible and light rare earth element enrichment and Nb–Ta depletion). It also had Sr–Nd–Hf isotopic compositions 87 86 Early Jurassic ( Sr/ Sr(i) = 0.7042 to 0.7044, εNd(t) = +2.5 to +3.5 and εHf(t) = +8.4 to +10.5) similar to those of modern NE China arc basalts. The parental magma was likely derived from 5 to 20% melting of a mantle wedge metasomatized by an addition of 3–4% hydrous sediment melt from the subducting Paleo-Pacific Oceanic slab. The Tumen mafic intrusive complex, together with other contemporaneous mafic intrusions, I-type granitoids, and felsic lavas, constitutes an Early Jurassic N–S-trending arc magmatic belt that was formed by westward subduction of the Paleo-PacificOcean. © 2015 Elsevier B.V. All rights reserved.

1. Introduction of converging plates and distribution of the related arc magmas are poorly constrained, such as the Mesozoic subduction zone in NE China. Mafic magmas in convergent plate margins are generally enriched in The Mesozoic geology of the East Asian continent is characterized by large ion lithophile elements (LILEs), light rare earth elements (LREEs) extensive magmatism, lithospheric deformation, metamorphism, and and depleted in high field strength elements (HFSEs), and have evolved metal mineralization. Tectonic reconstruction has suggested a predom- isotopic compositions relative to mid-oceanic ridge basalts (MORBs). inant role of the interaction between the subducted Paleo-Pacific Such geochemical differences have been widely attributed to mantle (Izanagi and Fallalon Plate) Oceanic Plate and the continental litho- enrichment caused by recycled crustal components (e.g., Arculus, sphere in eastern Asia (Engebretson et al., 1985; Faure and Natal'in, 1994; Macdonald et al., 2000; Tatsumi and Eggins, 1995). In subduction 1992). In NE China, the subduction of the Izanagi Plate has been zones, mafic magmas are mainly formed through the hydrous melting suggested to have triggered the Early Mesozoic blueschist-facies meta- of a mantle wedge, in which the mantle has been chilled by subduction morphism of the Heilongjiang metamorphic complex (Wu et al., 2007; of oceanic crust and the overlying sediments, and magma generation is Zhou and Wilde, 2013 and references therein), emplacement of volumi- attributed to lowering of the peridotite solidus by the introduction of nous I-type granitoids during 210–155 Ma (Wu et al., 2011), eruption of

H2O and other volatiles from the subducting slab (Arculus, 1994; intermediate-felsic volcanic rocks during 228–202 Ma (Wang et al., Davies and Stevenson, 1992; Fan et al., 2003; Tatsumi and Eggins, 2015; Xu et al., 2012, 2013), and widespread porphyry Cu-Mo mineral- 1995). Such magmas thus have high contents of water or volatiles and ization during 180–160 Ma (e.g., Ge et al., 2007; Pirajno and Santosh, crystallize hydrous mineral assemblages. Nevertheless, it is rather diffi- 2014; Zeng et al., 2013). Zhou et al. (2014) recently reported the zircon cult to identify paleo-subduction zones in areas where the configuration U–Pb ages of the Nadanhada accretionary complex in NE China, and suggested that the subduction of the Paleo-Pacific Ocean was initiated ⁎ Corresponding author. Tel.: +86 20 85290280; fax: +86 20 85290130. in the late Triassic to early Jurassic. Yet, the exact time that the Izanagi E-mail address: [email protected] (F. Guo). Plate began to subduct beneath the NE China continent is debatable.

Although those I-type granitoids have been considered to be genetically 1989; Safonova and Santosh, 2014; Sengör et al., 1993; Tang, 1990; linked with the subduction of the Izanagi Plate (e.g., Wu et al., 2011), Xiao et al., 2003; Zhou and Wilde, 2013). The history of tectonic evolu- subduction-related mafic magmas have been rarely documented. tion includes: (1) subduction and accretion of the Paleo-Asian Ocean Although a few Early Jurassic mafic intrusions were identified in NE during Paleozoic to form the giant Central Asian Orogenic Belt; and China (e.g., Yu et al., 2012), whether their origin was formed by the (2) northwestward subduction and accretion of the Paleo-Pacific Plate subduction of the Paleo-Pacific Ocean is poorly constrained. to the Eurasia Continent, thereby controlling the evolution of the In this paper, we present a comprehensive study (including zircon East Asian continental margin since the Mesozoic (Guo et al., 2007; U–Pb dating, petrologic, mineralogical, and geochemical data) on an Maruyama et al., 1989; Wu et al., 2011). Early Jurassic subduction-related mafic intrusive complex in the The Tumen area is located at the junction of China, Russia and Korea, Tumen area (NE China), and conduct additional Sr–Nd–Hf isotopic ~100 km west of the Japan Sea. It is considered to be part of the orogenic analyses on the other contemporaneous mafic intrusions reported in collage between the North China Craton to the south and the Jiamusi– Yu et al. (2012). These data will enable estimation of the compositions Khanka Massif to the northeast (Fig. 1a; Guo et al., 2007). Phanerozoic and natures of the parental magmas of these early Jurassic mafic granitoids are widely exposed over an area of more than 20,000 km2 intrusions and allow us to provide further insight into the dominant and occupy ~70% of the region. The emplacement of the Phanerozoic enrichment processes operating in the mantle wedge through slab granitoids occurred from the Late Paleozoic (285 Ma) to Early Creta- subduction. By integrating these results with previous petrologic, ceous (112 Ma), and the dominant emplacement ages are between geochemical, and geochronological studies on Mesozoic igneous rocks 210 and 155 Ma (Li et al., 2012; Wu et al., 2011; Zhang et al., 2004). In in the area (e.g., Guo et al., 2010; Wu et al., 2011; Yu et al., 2012), the addition, extensive eruption of predominant intermediate-felsic lavas framework of the Mesozoic arc magmatic belt related to the subduction occurred from the Late Triassic to Early Cretaceous (228–106 Ma) of the Paleo-Pacific Ocean can be outlined. (Li et al., 2007; Wang et al., 2015; Xu et al., 2009). The Tumen mafic intrusive complex is intruded by Jurassic granit- 2. Geological backgrounds oids. It was previously grouped as a Late Paleozoic mafic intrusion (BGMRJL, 1989). The rock types include hornblende-bearing olivine NE Asia comprises a series of subduction-related accretionary norite, gabbro, and diorite. The studied samples are fresh and have orogens distributing among the North China Craton, Siberian Craton poikilitic to granoblastic textures (Fig. 2). The olivine is fresh and and Pacific Plate (Fig. 1a, Guo et al., 2010; Li, 2006; Maruyama et al., euhedral, and it is enclosed in clinopyroxene and hornblende in local

Fig. 1. A tectonic sketch of the northeastern Asian continent (a, modified after Zhou and Wilde, 2013), distribution of Phanerozoic intrusions in NE China (b, modified after Wu et al., 2011 and Yu et al., 2012) and a simplified geological map showing the location of Tumen mafic intrusive complex (c, modified after BGMRJL, 1989; Li, 2012). 48 F. Guo et al. / Lithos 224–225 (2015) 46–60

Fig. 2. Microphotos of the Tumen mafic intrusive complex (+). Photos (a) and (b) — olivine norite (06TM-8), showing ferromagnesian accumulation of olivine and pyroxene (a) and olivine inclusion in the hornblende (b); photo (c) — gabbro (06TM-6), showing gabbroic texture with plagioclase and magnetite inclusions in orthopyroxene, indicating plagioclase and magnetite crystallization prior to orthopyroxene; and (d) — diorite (06TM-1). Mineral abbreviations: Ol — olivine; Opx — orthopyroxene; Cpx — clinopyroxene; Pl — plagioclase; Hb — hornblende; Mt — magnetite. areas, indicating that it is an early crystallized phase. The clinopyroxene Sciences (CAS). The normal operating conditions were 15 kV accelerat- is green and belongs to diopside, whereas the orthopyroxene belongs to ing voltage, 10 nA beam current, and 1–2 μm beam diameter. Extended hypersthene. The plagioclase is euhedral to subhedral and displays al- counting time was used to measure olivine following the method in bite twinning and zoning. Some plagioclase grains are enclosed within Sobolev et al. (2007). Based on repeated analyses of natural and orthopyroxene, indicating crystallization of the plagioclase prior to synthetic standards, the relative analytical uncertainty is b2% for orthopyroxene (Fig. 2c). The hornblende is subhedral with polychroism major elements and b5% for minor elements. The atomic number- and occurs as a major crystallizing phase in diorite and gabbroic diorite absorption-fluorescence (ZAF) correction procedure was used for data (Fig. 2d). reduction. The chemical compositions of the olivine, pyroxene, plagio- In the Lesser Hinggan Mountains and Zhangguangcailing Range clase, and hornblende are attached in Appendix 1. (Fig. 1b), several early Jurassic mafic intrusions also outcrop. The rock types include olivine norite, gabbro, hornblendite, and diorite (Yu 3.2. SHRIMP zircon U–Pb dating et al., 2012). The modal mineral compositions of the Tumen and other fi ma c intrusions are summarized in Table 1. Zircon U–Pb dating was conducted using a sensitive high-resolution ion microprobe (SHRIMP II) at the Beijing SHRIMP Center, Institute of 3. Analytical techniques Geology, Chinese Academy of Geological Sciences. Zircon was extracted from a fresh rock sample following normal separation methods, and in- 3.1. Mineral compositions dividual zircon crystals were then handpicked and mounted, together with the TEM standard (417 Ma), onto double-sided adhesive tape Back-scattered electron (BSE) images and chemical compositions of and enclosed in epoxy resin disks. Cathodoluminescence (CL) images minerals were measured with a JEOL LIXA-8100 Electron Microprobe at of the zircon were made prior to U–Th–Pb analyses to reveal the internal the Guangzhou Institute of Geochemistry (GIG), Chinese Academy of textures and to select the sites for the SHRIMP analyses. During the F. Guo et al. / Lithos 224–225 (2015) 46–60 49

Table 1 A summary of petrologic and mineralogical characteristics of the Early Jurassic maﬁc intrusions in NE China.

Sample Locality Lithology Major minerals Accessories

06TM-1 43°02′37″ N Diorite Hb (60%) + Pl (35%) Mag, Ilm, Ap, Zr 129°40′24″E 06TM-2 43°02′31″ N Gabbro Ol (17%) + Opx (19%) + Cpx (10%) + Pl (50%) Hb, Mag, Ap, Zr 06TM-4129°40′28″E Diorite Hb (55%) + Pl (40%) Mag, Ilm, Ap, Zr 06TM-6 43°02′31″ N Norite Ol (10%) + Opx (30%) + Cpx (10%) + Pl (45%) Hb, Mag, Ap 06TM-7129°40′28″E Diorite Hb (45%) + Pl (50%) Mag, Ilm, Ap, Zr 06TM-8 Norite Ol (30%) + Opx (22%) + Cpx (13%) + Pl (30%) Hb, Mag, Ap 06TM-9 Diorite Hb (60%) + Pl (35%) Mag, Ilm, Ap, Zr 06TM-11 Gabbro Ol (5%) + Opx (15%) + Cpx (25%) + Pl (50%) Hb, Mag, Ap HTW1-3* Xincun village, Yichun Norite Ol (15%) + Opx (10%) + Cpx (15%) + Pl (55%) Hb, Mag, Ap, Zr HTW1-4* Xincun village, Yichun Norite Ol (16%) + Opx (12%) + Cpx (12%) + Pl (55%) Hb, Mag, Ap, Zr HYC10* Xinhuo village, Gabbro Cpx (25–30%) + Hb (20–25%) + Pl (50%) Mag, Ap, Zr Yichun HTW6* Shuguang village, Yichun Gabbro Ol (7–10%) + Cpx (35–40%) + Pl (45%) + Mag (8–10%) Hb, Ap, Zr HYL1* Liuzhonggou village, Mulan hornblendite Hb (~90%) + Mag (8–10%) + Pl (2–3%) Ti, Ap, Zr HWC1* Pingfang village, Wuchang Diorite Hb (20–25%) + Pl (65–70%) + Bi (6–8%) Qz, Mag, Ap, Zr

The modal mineral compositions of maﬁc rock samples with star mark (*) are estimated from Yu et al. (2012). Mineral abbreviations: Ol — olivine; Opx — orthopyroxene; Cpx — clinopyroxene; Hb — hornblende; Pl — plagioclase; Bi — biotite; Qz — quartz; Ap — apatite; Zr — zircon; Ilm — ilmenite; Mag — magnetite; Ti — titanite. analyses, the analytical spot size averaged ~30 μm and each spot was The Sr and Nd isotopic analyses were performed with a Finnigan rastered for three minutes to remove common Pb on the zircon surface. Neptune multi-collector ICP-MS at GIG-CAS using the analytical proce- More detailed analytical procedures were described by Compston et al. dures described by Li et al. (2006). Powders were dissolved in HF-

(1984) and Liu et al. (2003). The decay constants used were those rec- HClO4, and the dissolution was conducted at 200 °C for one week. Sr ommended by Steiger and Jager (1977), and the common Pb correction and Nd were separated using conventional ion exchange columns, and used the measured 204Pb value. The zircon U–Pb results for the gabbro the Nd fractions were further separated with di-(2-ethylhexyl) phos- sample 06TM-2 are listed in Table 2. phoric acid (HDEHP)-coated Kef columns. The measured 87Sr/86Sr and 143Nd/144Nd ratios were normalized to 86Sr/88Sr = 0.1194 and 146Nd/144Nd = 0.7219, respectively. The reported 87Sr/86Sr and 3.3. Whole-rock elemental and Sr–Nd–Hf isotope analyses 143Nd/144Nd ratios were adjusted to the NBS SRM 987 standard (87Sr/86Sr = 0.710247 ± 8, 2σ) and the JNdi-1 standard All samples for whole-rock major oxides, trace elements and Sr–Nd– (143Nd/144Nd = 0.512115 ± 4, 2σ), respectively. Ten analyses of the Hf isotope analyses were crushed, and fresh chips were picked out and rock standard BHVO-2 yielded 87Sr/86Sr = 0.703496 ± 6 (2σ,n=10) washed in 0.05 N HCl and puriﬁed water. The cleaned chips were then and ten analyses of rock standards BHVO-2 and JB-3 gave ground in an agate mortar to less than 200 mesh. Whole-rock major 143Nd/144Nd = 0.512965 ± 4 (2σ, n = 10) and 143Nd/144Nd = oxides were determined by X-ray ﬂuorescence (XRF) spectrometry on 0.513041 ± 5 (2σ, n = 10), respectively. The procedural blanks were fused glass pellets at GIG-CAS. The accuracy and precision of the XRF b200 pg for Sr and ~30 pg for Nd. analyses were estimated to be 2% for major oxides. For the Hf isotopic analysis, about 0.5 g of sample powders and 1.0 g

Trace elements were determined with an inductively coupled of Li2B4O7 were mixed homogeneously. The mixtures were fused in Pt plasma-mass spectrometry (ICP-MS) at the Institute of Geochemis- crucibles at 1250 °C for 15 min in a high-frequency furnace. Then, try, CAS. The powders (~50 mg) were digested in 1 ml of HF and about 400 mg of the resulting glassy samples were dissolved in 2 N

0.5 ml of HNO3 in Teﬂon cups sealed in screw-top stainless steel HCl. The Hf fraction was separated using a modiﬁed single-column sep- bombs at 190 °C for 12 h. The analytical precision is generally better aration procedure involving ion exchange with Ln-Spec resin. Detailed than 5% for elements with concentrations N200 ppm and 5–10% for analytical procedure was also reported in Li et al. (2006).TheHf those b200 ppm. A detailed description of the analytical procedure isotopes were determined using a Finnigan Neptune MC-ICP-MS at was reported in Qi et al. (2000). The analytical results of major ox- GIGCAS. The 176Hf/177Hf ratios were normalized to 179Hf/177Hf = ides and trace elements of the international standards can be found 0.7325. During the Hf isotope analyses, the reported176Hf/177Hf ratios in Appendix 2. were adjusted to the standard solution JMC-475 (0.282163 ± 7, 2σ).

Table 2 SHRIMP zircon U–Pb dating results of the Tumen maﬁc intrusive complex (06TM-2).

Spot 206Pb U ppm Th ppm 232Th/238U 207Pb/235U % err 206Pb/238U % err 206Pb/238U age (Ma) 1σ

1.1 0.24 595 593 1.03 0.21 2.6 0.0296 1.2 188.3 2.3 2.1 0.08 422 286 0.70 0.23 14.3 0.0294 1.7 186.6 3.1 3.1 0.15 1308 1075 0.85 0.21 2.8 0.0297 1.2 188.7 2.2 4.1 0.82 364 217 0.62 0.18 6.2 0.0279 1.4 177.4 2.4 5.1 −0.10 618 400 0.67 0.22 2.6 0.0295 1.2 187.1 2.3 6.1 0.19 672 373 0.57 0.21 5.7 0.0296 1.2 188.3 2.3 7.1 0.72 431 418 1.00 0.19 5.1 0.0277 1.4 176.1 2.4 8.1 0.61 1119 825 0.76 0.20 5.2 0.0304 1.2 192.9 2.2 9.1 0.71 455 545 1.24 0.22 4.0 0.0292 1.3 185.8 2.4 10.1 0.66 859 865 1.04 0.20 5.1 0.0297 1.2 188.4 2.3 11.1 1.08 486 350 0.74 0.20 6.7 0.0289 1.3 183.9 2.4 12.1 0.03 1275 851 0.69 0.21 2.0 0.0293 1.1 186.0 2.1 13.1 −0.24 742 616 0.86 0.22 3.2 0.0289 1.2 183.7 2.2 14.1 0.66 450 526 1.21 0.21 4.9 0.0298 1.5 189.0 2.8 15.1 0.42 501 617 1.27 0.20 4.3 0.0270 1.4 171.7 2.3 50 F. Guo et al. / Lithos 224–225 (2015) 46–60

Ten analyses of the USGS reference material BHVO-2 yielded 100 × Mg/(Mg + ΣFe) in atomic ratio) of the sample (06TM-8, 176Hf/177Hf = 0.283087 ± 10 (2σ, n = 10) and five analyses of JB-3 Mg# = 74), the olivine has a relatively lower Fo content. Equilibra- gave 176Hf/177Hf = 0.283222 ± 10 (2σ, n = 5). The procedural blanks tion estimates between the olivine and the bulk rock yield an Fe–Mg were ~50 pg for Hf. The whole-rock major and trace element composi- exchange coefficient of 0.34, which is approximately equal to the tions and Sr–Nd–Hf isotopic compositions of the Tumen mafic intrusive upper limit between olivine and basaltic liquid (Ford et al., 1983; complex are listed in Table 3.TheSr–Nd–Hf isotopic compositions, to- Longhi et al., 1978). A likely explanation for such a decoupling is at- gether with the major and trace element compositions of other contem- tributed to olivine accumulation in the crust-level magma chamber. poraneous mafic intrusions in NE China reported in Yu et al. (2012),are The orthopyroxene is classified as hypersthene with a formula of listed in Appendix 3. Wo1.6–3.6 En75.7–77.2 Fs21.6–22.9 (Table 3 and Fig. 3a). The clinopyroxene has a CaO content ranging from 23.4% to 24.5% and is identified as Ca-

4. Results rich diopside with a formula of Wo46.6–49.1En40.8–43.5 Fs8.5–10.8 (Appen- dix 1 and Fig. 3a). Such high-Ca clinopyroxene is similar to that 4.1. Mineral compositions of the olivine norite (06TM-8) primarily observed in water-saturated basaltic magma (Fig. 3b). The plagioclase has an anorthite (An) content between 87.4 and Theolivinecrystalshaveaforsterite(Fo)contentof76–78 95.3 and belongs to bytownite and anorthite (Appendix 1 and (Appendix 1). Compared with the whole-rock Mg# (Mg# = Fig. 3c).

Table 3 Major oxide (wt.%), trace element (ppm) compositions and Sr–Nd–Hf isotopic compositions of the Tumen maﬁc intrusive complex.

Sample 06TM-1 06TM-2 06TM-4 06TM-6 06TM-7 06TM-8 06TM-9 06TM-11

Lithology Diorite Gabbro Diorite Norite Diorite Norite Diorite Gabbro

SiO2 43.15 46.61 45.23 50.36 46.03 42.96 45.19 49.99

TiO2 1.83 0.47 1.7 0.57 1.71 0.3 1.05 1.43

Al2O3 18.72 17.16 19.43 10.69 19.52 9.87 19.16 17.85

Fe2O3 12.45 10.07 11.14 12.7 11.21 14.36 10.58 10.76 MnO 0.13 0.13 0.11 0.2 0.13 0.19 0.14 0.15 MgO 7.71 10.85 6.58 16.35 6.31 19.98 7.82 5.83 CaO 11.61 10.23 11.05 6.19 11.26 9.34 13.49 9.57

Na2O 2.22 1.47 2.65 1.16 2.65 0.85 1.67 3.39

K2O 0.63 0.44 0.65 0.62 0.59 0.15 0.27 0.35

P2O5 0.09 0.04 0.12 0.02 0.22 0.03 0.06 0.16 LOI 1.07 2.34 0.99 0.78 0.46 1.57 0.44 0.7 TOTAL 99.62 99.81 99.67 99.63 100.08 99.59 99.86 100.18 Mg# 55.3 68.3 54.2 72 53 73.6 59.7 52 Sc 40 18 33 20 32 26 35 24 V 415 103 350 109 331 99 440 279 Cr 32 397 13 492 4 720 95 64 Co 43 60 41 76 38 93 42 31 Ni 10 137 8 240 3 292 33 25 Rb 14.7 14.4 10.3 20.6 5.9 3.9 2.1 7.5 Sr 652 544 577 253 607 271 560 573 Y 19 7.2 20.8 14.6 27.8 5.2 10.5 14 Zr 42 27 46.1 23.5 62.3 16.1 31.1 46.2 Nb 2.7 1.1 2.8 1.7 3.9 0.5 1.6 3.9 Ba 165 76 149 131 156 35 58 191 La 5.1 3.81 6.62 6.53 10.27 1.92 4.89 9.7 Ce 15.18 8.84 18.27 14.77 28.63 4.83 12.86 20.77 Pr 2.61 1.3 2.98 2.02 4.62 0.76 1.88 2.84 Nd 14.63 6 15.77 8.99 23.8 3.82 9.22 12.93 Sm 4.3 1.47 4.51 2.41 6.37 1.1 2.4 3.09 Eu 1.66 0.6 1.73 0.88 2.1 0.48 0.97 1.29 Gd 4.94 1.53 5.02 2.7 6.91 1.23 2.55 3.17 Tb 0.82 0.26 0.81 0.49 1.13 0.21 0.42 0.51 Dy 4.81 1.51 4.72 3.08 6.77 1.28 2.53 3.11 Ho 0.93 0.29 0.95 0.67 1.32 0.25 0.51 0.62 Er 2.35 0.8 2.4 1.96 3.49 0.68 1.35 1.63 Tm 0.32 0.12 0.33 0.33 0.48 0.1 0.19 0.25 Yb 1.93 0.73 2.01 2.17 2.99 0.56 1.21 1.48 Lu 0.29 0.11 0.3 0.36 0.44 0.09 0.17 0.24 Hf 1.83 0.82 1.85 1.04 2.55 0.6 1.36 1.34 Ta 0.23 0.14 0.22 0.16 0.27 0.05 0.14 0.32 Th 0.36 0.8 0.52 1.9 0.69 0.42 0.17 0.62 U 0.15 0.22 0.22 0.57 0.25 0.12 0.05 0.23 87Rb/86Sr 0.0653 0.0766 0.0517 0.2357 0.0281 0.0417 0.0109 0.0379 87Sr/86Sr ± 2σ 0.704451 ± 7 0.704429 ± 6 0.704452 ± 7 0.704943 ± 6 0.704342 ± 6 0.704273 ± 7 0.704310 ± 7 0.704453 ± 8 87Sr/86Sr(i) 0.70428 0.70423 0.70431 0.70432 0.70427 0.70416 0.70428 0.70435 147Sm/144Nd 0.1777 0.1481 0.1729 0.1621 0.1618 0.1741 0.1574 0.1445 143Nd/144Nd ± 2σ 0.512779 ± 3 0.512737 ± 3 0.512745 ± 4 0.512756 ± 3 0.512779 ± 3 0.512742 ± 4 0.512757 ± 4 0.512726 ± 4

εNd(t) 3.13 3.04 2.59 3.06 3.52 2.5 3.2 2.9 176Lu/177Hf 0.0221 0.0226 0.0482 0.024 0.0209 0.0174 0.0249 176Hf/177Hf ± 2σ 0.282998 ± 3 0.282974 ± 2 0.283040 ± 3 0.282989 ± 3 0.282975 ± 2 0.282960 ± 2 0.283010 ± 3

εHf(t) 9.3 8.4 7.6 10.4 8.8 10.55 9.5 Note: The initial Sr–Nd–Hf isotopic ratios are recalculated at 187 Ma. The present chondrite has: 147Sm/144Nd = 0.1967, 143Nd/144Nd = 0.512638, 176Lu/177Hf = 0.0332 and 176Hf/177Hf = 0.282772 (Blichert-Toft and Albarède, 1997). The decay constants are: λ =1.42×10−10/y for 87Rb, 6.54 × 10−12/y for 147Sm and 1.867 × 10−12/y for 176Lu. F. Guo et al. / Lithos 224–225 (2015) 46–60 51

Fig. 3. Classification of pyroxene (a), a Wo80-En + Fs-CaTs + CrTs80 triangle plot of clinopyroxene (b, after Gaetani et al., 1993), classification of plagioclase (c) and hornblende (d) of the Tumen mafic intrusive complex.

The hornblende belongs to pargasite (Appendix 1 and Fig. 3d). the hornblende crystallized in a crust-level magma chamber The hornblende xenocrysts are overgrown around the early crystal- (e.g., Chiaradia et al., 2011; Claeson and Meurer, 2004; Davidson lized olivine, which has a typical poikilitic texture, indicating that et al., 2007).

Fig. 4. CL images (a) and U–Pb concordia age of zircon (b) in 06TM-2. 52 F. Guo et al. / Lithos 224–225 (2015) 46–60

4.2. Zircon U–Pb age plot above the terrestrial array defined by Vervoort et al. (1999) in an Nd–Hf isotopic space (Fig. 6b), and have Sr–Nd–Hf isotopic variation The zircon grains separated from the gabbro sample 06TM-2 are ranges similar to those basalts from the Lesser Antilles and Sunda mostly clear, subhedral to euhedral, stubby prisms, with a length/ Arcs, where the mantle wedges were intensively metasomatized by width ratio of 1.2–3.0. Some grains show core-rim overgrowth and subducted sediments (e.g., Bouvier et al., 2010; Labanieh et al., 2010; oscillatory zoning textures in the CL images (Fig. 4a), indicating a Macdonald et al., 2000; Woodhead et al., 2001). magmatic origin. The zircon grains have a Th/U ratio of 0.57–1.27 (Table 2). Fifteen analyses yielded an apparent 206Pb/238U age ranging from 172 to 193 Ma, with a weighted mean age of 187 ± 1 Ma 5. Discussion (MSWD = 1.16, n = 12; Fig. 4b), which is interpreted to be the fi emplacement age of the Tumen mafic intrusion. The rocks of the Tumen ma c intrusive complex have mineral assemblages and geochemical compositions similar to arc gabbros or mafic accumulates (e.g., Claeson and Meurer, 2004; Grove et al., 4.3. Whole-rock geochemistry 2002). In the following section, we discuss the parental magma, fractionation processes, and source characteristics of the complex and pro- – – The major, trace elemental and Sr Nd Hf isotopic compositions pose a possible petrogenetic link to the subduction of the Paleo-Pacific fi of eight samples from the Tumen ma c intrusive complex are listed Ocean. in Table 3. These rocks have wide compositional ranges (e.g., SiO2 = 43.2–50.4% and MgO = 5.8–20.0%; Table 3). They also show subparallel chondrite-normalized REE patterns with variable light REE 5.1. Estimation of the parental magma enrichment and heavy REE depletion and variable Eu anomalies (Eu/Eu* = 0.97–1.26; Fig. 5a).Somesamples(e.g.,06TM-1,4,7 The Early Jurassic Tumen and other mafic intrusions in NE China and 8) have La/NdCN (CN denotes chondrite normalization) values show variable effects of crystal accumulation, so the whole-rock geo- of less than 1.0, which is a feature of hornblende or clinopyroxene chemistry cannot represent the parental magma composition; instead accumulation (Guo et al., 2007, 2009b).ThemiddleREEofthesam- it reflects the sum of composition of the accumulative crystals and ple 06TM-6 has a convex shape, indicating the role of hornblende trapped melts (Bédard, 1994). In the following paragraphs, we use the fractionation. The primitive mantle (PM)-normalized trace element method proposed by Bédard (1994, 2001) and Bédard et al. (2009) to spidergrams are characterized by LILE and LREE enrichments and estimate the parental magma compositions of the Early Jurassic mafic variable Nb–Ta–Zr–Hf depletions relative to neighboring REEs intrusions in NE China. (Fig. 5bandTable 3). The positive Sr and Eu anomalies are consistent The mass-balance equation (Eq. (1)) expresses the concentration of with the accumulation of plagioclase (Fig. 5b). an element ‘i’ in the cumulate rock as the sum of the products of the Compared with their large compositional variations, the samples of modal proportions φ (weight equivalents as fractions of 1) and the con- the Tumen mafic intrusive complex have narrow ranges of Sr–Nd–Hf centrations (C) of element ‘i’ in the constituent minerals and trapped 87 86 fi isotopic compositions: Sr/ Sr(i) = 0.7042 to 0.7044, εNd(t) = +2.5 melt (TM). In the Tumen ma c intrusive complex, the olivine gabbro to +3.5, and εHf(t) = +8.4 to +10.6 (Table 3). In contrast, rocks samples have the lowest REE concentrations and positive Sr and Eu from other contemporaneous mafic intrusions in NE China show anomalies, which indicate the accumulation of plagioclase and ferro- much larger Sr–Nd–Hf isotopic variations: 87Sr/86Sr(i) = 0.7039 to magnesian minerals. The modal mineral proportions are listed in

0.7073, εNd(t) = −1.9 to +2.9, and εHf(t) = +2.0 to +11.7. These Table 1. Therefore, the trace element concentration of the olivine- mafic rocks have two isotopic groups: one group (HWC1-1 and bearing gabbro-norite can be expressed as: 2) with Sr–Nd–Hf isotopic compositions similar to those of the Tumen mafic rocks; and the other (HTW1-1–4, HTW6-1–4 and HYC10-1–2) with higher 87Sr/86Sr(i) (0.7065–0.7073) and lower ε (t) (−1.9 to Nd rock ¼ φCpx Cpx þ φOpx Opx þ φOl Ol þ φPl Pl þ φMt Mt C i C i C i C i C i C i −0.4) and εHf(t) (+2.0–+5.2) values (Table 3 and Fig. 6). Similar to þ φAp Ap þ φTM TM ð Þ most arc magmas, the Early Jurassic mafic rocks in NE China generally C i C i 1

Fig. 5. Chondrite-normalized REE patterns and primitive mantle-normalized trace element spidergrams of the Tumen maﬁc intrusive complex. Normalization values for REE are from Taylor and McLennan (1985) and for trace elements are from Sun and McDonough (1989). F. Guo et al. / Lithos 224–225 (2015) 46–60 53

trapped melt to 1.0, the equation for the clinopyroxene solution has the following form, in which φ™ is the only unknown:

n Cpx ¼ rock = φCpx þðφOpx Opx=melt =Cpx=melt Þþ φOl Ol=melt =Cpx=melt C i C i Di Di Di Di

þðφPl Pl=melt =Cpx=melt ÞþðφMt Mt=melt =Cpx=melt Þ Di Di Di Di o þðφAp Ap=melt =Cpx=melt ÞþððφTM Cpx=melt Þ Di Di Di ð3Þ

Eq. (3) yields the clinopyroxene trace element concentration, and division by the partition coefficient (Cpx/meltD) gives the concen- melt tration of the element in the coexisting equilibrium liquid (C i = Cpx Cpx/melt C i/ Di). To simplify the calculation, the Tumen mafic intrusive complex is re- duced to two- or three-phase assemblages with a trapped melt fraction (TMF) of 10–15% using a non-modal melting backstripping procedure. The detailed calculation results with different TMFs for each sample are given in Appendix 4. The modal mineral compositions of the samples and partition coefficients used in calculation are listed in Table 1 and Appendix 5, respectively. Here, we select the two olivine gabbro samples (06TM-2 and 8) with the lowest REE concentrations (Fig. 5a) to estimate the composition of the parental magmas in equilibrium with the maficcumulates.As shown in Fig. 7a, the calculated parental magmas for both samples are enriched in LILE (Rb, Ba), Th–U, and LREE, but depleted in Nb–Ta and

P. The K2O concentration varies between 0.9% and 1.8%, showing calc- alkaline afﬁnities. Except for higher Zr and Hf, and lower Ba, K, Nb, and P concentrations (Fig. 7a), the calculated parental magmas for sample 06TM-2 and 8 have PM-normalized incompatible element patterns

similar to that of HCW1-1, which has the highest SiO2 concentration (55.7%) and Sr–Nd–Hf isotopic compositions similar to those of the Tumen maﬁc intrusion (Appendix 5 and Fig. 6). Similarly, we use the same method to calculate the composition of parental magmas in equilibrium with the Xincun olivine gabbros (HTW1–3 and 4), which have higher 87Sr/86Sr(i) ratios and lower

εNd(t) and εHf(t) values than those of the Tumen maﬁc intrusion. The results indicate that the parental magmas for sample HTW1–3and4are

also calc-alkaline, with a K2O concentration ranging from 0.7% to 1.1%, and show arc-type trace element features (i.e., Nb–Ta depletion and LILE and LREE enrichment without positive Sr and Eu anomalies; Fig. 7b) (e.g., Hawkesworth et al., 1993; Perfit et al., 1980). Once the effect of crystal accumulation was effectively removed, the estimation Fig. 6. Sr–Nd–Hf isotope diagrams for the Tumen mafic intrusive complex and other related mafic rocks in NE China. Data sources: Late Paleozoic maficrocks(Li, 2012); the Early results indicate that the parental magmas, which crystallized the Early Jurassic mafic rocks (this study); the range of Dashizhai basalts (Guo et al., 2009a); the Jurassic mafic intrusions from Tumen and other localities in NE China, variation field of I-type granitoids (Li et al., 2014); Marine sediments (Vervoort et al., were calc-alkaline series with arc-type trace element features. 2011); Mashan Group (Wu et al., 2000); Terrestrial array (Vervoort et al., 1999); Sunda arc (Woodhead et al., 2001); Lesser Antilles (Labanieh et al., 2010); the range of MORB 5.2. A water-saturated parental magma (Pearce et al., 1999).

The An content of plagioclase depends on the contents of CaO, Al2O3, H2O, CaO/Na2O, and Al2O3/SiO2, and the pressure of primitive magma (e.g., Panjasawatwong et al., 1995; Sisson and Layne, 1993). High-An plagioclase has been widely observed in hydrous arc high-Al basalts ﬁ By de nition: and gabbroic nodules, forearc boninites, and gabbros (e.g., Crawford et al., 1987; Falloon and Crawford, 1991; Stolz et al., 1988; Thy, 1987). Previous experimental studies have shown that both the An compo- Cpx Pl Cpx=melt Pl=melt Opx Cpx=melt Opx=melt C ¼ C D = D ¼ C D = D Ca–Na i i i i ii i nent of plagioclase and the K D (Ca relative to the Na partition coef- ¼ Ol Cpx=melt =Ol=melt ¼ Mt Cpx=melt =Mt=melt ﬁcient between plagioclase and melt) between plagioclase and basalt C i Di Di C i Di Di increase with an increase in the H O content of melt and decrease ¼ Ap Cpx=melt =Ap=melt ð Þ 2 C i Di Di 2 with crystallization pressure (e.g., Panjasawatwong et al., 1995; Sisson

and Layne, 1993; Takagi et al., 2005). This suggests that high H2Ocon- tent of the melt and low pressure are preferable for the crystallization The equilibrium distribution of trace elements among the con- of An-rich plagioclase. Sisson and Grove (1993) proposed that calcic stituent minerals can be calculated by substituting Eq. (2) into (An N 90) plagioclase crystallized from high-Al basalts with high H2O Eq. (1). By setting the crystal/liquid partition coefﬁcient (D) for the content but could not be formed from dry melt with normal Na2Oand 54 F. Guo et al. / Lithos 224–225 (2015) 46–60

Fig. 7. Calculated parental magma compositions of the Tumen and Xincun maﬁc intrusions (Yu et al., 2012). The description of the calculation method and partition coefﬁcients between minerals and basalt are respectively attached in Appendices 4 and 5. Normalization values of PM are from Sun and McDonough (1989).

CaO concentrations of arc magmas. At moderate to high pressures comparable with those of the clinopyroxene in hydrous arc basalts (5–10 kbar, corresponding to middle-lower crustal levels), highly calcic (Gaetani et al., 1993, Fig. 3b), indicating a water-saturated magma con- plagioclase (An = 90–100) could crystallize from either high-Al arc ducive to clinopyroxene crystallization. basalts with very high H2O contents (~6%) or melts with abnormally The mineral assemblage of calcic plagioclase, Ca-rich diopside, and high CaO/Na2O(N8) ratios (Panjasawatwong et al., 1995). In accordance hornblende in the Tumen mafic intrusion crystallized in water- with experimental studies on tholeiitic basalts, Takagi et al. (2005) saturated magma. Considering their arc-type trace element and Sr– proposed that the optimal condition for the crystallization of most An- Nd–Hf isotopic features, the Tumen mafic intrusive complex was prob- rich plagioclases was near H2O-saturation at 2–3 kbar. Although some ably formed through hydrous melting of a mantle wedge in response to calcic plagioclase crystals have been reported in melt inclusions of the subduction of the Paleo-Pacific Ocean.

H2O-undersaturated magmas (e.g., MORBs), plagioclase with An N 94 has been rarely found in these rocks (Panjasawatwong et al., 1995), except in cases where the parental magma has very high CaO/Na2O 5.3. Fractionation and accumulation of olivine, pyroxene and plagioclase (12–15) and Al2O3 contents (N18%). However, both the CaO/Na2O (11) ratio and Al2O3 content (~9.9%) of 06TM-8 are much lower than The narrow Sr–Nd–Hf isotopic variations over the wide ranges of the values at which the crystallization of high-An plagioclase is expected MgO and SiO2 concentrations in the Tumen mafic intrusion implies in a water-deficient magmatic system. Besides the existence of calcic that the effects of open-system magmatic processes, such as crustal plagioclase, the occurrence of hornblende in all rock types of the contamination and/or assimilation-fractional crystallization, are insig- Tumen mafic intrusive complex indicates that they crystallized in nificant. In contrast, the subparallel REE patterns, variable Eu and Sr water-saturated parental melt. Finally, the high-Ca diopside coexisting anomalies, and large Cr and Ni variations in these rocks indicate a with the plagioclase in 06TM-8 also contains Ti and Al contents predominant role of fractional crystallization and accumulation. F. Guo et al. / Lithos 224–225 (2015) 46–60 55

The positive correlation between Cr and Ni favors fractionation of the higher Th/Sm, Th/Ce, and Th/Yb ratios and less radiogenic Nd and clinopyroxene and olivine (Fig. 8a), whereas the positive correla- Hf isotopic compositions than MORBs or Early Paleozoic Dashizhai tions of Eu/Eu* with Sr/Y and Sr/Sr* and a negative correlation basalts (Guo et al., 2009a; Pearce et al., 1999) suggest that the mantle between Eu/Eu* and total REE indicate the role of plagioclase frac- source of the Tumen mafic intrusion experienced enrichment (fluid- tionation (Fig. 8b–d). As most of the samples have Eu/Eu* higher mediated, melt-dominated, or both) of subducted sediments. As illus- than 1.0, plagioclase accumulation was responsible for the positive trated in the plots of Th/Sm versus Th/Ce and Ba/La versus Th/Yb Sr and Eu anomalies. Accordingly, fractionation and/or accumulation (Fig. 9a and b), the rocks in the Tumen mafic intrusive complex show of olivine, pyroxene, and plagioclase in a crustal magma chamber varying trends of sediment input and a predominant sediment-melt were responsible for the formation of the wide spectrum of rock contribution to the mantle source. To further estimate the proportion types in the Tumen mafic intrusive complex. of the subducted sediment component in the mantle wedge, we assume

that the depleted mantle has: Nd = 1.2 ppm, Hf = 0.3 ppm, εNd(t) = 5.4. Nature of the mantle source +12, and εHf(t) = +18; in addition, we assume that the subducted sediment component is characterized by: Nd = 27 ppm, Hf = 4 ppm,

It is widely suggested that Th, Ba, and REEs exhibit different geo- εNd(t) = −8, and εHf(t) = −10. Results from Nd–Hf isotopic modeling chemical behaviors in slab-derived fluids and melts (e.g., Ayers, 1998; indicate that an addition of 3–4% subducted sediment into a depleted Kogiso et al., 1997; Spandler and Pirard, 2013; Woodhead et al., 2001), mantle could produce the observed Nd–Hf isotopic variations in these which can be effectively used to distinguish the possible role of fluid rocks (Fig. 9c). We also employ the variation trends of Ba/Nb and Ba/ and melt derived from subducted oceanic crust and its overlying La with εNd(t) and εHf(t) to determine the possible roles of the two sediments. In subduction zones, slab-derived fluids and melts are types of mantle metasomatism. As shown in Fig. 10, both sediment regarded as the most important agents for interpreting the LILE and melt and fluid might contribute to the melting source associated with LREE enrichment and HFSE depletion in arc magmas (e.g., Ayers, the Tumen mafic intrusive complex, and the contribution of sediment 1998; Grove et al., 2002; Handley et al., 2007; Jicha et al., 2004). Because melt is dominated. Ba, Th, La, Sm, and Th are incompatible in accumulative crystals With the aim of further estimating the respective contributions of (e.g., olivine, plagioclase and pyroxene with partition coefficients sediment melt and fluid to the melting mantle source, we conduct much less than 1.0; Rollinson, 1993), some elemental ratios (e.g., Ba/ trace element modeling in accordance with the source component pro- La, Ba/Nb, Th/Sm, and Th/Yb) in the studied mafic rocks only slightly portions inferred from the Hf–Nd isotopic modeling results (Fig. 9c). Be- vary from the calculated parental magmas (Fig. 7). As a consequence, cause the Tumen mafic rocks are mafic cumulates and their whole-rock

Fig. 8. Cr vs. Ni (a), Eu/Eu* vs. Total REE (b), Sr/Y (c) and Sr/Sr* (d) diagrams of the Early Jurassic maﬁc rocks in NE China, showing the roles of fractionation and accumulation of olivine, clinopyroxene and plagioclase in the magmatic evolution of the Tumen and other contemporaneous maﬁc intrusions in NE China (Yu et al., 2012). 56 F. Guo et al. / Lithos 224–225 (2015) 46–60

Fig. 9. Th/Sm vs. Th/Ce (a) and Th/Yb vs. Ba/La (b) plots and Nd–Hf isotopic modeling results (c) for the Tumen mafic intrusive complex and related mafic rocks. Data source: Late Paleozoic maficrocks(Li, 2012); GLOSS (Plank and Langmuir, 1998); Cenozoic adakitic andesites (Guo et al., 2007, 2009b), MORB (Sun and McDonough, 1989). The Hf–Nd isotopic compositions of depleted mantle (DM) in NE China are estimated from the mafic components of the Hegenshan ophiolite suite and the Dashizhai arc basalts in Inner Mongolia (Guo et al., 2009a; Miao et al., 2008). Hf and Nd concentrations of DM are assumed to be somewhat lower than primitive mantle (Sun and McDonough, 1989). The Nd–Hf isotopic compositions of the subducted sediment component are assumed from the global marine sediments (Vervoort et al., 2011), while the Nd and Hf concentrations are estimated from the average global subducted sediment (GLOSS, Plank and Langmuir, 1998). On the mixing curve between the sediment fluid and DM, the mobility of Nd and Hf in aqueous fluid is assumed to be 0.03 and 0.005 and the dehydration volume of sediment fluid is assumed to be 0.02 (Aizawa et al., 1999; Hanyu et al., 2006). See discussion in the text. The symbols are shown in the figure. trace element concentrations cannot represent the ‘true’ geochemical patterns (Yu et al., 2012). As discussed earlier, these mafic intrusions compositions of the mantle-derived magmas, we use the calculated have two isotopic groups. The Nd-Hf isotopic modeling results suggest parental magma compositions of 06TM-2 and 8 to model the melting that the melting mantle source contained a 3–4% sediment component degrees. The calculation parameters are shown in Appendix 6. The in the first group (e.g., HWC1-1 and 2), which has Sr–Nd–Hf isotopic results indicate that 5–20% melting of a metasomatized mantle through compositions similar to those of the Tumen mafic intrusion. The lower

3% and 1% additions of sediment melt and sediment fluid, respectively, MgO and higher SiO2 contents and the correspondingly higher concen- could result in the calculated parental magma compositions (Fig. 11a). trations of incompatible elements in these rocks could have been formed through fractionation of ferromagnesian minerals (Fig. 8). 87 86 5.5. Comparison with the other Early Jurassic maficintrusionsinNEChina The other group has higher Sr/ Sr(i) ratios and lower εNd(t) and εHf(t) values than those of the Tumen mafic intrusion. The Nd–Hf isoto- In NE China, five contemporaneous (182–186 Ma) mafic intrusions, pic modeling results suggest that the melting mantle source contained a besides the Tumen mafic intrusive complex, occur in the Lesser Hinggan 7–10% sediment component (Fig. 9c). Elemental and isotopic modeling Mountains and Zhangguangcailing Range (Yu et al., 2012, Fig. 1b). These results suggest the contributions of both sediment melts and fluids to mafic intrusions also contain hydrous mineral assemblages and show the melting sources (Fig. 11). Similarly, we use the calculated parental ferromagnesian mineral and plagioclase accumulation, as indicated magma compositions of samples HTW1–3 and 4 to model the melting from the positive Eu and Sr anomalies in the REE and trace element degrees. The results show that 5–20% melting of a metasomatized F. Guo et al. / Lithos 224–225 (2015) 46–60 57

Fig. 10. Ba/Nb and Ba/La vs. εHf(t) and εNd(t) diagrams showing the respective roles of sediment melt and fluid in modifying the mantle source of the Early Jurassic Tumen and other mafic intrusions in NE China. Nd and Hf concentrations and Nd–Hf isotopic compositions of the end-member components are the same as in Fig. 9c. During sediment melting, the bulk partition coefficients for Nd and Hf are 0.05 and 0.2, respectively. The other calculation parameters are listed in the figure. The bulk partition coefficients (KD) during sediment melting are estimated with a residual mineral assemblage of garnet + pyroxene + K-rich mineral (e.g., phengite) + Al-rich mineral (e.g., zoisite and allanite) with no or little plagioclase (Guo et al., 2014; Hermann and Spandler, 2008). Data of Late Paleozoic mafic rocks are from Li (2012). mantle wedge through 3% and 5% additions of fluids and melts, respec- accretionary complexes, magmatic activities, and metamorphic assem- tively, derived from the subducted sediments, can explain most of the blages have also been observed in the Sikhote-Alin area, Far East of trace elemental compositions of the calculated parental magmas of Russia (Ishiwatari and Tsujimori, 2003; Kojima, 1989; Safonova and HTW1–3 and 4, with the exception of Th (Fig. 11b). The low Th concen- Santosh, 2014). trations in these two samples are possibly attributed to zircon and Recently, Yu et al. (2012) reported an N–S-trending Early Jurassic apatite fractionation during magmatic evolution. The mafic intrusions (182–186 Ma) mafic intrusion belt from the Lesser Hinggan Mountains to the north of the Tumen area could also have been formed by the and Zhangguangcailing Range. They considered that these mafic rocks, melting of a mantle wedge through the addition of sediment melt and together with I-type granitoids, resulting from a bimodal magmatic fluid derived from the subducting Paleo-Pacific Oceanic slab. The wide event in these regions, were formed under a back-arc extension related compositional ranges and positive Eu and Sr anomalies in these mafic to the subduction of the Izanagi Plate (Yu et al., 2012). Most Jurassic I- intrusions were probably due to fractionation and/or accumulation of type granitoids have Sr–Nd isotopic compositions similar to those of olivine, pyroxene, and plagioclase (Fig. 8), similar to those observed in coeval maficintrusions(Fig. 7a, Li et al., 2014), and are probably either the Tumen mafic intrusion. differentiates of mantle-derived magmas or crustal remelting products with a predominant contribution from contemporaneous underplated 5.6. Geological responses to the subduction of the Paleo-PacificOcean arc basalts. Xu et al. (2013) and Wang et al. (2015) proposed a petrogenetic link between the volcanic eruption of the Early Jurassic In NE Asia, the first reported geological record of Paleo-PacificOcean intermediate-felsic rocks in the Jiamusi Massif and the subduction of subduction was the Early to Late Jurassic accretionary complex along the Paleo-Pacific Oceanic Plate. All these facts indicate subduction of the Mino–Tanba–Chichibu Belt in southwest Japan, where high-P/T the Paleo-Pacific Oceanic Plate that resulted in a collision between the metamorphic rocks and granite batholiths outcropped beneath the Xing'an Block (or Songnen block) and the Jiamusi Massif (Zhou and coeval volcanic arc (e.g., Maruyama, 1997). In NE China, thermal events Wilde, 2013 and references therein). The Mesozoic tectonic evolution recorded in Jurassic I-type granitoids, metamorphic complexes (e.g., the of NE China was mainly controlled by the subduction and rollback of blueschist-facies metamorphism in the Heilongjiang Group), and sub- the Paleo-Pacificslab(Wu et al., 2011). duction accretionary complexes also suggest a westward subduction The Early Jurassic mafic intrusions constitute an N–S-trending mafic of the Paleo-Pacific Oceanic slab during 210–155 Ma (Wu et al., 2007, magmatic belt, nearly parallel to the distribution of the Heilongjiang 2011; Zhou and Wilde, 2013; Zhou et al., 2014). Similar Jurassic metamorphic complexes (Fig. 1b). The coeval intermediate-felsic 58 F. Guo et al. / Lithos 224–225 (2015) 46–60

Fig. 11. PM-normalized spidergrams of incompatible trace element modeling results of the Early Jurassic mafic intrusions in NE China. (a) shows the results of the Tumen maficintrusive complex and (b) shows the results of Xincun mafic intrusion reported in Yu et al. (2012). The end-member components of DM and subducted sediment are the same as in Fig. 9c. The modeling parameters and captions are listed in Appendix 6. See discussion in the text. igneous rocks (mostly I-type granitoids and their eruptive counterparts, granitoids and felsic lavas represent the magmatic records of the Early Wu et al., 2011; Xu et al., 2013; Wang et al., 2015) in the area also dis- Jurassic subduction of the Paleo-Pacific Oceanic Plate. tribute around the metamorphic complexes in a north–south direction (Fig. 1b).BothofthemconstituteanN–S-trending magmatic belt over 6. Concluding remarks 200 km wide. To summarize the above discussion, we develop a tectonic model invoking the subduction of the Paleo-Pacific Ocean beneath the New petrologic and geochemical evidence from the Early Jurassic NE China continental margin (Fig. 12). In the course of the westward Tumen mafic intrusive complex indicated its crystallization in a water- subduction of the Paleo-PacificOcean,thefluids and melts dehydrated saturated parental magma that was comparable to that of the arc from the subducted sediments (possibly also from the altered oceanic mafic cumulates. The parental magmas of the Tumen and other contem- crust) modified the overlying mantle wedge. Hydrous melting of the poraneous mafic intrusions in NE China were calc-alkaline and showed metasomatized mantle wedge produced mafic magmas along the sub- similar trace elemental and Sr–Nd–Hf isotopic compositions to those of duction zone. Differentiation of the mafic magmas and remelting of modern arc basalts. These mafic magmas were derived from partial the preexisting juvenile crust (probably also the underplating arc ba- melting of the mantle wedge metasomatized by melts and fluids salts) in NE China formed the widespread I-type granitoids and felsic dehydrated or released from the subducted sediments of the Paleo- lavas (Li et al., 2014; Wang et al., 2015; Wu et al., 2011; Xu et al., Pacific Ocean. In NE China, these Early Jurassic mafic intrusions and 2013). Therefore, the N–S-trending mafic intrusions and coeval I-type contemporaneous I-type granitoids and felsic lavas constituted an N– F. Guo et al. / Lithos 224–225 (2015) 46–60 59

Fig. 12. A tectonic model showing a petrogenetic link between the Early Jurassic maﬁc intrusions and the subduction of the Paleo-Paciﬁc Ocean in NE China. See discussion in the text.

S-trending arc magmatic belt, which was closely related to the west- Bureau of Geology and Mineral Resources of Jilin Province (BGMRJL), 1989. Regional – fi Geology Of Jilin Province. Geological Press, Beijing, pp. 1 698 (in Chinese with ward subduction of the Paleo-Paci c Oceanic Plate. English summary). Supplementary data to this article can be found online at http://dx. Chiaradia, M., Muntener, O., Beate, B., 2011. Enriched basaltic andesites from mid-crustal doi.org/10.1016/j.lithos.2015.02.014. fractional crystallization, recharge, and assimilation (Pilavo volcano, Western Cordil- lera of Ecuador). Journal of Petrology 52, 1107–1141. Claeson, D.T., Meurer, W.P., 2004. Fractional crystallization of hydrous basaltic “arc-type” magmas and the formation of amphibole-bearing gabbroic cumulates. Contributions Acknowledgments to Mineralogy and Petrology 147, 288–304. Compston, W., Williams, I., Meyer, C., 1984. U–Pb geochronology of zircons form lunar We thank Y. Liu, L. Qi and Z.R. Ren for help with whole-rock geo- Breccia 73217 using a sensitive high mass-resolution ion microprobe. Journal of Geophysical Research 89, B525–B534. chemical analyses, and LL Chen for microprobe analyses, and DY Liu Crawford, A.J., Falloon, T.J., Eggins, S., 1987. The origin of island arc high-alumina basalts. for SHRIMP zircon U–Pb dating. Prof. B Xu and HF Zhang and YC Wang Contributions to Mineralogy and Petrology 97, 417–430. are thanked for helpful discussion. Dr. JH Bèdard is thanked for his gen- Davidson, J., Turner, S., Handley, H., Macpherson, C., Dosseto, A., 2007. Amphibole “sponge” in arc crust? Geology 35, 787–790. erous help in estimation of parental magma composition. Constructive Davies, J.H., Stevenson, D.J., 1992. Physical model of source region of subduction zone reviews and comments from two anonymous referees and the Editor volcanics. Journal of Geophysical Research 97, 2037–2070. Engebretson, D.C., Cox, A., Gordon, R.G., 1985. Relative motions between oceanic and Andrew Kerr are appreciated, which help greatly improve the manu- fi fi continental plates in the Paci c basins. Geological Society of America, Special Paper script. This study was nancially supported by the Ministry of Science 206, 1–59. and Technology of China (2013CB429804) and the Natural Science Falloon, T.J., Crawford, A.J., 1991. The petrogenesis of high-calcium boninite lavas dredged Foundation of China (41121002, 41330206 and 40673035). from the north Tonga ridge. Earth and Planetary Science Letters 102, 375–394. Fan, W.M., Guo, F., Wang, Y.J., Lin, G., 2003. Late Mesozoic calc-alkaline volcanism of post- orogenic extension in the northern Da Hinggan Mountains, northeastern China. Journal of Volcanology and Geothermal Research 121, 115–135. References Faure, M., Natal'in, B., 1992. The geodynamic evolution of the eastern Eurasian margin in Mesozoic times. Tectonophysics 208, 397–411. Aizawa, Y.Y., Tatsumi, Y., Yamada, H., 1999. Element transport by dehydration of Ford, C.E., Russell, D.G., Craven, J.A., Fisk, M.R., 1983. Olivine–liquid equilibria: tempera- subducted sediments: implication for arc and ocean island magmatism. Island Arc ture, pressure and composition dependence of the crystal/liquid cation partition 8, 38–46. coefficients for Mg, Fe2+, Ca and Mn. Journal of Petrology 24, 256–266. Arculus, R.J., 1994. Aspects of magma genesis in arcs. Lithos 33, 189–208. Gaetani, G.A., Grove, T.L., Bryan, W.B., 1993. The influence of water on the petrogenesis of Ayers, J., 1998. Trace element modeling of aqueous fluid–peridotite interaction in the subduction related igneous rocks. Nature 365, 332–334. mantle wedge of subduction zones. Contributions to Mineralogy and Petrology 132, Ge, W.C., Wu, F.Y., Zhou, C.Y., Zhang, J.H., 2007. Porphyry Cu–Mo deposits in the eastern 390–404. Xing'an–Mongolian Orogenic Belt: mineralization ages and their geodynamic impli- Bédard, J.H., 1994. A procedure for calculating the equilibrium distribution of trace cations. Chinese Science Bulletin 52, 3416–3427. elements among the minerals of cumulate rocks, and the concentration of trace Grove, T.L., Parman, S.W., Bowring, S.A., Price, R.C., Baker, M.B., 2002. The role of an H2O- elements in the coexisting liquids. Chemical Geology 118, 143–153. rich fluid component in the generation of primitive basaltic andesites and andesites Bédard, J.H., 2001. Parental magmas of Nain Plutonic Suite anorthosites and maficcumu- from the Mt. Shasta region, N California. Contributions to Mineralogy and Petrology lates: a trace element modeling approach. Contributions to Mineralogy and Petrology 142, 375–396. 141, 747–771. Guo, F., Nakamura, E., Fan, W.M., Kobayoshi, K., Li, C.W., 2007. Generation of Palaeocene Bédard, J.H., Leclerc, F., Harris, L.B., Goulet, N., 2009. Intra-sill magmatic evolution in the adakitic andesites by magma mixing; Yanji Area, NE China. Journal of Petrology 48, Cummings Complex, Abitibi greenstone belt: tholeiitic to calc-alkaline magmatism 661–692. recorded in an Archaean subvolcanic conduit system. Lithos 111, 47–71. Guo, F., Fan, W.M., Li, C.W., Miao, L.C., Zhao, L., 2009a. Early Paleozoic subduction of the Blichert-Toft, J., Albarède, F., 1997. The Lu–Hf isotope geochemistry of chondrites and the Paleo-Asian Ocean: evidence from the geochronology and geochemistry of Dashizhai evolution of the mantle crust system. Earth and Planetary Science Letters 148, basalts from the Nei Mongolia region, NE China. Science in China Series D-Earth 243–258. Sciences 52, 940–951. Bouvier, A.S., Deloule, E., Métrich, N., 2010. Fluid input to magma sources of St. Vincent Guo, F., Nakamura, E., Fan, W.M., Kobayoshi, K., Li, C.W., Gao, X.F., 2009b. Mineralogical and Grenada Lesser Antilles: new insights from trace element in olivine-hosted and geochemical constraints on magmatic evolution of Paleocene adakitic andesites melt inclusions. Journal of Petrology 51, 1597–1615. from the Yanji area, NE China. Lithos 112, 321–341. 60 F. Guo et al. / Lithos 224–225 (2015) 46–60

Guo, F., Fan, W.M., Gao, X.F., Li, C.W., Miao, L.C., Zhao, L., Li, H.X., 2010. Sr–Nd–Pb isotope Safonova, I.Y., Santosh, M., 2014. Accretionary complexes in the Asia-Paciﬁcregion: mapping of Mesozoic igneous rocks in NE China: constraints on tectonic framework tracing archives of ocean plate stratigraphy and tracking mantle plumes. Gondwana and Phanerozoic crustal growth. Lithos 120, 563–578. Research 25, 126–158. Guo, F., Fan, W.M., Wang, C.Y., Li, C.W., Li, H.X., Zhao, L., Li, J.Y., 2014. Hf–Nd–Oisotopic Sengör, A.M.C., Natal'in, B.A., Burtman, V.S., 1993. Evolution of the Altaid tectonic collage evidence for melting of recycled sediments beneath the Sulu Orogen, North China. and Paleozoic crustal growth in Eurasia. Nature 364, 299–307.

Chemical Geology 381, 243–258. Sisson, T.W., Grove, T.L., 1993. Experimental investigations of the role of H2O in calc- Handley, H., Macpherson, C., Davidson, J., Berlo, K., Lowry, D., 2007. Constraining ﬂuid and alkaline differentiation and subduction zone magmatism. Contributions to Mineralo- sediment contributions to subduction-related magmatism in Indonesia, Ijen Volcanic gy and Petrology 113, 143–166.

Complex. Journal of Petrology 48, 1155–1183. Sisson, T.W., Layne, G.D., 1993. H2O in basalt and basaltic andesite glass inclusions from Hanyu, T., Tatsumi, Y., Nakai, S., Chang, Q., Miyazaki, T., Sato, K., Tani, K., Shibata, T., four subduction-related volcanoes. Earth and Planetary Science Letters 117, 619–635. Yoshida, T., 2006. Contribution of slab melting and slab dehydration to magmatism Sobolev, A.V., Hofmann, A.W., Kuzmin, D.V., Yaxley, G.M., Arndt, N.T., Chung, S.L., in the NE Japan arc for the last 25 Myr: constraints from geochemistry. G-cube 7, Danyushevsky, L.V., Elliott, T., Frey, F.A., Garcia, M.O., 2007. The amount of recycled Q08002. http://dx.doi.org/10.1029/2005GC001220. crust in sources of mantle-derived melts. Science 316, 412–417. Hawkesworth, C.J., Gallagher, K., Hergt, J.M., McDermott, F., 1993. Mantle and slab contri- Spandler, C., Pirard, C., 2013. Element recycling from subducting slabs to arc crust: a butions in arc magmas. Annual Review of Earth and Planetary Sciences 21, 175–204. review. Lithos 170–171, 208–223. Hermann, J., Spandler, C.J., 2008. Sediment melts at sub-arc depths: an experimental Steiger, R.H., Jager, E., 1977. Subcommission on geochronology: convention on the use of study. Journal of Petrology 49, 717–740. decay constants in geo- and cosmo-chronology. Earth and Planetary Science Letters Ishiwatari, A., Tsujimori, T., 2003. Paleozoic ophiolites and blueschists in Japan and 36, 359–362. Russian Primorye in the tectonic framework of East Asia: a synthesis. Island Arc 12, Stolz, A.J., Varne, R., Wheller, G.E., Foden, J.D., Abbott, M.J., 1988. The geochemistry and 190–206. petrogenesis of K-rich alkaline volcanics from the Batu Tara volcano, eastern Sunda Jicha, B.R., Singer, B.S., Brophy, J.G., Fournelle, J.H., Johnson, C.M., Beard, B.L., Lapen, T.J., arc. Contributions to Mineralogy and Petrology 98, 374–389. Mahlen, N.J., 2004. Variable impact of the subducted slab on Aleutian Island Arc Sun, S.S., McDonough, W.F., 1989. Chemical and isotopic systematics of oceanic basalts, magma sources: evidence from Sr, Nd, Pb, and Hf isotopes and trace element implications for mantle composition and processes. Geological Society of London, abundances. Journal of Petrology 45, 1845–1875. Special Publications 42, 313–345. Kogiso, T., Tatsumi, Y., Nakano, S., 1997. Trace element transport during dehydration Takagi, D., Sato, H., Nakagawa, M., 2005. Experimental study of a low-alkali tholeiite at processes in the subducted oceanic crust: 1. Experiments and implications for the 1–5 kbar: optimal condition for the crystallization of high-An plagioclase in hydrous origin of oceanic island basalts. Earth and Planetary Science Letters 148, 193–205. arc tholeiite. Contributions to Mineralogy and Petrology 149, 527–540. Kojima, S., 1989. Mesozoic terrane accretion in Northeast China, Sikhote-Alin and Japan Tang, K.D., 1990. Tectonic development of Paleozoic fold belts at the north margin of the regions. Palaeogeography, Palaeoclimatology, Palaeoecology 69, 213–232. Sino–Korean craton. Tectonics 9, 249–260. Labanieh, S., Chauvel, C., Germa, A., Quidelleur, X., Lewin, E., 2010. Isotopic hyperbolas Tatsumi, Y., Eggins, S., 1995. Subduction Zone Magmatism. Blackwell Science Publication, constrain sources and processes under the Lesser Antilles arc. Earth and Planetary Cambridge (211 pp.). Science Letters 298, 35–46. Taylor, S.R., McLennan, S.M., 1985. The Continental Crust: Its Composition And Evolution. Li, J.Y., 2006. Permian geodynamic setting of Northeast China and adjacent regions, Blackwell, Oxford Press (312 pp.). closure of the Paleo-Asian Ocean and subduction of the Paleo-Pacific Plate. Journal Thy, P., 1987. Petrogenetic implications of mineral crystallization trends of Troodos of Asian Earth Sciences 26, 207–224. cumulates, Cyprus. Geological Magazine 124, 415–430. Li, H.X., 2012. Origins of Late Paleozoic–Early Mesozoic Mafic Intrusions In The Yanbian Vervoort, J.D., Blichert-Toft, J., Patchett, P.J., Albarede, F., 1999. Relationships between Lu– Area, Ne China: Implications For Tectonic Evolution. (PhD thesis), Guangzhou Insti- Hf and Sm–Nd isotopic systems in the global sedimentary system. Earth and Plane- tute of Geochemistry, Chinese Academy of Sciences (169 pp. (in Chinese with English tary Science Letters 168, 79–99. abstract)). Vervoort, J.D., Plank, T., Prytulak, J., 2011. The Hf–Nd isotopic composition of marine sed- Li, X.H., Li, Z.X., Wingate, M.T.D., Chung, S.L., Liu, Y., Lin, G.C., Li, W.X., 2006. Geochemistry iments. Geochimica et Cosmochimica Acta 75, 5903–5926. of the 755 Ma Mundine Well dyke swarm, northwestern Australia: part of a Wang, F., Xu, W.L., Xu, Y.G., Gao, F.H., Ge, W.C., 2015. Late Triassic bimodal igneous rocks Neoproterozoic mantle superplume beneath Rodinia? Precambrian Research 146, in eastern Heilongjiang Province, NE China: implications for the initiation of subduc- 1–15. tion of the Paleo-Pacific Plate beneath Eurasia. Journal of Asian Earth Sciences 97, Li, C.W., Guo, F., Fan, W.M., Gao, X.F., 2007. Ar–Ar geochronology of Late Mesozoic volca- 406–423. nic rocks from the Yanji area, NE China and tectonic implications. Science in China Woodhead, J.D., Hergt, J.M., Davidson, J.P., Eggins, S.M., 2001. Hafnium isotope evidence Series D-Earth Sciences 50, 505– 518. for ‘conservative’ element mobility during subduction zone processes. Earth and Li, H.X., Guo, F., Li, C.W., Zhao, L., 2012. Petrogenesis of Early Cretaceous tonalites from the Planetary Science Letters 192, 331–346. Xiaoxinancha Au–Cu deposit. Geochimica 41, 497–514 (in Chinese with English Wu, F.Y., Jahn, B.M., Wilde, S., Sun, D.Y., 2000. Phanerozoic crustal growth: U–Pb and Sr– abstract). Nd isotopic evidence from the granites in northeastern China. Tectonophysics 328, Li, J.Y., Guo, F., Li, C.W., Li, H.X., Zhao, L., 2014. Neodymium isotopic variations of Late Pa- 89 –113. leozoic to Mesozoic I- and Atype granitoids in NE China: implications for tectonic Wu, F.Y., Yang, J.H., Lo, C.H., Wilde, S.A., Sun, D.Y., Jahn, B.M., 2007. The Heilongjiang evolution. Acta Petrologica Sinica 30, 1995–2008 (in Chinese with English abstract). Group: a Jurassic accretionary complex in the Jiamusi Massif at the western Pacific Liu, D.Y., Jian, P., Zhang, Q., Zhang, F.Q., Shi, Y.R., Shi, G.H., Zhang, L.Q., Tao, H., 2003. margin of northeastern China. Island Arc 16, 156–172. SHRIMP dating of adakites in the Tulingkai ophiolite, Inner Mongolia: evidence for Wu, F.Y., Sun, D.Y., Ge, W.C., Zhang, Y.B., Grant, M.L., Wilde, S.A., Jahn, B.M., 2011. Geochro- the Early Paleozoic subduction. Acta Geologica Sinica 77, 317–327 (in Chinese with nology of the Phanerozoic granitoids in northeastern China. Journal of Asian Earth English abstract). Sciences 41, 1–30. Longhi, J., Walker, D., Hays, J.F., 1978. The distribution of Fe and Mg between olivine and Xiao, W.J., Windley, B.F., Hao, J., Zhai, M.G., 2003. Accretion leading to collision and the lunar basaltic liquids. Geochimica et Cosmochimica Acta 42, 1545–1558. Permian Solonker suture, Inner Mongolia, China: termination of the central Asian Macdonald, R., Hawkesworth, C.J., Heath, E., 2000. The Lesser Antilles volcanic chain: a orogenic belt. Tectonics 226, 1069. http://dx.doi.org/10.1029/2002TC001484. study in arc magmatism. Earth-Science Reviews 49, 1–76. Xu, W.L., Ji, W.Q., Pei, F.P., Meng, E., Yu, Y., Yang, D.B., Zhang, X.Z., 2009. Triassic volcanism Maruyama, S., 1997. Pacific-type orogeny revisited: Miyashiro-type orogeny proposed. in eastern Heilongjiang and Jilin provinces, NE China: chronology, geochemistry, and Island Arc 6, 91–120. tectonic implications. Journal of Asian Earth Sciences 34, 392–402. Maruyama, S., Liou, J.G., Seno, T., 1989. Mesozoic and Cenozoic evolution of Asia. In: Ben- Xu, W.L., Wang, F., Meng, E., Pei, F.P., Yu, J.J., Tang, J., 2012. Paleozoic-early Mesozoic tec- Avraham, Z. (Ed.), The Evolution Of The Pacific Ocean Margins. Oxford University tonic evolution in the eastern Heilongjiang province, NE China: evidence from igne- Press, New York, pp. 75–99. ous rock association and U–Pb geochronology of detrital zircons. Journal of Jilin Miao, L.C., Fan, W.M., Liu, D.Y., Zhang, F.Q., Jian, P., Guo, F., Tao, H., Shi, Y.R., 2008. Geochro- University — Earth Science Edition 42, 1378–1389 (in Chinese with English abstract). nology and geochemistry of the Hegenshan ophiolitic complex: implications for late- Xu, W.L., Pei, F.P., Wang, F., Meng, E., Ji, W.Q., Yang, D.B., Wang, W., 2013. Spatial–tempo- stage tectonic evolution of the Inner Mongolia–Daxinganling orogenic belt, China. ral relationships of Mesozoic volcanic rocks in NE China: constraints on tectonic Journal of Asian Earth Sciences 32, 348–370. overprinting and transformations between multiple tectonic regimes. Journal of Panjasawatwong, Y., Danyushevsky, L.V., Crawford, A.J., Harris, K.L., 1995. An experimen- Asian Earth Sciences 74, 167–193. tal study of the effect of melt composition on plagioclase–melt equilibria at 5 and Yu, J.J., Wang, F., Xu, W.L., Gao, F.H., Pei, F.P., 2012. Early Jurassic mafic magmatism in the 10 kbar: implications for the origin of magmatic high-An plagioclase. Contributions Lesser Xing'an–Zhangguangcai Range, NE China, and its tectonic implications: con- to Mineralogy and Petrology 118, 420–432. straints from zircon U–Pb chronology and geochemistry. Lithos 142–143, 256–266. Pearce, J.A., Kempton, P.D., Nowell, G.M., Noble, S.R., 1999. Hf–Nd element and isotope Zeng, Q.D., Liu, J.M., Qin, K.Z., Fan, H.R., Chu, S.X., Wang, Y.B., Zhou, L.L., 2013. Types, perspective on the nature and provenance of mantle and subduction components characteristics, and time–space distribution of molybdenum deposits in China. in western Pacific arc-basin systems. Journal of Petrology 40, 1579–1611. International Geology Review 55, 1311–1358. Perfit, M.R., Gust, D.A., Bence, A.E., Arculus, R.J., Taylor, S.R., 1980. Chemical characteristics Zhang, Y.B., Wu, F.Y., Wilde, S.A., Zhai, M.G., Lu, X.P., Sun, D.Y., 2004. Zircon U–Pb ages and of island-arc basalts: implications formantle sources. Chemical Geology 30, 227–256. tectonic implications of “Early Paleozoic” granitoids at Yanbian, Jilin Province, north- Pirajno, F., Santosh, M., 2014. Rifting, intraplate magmatism, mineral systems and mantle east China. Island Arc 13, 484–505. dynamics in central-east Eurasia: an overview. Ore Geology Reviews 63, 265–295. Zhou, J.B., Wilde, S.A., 2013. The crustal accretion history and tectonic evolution of the NE Plank, T., Langmuir, C., 1998. The chemical composition of subducting sediment and its China segment of the Central Asian Orogenic Belt. Gondwana Research 23, consequences for the crust and mantle. Chemical Geology 145, 325–394. 1365–1377. Qi, L., Hu, J., Gregoire, D.C., 2000. Determination of trace elements in granites by induc- Zhou, J.B., Cao, J.L., Wilde, S.A., Zhao, G.C., Zhang, J.J., Wang, B., 2014. Paleo-Pacific tively coupled plasma mass spectrometry. Talanta 51, 507–513. subduction-accretion: evidence from Geochemical and U–Pb zircon dating of the Rollinson, H.R., 1993. Using Geochemical Data: Evaluation, Presentation, Interpretation. Nadanhada accretionary complex, NE China. Tectonics 33, 2444–2466. http://dx.doi. Longman Scientific&Technical(352pp.). org/10.1002/2014TC003637.