Age and evolution of the lithospheric mantle beneath ...

Lithos 282–283 (2017) 326–338

Contents lists available at ScienceDirect

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

Age and evolution of the lithospheric mantle beneath the Khanka Massif: Geochemical and Re–Os isotopic evidence from Sviyagino mantle xenoliths Peng Guo a, Wen-Liang Xu a,b,⁎, Chun-Guang Wang a, Feng Wang a, Wen-Chun Ge a, A.A. Sorokin c, Zhi-Wei Wang a a b c

College of Earth Sciences, Jilin University, Changchun 130061, China State Key Laboratory of Geological Processes and Mineral Resources, China University of Geosciences, Wuhan 430074, China Institute of Geology and Nature Management, Far Eastern Branch of the Russian Academy of Sciences, 1 Relochny, Line, Blagoveshchensk 675000, Russia

a r t i c l e

i n f o

Article history: Received 7 December 2016 Accepted 13 March 2017 Available online 30 March 2017 Keywords: Mantle xenoliths REE-in-two-pyroxene thermometer Re-Os isotopes Khanka Massif Central Asian Orogenic Belt

a b s t r a c t New geochemical and Re–Os isotopic data of mantle xenoliths entrained in Cenozoic Sviyagino alkali basalts from the Russian Far East provide insights into the age and evolution of the sub-continental lithospheric mantle (SCLM) beneath the Khanka Massif, within the Central Asian Orogenic Belt (CAOB). These mantle xenoliths are predominantly spinel lherzolites with minor spinel harzburgite. The lherzolites contain high whole-rock concentrations of Al2O3 and CaO, with low forsterite content in olivine (Fo = 89.5–90.3%) and low Cr# in spinel (0.09– 0.11). By contrast, the harzburgite is more refractory, containing lower whole rock Al2O3 and CaO contents, with higher Fo (91.3%) and spinel Cr# (0.28). Their whole rock and mineral compositions suggest that the lherzolites experienced low-degree (1–4%) batch melting and negligible metasomatism, whereas the harzburgite underwent a higher degree (10%) of fractional melting, and experienced minor post-melting silicate metasomatism. Twopyroxene rare earth element (REE)-based thermometry (TREE) yields predominant equilibrium temperatures of 884–1043 °C, similar to values obtained from two-pyroxene major element-based thermometry (TBKN = 942–1054 °C). Two lherzolite samples yield high TREE relative to TBKN (TREE − TBKN ≥71 °C), suggesting that they cooled rapidly as a result of the upwelling of hot asthenospheric mantle material that underplated a cold ancient lithosphere. The harzburgite with a low Re/Os value has an 187Os/188Os ratio of 0.11458, yielding an Os model age (TMA) relative to the primitive upper mantle (PUM) of 2.09 Ga, and a Re depletion ages (TRD) of 1.91 Ga; both of which record ancient melt depletion during the Paleoproterozoic (~2.0 Ga). The 187Os/188Os values of lherzolites (0.12411– 0.12924) correlate well with bulk Al2O3 concentrations and record the physical mixing of ancient mantle domains and PUM-like ambient mantle material within the asthenosphere. This indicates that the SCLM beneath the Khanka Massif had been formed since at least the Paleoproterozoic (~2.0 Ga), and was replaced by juvenile (Phanerozoic) mantle material accreted from the asthenosphere. The synthesis of available TRD ages for mantle-derived rocks and sulfides in xenoliths is consistent with the prior existence of a common Paleoproterozoic (~2.0 Ga) SCLM beneath the eastern CAOB. Finally, comparing of mantle TRD ages and the ages of crustal rocks suggests temporal and genetic links between crust and mantle formation during the evolution of the CAOB. © 2017 Elsevier B.V. All rights reserved.

1. Introduction The Central Asian Orogenic Belt (CAOB) is located between the Siberian and North China cratons (Fig. 1), and was formed by the

⁎ Corresponding author at: 2199 Jianshe Street, College of Earth Sciences, Jilin University, Changchun 130061, China. E-mail addresses: [email protected] (P. Guo), [email protected] (W.-L. Xu), [email protected] (C.-G. Wang), [email protected] (F. Wang), [email protected] (W.-C. Ge), [email protected] (A.A. Sorokin), [email protected] (Z.-W. Wang).

http://dx.doi.org/10.1016/j.lithos.2017.03.015 0024-4937/© 2017 Elsevier B.V. All rights reserved.

accretion of various geological terranes during the Phanerozoic (e.g., Sengör et al., 1993; Windley et al., 2007). The belt contains voluminous granitic intrusions and their volcanic equivalents (Wu et al., 2011; Xu et al., 2013). To date, many researches related to the lithosphere beneath the eastern CAOB have presented the timing of crustal growth and reworking (e.g., Jahn, 2004; Jahn et al., 2000; Wu et al., 2011); however, the age and evolution of the sub-continental lithospheric mantle (SCLM) beneath the crust remain unclear (Wu et al., 2003a, 2006; Zhang et al., 2011, 2012; Zhou et al., 2010). The eastern CAOB contains several micro-continental massifs, including, from west to east, the Erguna, Xing'an, Songnen-Zhangguangcai

P. Guo et al. / Lithos 282–283 (2017) 326–338

327

Fig. 1. Simplified distribution map of Cenozoic basalts and peridotite xenoliths in the CAOB and the North China Craton (NCC). The location of Sviyagino peridotite is marked with red pentagram. CAOB: Central Asian Orogenic Belt; NE China: Northeastern China. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Range, Jiamusi-Bureya, and Khanka massifs, all of which are separated by major faults (Fig. 1; Li, 2006; Sengör et al., 1993). Most available Re–Os isotopic and geochemical data for peridotite xenoliths within the eastern CAOB are from the Xing'an and Songnen–Zhangguangcai Range massifs in NE China. These peridotites range in fertility from fertile to moderately depleted (Wu et al., 2003a, 2006; Zhang et al., 2012; Zhou et al., 2010) and they have 187Os/188Os values within the range of compositions of the modern convecting mantle (0.122–0.130; Brandon et al., 2000; Harvey et al., 2006), in contrast to the highly refractory mantle beneath the Siberian Craton (e.g., Ionov et al., 2015a, 2015b). Fertile xenolith suites within the eastern CAOB are thought to represent young (Phanerozoic) continental lithospheric mantle that formed by the addition of asthenospheric mantle material (Griffin et al., 1999). This interpretation is consistent with the prevailing hypothesis that the majority of the crust within the CAOB is juvenile, with the CAOB representing the largest known area of Phanerozoic crustal growth on Earth (Jahn, 2004; Jahn et al., 2000; Sengör et al., 1993; Wu et al., 2003b). However, other studies have identified mantle xenoliths within the CAOB with low 187 Os/188Os and high 143Nd/144Nd ratios that yield Mesoproterozoic to Paleoproterozoic depletion ages (Deng and Macdougal, 1992; Xu et al., 2008; Zhang et al., 2011). Such isotope compositions suggest the presence of a minor ancient mantle component beneath the eastern CAOB. In addition, the Re–Os characteristics of abyssal peridotites are indicative of widely distributed ancient mantle domains within the asthenosphere that have not been homogenized by convective stirring (e.g., Harvey

et al., 2006; Liu et al., 2008a). However, it remains unclear whether the mantle xenoliths from the CAOB with ancient Os and Nd model ages provide evidence for the Proterozoic formation age of the lithospheric mantle beneath the eastern CAOB, or whether the entirety of the lithospheric mantle beneath the eastern CAOB is juvenile and was recently accreted from a heterogeneous asthenosphere (Zhang et al., 2012; Zhou et al., 2010). The Khanka Massif provides evidence of the age of formation and the composition of the SCLM beneath the CAOB. However, the SCLM beneath the Khanka Massif remains relatively poorly studied (Wang et al., 2015a), meaning in turn that the characteristics and evolution of the SCLM beneath the CAOB remain unclear. Here, we present new mineral, whole-rock, and Re–Os isotopic geochemical data for mantle xenoliths entrained within the Cenozoic Sviyagino alkali basalts of the Khanka Massif. These data provide direct constraints on the formation age of the SCLM beneath the Khanka Massif, providing insights into the deep mantle processes recorded in this area and allowing the modeling of genetic links between the mantle and crust during the formation and evolution of the CAOB. 2. Geologic background The eastern CAOB is composed of four micro-continental massifs, namely the Erguna, Xing'an, Songnen–Zhangguangcai Range, JiamusiBureya, and Khanka massifs (Fig. 1). Zircon U–Pb dating has identified

328

P. Guo et al. / Lithos 282–283 (2017) 326–338

deformed ancient (0.7–1.8 Ga) basement components within these massifs (e.g., Gou et al., 2013; Jia et al., 2004; Pei et al., 2007; Tang et al., 2013). In addition, several Precambrian supracrustal successions have been identified within the eastern CAOB, including the Jiageda Formation (Zhao et al., 2016) and the Xinghuadukou Group (Wu et al., 2012), both of which crop out within the Erguna Massif; the Dongfengshan and Tadong groups within the eastern margin of the Songnen–Zhangguangcai Range Massif (Wang et al., 2014); and the Kimkanskaya and Majiajie groups within the Jiamusi-Bureya Massif (Luan et al., in press). The Paleozoic tectonic evolution of the eastern CAOB was dominated by the amalgamation of micro-continental massifs, and the final closure of the Paleo-Asian Ocean (Jahn, 2004; Li, 2006; Sengör et al., 1993), whereas the Mesozoic tectonic evolution of the eastern CAOB was characterized by the overprinting of the circum-Pacific and Mongol–Okhotsk tectonic systems (Tang et al., 2015; Xu et al., 2013). The Khanka Massif is located at the eastern edge of the CAOB and is exposed mainly in the Russian Far East, with a small part cropping out in NE China (Fig. 1). It is separated from the Jiamusi–Bureya Massif to the north by the Dunhua–Mishan Fault and is bounded to the east by the Sikhote–Alin accretionary belt. Previous research suggests that the Khanka and the Jiamusi–Bureya massifs have formed a contiguous crustal unit since at least the Proterozoic and experienced granulite facies metamorphism at ~500 Ma (Wilde et al., 2003). The Khanka Massif also records several Paleozoic–Mesozoic tectonic and magmatic events, as evidenced by the presence of widespread granitic and more sporadic gabbro intrusions (Yang et al., 2014). Cenozoic alkali basalts are also widespread throughout the Khanka Massif and other regions of the eastern Eurasian continent and are thought to have formed as a result of rollback of the subducted paleo-Pacific Plate (Okamura et al., 1998; Xu et al., 2012). The host alkali basalts have an eruption age of ~12 Ma and have a within-plate basaltic geochemical signature that similar to alkali basalts in NE China (Okamura et al., 1998). More importantly, the eruption age of these alkali basalts postdates the early Miocene opening of the Japan Sea, which formed as a result of back-arc spreading associated with subduction of the Paleo-Pacific Plate (Okamura et al., 1998).

3. Xenolith petrography Eight round-shaped peridotite xenoliths ranging in size from ~6 × 5 × 4 cm to ~ 3 × 2 × 2 cm were analyzed during this study. The modal abundances of minerals within these samples were calculated using a least squares method employing whole-rock and average mineral compositions (Table 1). These modal abundances were

then combined with the International Union of Geological Sciences (IUGS) classification system for ultramafic rocks (Le Maitre, 1982) to determine that all but one of the samples are spinel lherzolites, with the remaining sample classified as a spinel harzburgite (sample R42-6). The lherzolites contain olivine (51–67%), orthopyroxene (18–29%), clinopyroxene (12–17%), and minor amounts of spinel (2–3%), whereas the harzburgite contains olivine (72%), orthopyroxene (22%), and minor amounts of clinopyroxene (4%) and spinel (2%). These samples are undeformed, fresh, and exhibit medium- to coarse-grained protogranular textures. Triple-junction (~ 120°) grain boundaries between olivine and pyroxenes are present in the harzburgite and in some of the lherzolites (Fig. 2a). Olivine and pyroxene have similar sizes and curvilinear grain boundaries (Fig. 2b), whereas spinel occurs either as an interstitial phase disseminated between other minerals (Fig. 2c) or as an inclusion within both olivine and pyroxene. Exsolved orthopyroxene lamellae are present within the clinopyroxene in the lherzolite (Fig. 2d), and no hydrous minerals were identified in any of the samples. 4. Analytical methods Fresh xenoliths were sawn from their basalt hosts and made into thin section for mineral analyses. The rest of specimens were then crushed to powders with sizes of ca. 200 mesh for analyses of whole rock. All analyses were performed at the State Key Laboratory of Lithospheric Evolution, Institute of Geology and Geophysics, Chinese Academy of Sciences. 4.1. Mineral major element and trace elements Major element compositions and BSE images of Sviyagino xenoliths were determined on a JEOL JXA-8100 Electron Probe, using a 15 kV accelerating voltage and a 10 nA beam current. Natural minerals and synthetic oxides were used as standards. Trace element concentrations of pyroxenes were analyzed using a laser ablation inductively coupled plasma mass spectrometer (LA-ICPMS). The LA-ICP-MS system consists of a GEOLAS 2005 pulsed ArF excimer laser coupled to an Agilent 7500a ICP-MS. The analyses were performed by laser ablating with a spot size of 60 μm for clinopyroxene and 90 μm for orthopyroxene at 5 Hz. NIST SRM 610 was used as the external standard in conjunction with internal standardization using Ca measured by electron microprobe for clinopyroxene and Si for olivine and orthopyroxene. The analytical uncertainty of the LA-ICP-MS method is less than 10% and data were processed using the GLITTER (ver 4.0) program.

Table 1 Petrography and temperatures of the peridotite xenoliths from Sviyagino, Khanka Massif. Temperatures ( ℃ )

Modal abundances (wt.%) Sample

Lithology

Ol

Opx

Cpx

Sp

Fo in Ol

Cr# in Sp

TP37

TWells

TBKN

Tavg

TREE

σ

R42-5 R42-6 R42-7 R42-8 R42-9 R42-10 R42-11 R42-12

Sp lher Sp harz Sp lher Sp lher Sp lher Sp lher Sp lher Sp lher

51 71 58 65 67 60 60 57

29 22 26 19 18 22 26 25

18 4 13 14 13 14 12 16

3 2 3 3 2 3 3 3

89.5% 91.3% 89.5% 90.3% 89.8% 89.7% 89.6% 89.6%

0.10 0.28 0.10 0.11 0.10 0.09 0.10 0.11

991 983 975 929 1029 924 993 977

978 982 978 909 992 913 982 975

1015 1017 1012 943 1054 942 1023 1008

995 994 988 927 1025 926 999 987

1023 1043 1029 884 992 982 1192 1079

45 9 44 23 23 34 4 20

Ol — olivine; Opx — orthophyroxene; Cpx — clinopyroxene; Sp — spinel; lher — lherzolite, harz — harzburgite; Fo — fosterite content. Fo = [Mg / (Mg + Fe)]; Cr# = [Cr / (Cr + Al)]. TP37: two-pyroxene thermometer of Putirka (2008, his Eq. 37), assuming a pressure of 1.5 GPa. TWells: two-pyroxene thermometer of Wells (1977). TBKN: two-pyroxene thermometer of Brey and Köhler (1990), assuming a pressure of 1.5 GPa. Tavg: average value of TBKN, TWells, and TP37. TREE: REE-in-two-pyroxene thermometer of Liang et al. (2013); σ: standard deviation of calculated TREE.

P. Guo et al. / Lithos 282–283 (2017) 326–338

329

Fig. 2. Petrography of the Sviyagino peridotites. (a) 120° triple junctions between olivine and pyroxene grains in spinel harzburgite (sample R42-6); (b) Curvilinear grain boundaries between olivine and pyroxene grains (R42-7); (c) Exsolved orthopyroxene within clinopyroxene (R42-8); (d) Protogranular microstructure of sample R42-12. Cpx: clinopyroxene; Ol: olivine; Opx: orthopyroxene; Sp: spinel; Sul: sulfide.

4.2. Whole rock major elements Major element compositions of whole rocks were determined by X-ray fluorescence (XRF; Rigaku RIX 2100 spectrometer) on fused glass disks, with an analytical uncertainty ranging from 1 to 3%. The detail procedures are the same as described by Liu et al. (2008b).

procedural blank has an Os content of 1 pg and an 187Os/188Os ratio of ~0.15. Total procedural blanks were about 9 pg for Re, 4 pg for Ir, 22 pg for Ru, 15 pg for Pt and 32 pg for Pd. The blank corrections were negligible (b1%) for Ir, Ru, Pt and Pd, but up to 3% for Re. 5. Results 5.1. Mineral major and trace element compositions

4.3. Highly siderophile elements (HSE) and Re–Os isotopes Whole rock Re–Os isotopes were analyzed by isotope dilution method. About 2 g of powder, together with Re–Os (i.e., 187Re and 190Os) and HSE (99Ru, 105Pd, 191Ir and 194Pt) isotope tracers, was digested with reverse aqua regia (i.e., 3 ml 12 N HCl and 6 ml 16 N HNO3) in a Carius Tube at 240 °C for ca. 48–72 h. Osmium was extracted from the aqua regia solution by solvent extraction into CCl4 and further purified by micro-distillation. Afterwards, Ru, Pd, Re, Ir and Pt were sequentially separated from the solution by anion exchange resin (AG-1 × 8, 100–200 mesh). Osmium isotopes were measured by N-TIMS on a GV Isoprobe-T instrument in a static mode using Faraday cups. To increase the ionization efficiency, Ba(OH)2 solution was used as an ion emitter. The measured Os isotopes were corrected for mass fractionation using the 192Os/188Os ratio of 3.0827. The Nier oxygen isotope composition (17O/16O = 0.0003708 and 18O/16O = 0.002045) has been used for oxide correction. The in-run precisions for Os isotopic measurements were better than 0.2% (2σ) for all the samples. Johnson-Matthey standard of UMD was used as an external standard and gave 187Os/188Os ratio of 0.11378 ± 2. The concentrations of other HSE were measured on a Neptune MCICPMS in peak-jumping mode or static mode, according to their measured signal intensities. In-run precisions for 185Re/187Re, 191Ir/193Ir, 99 Ru/101Ru, 194Pt/196Pt and 105Pd/106Pd were 0.1–0.3% (2δ). The total

The major element compositions of minerals within the Sviyagino peridotites are given in Supplementary material (Table S1), and trace element data for clinopyroxene and orthopyroxene are presented in Table S2. Core-to-rim compositional variations were not evident in any mineral in any of the samples analyzed; consequently, average values are given in Tables S1 and S2. 5.1.1. Olivine The forsterite content [Fo = Mg/(Mg + Fe)] of the olivine ranges from 89.5% to 91.3%, similar to the compositions of olivine within other Cenozoic xenoliths from the Khanka Massif (Wang et al., 2015a). The olivine within the lherzolites has lower Fo contents (89.5–90.3%) than the olivine within the harzburgite (Fo = 91.3%). 5.1.2. Orthopyroxene Harzburgite-hosted orthopyroxene has a higher Mg# (100 × Mg/(Mg + Fe); 91.6) and contain higher concentration of Cr2O3 (0.61 wt.%) and lower concentration of Al2O3 (3.64 wt.%) contents than orthopyroxene within the lherzolites (Mg# = 90.0–90.4, Cr2O3 = 0.29–0.38 wt.%, Al2O3 = 3.87–4.94 wt.%, Fig. 3). All of the orthopyroxenes have light rare earth element (LREE) depleted chondrite-normalized REE diagram patterns (Fig. 4a; normalized to the chondrite composition of Boynton, 1984) and yield primitive mantle

330

P. Guo et al. / Lithos 282–283 (2017) 326–338

Fig. 3. Variations of oxide abundances as a function of Mg# in orthopyroxene (Opx) (a–c) and clinopyroxene (Cpx) (d–f) in peridotite xenoliths from the Khanka Massif. Colored symbols are samples included from the Khanka Massif: blue and red circles are samples in this study; yellow circles are peridotite xenoliths from other papers (Ionov et al., 1995; Yamamoto et al., 2009). For comparison, voluminous xenoliths across the NCC compiled by Wang et al. (2013) are also shown: lighter gray dots are peridotites with relatively fertile compositions (Mg# in Opx = 88–91); darker gray dots are peridotites with refractory compositions (Mg# in Opx N 91); and black dots are peridotites and pyroxenites affected by peridotite and siliceous melt interaction. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

(PM)-normalized multi-element variation diagrams that show positive Ti, Ta, and U anomalies (Fig. 4c).

whereas the harzburgite-hosted clinopyroxene contains higher concentrations of Nb and Ta (Fig. 4b).

5.1.3. Clinopyroxene Lherzolite-hosted clinopyroxenes have Mg# values of 90.5–91.2 and contain 6.08–7.01 wt.% Al2O3 and 0.66–0.80 wt.% Cr2O3, whereas harzburgite-hosted clinopyroxene has higher Mg# values (92.6) and contains more Cr2O3 (1.37 wt.%) and less Al2O3 (4.84 wt.%; Fig. 3). The lherzolite-hosted clinopyroxenes have variably LREE-depleted chondrite-normalized REE patterns that yield [La/Yb]N ratios of 0.06–0.24 (Fig. 4a), whereas the harzburgite-hosted clinopyroxene has a flat REE pattern with a [La/Yb]N ratio of 0.67 (Fig. 4a). The lherzolitehosted clinopyroxenes also have higher total heavy rare earth element (∑ HREE) contents (10.36–11.65 ppm) than the harzburgite-hosted clinopyroxene (∑ HREE = 4.88 ppm). In addition, the lherzolitehosted clinopyroxenes are significantly depleted in Nb, Ta, Zr, and Ti,

5.1.4. Spinel The Cr# values [Cr# = Cr/(Cr + Al)] of spinels from the study area indicate that the lherzolite units contain Al-spinel and the harzburgite contains Cr-spinel (Table 1). The Al-spinels within the lherzolites have Cr# values of 0.09–0.11, which are at the low-Cr# end of the compositional range of lherzolitic spinels from the eastern North China Craton (NCC) (Fig. 5a). These Al-spinels have Al2O3 and Cr2O3 concentrations that vary from 57.0 to 59.0 wt.% and from 8.6 to 10.1 wt.%, respectively. In comparison, the Cr-spinel has a higher Cr# value (0.28) and contain higher concentrations of Cr2O3 (25.1 wt.%), but a lower Al2O3 (44.0 wt.%), indicating it is similar to spinels in moderately refractory peridotite from the NCC (Fig. 5a). All of these spinels plot within the olivine–spinel mantle array (Fig. 5b; Arai, 1994).

P. Guo et al. / Lithos 282–283 (2017) 326–338

331

Fig. 5. Variations of Cr# in spinel versus (a) Mg# in spinel and (b) Fo in olivine in peridotite xenoliths from the Khanka Massif. Symbols are identical to those in the Fig. 3. The pale yellow field in (b) marks the olivine-spinel mantle array (Arai, 1994). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

low loss on ignition (LOI) values (0.06–0.34 wt.%). The lherzolites contain 2.96–4.08 wt.% Al2O3 and 2.70–3.82 wt.% CaO and have Mg# values of 89.4–90.1, whereas the harzburgite contains lower concentrations of Al2O3 (1.92 wt.%) and CaO (1.04 wt.%), but has a higher Mg# (91.1). All of the lherzolites plot along the oceanic trend defined by Boyd (1989) in a Cao vs. Al2O3 diagram (Fig. 6a), whereas the harzburgite plots close to cratonic mantle field in this diagram. In addition, Mg# values decrease with increasing Al2O3 concentrations from the harzburgite to the lherzolites (Fig. 6b).

Fig. 4. (a) Chondrite-normalized REE patterns for clinopyroxenes (filled symbols) and orthopyroxenes (open symbols) in Sviyagino peridotite xenoliths. (b), (c) Primitive mantle-normalized trace element diagrams for clinopyroxenes and orthopyroxenes, respectively. Chondrite and primitive mantle values are from Boynton (1984) and Sun and Mcdonough (1989), respectively. PM: primitive mantle.

5.2.2. Re–Os isotopic compositions and HSE abundances The HSE and Re–Os isotopic compositions of the samples are given in Table S3. The Os concentrations (1.57–2.50 ppb) of the seven peridotites analyzed are relatively invariable, whereas the Re concentrations range from 0.03 to 0.34 ppb. The primitive upper mantle (PUM)-normalized HSE patterns for these samples are depleted in Pd relative to Ir (Fig. 7), yielding [Pd/Ir]N values of 0.32–0.64. In addition, the harzburgite sample and four of the lherzolites are depleted in Re relative to Pt and Pd, but two of the other lherzolite samples show positive Re anomalies. The 187Os/188Os values of the Khanka mantle xenoliths do not correlate well with their 187Re/188Os values (Fig. 8a), which vary from 0.06 to 0.77 and are generally lower than the value of the primitive upper mantle (0.43; Meisel et al., 2001), with the exception of lherzolite samples RF42-5 (0.45) and RF42-10 (0.77). The lherzolites have 187Os/188Os ratios of 0.12411–0.12924 that are within the range of the convecting mantle (0.122–0.130; Brandon et al., 2000; Harvey et al., 2006; Liu et al., 2008a). In comparison, the harzburgite sample has a lower 187 Os/188Os value (0.11458) that is similar to the composition of Siberian cratonic mantle peridotites (Fig. 8a; Ionov et al., 2015a, 2015b). Finally, the 187Os/188Os ratios of the lherzolite samples positively correlate with their whole-rock Al2O3 concentrations, but this relationship is not observed for the harzburgite samples (Fig. 8b).

5.2. Whole-rock geochemistry 5.3. Peridotite thermometry 5.2.1. Major elements The whole-rock major element compositions of the samples are given in Table 2. All of the analyzed samples are fresh and have very

Equilibrium temperatures of the Khanka mantle xenoliths were calculated using the major-element-based two-pyroxene thermometers of

332

P. Guo et al. / Lithos 282–283 (2017) 326–338

Table 2 Whole rock major trace elements (wt%) of Sviyagino mantle xenoliths. Sample

RF42-5

RF42-6

RF42-7

RF42-8

RF42-9

RF42-10

RF42-11

RF42-12

SiO2 TiO2 Al2O3 Fe2O3 FeO MgO MnO CaO Na2O K2O P2O5 LOI Mg#

45.73 0.12 4.08 1.18 6.67 36.62 0.14 3.82 0.2 0.015 0.044 0.22 89.4

44.23 0.032 1.92 0.63 6.96 43.21 0.13 1.04 0.064 0.014 0.005 0.34 91.1

44.92 0.092 3.45 0.8 7.33 38.61 0.14 2.82 0.15 0.02 0.072 0.14 89.5

44.37 0.097 3.33 0.71 7.2 40.05 0.14 2.81 0.23 0.008 0.13 0.11 90.1

44.48 0.097 2.96 0.66 7.49 40.1 0.14 2.74 0.22 0.038 0.23 0.09 89.8

44.65 0.12 3.79 0.86 7.18 38.81 0.14 3.02 0.22 0.012 0.078 0.13 89.7

45.1 0.095 3.45 0.99 7.31 39.19 0.14 2.7 0.14 0.014 0.072 0.06 89.5

45.41 0.1 3.69 0.98 6.98 38.19 0.14 3.43 0.18 0.015 0.055 0.12 89.6

Mg# = 100 × [Mg / (Mg + Fe)]

Brey and Köhler (1990), Wells (1977), and Putirka (2008; eq. 37), and the REE-in-two-pyroxene thermometer of Liang et al. (2013; here designated as TREE). Temperatures were calculated using average mineral core and rim compositions (Table 1; details of the TREE inversions are shown in the Fig. S1–S8). All of the TREE results have 1σ uncertainties less than ±50 °C after excluding LREE data outliers. The TREE values do not correlate with mineral compositions barring a positive correlation with CaO concentrations in orthopyroxene (Fig. S9), and the calculated TREE values for the harzburgite sample are similar to those for the lherzolite samples. The calculated TREE values range from 884 °C to 1192 °C, with the three major-element-based thermometers yielding consistent temperatures that have average values (Tavg) that fall within a relatively narrow range (926 °C–1025 °C). Fig.9 plots the calculated TREE against temperatures calculated using the two-pyroxene thermometers of Brey and Köhler (1990, TBKN), Wells (1977, TWells), and Putirka (2008; eq. 37 in that study, TP37), as well as Tavg. The pink field in each panel is defined by temperatures of well-equilibrated subcratonic peridotites from Liang et al. (2013), with six of the samples analyzed during this study plotting in this field and the remaining two lhertzolite samples (R42–12 and R42–11) having relatively high (TREE − TBKN) values, similar to abyssal peridotites.

6. Discussion 6.1. Partial melting and metasomatism in the lithospheric mantle beneath the Khanka Massif The studied xenoliths have variable mineral modes that range from fertile lherzolite to moderate refractory harzburgite. The mineral and clinopyroxene trace element compositions of these samples are consistent with a genetic model involving variable degree partial melting and the extraction of melts from the mantle, as inferred from the following lines of evidence: 1) a decrease in Al2O3 and increase in Cr2O3 concentrations in pyroxene with increasing pyroxene Mg# values (Fig. 3); 2) an increase in spinel Cr# values with increasing olivine Fo contents (Fig. 5b); 3) linear trends between whole rock Al2O3 and CaO concentrations, and between Al2O3 concentrations and Mg# values (Fig. 6); 4) the variable LREE depletion in clinopyroxene (e.g., [La/Yb]N = 0.06–0.67) (Fig. 4). In contrast, the relatively flat REE pattern of the clinopyroxene within the harzburgite provides evidence of cryptic metasomatism at a given extent of partial melting. Clinopyroxene is the major repository of incompatible elements within anhydrous spinel peridotites, although some high field strength

Fig. 6. Plots of whole rock major Al2O3 against CaO (a) and Mg# (b). Literature data from the Siberian Craton and NE China are also presented for comparison. The oceanic trend is from Boyd (1989). Values of the primitive mantle (PM) are from McDonough and Sun (1995). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) Date sources: the Siberian Craton (Doucet et al., 2013; Ionov et al., 2015a, 2015b); NE China (Pan et al., 2015; Wu et al., 2003a, 2006; Zhang et al., 2011; Zhang et al., 2012; Zhou et al., 2010).

P. Guo et al. / Lithos 282–283 (2017) 326–338

Fig. 7. Highly siderophile elements (HSE) patterns of Sviyagino peridotites. HSE concentrations were normalized to Primitive Upper Mantle (PUM, Becker et al., 2006).

elements (HFSE; e.g., Zr and Ti) and HREE (e.g., Dy, Er, and Yb) have an affinity to orthopyroxene (McDonough et al., 1992; Rampone et al., 1991). The clinopyroxene versus orthopyroxene partition ratios for some of the incompatible elements in our samples are relatively high (e.g., ZrCpx/ZrOpx = 11.3–33.9, YbCpx/YbOpx = 5.0–8.3, DyCpx/DyOpx = 12.1–49.7). This means that clinopyroxene trace element compositions can be used to identify mantle processes such as partial melting and post-melting metasomatism (Johnson et al., 1990). We estimated the degree of melt extraction and the mode of melting (i.e., fractional or batch melting) using clinopyroxene trace element systematic, and melting parameters and curves are from Norman (1998). We also assumed that clinopyroxene was the only residual mineral phase that contributed to trace element bulk partition coefficients (Fig. 10). Plotting these data on Y vs. Yb diagrams indicates that the lherzolites formed as a result of 1–3% fractional melting (Fig. 10a) or 1–4% batch melting (Fig. 10b). This low degree of partial melting means that Y and HREE concentrations are insensitive to the mode of melting (Hellebrand et al., 2001; Norman, 1998). However, the lhertzolite in the study area contain higher concentrations of highly incompatible elements (e.g., Nb and LREE) than would be expected for such a low degree of fractional melting, suggesting that these rocks are batch melting residues (see Fig. 10 in Norman, 1998). The harzburgite most likely formed as a result of either 10% fractional melting (Fig. 10a) or 20% batch melting (Fig. 10b). The latter appears to be inconsistent with the clinopyroxene Mg# of 92.6 (Norman, 1998), and more indicative of

333

10% fractional melting as established from the function between the extent of melting (F) and spinel Cr# (Hellebrand et al., 2001), i.e., F = 10 × ln(Cr#) + 24. Volatile-bearing metasomatic minerals such as hornblende, phlogopite, and apatite, were not observed in the samples analyzed during this study. This, combined with the presence of clinopyroxene with LREEdepleted and HREE-flat REE patterns, suggests that the lherzolites in this region underwent minimal or no metasomatism. This suggests that the concentrations of Nb, Sr, and Zr within the clinopyroxene in these lherzolites are a natural result of melting within the upper mantle rather than any other additional processes (Norman, 1998; Fig. 10c and d). The Ti depletions within these clinopyroxene can also be explained by the coincident Ti enrichments present within coexisting orthopyroxene (Rampone et al., 1991; Fig. 4c). Harzburgite-hosted clinopyroxenes have Zr and Sr concentrations that plot below the calculated melting trends shown in Fig. 10c and d, indicating that these elements were added after the partial melting event. In addition, the LREE concentrations within the clinopyroxenes in the harzburgite are greater than those expected for a melting residue after 10% fractional melting (Fig. 4a), suggesting that the LREE were added during postmelting metasomatism. The enrichment of Nb relative to Ba (Fig. 4b) indicates that this metasomatism was not caused by hydrous fluids, as Nb is relatively insoluble in these fluids compared with Ba (Keppler, 1996). This is also consistent with the lack of hydrous minerals in the harzburgite and strongly suggests that the metasomatic agent was a silicate or carbonatite melt. The low [La/Yb]N (0.67) and Zr/Hf (36), and high Ti/Eu (4250) ratios of the clinopyroxene within the harzburgite are consistent with metasomatism by a silicate rather than carbonatite melt (Coltori et al., 1999; Rudnick et al., 1993). In summary, the lherzolites in this study formed by low-degree (1–4%) batch melting with a negligible metasomatic overprint, whereas the harzburgite underwent relatively high-degree (10%) fractional melting and subsequent minor siliceous metasomatism that increased the contents of LREEs, Sr, and some of the HFSEs such as Zr. 6.2. Thermometric constraints on the rapid cooling of newly accreted lithospheric mantle The REE-in-two-pyroxene thermometer and major-element-based thermometers have been used to reveal the thermal evolution of mafic and ultramafic rocks (Liang et al., 2013). It has been shown that TREE temperature estimates agree well with those from major element-based thermometers for well-equilibrated sub-continental lithospheric mantle peridotites, but are 50–300 °C higher than the

Fig. 8. Diagrams of 187Os/188Os versus 187Re/188Os (a) and bulk Al2O3 contents (b). Symbols are identical to those in the Fig. 6. The mixing trajectories are made between the PUM (Al2O3 = 4.5 wt.%, 187Os/188Os = 0.1296) and an ancient mantle (Al2O3 = 0.7 wt.%, 187Os/188Os = 0.11457) with different Os contents of 0.975 ppb, 1.975 ppb, 3.9 ppb and 5.85 ppb. The PUM data are from Meisel et al. (2001). The data of global abyssal peridotites are shown as gray field (Brandon et al., 2000; Harvey et al., 2006; Liu et al., 2008a).

334

P. Guo et al. / Lithos 282–283 (2017) 326–338

Fig. 9. Correlations of calculated temperatures between REE-based and major element-based two-pyroxene thermometers for mantle xenoliths from the CAOB and the NCC. Error bars are 1σ uncertainties in TREE. Dashed lines are ±50 °C deviations from the 1:1 correlation lines. Temperatures of well-equilibrated subcratonic mantle (pink fields) are from Liang et al. (2013). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) Date sources: NE China (Pan et al., 2015; Zhang et al., 2000), the NCC (Liu et al., 2012; Wang et al., 2015b), and Vitim (Ionov et al., 2005).

major-element-derived temperatures for abyssal peridotites and pyroxenites (Liang et al., 2013; Wang et al., 2015b). Liang et al. (2013) attributed this discrepancy to differences in diffusion rate (and hence closure temperatures) at a given cooling rate between the trivalent REE and divalent Fe, Mg, and Ca in pyroxene. Therefore, comparison of temperatures calculated by REE- and major-element-based two-pyroxene thermometers can provide important insights into the thermal evolution of peridotite xenoliths. The concentrations of Al2O3, MgO, and Cr2O3, and trace elements within pyroxenes in this study were controlled mainly by melting and melt–rock interaction, and do no correlation with TREE, as also reported by Dygert and Liang (2015) and Wang et al. (2015b). There is also no correlation between TREE and mineral modal abundance, and the difference in TREE between harzburgite and lherzolite is negligible. In contrast, TREE correlates positively with orthopyroxene CaO concentration (Fig. S9). The CaO content of pyroxene is known to be strongly dependent on temperature, with CaO diffusing out of orthopyroxene and into clinopyroxene during cooling and sub-solidus re-equilibration (Nickel et al., 1985). The peridotites analyzed during this study mainly overlap with the composition of well-equilibrated continental peridotites analyzed by Liang et al. (2013) in a TREE vs. TBKN diagram (Fig. 9a). In addition, Liang et al. (2013) suggested that temperature estimates from REE- and major element-based thermometers for peridotites formed in a stable thermal regime are expected to be similar (or identical). This, combined with the fact that the six well-equilibrated peridotite samples of this study yield similar TREE and TBKN, suggests

that the samples were probably derived from the SCLM under relatively stable thermal conditions. Two of the lherzolite samples (R42–11 and R42–12) have relatively high (TREE − TBKN) values (169 °C and 71 °C, respectively) that are comparable to temperature discrepancies recorded in abyssal peridotites (Liang et al., 2013). These discrepancies may relate to variation in grain size, cooling rate, and/or possibly melt–rock interaction in the upper mantle (Liang et al., 2013; Yao and Liang, 2015). The pyroxenes within all of the samples are medium- to coarse-grained and have chemically re-equilibrated from core to rim, thereby excluding grain size as a factor in explaining the temperature discrepancy. The fact that samples R42-11 and R42-12 are both the products of partial melting and have negligible metasomatic overprints means that the discrepancies in temperature estimates probably resulted from differences in cooling rate. Moderate to high cooling rates mean that the REE-based thermometer is more likely to produce a higher temperature estimate than the major-element-based thermometer (Dygert and Liang, 2015; Liang et al., 2013; Yao and Liang, 2015), suggesting the two lherzolites with higher TREE than TBKN underwent relatively rapid cooling. Recent seismic imaging and three-dimensional waveform modeling have identified a slow but continuous seismic anomaly beneath the southeastern edge of the CAOB that extends from slightly deeper than 660 km to the surface (Tang et al., 2014). This widespread lowvelocity zone has been interpreted as a large volume of hot asthenospheric material upwelling after breaking through a gap in the subducting Pacific slab (Guo et al., 2016; Tang et al., 2014). We propose

P. Guo et al. / Lithos 282–283 (2017) 326–338

335

Fig. 10. Diagrams of fractional (a) and batch (b) partial melting modeling with an assumption of a primitive mantle source using trace elements in clinopyroxenes from Sviyagino peridotites. Subscript N indicate normalization to primitive mantle (Sun and Mcdonough, 1989). Plots of (c) and (d) imply later addition of Zr and Sr after melting, respectively. The detailed parameters and melting curves are from Norman (1998). Symbols are identical to those in the Fig. 3.

that rapid cooling of the two lherzolite samples occurred when upwelling hot asthenospheric mantle material underplated the overlying cold ancient lithosphere. 6.3. Age of the SCLM beneath the Khanka Massif The timing of melt extraction from peridotites can be constrained using Re–Os isotope systematic to determine the formation age of lithospheric mantle (e.g., Walker et al., 1989; Rudnick and Walker, 2009). As is the case for most geochronological techniques, the majority of ages are derived from isochrons. The Sviyagino mantle xenoliths have a relatively scattered distribution in an 187Re/188Os vs. 187Os/188Os diagram (Fig. 8a), with two lherzolites having 187Re/188Os ratios that exceed the estimated composition of the primitive mantle (Meisel et al., 2001), meaning that an isochron cannot be constructed for these samples. However, although a true Re–Os isochron cannot be constructed, age information can still be extracted from the Re–Os compositions through model ages (Walker et al., 1989). One method of constraining the timing of melt depletion recorded by mantle peridotites is Re-Os model age (TMA), where measured 187Re/188Os ratios are used to calculate the time of separation of a given sample from the mantle reservoir (Shirey and Walker, 1998). The TMA ages for the six lherzolites analyzed during this study range from 0.05 to 0.77 Ga, whereas the harzburgite yield a TMA age of 2.09 Ga (Table S3). The timing of melt depletion of a mantle peridotite can also be constrained using the Re depletion age (TRD) that is calculated assuming that the peridotite lost all Re during melt extraction. This approach ignores the inherent increase in 187Os due to Re in the sample, and provides a minimum estimate of the timing of melt depletion. The TRD ages of the studied lherzolites range from 0.09 to 0.52 Ga, with the exception of sample RF42-5 that yields a future age of − 0.25 Ga. The harzburgite has the lowest 187Os/188Os value (0.11458), yields a TRD age of 1.91 Ga, and has a PUM-normalized HSE pattern that is characterized by depletions in Pd and Re relative to Ir

(Fig. 7). This suggests that the HSE and Re–Os contents of this sample were not significantly affected by late-stage metasomatism, especially the alteration of sulfides. The refractory nature and low Re/Os value of the harzburgite mean that the TRD age of this sample should closely approximate the true timing of melt depletion, and the consistency of TMA and TRD ages for this sample suggest that ancient melt depletion occurred during the Paleoproterozoic (~ 2.0 Ga). It should also be noted that the TMA and TRD ages of the lherzolites are unreliable due to the retention of significant amounts of Re and associated elevated Re/Os values (Rudnick and Walker, 2009). An alternative approach to estimating mantle melt depletion ages is to plot measured 187Os/188Os values against whole-rock Al2O3 concentrations to produce a “alumina-chron” (Reisberg and Lorand, 1995). This yields a strong correlation between the 187Os/188Os values and Al2O3 concentrations of the lherzolites but not the harzburgite, suggesting that the former may not be genetically related to the latter (Fig. 8b). The consumption of Re during melting is thought to leave 0.7 wt.% Al2O3 within the resulting mantle residue (Handler and Bennett, 1999). Extrapolation of the lherzolite alumina-chron to 0.7 wt.% Al2O3 yields an 187 Os/188Os value ~0.11457 and a model age of 1.9 Ga, suggesting that the lherzolites may also record Paleoproterozoic melt-depletion events. There is also growing evidence that ancient mantle domains that have not been homogenized by convection are widely distributed throughout the asthenosphere (e.g., Harvey et al., 2006; Liu et al., 2008a). The linear correlation between 187Os/188Os values and Al2O3 concentrations may be attributed to the physical mixing of ancient mantle with surrounding juvenile mantle material in the asthenosphere, a mechanism that may explain the wide spectrum of Os isotopic compositions observed in abyssal peridotites (Liu et al., 2015). The variability of the Os isotope compositions of the studied lherzolites can be modeled by mixing a Paleoproterozoic mantle end-member (Al2O3 = 0.7 wt.%, 187Os/188Os = 0.11457) with a PUM-like surrounding mantle (Al2O3 = 4.5 wt.%, 187Os/188Os = 0.1296) that contains variable

336

P. Guo et al. / Lithos 282–283 (2017) 326–338

amounts of Os (Fig. 8b). In this model, the lherzolites represent juvenile (Phanerozoic) mantle that recently accreted from a region of the asthenosphere containing ancient mantle domains that were well mixed with the surrounding convecting mantle. This model is consistent with the observed fertile mineral and whole-rock compositions of these lherzolites. However, we interpret the harzburgite to represent relic Paleoproterozoic SCLM material that formed prior to the accretion of the fertile mantle material. The harzburgite contains elevated concentrations of Al2O3 (1.9 wt.%) and has a 187Os/188Os value of 0.11457, plotting right of the alumina-chron (Fig. 8b). This is inconsistent with this harzburgite representing ancient mantle domains preserved in the juvenile mantle that were recently accreted from the asthenosphere, primarily as any recent melt extraction associated with the formation of the lithospheric mantle would reduce the Al2O3 content while having negligible effect on 187Os/188Os values (Liu et al., 2015). The elevated Al2O3 content of the harzburgite can be attributed to the addition of a minor (b 5%) amount of basalt (typically low-Os) to the system after melt depletion (Reisberg and Lorand, 1995; Rudnick and Walker, 2009). Whole rock Re–Os isotopic data suggest that the SCLM beneath the Khanka Massif began forming by at least the Paleoproterozoic, a hypothesis that is consistent with the Nd model ages of metamorphic rocks in the eastern Khanka Massif (Kruk et al., 2014). In addition, Permian granitoids emplaced within the massif yield Paleoproterozoic Hf model ages (Yang et al., 2014). This correlation of mantle and crustal ages corroborates the hypothesis that both the differentiation of pristine mantle material and the evolution of the crust in the Khanka lithosphere occurred during the Paleoproterozoic. The presence of numerous fertile lherzolite mantle xenoliths within the Khanka Massif, combined with the geothermometric data obtained during this study, suggests that the cold Paleoproterozoic lithospheric mantle in this region was replaced relatively recently by juvenile (i.e., Phanerozoic) mantle that accreted from upwelling hot asthenospheric mantle.

6.4. Implications for the lithospheric architecture of the eastern CAOB The new data of this study suggest that the pristine SCLM underlying the Khanka Massif formed during the Paleoproterozoic at ~2.0 Ga. Ancient SCLM has also been identified in other massifs within the eastern CAOB. For instance, refractory harzburgites within the Keluo area of the Xing'an Massif yield TRD ages of ~2.0 Ga (Zhang et al., 2011), indicating that SCLM beneath this massif also formed during the Paleoproterozoic. Refractory peridotite xenoliths from the Shuangliao, Yitong, Jiaohe, and Wangqing areas of the Songnen–Zhangguangcai Range Massif have also yielded ages of 1.3–1.6 Ga (Wu et al., 2003a; Zhou et al., 2010), slightly younger than the TRD ages obtained from adjacent massifs. However, recent research suggests that the basalts that host these mantle xenoliths were derived from a source containing recycled oceanic crustal material (Xu et al., 2012; Yu et al., 2010). The melts produced by this young oceanic crustal material had high 187 Os/188Os values (Brandon et al., 1998) and percolated upwards, reacting with solid ancient mantle material at shallower depths (e.g., Kelemen et al., 1992). This meant that ancient, unradiogenic 187 Os/188Os ratios were able to re-equilibrate with more radiogenic Os within melts generated by extensive partial melting in a back-arc setting (Meibom and Frei, 2002). Thus, due to the mass balance between radiogenic and unradiogenic Os components, the mantle xenoliths in this area yield elevated 187Os/188Os values relative to pristine SCLM. The TRD ages obtained for these xenoliths also provide minimum estimates of true melt depletion ages. This, combined with the earlier discovery of Paleoproterozoic (~ 1.8 Ga) basement rocks from the overlying crust in these regions (Pei et al., 2007), indicates that the SCLM underlying the Songnen–Zhangguangcai Range Massif may also have formed during the Paleoproterozoic. These data are consistent with the prior existence of a common ~2 Ga Paleoproterozoic SCLM beneath the eastern CAOB.

Fig. 11. Age distribution plots for ancient lithospheric materials from the eastern CAOB. (a) TRD ages for mantle-derived rocks (Wu et al., 2003a, 2006; Zhang et al., 2011; Zhang et al., 2012; Zhou et al., 2010) and sulfides in xenoliths (Wang et al., 2015a; Xu et al., 2008); (b) U–Pb ages of detrial zircons from Precambrian strata (Luan et al., in press; Wang et al., 2014; Wu et al., 2012; Zhao et al., 2016); (c) Nd model ages of Mesozoic granites and Paleozoic metamorphic rocks (Kruk et al., 2014; Wu et al., 2003b and references therein) and a compile of published Hf model ages of granites with crystallization ages vary from the Neoproterozoic to the Mesozoic.

Fig. 11 compares the TRD ages of mantle-derived rocks (with preference given to refractory peridotites with low Al2O3 concentrations and high Mg# values) and sulfides in xenoliths with the ages of crustal rocks from the eastern CAOB, including detrital zircon U–Pb ages for Precambrian supracrustals, and Nd and Hf model ages (TDM) for magmatic and metamorphic rocks. Crustal rock ages have a common peak at 1.8–1.9 Ga, indicating that the Paleoproterozoic was an important period of crustal growth in the eastern CAOB. This age of early crustal growth was coeval with the formation of pristine SCLM in this area. The whole rock TRD and Hf model ages peak at 1.2–1.3 Ga could record the accretion of new SCLM and associated crust. The TRD ages of mantlederived rocks and sulfides in xenoliths both peak at 0.75–0.85 Ga, identical (within uncertainty) to the peaks in the U–Pb and Nd model ages. This peak records either the accretion of new SCLM and associated crust (e.g., Jahn, 2004; Jahn et al., 2000; Wu et al., 2003b), or a Neoproterozoic magmatic event (e.g., Tang et al., 2013; Wang et al., 2014) that affected both the SCLM and the overlying crust. However, the multiple tectonic events since the Paleozoic and recent asthenospheric upwelling have stripped away most of the old lithosphere, leaving residues of ancient SCLM and crust. A comparison of mantle TRD and crustal rock ages suggests temporal coupling and genetic links between the formation of the crust and mantle during the evolution of the CAOB. 7. Conclusions 1. The lithospheric mantle beneath the Khanka Massif is composed predominantly of spinel lherzolite with minor spinel harzburgite. The lherzolites experienced low-degree (1–4%) batch melting with a negligible metasomatic overprint, whereas the harzburgite underwent a higher degree (10%) of fractional melting and minor post-melting siliceous metasomatism. 2. Two lherzolites with high TREE, relative to TBKN, underwent relatively rapid cooling when upwelling hot asthenospheric mantle underplated the cold ancient lithosphere.

P. Guo et al. / Lithos 282–283 (2017) 326–338

3. The lithospheric mantle beneath the Khanka Massif was formed at least in the Paleoproterozoic (~2.0 Ga), and was recently replaced by juvenile (Phanerozoic) mantle that accreted from the asthenosphere. 4. A synthesis of available TRD ages for mantle-derived rocks and sulfides in xenoliths is consistent with the prior existence of a common ~2.0 Ga Paleoproterozoic SCLM beneath the eastern CAOB. The comparison of TRD ages of mantle materials with the ages of crustal rocks suggests temporal coupling and genetic links between crust and mantle formation in the building of the CAOB.

Acknowledgements The authors would like to thank the journal Editor in-chief Dr. Andrew Kerr and two anonymous reviewers for their constructive comments and suggestions. We also thank Yu-Guang Ma, Yue-Heng Yang, Chang Zhang, Zhu-Yin Chu and the staff members from the State Key Laboratory of Lithospheric Evolution, Institute of Geology and Geophysics, Chinese Academy of Sciences for their help in geochemical analyses. This work was financially supported by the National Natural Science Foundation of China (Grant 41330206), the National Key Basic Research Program of China (2013CB429803) and the Graduate Innovation Fund of Jilin University (Project 2016032).

Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.lithos.2017.03.015. References Arai, S., 1994. Characterization of spinel peridotites by olivine–spinel compositional relationships: review and interpretation. Chemical Geology 113, 191–204. Becker, H., et al., 2006. Highly siderophile element composition of the Earth's primitive upper mantle: constraints from new data on peridotite massifs and xenoliths. Geochimica et Cosmochimica Acta 70 (17), 4528–4550. Boyd, F.R., 1989. Compositional distinction between oceanic and cratonic lithosphere. Earth and Planetary Science Letters 96, 15–26. Boynton, W.V., 1984. Cosmochemistry of the Rare Earth Elements: Meteorite Studies. In: Henderson, P. (Ed.), Rare Earth Element Geochemistry. Elsevier Science Publishing Company Inc., New York, pp. 63–114. Brandon, A.D., Creaser, R.A., Shirey, S.B., Carlson, R.W., 1998. Osmium recycling in subduction zones. Science 272, 861–864. Brandon, A.D., Snow, J.E., Walker, R.J., Morgan, J.W., Mock, T.D., 2000. 190Pt-186Os and 187 Re-187Os systematics of abyssal peridotites. Earth and Planetary Science Letters 177 (3–4), 319–335. Brey, G.P., Köhler, T., 1990. Geothermobarometry in four-phase lherzolites II. New thermobarometers, and practical assessment of existing thermobarometers. Journal of Petrology 31, 1353–1378. Coltori, M., Bonadiman, C., Hinton, R.W., Siena, F., Upton, B.G., 1999. Carbonate metasomatism of the oceanic upper mantle: evidence from clinopyroxenes and glasses in ultramafic xenoliths of Grande Comore, Indian Ocean. Journal of Petrology 40, 133–165. Deng, F.L., Macdougal, J.D., 1992. Proterozoic depletion of the lithosphere recorded in mantle xenoliths from Inner Mongolia. Nature 360, 333–336. Doucet, L., Ionov, D., Golovin, A., 2013. The origin of coarse garnet peridotites in cratonic lithosphere: new data on xenoliths from the Udachnaya kimberlite, central Siberia. Contributions to Mineralogy and Petrology 165, 1225–1242. Dygert, N., Liang, Y., 2015. Temperatures and cooling rates recorded in REE in coexisting pyroxenes in ophiolitic and abyssal peridotites. Earth and Planetary Science Letters 420, 151–161. Gou, J., Sun, D.Y., Ren, Y.S., Liu, Y.J., Zhang, S.Y., Fu, C.L., Wang, T.H., Wu, P.F., Liu, X.M., 2013. Petrogenesis and geodynamic setting of Neoproterozoic and Late Paleozoic magmatism in the Manzhouli-Erguna area of Inner Mongolia, China: geochronological, geochemical and Hf isotopic evidence. Journal of Asian Earth Sciences 67–68, 114–137. Griffin, W.L., O'Reilly, S.Y., Ryan, C.G., 1999. The Composition and Origin of Subcontinental Lithospheric Mantle. In: Fei, Y., Berka, C., Mysen, B.O. (Eds.), Mantle Petrology: Field Observations and High-Pressure Experimentation. The Geochemical Society, Stony Brook, N.Y, pp. 13–45. Guo, Z., Chen, Y.J., Ning, J.Y., Yang, Y.J., Afonso, J.C., 2016. Seismic evidence of on-going sublithosphere upper mantle convection for intra-plate volcanism in Northeast China. Earth and Planetary Science Letters 433, 31–43. Handler, M.R., Bennett, V.C., 1999. Behaviour of platinum-group elements in the subcontinental mantle of eastern Australia during variable metasomatism and melt depletion. Geochimica et Cosmochimica Acta 63 (21), 3597–3618.

337

Harvey, J., Gannoun, A., Burton, K.W., Rogers, N.W., Alard, O., Parkinson, I.J., 2006. Ancientmelt extraction fromthe oceanic upper mantle revealed by Re–Os isotopes in abyssal peridotites from the mid-Atlantic ridge. Earth and Planetary Science Letters 244, 606–621. Hellebrand, E., Snow, J.E., Dick, H.J.B., Hofmann, A.W., 2001. Coupled major and trace elements as indicators of the extent of melting in mid-ocean-ridge peridotites. Nature 410, 677–681. Ionov, D.A., Prikhod'Ko, V.S., O'Reilly, S.Y., 1995. Peridotite xenoliths in alkali basalts from the Sikhote-Alin, southeastern Siberia, Russia: trace-element signatures of mantle beneath a convergent continental margin. Chemical Geology 120, 275–294. Ionov, D.A., Ashchepkov, I., Jagoutz, E., 2005. The provenance of fertile off-craton lithospheric mantle: Sr–Nd isotope and chemical composition of garnet and spinel peridotite xenoliths from Vitim, Siberia. Chemical Geology 217, 41–75. Ionov, D.A., Doucet, L.S., Carlson, R.W., Golovin, A.V., Korsakov, A.V., 2015a. Post-Archean formation of the lithospheric mantle in the central Siberian craton: Re–Os and PGE study of peridotite xenoliths from the Udachnaya kimberlite. Geochimical et Cosmochimica Acta 165, 466–483. Ionov, D.A., Carlson, R.W., Doucet, L.S., Golovin, A.V., Oleinikov, O.B., 2015b. The age and history of the lithospheric mantle of the Siberian craton: Re–Os and PGE study of peridotite xenoliths from the Obnazhennaya kimberlite. Earth and Planetary Science Letters 428, 108–119. Jahn, B.M., Wu, F.Y., Chen, B., 2000. Massive granitoid generation in central Asia: Nd isotopic evidence and implication for continental growth in the Phanerozoic. Episodes 23, 82–92. Jahn, B.M., 2004. The Central Asian Orogenic Belt and Growth of the Continental Crust in the Phanerozoic. In: Malpas, J., Fletcher, C.J.N., Ali, J.R., Aitchison, J.C. (Eds.), Aspects of the Tectonic Evolution of China. Geological Society London Special Publication Vol. 226, pp. 73–100. Jia, D.C., Hu, R.Z., Lu, Y., Qiu, X.L., 2004. Collision belt between the Khanka block and the North China block in the Yanbian Region, Northeast China. Journal of Asian Earth Sciences 23, 211–219. Johnson, K.T.M., Dick, H.J.B., Shimizu, N., 1990. Melting in the upper mantle: an ion probe study of diopsides in abyssal peridotites. Journal of Geophysical Research 95, 2661–2678. Kelemen, P.B., Dick, H.J.B., Quick, J.E., 1992. Formation of harzburgite by pervasive melt/ rock reaction in the upper mantle. Nature 358, 635–641. Keppler, H., 1996. Constraints from partitioning experiments on the composition of subduction-zone fluids. Nature 380, 237–239. Kruk, N.N., Kovach, V.P., Golozubov, V.V., Kasatkin, S.A., Lavrent'eva, L.B., Lavrik, S.N., 2014. Nd-isotope systematic of metamorphic rocks of the south of Russian Far East. Doklady Earth Sciences 455 (1), 233–237. Le Maitre, R.W., 1982. A Classification of Igneous Rocks and Glossary of Terms. Blackwell Scientific Publications, London. Li, J.Y., 2006. Permian geodynamic setting of Northeast China and adjacent regions: closure of the Paleo-Asian Ocean and subduction of the Paleo-Pacific plate. Journal of Asian Earth Sciences 26, 207–224. Liang, Y., Sun, C., Yao, L., 2013. A REE-in-two-pyroxene thermometer for mafic and ultramafic rocks. Geochimica et Cosmochimica Acta 102, 246–260. Liu, C.Z., Snow, J.E., Hellebrand, E., Brugmann, G., Handt, A., Buchl, A., Hofmann, A.W., 2008a. Ancient, highly heterogeneous mantle beneath Gakkel ridge, Arctic Ocean. Nature 452, 311–316. Liu, J.G., Carlson, R.W., Rudnick, R.L., Walker, R.J., Gao, S., Wu, F.Y., 2012. Comparative Sr– Nd–Hf–Os–Pb isotope systematics of xenolithic peridotites from Yangyuan, North China Craton: additional evidence for a Paleoproterozoic age. Chemical Geology 332, 1–14. Liu, J.G., Scott, J.M., Martin, C.E., Pearson, D.G., 2015. The longevity of Archean mantle residues in the convecting upper mantle and their role in young continent formation. Earth and Planetary Science Letters 424, 109–118. Liu, Y.S., Zong, K.Q., Kelemen, P.B., Gao, S., 2008b. Geochemistry and magmatic history of eclogites and ultramafic rocks from the Chinese continental scientific drill hole: subduction and ultrahigh-pressure metamorphism of lower crustal cumulates. Chemical Geology 247, 133–153. Luan, J.P., Wang, F., Xu, W.L., Ge, W.C., Sorokin, A.A., Wang, Z.W., Guo, P., 2017. Formation timing and provenance of Neoproterozoic strata in the Jiamusi massif and its tectonic implications: evidence from detrital zircon U–Pb ages. Precambrian Research In press. McDonough, W.F., Stosch, H.G., Ware, N.G., 1992. Distribution of titanium and the rare earth elements between peridotitic minerals. Contributions to Mineralogy and Petrology 110, 321–328. McDonough, W.F., Sun, S.S., 1995. The composition of the Earth. Chemical Geology 120 (3–4), 223–253. Meibom, A., Frei, R., 2002. Evidence for an ancient osmium isotopic reservoir in earth. Science 296, 516–518. Meisel, T., Walker, R.J., Irving, A.J., Lorand, J.P., 2001. Osmium isotopic compositions of mantle xenoliths: a global perspective. Geochimica et Cosmochimica Acta 65 (8), 1311–1323. Nickel, K.G., Brey, G.P., Kogarko, L., 1985. Orthopyroxene–clinopyroxene equilibria in the system CaO–MgO–Al2O3–SiO2 (CMAS): new experimental results and implications for two-pyroxene thermometry. Contributions to Mineralogy and Petrology 91, 44–53. Norman, M.D., 1998. Melting and metasomatism in the continental lithosphere: laser ablation ICPMS analysis of minerals in spinel lherzolites from eastern Australia. Contributions to Mineralogy and Petrology 130, 240–255. Okamura, S., Martynov, Y.A., Furuyama, K., Nagao, K., 1998. K–Ar ages of the basaltic rocks from Far East Russia: constraints on the tectono-magmatism associated with the Japan Sea opening. Island Arc 7, 271–282. Pan, S.K., Zheng, J.P., Griffin, W.L., Xu, Y., Li, X., 2015. Nature and evolution of the lithospheric mantle beneath the eastern central asian orogenic belt: constraints from

338

P. Guo et al. / Lithos 282–283 (2017) 326–338

peridotite xenoliths in the central part of the Great Xing'an Range, NE China. Lithos 238, 52–63. Pei, F.P., Xu, W.L., Yang, D.B., Zhao, Q.G., Liu, X.M., Hu, Z.C., 2007. Zircon U–Pb geochronology of basement metamorphic rocks in the Songliao Basin. Chinese Science Bulletin 52 (7), 942–948. Putirka, K.D., 2008. Thermometers and barometers for volcanic systems. Reviews in Mineralogy and Geochemistry 69, 61–120. Rampone, E., Bottazzi, P., Ottolini, L., 1991. Complementary Ti and Zr anomalies in orthopyroxene and clinopyroxene from mantle peridotites. Nature 354, 518–520. Reisberg, L., Lorand, J.P., 1995. Longevity of sub-continental mantle lithosphere from osmium isotope systematics in orogenic peridotite massifs. Nature 376, 159–162. Rudnick, R.L., McDonough, W.F., Chappell, B.W., 1993. Carbonatite metasomatism in the northern Tanzanian mantle: petrographic and geochemical characteristics. Earth and Planetary Science Letters 114, 463–475. Rudnick, R.L., Walker, R.J., 2009. Interpreting ages from Re–Os isotopes in peridotites. Lithos 112S, 1083–1095. Sengör, A., Natal'in, B.A., Burtman, V.S., 1993. Evolution of the Altaid tectonic collage and Palaeozoic crustal growth in Eurasia. Nature 364, 299–307. Shirey, S.B., Walker, R.J., 1998. Re–Os isotopes in cosmochemistry and high-temperature geochemistry. Annual Review of Earth and Planetary Sciences 26, 423–500. Sun, S.S., McDonough, W.F., 1989. Chemical and isotopic systematics of ocean basalts: implications for mantle composition and processes. In: Saunders, A.D., Norry, M.J. (Eds.), magmatism in the ocean basins. Geological Society, London, Special Publications 42 (1), 313–345. Tang, J., Xu, W.L., Wang, F., Wang, W., Xu, M.J., Zhang, Y.H., 2013. Geochronology and geochemistry of Neoproterozoic magmatism in the Erguna massif, NE China: Petrogenesis and implications for the breakup of the Rodinia supercontinent. Precambrian Research 224, 597–611. Tang, J., Xu, W.L., Wang, F., Zhao, S., Li, Y., 2015. Geochronology, geochemistry, and deformation history of Late Jurassic–Early Cretaceous intrusive rocks in the Erguna Massif, NE China: constraints on the late Mesozoic tectonic evolution of the Mongol-Okhotsk orogenic belt. Tectonophysics 658, 91–110. Tang, Y., Obayashi, M., Niu, F., Grand, S.P., Chen, Y.J., Kawakatsu, H., Tanaka, S., Ning, J., Ni, J.F., 2014. Changbaishan volcanism in northeast China linked to subduction-induced mantle upwelling. Nature Geoscience 7 (6), 470–475. Walker, R.J., Carlson, R.W., Shirey, S.B., Boyd, F.R., 1989. Os, Sr, Nd, and Pb isotope systematic of southern African peridotite xenoliths: implications for the chemical evolution of subcontinental mantle. Geochimica et Cosmochimica Acta 53, 1583–1595. Wang, C., Liang, Y., Xu, W.L., 2015b. On the significance of temperatures derived from major element and ree based two-pyroxene thermometers for mantle xenoliths from the north china craton. Lithos 224–225, 101–113. Wang, C., Liang, Y., Xu, W.L., Dygert, N., 2013. Effect of melt composition on basalt and peridotite interaction: laboratory dissolution experiments with applications to mineral compositional variations in mantle xenoliths from the North China craton. Contributions to Mineralogy and Petrology 166, 1469–1488. Wang, F., Xu, W.L., Gao, F.H., Zhang, H.H., Pei, F.P., Zhao, L., Yang, Y., 2014. Precambrian terrane within the Songnen-Zhangguangcai range massif, NE China: evidence from U-Pb ages of detrital zircons from the Dongfengshan and Tadong groups. Gondwana Research 26, 402–413. Wang, K., Vladimir, P., O'Reilly, S.Y., Griffin, W.L., Pearson, N.J., Victor, K., 2015a. Ancient mantle lithosphere beneath the Khanka massif in the Russian Far East: in situ Re–Os evidence. Terra Nova 27 (4), 277–284. Wells, P.R., 1977. Pyroxene thermometry in simple and complex systems. Contributions to Mineralogy and Petrology 62, 129–139. Wilde, S.A., Wu, F.Y., Zhang, X.Z., 2003. Late pan-African magmatism in northeastern China: SHRIMP U–Pb zircon evidence for igneous ages from the Mashan complex. Precambrian Research 122, 311–327. Windley, B.F., Alexeiev, D., Xiao, W., Kröner, A., Badarch, G., 2007. Tectonic models for accretion of the Central Asian Orogenic Belt. Journal of the Geological Society 164, 31–47.

Wu, F.Y., Walker, R.J., Ren, X., Sun, D., Zhou, X., 2003a. Osmium isotopic constraints on the age of lithospheric mantle beneath northeastern China. Chemical Geology 196 (1–4), 107–129. Wu, F.Y., Jahn, B.M., Wilde, S.A., Lo, C.H., Yui, T.F., Lin, Q., Ge, W.C., Sun, D.Y., 2003b. Highly fractionated I-type granites in NE China (II): isotopic geochemistry and implications for crustal growth in the Phanerozoic. Lithos 67, 191–204. Wu, F.Y., Sun, D.Y., Ge, W.C., Zhang, Y.B., Grant, M.L., Wilde, S.A., Jahn, B.M., 2011. Geochronology of the Phanerozoic granitoids in northeastern China. Journal of Asian Earth Sciences 41, 1–30. Wu, F.Y., Walker, R.J., Yang, Y.H., Yuan, H.L., Yang, J.H., 2006. The chemical–temporal evolution of lithospheric mantle underlying the North China craton. Geochimica et Cosmochimica Acta 70, 5013–5034. Wu, G., Chen, Y.C., Chen, Y.J., Zeng, Q.T., 2012. Zircon U–Pb ages of the metamorphic supracrustal rocks of the Xinghuadukou group and granitic complexes in the Argun massif of the northern Great Hinggan Range, NE China, and their tectonic implications. Journal of Asian Earth Sciences 49, 214–233. Xu, W.L., Pei, F.P., Wang, F., Meng, E., Ji, W.Q., Yang, D.B., Wang, W., 2013. Spatial– temporal relationships of Mesozoic volcanic rocks in NE China: constraints on tectonic overprinting and transformations between multiple tectonic regimes. Journal of Asian Earth Sciences 74, 167–193. Xu, X.S., Griffin, W.L., O'Reilly, S.Y., Pearson, N.J., Geng, H.Y., Zheng, J.P., 2008. Re–Os isotopes of sulfides in mantle xenoliths from eastern China: progressive modification of lithospheric mantle. Lithos 102, 43–64. Xu, Y.G., Zhang, H.H., Qiu, H.N., Ge, W.C., Wu, F.Y., 2012. Oceanic crust components in continental basalts from Shuangliao, Northeast China: derived from the mantle transition zone? Chemical Geology 328, 168–184. Yamamoto, J., Nakai, S., Nishimura, K., Kaneoka, I., Kagi, H., Sato, K., et al., 2009. Intergranular trace elements in mantle xenoliths from russian far east: example for mantle metasomatism by hydrous melt. Island Arc 18 (1), 225–241. Yang, H., Ge, W., Zhao, G., Yu, J., Zhang, Y., 2014. Early permian–late triassic granitic magmatism in the Jiamusi–Khanka Massif, eastern segment of the central asian orogenic belt and its implications. Gondwana Research 27 (4), 1509–1533. Yao, L., Liang, Y., 2015. Closure temperature in cooling bi-mineralic systems: I. Definition and with application to REE-in-two-pyroxene thermometer. Geochimica et Cosmochimica Acta 162, 137–150. Yu, S.Y., Xu, Y.G., Ma, J.L., Zheng, Y.F., Kuang, Y., Hong, L.B., Ge, W.C., Tong, L.X., 2010. Remnants of oceanic lower crust in the subcontinental lithospheric mantle: trace element and O isotope evidence from aluminous garnet pyroxenite xenoliths from Jiaohe, Northeast China. Earth and Planetary Science Letters 297, 413–422. Zhang, M., Suddaby, P., O'Reilly, S.Y., Norman, M., Qiu, J., 2000. Nature of the lithospheric mantle beneath the eastern part of the Central Asian fold belt: mantle xenolith evidence. Tectonophysics 328, 131–156. Zhang, M., Yang, J.H., Sun, J.F., Wu, F.Y., Zhang, M., 2012. Juvenile subcontinental lithospheric mantle beneath the eastern part of the Central Asian Orogenic Belt. Chemical Geology 328, 109–122. Zhang, Y.L., Liu, C.Z., Ge, W.C., Wu, F.Y., Chu, Z.L., 2011. Ancient sub-continental lithospheric mantle (SCLM) beneath the eastern part of the Central Asian Orogenic Belt (CAOB): implications for crust–mantle decoupling. Lithos 126, 233–247. Zhao, S., Xu, W.L., Tang, J., Li, Y., Guo, P., 2016. Timing of formation and tectonic nature of the purportedly Neoproterozoic Jiageda Formation of the Erguna massif, NE China: constraints from field geology and U–Pb geochronology of detrital and magmatic zircons. Precambrian Research 281, 585–601. Zhou, Q., Wu, F.Y., Chu, Z.Y., Ge, W.C., 2010. Isotopic compositions of mantle xenoliths and age of the lithspheric mantle in Yitong, Jilin Province. Acta Petrologica Sinica 26, 1241–1264.