Journal of Petrology Advance Access originally published online on February 2, 2007
Journal of Petrology 2007 48(4):661-692; doi:10.1093/petrology/egl077
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Generation of Palaeocene Adakitic Andesites by Magma Mixing; Yanji Area, NE China
1Key Laboratory of Marginal Sea Geology, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Wushan, Guangzhou, 510640, China
2Institute for Study of the Earth's Interior (ISEI), Okayama University, Yamada 827, Misasa, Tottori 682-0193, Japan
RECEIVED JUNE 16, 2006; ACCEPTED DECEMBER 14, 2006
 |
ABSTRACT |
|---|
TOP
ABSTRACT
INTRODUCTIONPalaeocene (c. 55–58 Ma) adakitic andesites from the Yanji area, NE China, are typically clinopyroxene-bearing sodic andesites containing 60· 9–62· 2% SiO2 and 4· 02–4· 36% MgO, with high Mg-number [100 Mg/(Mg +
Fe) atomic ratio] from 65· 5 to 70· 1. Whole-rock geochemical features include high Cr (128–161 ppm) and Ni (86–117 ppm) concentrations, extremely high Sr (2013–2282 ppm), low Y (10–11 ppm) and heavy rare earth elements (HREE; e.g. Yb = 0· 79–1· 01 ppm), and mid-ocean ridge basalt (MORB)-like Sr–Nd–Pb isotopic compositions [e.g. 87Sr/ 86Sr(i) = 0· 70298–0· 70316, 
Nd(t) = +3· 8 to +6· 3 and 206Pb/ 204Pb = 17· 98 – 18· 06], analogous to high-Mg adakites occurring in modern subduction zones. However, mineralogical evidence from clinopyroxene phenocrysts and microcrystalline plagioclase clearly points to magma mixing during magma evolution. Iron-rich clinopyroxene (augite) cores with low Sr, high Y and heavy REE contents, slightly fractionated REE patterns and large negative Eu anomalies probably crystallized along with low-Ca plagioclase from a lower crustal felsic magma. In contrast, high Mg-number clinopyroxene (diopside and endiopside) mantles and rims have higher Sr and lower HREE and Y concentrations, highly fractionated REE patterns (high La/Yb) and negligible Eu anomalies, similar to those found in adakites from subduction zones. The Yanji adakitic andesites can be interpreted as a mixture between a crust-derived magma having low Mg-number and Sr, and high Y and HREE, and a mantle-derived high Mg-number adakite having high Sr and low Y and HREE concentrations. During storage and/or ascent, the mixed magma experienced further crustal contamination to capture zircons, of a range of ages, from the wall rocks. The absence of coeval arc magmatism and an extensional tectonic regime in the Yanji area and surrounding regions suggest that these Palaeocene adakitic andesites were formed during post-subduction extension that followed the late Cretaceous Izanagi–Farallon ridge subduction. Generation of these adakitic andesites does not require contemporaneous subduction of a young, hot oceanic ridge or delamination of eclogitic lower crust as suggested by previous models. KEY WORDS: magma mixing; adakitic andesites; Palaeocene; NE China
|
INTRODUCTION |
|---|
High Mg-number [100Mg/(Mg +
Fe) atomic ratio] andesites and their intrusive equivalents are a class of intermediate and felsic igneous rocks with Mg-number >45, a feature that distinguishes them from normal andesites that typify arc magmatism (e.g. Kay, 1978
; Kelemen, 1995; Tatsumi, 2006). This characteristic suggests that some high Mg-number andesites might represent near-primary melts in equilibrium with the Earth's mantle (Kushiro, 1969; Hirose, 1997). The petrogenesis of high Mg-number andesite has been extensively studied because of its unique geochemical characteristics and geological significance (e.g. Kay, 1978; Rogers et al., 1985; Yogodzinski et al., 1995; Shimoda et al., 1998; Wang et al., 2002a; Tsuchiya et al., 2005; Tatsumi, 2006). A variety of petrogenetic models have been proposed, including differentiation of high-Mg basalt (Wang et al., 2002a), reaction between slab melt and the overlying mantle wedge (Yogodzinski et al., 1995; Bindeman et al., 2005; Tsuchiya et al., 2005), and melt derivation from a lithospheric mantle metasomatized by silica-rich melt or fluid (e.g. high-Mg andesites and bajaites, Prouteau et al., 2001; Hanyu et al., 2002; Calmus et al., 2003; Martin et al., 2005; Tatsumi, 2006). Most ‘arc andesites’, in contrast, are derived from basalt magmas via complex and variable processes such as fractional crystallization, crustal contamination, anatexis of pre-existing basaltic crust, and magma mixing. Although high Mg-number andesites have rather primitive compositions, they also show many compositional similarities to bulk continental crust, a geochemical reservoir highly enriched in incompatible elements. Indeed, it has been suggested that the petrogenesis of these rocks could have been an important crust-forming process in the early Earth (Kelemen, 1995; Rudnick & Fountain, 1995; Taylor & McLennan, 1995).
Many adakites, which are andesitic to dacitic rocks formed at modern subduction zones, also have high Mg-number as well as high Sr, low Y and heavy rare earth element (HREE) concentrations and highly fractionated REE patterns (e.g. Kay, 1978; Rogers et al., 1985; Saunders et al., 1987; Defant & Drummond, 1990; Kay et al., 1993; Yogodzinski et al., 1995; Stern & Kilian, 1996). Using a database comprising >340 analyses of adakites in modern subduction zones, Martin et al. (2005) summarized and classified modern adakites into two subgroups: high-SiO2 and low-SiO2. They further proposed that the low-SiO2 adakites, like bajaites from Baja California, Adak-type magnesian andesites from the Aleutian and Andean Arcs, and probably also Archaean sanukitoids, owe their high Mg-number, Sr, Cr and Ni, and low Y and HREE concentrations to derivation from incompatible element enriched mantle that was metasomatized by high-pressure slab melts. High-SiO2 adakites, in contrast, are considered to be produced by reactions between high-pressure slab melts and the overlying mantle wedge, and typically range to lower MgO, Mg-number, Cr and Ni (e.g. Kay, 1978; Rogers et al., 1985; Defant & Drummond, 1990; Yogodzinski et al., 1995; Calmus et al., 2003; Martin et al., 2005; Xiong et al., 2006). The andesites of the Yanji area, which form the focus of the present study, are compositionally similar to low-SiO2 adakites except for their higher SiO2 (>60%) (Martin et al., 2005). However, they were produced in a post-subduction tectonic setting with no contemporaneous subduction of young, hot oceanic lithosphere. These rocks are not strictly adakites as originally defined by Defant & Drummond (1990), and so we refer to them here as the Yanji ‘adakitic andesites’.
Because of the near primary compositions of many adakites and high Mg-number andesites, many previous studies have concentrated mainly on their geodynamic significance (slab melt–mantle reaction, melt metasomatism and crustal growth, etc.) rather than on processes that might influence the compositional evolution of the parental magmas. However, ion microprobe analyses of clinopyroxene phenocrysts in ‘primitive’ adakites from the Aleutian Arc suggest that magma mixing occurred during magma ascent (Yogodzinski & Kelemen, 1998). Also, some adakites from the Austral volcanic zone (e.g. at Cerro Pampa and Cook Island) have experienced crustal contamination or assimilation via fractional crystallization (AFC) processes (e.g. Kay et al., 1993; Stern & Kilian, 1996; Motoki et al., 2003). In addition, Macpherson et al. (2006) proposed that some adakites from Mindanao in the Philippines were produced through high-pressure fractional crystallization of arc basalts. Such studies suggest that the magmatic evolution of many adakites and other high Mg-number andesites might be far from simple and this needs to be carefully examined before the petrogenetic or geodynamic significance of these rocks can be fully appreciated.
In this paper, we present geochronological [Ar–Ar dating and sensitive high-resolution ion microprobe (SHRIMP) zircon U–Pb dating results], mineral (major and trace element) and whole-rock geochemical and isotopic (Sr–Nd–Pb and zircon Hf isotope) data form Palaeocene adakitic andesites from the Yanji area of NE China. The study is primarily focused on the processes that control the magmatic compositional evolution. Despite retaining Mg-number as high as 70, these rocks preserve evidence for magma mixing between a mantle-derived melt and an andesitic–dacitic melt produced by melting of the lower crust.
The time of subduction of the Pacific plate and its effect on the tectonic evolution of the East Asian continental margin has been long debated. Another aim of this study is to investigate possible links between the adakitic magmatism and the subduction of the Pacific plate, which began to subduct beneath the East Asian continental margin at 
100 Ma according to palaeomagnetic observations on the Pacific Ocean and its neighbouring Eurasia continents (Engebretson et al., 1985; Northrup et al., 1995).
|
GEOLOGICAL BACKGROUND |
|---|
Northeastern Asia comprises a series of accretionary orogens between the North China Block and Siberian Craton and the Pacific plate (Fig. 1a, Maruyama et al., 1989
; Sengör et al., 1993). The history of tectonic evolution includes: (1) the subduction of the Palaeo-Asian Ocean from the early Palaeozoic to the early Mesozoic to form the Central Asian Orogenic Belt; (2) northwestward subduction of the Pacific plate controlling the evolution of the East Asian continental margin since the late Mesozoic. These multiple-stage plate interactions have led to periodic continental growth and tectono-magmatic events in the Phanerozoic.
|
The Yanji area is located at the border of China, Russia and Korea, about 100 km west of the Japan Sea. It was considered part of the orogenic collage between the North China Block in the south and the Jiamusi–Khanka Massifs in the NE (Fig. 1b; Peng & Su, 1997
; Xiao et al., 2003; Jia et al., 2004; Zhang et al., 2004). To the south, there occur Archaean (2· 5 Ga) greenschist- and amphibolite-facies metamorphosed granitic rocks (JBGMR, 1988).
Palaeozoic strata are widely distributed in the region, and these have undergone variable degrees of metamorphism and deformation. Previous studies suggested that these Palaeozoic rocks might form tectonic mélanges related to the accretion of the NE Asian continent (e.g. Shao & Tang, 1995). Phanerozoic granitoids are widely exposed over an area of more than 20 000 km2 and occupy about 70% of the region. Available zircon U–Pb data indicated that emplacement of these Phanerozoic granitoids extended from late Palaeozoic (285 Ma) to early Cretaceous (116 Ma) times (Zhang et al., 2004). Zhang et al. (2004) related the multi-stage granitoid emplacement events to the evolution of the eastern segment of the Central Asian Orogenic Belt. The Mesozoic geological evolution was characterized by the eruption of voluminous lavas, mainly in the early Cretaceous (118–106 Ma, our unpublished Ar–Ar ages), granitoid emplacement (Zhang et al., 2004), and deposition of terrestrial sediments as a result of post-orogenic lithospheric extension (Wang et al., 2002b; Fan et al., 2003).
Regional evolution since the late Cretaceous was probably related to subduction of the Farallon–Izanagi and Kula–Pacific ridges toward the NE Asian continental margin (Engebretson et al., 1985; Faure & Natal’in, 1992; Northrup et al., 1995; Taira, 2001; Wang et al., 2002b). Subduction-related magmatism is observed in the southwestern and northern Japan Arcs (Kinoshita, 1995; Maeda & Kagami, 1996; Taira, 2001). However, coeval arc magmatism is scarce in the Yanji area and in the surrounding regions, the only possible example being the sporadic extrusion of the Palaeocene Yanji adakitic andesites (Fig. 1c). In contrast to the rare magmatism in the Yanji area, voluminous Palaeocene tholeiitic to calc-alkaline basalts erupted in fault-rift basins around the Bohai Bay area (Fig. 1a) and along the Tan-Lu fault in response to lithospheric extension (Chen et al., 1992; Liu, R. X., et al., 1992; Liu, C. Q., et al., 1994; Liu, J. Q., et al., 2001; Xu et al., 1995), suggesting a similar extensional regime during the extrusion of the adakitic andesites in the Yanji area.
Rifting of the northeastern margin of the Asian continent started at 22 Ma and was followed by development of the Japan Sea basin, which separated from the Asian continent between 22 and 15 Ma (Nakamura et al., 1990; Yamaji, 1990; Lee et al., 1999; Taira, 2001). Back-arc extension continued into the Quaternary, associated with the eruption of Cenozoic–Quaternary basaltic lavas in NE China (e.g. Basu et al., 1991; Liu, R. X., et al., 1992; Liu, C. Q., et al., 1994; Liu, J. Q., et al., 2001; Zhang et al., 1995).
|
FIELD AND PETROGRAPHIC CHARACTERISTICS |
|---|
The Palaeocene adakitic andesites of the Yanji area occur as low hills up to 0· 5 km2 in area and form a WNW–ESE-trending chain
25 km long. The volcanic units show well-developed columnar joints, dip very shallowly (<10°), and overlie Upper Cretaceous (K2, Longjing Formation) sediments. Rare crustal xenoliths occur in the volcanic rocks and include granodiorite, gneiss, metagabbro and amphibolite. Two types of adakitic andesites can be identified in the Yanji area: one contains only clinopyroxene phenocrysts (clinopyroxene-andesite); the other (e.g. the outcrop near Helong town, Fig. 1c) is an amphibole-andesite with predominantly amphibole phenocrysts and subordinate clinopyroxene phenocrysts.
The studied samples were collected from the largest outcrop of clinopyroxene-andesites, located about 10 km NW of Longjing town (Fig. 1c). These rocks are fresh, with 10% clinopyroxene phenocrysts up to 10 mm in size. The groundmass consists mainly of fine-grained (<0· 5 mm) plagioclase, clinopyroxene and opaque oxides. Clinopyroxene is typically euhedral to subhedral (Fig. 2). In addition to the weakly zoned clinopyroxene (as shown in Fig. 2a), many clinopyroxene phenocrysts show complex zonation with plagiolcase inclusions (Fig. 2b). In some clinopyroxene phenocrysts, orthopyroxene inclusions can also be found (Fig. 3).
|
|
|
ANALYTICAL TECHNIQUES |
|---|
Major element compositions of minerals were determined using a JXA-8100 electron microprobe at Guangzhou Institute of Geochemistry, Chinese Academy of Sciences (CAS). An accelerating voltage of 15 kV, a specimen current of 3· 0 x 10–8 A, and a beam size of 2 µm were employed. The analytical errors are generally less than 2%.
Trace element compositions of minerals (clinopyroxene and zircon) were analyzed by secondary ion mass spectrometry (SIMS) at the Pheasant Memorial Laboratory of the Institute for Study of the Earth's Interior (ISEI), Okayama University, Japan, using a Cameca ims-5f system. Analytical methods were similar to those used by Nakamura & Kushiro (1998). A primary negative oxygen beam of 10 nA intensity was used to sputter samples. The spot sizes were about 10 µm. Secondary ions were collected using a 4500 V accelerating voltage and measured by ion counting. To reduce the mass interference effect, –45 V of sample offset voltage was applied. The calculation factors for clinopyroxene and zircon were obtained by repeatedly analyzing standard clinopyroxenes from KLB-1 and SAG30 mantle peridotites and standard Sri Lanka zircon, respectively. Analytical errors (RSD%) were between 5 and 10% (1
).
Argon step-heating analyses on one whole-rock sample (04YJ-3) were performed at the Institute of Geology and Geophysics, CAS on a GV5400 mass spectrometer operating in static mode. Samples were firstly crushed to 40–60 mesh, and fresh matrix, free of phenocrysts, was handpicked under a binocular microscope. The detailed analytical technique and age correction method were reported by Wang et al. (2004). The reference age for the GA-1550 biotite standard is 98· 79 ± 0· 96 Ma (Renne et al., 1998). Ca and K correction factors were calculated from the CaF2 and K-glass monitors: (36Ar/37Ar)Ca = 2· 6088 x 10–4 ± 1· 1418 x 10–5, (39Ar/37Ar)Ca = 7· 236 x 10–4 ± 2· 814 x 10–5, (40Ar/39Ar)K= 2· 648 x 10–2 ± 2· 254 x 10–4. Three other samples for Ar–Ar dating were analyzed using a GV5400Ar mass spectrometer attached to a MIR10 CO2 laser at Guangzhou Institute of Geochemistry, CAS. The laser beam is around 3 mm in diameter. A detailed description of the analytical technique was reported by Qiu (2006). The Ca and K correction factors are: (38Ar/36Ar)a = 0· 1869, (38Ar/37Ar)Ca = 0, (36Ar/37Ar)Ca = 2· 69 x 10–4, (39Ar/37Ar)Ca = 7· 09 x 10–4, (40Ar/39Ar)K = 0· 00165. The measured Ar–Ar results were processed using ArArCALC ver2.2c software (Koppers, 2002). The biotite standard ZBH2506 yields an Ar–Ar plateau age of 132· 5 ± 0· 20 Ma.
Zircon U–Pb dating was conducted using a sensitive high-resolution ion microprobe (SHRIMP II) at the Beijing SHRIMP Laboratory, Institute of Geology, Chinese Academy of Geological Science. Zircons were concentrated from fresh rock samples following normal separation methods, and were then handpicked and mounted together with standard TEM (417 Ma) onto double-sided adhesive tape and enclosed in epoxy resin discs. The discs were polished, so as to effectively section the zircons nearly in half, and then cleaned, photographed and gold coated. Cathodoluminescence (CL) images were made on zircons prior to U–Pb analyses to reveal the internal textures and to guide the SHRIMP analyses. Analytical spot size averaged 30 µm and each spot was rastered over 120 µm for 3 min to remove common Pb on the surface of the zircon. More detailed analytical procedures have been described by Compston et al. (1984, 1992), Williams (1998) and Liu et al. (2003). Decay constants used are those recommended by Steiger & Jager (1977), and common Pb correction used the 204Pb methods of Compston et al. (1984). The SHRIMP analytical data presented here are the mean values of five consecutive scans for each analytical spot. Errors are 1, whereas the weighted mean ages of samples are quoted at the 2 level. The 206Pb/238U age is adopted in this study, as it is considered the most reliable measurement for concordant Phanerozoic zircons (Compston et al., 1992).
Whole-rock major and trace element analyses were performed at Guangzhou Institute of Geochemistry, CAS, by X-ray fluorescence (XRF) spectrometry (Rigaku RIX-3000) and inductively coupled plasma mass spectrometry (ICP-MS), respectively. For major elements, the analytical errors are less than 2%. The analytical procedure for trace element determination was described by Liu et al. (1996). Precision for REE and high field strength elements (HFSE) is estimated to be 5% and about 10% for other elements from repeatedly analyzed US Geological Survey (USGS) standards BHVO-1 (basalt) and W-2 (diabase). The measured values for the standards were listed by Xu (2002).
Sr, Nd and Pb isotopic analyses were performed at Institute of Geology and Geophysics, CAS, on a MAT 262 mass spectrometer. Rock chips (< 20 mesh) were leached in purified 0· 1 N HCl for 24 h at room temperature to avoid the influence of surface alteration or weathering, especially for Sr and Pb isotopes. Powders were dissolved in HF–HClO4 for Sr and Nd isotope analyses, and in HF–HNO3 for Pb isotope analysis. Dissolution was at 200°C for 1 week. Detailed analytical and correction techniques were described by Chen et al. (2002). The procedure blanks were <200 pg for Sr, <50 pg for Pb, and 30 pg for Nd. Pb standard NBS 981 was used to determine thermal fractionation, and the measured isotopic ratios of samples were corrected with a value of 0· 11% per atomic mass unit. 86Sr/88Sr ratios were normalized to 86Sr/88Sr = 0· 1194 and 146Nd/144Nd ratios were normalized to 146Nd/144Nd = 0· 7219. The Ames standard yielded 143Nd/144Nd = 0· 512149 ± 3 (n = 98) and NBS 987 gave 87Sr/86Sr = 0· 710244 ± 4 (n = 100) for the last period of 6 months. Analytical errors for Sr, Nd and Pb isotopic ratios are given as 2. The 87Rb/86Sr, 147Sm/144Nd, U/Pb and Th/Pb ratios were calculated using the Rb, Sr, Sm, Nd, U, Th and Pb abundances obtained by ICP-MS. The initial Sr, Nd and Pb isotopic ratios were corrected using the Ar–Ar age of 55 Ma.
Hf isotope ratios were analyzed on zircon with a Neptune laser ablation multi-collector ICP-MS system attached to a Geolas 193 nm CQ ArF excimer laser system at the Institute of Geology and Geophysics, CAS. Analyses reported here were carried out with a laser beam diameter of 63 µm. The isobaric interference of 176Yb on 176Hf was corrected using a laboratory-established formula: ßYb = 0· 8725ßHf. A detailed description of the zircon Hf isotope analytical technique has been given by Griffin et al. (2002) and Xu et al. (2004). Twenty-five analyses of standard zircon 95100 gave a 176Hf/177Hf ratio of 0· 282665 ± 13 (n = 25). The 176Hf/177Hf ratios for sample zircon grains were corrected using the ratio obtained from standard 95100. The 176Lu/177Hf and 176Hf/177Hf ratios for chondrites are 0· 0332 and 0· 282772, respectively, and 
is 1· 865 x 10–11 year–1.
|
MINERAL CHEMISTRY |
|---|
Clinopyroxene
Clinopyroxene is the only phenocryst in the studied adakitic andesites. It is pale green with an euhedral to subhedral shape (Fig. 3a–f). Representative electron microprobe and ion microprobe analyses for clinopyroxene are listed in Table 1. They span a large range in SiO2 (50· 6–55· 5%) and MgO (11· 85–18· 9%) contents and in Mg-number (66· 2–94· 0). Most clinopyroxene crystals are diopside and endiopside (Fig. 4) with relatively high Mg-number (> 80); however, the bright cores seen in the back-scattered electron (BSE) images (Fig. 3c) are Fe-rich augite.
|
|
Geochemical data from a wide range of rock types show that clinopyroxenes from peridotite xenoliths, ophiolites, basalts, high-Mg andesites, gabbros and granulites can be roughly discriminated using a Mg-number vs Na2O diagram (Fig. 5a). Compared with clinopyroxene from basaltic rocks, clinopyroxenes from peridotites have a high Mg-number. Clinopyroxenes from boninites are depleted in Na2O, and those from high-Mg andesites have relatively low Na2O and high Mg-number extending to the field for clinopyroxene in peridotites. Clinopyroxenes from the Yanji adakitic andesites have Mg-number and Na2O contents comparable with those in high-Mg andesites from modern subduction zones (Fig. 5b), except for the Fe-rich augite cores shown in Fig. 3c, which have high Na2O (0· 79%) and very low Mg-number (
66) and plot well away from the field for high-Mg andesites.
|
Most clinopyroxene phenocrysts in the Yanji adakitic andesites exhibit compositional zonation (Fig. 3a–f). Four types of clinopyroxene phenocrysts can be identified based on BSE imagery (Fig. 3): weakly, normally, reversely and complex zoned phenocrysts.
Weakly zoned clinopyroxene
This type of clinopyroxene is euhedral (Figs 2a and 3a). Rims have relatively low Mg-number (79–84) and are augite with an Fs component >10, whereas the core is composed of high-Mg number (Mg-number = 87–89) diopside and endiopside. Relative to the high Mg-number cores, rims have high incompatible element concentrations, and both rims and cores show similar primitive mantle-normalized trace element patterns, characterized by significant Nb, Zr and Ti depletion relative to REE and by very small Sr–Eu anomalies (Fig. 6a). Such major and trace element variations from core to rim are typical of normal magma differentiation during crystal overgrowth.
|
Normally zoned clinopyroxene
This type of clinopyroxene is also euhedral, with dark cores and light rims (Fig. 3b). The cores are high in Cr2O3 (> 0· 5%) and NiO (0· 17–0· 23%) and have high Mg-number (>93), and are classified as Cr-diopside (Table 1). Rims have slightly lower MgO and SiO2. Both cores and rims show similar trace element patterns to the weakly zoned clinopyroxene, with higher incompatible element concentrations in the rim (Fig. 6b). Such major and trace element variations possibly reflect overgrowth around a xenocrystic clinopyroxene.
Reversely zoned clinopyroxene
This type of clinopyroxene is subhedral. The bright core shown in Fig. 3c has very low MgO (11· 8%) and SiO2 (50· 7%), and high FeO (12%), whereas the dark rim has higher MgO (17%) and SiO2 (53· 5%), and lower FeO (4· 5%). Relative to the dark rim, the bright core has much higher REE, Zr and Y but much lower Sr concentrations. In primitive mantle-normalized trace element diagrams, apart from the Nb, Zr and Ti depletion, the Fe-rich cores also show strong negative Sr and Eu anomalies with Eu/Eu* about 0· 30 (Fig. 6c and Table 1). This contrasts with the high-MgO rims, which have negligible negative Sr and Eu anomalies (Fig. 6c and Table 1). The core shown in Fig. 6d also shows a progressive outwards increase in REE content and decrease in Sr/Y, Nd/YbCN and Eu/Eu* ratios (i.e. richer in Fe, higher REE concentrations but lower Sr/Y, Eu/Eu* and Nd/YbCN). The core and mantle also exhibit very different trace element patterns (Fig. 6d), as observed in Fig. 6c. Such reverse zonations in clinopyroxene major and trace element compositions suggest magma mixing during clinopyroxene overgrowth.
Complex zoned clinopyroxene
Some clinopyroxene phenocrysts show complex zonation patterns as illustrated in Fig. 3e and f. From the core to the rim, these phenocrysts show a stage of reverse zonation following by oscillatory zonation. Like that in Fig. 6c, the Fe-rich core in Fig. 6e shows similar trace element patterns with slight REE fractionation and negative Sr and Eu anomalies. The higher Mg-number clinopyroxene corresponds to lower incompatible element concentrations, suggesting a normal magma differentiation trend. Nevertheless, there also occur Cr-diopside fragments with Cr2O3 up to 1· 03% and NiO at 0· 28% (spot 2 in Fig. 3f, Table 1), as observed in spots 4–6 in Fig. 3b. It is possible that the Cr-diopside fragments with similar trace element features to other high Mg-number clinopyroxene were xenocrystic from mantle peridotite that had been metasomatized by slab melt.
Clinopyroxene phenocrysts from the Yanji adakitic andesites show positive correlations between Sr/Y, Eu/Eu*, Nd/YbCN and Mg-number (Fig. 7a–c), and between Sr and Nd/Yb (Fig. 7d). Such variations do not reflect normal magmatic differentiation but point to mixing between a high Mg-number melt having high Eu/Eu*, Sr/Y and Nd/YbCN and a low Mg-number magma characterized by lower Eu/Eu*, Sr/Y and Nd/YbCN (e.g. Yogodzinski & Kelemen, 1998). Two-component mixing calculations show that most analyzed spots plot exactly on or adjacent to the mixing curves with the proportion of the Fe-rich core <10%. Analyses that plot away from the mixing lines were of clinopyroxenes that possibly crystallized during magmatic evolution (Fig. 7).
|
Orthopyroxene
Orthopyroxene phenocrysts are absent in the Yanji adakitic andesites. However, rare anhedral orthopyroxene crystals (0· 2–0· 5 mm) are enclosed in clinopyroxene phenocrysts (e.g. Fig. 3f). These have high SiO2 (56· 7–57· 4%) and Mg-number (86· 7–90· 4), spanning an enstatite component range of 85· 7–88· 2, and are classified as bronzites in an Fs–En–Wo discrimination diagram (Table 2; Fig. 4).
|
Plagioclase
Plagioclase phenocrysts are also absent in the Yanji adakitic andesites. However, plagioclase forms the microcrystalline groundmass as well as anhedral inclusions in clinopyroxene phenocrysts (Fig. 3b, d, e and f). Groundmass plagioclase is labradorite, with an anorthite (An) content between 50· 1 and 56· 7, whereas plagioclase inclusions hosted in clinopyroxene phenocrysts are andesine, with lower An contents (31· 2–43· 1) (Table 3). Such compositional differences in plagioclase also contrast with normal magmatic differentiation trends, but like the reverse zonation patterns in the clinopyroxene phenocrysts are consistent with magma mixing during crystal growth.
|
|
GEOCHRONOLOGY AND ZIRCON REE CHEMISTRY |
|---|
Ar–Ar dating results
Four samples (04YJ-3, 05YJ-1, 05YJ-20 and 05YJ-64) were selected for Ar–Ar dating to determine the eruption age of the Yanji adakitic andesites (see Fig. 1c). Samples 04YJ-3 and 05YJ-1 were collected from the same locality. 04YJ-3 was sampled from the upper part of the volcanic sequence and 05YJ-1 was sampled from the lower part. The whole-rock Ar–Ar analytical results are listed in Tables 4 and 5, and are shown in Fig. 8. Sample 04YJ-3 shows a flat age spectrum. During 11 of 14 consecutive steps (from 850°C to 1400°C), over 95% 39Ar was released, defining a plateau age of 54· 9 ± 0· 3 Ma (2
), consistent with an isochron age of 55· 5 ± 0· 4 Ma with MSWD = 4· 7 (n = 11) (Fig. 8). The initial 40Ar/36Ar (287· 3 ± 3· 7) is consistent with that of atmosphere, precluding significant excess argon contamination (Fig. 8b). Thus, the plateau age of 54· 9 ± 0· 3 Ma can be interpreted as the minimum eruption age of the Yanji adakitic andesites. Sample (05YJ-1), from the base of the volcanic sequence, yields a plateau age of 56· 3 ± 0· 3 Ma (2), consistent with an isochron age of 55· 9 ± 0· 7 Ma with MSWD = 0· 81 (n = 13) and initial 40Ar/36Ar at 301· 9 ± 11· 4 (Fig. 8c and d). These two ages are almost identical within the analytical errors.
|
|
|
The other two samples (05YJ-20 and 05YJ-64) also yield Palaeocene ages (Fig. 1). Sample 05YJ-20 yields a plateau age of 57· 2 ± 0· 5 Ma and an isochron age of 56· 3 ± 1· 8 Ma (MSWD = 0· 48, n = 16) (Fig. 8e and f). Sample 05YJ-64 gives a plateau age of 58· 5 ± 0· 6 Ma and an isochron age of 58· 5 ± 2· 3 Ma (MSWD = 0· 08, n = 14) (Fig. 8g and h). The range of initial 40Ar/36Ar for both samples is well within the range of the atmosphere. Therefore, the two plateau ages are interpreted to represent the eruption age of the adakitic andesites.
In summary, the Ar–Ar dating results for the Yanji adakitic andesites give an eruption age of 55–58 Ma, corresponding to the Palaeocene epoch. These ages are consistent with the field observations that Upper Cretaceous sediments are entrained in, and interbedded with, the volcanic layers.
Zircon U–Pb dating results and REE chemistry
Zircon occurs as a very rare accessory in the Yanji adakitic andesites. The crystals are mostly clear, subhedral to euhedral stubby prisms, with length/width ratios about 1–3. They contrast with other volcanic zircons, which are usually slender prisms (Fig. 9a and b). Some CL images show core–mantle overgrowth relationships and oscillatory zoning textures, indicating a magmatic origin.
|
Twenty analyses of zircon span an apparent 206Pb/238U age range of 170–338 Ma (Table 6), much older than the Ar–Ar age of 55–58 Ma. When plotted on a 207Pb/238U vs 206Pb/238U diagram (Fig. 9c), they indicate a mean age of 190 ± 4 Ma with MSWD of 2· 1 (n = 14). Except for one high-U and one high-Th zircon, the zircons have reasonably high Th/U ratios around 0· 6 (Fig. 9d), also suggesting a magmatic origin. The mean age of 190 ± 4 Ma and the 170–338 Ma age range are comparable with the age range for granitoid intrusions in the region (e.g. Zhang et al., 2004
).
|
Trace element compositions of zircons from sample 04YJ-6 are listed in Table 7. The zircons of different ages show similar REE patterns with LREE depletion and HREE enrichment and significant positive Ce anomalies and negative Eu anomalies (Fig. 9e and f). Zircon crystals with an apparent 238U/206Pb age between 183 and 195 Ma have a Ce/Ce* range of 12· 3–48· 6 and a Eu/Eu* range of 0· 19–0· 27, except for one crystal (spot 13), which has a higher Eu/Eu* of 0· 53. Other crystals with an apparent 238U/206Pb age either older than 200 Ma or younger than 180 Ma have a Ce/Ce* range of 11· 4–44· 6 and a Eu/Eu* range of 0· 05–0· 37.
|
The zircons span a large range of Y concentrations from 313 to 1777 ppm, but a narrow range of Hf concentrations from 2688 to 4632 ppm. Such features are very different from those of zircon populations from ultra-high pressure eclogites, which generally have low Y (<100 ppm) and high Hf (>9000 ppm) concentrations as well as smaller negative Eu anomalies (e.g. Rubatto, 2002
). The high Y and low Hf concentrations, and significant negative Eu anomalies, suggest that the zircons in the Yanji adakitic andesites crystallized in the presence of plagioclase but not garnet. |
GEOCHEMISTRY |
|---|
Whole-rock major and trace elements
The studied Yanji adakitic andesites span a narrow range of SiO2 from 60· 9 to 62· 2% and MgO from 4· 02 to 4· 36%, with a Mg-number range of 65· 5–70· 1 (Table 8). They have much higher MgO (and Mg-number) at given SiO2 than experimental melts derived from metabasites (e.g. Rapp & Watson, 1995
; Rapp et al., 1999; Xiong et al., 2005), but are similar to bajaites and high-Mg andesites and adakites in modern arcs such as the Japan and Aleutian Arcs and the Austral volcanic zone in South America (Fig. 10a; Kay et al., 1993; Yogodzinski et al., 1995; Stern & Kilian, 1996; Shimoda et al., 1998; Calmus et al., 2003; Tatsumi et al., 2003; Tatsumi, 2006). Similarly, they also show rather high Cr (128–161 ppm) and Ni (86–117 ppm) concentrations, similar to those of bajaites from Baja California and high-Mg andesites and adakites in the Aleutian and Andean Arcs (Fig. 10b and c). A remarkable feature of the Yanji adakitic andesites is their extremely high Sr contents, from 2013 to 2282 ppm (Table 8), analogous to those of bajaites from Baja California and some Adak-type magnesian andesites from the Aleutian and Andean Arcs (Rogers et al., 1985; Saunders et al., 1987; Kay et al., 1993; Yogodzinski et al., 1995; Stern & Kilian, 1996; Calmus et al., 2003), but much higher than the Sr range of high-Mg andesites or diorites from the adjacent Japan Arc (Fig. 10d; Shimoda et al., 1998; Kamei et al., 2004; Tatsumi, 2006). In common with adakites from modern subduction zones (Defant & Drummond, 1990; Smithies, 2000; Martin et al., 2005), the Yanji adakitic andesites also have low Y (10· 4–11· 0 ppm) and HREE concentrations (e.g. Yb is only 3–4 times chondritic values). Such features are reflected in a Y vs Sr/Y plot (Fig. 11), in which the Yanji adakitic andesites have Sr/Y values comparable with those of high Mg-number adakites and bajaites but much higher than those of the Japan Arc high-Mg andesites (e.g. Kamei et al., 2004; Tatsumi, 2006).
|
|
|
Like adakites worldwide (e.g. Defant & Drummond, 1990
; Yogozinshi et al., 1995; Martin et al., 2005), the Yanji adakitic andesites show highly fractionated chondrite-normalized REE patterns (Fig. 12a), having a La/Yb range of 44–56 and negligible Eu anomalies (Eu/Eu* = 0· 93–1· 03). Their primitive mantle-normalized trace element patterns are characterized by large ion lithophile element (LILE; Ba, Th, K, Pb and Sr) and LREE enrichment and Nb–Ta, P, Y and HREE depletion (Fig. 12b).
|
Whole-rock Sr–Nd–Pb isotope compositions
The measured and age-corrected Sr–Nd–Pb isotopic ratios of the studied Yanji adakitic andesites are given in Table 9. In general, these rocks have unradiogenic Sr [87Sr/86Sr(i) = 0· 70298–0· 70316] and radiogenic Nd [
Nd(t) = +3· 8 to +6· 3) and Pb isotopic compositions (206Pb/204Pbi = 17· 98–18· 06; 207Pb/204Pbi = 15· 49–15· 55, 208Pb/204Pbi = 37· 73–37· 89). Compared with modern adakites and bajaites in subduction zones (e.g. Rogers et al., 1985; Kay et al., 1993; Yogodzinski et al., 1995; Stern & Kilian, 1996), the Yanji adakitic andesites have Sr–Nd isotopic compositions within the field of modern analogues but have lower Pb isotope ratios and plot within the array of global mid-ocean ridge basalt (MORB) (Fig. 13).
|
|
Compared with the early Cretaceous basalts in the region, the Yanji adakitic andesites have less radiogenic Sr and Pb, and more radiogenic Nd isotopic compositions (Fig. 13, our unpublished data). They have low TDM, ranging from 284 to 442 Ma, much older than the eruptive age but overlapping the SHRIMP U–Pb age range for zircon in these rocks.
Zircon Hf isotope data
Hf isotopic ratios of zircons are listed in Table 10. The zircon crystals have positive Hf(t) values ranging from +2· 0 to 12· 3, with a mean of +8· 4 (Fig. 14a). Depleted mantle model Hf ages [TDM(Hf)] range from 369 to 771 Ma, with a peak age at 400–500 Ma (Fig. 14b), slightly older than the whole-rock TDM(Nd).
|
|
The U–Pb zircon data resemble those of magmatic zircons from intermediate–felsic intrusions, whereas the zircon Hf isotope data suggest that the zircons crystallized from magmas that were derived from juvenile crustal materials.
|
PETROGENESIS |
|---|
In the following discussion, we focus on the constraints that mineral compositions place on magmatic processes and on the possible magma components involved in the petrogenesis of the Palaeocene Yanji adakitic andesites.
Mineralogical evidence for magma mixing
The complex compositional zonation of the clinopyroxene phenocrysts in the Yanji adakitic andesites suggest that these minerals experienced a complex overgrowth history. Most of the clinopyroxene phenocrysts experienced conditions that caused some degree of reverse zonation. Such a phenomena suggests that these clinopyroxene phenocrysts were neither products of a normal magmatic fractionation process nor were they produced by assimilation and fractional crystallization of a mafic magma, as both processes should result in normal zonation in the clinopyroxene. Reverse zonation can be formed as a result of a change in P–T conditions, magma mixing or by melt–mantle interaction, or by any combination of these processes.
P–T change?
High pressure will retard plagioclase fractionation, and in a hydrous andesitic magma, plagioclase will not crystallize when P is >20 kbar; in this case, the fractionating assemblage will be pyroxene + garnet + amphibole + quartz (Green, 1982). When T is >1000°C and P > 5 kbar, the fractionating phases include only clinopyroxene and orthopyroxene. If the common reverse zonation in clinopyroxene and the high An content in the microcrystalline groundmass plagioclase are a result of a P–T change, then a rapid increase in temperature of the magma is required (e.g. Green, 1982). Possible ways to increase the magma temperature are by addition of another high-T magma, by hybridization with hot mantle or by rapid decompressive crystallization.
Melt–mantle interaction?
Melt–mantle interaction might be an alternative way of creating the compositional zonations observed in clinopyroxene. Such a model has been widely applied to interpret the high Mg-number, and high Cr and Ni contents, in most of the modern adakites occurring at convergent plate margins and in experimental slab melts hybridized by mantle peridotite (Kay, 1978; Yogodzinski et al., 1995; Rapp et al., 1999; Tsuchiya et al., 2005; Guo et al., 2006; Xiong et al., 2006). Decomposition of olivine and possibly orthopyroxene in mantle peridotite lead to an increase of MgO, Cr and Ni in the hybridized slab melts, and thus can result in reverse zonations in crystallizing minerals. If such a petrogenetic process was the case for the Yanji adakitic andesites, the high-Fe clinopyroxene cores would reflect crystallization from the slab melt (before assimilation of peridotite), whereas the high Mg-number clinopyroxene rims would reflect interaction between that melt and the surrounding mantle peridotite.
In general, high-pressure slab melts (e.g. adakites), derived from eclogitic protoliths with residual pyroxene, garnet, amphibole and rutile are characterized by high Sr, low Y and HREE (Defant & Drummond, 1990; Smithies, 2000; Martin et al., 2005). During the ascent of these melts, clinopyroxene crystallizes without significant coexisting plagioclase. However, the Fe-rich clinopyroxene cores in the Yanji adakitic andesites show large negative Eu anomalies combined with low Sr and Nd/YbCN (Figs 6 and 7). These compositional features suggest crystallization accompanied by plagioclase—a paragenesis very unlike that of an ascending adakitic slab melt. In addition, if the high Mg-number clinopyroxene rims formed by melt–peridotite interaction, they should have lower incompatible element (e.g. Sr) concentrations than those of the Fe-rich cores, as the mantle wedge should have much lower incompatible element concentrations than a slab melt. Experimental results show that mantle hybridization does not notably increase the incompatible element compositions of hybridized melts (Rapp et al., 1999).
Accordingly, although a melt–mantle interaction model could explain the observed major element variations in the clinopyroxene phenocrysts in the Yanji adakitic andesites, it is highly unlikely that it could account for their trace element characteristics. Furthermore, it appears unlikely that the Fe-rich clinopyroxene cores crystallized from an adakitic slab melt.
Magma mixing?
Our ion microprobe data for clinopyroxene are similar to those presented by Yogodzinski & Kelemen (1998), who showed that clinopyroxene in some ‘primitive’ adakites from the Aleutian Arc had low Mg-number cores with lower Sr and Nd/Yb than the high Mg-number rims. To explain these features, they proposed mixing between a primitive adakite with high Mg-number, Sr and Nd/Yb, and an evolved arc-basalt or basaltic andesite with relatively low Mg-number, Sr and Nd/Yb (Yogodzinski & Kelemen, 1998).
High Mg-number clinopyroxene phenocrysts in the primitive Aleutian adakites contain inclusions of hypersthene, which probably crystallized from an evolved arc basalt or basaltic andesite. Clinopyroxene phenocrysts in the Yanji adakitic andesites contain inclusions of bronzite and low-An plagioclase. Bronzite is a common early-crystallizing phase in high Mg-number sanukitoid magmas interpreted to be melts of a peridotite previously metasomatized via addition of slab melt (Shirey & Hanson, 1984; Tatsumi, 2006). In these cases, both the residual bronzite and the high Mg-number clinopyroxene are likely to be fractionating phases in the high Mg-number primitive magma. It is possible that the clinopyroxene phenocrysts and bronzite inclusions in the Yanji adakitic andesites can be interpreted in the same way.
In contrast, the low An content of the plagioclase inclusions in the Yanji clinopyroxene phenocrysts implies a more felsic magmatic source, possibly more consistent with the high-Fe compositions of some clinopyroxene cores, as both Fe-rich clinopyroxene and plagioclase with an An content range of 35–41 are common fractionating phases in andesitic–dacitic melts. Experimental results show that low-An plagioclase is unstable at high temperatures (e.g. within the hot mantle wedge), and so the low-An plagioclase inclusions hosted in clinopyroxene phenocrysts argue further against an origin from a felsic slab melt—mantled by high Mg-number clinopyroxene (e.g. Green, 1982). We suggest that the low-An plagioclase and the Fe-rich clinopyroxene (cores) coexisted as early fractionating phases in a felsic magma—but not a slab melt.
Further, compared with the Fe-rich clinopyroxene cores, the high Mg-number clinopyroxenes generally have higher Sr, Sr/Y and Nd/YbCN, and negligible Eu anomalies, and so are unlikely to have coexisted with plagioclase during crystal growth. This is consistent with petrographic observations that no plagioclase phenocrysts have been found.
We envisage that the clinopyroxene and plagioclase compositional relationships and zoning patterns developed in a high Mg-number adakitic magma that ascended from the mantle and mixed with a felsic lower crustal melt that was already crystallizing Fe-rich clinopyroxene and low-An plagioclase.
Constraints on the possible magma mixing components
The above discussion, based on mineralogical evidence from clinopyroxene phenocrysts and microcrystalline plagioclase, strongly suggests that magma mixing influenced the evolution of the Yanji adakitic andesites. In the following section, we discuss the possible magma components involved in this mixing process.
The felsic magma
This magma might be either a differentiate of a mantle-derived basalt or a crust-derived melt. As illustrated in Fig. 7, the felsic magmatic component is characterized by low MgO and Sr but high Y and HREE contents, very different from adakites occurring in modern subduction zones (e.g. Defant & Drummond, 1990; Martin et al., 2005). In the Yanji area and in the adjacent regions, the lack of basaltic magmas contemporaneous with the Palaeocene adakitic andesites suggests that the felsic magma component was derived from the local crust (e.g. Liu et al., 2001).
Using experimental REE partition coefficients for clinopyroxene in basaltic–andesitic, andesitic–dacitic and low-Si and high-Si rhyolitic melts, respectively (e.g. Luhr & Carmichael, 1980; Dostal et al., 1983; Nash & Crecraft, 1985; Sisson, 1991), we have recalculated the possible REE contents of the melts in equilibrium with the Fe-rich clinopyroxene cores and high Mg-number clinopyroxene, respectively. From these calculations, it becomes obvious that mixing between a rhyolitic melt (whether low-Si or high-Si in composition) coexisting with the Fe-rich clinopyroxene cores and a mantle-derived high Mg-number magma coexisting with the high Mg-number clinopyroxene could not have produced the whole-rock REE contents of the Yanji adakitic andesites. However, the whole-rock REE contents in these rocks can be successfully modelled by mixing between an andesitic–dacitic melt coexisting with the Fe-rich clinopyroxene cores and a mantle-derived high Mg-number magma (Fig. 15; see also Table 11).
|
|
The high Y concentrations and weakly fractionated REE patterns in the Fe-rich clinopyroxene cores point to protoliths with little or no residual garnet. In contrast, Sr is compatible in plagioclase, and therefore low Sr in Fe-rich clinopyroxene might be attributed either to low Sr concentration in the magma in equilibrium with residual plagioclase or to plagioclase fractionation during clinopyroxene crystallization. Therefore, we can summarize the characteristics of the felsic magma component as follows: (1) it was andesitic–dacitic in composition with high Y and a significant negative Eu anomaly; (2) the melt was in equilibrium with plagioclase (either as a fractionating phase in the melt or as a residual phase during melting) but not with garnet, arguing against an eclogitic protolith.
The high Mg-number melt
The other mixing component must have low SiO2, Y and HREE, and high MgO (or Mg-number), Sr and Sr/Y ratios, analogous to a high Mg-number adakite (e.g. Yogodzinski et al., 1995; Martin et al. 2005). Mass balance considerations suggest that such high Mg-number adakites could not be derived from melting of an eclogite-facies basaltic lower crust (Petford & Atherton, 1996; Yogodzinski & Kelemen, 1998). As discussed above, slab melt–mantle interaction cannot account for the trace element characteristics of the clinopyroxene phenocrysts. Therefore, we suggest that the high Mg-number melt was mantle-derived (Yogodzinski & Kelemen, 1998; Calmus et al., 2003; Martin et al., 2005).
The high Mg-number clinopyroxenes have lower Na2O and higher Mg-number than clinopyroxene phenocrysts in basalts (Fig. 4b). They also have higher Sr/Y and Nd/YbCN ratios than clinopyroxene from basalts in the Aleutian Arc (e.g. Yogodzinski & Kelemen, 1998). Accordingly, we suggest that the mantle-derived melt from which the clinopyroxene crystallized and that eventually mixed with the lower crustal andesitic–dacitic magma was not an arc basalt. Instead, the general positive correlation between Sr (not shown), Sr/Y, Nd/YbCN and Mg-number (Fig. 7) suggests that the mantle-derived melt had extremely high Sr, high MgO (Mg-number) and LREE/HREE ratio, but low Y and HREE concentrations, features typical of bajaites, Aleutian Adak-type magnesian andesites and other low-Si (high Mg-number) adakites (Rogers et al., 1985; Kay et al., 1993; Yogodzinski et al., 1995; Stern & Kilian, 1996; Calmus et al., 2003; Martin et al., 2005).
Thus, despite the strong evidence for magma mixing with a crustal melt, both the parental mantle-derived melt, and the post-mixing derivative Yanji adakitic andesite were high Mg-number adakitic magmas, as suggested by mixing calculation results that show that <10% felsic melt was involved in the mixed magma (Fig. 7). This is also reflected in the whole-rock isotopic data, which show MORB-like Sr, Nd and Pb isotopic compositions, suggesting that the mantle source for the high Mg-number melt was characterized by unradiogenic Sr and radiogenic Nd and Pb isotopic compositions, and that the addition of the felsic component had only a small effect on these compositions. Considering the younger bulk TDM(Nd) (284–442 Ma), the mantle source of these magmas was probably influenced by either contemporaneous or recent oceanic slab subduction.
Crustal contamination recorded by zircon
As recorded by the weak and normal zoning in clinopyroxene phenocrysts, the mixed magma probably accumulated in a magma chamber where it underwent differentiation before eruption. During storage or ascent through the crust, the mixed magma experienced further crustal contamination, as reflected in the zircon data, as follows (1) All zircons have much older apparent 206Pb/238U ages (170–338 Ma) than the whole-rock Ar–Ar age (55–58 Ma), suggesting they are inherited from the wall-rocks instead of crystallizing in any kind of melt involved in the magma mixing process. Zircon U–Pb dating results show that the emplacement ages of granitoid intrusions in the Yanji region spanned a range from 116 to 285 Ma (Zhang et al., 2004). (2) With their reasonably high Th/U ratio (0· 6), the zircons are likely to be magmatic in origin, but the stubby morphology is typical of plutonic rocks rather than of volcanic rocks. (3) The REE patterns of the zircons of different ages are LREE-depleted and HREE-enriched with significant negative Eu anomalies, arguing against inheritance from the eclogitic protolith that produced the adakitic melt. (4) These zircons have high Hf(190 Ma) (+2· 0 to +12· 3, with an average of +8· 4) and TDM(Hf) (369–771 Ma), suggesting that there was a considerable volume of juvenile material within the crust sampled by the high Mg-number magmas—as observed in the Central Asian Orogenic Belt (our unpublished data). A similar case for crustal contamination, inferred by the presence of old inherited zircons, was proposed for the Cerro Pampa adakites in Argentine Patagonia (Motoki et al., 2003). Thus, the clinopyroxene and plagioclase data and the inherited zircons appear to identify two independent sources of felsic material within the Yanji adakitic andesites: an early andesitic–dacitic lower crustal melt, and later, higher level contamination by wall-rock granitoids. Despite the MORB-like Sr–Nd–Pb isotopic compositions, the later crustal contamination might also contribute to the 87Sr/86Sr(i) increase from 0· 70298 to 0· 70342 and 206Pb/204Pb(i) increase from 17· 83 to 18· 07, and Nd(t) decrease from +6· 93 to +3· 42 in the clinopyroxene-andesites across the Yanji area (our unpublished data). Other high Mg-number adakites, such as those at Cerro Pampa and Cook Island in the Austral Volcanic Zone in the Andean Arc (Kay et al., 1993; Stern & Kilian, 1996), similarly underwent moderate amounts of crustal contamination. This possibility should be carefully assessed before such rocks are interpreted as direct partial melts of either the subducting slab or slab melt-metasomatized mantle. Indeed, where contamination is more extensive or where the contaminating crust is isotopically more evolved, clear evidence for an adakitic parental magma might be lost.
To summarize the above discussion, we can depict the magmatic evolution of the Yanji adakitic andesites in the following stages (Fig. 16). (1) Generation of two kinds of magmas derived from slab-metasomatized mantle and crust, respectively. The mantle-derived magma crystallized high Mg-number clinopyroxene (possibly some Cr-diopside fragments were captured from the melting mantle source) whereas Fe-rich clinopyroxene cores and low-An plagioclase were derived from the crustal-sourced felsic melt. (2) Mixing between the two types of magmas in a crustal magma chamber led to formation of the reverse zonation in clinopyroxene. (3) Magma storage and differentiation generated the weak to normal zonation in clinopyroxene with increasing incompatible element concentrations at the outermost rim. (4) During magma storage or ascent, the mixed magma was further contaminated by plutonic wall rocks, leading to entrainment of zircons of a range of different ages, which possibly led to the Sr–Nd–Pb isotopic variations in these rocks.
|
|
POSSIBLE LINK WITH LATE CRETACEOUS–PALAEOCENE PACIFIC SUBDUCTION |
|---|
Although the Palaeocene Yanji adakitic andesites were generated by magma mixing, the primary high Mg-number adakitic magma was derived from a LILE- and LREE-enriched mantle source metasomatized by slab melt, as envisaged for low-Si adakites defined by Martin et al. (2005
). How and when the slab-modified mantle was formed remains unclear, as the coeval arc magmas that might place robust constraints on the enrichment event are lacking. As discussed above, the Yanji area experienced a complex evolution from the late Palaeozoic to Cenozoic, and there are three possible subduction events that might have generated the slab-modified mantle reservoir from which the primary high Mg-number adakites were derived: (1) subduction of the Palaeo-Asian ocean during the late Palaeozoic to early Mesozoic (Robison et al., 1999; Jia et al., 2004); (2) the late Cretaceous (95–65 Ma) Izanagi–Farallon ridge subduction associated with the formation of a Late Cretaceous accretionary prism and arc magmatism in southwestern Japan [e.g. the Shikoku accretion mélange that was subducted at around 70 Ma (Taira, 2001, and references therein)]; (3) the Kula–Pacific ridge subduction during Palaeocene times (55 Ma), associated with the formation of the Shimanto and western Kokkaido accretionary prisms and the contemporaneous emplacement of voluminous granitic intrusions in SW and northern Japan (Maeda & Kagami, 1996; Taira, 2001). The first possibility can be ruled out because the Palaeocene adakitic andesites have less radiogenic Pb and Sr, and more radiogenic Nd isotopic compositions than the early Cretaceous basalts (Fig. 13). Considering the effect of crustal contamination on Palaeocene adakitic magma evolution, the mantle source for the primary high Mg-number adakites must have had much lower Sr and Pb and higher Nd isotope ratios than the source for the early Cretaceous basalts. A LILE- and LREE-enriched mantle source metasomatized by slab melt will have higher Rb/Sr and lower Sm/Nd and U/Pb ratios than chondrite or Bulk Silicate Earth, and will develop more radiogenic Sr, and less radiogenic Nd and Pb isotopic compositions with time. This is the opposite of what we observe in the Yanji area. It is therefore unlikely that the Palaeocene primary high Mg-number adakitic magma originated from the same mantle source as that producing the early Cretaceous basalts, even if the Palaeo-Asian oceanic slab-modified mantle survived into the Palaeocene.
The Palaeocene Kula–Pacific ridge subduction toward the Hidaka magmatic zone of central Hokkaido in northern Japan could have influenced the formation of the Yanji adakitic andesites (Maeda & Kagami, 1996). However, as well as the absence of coeval arc magmas in the Yanji area, field observations of Upper Cretaceous and Palaeocene sediments in this area, and in the surrounding regions, suggest that they are terrestrial deposits with coal layers, distinct from the accretionary prism deposits along the Japan Arc, which comprise mainly oceanic-plate mélanges (e.g. turbidite units, pillow lavas, radiolarian cherts, serpentinites, and high-temperature metamorphic belts; Taira, 2001, and references therein). Additionally, previous studies of the Palaeocene geology inferred an extensional regime for the Yanji area, and the surrounding regions; for example, the opening of Bohai Bay coinciding with eruption of asthenosphere-derived tholeiitic basalts in the regions along the Tanlu fault, and the rifting of the Songliao basin (Chen et al., 1992; Liu, R. X., et al., 1992; Liu, J. Q., et al., 2001; Xu et al., 1995; Wang et al., 2002b). All these points tend to negate a direct link between the Palaeocene Yanji adakitic generation and the contemporaneous Kula–Pacific ridge subduction towards northern Japan.
Rather, we prefer to relate the generation of the Palaeocene Yanji adakitic andesites to Izanagi–Farallon ridge subduction during the late Cretaceous. According to this model, the mantle source for the primary high Mg-number adakites was formed during slab subduction, but the mantle melting event postdated the ridge subduction (Fig. 17a), analogous to the post-subduction production of magmas in Baja California (e.g. Rogers et al., 1985; Saunders et al., 1987). The evidence for an extensional tectonic regime in the Yanji area, and in surrounding regions, during the Palaeocene Epoch might suggest that lithospheric extension was the cause of mantle melting, production of the parental high Mg-number adakite and melting of the lower crust to form felsic melts with which the adakite magmas mixed (Fig. 17b). Such an extensional regime is also different from the ridge–trench collision setting in the Austral volcanic zone of Andean Arc, where contemporaneous slab melting occurred to form the Neogene Cerro Pampa adakites (Kay et al., 1993).
|
Despite the proposed genetic link with the late Cretaceous (
70 Ma) Farallon–Izanagi ridge subduction that led to the LILE and LREE enrichment of the mantle source (e.g. Taira, 2001), generation of the Palaeocene Yanji adakitic andesites does not necessitate either contemporaneous subduction of a young and hot oceanic ridge and associated slab melt metasomatism, or delamination of eclogitic lower crust as suggested by previous hypotheses. |
CONCLUSIONS |
|---|
Palaeocene (55–58 Ma) adakitic andesites from the Yanji area, NE China, are characterized by high MgO (Mg-number), Cr and Ni, extremely high Sr, low Y and HREE concentrations, and MORB-like Sr–Nd–Pb isotopic compositions. Mineralogical evidence from clinopyroxene phenocrysts and microcrystalline plagioclase indicates an origin through mixing between a crust-derived felsic magma having high Y and HREE concentrations and low Sr, Sr/Y and Nd/YbCN ratios, and a mantle-derived high Mg-number adakitic melt. During magma storage or ascent to surface, the mixed magma experienced crustal contamination to capture zircons from the wall-rock granitoid plutons, which possibly led to the whole-rock Sr–Nd–Pb isotopic variations. Our results support previous studies that suggest that adakitic magmas with high MgO (or Mg-number) are not necessarily near-primary melts in equilibrium with the mantle, and magmatic process including magma mixing and crustal contamination should be investigated in detail before the wider geodynamic significance of such rocks is evaluated.
The absence of coeval arc magmatism and an associated accretionary prism assemblage, together with an extensional tectonic regime in the Yanji area during the Palaeocene Epoch, suggest that adakite generation postdated the late Cretaceous Fallaron–Izanagi ridge subduction, and is thus considered to have been produced under post-subduction extension. Generation of the Yanji adakitic andesites does not necessitate contemporaneous subduction of a young and hot oceanic ridge, slab melt metasomatism or delamination of eclogitic lower crust as suggested by previous hypotheses.
|
ACKNOWLEDGEMENTS |
|---|
The authors wish to thank R. H. Smithies for providing constructive comments and suggestions and language help on the early version. H. F. Zhang and X. L. Xiong are thanked for discussion. F. Wang, F. K. Chen, H. N. Qiu, Y. Liu and X. F. Gao are thanked for technical assistance in mineralogical and geochemical analyses, and X. Y. Li for help with fieldwork. Thorough and constructive reviews by S. M. Kay, B. Leeman, M. Streck and an anonymous reviewer helped to improve the manuscript. Editorial handling was by John Gamble. This study was financially supported by National Natural Science Foundation of China (40373018 and 40334043), and Cooperative Research program COE-21 by the Government of Japan.
* Corresponding author. Telephone: + 86-20-85290280. Fax: + 86-20-85290130. E-mail: guofengt{at}263.net
|
REFERENCES |
|---|
Basu AR, Wang JW, Huang WK, Xie GH, Tatsumoto M. (1991) Major element, REE, and Pb, Nd and Sr isotopic geochemistry of Cenozoic volcanic rocks of eastern China: implications for their origin from suboceanic-type mantle reservoirs. Earth and Planetary Science Letters 105:149–169.[CrossRef][Web of Science]
Bindeman IN, Eiler JM, Yogodzinski GM, Tatsumi Y, Stern CR, Grove TL, Portnyagin M, Hoernle K, Danyushevsky LV. (2005) Oxygen isotope evidence for slab melting in modern and ancient subduction zones. Earth and Planetary Science Letters 235:480–496.[CrossRef][Web of Science]
Calmus T, Aguillón-Robles A, Maury RC, Bellon H, Benoit M, Cotton J, Bourgois J, Michaud F. (2003) Spatial and temporal evolution of basalts and magnesian andesites (‘bajaites’) from Baja California, Mexico: the role of slab melts. Lithos 66:77–105.[CrossRef][Web of Science]
Chen FK, Satir M, Ji J, Zhang D. (2002) Nd–Sr–Pb isotopes of Tengchong Cenozoic volcanic rocks from western Yunnan, China: evidence for an enriched mantle source. Journal of Asian Earth Sciences 21:39–45.[CrossRef][Web of Science]
Chen WJ, Li DM, Li Q, Shen SW, Liang HD, Zhou XH, Liu RX, Wang X, Liu XT, Zheng JW. (1992) The age and geochemistry of basin basalts at Xialiaohe rift. In Liu R (Ed.). The Age and Geochemistry of Cenozoic Volcanic Rocks in China(Seismological Press, Beijing) pp. 44–80 (in Chinese).
Compston W, Williams IS, Mayer C. (1984) U–Pb geochronology of zircons from Lunar Breccia 73217 using a Sensitive High Resolution Ion Microprobe. Proceedings of the XIV Lunar and Planetary Science Conference. Journal of Geophysical Research 89:Supplement, B525–B534.
Compston W, Williams IS, Kirschvink JL, Zhang ZC, Ma GG. (1992) Zircon U–Pb ages for the Early Cambrian time scale. Journal of the Geological Society, London 149:171–184.
Defant MJ and Drummond MS. (1990) Derivation of some modern arc magmas by melting of young subducted lithosphere. Nature 347:662–665.[CrossRef]
Dostal J, Dupuy C, Carron JP, Le Guen de Kerneizon M, Maury RC. (1983) Partition coefficients of trace elements: application to volcanic rocks of St Vincent, West Indies. Geochimica et Cosmochimica Acta 47:525–533.[CrossRef][Web of Science]
Engebretson DC, Cox A, Gordon RG. (1985) Relative motions between oceanic and continental plates in the Pacific basins. Geological Society of America, Special Papers 206:1–59.
Fan WM, Guo F, Wang YJ, 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.[CrossRef][Web of Science]
Faure M and Natal’in B. (1992) The geodynamic evolution of the eastern Eurasian margin in Mesozoic times. Tectonophysics 208:397–411.[CrossRef][Web of Science]
Green TH. (1982) Anatexis of mafic crust and high pressure crystallization of andesites. In Thorpe RS (Ed.). Andesites: Orogenic Andesites and Related Rocks(John Wiley, Chichester) pp. 465–487.
Griffin WL, Wang X, Jackson SE, Pearson NJ, O'Reilly SY, Xu XS, Zhou XM. (2002) Zircon chemistry and magma mixing, SE China: in-situ analysis of Hf isotopes, Tonglu and Pingtan igneous complexes. Lithos 61:237–269.[CrossRef][Web of Science]
Gvirtzman Z and Nur A. (1999) The formation of Mount Etna as the consequence of slab rollback. Nature 401:782–785.[CrossRef]
Guo F, Fan WM, Li CW. (2006) Geochemistry of late Mesozoic adakites from the Sulu belt, eastern China: magma genesis and implications for crustal recycling beneath continental collisional orogens. Geological Magazine 143:1–13.
Hanyu T, Tatsumi Y, Nakai A. (2002) A contribution of slab melts to the formation of high-Mg andesite magmas: Hf isotopic evidence from SW Japan. Geophysical Research Letters 29:81–84.
Hart SR. (1984) A large-scale isotope anomaly in the southern hemisphere mantle. Nature 309:753–757.[CrossRef]
Hirose K. (1997) Melting experiments on lherzolite KLB-1 under hydrous conditions and generation of high-magnesian andesitic melts. Geology 25:42–44.
Jia DC, Hu RZ, Lu Y, Qiu XL. (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.
Jilin Bureau of Geology and Mineral Resources (JBGMR). (1988) Regional Geology of Jilin Province(Geological Publishing House, Beijing) pp. 698 (in Chinese with English summary).
Kamei A, Owada M, Nagao T, Shiraki K. (2004) High-Mg diorites derived from sanukitic HMA magmas, Kyushu Island, southwest Japan arc: evidence from clinopyroxene and whole rock compositions. Lithos 75:359–371.[CrossRef][Web of Science]
Kay RW. (1978) Aleutian magnesian andesites-melts from subducted Pacific ocean crust. Journal of Volcanology and Geothermal Research 4:117–132.[CrossRef][Web of Science]
Kay SM, Ramos CA, Marquez M. (1993) Evidence in Cerro Pampa volcanic rocks for slab-melting prior to ridge–trench collision in southern South America. Journal of Geology 101:703–714.[Web of Science]
Kelemen PB. (1995) Genesis of high-Mg-number andesites and the continental crust. Contributions to Mineralogy and Petrology 120:1–19.[Web of Science]
Kinoshita O. (1995) Migration of igneous activities related to ridge subduction in Southwest Japan and the East Asian continental margin from the Mesozoic to the Paleogene. Tectonophysics 245:25–35.[CrossRef][Web of Science]
Koppers AAP. (2002) ArArCALC—software for 40Ar/39Ar age calculations. Computers and Geosciences 28:605–619.
Kushiro I. (1969) The system forsterite–diopside–silica with and without water at high pressures. American Journal of Sciences 267:269–294.
Lee YS, Ishikawa N, Kim WK. (1999) Paleomagnetism of Tertiary rocks on the Korea Peninsula: tectonic implications for the opening of the East Sea (Sea of Japan). Tectonophysics 304:131–149.[CrossRef][Web of Science]
Li CW, Guo F, Fan WM, Gao XF. (2007) Ar–Ar geochronology on Late Mesozoic volcanic rocks from the Yanji area, NE China and tectonic implications. Science in China (Series D) (in press).
Liu CQ, Masuda A, Xie GH. (1994) Major- and trace-element compositions of Cenozoic basalts in eastern China: petrogenesis and mantle source. Chemical Geology 114:19–42.[CrossRef][Web of Science]
Liu DY, Jian P, Zhang Q, Zhang FQ, Shi YR, Shi GH, Zhang LQ, Tao H. (2003) SHRIMP dating of adakites in the Tulingkai ophiolite, Inner Mongolia: evidence for the Early Paleozoic subduction. Acta Geologica Sinica 77:317–327 (in Chinese with English abstract).
Liu JQ, Han JT, Fyfe WS. (2001) Cenozoic episodic volcanism and continental rifting in northeast China and possible link to Japan Sea development as revealed from K–Ar geochronology. Tectonophysics 339:385–401.[CrossRef][Web of Science]
Liu RX, Chen WJ, Sun JZ, Li DM. (1992) The K–Ar age and tectonic environment of Cenozoic volcanic rocks in China. In Liu R (Ed.). The Age and Geochemistry of Cenozoic Volcanic Rock in China(Seismology Press, Beijing) pp. 1–43 (in Chinese).
Liu Y, Liu HC, Li XH. (1996) Simultaneous and precise determination of 40 trace element elements using ICP-MS. Geochimica 25:552–558 (in Chinese with English abstract).
Luhr JF and Carmichael ISE. (1980) The Colima volcanic complex, Mexico. I: Post-caldera andesites from Volcan Colima. Contributions to Mineralogy and Petrology 71:343–372.[CrossRef][Web of Science]
Macpherson CG, Dreher ST, Thirlwall MF. (2006) Adakites without slab melting: high pressure differentiation of island arc magma, Mindanao, the Philippines. Earth and Planetary Science Letters 243:581–593.
Maeda J and Kagami H. (1996) Interaction of a spreading ridge and an accretionary prism: implications from MORB magmatism in the Hidaka magmatic zone, Hokkaido, Japan. Geology 24:31–34.
Martin H, Smithies RH, Rapp R, Moyen J-F, Champion D. (2005) An overview of adakite, tonalite–trondhjemite–granodiorite (TTG), and sanukitoid: relationships and some implications for crustal evolution. Lithos 79:1–24.[CrossRef][Web of Science]
Maruyama S, Liou JG, Seno T. (1989) Mesozoic and Cenozoic evolution of Asia. In Ben-Avraham Z (Ed.). The Evolution of the Pacific Ocean Margins. Oxford Monography of Geology and Geophysics 8: pp. 75–99.
Motoki A, Orihashi Y, Cario FD, Hirata D, Haller MJ, Ramos VA, Kawano H, Watanabe Y, Schilling M, Iwano H, Anma R. (2003) U–Pb dating for single grain zircon using laser ablation ICP mass spectrometer and fission track ages for zircon grains of back-arc adakitic bodies, Argentine Patagonia. The Fourth International Symposium of Isotope Geology, Abstracts 219–220.
Nash WP and Crecraft HR. (1985) Partition coefficients for trace elements in silicic magmas. Geochimica et Cosmochimica Acta 49:2309–2322.[CrossRef][Web of Science]
Nakamura E and Kushiro I. (1998) Trace element diffusion in jadeite and diopside melts at high pressures and its geochemical implication. Geochimica et Cosmochimica Acta 62:3151–3160.[CrossRef][Web of Science]
Nakamura E, McCulloch MT, Campbell IH. (1990) Chemical geodynamics in the back-arc region of Japan based on the trace element and Sr–Nd isotopic compositions. Tectonophysics 174:207–233.[CrossRef][Web of Science]
Northrup CJ, Royden LH, Burchfiel BC. (1995) Motion of the Pacific plate relative to Euroasia and its potential relation to Cenozoic extrusion along the eastern margin of Eurasia. Geology 23:719–722.
Peng YJ and Su YZ. (1997) Geotectonic characteristics of central Jilin Province. Memoirs of Shenyang Institute of Geology and Mineral Resources, Chinese Academy of Geological Sciences 5–6:335–376 (in Chinese).
Petford N and Atherton MP. (1996) Na-rich partial melts from newly underplated basaltic crust: the Cordillera Blanca Batholith, Peru. Journal of Petrology 37:1491–1521.
Prouteau G, Scaillet B, Pichavant M, Maury RC. (2001) Evidence for mantle metasomatism by hydrous silicic melts derived from subducted oceanic crust. Nature 410:197–200.[CrossRef]
Qiu HN. (2006) Construction and development of new Ar–Ar laboratories in China: insight from VG-5400 Ar–Ar laboratory in Guangzhou Institute of Geochemistry, Chinese Academy of Sciences. Geochimica 35:133–140 (in Chinese with English abstract).
Rapp RP and Watson EB. (1995) Dehydration melting of metabasalt at 8–32 kbar: implications for continental growth and crust–mantle recycling. Journal of Petrology 36:891–931.
Rapp RP, Shimizu N, Norman MD, Applegate GS. (1999) Reaction between slab-derived melts and peridotite in the mantle wedge: experimental constraints at 3· 8 GPa. Chemical Geology 160:335–356.[CrossRef][Web of Science]
Renne PR, Swisher CC, Deino AL, Owens TL, DePaolo DJ, Karner DB. (1998) Intercalibration of standards, absolute ages and uncertainties in 40Ar/39Ar dating. Chemical Geology 145:117–152.[CrossRef][Web of Science]
Robinson PT, Zhou MF, Hu XF, Reynolds P, Bai WJ, Yang J. (1999) Geochemical constraints on the origin of the Hegenshan ophiolite, Inner Mongolia, China. Journal of Asian Earth Sciences 17:423–442.[CrossRef][Web of Science]
Rogers G, Saunders AD, Terrell DJ, Verma SP, Marriner GF. (1985) Geochemistry of Holocene volcanic rocks associated with ridge subduction in Baja California, Mexico. Nature 315:389–392.[CrossRef]
Rubatto D. (2002) Zircon trace element geochemistry: partitioning with garnet and the link between U–Pb ages and metamorphism. Chemical Geology 184:123–138.[CrossRef][Web of Science]
Rudnick RL and Fountain DM. (1995) Nature and composition of the continental crust: a lower crustal perspective. Reviews of Geophysics 33:267–309.[CrossRef][Web of Science]
Saunders AD, Rogers G, Marriner GF, Terrell DJ, Verma SP. (1987) Geochemistry of Cenozoic volcanic rocks, Baja California, Mexico: implications for the petrogenesis of post-subduction magmas. Journal of Volcanology and Geothermal Research 32:223–245.[CrossRef][Web of Science]
Sengör AMC, Natal’in BA, Burtman VS. (1993) Evolution of the Altaid tectonic collage and Paleozoic crustal growth in Eurasia. Nature 364:299–307.[CrossRef]
Shao JA and Tang KD. (1995) Terranes in Northeast China and Evolution of Northeast Asia Continental Margin(Seismological Press, Beijing) pp. 64–70 (in Chinese with English summary).
Shimoda G, Tatsumi Y, Nohda S, Ishizaka K, Jahn BM. (1998) Setouchi high-Mg andesites revisited: geochemical evidence for melting of subducting sediments. Earth and Planetary Science Letters 160:479–492.[CrossRef][Web of Science]
Shirey SB and Hanson GN. (1984) Mantle derived Archaean monzodiorites and trachyandesites. Nature 310:222–224.[CrossRef]
Sisson TW. (1991) Pyroxene–high-silica rhyolite trace element partition coefficients measured by ion microprobe. Geochimica et Cosmochimica Acta 55:1575–1585.[CrossRef][Web of Science]
Smithies RH. (2000) The Archaean tonalite–trondhjemite–granodiorite (TTG) series is not analogue of Cenozoic adakites. Earth and Planetary Science Letters 182:115–125.[CrossRef][Web of Science]
Steiger RH and Jager E. (1977) Subcommission on geochronology: convention on the use of decay constants in geo- and cosmo-chronology. Earth and Planetary Science Letters 36:359–362.[CrossRef][Web of Science]
Stern CR and Kilian R. (1996) Role of the subducted slab, mantle wedge and continental crust in the generation of adakites from the Austral volcanic zone. Contributions to Mineralogy and Petrology 123:263–281.[CrossRef][Web of Science]
Sun SS and McDonough WF. (1989) Chemical and isotopic systematics of oceanic basalts: implication for mantle composition and processes. In Saunders AD and Norry MJ (Eds.). Magmatism in the Ocean Basins. Geological Society London: Special Publications 42: pp. 313–345.
Taira A. (2001) Tectonic evolution of the Japanese Island Arc system. Annual Review of Earth and Planetary Sciences 29:109–134.[CrossRef][Web of Science]
Tatsumi Y. (2006) High-Mg andesites in the Setouchi volcanic belt, southwestern Japan: analogy to Archean magmatism and continental crust formation? Annual Review of Earth and Planetary Sciences 34:467–499.[CrossRef][Web of Science]
Tatsumi Y, Shukuno H, Sato K, Shibata T, Yoshikawa M. (2003) The petrology and geochemistry of high-Mg andesites at the western tip of the Setouchi volcanic belt, SW Japan. Journal of Petrology 44:1561–1578.
Taylor SR and McLennan SM. (1995) The geochemical evolution of the continental crust. Reviews in Geophysics 33:241–265.
Tsuchiya N, Suzuki S, Kimura JI, Kagami H. (2005) Evidence for slab melt/mantle reaction: petrogenesis of Early Cretaceous and Eocene high-Mg andesites from the Kitakami Mountains, Japan. Lithos 79:179–206.[CrossRef][Web of Science]
Wang F, Zhu RX, Li Q, He HY, Luo QH, Lu XX, Sang HQ, Wang YL. (2004) A differential uplifting of Qinling orogeny belt evidences from 40Ar/39Ar thermochronology of granites. Earth Science Frontier 11:445–459 (in Chinese with English abstract).
Wang KL, Chung SL, Chen CH, Chen CH. (2002a) Geochemical constraints on the petrogenesis of high-Mg basaltic andesites from the Northern Taiwan Volcanic Zone. Chemical Geology 182:513–528.[CrossRef][Web of Science]
Wang PJ, Liu WZ, Wang SX, Song WH. (2002b) 40Ar/39Ar and K/Ar dating on the volcanic rocks in the Songliao basin, NE China: constraints on stratigraphy and basin dynamics. International Journal of Earth Sciences 91:331–340.[CrossRef][Web of Science]
Williams IS. (1998) U–Th–Pb geochronology by ion microprobe. In McKibben MA, Shanks WC, Ridley WI (Eds.). Applications of Microanalytical Techniques to Understanding Mineralizing Processes. Reviews in Economic Geology 7: pp. 1–35.
Xiao WJ, Windley B, Hao J, Zhai MG. (2003) Accretion leading to collision and the Permian Solonker suture, Inner Mongolia, China: termination of the Central Asian Orogenic Belt. Tectonics 22:61069 doi:10.1029/2002TC001484.
Xiong XL, Adam J, Green TH. (2005) Rutile stability and rutile/melt HFSE partitioning during partial melting of hydrous basalt: implications for TTG genesis. Chemical Geology 218:339–359.[CrossRef][Web of Science]
Xiong XL, Xia B, Xu JF, Niu HC, Xiao WS. (2006) Na depletion in modern adakites via melt/rock reaction within the sub-arc mantle. Chemical Geology 229:273–292.
Xu P, Wu FY, Xie LW, Yang YH. (2004) Hf isotopic compositions of the standard zircons for U–Pb dating. Chinese Science Bulletin 49:1642–1648.[CrossRef][Web of Science]
Xu YG. (2002) Evidence for crustal components in the mantle and constraints on crustal recycling mechanisms: pyroxenite xenoliths from Hannuoba, North China. Chemical Geology 182:301–322.[CrossRef][Web of Science]
Xu YG, Fan WM, Lin G. (1995) Lithosphere–asthenosphere interaction: a comparative study on Cenozoic and Mesozoic basalts around Bohai Area. Geotectonica et Metallogenia 19:1–13.
Yamaji A. (1990) Rapid intra-arc rifting in Miocene Northeast Japan. Tectonics 9:365–378.[Web of Science]
Yogodzinski GM and Kelemen PB. (1998) Slab melting in the Aleutians: implications of an ion probe study of clinopyroxene in primitive adakite and basalt. Earth and Planetary Science Letters 158:53–65.[CrossRef][Web of Science]
Yogodzinski GM, Kay RW, Volynets ON, Koloskov AV, Kay SM. (1995) Magnesian andesite in the western Aleutian Komandorsky region: implications for slab melting and processes in the mantle wedge. Geological Society of America Bulletin 107:505–519.
Zhang M, Suddaby P, Thompson RN, Thirlwall MF, Menzies MA. (1995) Potassic volcanic rocks in NE China: geochemical constraints on mantle source and magma genesis. Journal of Petrology 36:1275–1303.
Zhang YB, Wu FY, Wilde SA, Zhai MG, Lu XP, Sun DY. (2004) Zircon U–Pb ages and tectonic implications of ‘Early Paleozoic’ granitoids at Yanbian, Jilin Province, northeast China. Island Arc 13:484–505.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  L. V. Danyushevsky, T. J. Falloon, A. J. Crawford, S. A. Tetroeva, R. L. Leslie, and A. Verbeeten High-Mg adakites from Kadavu Island Group, Fiji, southwest Pacific: Evidence for the mantle origin of adakite parental melts Geology, June 1, 2008; 36(6): 499 - 502. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||