doi:10.1016/j.gca.2004.04.015
Multiple sources for the origin of granites: Geochemical and Nd/Sr isotopic evidence from
the Gudaoling granite and its mafic enclaves, northeast China
JIN-HUIYANG,1,* FU-YUANWU,1SUN-LINCHUNG,2SIMONA. WILDE,3and MEI-FEICHU2
1Institute of Geology and Geophysics, Chinese Academy of Sciences, P.O. Box 9825, Beijing 100029, China 2Department of Geosciences, National Taiwan University, Taipei, Taiwan
3Department of Applied Geology, Curtin University of Technology, P.O. Box U1987, Perth, Western Australia 6845, Australia
(Received January 5, 2004; accepted in revised form April 7, 2004)
Abstract—Geochemical and Sr- and Nd-isotopic data have been determined for mafic to intermediate microgranular enclaves and host granitoids from the Early Cretaceous Gudaoling batholith in the Liaodong Peninsula, NE China. The rocks include monzogranite, porphyric granodiorite and quartz diorite. Monzo-granites have relatively high87Rb/86Sr ratios (0.672– 0.853), low initial87Sr/86Sr ratios (0.7052– 0.7086) and Nd(t) values (⫺18.5 to ⫺20.9) indicating that they were mainly derived from a newly underplated crustal source with a short crustal residence time. Quartz diorites have high initial87Sr/86Sr ratios (0.7118 – 0.7120) and negativeNd(t) values (⫺13.2 to ⫺18.1) coupled with high Al2O3and MgO contents, indicating they were derived from enriched lithospheric mantle with contributions of radiogenic Sr from plagioclase-rich metagreywackes or meta-igneous rocks, i.e., ancient lower crust. Two groups of enclaves with igneous textures and abundant acicular apatites are distinguished: dioritic enclaves and biotite monzonitic enclaves. Dioritic enclaves have low Al2O3(13.5–16.4 wt%) and high MgO (Mg#⫽ ⬃72.3) concentrations, low initial 87Sr/86Sr ratios (0.7058 – 0.7073) and negative
Nd(t) values (⬃⫺7.2), and are enriched in LILEs and LREEs and depleted in HFSEs, suggesting they were derived from an enriched lithospheric mantle source. Biotite monzonitic enclaves have Sr and Nd isotopic compositions similar to the monzogranites, indicating they were crystal cumulates of the parental magmas of these monzogranites. Granodiorites have transitional geochem-istry and Nd- and Sr-isotopic compositions, intermediate between the monzogranites, quartz diorites and the enclaves.
Geochemical and Sr- and Nd-isotopic compositions rule-out simple crystal-liquid fractionation or restite unmixing as the major genetic link between enclaves and host rocks. Instead, magma mixing of mafic mantle-derived and juvenile crustal-derived magmas, coupled with crystal fractionation and assimilation of ancient lower crust, is compatible with the data. This example shows that at least some calc-alkaline granitoids are not produced by pure intracrustal melting, but formed through a complex, multi-stage hybridization process, involving mantle- and crustal-derived magmas and several concomitant magmatic processes (crystal fractionation, crustal assimilation and crustal anatexis). Copyright © 2004 Elsevier Ltd
Keywords—Mafic microgranular enclave, Origin for granites, Magma mixing, Sr-Nd isotopes, China
1. INTRODUCTION
Since the introduction of the terms I- and S-type granites, the source of granitic magma has become one of the most hotly debated subjects in granite petrology. Originally, I-type gran-ites were believed to have formed by the partial melting of older igneous rocks at intracrustal levels, and that composi-tional variation was the result of restite composition or crystal fractionation (e.g., Chappell et al., 1987; Chappell and Ste-phens, 1988; Chappell and White, 1992). Although crustal recycling seems to have controlled the composition of many of these rocks, it is commonly suggested that mantle-derived magmas play a significant role in providing heat and/or mass input which contributed to the growth of continental crust (e.g., DePaolo, 1981; Holden et al., 1987; Hildreth and Moorbath, 1988; Collins, 1998; Snyder and Tait, 1998; Clynne, 1999; Millar et al., 2001). The evidence for direct contribution of mantle-derived magmas in granitoid production is often cir-cumstantial. Intracrustal melting of older protoliths, formed by a mixture of metasedimentary and juvenile metavolcanics, or
by metaigneous rocks, could also account for the observed hybrid geochemical characteristics (e.g., Clemens and Viel-zeuf, 1987; Eberz et al., 1990; Turpin et al., 1990; Roberts and Clemens, 1993; Maas et al., 1997). The fact that isotopic compositions of most basic rocks related to these granitoids are heterogeneous and are rarely indicative of depleted mantle sources poses another difficulty in the assessment of this ques-tion. Two main hypotheses are envisaged for the origin of such rocks: (1) depleted mantle-derived magmas that suffered con-tamination by crustal rocks; or (2) enriched mantle-derived magmas.
Mafic microgranular enclaves (MMEs; see review of Didier and Barbarin, 1991) are common in intermediate to felsic granitoids and can provide significant information on the nature of the source rocks, the mechanism of production of granitic melt, the geody-namic setting of the granites, as well as evidence of interaction between continental crust and mantle. However, there are consid-erable discrepancies between the models proposed to explain the origin of MMEs as to whether they are restites (e.g., Chappell et al., 1987; Chen et al., 1989), or inclusions of basic magma derived from the mantle (e.g., Vernon, 1984; Bacon, 1986; Holden et al., 1987; Collins, 1998) or from the lower crust (e.g., Eberz et al., 1990; Maas et al., 1997).
* Author to whom correspondence should be addressed ((jinhui@mail.igcas.ac.cn).
Pergamon
Geochimica et Cosmochimica Acta, Vol. 68, No. 21, pp. 4469-4483, 2004 Copyright © 2004 Elsevier Ltd Printed in the USA. All rights reserved 0016-7037/04 $30.00⫹ .00
In our contribution, we present a detailed geochemical and Nd-Sr isotopic study of a suite of mafic microgranular enclaves and host granitoids from the Early Cretaceous Gudaoling batholith in the Liaodong Peninsula, northeastern China, and use these data to constrain the role of MMEs in granitic plutonism. The results suggest that the Guodaoling granitoids are a mixture of melts of enriched lithospheric mantle and both ancient and newly underplated crust. This argues against the restite model proposed by Chappell et al. (1987) and Chen et al. (1989), indicating that these granitoids were not produced by pure intracrustal melting, but require heat and melt input from mantle-derived magmas (see also Vernon, 1984; Holden et al., 1987; Collins, 1998). These results also have important impli-cations for the lithospheric evolution and crustal architecture of the eastern segment of the North China Craton.
2. GEOLOGICAL SETTING AND GEOLOGY OF THE GUDAOLING PLUTON
2.1. Geological Setting
East China is composed of the Xing’an-Mongolia (Xing-Meng) Orogenic belt in the north, the North China Craton (NCC) in the center and the Dabie-Sulu ultrahigh-pressure orogenic belt in the south and east (Fig. 1, Wang and Mo, 1996). The Liaodong Peninsula is located in the eastern seg-ment of the NCC and consists of Archean to Paleoproterozoic basement rocks overlain by unmetamorphosed
Mesoprotero-zoic to PaleoMesoprotero-zoic sediments and MesoMesoprotero-zoic to CenoMesoprotero-zoic sedi-mentary and volcanic rocks (Fig. 1). Early Archean basement rocks, which range in age from 3.85 to 3.2 Ga, have been reported near Anshan, north of the Liaodong Peninsula (Fig. 1, Liu et al., 1992; Song et al., 1996). Deformed late Archean diorites, tonalites and granodiorites were emplaced at about ⬃2500 Ma (LBGMR, 1989). In the Paleoproterozoic, the Li-aohe Group was deposited and then metamorphosed during the 1.85 Ga collisional event that is considered to mark cratoniza-tion of the NCC (Zhao et al., 2001). Subsequently, the Liao-dong Peninsula was covered by a thick sequence of Meso- to Neoproterozoic and Paleozoic sediments (Wang and Mo, 1996). Paleozoic diamond-bearing kimberlite (LBGMR, 1989), Late Triassic basalt (Chen and Chen, 1997), nepheline syenite (Lin et al., 1992; Jing et al., 1995) and Cenozoic gabbro have also been found here.
In the Late Mesozoic, the Liaodong Peninsula became an important part of the circum-Pacific tectono-magmatic zone. About 20,000 km2 of intrusive rocks have been identified, along with minor volcanic rocks (LBGMR, 1989), and these have been interpreted to result from Late Mesozoic lithospheric thinning and crustal extension in the eastern NCC (Menzies et al., 1993; Griffin et al., 1998; Wu et al., 2003; Yang et al., 2003). According to our recent work, these intrusions can be divided into two groups: 1) Jurassic (180 –153 Ma) tonalite, diorite and gneissic two-mica monzogranite that have experi-Fig. 1. Location and setting of the Gudaoling batholith in the Liaodong Peninsula, northeastern China. Inset map shows
the main tectonic subdivisions of northeastern China, with study area indicated by a rectangle.
enced ductile deformation, and 2) undeformed to slightly de-formed Early Cretaceous (131–120 Ma) diorite, granodiorite, monzogranite and syenogranite.
2.2. Petrography of the Gudaoling Batholith
The Cretaceous Gudaoling batholith is a syn-extensional granitoid surrounded by mylonites of the Liaonan metamorphic core complex (Fig. 1). The batholith consists mainly of monzo-granite, with minor quartz diorite at Maoling and biotite grano-diorite at Wanjialing (Fig. 1). The mineralogy of monzogranite is plagioclase, alkali-feldspar, quartz and biotite, with or with-out hornblende. Accessory minerals include apatite, zircon, titanite and Fe-Ti oxides. Granodiorites are medium- to coarse-grained with hypidiomorphic inequigranular textures due to the presence of K-feldspar megacrysts. The main minerals are quartz, plagioclase, K-feldspar, biotite and hornblende, with accessory iron oxides, apatite, zircon, and local titanite and allanite. Quartz diorites are weakly deformed, with a mineral assemblage of quartz, plagioclase and hornblende, with or without K-feldspar and biotite. Field mapping reveals the fol-lowing sequence of magmatic events: (1) intrusion of quartz diorite; (2) emplacement of granodiorite; and (3) intrusion of monzogranite into the granodiorite.
MMEs have been observed locally in the Gudaoling monzo-granite. They range from submillimeter up to tens of meters in size. Those recognized in the field commonly have sharp con-tacts with their host, are angular to oval in shape (Fig. 2) and locally occur as dike-like trails that become progressively thin-ner towards their termination with the host granitoid. Some mafic enclaves contain small monzogranitic enclaves (Fig. 2c), indicating that the mafic and felsic magmas were coeval. The enclaves generally have fine-grained, equigranular, hypidiom-orphic textures, although ophitic textures are present in more basic varieties. Two groups of enclaves are distinguished on the basis of mineral assemblage: (a) diorite to quartz diorite and (b) monzonite enclaves. The main minerals in the dioritic enclaves are plagioclase, hornblende and quartz, with or without biotite and K-feldspar, with accessory iron oxides, apatite, zircon and local titanite and allanite. Apatite displays euhedral acicular habit. The biotite monzonite enclaves have a mineral assem-blage of quartz, plagioclase, K-feldspar and biotite, with or without hornblende, different from the monzogranites with biotite contents up to 20 –30%.
LA-ICP-MS U-Pb dating of zircons indicates emplacement ages for the components of the Gudaoling batholith as follows: 127 ⫾ 4 Ma for the quartz diorite; 124 ⫾ 2 Ma for the granodiorite; and 122⫾ 2 and 118 ⫾ 3 Ma for the biotite monzogranite.40Ar/39Ar dating of hornblende in the dioritic enclaves gives an age of 118.4⫾ 0.1 Ma, consistent with the emplacement age of the host biotite monzogranite. These re-sults are also consistent with the field observations, where the quartz diorite is intruded and enclosed by the granodiorite and monzogranite, and the monzogranite has mingled with the mafic enclaves (Fig. 2).
3. ANALYTICAL METHODS
3.1. Major and Trace Elements
After petrographic examination, 27 of the freshest samples were selected for geochemical analysis and crushed in a
hard-ened jaw crusher and then powdered in an agate mill to⬍ 200 mesh (75 m). Chemical analyses were carried out at the Department of Geosciences, National Taiwan University. Ma-jor element oxides were determined by X-ray fluorescence techniques on fused glass beads using Rigaku RIX-2000 spec-trometers. Trace elements were measured by inductively cou-pled plasma-mass spectrometry (ICP-MS, Perkin Elmer Elan-6000 spectrometer). Fused glass beads were powdered and dissolved using superpure HF and HNO3 (10:1) mixture in screw-top Teflon beakers for 12 h at⬃100 °C, followed by evaporation to dryness, and then refluxing in 7N HNO3. After Fig. 2. Examples of (a and b) mafic microgranular enclaves (MMEs) occurring within monzogranite and (c) monzogranitic enclaves within an MME (c) in the Gudaoling batholith. Hammer is 35 cm long and pen is 14 cm long.
4471 Multiple sources for the origin of the Gudaoling granites
evaporating to dryness, 2 mL HNO3was added, together with 10 ppb Rh as an internal standard. The internal standard was used for monitoring the signal shift during ICP-MS measure-ment: this showed good stability, with ⬃5% variation. The analytical errors are generally better than⫾ 5% (2) as shown by the statistics of duplicate analyses on five rock standards AGV-1, AGV-2, GSP-1, JB-1 and JG-1.
3.2. Nd and Sr Isotopes
Samples for isotopic analysis were dissolved in Teflon bombs after being spiked with84Sr, 87Rb, 150Nd and147Sm tracers before HF⫹ HNO3(with a ratio of 2:1) dissolution. Rb, Sr, Sm and Nd were separated using conventional ion exchange procedures as described by Yang et al. (2004), and measured using a Finnigan MAT 262 multi-collector mass spectrometer at the Institute of Geology and Geophysics, Chinese Academy of Sciences, Beijing, China. Procedural blanks were⬍ 100 pg for Sm and Nd and⬍ 500 pg for Rb and Sr.143Nd/144Nd ratios were corrected for mass fractionation by normalization to 146Nd/144Nd⫽ 0.7219 and87Sr/86Sr ratios were normalized to 86Sr/88Sr⫽ 0.1194. Typical within-run precision (2) for Sr and Nd was estimated to be⫾0.000015. The measured values for the La Jolla and BCR-1 Nd standards and NBS-607 Sr standard were143Nd/144Nd⫽ 0.511853 ⫾ 7 (2
n, n⫽ 3) and 0.512604⫾ 7 (2n, n⫽ 3) and87Sr/86Sr⫽ 1.20042 ⫾ 2 (2n, n⫽12) during the period of data acquisition.
4. RESULTS
4.1. Major and Trace Elements
Granitoids of the Early Cretaceous Gudaoling batholith form part of a calc-alkaline suite that ranges in composition from quartz diorite to granodiorite and monzogranite (Fig. 3a), with an SiO2range from 54.2 to 72.9 wt%. The quartz diorites have higher Al2O3(15.8 –17.8 wt%), P2O5(0.21– 0.27 wt%), TiO2 (0.70 – 0.89 wt%) and higher Mg# [molar 100⫻ MgO/(MgO ⫹ FeOT)⫽ 51.7–60.8] than the monzogranites (Table 1). Biotite monzogranites and granodiorites are peraluminous, with A/CNK [molar ratios Al2O3/(CaO⫹ Na2O⫹ K2O)] of 1.00 – 1.09 (with the exception of sample FW01-21), while quartz diorites are metaluminous with A/CNK of 0.81– 0.91 (Fig. 3b). Quartz diorites have higher V, Cr, Ni, and Rb concentrations, and lower Ba concentrations and Rb/Sr ratios than either the monzogranites or the granodiorites (Table 1).
Enclaves in the Gudaoling batholith are mafic to intermediate in composition (SiO2 51.7– 60.7 wt%), corresponding to gabbro, diorite and syenodiorite in composition (Fig. 3a). The biotite monzonitic enclaves have high Al2O3 concentrations (⬎19.0 wt%) and low Mg# (27.2–32.7) and are peraluminous (A/CNK⫽ 1.06 –1.08), whereas, the dioritic enclaves have lower Al2O3 con-centrations (⬍19.0 wt%) and high Mg# (44.9–72.3) and are metaluminous (A/CNK⫽ 0.59–0.89) (Fig. 3b). The dioritic en-claves have higher V, Cr, Ni, and Rb concentrations, and lower Ba concentrations and Rb/Sr ratios than the biotite monzogranitic enclaves (Table 1).
MgO has been chosen as the abscissa for the variation diagrams, because it more effectively discriminates between these granitic to mafic rocks than SiO2does. Monzogranites, granodiorites and biotite monzonitic enclaves commonly define
a single sequence while dioritic enclaves define another sepa-rate trend (Fig. 4). For SiO2, TiO2, Al2O3, CaO, Cr, Na2O⫹ K2O, Rb/Sr, Sr and Sr/Y, the quartz diorites plot along a line that is the extension of the dioritic enclave data.
Chondrite-normalized REE patterns are shown in Figure 5. Both enclaves and host rocks have high total REE contents and are enriched in light REE. Quartz diorites have similar total REE concentrations (118 –273 times chondrite) and subparallel REE patterns to monzogranites and granodiorites (Figs. 5a,b), but they have high La/Yb values and negligible negative Eu anomalies. The granodiorites and quartz diorites have higher total REE contents than the biotite monzogranites, especially for the HREEs. The (La/Yb)Nratios of quartz diorites, grano-diorites, and monzogranites are high (14 –76) due to HREE fractionation [(Gd/Yb)N⫽ 1.8–5.9] that suggests equilibrium partial melting with a contemporary garnet-rich residuum. En-claves have higher⌺REE contents (Table 1 and Figs. 5c,d) than Fig. 3. Plots of (a) Na2O⫹ K2O vs. SiO2and (b) A/NK [Al2O3/
(Na2O⫹ K2O)] vs. A/CNK [molar ratio Al2O3/(CaO⫹ Na2O⫹ K2O)]
for mafic enclaves and host granitoids from the Gudaoling batholith in the Liaodong Peninsula, northeastern China. The field of Early Creta-ceous volcanic rocks in Figure 3a is from Chen and Chen (1997).
Table 1. Major and trace element analyses of mafic microgranular enclaves and host granitoids of the Gudaoling batholith.
Element
FW01-350 FW01-38 FW01-18 FW01-353 FW01-21 FW01-12 JH-37 JH-38 JH-39 JH-43-2 JH-45 FW01-22 FW01-25 FW01-26
Quartz diorite Granodiorite Monzogranite
SiO2 54.16 60.14 61.86 72.86 69.20 68.43 71.56 71.54 71.44 72.56 70.87 72.42 72.25 72.00 TiO2 0.98 0.75 0.70 0.21 0.38 0.49 0.19 0.18 0.17 0.15 0.20 0.14 0.18 0.18 Al2O3 17.81 15.79 16.39 14.87 15.25 15.19 15.06 15.21 14.89 14.45 15.42 14.49 14.43 14.35 Fe2O3 7.77 5.78 5.43 1.38 2.77 3.71 1.44 1.38 1.40 0.73 1.33 1.49 2.00 1.77 MnO 0.10 0.08 0.07 0.02 0.04 0.04 0.03 0.03 0.02 0.02 0.02 0.02 0.03 0.03 MgO 4.88 4.49 2.91 0.55 0.87 0.65 0.34 0.30 0.27 0.31 0.16 0.36 0.33 0.42 CaO 7.18 5.28 4.85 1.76 2.34 2.42 1.42 1.37 1.35 1.47 1.74 1.48 1.36 1.37 Na2O 4.14 3.74 3.86 4.12 4.14 3.61 3.96 3.94 3.95 3.92 4.47 3.93 3.86 3.95 K2O 1.91 2.87 2.69 4.25 4.03 4.41 4.45 4.57 4.60 4.27 3.39 4.89 4.70 4.86 P2O5 0.27 0.21 0.21 0.09 0.14 0.17 0.07 0.06 0.06 0.05 0.05 0.07 0.07 0.07 TOT 99.70 99.58 99.54 100.38 99.56 99.53 98.51 98.58 98.16 97.91 97.64 99.61 99.51 99.62 Mg# 55.7 60.8 51.7 44.4 38.6 25.9 31.9 30.2 27.9 45.5 19.3 32.6 24.8 32.2 V 128 92 69 11.3 24 17 26 27 25 22 21 13 8.6 16 Cr 131 188 59 4.6 7.9 5.9 7.8 6.3 6.4 5.7 3.6 7.2 2.8 5.1 Ni 29 51 19 2.7 1.9 1.9 4.8 4.8 3.8 4.2 2.6 3.3 0.7 2.2 Ga 23 20 20 16 19 23 18 18 17 17 21 15 19 16 Rb 61.7 95.4 72.3 89.9 124 100 143 141 148 111 142 127 176 171 Sr 1005 673 649 551 602 516 490 487 477 447 838 555 346 404 Y 24.3 16.7 18.5 7.35 13.2 17.7 8.20 8.78 8.13 6.60 4.83 3.21 8.72 8.80 Zr 197 183 202 122 220 441 153 152 143 122 132 159 189 189 Nb 12.5 13.6 11.1 8.58 14.4 26.0 8.15 9.28 7.00 7.75 6.82 4.97 16.8 8.62 Cs 0.89 1.89 0.96 1.03 1.06 0.55 1.77 1.69 1.66 1.04 2.99 1.00 1.89 1.45 Ba 1082 952 1215 2129 1523 1987 1850 1921 1898 1449 2007 1688 1177 1531 La 43.8 54.1 31.4 76.0 61.6 95.1 42.8 34.6 33.0 32.4 28.0 34.9 64.7 54.0 Ce 93.0 92.9 61.6 140 102 169 73.7 58.9 55.2 55.1 50.2 53.2 109 85.1 Pr 11.6 10.6 7.98 12.2 10.9 19.6 7.02 5.60 5.13 5.22 5.04 5.13 11.5 8.71 Nd 47.7 37.5 32.0 39.9 36.1 70.6 23.2 18.7 17.1 17.5 17.7 15.0 37.1 27.1 Sm 7.86 5.61 5.72 4.55 5.20 11.0 3.25 2.75 2.47 2.47 2.54 1.50 5.18 3.53 Eu 2.03 1.51 1.59 1.11 1.20 2.67 0.84 0.79 0.76 0.66 0.93 0.69 0.86 0.81 Gd 6.73 5.03 4.92 4.17 4.52 9.25 2.77 2.46 2.16 2.10 2.06 1.41 4.34 3.15 Tb 0.88 0.61 0.67 0.37 0.50 1.00 0.34 0.33 0.28 0.26 0.25 0.10 0.45 0.31 Dy 4.41 3.20 3.53 1.59 2.53 4.26 1.53 1.56 1.39 1.15 1.04 0.51 1.96 1.58 Ho 0.84 0.57 0.64 0.25 0.43 0.59 0.27 0.29 0.26 0.22 0.17 0.09 0.27 0.28 Er 2.09 1.55 1.68 0.68 1.16 1.42 0.82 0.84 0.84 0.67 0.45 0.32 0.71 0.80 Tm 0.30 0.22 0.25 0.09 0.17 0.17 0.13 0.13 0.15 0.11 0.06 0.05 0.09 0.12 Yb 1.99 1.44 1.57 0.63 1.10 1.06 0.81 0.86 0.99 0.71 0.38 0.39 0.61 0.81 Lu 0.30 0.23 0.25 0.10 0.18 0.15 0.13 0.13 0.16 0.11 0.05 0.07 0.09 0.14 Hf 5.17 4.63 5.29 3.22 5.47 10.9 4.44 4.45 4.09 3.78 3.78 4.33 5.22 4.87 Ta 0.58 0.91 0.59 0.83 0.96 1.40 0.79 1.04 0.94 1.04 0.50 0.23 0.80 0.73 Pb 10 18 14 20 20 15 27 29 27 25 27 24 28 22 Th 4.8 16 5.2 19 16 12 14 12 10 9.7 5.0 9.3 19 12 U 0.90 2.55 1.02 3.21 2.08 2.05 2.03 3.77 1.34 3.35 2.16 1.26 3.63 1.51 4473 Multiple sources for the origin of the Gudaoling granites
Table 1. (Continued)
Element
JH-34 JH-35 JH-40 JH-41 JH-42 JH-44 JH-43-1 FW01-14 JH-36 JH-48 JH-46 JH-47 JH-49
Dioritic enclave Monzonitic enclave
SiO2 52.71 53.66 53.25 60.09 52.73 51.62 56.74 51.25 52.40 55.36 60.23 59.46 60.74 TiO2 0.89 0.86 0.87 0.67 0.85 0.92 0.73 0.86 0.93 0.87 0.64 0.68 0.65 Al2O3 15.76 15.92 15.91 15.51 14.79 14.77 14.96 13.45 16.40 16.26 19.22 19.66 19.05 Fe2O3 6.68 6.27 6.50 4.44 7.02 7.03 5.78 8.21 6.67 6.76 4.70 5.41 5.10 MnO 0.11 0.10 0.11 0.07 0.12 0.15 0.11 0.13 0.10 0.10 0.07 0.08 0.08 MgO 6.70 6.21 6.64 3.99 7.50 7.21 5.67 10.72 6.30 2.75 1.14 1.01 1.02 CaO 7.56 7.63 7.70 5.26 8.12 7.66 6.43 8.97 8.11 4.38 5.13 4.83 4.28 Na2O 3.65 3.82 3.68 3.68 3.34 3.29 3.78 2.84 4.19 3.96 4.23 4.59 4.93 K2O 2.74 2.80 2.67 3.48 2.55 2.69 2.67 1.69 1.86 3.54 1.73 1.79 1.81 P2O5 0.57 0.56 0.57 0.34 0.51 0.56 0.40 0.39 0.64 0.66 0.22 0.22 0.23 TOT 97.37 97.82 97.90 97.52 97.51 95.89 97.28 99.64 97.60 94.65 97.33 97.71 97.90 Mg# 66.7 66.5 67.2 64.3 68.1 67.2 66.2 72.3 65.4 44.9 32.7 27.2 28.6 V 121 115 118 93 126 119 107 136 125 91 48 40 42 Cr 185 161 185 108 221 181 165 493 154 41 6.9 5.1 7.0 Ni 114 107 114 53 134 115 91 181 106 28 4.9 5.1 5.7 Ga 17 17 17 17 16 16 17 15 18 20 25 26 25 Rb 87.4 68.9 77.8 101 68.5 130 101 43.0 70.0 97.4 147 156 181 Sr 1085 1088 1127 852 995 916 783 814 1143 646 1307 1468 1130 Y 22.5 22.1 22.7 21.9 21.8 22.1 19.5 20.3 23.2 23.9 16.4 13.4 14.9 Zr 221 210 209 170 192 208 197 169 207 301 269 228 175 Nb 12.0 13.0 11.9 14.4 11.1 11.4 11.4 8.90 12.0 16.0 10.8 12.1 12.7 Cs 2.29 1.35 1.72 1.73 1.67 5.06 2.35 0.77 1.49 2.95 4.98 4.43 6.34 Ba 1433 1532 1488 1566 1412 1073 1321 854 713 1649 828 1053 627 La 51.1 53.6 55.5 39.7 53.7 47.9 41.7 50.5 65.8 72.0 42.8 53.6 40.3 Ce 100 103 104 76.7 99.1 91.3 80.9 94.8 117 139 86.0 104 79.7 Pr 10.9 11.0 11.1 8.42 10.5 9.87 8.77 11.8 11.9 14.9 9.72 11.2 8.78 Nd 42.6 42.1 42.6 32.7 40.3 38.7 33.9 45.0 45.3 56.6 38.7 41.8 34.1 Sm 7.11 6.87 6.96 5.71 6.69 6.39 5.69 7.54 7.31 8.59 6.51 6.02 5.51 Eu 1.99 1.95 1.92 1.48 1.86 1.74 1.60 2.08 1.88 2.11 1.63 1.56 1.15 Gd 6.17 5.93 6.12 5.02 5.76 5.55 4.97 6.51 6.39 7.22 5.10 4.81 4.53 Tb 0.88 0.85 0.86 0.77 0.83 0.81 0.72 0.81 0.91 0.96 0.70 0.60 0.62 Dy 4.38 4.18 4.28 3.96 4.10 4.08 3.62 4.09 4.46 4.47 3.15 2.59 2.86 Ho 0.82 0.79 0.81 0.75 0.77 0.76 0.68 0.73 0.84 0.82 0.56 0.45 0.49 Er 2.37 2.25 2.35 2.17 2.22 2.22 1.96 1.93 2.37 2.35 1.58 1.27 1.33 Tm 0.35 0.34 0.35 0.33 0.34 0.34 0.29 0.27 0.35 0.35 0.24 0.18 0.18 Yb 2.21 2.15 2.17 2.08 2.08 2.07 1.87 1.75 2.22 2.21 1.47 1.15 1.14 Lu 0.34 0.33 0.33 0.30 0.31 0.31 0.29 0.27 0.34 0.35 0.23 0.18 0.17 Hf 5.60 5.27 5.28 4.85 4.91 5.09 5.13 4.11 5.24 6.82 6.32 5.57 4.38 Ta 0.86 0.90 0.85 1.75 0.77 0.76 0.92 0.46 0.90 0.99 0.90 0.84 1.05 Pb 8.6 9.5 8.6 14 8.6 8.4 9.7 4.2 8.0 6.6 10 15 21 Th 8.5 9.3 9.4 9.4 9.2 8.8 10 6.3 10.3 5.4 4.3 6.1 4.4 U 2.75 3.48 2.76 3.23 2.15 3.15 6.63 1.15 3.16 0.94 1.35 1.59 1.83 4474 J.-H. Yang et al.
the monzogranites. They also have markedly higher HREE contents than the monzogranites.
In the primitive mantle (PM)-normalized trace element pat-terns (Fig. 6), all of these rocks are enriched in large ion lithophile elements (LILEs, such as Rb, Ba, Sr) and light rare earth elements (LREEs), and depleted in high field strength elements (HFSEs, such as Nb, Ta, P, Ti). Biotite monzonitic enclaves have similar characteristics to the host rocks, includ-ing positive Sr anomalies (Figs. 6a– c). The biotite monzonitic enclaves have relatively elevated Rb values compared to the dioritic enclaves (Figs. 6c,d).
4.2. Sr-Nd Isotope Results
Rubidium, Sr, Sm and Nd concentrations,143Nd/144Nd and 87Sr/86Sr ratios, and T
DMages for MMEs and granitoids are listed in Table 2. The initial87Sr/86Sr ratios and
Nd(t) values have been calculated at 120 Ma on the basis of zircon U-Pb and hornblende40Ar/39Ar dating of rocks from the batholith. De-pleted mantle model ages (TDM) are reported using the model of DePaolo (1981). The data are shown in a plot of Nd(t) versus (87Sr/86Sr)
iin Figure 7, and compared on that diagram with published compositional fields for Late Triassic nepheline Fig. 4. Various oxide plots [(a), SiO2,(b), TiO2, (c), Al2O3, (d), CaO and (e), K2O⫹ Na2O vs. MgO (all expressed in
wt%)] and trace element plots [(f), Cr, (g), Rb/Sr and (h), Sr/Y] vs. MgO diagrams (expressed in ppm) for mafic enclaves and host rocks from the Gudaoling batholith in the Liaodong Peninsula, northeastern China. Solid lines indicate evolu-tionary trends for the mafic dioritic enclaves and dashed lines indicate trends shown by the granodiorites, monzogranites and biotite monzonitic enclaves. Quartz diorites consistently plot near the dioritic enclaves.
4475 Multiple sources for the origin of the Gudaoling granites
syenites (Jing et al., 1995) and basalts (Chen and Chen, 1997) from the Liaodong Peninsula and Late Jurassic and Early Cretaceous granites and Cenozoic gabbros from the area.
The mafic enclaves and host granitoids show a large range in (87Sr/86Sr)
iratios (0.7055– 0.7124) andNd(t) values (⫺7.2 to ⫺20.9). The monzogranites are characterized by high
87Rb/86Sr ratios and low initial 87Sr/86Sr ratios (0.70552– 0.70864) and strongly negativeNd(t) values (⫺18.5 to⫺20.9). In contrast, the quartz diorites have high (87Sr/86Sr)
i ratios (0.71185– 0.71240) and relatively low 87Rb/86Sr ratios (0.1773– 0.4101). The granodiorites have transitional isotopic compositions between monzogranites and quartz diorites with Fig. 5. Chondrite-normalized REE patterns for (a) monzogranites and granodiorites, (b) quartz diorites, (c) monzonitic
enclaves, and (d) dioritic enclaves of the Gudaoling batholith. The chondrite values are from Sun and McDonough (1989).
Fig. 6. Primitive mantle (PM) normalized trace element patterns, elements arranged in the order of decreasing incompatibility from left to right. (a) Monzogranites and granodiorites, (b) quartz diorites, (c) monzonitic enclaves, and (d) dioritic enclaves of the Gudaoling batholith. The PM values are from Sun and McDonough (1989).
Table 2. Rb-Sr and Sm-Nd isotopic results of mafic microgranular enclaves and host granitoids of the Gudaoling batholith.
Sample Rock type Rb (ppm) Sr (ppm) 87Rb/86Sr 87Sr/86Sr 2 Sm (ppm) Nd (ppm) 147Sm/144Nd 143Nd/144Nd 2 (87Sr/86Sr) ia Nd(t)a,b TDM (Ma)a,b f Sm/Nd JH-37 Monzogranite 140 486 0.8323 0.707130 11 3.10 22.7 0.0827 0.511502 10 0.70571 ⫺20.4 1912 ⫺0.58 JH-38 Monzogranite 145 491 0.8532 0.706979 15 2.72 18.7 0.0880 0.511494 7 0.70552 ⫺20.6 2001 ⫺0.55 JH-39 Monzogranite 142 481 0.8527 0.707106 11 2.34 16.7 0.0848 0.511475 8 0.70565 ⫺20.9 1974 ⫺0.57 JH-43-2 Monzogranite 105 448 0.6796 0.706783 13 2.22 16.0 0.0840 0.511513 11 0.70562 ⫺20.2 1917 ⫺0.57 FW01-22 Monzogranite 126 542 0.6716 0.709786 15 1.33 20.3 0.0396 0.511563 10 0.70864 ⫺18.5 1387 ⫺0.80 FW01-26 Monzogranite 195 462 1.223 0.708998 18 3.34 26.6 0.0761 0.511522 8 0.70691 ⫺19.9 1798 ⫺0.61 FW01-353 Granodiorite 90.5 530 0.5419 0.709305 19 4.09 31.9 0.0775 0.511647 12 0.70838 ⫺17.4 1678 ⫺0.61 FW01-353c Granodiorite 89.9 523 0.5419 0.709323 14 3.97 32.3 0.0750 0.511603 10 0.70826 ⫺17.8 1661 ⫺0.63 FW01-21 Granodiorite 114 581 0.5699 0.710414 17 5.57 34.0 0.0992 0.511931 11 0.70944 ⫺12.2 1619 ⫺0.50 FW01-18 Qz diorite 77.7 669 0.3363 0.712969 20 4.59 25.7 0.1082 0.511638 9 0.71240 ⫺18.1 2176 ⫺0.45 FW01-350 Qz diorite 99.3 984 0.1773 0.712335 18 7.19 41.0 0.1059 0.511636 11 0.71203 ⫺18.1 2133 ⫺0.46 FW01-38 Qz diorite 92.2 651 0.4101 0.712552 17 5.19 41.28 0.0760 0.511865 19 0.71185 ⫺13.2 1420 ⫺0.61 JH-34 Dioritic enclave 84.0 1072 0.2266 0.706682 14 6.78 41.6 0.0987 0.512183 8 0.70630 ⫺7.3 1280 ⫺0.50 JH-35 Dioritic enclave 67.7 1050 0.1865 0.706579 11 7.04 44.0 0.0968 0.512189 6 0.70626 ⫺7.2 1251 ⫺0.51 JH-43-1 Dioritic enclave 95.2 713 0.3863 0.706451 10 5.47 32.8 0.1010 0.512149 16 0.70579 ⫺8.0 1352 ⫺0.49 FW01-14 Dioritic enclave 47.1 795 0.1714 0.707597 20 6.92 39.8 0.1051 0.512172 10 0.70730 ⫺7.6 1371 ⫺0.47 JH-48 Dioritic enclave 91.4 640 0.4132 0.705925 11 8.83 58.9 0.0906 0.511649 7 0.70522 ⫺17.6 1853 ⫺0.54 JH-46 Monzonitic enclave 138 1239 0.3235 0.708958 9 7.17 44.5 0.0975 0.511673 5 0.70841 ⫺17.2 1931 ⫺0.50 JH-47 Monzonitic enclave 147 1366 0.3119 0.708944 11 6.12 42.8 0.0864 0.511707 8 0.70841 ⫺16.4 1724 ⫺0.56 at⫽ 120 Ma.
bThe143Nd/144Nd and147Sm/144Nd ratios of chondrite and depleted mantle at present day are 0.512638 and 0.1967, and 0.51315 and 0.222, respectively. cDuplicate. 4477 Multiple sources for the origin of the Gudaoling granites
87Rb/86Sr ratios ranging from 0.4939 to 0.5699, initial87Sr/86Sr ratios ranging from 0.70848 to 0.70944, and Nd(t) values ranging from ⫺12.2 to ⫺18.2. The dioritic enclaves have a narrow range in initial 87Sr/86Sr ratios (0.70579 to 0.70730) and relatively highNd(t) values (⫺7.2 to ⫺8.0) except for one sample (JH-48) which has a low (87Sr/86Sr)
iratio andNd(t) value, clearly different from those of the biotite monzonitic enclaves [(87Sr/86Sr)
i ⫽ 0.70841 and Nd(t) ⫽ ⫺16.4 to ⫺17.2].
TheNd(t)-(87Sr/86Sr)
iplot (Fig. 7) highlights the following points: (1) Four of the dioritic enclaves (JH-34,⫺35, ⫺43–1 and FW01-14) plot in a narrow field [Nd(t): ⫺8.0 to ⫺7.2; (87Sr/86Sr)
i: 0.7058 to 0.7073]; (2) four of the monzogranites (JH-37, ⫺38, ⫺39 and ⫺43–2) also plot in a narrow field [Nd(t):⫺20.9 to ⫺20.2; (87Sr/86Sr)
i: 0.70552 to 0.70571]; the other two samples have higher (87Sr/86Sr)
i ratios (0.7069 – 0.7086) andNd(t) values (⫺19.9 to ⫺18.5); (3) quartz diorites have the highest (87Sr/86Sr)
iratios (0.7119 – 0.7124) and have negativeNd(t) values (⫺13.2 to ⫺18.1); (4) the biotite mon-zonitic enclaves have isotopic compositions similar to the high-est (87Sr/86Sr)
i monzogranite sample (FW01-22) and lowest (87Sr/86Sr)
igranodiorite sample (FW01-353).
5. DISCUSSION
5.1. Restite or Magma Mixing?
With the development of the restite unmixing theory (e.g., Chappell et al., 1987; Chappell and White, 1992), mafic mi-crogranular enclaves were then incorporated in genetic models of granitoids. In this model, they are thought to represent residual material (restite) which successively unmixes from the
melt during the rise of a crystal mush from its source region. A critical feature of the restite model is linear chemical variations, which are observed in many calc-alkaline plutonic suites (e.g., Chappell et al., 1987; Chappell and White, 1992; Collins, 1998). However, these features are not observed in the Guda-oling suite. In the selected oxides and trace elements versus MgO diagrams (Fig. 4), there are no converging linear trends shown that incorporate the enclaves, monzogranites and quartz diorites.
The isotopic data presented here place important constraints on the origin of microgranular enclaves. The large isotopic differences between enclaves and host rocks clearly rule out any strictly cogenetic origin. Restite derived from the source rock of the granitoids should be in isotopic equilibrium with its enclosing melt. It is therefore unlikely that the mafic micro-granular enclaves from the Gudaoling batholith are restitic in origin. This view is supported by detailed microstructural ob-servations, since the enclave textures are igneous and not metamorphic (cf. Eichelberger, 1980; Vernon, 1984).
The heterogeneous isotopic compositions of this granitoid-mafic enclave association preclude a simple, common evolution by closed-system fractionation processes. Increasing recogni-tion that many magma chambers are open systems that may regularly be fed with more primitive magma (which can be related or unrelated to magma already in the chamber) caused a shift in the role assigned to microgranular enclaves (e.g., Vernon, 1984; Collins, 1998). Accordingly, they may also be considered to represent remnants of a mafic component added to intermediate to felsic magma chambers (e.g., Holden et al., 1987; Didier and Barbarin, 1991; Collins, 1998). In addition, the fact that they do not fit a single trend in the Nd(t) vs. Fig. 7.Nd(t) vs. (87Sr/86Sr)iplot of mafic enclaves and host granitoids of the Gudoaling batholith, Liaodong Peninsula.
The data for Triassic syenites and basalts are from Jing et al. (1995) and Chen and Chen (1997), respectively. The data for Jurassic granitoids (two-mica granites, diorites and tonalite), Early Cretaceous I-type granites and Cenozoic gabbros are from Wu F.-Y. (unpublished data). Superimposed on these fields are isotopic mixing curves with the dots along the three lines that link up the end members at 10% intervals. The fields were calculated as follows:
Sources Sr (ppm) (87Sr/86Sr)
i Nd (ppm) 143Nd/144Nd Nd(t)
Melt from lithospheric mantle 1070 0.7065 42 0.51217 ⫺6.0 Melt from juvenile lower crust 480 0.7050 18 0.51140 ⫺21.0 Melt from ancient lower crust 300 0.7200 22 0.51135 ⫺22.5
(87Sr/86Sr)
idiagram (Fig. 7), together with the lack of evidence for converging linear geochemical trends for the enclaves, monzogranites and quartz diorites (Fig. 4) precludes a simple two-component mixing model. The enclaves may therefore represent additional components that need to be considered when interpreting the generation of their host rocks. The iso-topic data (Fig. 7) may be interpreted as a result of three-component mixing among mafic [member 1: (87Sr/86Sr)
i: ⬃0.7058, and Nd(t):⬃⫺7.2], intermediate or felsic [member 2: (87Sr/86Sr)
i: ⬎0.7120, and Nd(t): ⬍⫺18.2], and felsic [member 3: (87Sr/86Sr)
i:⬃0.7055, and Nd(t):⬃⫺20.9] isoto-pically evolved end-members, which are represented by the isotopic compositions of the dioritic enclaves, quartz diorites and monzogranites, respectively.
5.2. Mafic Magma Sources
The dioritic enclaves have increasing SiO2, Al2O3, K2O⫹ Na2O concentrations and decreasing CaO and Cr concentra-tions (Fig. 4), and constant initial87Sr/86Sr ratios and
Nd(t) values with decreasing MgO contents (Fig. 8). They have negativeNd(t) values (⬃⫺7.2) and low Nb/U (1.7–7.7) and Ce/Pb (5.4 –22.4) ratios, similar to those of continental crust (Nb/U⫽ 6.2 and Ce/Pb ⫽ 3.9; Rudnick and Fountain, 1995), but distinct from those of mid-ocean ridge and oceanic island
arc basalts (MORB and OIB, Nb/U ⫽ 47 and Ce/Pb ⫽ 27, Hofmann et al., 1986). In addition, the dioritic enclaves are depleted in high field strength elements (HFSEs) with Nb and Ta anomalies and enriched in large ion lithophile elements (LILEs) and light REEs (Fig. 6d). All these features suggest that the parental magmas have been contaminated by continen-tal crust in the source or during magma ascent, accompanied by mafic mineral fractionation.
The dioritic enclaves have high MgO (up to 10.72%), Mg# (up to 72.3), Cr (up to 493 ppm) and Ni (up to 181 ppm) at basic to intermediate silica contents, suggesting the dioritic enclaves contain a mantle component. The TiO2contents and Nd(t) values of samples with high Mg#, Cr and Ni contents are relatively constant with variable MgO concentrations (Figs. 4b and 8b), suggesting these samples may represent primary magma compositions that have not experienced Fe-Ti oxide fractionation and crustal assimilation during magma ascent. The low FeO and TiO2 contents of the high-Mg# dioritic enclaves, corresponding to the component of experimental melts from depleted peridotite (Falloon et al., 1988), indicate a dominantly refractory mantle (low TiO2, CaO/Al2O3, HREEs) contribution to the melt. The high-Mg# dioritic enclaves have trace element characteristics of a subduction-zone environment (Fig. 6d), for which there is no evidence in the study area. The contemporaneous formation of the Liaonan metamorphic core complex, pull-apart basins and emplacement of A-type granites with emplacement of the Gudaoling intrusion in the Liaodong Peninsula indicate an intraplate extensional environment during magma generation. Therefore, the incompatible element en-richment and subduction signatures may have been inherited from earlier subduction events, in which lithospheric mantle was metasomatized by slab-derived aqueous fluids, silicic melts or sediments (Hickey and Frey, 1982; Crawford et al., 1989; Pearce et al., 1992; Kelemen, 1995; Shimoda et al., 1998). This is similar to the formation of high-Mg boninites (e.g., Crawford et al., 1989), boninitic diorite intrusions (e.g., Kemp, 2003) and high-Mg andesites (e.g., Tatsumi and Maruyama, 1989).
Enriched isotopic and geochemical signatures are not re-stricted to the dioritic enclaves in the Gudaoling batholith. Similar negativeNd(t) values (⫺5 to ⫺8) have been reported from Early Cretaceous mafic volcanics of the Liaodong Penin-sula (Chen and Chen, 1997), indicating their derivation from similar mantle reservoirs. Furthermore, melts derived from subcontinental lithosphere beneath the Liaodong Peninsula are high-K calc-alkaline to shoshonitic in composition (Chen and Chen. 1997), which is similar to the composition of the en-claves (Fig. 3a). Therefore, we propose that the parental mag-mas of dioritic enclaves were derived by partial melting of enriched lithospheric mantle.
5.3. Intermediate Magma Sources
The quartz diorites are notable for having high Al2O3 concen-trations and (87Sr/86Sr)
i ratios but low Rb/Sr ratios, (Tables 1 and 2), suggesting involvement of melts from metagreywackes or metaigneous rocks with a near-complete breakdown of plagioclase at high pressure (Patinˇo Douce, 1999). However, the high MgO concentrations for both quartz diorites and granodiorites, together with their isotopic array extrapolating towards the field of dioritic enclaves (Figs. 4, 7, and 8), suggest that they resulted from the Fig. 8. Plots of (a) (87Sr/86Sr)
iand (b)Nd(t) vs. MgO for mafic
microgranular enclaves and host granitoids of the Gudaoling batholith showing crystal fractionation, crustal assimilation and magma mixing trends.
4479 Multiple sources for the origin of the Gudaoling granites
reaction of metagreywacke or metaigneous rock with basaltic magma at high pressure. The existence of ancient metagreywackes or metaigneous rocks beneath the Liaodong Peninsula is supported by the isotopic compositions of the Jurassic gneissic granites and Archean metamorphic rocks and granites that outcrop in the area, which have high (87Sr/86Sr)
iratios (⬃0.715) and low Nd(t) values (⬃⫺22) and are interpreted to be derived from an ancient crustal source (Wan et al., 1998, 1999). Therefore, high (87Sr/86Sr)
i metagreywackes or metaigneous rocks represent a potential crustal end-member; probably from an ancient lower crustal source.
5.4. Felsic Magma Sources
The monzogranites, granodiorites and quartz diorites of the Gudaoling batholith have Sr-Nd isotopic signatures reflecting crustal derivation, and their concave-up REE patterns with respect to MREE-HREE (Figs. 5a,b) suggest equilibrium with garnet and amphibole. Geochemical data from TTG/adakites and Na-rich granitoids from other locations (e.g., Defant and Drummond, 1990; Atherton and Petford, 1993; Martin, 1999) show similarities with the Early Cretaceous Gudaoling grani-toids in terms of high Sr, low Yb and Y concentrations and high Sr/Y and La/Yb ratios (Table 1). Considering that partial melt-ing of subducted oceanic crust or newly underplated lower crust under garnet stability conditions played an important role in the generation of TTG/adakites and Na-rich granitoids (e.g., Defant and Drummond, 1990; Atherton and Petford, 1993; Martin, 1999), these trace element characteristics can originate from partial melting of a crustal source with a high-pressure, plagioclase-poor and garnet-rich residual assemblage (Atherton and Petford, 1993; Patinˇo Douce, 1999).
In addition, the trace element variations can be explained by an assimilation-fractional crystallization (AFC) process involv-ing basaltic magma (e.g., Castillo et al., 1999). The isotopic evolutionary trend of the granitoids, i.e., increasingly negative Nd(t) (⫺12.2 to ⫺20.9) from intermediate quartz diorites to granodiorites and monzogranites (Table 2), is also consistent with an AFC process, combined with an increasing degree of differentiation. However, the large range in (87Sr/86Sr)
iratios (0.70552 to 0.71240) and theNd(t)-(87Sr/86Sr)i (Fig. 7) and (87Sr/86Sr)
i-MgO plots (Fig. 8) show that, except for the en-riched mantle source, two distinct crustal sources, i.e., a high-(87Sr/86Sr)
iand a low-(
87Sr/86Sr)
isource, can be distinguished. The monzogranite samples are characterized by strongly negative Nd(t) values (⬃⫺21) combined with a relatively nonradiogenic Sr isotopic signature [(87Sr/86Sr)
i⬇ 0.705] (Ta-ble 2 and Fig. 7) and show little interaction with other magmas parental to the mafic enclaves and quartz diorites. The samples with lower initial87Sr/86Sr ratios (⬃0.705) have high Rb/Sr ratios (0.29 – 0.51), which are usually attributed to partial melt-ing processes involvmelt-ing mica breakdown and/or late plagio-clase fractionation. However, the absence of negative Eu anom-alies in samples with lower initial87Sr/86Sr ratios indicates the parental magmas have not experienced plagioclase fraction-ation. Recently, detailed studies of granitic rocks in a number of orogenic belts have identified broad positive arrays between Rb/Sr ratios of individual granite samples and the time-inte-grated Rb/Sr ratios of their source rocks inferred from their model neodymium ages. It means that the Rb/Sr ratios of granites can reflect the Rb/Sr ratios of their source rocks (Kemp
and Hawkesworth, 2003). Therefore, the source of the monzo-granites would have high Rb/Sr ratios, but low initial87Sr/86Sr ratios, indicating it is a mica-bearing source with a short resi-dence time. The monzogranites show elemental characteristics (e.g., high Sr, Ba and LREE, low Y and HREE, elevated La/Yb and Sr/Y ratios) that are typical of adakites, Archean TTG suites and Na-rich granitoids (Defant and Drummond, 1990; Atherton and Petford, 1993; Martin, 1999). Therefore, they have been envisaged by some workers as adakites that were derived by lower crustal melting under high pressure with a garnet-bearing residue (Zhang et al., 2001; Defant et al., 2002). We conclude that the parental magmas of the monzogranites were produced by partial melting of a juvenile basaltic lower crust. The newly underplated basaltic magmas should have a geochemistry and isotopic composition similar to that of en-riched mantle, with highly negativeNd(t) values (⬍⫺15). The Sr-Nd isotopic data for the Late Triassic nepheline syenites [(87Sr/86Sr)
i⫽ 0.705–0.706 and Nd(t) values⫽ ⫺13 to ⫺15 (Jing et al., 1995)] (Fig. 7) are similar to those of the monzo-granites. It can therefore be interpreted that the source of the newly underplated magmas was similar to that of the nepheline syenites.
5.5. Genesis of the Gudaoling Batholith
Based on the magma sequence, petrography, major and trace element geochemical data and Sr and Nd isotopic compositions of mafic microgranular enclaves and host granitic rocks, a complex, multi-stage process involving magma mixing, crystal fractionation, wall-rock assimilation, and crustal anatexis of three distinct sources is proposed for the formation of the Gudaoling granitoids.
The monzogranites with low initial 87Sr/86Sr ratios and highly negativeNd(t) values were mainly derived by partial melting of juvenile basaltic lower crust coupled with crystal fractionation and little or no involvement of other components, such as enriched lithospheric mantle or ancient lower crust. However, the samples with higher initial87Sr/86Sr ratios (up to 0.70864) andNd(t) values up to⫺18.5, most likely originated from partial melting of 80 –90% juvenile lower crust, followed by crystal fractionation, but mixed with mantle-derived mag-mas and assimilated ancient crustal materials (Fig. 7).
The granodiorites and quartz diorites were the result of interaction between a mantle-derived magma and ancient lower crustal materials, with involvement of juvenile lower crustally derived magma. Sr-Nd isotopic calculations indicate that 60 – 70% mantle, 5–10% juvenile lower crust and 20 –30% ancient lower crustal material is compatible with the data.
The dioritic enclaves with Nd(t) values of ⬃⫺7.2 were derived from an enriched lithospheric mantle with fractional crystallization and little involvement of crustal materials. The biotite monzonitic enclaves have similar isotopic compositions to the monzogranite (e.g., FW01-22), indicating that they were mainly produced by fractional crystallization of magmas pa-rental to the monzogranites.
5.6. Implications for Crustal Architecture
There is a progressive change in composition of plutonic rocks in the Liaodong Peninsula with time (Fig. 9), reflecting a
change from continental crust/ancient lithospheric mantle sources in the Late Triassic to juvenile continental crust/litho-spheric mantle sources in the Early Cretaceous. This may be explained by the replacement of ancient lithospheric mantle by juvenile lithospheric mantle during the Late Jurassic to Early Cretaceous, which is consistent with the fact that lithospheric mantle beneath the eastern North China Craton has been thinned considerably since the Phanerozoic, as evidenced by geophysical and geochemical data (Menzies et al., 1993; Grif-fin et al., 1998; Wu et al., 2003; Yang et al., 2003).
We provide the following synthesis of major events in the region during this time period:
(i) Metagreywackes and/or metaigneous rocks represent a common lower crustal component of the Liaodong Peninsula before underplating by mantle-derived magma from melting of an ancient lithospheric source prior to the formation of the Gudaoling granitoids. The underplate contributed substantially to crustal thickness and these magmas had low (87Sr/86Sr)
iand Nd(t) values similar to the Late Triassic nepheline syenites.
(ii) Ancient lithospheric mantle was replaced by younger, less refractory lithospheric mantle in the eastern part of the North China Craton, as evidenced by high surface heat flow, uplift and basin development, slow seismic wave velocities in the upper mantle, and changes in the Nd isotopic compositions of mantle-derived rocks and in the character of mantle xeno-liths sampled by Paleozoic to Cenozoic magmas (Griffin et al., 1998; Wu et al., 2003; Yang et al., 2003).
(iii) Basaltic magmas derived from juvenile lithospheric mantle underwent fractional crystallization during uprise into the continental crust and induced anatexis of the lower crust. Both mantle-derived and anatectic magmas mixed together, and coupled with AFC processes, gave rise to magmas ranging from quartz diorite to biotite monzonite in composition. The
dioritic enclaves represent the tangible remains of these mantle-derived magmas, whereas the biotite monzonitic enclaves are products of crystal fractionation of the magmas which were parental to the monzogranites.
6. CONCLUDING REMARKS
Mafic microgranular enclaves and host granitoids of the Gudoaling batholith in the Liaodong Peninsula of NE China provide direct evidence of the involvement of mantle-derived magmas in granitoid genesis. Geochemical and Sr and Nd isotopic compositions support multiple sources for their origin. Basic magmas were emplaced penecontemporaneously with granitoids and were contaminated during their ascent by crustal components. Although it is difficult to deduce precise estimates for the isotopic signature of the basaltic component, the avail-able data suggest that a magma with high MgO, negativeNd(t) and high (87Sr/86Sr)
i, i.e., partial melts of enriched refractory lithospheric mantle and its evolved products, were involved in the hybridization process. Monzogranites with low (87Sr/86Sr)
i ratios and negativeNd(t) values are interpreted as representa-tives of a crustal end-member generated during this process. These were derived by fractionation or partial melting of sources with a short time-integrated Rb/Sr ratio, such as newly underplated crust. Another crustal end-member, ancient lower crust, is identified in the quartz diorites and biotite monzonitic enclaves that were the result of interaction between meta-greywackes and/or metaigneous rocks and ascending basalt magma.
The Gudaoling plutonic association thus illustrates that hy-bridization of mantle-derived magmas and crustal components is a complex, multi-stage process involving several different mechanisms (magma mixing, crystal fractionation, and wall-rock assimilation) which all operated together during the Late Mesozoic in the Liaodong Peninsula of NE China.
Acknowledgments—J. H. Yang benefited from a 1-yr stay in the De-partment of Geosciences, National Taiwan University. We thank Profs. Guang-Sheng Qiao and Ren-Hu Zhang and Drs. Zhu-Yin Chu and Chao-Feng Li for helping with Sr and Nd isotopic analyses. Grahame Oliver commented on an earlier draft and we thank Charles R. Bacon and P. T. Leat for their constructive reviews and Martin Menzies (AE), Frank A. Podosek (EE), and Linda Trower (EM) for handling the manuscript. This study was supported by grants NSFC-40325006, 40132020 and 40133020, the “Funds for Hundred Outstanding Talents Plan,” and KZCX1-07.
Associate editor: M. Menzies
REFERENCES
Atherton M. P. and Petford N. (1993) Generation of sodium-rich magmas from newly underplated basaltic crust. Nature 362, 144 – 146.
Bacon C. R. (1986) Magmatic inclusions in silicic and intermediate volcanic rocks. J. Geophys. Res. 91, 6091– 6112.
Castillo P. R., Janney P. E., and Solidum R. (1999) Petrology and grechemistry of Camiguin Island, southern Philippines: Insight into the source of adakite and other lavas in a complex arc tectonic setting. Contrib. Minerl. Petrol. 134, 33–51.
Chappell B. W. and Stephens W. E. (1988) The origin of infracrustal (I-type) granite magmas. Trans. R. Soc. Edinburgh Earth Sci. 79, 71– 86.
Fig. 9. Plot of Nd(t) vs. Age (Ma) for mafic enclaves and host
granitoids from the Gudaoling batholith, compared with Jurassic gran-ites, shows the lithospheric evolution of the Liaodong Peninsula. Fields of Triassic syenites and basalts, Jurassic granitoids, and Cenozoic gabbros as defined in Figure 7. The data for Precambrian TTG suites and granites are from Song et al. (1996) and Wan et al. (1998, 1999). The inset map showing the evolution of depleted mantle and ancient crust with time.
4481 Multiple sources for the origin of the Gudaoling granites
Chappell B. W., White A. J. R., and Wyborn D. (1987) The importance of residual source material (restite) in granite petrogenesis. J. Petrol. 28, 1111–1138.
Chen Y. D., Price R. C., White A. J. R., and Chappell B. W. (1989) Inclusions in three S-type granites from southeastern Australia. J. Petrol. 30, 1181–1218.
Chappell B. W. and White A. J. R. (1992) I- and S-type granites in the Lachlan Fold Belt. Trans. R. Soc. Edinburgh Earth Sci. 83, 1–26. Chen Y. X. and Chen W. J. (1997) Mesozoic Volcanic Rocks in West Liaoning and Adjacent Areas. Seismological Publishing House, Beijing.
Clemens J. D. and Vielzeuf D. (1987) Constraints on melting and magma production in the crust. Earth Planet. Sci. Lett. 86, 287– 306.
Clynne M. A. (1999) A complex magma mixing origin of rock erupted in 1915, Lassen Peak, California. J. Petrol. 40, 105–132. Collins W. J. (1998) Evaluation of petrogenetic models for Lachlan
Fold Belt granitoids: Implications for crustal architecture and tec-tonic models. Austr. J. Earth Sci. 45, 483–500.
Crawford A. J., Falloon T. J. and Green D. H. (1989) Classification, petrogenesis and tectonic setting of boninites. In Boninites and Related Rocks (ed. A. J. Crawford), pp. 2– 44. Unwin Hyman. Defant M. J. and Drummond M. S. (1990) Derivation of some modern
arc magmas by melting of young subducted lithosphere. Nature
347, 662– 665.
Defant M. J., Xu J. F., Kepezhinskas P., Wang Q., Zhang Q., and Xiao L. (2002) Adakites: Some variations on a theme. Acta Petrol. Sinica
18, 129 –142.
DePaolo D. J. (1981) Neodymium isotopes in the Colorado Front Range and crust-mantle evolution in the Proterozoic. Nature 291, 193–196.
Didier J. and Barbarin B. (1991) Enclaves and Granite Petrology, Developments in Petrology. Elsevier Science.
Eberz G. W., Nicholls I. A., Maas R., McCulloch M. T., and Whiteford D. J. (1990) The Nd- and Sr-isotopic composition of microgranitoid enclaves and their host rocks from the Swifts Creek Pluton, south-east Australia. Chem. Geol. 85, 119 –134.
Eichelberger J. C. (1980) Vesiculation of mafic magma during replen-ishment of silicic magma reservoirs. Nature 288, 446 – 450. Falloon T. J., Green D. H., Hatton C. J., and Harris K. L. (1988)
Anhydrous partial melting of a fertile and depleted peridotite from 2 to 30 kba and application to basalt petrogenesis. J. Petrol. 29, 1257–1282.
Griffin W. L., Zhang A., O’Reilly S. Y., Ryan C. G. (1998) Phanero-zoic evolution of the lithosphere beneath the Sino-Korean Craton. In Mantle Dynamics and Plate Interaction in East Asia (eds. M. F. J. Flower, S. L. Chung, C. H. Lo and T. Y. Lee), 107–126. Geodynamics Series 27.
Hickey R. and Frey F. A. (1982) Geochemical characteristics of bon-inite series volcanics: Implications for their source. Geochim. Cos-mochim. Acta 49, 1797–1811.
Hildreth W. and Moorbath S. (1988) Crustal contributions to arc magmatism in the Andes of central Chile. Contrib. Mineral. Petrol.
98, 455– 489.
Hofmann A., Jochum K., Seufert M., and White M. (1986) Nb and Pb in oceanic basalts: New constraints on mantle evolution. Earth Planet. Sci. Lett. 33, 33– 45.
Holden P., Halliday A. N., and Stepens W. E. (1987) Neodymium and strontium isotope content of microdiorite enclaves points to mantle input to granitoid production. Nature 330, 53–56.
Jing Y. Z., Guo Y. J., and Ding C. X. (1995) Geochronology and origin of Saima alkaline rocks in Liaoning Province. Liaoning Geol. 4, 258 –271.
Kelemen P. B. (1995) Genesis of high Mg# andesites and the conti-nental crust. Contrib. Mineral. Petrol. 120, 1–19.
Kemp A. I. S. (2003) Plutonic boninite-like rocks in an anatectic setting: Tectonic implications for the Delamerian orogen in south-eastern Australia. Geology 31, 371–374.
Kemp A. I. S. and Hawkesworth C. J. (2003) Granitic perspectives on the generation and secular evolution of the continental crust. In Treatise on Geochemistry, Vol. 3, The Crust (ed. R. L. Rudnick), pp. 349 – 410. Elsevier Pergamon.
LBGMR (Liaoning Bureau of Geology and Mineral Resources) (1989) Regional Geology of Liaoning Province (in Chinese with English abstract). Geological Publishing House, Beijing.
Lin J. Q., Tan D. J., Chi X. G., Bi L. J., Xie C. F., and Xu W. L. (1992) Mesozoic Granites in Jiao-Liao Peninsula (in Chinese with English abstract). Science Press, Beijing.
Liu D. Y., Nutman A. P., Compston W., Wu J. S., and Shen Q. H. (1992) Remmants ofⱖ3800 Ma crust in the Chinese part of the Sino-Korean Craton. Geology 20, 339 –342.
Maas R., Nicholls I. A., and Legg C. (1997) Igneous and metamorphic encalves in the S-type Deddick Granodiorite, Lachlan Fold Belt, SE Australia: Petrographic, geochemical and Nd-Sr isotopic evidence for crustal melting and magma mixing. J. Petrol. 38, 815– 841. Martin H. (1999) The adakitic magmas: Modern analoge of Archean
granitoids. Lithos 46, 411– 429.
Menzies M. A., Fan W. M., and Zhang M. (1993) Palaeozoic and Cenozoic lithoprobe and the loss of⬎120 km of Archean litho-sphere, Sino-Korean craton, China. In Magmatic Processes and Plate Tectonics (eds. H. M. Prichard, T. Alabaster, N. B. W. Harris and C. R. Neary), pp. 71– 81. Spec. Publ. 76. Geol. Soc. Lond. Millar I. L., Willan R. C. R., Wareham C. D., and Boyce A. J. (2001)
The role of crustal and mantle sources in the genesis of granitoids of the Antarctic Peninsula and adjacent crustal rock. J. Geol. Soc. Lond. 158, 855– 867.
Patinˇo Douce A. E. (1999) What do experiments tell us about the relative contribution of crust and mantle to the origin of granitic magmas? In Understanding Granites: Integrating New and Clas-sical Techniques (eds. A. Catro, C. Fernández, and J. L. Vignersse), pp. 55–76. Spec. Publ. 168. Geol. Soc. Lond.
Pearce J. A., van der Laan S. R., Arculus R. J., Murton B. J., Ishii T., Peate D. W., and Parkinson I. J. (1992) Boninite and harzburgite from Leg 125 (Bonin-Mariana forearc): A case study of magma genesis during the initial stages of subduction. In Proceedings of the Ocean Drilling Program 94 (eds. P. Freyre and J. A. Pearce), pp. 304 –316. Ocean Drilling Program.
Roberts M. P. and Clemens J. (1993) Origin of high-potassium, calc-alkaline, I-type granitoids. Geology 21, 825– 828.
Rudnick R. L. and Fountain D. M. (1995) Nature and composition of the continental crust: A lower crustal perspective. Rev. Geophys.
33, 267–309.
Shimoda G., Tatsumi Y., Nohda S., Ishizaka K., and Jahn B. M. (1998) Setouchi high-Mg andesites revisited: Geochemical evidence for melting of subducted sediments. Earth Planet. Sci. Lett. 160, 479 – 492.
Snyder D. and Tait S. (1998) The imprint of basalt on the geochemistry of silicic magmas. Earth Planet. Sci. Lett. 160, 433– 445. Song B., Nutman A. P., Liu D. Y., and Wu J. S. (1996) 3800 –2500 Ma
crustal evolution in the Anshan area of Liaoning Province, north-eastern China. Precam. Res. 78, 79 –94.
Sun S. S. and McDonough W. F. (1989) Chemical and isotopic sys-tematics of oceanic basalts: Implications for mantle composition and processes. In Magmatism in the Ocean Basins (eds. A. D. Saunders and M. J. Norry), pp. 313–345. Spec. Publ. 42. Geol. Soc. Tatsumi Y. and Maruyama S. (1989) Boninites and high-Mg andesites: Tectonics and petrogenesis. In Boninites (ed. A. J. Crawford), pp. 50 –71. Unwin Hyman, London.
Turpin L., Cuney M., Friedrich H., Bouchez J. L., and Aubertin M. (1990) Meta-igneous origin of Hercynian peraluminous granites in NW French Massif Central: Implications for crustal history recon-struction. Contrib. Mineral. Petrol. 104, 163–172.
Vernon R. H. (1984) Microgranitoid enclaves: Globules of hybrid magma quenched in a plutonic environment. Nature 304, 438 – 439. Wan Y. S., Liu D. Y., Wu J. S., Zhang Z. Q., and Song B. (1998) The origin of Mesoarchean granitic rocks from Anshan-Benxi area: Constraints of Geochemistry and Nd isotope. Acta Petrol. Sinica
14, 278 –288.
Wan Y. S., Song B., Wu J. S., Liu D. Y., and Zhang Z. Q. (1999) Geochemical and Nd and Sr isotopic compositions of 3.8 Ga trondhjemitic rocks from the Anshan area and their significance. Acta Geol. Sinica 73, 25–36.
Wang H. Z. and Mo X. X. (1996) An outline of the tectonic evolution of China. Episodes 18, 6 –16.
Wu F. Y., Walker R. J., Ren X. W., Sun D. Y., and Zhou X. H. (2003) Osmium isotopic constraints on the age of lithospheric mantle beneath northeastern China. Chem. Geol. 196, 107–129. Yang J. H., Chung S. L., Zhai M. G., Zhou X. H. (2004) Geochemical
and Sr-Nd-Pb isotopic compositions of mafic dikes from the Jiaodong Peninsula, China: Evidence for vein-plus-peridotite melt-ing in the lithospheric mantle. Lithos 73, 145–160.
Yang J. H., Wu F. Y., Wilde S. A. (2003) A review of the geodynamic setting of large-scale Late Mesozoic gold mineralization in the
North China Craton: An association with lithospheric thinning. Ore Geol. Rev. 23, 125–152.
Zhang Q., Qian Q., Wang E., and Wang Y. (2001) Existence of East China plateau in mid-late Yanshan period: Implication from ada-kites. Sci. Geol. Sinica 36, 248 –255.
Zhao G. C., Wilde S. A., Cawood P. A., and Sun M. (2001) Archean blocks and their boundaries in the North China Craton: Lithologi-cal, geochemiLithologi-cal, structural and P-T path constraints and tectonic evolution. Precam. Res. 107, 45–73.
4483 Multiple sources for the origin of the Gudaoling granites
