1. Introduction
The South China Block comprises the Archaean–Proterozoic Yangtze Block and the Palaeoproterozoic to Mesoproterozoic Cathaysia (or Huanan) Block (Fig. 1; Li, Reference Li1997, Reference Li, Martin, Chung, Lo and Lee1998; Chen & Jahn, Reference Chen and Jahn1998; Gao, Lin & Qiu, Reference Gao, Lin and Qiu1999; Shen et al. Reference Shen, Ling, Li and Wang2000). It is a consequence of a series of intercontinental and intracontinental collisions and deformations that have occurred since Proterozoic times (e.g. Li & McCulloch, Reference Li and McCulloch1996; Charvet et al. Reference Charvet, Shu, Shi, Guo and Faure1996; Li, Reference Li, Martin, Chung, Lo and Lee1998). The early Palaeozoic orogen (Fig. 1) in south China occupies most or part of the southern provinces of mainland China and could have extended into the Korean peninsula and the Indochina Block (Guo et al. Reference Guo, Liu, Li, Xu and Ye2009; Li et al. Reference Li, Li, Wartho, Clark, Li, Zhang and Bao2010 and references therein). Its formation may be linked to interactions between the Cathaysia margin of the South China Block and the Australian–Indosinian margin of Gondwanaland that rotated clockwise during Ordovician–Silurian times (Li, Reference Li, Martin, Chung, Lo and Lee1998; Li & Powell, Reference Li and Powell2001). Therefore, knowledge of the orogen, including its nature, extent and tectonothermal evolution, is essential to understanding the formation of East Asia and to constraining early Palaeozoic global geodynamics and palaeography (Shu, Reference Shu2006; Li et al. Reference Li, Li, Wartho, Clark, Li, Zhang and Bao2010).
Figure 1. Tectonic sketch map of the early Palaeozoic Wuyun (Wuyi-Yunkai) Orogen within the South China Block (the combination of the Yangtze and Cathaysia blocks), with regional distribution of the felsic gneisses and (post-)Wuyun granites. Box indicates the location of Figure 2. Modified from Li (Reference Li, Martin, Chung, Lo and Lee1998), Zhang & Wu (Reference Zhang, Wu and Ma2002) and Sun (Reference Sun2006). Inset shows position of the map field area in China.
The early Palaeozoic orogenic event in south China has been referred to as the South China Caledonian Orogeny since its recognition in the 1960s (e.g. Huang, Reference Huang1960; Ren, Reference Ren1964; Guo, Yu & Shi, Reference Guo, Yu and Shi1965). In the western geological literature, however, the word ‘Caledonia’ is commonly employed to describe a geographically restricted orogeny that took place within, and on the borders of, the Iapetus Ocean (McKerrow, Niocaill & Dewey, Reference McKerrow, Niocaill and Dewey2000). To avoid confusion, we here use the term ‘the Wuyun (Wuyi-Yunkai) Orogeny (and for the orogen, the Wuyun Orogen)’, recommended by Li et al. (Reference Li, Li, Wartho, Clark, Li, Zhang and Bao2010), to encompass all the tectonic events affecting the South China Block in the early Palaeozoic (Late Ordovician? to Silurian).
Inliers of felsic gneissic rocks are common in the Wuyun Orogen. They are traditionally considered to be uplifted Precambrian (pre-Sinian c. 680 Ma) fragments of continental crust that represented the basement of the Cathaysia Block (e.g. Shu, Reference Shu2006). Geochronology has constrained the origin and evolution of the felsic gneisses with unprecedented resolution in some parts of the Yunkai massif and the Wuyishan metamorphic belt (Fig. 1; Table 1), the two best-studied areas with abundant occurrences of gneissic rocks. It is now realized that the gneiss exposed in the Wuyishan belt may mostly represent Proterozoic basement rocks which experienced extensive overprinting by high-grade metamorphism, migmatization and anatexis during the Wuyun Orogeny (Ye et al. Reference Ye, Lan, Chen and Edward1994; Shu et al. Reference Shu, Lu, Jia, Charvet and Faure1999; Yu et al. Reference Yu, Zhou, O'Reilly, Zhao, Griffin, Wang, Wang and Chen2005; Wan et al. Reference Wan, Liu, Xu, Zhuang, Song, Shi and Du2007; Liu et al. Reference Liu, Zhang, Zhou, Zhong, Zeng, Xiang, Jin, Lu and Li2008; Zeng et al. Reference Zeng, Zhang, Zhou, Zhong, Xiang, Liu, Jin, Lu and Li2008; Li et al. Reference Li, Li, Wartho, Clark, Li, Zhang and Bao2010). For most of the felsic gneisses from the Yunkai massif in western Guangdong, however, several recent high-precision geochronological studies indicate that they are actually of Caledonian-aged anatectic origin (Wang et al. Reference Wang, Fan, Zhao, Ji and Peng2007a; Peng et al. Reference Peng, Jin, Liu, Fu, He, Cai and Wang2006). This present situation highlights the necessity of re-evaluating the gneisses of the Baiyunshan Mountains region in central Guangdong. This region has excellent surface exposures of felsic gneisses and offers an important spatial connection between the Yunkai massif to the southwest and the Wuyishan belt to the northeast (Fig. 1). Nevertheless, there are few published age determinations for this region. Existing geochronological data include several Meso-Neoproterozoic and early Palaeozoic single-zircon U–Pb and Pb–Pb evaporation ages (Wu, Reference Wu1991; Liu & Zhuang, Reference Liu and Zhuang2003), as well as a middle Carboniferous K–Ar age of 312 Ma (GBGMR, 1988), and no analytical details were provided from these early studies.
Table 1. Summary of published high-precision geochronological data for the Wuyun Orogen
*Dating method: A – SHRIMP U–Pb; B – LA-ICPMS U–Pb; C – 40Ar–39Ar; D – TIMS U—Pb.
We present here for the first time reliable geochronological data for the granitic gneiss and some other metamorphic rocks exposed in the Baiyunshan region (corresponding geographically to Guangzhou, the largest city of southern China, and its surrounding counties townships). Based on composite Sensitive High Resolution Ion Microprobe (SHRIMP) U–Pb zircon and 40Ar–39Ar biotite-hornblende dating, this paper aims to provide constraints on the timing and origin of the high-grade metamorphic rocks and to unravel their tectonothermal history. The time-scale used in this study is GTS2004 (Gradstein et al. Reference Gradstein, Ogg, Smith, Bleeker and Lourens2004).
2. Geological setting
2.a. Regional setting
The early Palaeozoic Wuyun Orogen corresponds spatially to the area of the Neoproterozoic Nanhua failed rift system that developed within the South China Block (Li, Reference Li, Martin, Chung, Lo and Lee1998; Wang & Li, Reference Wang and Li2003). The former Nanhua rift represents a tectonically weak zone that was a domain of extremely active tectonics and magmatism during the Wuyun Orogeny (Wang et al. Reference Wang, Fan, Zhao, Ji and Peng2007a). This orogeny led to an angular unconformity between the folded early Palaeozoic basement and the Upper Devonian conglomerate and coarse sandstone in the entire south China area (Shu, Reference Shu2006).
The tectonic environment and geodynamic driving force for the Wuyun Orogeny is still poorly known. It has been claimed to be genetically related to oceanic subduction and subsequent continental or arc collisions or post-collision lithospheric delamination (Guo et al. Reference Guo, Shi, Lu, Ma and Dong1989; Hsü, Reference Hsü1994; Charvet et al. Reference Charvet, Shu, Shi, Guo and Faure1996; Peng et al. Reference Peng, Jin, Liu, Fu, He, Cai and Wang2006). The presently available geological data demonstrating that no evidence of a major early Palaeozoic ocean basin is evident in south China, however, tend to argue for an intracontinental tectonic regime for the Wuyun Orogeny (this does not preclude the possibility of the existence of a small oceanic basin; Li, Reference Li, Martin, Chung, Lo and Lee1998; Li et al. Reference Li, Li, Wartho, Clark, Li, Zhang and Bao2010; Wang & Li, Reference Wang and Li2003; Wang et al. Reference Wang, Fan, Zhao, Ji and Peng2007a). These data include: (1) the Wuyun Orogeny was accompanied by widespread granitoid intrusions of middle Palaeozoic age, coupled with the absence of coeval (arc) volcanic rocks (Shu, Reference Shu2006; Sun, Reference Sun2006); (2) geochemical and geochronological data presented in the literature reveal that the Wuyun granites are mostly anatectic products of Precambrian continental crust, with little evidence for material contribution from a subducting oceanic plate (e.g. Jahn, Zhou & Li, Reference Jahn, Zhou and Li1990; Chen & Jahn, Reference Chen and Jahn1998; Zeng et al. Reference Zeng, Zhang, Zhou, Zhong, Xiang, Liu, Jin, Lu and Li2008; Guo et al. Reference Guo, Liu, Li, Xu and Ye2009); and (3) sedimentary facies distribution shows that an apparent exchange of sedimentary sources occurred between the Yangtze and Cathaysia blocks on both sides of the failed Nanhua basin in Cambrian times when the basin was at its broadest (Liu & Xu, Reference Liu and Xu1994; Li, Reference Li, Martin, Chung, Lo and Lee1998; Li et al. Reference Li, Li, Wartho, Clark, Li, Zhang and Bao2010).
In southern China, the Wuyun Orogen also almost completely overlaps in space with a tectonomagmatic zone of Late Permian through Triassic age (‘Indosinian period’) (Fig. 1). This is partly due to the similar intracontinental tectonic setting of the Wuyun and Indosinian orogenies (Wang et al. Reference Wang, Fan, Zhao, Ji and Peng2007a,Reference Wang, Fan, Sun, Liang, Zhang and Pengb).
During the late Mesozoic (‘Yanshanian period’), the palaeo-Pacific plate was somehow subducted under the Eurasian plate (Jahn, Zhou & Li, Reference Jahn, Zhou and Li1990; Charvet, Lapierre & Yu, Reference Charvet, Lapierre and Yu1994; Lapierre et al. Reference Lapierre, Jahn, Charvet and Yu1997; Zhou & Li, Reference Zhou and Li2000; Li & Li, Reference Li and Li2007; Li et al. Reference Li, Li, Li, Liu, Yuan, Wei and Qi2007), resulting in a widespread continental margin magmatic zone, as much as > 1000 km wide and > 3000 km long, along the eastern margin of China. It is now widely accepted that the Mesozoic Yanshanian magmatism in eastern China is attributable to intracontinental lithospheric extension in response to asthenospheric mantle upwelling (e.g. Li, Reference Li2000; Wang et al. Reference Wang, Fan, Guo, Peng and Li2003c; Zhou et al. Reference Zhou, Sun, Shen, Shu and Niu2006). The broad Yanshanian magmatic zone overprints all previous events in the South China Block (Fig. 1).
Exposures of possible Wuyun-aged rocks within the South China Block are mainly restricted to the Yunkai massif, the Wuyishan Mountains metamorphic belt (in this work the belt is defined to cover the Mountains region and the Chencai Complex and adjacent regions because of age similarity in the metamorphic rocks and their protoliths; Li et al. Reference Li, Li, Wartho, Clark, Li, Zhang and Bao2010), the Baiyunshan Mountains region, the Zhuguangshan and Wugongshan areas, respectively, at the Jiangxi–Hunan–Guangdong and Jiangxi–Hunan provincial boundaries, all of which are located in the Cathaysia Block (Fig. 1). Additionally, in the Jiangnan Proterozoic Orogen, which is classically regarded as the southern edge of the Yangtze Block (Charvet et al. Reference Charvet, Shu, Shi, Guo and Faure1996), particularly in the Jiangxi and Hunan segments of the belt, deformation and magmatism associated with late early Palaeozoic compressive reworking has been well documented (Xu, Guo & Shi, Reference Xu, Guo and Shi1992; Deng & Zhang, Reference Deng and Zhang1996). Also, in recent years some Wuyun granites have been reported in the Laojunshan–Song Chay metamorphic dome on the border between southeastern Yunnan (China) and northern Vietnam (Roger et al. Reference Roger, Leloup, Jolivet, Lacassin, Trinh, Brunel and Seward2000; Yan, Zhou & Wang, Reference Yan, Zhou and Wang2006; Guo et al. Reference Guo, Liu, Li, Xu and Ye2009). The main sectors of the Wuyun Orogen are roughly delimited to the north by the Jiangshan–Shaoxing fault zone and to the southeast by the Zhenghe–Dapu fault zone (Fig. 1). An early Neoproterozoic ophiolitic complex zone occurs along the Jiangshan–Shaoxing fault; it is therefore thought that the latter marks the collisional suturing between the Jiangnan ancient arc and the Cathaysia Block (Jinning or Grenville Orogeny: Shui et al. Reference Shui, Xu, Liang and Qiu1986; Chen et al. Reference Chen, Foland, Xing, Xu and Zhou1991; Zhou & Zhu, Reference Zhou and Zhu1992; Li & McCulloch, Reference Li and McCulloch1996). The tectonics of the Zhenghe–Dapu fault zone, however, remains an interesting challenge. A suite of metavolcanic rocks with mafic and ultramafic small intrusions occurs along the fault zone, which has recently been dated at c. 1750 Ma (Sm–Nd method; Wang et al. Reference Wang, Ling, Zhou, Yang and Wang2003a), at variance with the viewpoint of Guo (1989) that these meta-igneous rocks are a Caledonian-aged ophiolite assemblage.
The structural and tectonostratigraphic region between the Jiangshan–Shaoxing and Zhenghe–Dapu fault zones is characterized by the presence of inliers of felsic gneisses which are commonly considered to be remnants of microcontinental fragments of the Cathaysia Block (Fig. 1). The gneissic rocks are mainly distributed along several coastal provinces comprising Zhejiang, Fujian and Guangdong, generally metamorphosed to the greenschist to upper amphibolite facies. Locally in the Gaozhou–Yunlu zone of the Yunkai massif and in the southern part of the Wuyishan Mountains, felsic gneisses also record low-pressure (4–5.2 kbar; Chen, Reference Chen1992; Zhou et al. Reference Zhou, You, Zhong and Han1994b) or high-pressure (~ 1.1 GPa; Yu et al. Reference Yu, Zhou, Zhao and Chen2003) granulite-facies metamorphic conditions. Widespread regional metamorphism leading to most of the felsic gneissic rocks has been ascribed principally to the early Palaeozoic Wuyun evolution (GBGMR, 1988; Chen, Reference Chen1992; Zhang & Wu, Reference Zhang, Wu and Ma2002; Yu et al. Reference Yu, Zhou, O'Reilly, Zhao, Griffin, Wang, Wang and Chen2005; Xu et al. Reference Xu, O'Reilly, Griffin, Deng and Pearson2005; Wan et al. Reference Wan, Liu, Xu, Zhuang, Song, Shi and Du2007; Li et al. Reference Li, Li, Wartho, Clark, Li, Zhang and Bao2010), although alternative interpretations suggest that the Wuyun-aged igneous rocks within the Yunkai massif might have been metamorphosed to gneissic rocks during Indosinian reactivation (Wang et al. Reference Wang, Fan, Zhao, Ji and Peng2007a).
2.b. Local geology
In the the eastern and northern parts of the Baiyunshan region (Figs 1, 2), metamorphic rocks composed of predominantly felsic gneiss with minor schist, quartzite, metasandstone and migmatite are widespread, whereas in other areas the geology is largely dominated by Cenozoic sediments. Metamorphic basement rocks are unconformably overlain by or in fault contact with unmetamorphosed Devonian to Cenozoic sedimentary cover rocks. The felsic gneiss of Baiyunshan, once considered to be Wuyun intrusive rocks (e.g. the rocks cropped out in the Maofengshan area and to the northwest of the Xintang–Luofushan fault) or Lower Palaeozoic metasediments (Geological Map of Guangzhou, 1968; Geological Map of Heyuan, 1968), is now interpreted as Sinian or Meso- and Neoproterozoic rocks (GBGMR, 1988; Liu & Zhuang, Reference Liu and Zhuang2003). Pervasive foliation in the gneisses is generally concordant with the country rock structures, defined principally by discontinuous biotite flakes (Fig. 3) and, as a whole, oriented in a northeasterly direction.
Figure 2. Geological map of the eastern and northeastern parts of the Baiyunshan region showing numbers and sites of the samples used in this study. Other parts of this region are predominantly filled by Cenozoic sediments and therefore omitted. The map is compiled from earlier published geological maps (Geological Map of Guangzhou, 1968; Geological Map of Heyuan, 1968). Note that the previously mapped metasedimentary rocks consist of felsic gneiss, schist, quartzite and metasandstone, but much of the ‘paragneiss’ may well be orthogneiss based on our geochronological result from sample BY004-1 (see text for details).
Figure 3. Examples of biotite-defined gneissic fabric developed within the Baiyunshan gneiss. (a) Field photograph taken at the site sampling BY004-1. (b) Crossed nicols photomicrograph of sample BY005-1. See Figure 2 for sample location.
The minerals included in gneisses are diagnostic of the amphibolite-facies mineral assemblage (plagioclase + quartz + biotite ± garnet ± hornblende ± K-feldspar). Thin discontinuous lenses of massive amphibolite are found to occur in parts of the Baiyunshan gneiss. Retrograde alteration shown by variably chloritized biotite and sericitized feldspar has been locally observed. In some locations, leucogranites occur more or less as < 0.5 m thick, concordant, or slightly discordant veins that are folded or undeformed. Petrographically, the leucogranites consist of quartz, plagioclase, K-feldspar, biotite and accessory minerals, and attest to anatexis during low-pressure–high-temperature metamorphism.
The Baiyunshan metamorphic region is intruded by extensive Mesozoic granitic plutons (dominantly early Yanshanian), and in the north it adjoins the EW-trending Fogang batholith in the southern segment of the famous Nanling Ranges granitoids. Zircons from one granite sample collected near the town of Luogang (Fig. 2) have been dated by LA-ICPMS at 153 ± 6 Ma (D. S. Yang, unpub. data). This age broadly resembles that of the Fogang I-type and Nankunshan A-type granites belonging to the Fogang batholith (Liu et al. Reference Liu, Chen, Wang, Zhang and Hu2005; Li et al. Reference Li, Li, Li, Liu, Yuan, Wei and Qi2007). The regional magmatic event of the same age has been hypothesized to be the result of foundering of an early Mesozoic subducted flat-slab beneath the SE China continent (Li et al. Reference Li, Li, Li, Liu, Yuan, Wei and Qi2007).
A significant normal fault system, herein informally named the Xintang–Luofushan fault zone, has been recognized in the area of investigation which merges easterly into the crustal-scale Heyuan (–Shaowu) fault and extends from Sanshui city, approximately 30 km west of Guangzhou, through Xintang to Boluo and Mt Luofushan (Figs 1, 2). The fault is thought to be an extensional detachment fault which was initiated in Late Jurassic times (Wu, Liu & Liao, Reference Wu, Liu and Liao2001); it strikes variably from ESE 100° to ENE 70–80° and dips moderately south or southwest. Along the fault trace, mylonites and mylonitic rocks derived from granitic and gneissic protoliths are exposed at or near Xintang, the Xiangang Reservoir and Boluo (Fig. 2), and the production of the mylonites could be due to the detachment motion (Wu, Liu & Liao, Reference Wu, Liu and Liao2001). The timing of ductile deformation on the shear zone is constrained at Xintang by 40Ar–39Ar dates of 217 Ma and 172 Ma obtained for biotite (Zou et al. Reference Zou, Qiu, Zhuang and Shao2001). However, our unpublished geochronological data (see above) suggest that all or part of the observed deformation should have occurred later than 155 Ma because of a cross-cutting relationship of the Yanshanian Luogang granite with the shear zone.
3. Analytical methods
3.a. Zircon U–Pb dating
Zircons were concentrated using standard gravimetric and magnetic separation techniques. Representative grains were then extracted by hand-picking under a binocular microscope. Zircons from the samples described below were cast, along with fragments of a standard zircon (TEMORA), in an epoxy mount. The mount was polished with diamond compound to reveal zircon mid-sections, and all sectioned zircons were documented with transmitted and reflected light micrographs as well as cathodoluminescence (CL) images to identify their internal structures (including growth zoning, recrystallization/overgrowth, alteration and cracks) and select spots for analysis. The in situ U–Pb ages were obtained with the use of the SHRIMP II ion microprobe at the Beijing SHRIMP Centre, Chinese Academy of Geological Sciences. Operating conditions and data processing procedures were similar to those described in detail by Williams (Reference Williams, McKibben, Shanks III and Ridley1998). Calibration concentrations of U and Th were based on analyses of zircon SL13. Isotopic compositions were calibrated by replicate analyses of the TEMORA standard (c. 417 Ma, 206Pb/238U = 0.06683; Black et al. Reference Black, Kamo, Allen, Aleinikoff, Davis, Korsch and Foudoulis2003). Final 204Pb-corrected ratios and ages are reported with 1σ analytical errors (68% confidence levels), as are the error ellipses shown in the concordia diagrams presented below. Nonetheless, calculated weighted mean ages in the text are presented at 95% (2σ) confidence limits.
3.b. 40Ar–39Ar dating
High-purity (> 99%) mineral separates of biotite and hornblende were obtained by crushing, washing in de-ionized water, sieving to 40 to 80 meshes, and hand-picking. The amount of biotite used for analysis was between 0.2 and 0.3 mg, whereas 5 mg was used for hornblende. The mineral concentrates were wrapped in aluminium foil packets, encapsulated in sealed quartz vials, and irradiated in a nuclear reactor at the China Institute of Atomic Energy in Beijing. Variations in the flux of neutrons along the length of the irradiation assembly were monitored with mineral standards DRA1 (sanidine with age of 25.26 ± 0.07 Ma) and ZBH25 (biotite with age of 132.9 ± 1.3 Ma). The two standards are very homogeneous and have excellent reproducibility (Wijbrans et al. Reference Wijbrans, Pringle, Koppers and Scheveers1995; Sang et al. Reference Sang, Wang, He, Wang, Yang and Zhu2006).
Following fast neutron irradiation, argon isotopes were measured at the GIG-CAS Ar/Ar Isotope Laboratory. Mineral separates were step-heated in 12 to 18 steps at incrementally higher powers in the defocused beam of a 50W CO2 laser (MIR10, ®New Wave Research, Inc.). The gas evolved from each step was analysed by a high sensitivity GV5400 mass spectrometer equipped with an ion-counting electron multiplier. Details of analytical and data reduction procedures can be found in Qiu & Wijbrans (Reference Qiu and Wijbrans2006) and Qiu (Reference Qiu2006). Ages for individual temperature steps were calculated by assuming an initial atmospheric 40Ar/36Ar ratio equal to 295.5 (Steiger & Jäger, Reference Steiger and Jäger1977). Errors quoted on the ages and Ar isotope ratios are at the 2σ levels. The 40Ar–39Ar dating results were treated and calculated using the ArArCALC software package (http://earthref.org/tools/ararcalc/index.html) (Koppers, Reference Koppers2002).
4. SHRIMP U–Pb zircon geochronology
Mineralogically and structurally similar gneissic rocks occur in large areas of Baiyunshan. In this study, U–Pb SHRIMP dating of zircon crystals was carried out on three felsic gneiss samples (BY005-1, BY022-3 and BY004-1) which are foliated. Sample locations are indicated in Figure 2. The dated rocks include ortho- and paragneisses previously mapped and are typical for the gneiss within the Baiyunshan region. Combined with zircon CL images, a number of different zircon fractions from these samples were targeted to constrain the origin, age and tectonothermal overprinting. Efforts were made to avoid analysing areas with cracks and inclusions. Figure 4 shows morphologies and CL zoning patterns of some representative zircon crystals, as well as locations of the analytical pits and the resulting apparent ages. Isotopic data for each sample are presented in online Appendix Table A1 at http://journals.cambridge.org/geo and plotted in concordia diagrams (Fig. 5a–d). The results of the age determinations, along with 40Ar–39Ar age data (see next Section), are summarized in Table 2. In this investigation, zircons younger than c. 1.2 Ga were dated using the 206Pb/238U age, whereas older ones used the 207Pb/206Pb age.
Table 2. Summary of U–Pb and 40Ar–39Ar age data obtained in this study
*Abbreviations of minerals dated: Hb – hornblende, Bi – biotite.
Figure 4. Representative cathodoluminescence (CL) images of zircon grains from dated felsic gneiss samples. Small circles point to analytical pits, in all cases about a 30 μm diameter; grain and spot numbers, and resulting 206Pb/238U ages (±1σ uncertainties) are indicated and correspond to those listed in online Appendix Table A1 at http://journals.cambridge.org/geo. All scale bars equal 100 μm. Note rounded cores with or without platy zoning that yield old ages in samples BY004-1 and BY022-3.
Figure 5. Concordia diagrams showing U–Pb SHRIMP data (online Appendix Table A1) for zircons from dated felsic gneiss samples (a) BY004-1; (b) BY005-1 (Maofengshan pluton) and (c, d) BY022-3. Data are shown as 1σ error ellipses, but calculated mean ages are ±2σ.
4.a. Sample BY004-1: granitic biotite gneiss
Sample BY004-1 is from an abandoned quarry of felsic gneiss in the northwestern portion of the study area, about 15 km northeast of Baiyunshan Park (23°15′04″N/113°22′09″E) (Fig. 2). It consists of approximately 46% plagioclase (An35–40), 35% quartz, 18% biotite and 1% K-feldspar. Foliation of the investigated gneiss is moderate to weak, defined principally by discontinuous, subparallel alignment of biotite flakes (Fig. 3a). Quartz occasionally shows undulatory extinction and sutured margins or is polycrystalline in form, features which suggest post-crystallization deformation.
The sample yielded a single population of elongate (2:1–4:1), clear, colourless to slightly brown, euhedral zircon, characteristic of magmatic origin (Hanchar & Miller, Reference Hanchar and Miller1993; Da Silva et al. Reference Da Silva, Hartmann, Mcnaughton and Fletcher2000). The zircons range in size between 100 and 450 μm. Relict cores with or without planar zoning are often present (Fig. 4a). Most zircons are overgrown by thin white rims that were usually too small to allow analysis and clearly identify their origin; however, scanned CL imaging presents a grain with a truncated concentric zoning pattern (point 8.1 in Fig. 4a), indicative of the metamorphic origin of this and possibly other thin white rims (Hoskin & Black, Reference Hoskin and Black2000).
A total of 18 SHRIMP analyses were conducted on 17 different zircons, including one white metamorphic rim and three inherited cores. Fourteen analyses for magmatic zircons showing oscillatory zoning cluster on concordia (Fig. 5a); all of these analyses define a well-constrained mean 206Pb/238U age of 446 ± 7 Ma (mean square of weighted deviates (MSWD) = 1.0) that is considered the best estimate for the timing of crystallization of the corresponding intrusion. Three inherited cores (points 3.1, 5.1 and 13.1) plot on or very near concordia and yield Proterozoic ages ranging from 1017 to 1030 Ma. The single rim analysis (point 8.1) gave a 206Pb/238U age of 212 ± 12 Ma and may reflect new zircon growth during a later tectonothermal event. It is noteworthy that the analysis of the rim suffers from poor precision because of its extremely high common Pb content (ƒ206 = 12) and low U (36 ppm) and Th (0.003 ppm) contents (online Appendix Table A1), although it plots on concordia.
4.b. Sample BY005-1: Maofengshan granitic biotite gneiss
Sample BY005-1 was taken from the northern part of the Maofengshan Park area of Guangzhou (23°18′23″N/113°27′23″E) (Fig. 2). This gneiss is composed of approximately 45% plagioclase (An35–40), 32% quartz, 5% K-feldspar, 13% biotite and 5% sericite, showing an alternation of coarse- and fine-grained mineral bands (Fig. 3b). Biotite is mostly subhedral and fine-grained and sometimes variably chloritized. It makes up weakly orientated flakes, associated with fine-grained feldspar and quartz and defining the main foliation of the rock. Such a microtexture presumably reflects syntectonic recrystallization under the condition of anistropic stress (Kornprobst, Reference Kornprobst2002).
This sample contains predominantly elongate zircon grains with length to width ratios between 2:1 and 4:1 (Fig. 4b); approximately 2% have ratios of 3:2 or are near equant. The zircons range in size between 100 and 350 μm, and are clear and colourless to slightly brown. The CL images reveal concentrically zoned zircon domains of igneous origin, with or usually without relict cores. Many of these oscillatory zoned crystals have a thin white rim (of possibly metamorphic origin) that is inaccessible to ion microprobe spot analysis due to its small scale.
Seventeen analyses were conducted on 17 separate zircon grains with concentric, oscillatory zoning, which is indicative of magmatic crystallization. The analyses are all concordant and have 206Pb/238U ratios that are in good agreement within analytical precision (Fig. 5b); the analyses yield a weighted average age of 453.5 ± 7.8 Ma (MSWD = 1.4). We interpret this as the crystallization age of the granitic protolith at Maofengshan.
4.c. Sample BY022-3: granitic biotite gneiss
Sample BY022-3 was collected from an outcrop exposed between Henghe and Ping'an townships in the northeastern corner of the study area (Fig. 2; 23°19′22″N/114°11′39″E), along the northern side of the road to the village of Xiadong. The sample analysed here consists of approximately 35% quartz, 36% plagioclase (An35–40), 10% K-feldspar, 18% biotite, and 1% sericite. Foliation is defined by biotite that occurs as isolated equant flakes or as partly interconnected networks of flakes. As an effect of retrograde reactions, feldspar is partly replaced by sericite, and biotite by chlorite.
Zircons in the sample are mainly colourless to light brown, but a small proportion of darker brown grains are also present. The size of the crystals varies from less than 100 μm to nearly 420 μm, with the majority of the crystals less than 350 μm in the longest dimension (Fig. 4c). The zircon population of sample BY022-3 is relatively heterogeneous: euhedral, short-prismatic to elongate crystals with relatively sharp edges are most common, but subhedral or anhedral, short-prismatic to equidimensional grains also occur. The CL images show that the euhedral, short-prismatic to elongated grains usually record concentric zoning indicative of igneous growth, whereas subhedral or anhedral grains may be zircons with (e.g. points 6.1 and 15.1, Fig. 4c) or without (e.g. point 9.1, Fig. 4c) a core–rim relationship. Relict cores are more often present, compared to zircons from samples BY004-1 and BY005-1. Very old components (point 1.1, online Appendix Table A1), despite their rarity, are also discovered as unzoned subround whole grains.
Sixteen analyses were performed on 16 zircon grains and include one detrital grain. The complexity of the zircon population is reflected in the ages obtained (Fig. 5c, d; Table 2). Four concordant or near-concordant core analyses (points 6.1, 7.1, 15.1, 16.1) yield Proterozoic ages of 751 to 1190 Ma. Analysis (1.1) plots well away from the concordia line and records a 207Pb/206Pb age of 3027 ± 40 Ma, which is the oldest age reported in this region and can be regarded as its minimum age. Of the other 11 analyses, eight were determined on oscillatory-zoned domains of euhedral crystals. They defined a well-constrained concordant mean 206Pb/238U age of 439 ± 9 Ma (MSWD = 1.7) (see inset in Fig. 5d), which is regarded as the best estimate for the timing of crystallization of the granitic protolith. The three remaining analyses were determined on bright overgrowths of unzoned zircons (points 8.1, 9.1 and 11.1) that originated possibly from fluid activity during magma crystallization (?), with concordant ages ranging from 401 to 479 Ma. Although similar in Th/U (0.12–0.68) and age to normal magmatic zircons, the three analyses are excluded from the calculation of the weighted mean due to their uncertain significance.
5. 40Ar–39Ar thermochronology
K-bearing mineral concentrates from seven rock samples collected from the study area, which include six biotite fractions and one hornblende separate, were used for 40Ar–39Ar dating. Hornblende (BY017-22) was derived from a lens of massive amphibolite, whereas biotite samples were obtained from a variety of rock types, including felsic gneiss, micaschist and leucosome. The general sample locations within the Baiyunshan region are shown in Figure 2. It is of note that among the biotite samples, BY005-1 and BY022-3 were extracted from the same hand-specimen as those for U–Pb analyses, and BY017-19 and BY017-25 came from the same quarry site as the analysed hornblende.
The 40Ar/39Ar dating results and full analytical data are listed in Table 2 and Appendix Table A2, respectively. Apparent age spectra and, where appropriate, corresponding isotope correlation diagrams (inverse isochrons) are illustrated in Figures 6 and 7 and Appendix Figure A1. The appendices are available as supplementary material online at http://journals.cambridge.org/geo. In the case of hornblende 40Ar/39Ar analysis, the K/Ca plot is also included to assist in the interpretation of the spectra.
Figure 6. 40Ar–39Ar age spectrum and inverse isochron correlation diagram for hornblende from an amphibolite (BY017-22) collected at Nanxiangshan, showing calculated plateau and isochron ages. Apparent K/Ca spectrum is also presented for this sample. Steps used in the weighted average plateau age are indicated. The steps included in the isochron calculations are the same as those in plateau ages.
Figure 7. 40Ar–39Ar age spectrum for biotite separated from a mica schist (sample BY026-6) of the Baiyunshan region.
The standard criterion for identification of plateaux (that is, undisturbed portions of the 40Ar/39Ar spectra) is commonly the existence of at least three contiguous steps with concordant ages which contain a significant proportion (> 50%) of the 39Ar released (Dalrymple & Lanphere, Reference Dalrymple and Lanphere1974; Dallmeyer & Lecorche, Reference Dallmeyer and Lecorche1990; Lee et al. Reference Lee, Onstott, Cashman, Cumbest and Johnson1991; McDougall & Harrison, Reference McDougall and Harrison1999). However, a final criterion for defining a ‘true plateau’ age is based on concordance of inverse-isochron ages defined by isochronous (that is, low MSWD) data and plateau ages (Dalrymple & Lanphere, Reference Dalrymple and Lanphere1974). In this study, the isochron ages are preferred over the plateau ages, since they combine quantitative estimates of analytical precision plus the internal disturbance of the sample and they make no assumption of a present-day 40Ar/36Ar ratio concerning the trapped argon component (e.g. Singer & Pringle, Reference Singer and Pringle1996).
5.a. Hornblende
Hornblende from one sample of amphibolite (BY017-22) occurring within the plagioclase–quartz–biotite gneiss in the Nanxiangshan Hill displays discordant apparent age spectra (Fig. 6). The low-temperature (or low-laser-power) gas fractions of the spectrum (steps 1–4), comprising < 9.5% of the 39Ar released, record considerable age variation from c. 1903 Ma to < 200 Ma. This situation could occur because of excess 40Ar, phase mixing, and/or 39Ar recoil loss from grain boundaries during irradiation, which produces a high 40Ar/39Ar ratio. In view of the large fluctuation in apparent K/Ca ratios in the low temperature steps (Fig. 6), we conclude that experimental evolution of argon stemmed dominantly from compositionally distinct, relatively Ar-non-retentive phases. Such phases may be represented by (1) very minor, optically undetectable mineralogical contaminants in the hornblende concentrates; (2) petrographically unresolvable exsolution or compositional zonation within constituent hornblende grains (e.g. Siebel, Henjes-Kunst & Rhede, Reference Siebel, Henjes-Kunst and Rhede1998); (3) minor chloritic replacement of hornblende; and/or (4) intracrystalline fluid inclusions (Kelley, Reference Kelley2002; Harrison, Heizler & Lovera, Reference Harrison, Heizler and Lovera1993; Harrison et al. Reference Harrison, Heizler, Lovera, Chen and Grove1994).
For intermediate- and high-temperature gas fractions that contain 12 of 18 steps and 78.5% of the 39Ar released, relatively consistent intrasample apparent 40Ar–39Ar ages are obtained, which define a pseudo-plateau date of 172 ± 3 Ma (see below). Because there is little variation in corresponding apparent K/Ca ratios (Fig. 6), the experimental evolution of this part of the extracted gas fractions probably resulted from compositionally uniform sites. These gas fractions generate a 36Ar/40Ar v. 39Ar/40Ar inverse isochron age of 155 ± 4 Ma (MSWD = 2.2) with a poorly constrained initial 40Ar/36Ar ratio of 494 ± 48. The initial 40Ar/36Ar ratio is higher than the present-day atmospheric value (~ 295.5), indicating that small amounts of excess argon were released between steps 7 and 18 and that therefore the apparent ages (calculated assuming 40Ar/36Ar = 295.5) and the resultant plateau age in Figure 6 are not correct. Consequently, only the c. 155 Ma isochron age is here considered a geologically significant age. It is interpreted to date the last cooling through temperatures required for intracrystalline retention of argon in hornblende (~ 500 to 550 °C; e.g. McDougall & Harrison, Reference McDougall and Harrison1999).
5.b. Biotite
Sample BY005-1 was collected from the Maofengshan orthogneiss. Biotite from this sample gave a 12-step plateau age of 148.76 ± 0.54 Ma (Fig. A1, online Appendix at http://www.cambridge.org/journals/geo). The same temperature steps define an inverse isochron age of 150.3 ± 4.3 Ma (MSWD = 1.7) with an 40Ar/36Ar intercept of 255.7 ± 115.8. The ischron age overlaps with the plateau age, taking uncertainties into account, and thus dates the last Ar isotopic closure of the biotite.
Incremental heating of biotite concentrates from three felsic gneisses (BY022-3, BY023-1 and BY017-25) and one leucogranite (BY017-19, locally containing abundant aggregates of biotite) yielded simple apparent age spectra with plateau ages varying between 93.2 ± 0.3 and 97.7 ± 0.4 Ma (Fig. A1, online Appendix). Inverse isochron plots for the biotite analyses result in well-defined ages of 94.0 ± 0.14 Ma (40Ar/36Ar intercept = 282.4 ± 8.3; MSWD = 0.4), 93.6 ± 0.5 Ma (40Ar/36Ar intercept = 281.5 ± 16.6; MSWD = 0.2), 97.5 ± 0.5 Ma (40Ar/36Ar intercept = 304 ± 24; MSWD = 1.5) and 96.3 ± 1.2 Ma (40Ar/36Ar intercept = 257 ± 150; MSWD = 2.7), which are consistent with their respective plateau ages and therefore date the last Ar isotopic closure of the biotite.
Biotite BY026-6 was separated from one mylonitic micaschist from the eastern study area (see Fig. 2 for sample location). A corresponding incremental heating experiment yielded a rising staircase spectrum with no age significance, in which apparent ages increase from 128 to 150 Ma with increasing temperature (Fig. 7; Table A2, online Appendix at http://www.cambridge.org/journals/geo). An insufficient spread in 39Ar/40Ar ratios precluded isochron analysis for this sample. Even so, the biotite spectrum is still informative and is in agreement with the diffusional loss of argon (reheating) at ≤ 128 Ma from a biotite that was most likely isotopically reset at ≥ 150 Ma (this time corresponds to a regional tectonothermal event; see Section 6.c). Petrographic observations indicate that the phenocrystic biotite in mylonitic micaschist is strongly deformed and altered to sericite. The deformation and alteration were processes probably responsible for the Ar loss or partial resetting.
6. Discussion
6.a. Intrusion ages of the protoliths of the Baiyunshan gneiss and their implications for the Wuyun Orogeny
U–Pb dating and CL imaging suggest that the zircons in three gneisses from the Baiyunshan Mountains largely formed under magmatic conditions in Late Ordovician to Early Silurian times (c. 456–439 Ma, Table 2). This implies a Wuyun igneous origin for the protoliths of most of the felsic gneisses in the region of study. Consequently, the Baiyunshan gneiss cannot represent basement rocks of the Cathaysia Block as previously thought. Our geochronological data contrast with presently widely used geological maps where most of the Baiyunshan gneisses were identified as Sinian or Meso- and Neoproterozoic rocks (Liu & Zhuang, Reference Liu and Zhuang2003; GBGMR, 1988), but coincides partly with data documented in some earlier literature (Geological Map of Guangzhou, 1968; Geological Map of Heyuan, 1968). On geological maps published in 1968, although the gneiss units (corresponding to BY004-1) are mapped as pre-Devonian metasediments, those (corresponding to BY005-1 and BY022-3) at Maofengshan and some other areas such as Henghe, Ping'an and Boluo are shown as Wuyun intrusions (Fig. 2).
The zircon population in the Maofengshan orthogneiss (BY005-1) constrains the emplacement age of the magmatic protolith to 455.8 ± 6.4 Ma; the granitic orthogneiss thus constitutes the oldest known igneous unit confirmed from Baiyunshan. This emplacement age agrees within error with the age determined for sample BY004-1 of 446 ± 7 Ma; the latter age is also indistinguishable from the zircon formation age (439 ± 9 Ma) for the BY022-3 gneiss from the eastern study area. Therefore, granitic protoliths of the Baiyunshan gneisses can be interpreted to form within a single, continuous, protracted tectonothermal event lasting about 17 Ma. It should be emphasized that the 439 ± 9 Ma age for BY022-3 is slightly younger than that for the foliated Maofengshan granite. Interestingly, BY022-3 is also the sample that contains abundant detrital zircons or inherited cores relative to the other samples. We thus consider it most likely that the 439 ± 9 Ma time corresponds to the waning phase of the Wuyun tectonothermal event.
Our new age constraint for Wuyun magmatism in the Baiyunshan region (the middle segment of the Wuyun Orogen), combined with available geochronological data from southwestern and northeastern sectors of the orogen (Table 1; Fig. 1), indicates that the ages of large amounts of the Wuyun granitoids cluster at c. 460–420 Ma in the entire Wuyun Orogen (Table 1; Li et al. Reference Li, Tatsumoto, Premo and Gui1989; Li, Reference Li1994; Lou et al. Reference Lou, Shen, Wang, Shu, Wu, Zhang and Yu2005; Wang et al. Reference Wang, Fan, Zhao, Ji and Peng2007a; Zhou et al. Reference Zhou, You, Zhong and Han1994a; Peng et al. Reference Peng, Jin, Liu, Fu, He, Cai and Wang2006; Li et al. Reference Li, Li, Wartho, Clark, Li, Zhang and Bao2010; this study). This broad age cluster can be reasonably interpreted to reflect orogen-wide magmatism during this period. Most recent studies have demonstrated that metamorphic overprinting was widespread during the same time interval (Yu et al. Reference Yu, Zhou, O'Reilly, Zhao, Griffin, Wang, Wang and Chen2005; Xu et al. Reference Xu, O'Reilly, Griffin, Deng and Pearson2005; Wan et al. Reference Wan, Liu, Xu, Zhuang, Song, Shi and Du2007; Li et al. Reference Li, Li, Wartho, Clark, Li, Zhang and Bao2010). In addition, it is noted that a transpressive regime initiated its main activity by c. 420 Ma in the Wuyi Mountains area (Shu et al. Reference Shu, Lu, Jia, Charvet and Faure1999) and the Jiangnan belt (see Charvet et al. Reference Charvet, Shu, Shi, Guo and Faure1996). Therefore, we conclude that the 460–420 Ma magmatism is for the most part syn-orogenic relative to the Wuyun Orogeny.
Syn- to late orogenic magmatic activity in the western segment of the Wuyun Orogen, extending to 402 Ma in the Laojunshan–Song Chay metamorphic dome, appears to terminate later than in its eastern section (Table 1; Li et al. Reference Li, Li, Wartho, Clark, Li, Zhang and Bao2010). On the orogen scale, however, it is unknown whether Wuyun magmatism was continuous or was composed of discrete magmatic pulses. We want to emphasize that although many ages have been obtained from the Wuyun Orogen, details of the history of magmatic activity are far from being resolved. Further work is needed to refine the timing, duration and migration of magmatic activity of the Wuyun Orogen.
6.b. Sources of Wuyun magmatism
Relict cores in zircon from the Baiyunshan gneiss are abundant and define two clusters of (near-) concordant ages, one with latest Mesoproterozoic ages, including 1030, 1023, 1189 and 1017 Ma, the other with middle Neoproterozoic ages of 762, 751 and 809 Ma. When pooled together, the old age data (excluding 1189 Ma) give a well-constrained weighted 206Pb/238U date of 1023 ± 17 Ma (MSWD = 0.21), whereas the younger ones yield a weighted mean age of 772 ± 28 Ma (MSWD = 1.09). Moreover, one of the zircon grains (Table A1, online Appendix at http://www.cambridge.org/journals/geo, analysis 1.1 for BY022-3) yields a discordant SHRIMP analysis, showing that it is a detrital component with minimum age of 3027 Ma. These older zircon components are apparently exotic relative to the Wuyun magmatic stage. They may reflect an anatectic origin of the crustal source rock or country rock assimilation during magma ascent. None the less, the fact that the number of inherited cores is generally large does not support the latter possibility; the amount of assimilation required would demand very high extents of superheating of the magma.
Inherited zircons represent earlier igneous and/or metamorphic phases, the latter of which may be due to reactions (Fraser, Ellis & Eggins, Reference Fraser, Ellis and Eggins1997) or recrystallization (Hoskin & Black, Reference Hoskin and Black2000). Associated grains dated in this study have variable and high Th/U (0.05–1.34), characteristics that are not typical of zircon grown during prograde metamorphism (online Appendix Table A1; Rubatto, Reference Rubatto2002; Hoskin & Schaltegger, Reference Hoskin and Schaltegger2003). We note that the 1189 Ma to 1017 Ma age range of inherited zircons corresponds to the time of the Sibao Orogeny (c. 1.1–0.9 Ga) that formed the South China Block (e.g. Li et al. Reference Li, Li, Zhou and Kinny2002, Reference Li, Li, Wingate, Chung, Liu, Lin and Li2006; Ye et al. Reference Ye, Li, Li, Liu and Li2007), and the youngest age cluster at c. 772 Ma of inherited zircons is consistent with the period of subsequent intracontinental rifting (830–600 Ma) in the South China Block (e.g. Li et al. Reference Li, Li, Kinny and Wang1999, Reference Li, Li, Kinny, Wang, Zhang and Zhou2003; Wang et al. Reference Wang, Li, Duan, Liu, Song, Li and Gao2003b; Zhou et al. Reference Zhou, Li, Ge and Li2007). Thus, despite limited data from the present study, the Baiyunshan orthogneiss can be inferred to have sampled a crustal basement containing significant igneous (or recycled?) components linked to the Rodinia amalgamation and break-up, up to now clearly demonstrated in some other parts of the Wuyi Orogen (e.g. Xu et al. Reference Xu, O'Reilly, Griffin, Deng and Pearson2005; Wan et al. Reference Wan, Liu, Xu, Zhuang, Song, Shi and Du2007). This inference is compatible with the intracontinental nature of the Wuyun Orogeny (e.g. Li et al. Reference Li, Li, Wartho, Clark, Li, Zhang and Bao2010).
6.c. Late and post-Wuyun tectonothermal history
In the present state of knowledge, it is unclear whether or not the granitic protoliths of the Baiyunshan gneiss underwent a late-Wuyun history of metamorphism and deformation. However, as regional metamorphism and deformation probably continued until c. 370 Ma in the Wuyun Orogen (Table 1), we cannot rule out the possibility that solidified Wuyun igneous rocks were syntectonically metamorphosed. By contrast, the Indosinian metamorphic history of the studied felsic gneisses can be constrained by zircon geochronology. Despite the general occurrence of white thin rims that are characteristic of metamorphic zircons (Hoskin & Black, Reference Hoskin and Black2000), these could not be analysed in most cases. One exception is in spot 8.1 of sample BY004-1 (Fig. 4a), which gave a 206Pb/238U age of 212 ± 12 Ma that may represent the timing of new zircon growth during a later Indosinian tectonothermal event. It therefore remains possible that some of the thinner CL-white rims are also Indosinian in age, as found in (pre-)Wuyun crystalline rocks in the Yunkai Massif and Wuyishan metamorphic belt (Wang et al. Reference Wang, Fan, Zhao, Ji and Peng2007a; Yu et al. Reference Yu, Zhou, O'Reilly, Zhao, Griffin, Wang, Wang and Chen2005, Reference Yu, Wang, Zhou, Jiang, Wang, Xu and Qiu2006; Wan et al. Reference Wan, Liu, Xu, Zhuang, Song, Shi and Du2007). Thus, the present metamorphic age of 212 Ma seems to testify further to a significant influence of the Indosinian tectonothermal event on the continental crust of south China.
Within analytical uncertainty, the biotite 40Ar–39Ar age (150 ± 4 Ma, BY005-1) of the Maofengshan pluton overlaps with that of hornblende (155 ± 4 Ma, BY017-22) from an amphibolite lens in Nanxiangshan Hill (Table 2). The 40Ar–39Ar age of 155–150 Ma is consistent with the precise zircon U–Pb dates for widely developed early Yanshanian granitic intrusions in Baiyunshan and its neighbouring areas (Figs 1, 2; Liu et al. Reference Liu, Chen, Wang, Zhang and Hu2005; Li et al. Reference Li, Li, Li, Liu, Yuan, Wei and Qi2007; D. S. Yang, unpub. data). Hence, we conclude that the regional magmatic heat in Late Jurassic times caused widespread thermal resetting of the K–Ar isotope system within older high-grade rocks in the Baiyunshan region. This conclusion is also in accord with a 150 Ma apparent age in the last heating step of biotite BY026-6 from mylonitic micaschist that possibly reflects a relict record of a ≥ 150 Ma tectonothermal event, as noted earlier (Fig. 7).
Four younger but similar biotite 40Ar–39Ar ages of 94.0 ± 0.14, 93.6 ± 0.5, 97.5 ± 0.5 and 96.3 ± 1.2 Ma have been obtained from gneisses BY022-3, BY023-1 and BY017-25 and leucosome BY017-19 (the latter two samples were from the same quarry site at Nangxiangshan as the hornblende BY017-22) (Fig. 2; Table 2). Regionally, 90–100 Ma volcanic-intrusive rocks are widespread in southeastern China (e.g. Li, Reference Li2000; Zhou & Li, Reference Zhou and Li2000). They have been identified confidently in western and northern Guangdong (e.g. GBGMR, 1988; Geng et al. Reference Geng, Xu, O'Reilly, Zhao and Sun2006), but have seldom been investigated in central Guangdong. It remains unclear whether the 90–100 Ma magmatism has significantly affected the Baiyunshan region. Thus, the younger 40Ar–39Ar age group can be attributed to either cooling through the biotite closure temperature of approximately 300 to 350 °C (McDougall & Harrison, Reference McDougall and Harrison1999) or a second resetting of biotite Ar isotopes at c. 94–98 Ma due to contemporaneous magmatic activity. Note that in the latter case, the argon system in hornblende cannot have been reset, given the large age difference between hornblende and biotites from Nangxiangshan.
Our present 40Ar–39Ar data reveal that the orthogneiss unit at Maofengshan had already cooled down to about 300 °C at c. 150 Ma, but the other studied metamorphic rocks were still at temperatures above ~ 300 °C until about 90–100 Ma, or were sufficiently heated to reset Ar–Ar ages in biotites at this time, with subsequent rapid cooling to below ~ 300 °C. The period from 155 Ma to 90 Ma coincides with Late Jurassic to Tertiary extensional shearing along the Xintang–Luofushan detachment fault system (Fig. 2; Wu, Liu & Liao, Reference Wu, Liu and Liao2001), implying extensional exhumation of mid-crustal amphibolite-facies rocks during this period. On the basis of the present age data and previous structural investigation (Wu, Liu & Liao, Reference Wu, Liu and Liao2001), we suggest that the Maofengshan orthogneiss was exhumed to 8 to 10 >km crustal levels beneath Baiyunshan at c. 150 Ma, whereas the eastward components of the gneissic rock masses might have passed upward through the same crustal depth synchronously or later (by c. 94 Ma).
Exhumation of middle crustal-level rocks in Baiyunshan since c. 155 Ma is roughly coeval with exhumation of gneissic rocks from elsewhere in the Cathaysia Block. To the southwest, a recent fission track analysis of zircon and apatite from the Yunkai terrane shows that the old crystalline basements have been exhumed largely owing to orogenic uplift since Early Cretaceous (Li et al. Reference Li, Wang, Tan and Peng2005). Based on the P–T–t path of isothermal decompression, exhumation of the granulite-facies metamorphic rocks from the Wuyi Mountains area in the eastern Nanling Ranges is considered to be related to Mesozoic extension (Yu et al. Reference Yu, Zhou, Zhao and Chen2003). This synchroneity reflects the regional Yanshanian tectonic regime characterized by intracontinental lithospheric extension (Li, Reference Li2000; Wang et al. Reference Wang, Fan, Guo, Peng and Li2003c).
7. Conclusions
Our SHRIMP zircon U–Pb geochronological data indicate that most of the felsic gneisses from the Baiyunshan region cannot represent basements rocks of the Cathaysia Block as previously thought, because of their Late Ordovician–Early Silurian (454–439 Ma) ages, interpreted as the crystallization age of magmatic zircon. Felsic protoliths of the investigated gneisses may have formed within a single, continuous, protracted tectonothermal event lasting about 17 Ma. An integration of available age data for the Wuyun events reflects the emergence of orogen-wide magmatism that could be syn-orogenic and have occurred mainly between 460 and 420 Ma. Inherited zircon studies reveal that the protolith of the Baiyunshan orthogneiss was derived from a crustal basement containing significant igneous or recycled components linked to the Rodinia amalgamation and break-up. A SHRIMP date of 212 Ma from a metamorphic rim of zircon provides evidence for overprinting of a Indosinian tectonothermal event on the Baiyunshan gneiss.
Incremental heating experiments of biotite (6) and hornblende (1) separates from a variety of metamorphic rocks yielded two distinct 40Ar–39Ar age groups: 150–155 and 94–98 Ma. The older ages most likely reflect Late Jurassic magmatism-induced thermal resetting of the K–Ar system, whereas the young age records cooling through the biotite closure temperature of about 300–350 °C or a second resetting of biotite Ar isotopes at c. 94–98 Ma. The Maofengshan orthogneiss was exhumed to 8–10 km crustal levels at c. 150 Ma, whereas the eastward components of gneissic rock masses appear to have passed upward through the same crustal depth later (by c. 94 Ma). Exhumation of middle crustal-level rocks in Baiyunshan since c. 155 Ma was roughly coeval with exhumation of gneissic rocks from elsewhere in the Wuyun Orogen, suggesting a large-scale mechanism for the exhumation pulse in Yanshanian times.
Acknowledgements
We thank He-Ping Zou for assistance in the field and Hua-Ning Qiu, Yu-Ruo Shi, Xuan-Ce Wang and Hui-Qing Huang for analytical support. This study was supported by the Chinese National 973 Program (grant 2007CB411403) and the National Natural Science Foundation of China (grants 40728002 and 40825010). This is contribution No. IS-1131 from GIGCAS.
