1. Introduction
The Central Asian Orogenic Belt (CAOB), surrounded by the Siberia, Baltica, Tarim, and North China cratons, is a collage of microcontinents, oceanic-island arcs, seamounts, oceanic plateaus and accretionary complexes over ca. 800 million years (Sengör et al. Reference Şengör, Natal’in and Burtman1993; Jahn et al. Reference Jahn, Windley, Natal’in and Dobretsov2004; Kröner et al. Reference Kröner, Kovach, Belousova, Hegner, Armstrong, Dolgopolova, Seltmann, Alexeiev, Hofmann, Wong, Sun, Cai, Wang, Tong, Wilde, Degtyarev and Rytsk2014). It has been widely accepted that the CAOB, with the largest area of crustal growth in the Phanerozoic, is one of the largest accretionary orogens on Earth (Sengör et al. Reference Şengör, Natal’in and Burtman1993; Jahn et al. Reference Jahn, Wu and Chen2000, Reference Jahn, Windley, Natal’in and Dobretsov2004; Windley et al. Reference Windley, AlexeIev, Xiao, Kröner and Badarch2007; Xiao et al. Reference Xiao, Windley, Allen and Han2013; Safonova, Reference Safonova2017). The identification of many intro-oceanic arcs, accretionary complexes, oceanic plate strata, and MORB-OIB derived blueschist belts within the CAOB, and positive Nd-Hf isotope character of magmatic rocks are key evidence for its juvenile character (Jahn et al. Reference Jahn, Wu and Chen2000, Reference Jahn, Windley, Natal’in and Dobretsov2004; Safonova, Reference Safonova2017). However, some researchers stated that the volume of truly juvenile crustal material in the CAOB has been overestimated (Kröner et al. Reference Kröner, Kovach, Belousova, Hegner, Armstrong, Dolgopolova, Seltmann, Alexeiev, Hofmann, Wong, Sun, Cai, Wang, Tong, Wilde, Degtyarev and Rytsk2014, Reference Kröner, Kovach, Alexeiev, Wang, Wong, Degtyarev and Kozakov2017). One reason is that there are many microcontinents with Precambrian basement in the CAOB (Kröner et al. Reference Kröner, Kovach, Belousova, Hegner, Armstrong, Dolgopolova, Seltmann, Alexeiev, Hofmann, Wong, Sun, Cai, Wang, Tong, Wilde, Degtyarev and Rytsk2014; Long & Huang, Reference Long and Huang2017). Even in juvenile crustal regions such as the Chinese Altai, ancient crustal materials were identified (Xu et al. Reference Xu, Li, Jiang, Li, Qu, Yang, Zhou and Dong2015; Zhang et al. Reference Zhang, Wang, Tong, Zhang, Song, Zhang, Huang, Guo and Hou2017), and the Hf-in-zircon isotope data reveal recycled crust (Kröner et al. Reference Kröner, Kovach, Alexeiev, Wang, Wong, Degtyarev and Kozakov2017). Another reason might be whether the intra-oceanic arcs were correctly recognized (Kröner et al. Reference Kröner, Kovach, Alexeiev, Wang, Wong, Degtyarev and Kozakov2017). Therefore, the proportion of juvenile and recycled crust for crust growth during the late-stage evolution of the CAOB is still an open question.
Pacific-type convergent margins are major sits of continental growth and crustal recycling. The proportions of juvenile and recycled crust are generally estimated by the amount and geochemical signature of granitoids in the supra-subduction zone. However, the underestimation of crust growth caused by erosion of magmatic rocks during geological processes is generally ignored. These eroded materials were transported and deposited in the trench and adjacent for-arc and back-arc basins, forming characteristic clastic sedimentary rocks in accretionary complexes (Safonova et al. Reference Safonova, Perfilova, Savinskiy, Kotler, Sun and Wang2022a) These clastic rocks are compositionally similar to their source rocks and contain detrital zircons, recording crystallization ages and source characters information of their parental igneous rocks. If sediments were derived mainly from eroded intra-arc crust, then they can be regarded as contribution of continental growth. Therefore, the study of sediment recycling might be important in evaluating the proportion of continental growth. Despite a few recent studies (Safonova et al. Reference Safonova, Perfilova, Savinskiy, Kotler, Sun and Wang2022a, Reference Safonova, Perfilova, Obut, Kotler, Aoki, Komiya, Wang and Sunb), accretionary complexes, as major component of the CAOB, lack detailed study especially on assessing CAOB late-stage crustal growth.
The Chinese West Tianshan Orogen is located in the southwest part of the CAOB (Fig. 1a). The northern part of the Chinese West Tianshan Orogen is composed of late Paleozoic accretionary complex, including flysch type volcano-sedimentary rocks and some ophiolite fragments, in response to evolution of the North Tianshan Ocean. Previous studies mainly focus on the petrology, geochronology and geochemistry of the ophiolites (Xiao et al. Reference Xiao, Tang, Feng, Zhu, Li and Zhao1992; Xu et al. Reference Xu, Xia, Ma, Wang, Xia, Li and Wang2006a, Reference Xu, Li, Ma, Xia, Xia and Pengb; Li et al. Reference Li, Xiao, Han, Zhou, Zhang and Zhang2015; Feng & Zhu, Reference Feng and Zhu2018; Zheng et al. Reference Zheng, Zhong, Wang, Yang, Kapsiotis, Xiao and Wan2019), but studies on the sedimentary rocks are weak. Recently, our group reported detrital zircon U-Pb geochronology and Lu-Hf isotope characteristics of the turbidites, suggesting significant Carboniferous magmatism and continental growth (Wang et al. Reference Wang, Zhang, Pei, Liu, Zhang and Chen2018a). In this paper, we present whole rock geochemistry and Sr-Nd isotope for the turbidites in the North Tianshan Accretionary Complex (NTAC) (Fig. 2) to further constrain their provenance and tectonic setting of their parental igneous rocks and evaluate their contribution for continental growth.
(a) Tectonic framework of the Central Asian Orogenic Belt (modified after Şengör et al. Reference Şengör, Natal’in and Burtman1993 and Xiao et al. Reference Xiao, Windley, Allen and Han2013). (b) Geological map of the Chinese West Tianshan Orogen and adjacent regions (modified after Gao et al. Reference Gao, Long, Klemd, Qian, Liu, Xiong, Su, Liu, Wang and Yang2009). Numbers in circle refer to the tectonic boundaries: 1) North Tianshan Suture Zone; 2) Nikolaev line-North Nalati Suture Zone; 3) Southern Central Tianshan Suture Zone; and 4) North Tarim Thrust. Abbreviations for the tectonic unites are: NTAC = North Tianshan Accretionary Complex; YB = Yili Block; CTB = Central Tianshan Block; STB = South Tianshan Belt.
2. Geological background
The CAOB, also known as the modified Altaids, occupies an immense area in central Asia (Fig. 1a). The Tianshan orogenic belt, presently extending for more than 2500 km from west to east through Uzbekistan, via Tajikistan, Uzbekistan, and Kyrgyzstan to northwestern China, is an important constituent part of the CAOB. The Chinese Tianshan can be subdivided into the Western Tianshan and East Tianshan, which are different both in topography and evolutionary history, roughly along the longitude ∼90°E (Xiao et al. Reference Xiao, Windley, Allen and Han2013). The Chinese West Tianshan Orogen can be further subdivided into the North Tianshan Accretionary Complex (NTAC), the Yili Block (YB), the Central Tianshan Block, and the South Tianshan Belt, which are separated by the North Tianshan, North Narat, and South Tianshan faults (Fig. 1b).
The YB is a triangular shaped micro-continent which mainly Neoproterozoic basement rocks and extends over 1000 km across NW China, Kazakhstan, and Kyrgyzstan (Wang et al. Reference Wang, Faure, Cluzel, Shu, Charvet, Meffre and Ma2006). The center of the YB is covered by Cenozoic sediments. The basement rocks are exposed in the northern and southern side of the YB. Ordovican to Silurian intrusive rocks which might be related to the early stage subduction of the NTSO, are found in the Wenquan area of the northwestern YB (Wang et al. Reference Wang, Jahn, Shu, Li, Chung and Liu2012; Huang et al. Reference Huang, Long, Kröner, Yuan, Wang, Sun, Zhao and Wang2013). Late Paleozoic volcano-sedimentary rocks and intrusive rocks are widely distributed in both sides of the block and were suggested to be related to the subduction of the North Tianshan and South Tianshan oceans (Wang et al. Reference Wang, Zhang, Ren, Pei, Zhang, Chen, Li and Ge2022 and references therein).
The Central Tianshan Block is a strip-like arc terrane with Precambrian basement (Gao et al. Reference Gao, Long, Klemd, Qian, Liu, Xiong, Su, Liu, Wang and Yang2009). It approximately corresponds to the Narat Range in the west and extend from the Baluntai area to Xingxingxia area in the east. The basement rocks are mainly Meso- to Neo-proterozoic schists, paragneiss, and granitic gneisses (Hu et al. Reference Hu, Wei, Jahn, Zhang, Deng and Chen2010; Wang et al. Reference Wang, Liu, Shu, Jahn, Chung, Zhai and Liu2014; Gao et al. Reference Gao, Wang, Klemd, Jiang, Qian, Mu and Ma2015; Huang et al. Reference Huang, Cawood, Ni, Hou, Shi and Hu2017). Paleozoic volcanic and intrusive rocks are widely exposed and were formed due to subduction and closure of the Terskey and South Tianshan oceans (Xu et al. Reference Xu, Wang, Li, Chen, Ma, Zhu, Wang and Dong2013; Wang et al. Reference Wang, Zhang, Zhang, Liu and Ge2016a, Reference Wang, Zhang, Ren, Pei, Zhang, Chen, Li and Ge2022; Zhong et al. Reference Zhong, Wang, Alexeiev, Cao, Biske, Liu, Zhai and Xing2017).
The South Tianshan Belt is a wide tectonic unit bounded by the southern Central Tianshan Fault and the northern Tairm Fault (Xiao et al. Reference Xiao, Windley, Allen and Han2013). It was formed as a result of the consumption of the South Tianshan ocean and subsequent collision between the Tarim Craton and Yili-Central Tianshan Block (Charvet et al. Reference Charvet, Shu, Laurent-Charvet, Wang, Faure, Cluzel, Chen and De Jong2011; Xiao et al. Reference Xiao, Windley, Allen and Han2013; Wang et al. Reference Wang, Zhai, Kapp, de Jong, Zhong, Liu, Ma, Gong and Geng2018b; Huang et al. Reference Huang, Wang, Tong, Qin, Ma and Yin2020). The South Tianshan Belt is composed of Cambrian to Carboniferous limestone, siliciclastic turbidites, cherts, and interlayered volcanics (Jiang et al. Reference Jiang, Gao, Klemd, Qian, Zhang, Xiong, Wang, Tan and Chen2014). Permian unconformably overlie the former strata (Wang et al. Reference Wang, Liu, Shu, Jahn, Chung, Zhai and Liu2014). Several ophiolite fragments with ages ranging from 590 to 320 Ma have been identified within the South Tianshan Belt (Jiang et al. Reference Jiang, Gao, Klemd, Qian, Zhang, Xiong, Wang, Tan and Chen2014; Wang et al. Reference Wang, Shu, Faure, Jahn, Cluzel, Charvet, Chung and Meffre2011, Reference Wang, Zhai, Kapp, de Jong, Zhong, Liu, Ma, Gong and Geng2018b and references therein). Ophiolites that experienced high pressure metamorphism were also reported in the Akeyazi, NW China (Gao et al. Reference Gao, He and Li1995), and Atbashi, Kyrgyzstan (Hegner et al. Reference Hegner, Klemd, Kröner, Corsini, Alexeiev, Iaccheri, Zack, Dulski, Xia and Windley2010).
The NTAC located between the Junggar terrane and the YB, extends over 300 km along the northern margin of the North Tianshan. It consists mainly of two of two different lithologies: 1) the ophiolitic remnants (Wang et al. Reference Wang, Faure, Cluzel, Shu, Charvet, Meffre and Ma2006; Xu et al. Reference Xu, Xia, Ma, Wang, Xia, Li and Wang2006a; Han et al. Reference Han, Guo, Zhang, Zheng, Chen and Song2010) and 2) the Carboniferous volcano-sedimentary rocks which were regarded as turbidites (Wang et al. Reference Wang, Faure, Cluzel, Shu, Charvet, Meffre and Ma2006). The ophiolites exposed discontinuously over an area of ∼250 km in length and 5–15 km in width (Fig. 1(b)), and the Bayingou ophiolite is one of the best outcrops (Xu et al. Reference Xu, Xia, Ma, Wang, Xia, Li and Wang2006a). Famennian-Visean conodonts and radiolarians had been found from the cherts within the NTAC (Xiao et al. Reference Xiao, Tang, Feng, Zhu, Li and Zhao1992). Gabbro and plagiogranite from the Bayingou ophiolite have SHRIMP U-Pb age of 344 Ma and 325 Ma (Xu et al. Reference Xu, Xia, Ma, Wang, Xia, Li and Wang2006a, b). Plagiogranite from the Kuitun River section were dated at 343 Ma (Li et al. Reference Li, Xiao, Han, Zhou, Zhang and Zhang2015). Layered and massive gabbros in the Jinghe ophiolite were dated to be 381 and 382 Ma (Zheng et al. Reference Zheng, Zhong, Wang, Yang, Kapsiotis, Xiao and Wan2019). There exist several different perspectives about the nature of the North Tianshan ophiolites, i.e. 1) remnant of a ‘Red Sea type’ oceanic basin (Xia et al. Reference Xia, Xu, Xia, Li, Ma and Wang2004; Xu et al. Reference Xu, Xia, Ma, Wang, Xia, Li and Wang2006a, b), 2) remnants of a long-lived wide ocean (Yang et al. Reference Yang, Li, Kerr and Tong2018), 3) products related to ridge subduction (Li et al. Reference Li, Xiao, Han, Zhou, Zhang and Zhang2015), and 4) ophiolite formed in forearc setting (Feng & Zhu, Reference Feng and Zhu2018).
The turbiditic sequence in the NTAC consist mainly of tuffaceous sandstone, siltstone, tuff, and black argillite alternations (Fig. 3). The tuffaceous sandstone ranges in thickness from a few centimeters to 50 centimeters, but mostly around 10–40 cm (Fig. 3a, b). Some deep-water ichnofossils, like Chondrites sp., Cosmorphaphe sp., Helminthopsis sp. et al., and deep-water microfossils were found in the turbidites (Jin et al. Reference Jin, Li and Li1989), indicating in a bathyal to deep-sea environment (Wang et al. Reference Wang, Faure, Cluzel, Shu, Charvet, Meffre and Ma2006). The turbiditic sequence near the North Tianshan Fault underwent low-grade metamorphism and display a subvertical slaty cleavage (Figs. 2 and 3c), and some siltstones had turned out to be andalusite schists (Wang et al. Reference Wang, Faure, Cluzel, Shu, Charvet, Meffre and Ma2006). Age of the turbidites were constrained to be Devonian to Carboniferous based on fossils discovered in the chert and limestone (XBGMR, 1993). However, our recent study shows that the turbidites have the youngest DZ ages of 312–303 Ma, constraining the maximum depositional time of the turbidites to the late Carboniferous (Wang et al. Reference Wang, Zhang, Pei, Liu, Zhang and Chen2018a; Bai et al. Reference Bai, Chen, Song, Xiao, Windley, Ao, Li and Xiang2020).
Geological map of the study area (modified from XBGMR, 1993), showing sample locations. The sample locations for detrital zircon U-Pb dating are from Wang et al. (Reference Wang, Zhang, Pei, Liu, Zhang and Chen2018a).
Representative outcrop (a, b; from Wang et al. Reference Wang, Zhang, Pei, Liu, Zhang and Chen2018a) and microphotographs (c, d) of the turbidites from the NTAC.
Volcanic rocks, represented by the Arbasay Formation, also exist within the NTAC. The Arbasay Formation, which is mainly composed of volcanic rocks in the lower part and volcaniclastic rocks in the upper part, exposes to the north of the Bayingou ophiolite (Wang et al. Reference Wang, Zhang, Zhang, Liu and Ge2016a). The Arbasay Formation was previously regarded as early Permian in age, because it unconformably overlies the upper Carboniferous Qianxia Formation and is covered by the upper Permian Quanzijie Formation (XBGMR, 1993). However, zircons from the volcanic rocks and tuff of the Arbasay Formation were dated at 315–305 Ma, indicating a Late Carboniferous age (Liu et al. Reference Liu, Cheng, Guo, Jolivet and Song2015; Wang et al. Reference Wang, Wu, Li, Zhu, Chen, Li, Wu, Deng and Chen2017; Bai et al. Reference Bai, Chen, Song, Xiao, Windley, Ao, Li and Xiang2020). The volcanic rocks from the Arbasay Formation include basalt, andesite, dacite, and rhyolite, dominated by andesites (Liu et al. Reference Liu, Cheng, Guo, Jolivet and Song2015; Bai et al. Reference Bai, Chen, Song, Xiao, Windley, Ao, Li and Xiang2020). The volcanic rocks display arc-like geochemical signatures, indicating that they are subduction related and might have formed during the tectonic transition from compression to extension (Wang et al., Reference Wang, Wu, Li, Zhu, Chen, Li, Wu, Deng and Chen2017; Bai et al. Reference Bai, Chen, Song, Xiao, Windley, Ao, Li and Xiang2020).
3. Sampling and analytical methods
Fifteen clastic rock samples, mainly fine-grained sandstones and siltstones, and 4 slate samples were collected from the NTAC for geochemical studies. We also selected 6 sandstones to determine their Rb-Sr and Sm-Nd isotope compositions. The detailed sampling locations are illustrated in Fig. 2. Representative petrographic characteristics of the studied samples are shown in Fig. 3c-d.
Whole-rock major and trace element analyses were conducted at the SKLCD, Norwest University, China. The rock samples were first crushed into small fragment and cleaned in distilled water, dried, and then ground into powder. Major elements were determined by XRF method, while trace and rare earth elements were analyzed using an ICP-MS (PE 6100 DRC). Analytical precision and accuracy are better than 5% for major elements, 2%for most trace elements and 10% for transition metals. Detailed procedures and working conditions are described in Dong et al. (Reference Dong, Zhang, Neubauer, Liu, Hauzenberger, Zhou and Li2011).
The Rb-Sr and Sm-Nd isotopes of the samples were analyzed at the Wuhan SampleSolution Analytical Technology Co., Ltd., Wuhan, China. All chemical preparations were performed on class 100 work benches within a class 1000 over-pressured clean laboratory. Whole rock Sr-Nd isotope analyses were performed on a Neptune Plus MC-ICP-MS (Thermo Fisher Scientific, Dreieich, Germany). Analyses of the NIST SRM 987 standard solution yielded 87Sr/86Sr ratio of 0.710243 ± 4 (2 SD, n = 7), which is the same with published value of0.710241±12within error (Zhang et al. Reference Zhang, Hu and Liu2020). The USGS reference materials BCR-2 (basalt) and RGM-2 (rhyolite) yielded 87Sr/86Sr ratios of 0.704990 ± 6 and 0.703960 ± 6, respectively, consistent with the published values (Zhang et al. Reference Zhang, Hu and Liu2020; Li et al. Reference Li, Li, Li, Guo and Yang2012). Analyses of the GSB 04-3258-2015 standard yielded 143Nd /144Nd ratio of 0.512441 ±2 (2SD, n = 7), which is identical within error to the published values (0.512439 ± 10 (2SD), Li et al. Reference Li, Tang, Zhu and Pan2017). In addition, the USGS reference materials BCR-2 (basalt) and RGM-2 (rhyolite) yielded 143Nd/144Nd ratios of 0.512644 ± 7 and 0.512802 ± 6, respectively, which are within error to their published values (Zhang et al. Reference Zhang, Hu and Liu2020).
4. Analytical results
4.a. Major element
Analytical results of whole-rock major and trace elements are listed in Table 1. The sandstone/siltstones display a wide range of SiO2 contents varying from 49.73 wt %–71.25 wt %, but mostly lower than 65 wt % (average 60.28 wt %), which are below or slightly higher than that of average upper continental crust (UCC; 66.6 wt %) (Taylor et al. Reference Taylor, McLennan, Armstrong and Tarney1981) and post-Archaean Australian shales (PAAS; 62.4 wt %) (Taylor and McLennan, Reference Taylor and McLennan1985). They also have variable Na2O (2.30–5.58 wt %) and K2O (0.52–4.85), with Na2O/K2O ratios between 0.6 and 9.2. The samples have high Fe2O3T (av. 5.31 wt %), Al2O3(av. 16.05 wt %) and MgO (av. 2.15 wt %) contents, suggesting a more mafic to intermediate source. The slates show higher SiO2 contents of 66.78 wt %–73.21 wt %, higher than those in UCC and PAAS. They also have lower Fe2O3T (av. 3.57 wt %), MgO (1.44 wt %), and Al2O3 (av. 13.56 wt %) than the sandstones, suggesting more quartz and less mafic minerals in the source of the slates. Although the sandstones and the slates have different major element compositions, they all plot in the greywacke field in the rock type discrimination diagram (Fig. 4). The Na2O/Al2O3 ratios of 0.17–0.36 for the studied samples are also typical of greywackes. Negative correlations can be observed for all the samples between SiO2 and TiO2, Al2O3, Fe2O3, and MgO (Figures not shown), indicating weak major element fractionation during sedimentation, i.e., they are first cycle sediments (Safonova et al. Reference Safonova, Perfilova, Savinskiy, Kotler, Sun and Wang2022a).
Whole rock geochemical composition of the Carboniferous turbidites in the NTAC
Note: CIA=[Al2O3/(Al2O3+CaO*+Na2O+K2O)]×100 and PIA=[(Al2O3–K2O)/(Al2O3+CaO*+Na2O–K2O)]×100, where CaO* represents Ca in silicate-bearing minerals only and all in molecular proportions; ICV= (Fe2O3+K2O+Na2O+CaO+MgO+TiO2)/Al2O3.
Sr-Nd isotopic composition of the Carboniferous turbidites in the NTAC
Unit of model age (
${\rm{T}}_{{\rm{DM}}}^1$
and
${\rm{T}}_{{\rm{DM}}}^2$
) is Ma.
εNd(t)=[(143Nd/144Nd)S/(143Nd/144Nd)CHUR-1]×10,000.
${\rm{T}}_{{\rm{DM}}}^1$
=1/λSm×ln{[(143Nd/144Nd)S-0.51315]/[(147Sm/144Nd)S-0.2137]}.
${\rm{T}}_{{\rm{DM}}}^2$
=
${\rm{T}}_{{\rm{DM}}}^1$
-(
${\rm{T}}_{{\rm{DM}}}^1$
-t)×[(fCC-fS)/ (fCC-fDM)].
where S= sample, (143Nd/144Nd)CHUR=0.512638 and (147Sm/144Nd)CHUR=0.1967. The fCC, fS and fDM are fSm/Nd values of the continental crust, the sample and the depleted mantle, respectively.
Geochemical classification diagram of the turbidites in the NTAC (after Herron, Reference Herron1988).
4.b. Trace and rare earth element
Most of the studied samples have lower rare earth and other incompatible element concentrations than PAAS (Fig. 5). Most of the sandstone/siltstone samples have low rare earth element (REE) contents of from 60 ppm to 99 ppm, except for one sample TS17-70 which has total REE values of 133 ppm. The slates have relatively higher ΣREE values of 110 ppm–121 ppm, with one exception (TS17-44) of 76 ppm. Both the sandstone/siltstone and slate samples display are enriched in LREE and depleted in HREE (Fig. 5c, d), with (La/Yb)N ratios of 2.09–6.58 and 4.88–7.00, respectively (Table 1). Most sandstone/siltstone samples display weak positive Eu anomalies ((Eu/Eu*) = 1.01–1.3), and two samples show weak negative Eu anomalies ((Eu/Eu*) = 0.98 and 0.88), similar to that of the Late Carboniferous basaltic rocks in the NTAC (Fig. 5c). In comparison, the slate samples have moderate negative Eu anomalies ((Eu/Eu*) = 0.77–0.87, with one exception of 1.02), which is a feature of intermediate to felsic magmatic rocks that undergone fractional crystallization of plagioclase.
Chondrite-normalized REE patterns and upper crust-normalized spider diagrams for the Paleozoic graywackes. Symbols are the same as in Fig. 4. Chondrite and PM normalizing values are from Sun and McDonough (Reference Sun and McDonough1989) and upper crust-normalizing data are from Taylor and McLennan (Reference Taylor and McLennan1985). Geochemical data for volcanic rocks in the NTAC are from Liu et al. (Reference Liu, Guo, Zhang and Wu2012), Wang et al. (Reference Wang, Wu, Li, Zhu, Chen, Li, Wu, Deng and Chen2017) and Bai et al. (Reference Bai, Chen, Song, Xiao, Windley, Ao, Li and Xiang2020).
In the primitive mantle normalized multielement spectra (Fig. 5a, b), all the samples show clear Nb depletion compared to La (Nb/Lapm = 0.2-0.8) and Th (Nb/Thpm = 0.1-0.2), suggesting their derivation from supra-subduction/arc volcanic rocks (Briqueu et al. Reference Briqueu, Bougault and Joron1984; Safonova et al. Reference Safonova, Perfilova, Savinskiy, Kotler, Sun and Wang2022a)
4.c. Sr-Nd isotope
We conducted Rb-Sr and Sm-Nd isotope analyses for six sandstone samples from the NTAC. The samples display 147Sm/144Nd ratios between 0.129367 and 0.163533, and 143Nd/144Nd ratios between 0.512543 and 0.512975 (Table 2). Most samples have high positive εNd(t) values (+5.5 to +7.9), with only one exception(εNd(t) = +0.8) (Fig. 6), and the TDM2 values are mainly from 0.63 to 0.45 Ga (with one exception at ∼1.0 Ga) (Table 2).
εNd(t) versus age diagram for the Paleozoic igneous rocks and sediments in the northern Chinese West Tianshan. Nd isotopic Data for the igneous rocks can be found in Supplementary Table S1.
5. Discussion
5.a. Influence of chemical weathering and recycling
The geochemical compositions of clastic rocks are dominated controlled by, and thus can used to constrain weathering conditions and recycling of, their source rocks (Nesbitt and Young, Reference Nesbitt and Young1984; Mclennan et al. Reference McLennan, Hemming, Mcdaniel and Hanson1993; Fedo et al. Reference Fedo, Nesbitt and Young1995). Some parameters, such as the chemical index of alteration (CIA), the plagioclase index of alteration (PIA), and index of compositional variability (ICV), are frequently used to quantify the degree of chemical weathering and composition maturity of the source rocks (Nesbitt and Young, Reference Nesbitt and Young1984; Cox & Lowe, Reference Cox and Lowe1995).
The CIA values of the sandstone/siltstone samples are 35–56 (average 52) and slate samples are 56–63 (average 59). However, the CIA values could be influenced by potassium metasomatism during diagenesis, so it is necessary to use the A-CN-K triangle diagram to identify it and make corrections. In the A-CN-K diagram (Fig. 7a), the sandstone/siltstone samples define a weathering trend line that does not parrel to the A-CN line, and reflect potassium metasomatism. The point of intersection between the weathering trend line and the Pl-Kfs line is near the plot of the tonalite (Fig. 7a). This means that the average composition of the source rocks is equal to intermediate rocks. The corrected initial CIA values of the sandstones and siltstones were slightly higher than the calculated CIA, but still lower than 60 (Fig. 7a). The slates plot near the granite field with limited weathering towards illite (Fig. 7a), reflecting a dominate felsic rock source. The corrected CIA values of the sandstone/siltstones are close to the CIA values of NASC and Archean greywacke sandstones (58; Condie, Reference Condie1993), but much lower than that of PAAS (70; Taylor and McLennan, Reference Taylor and McLennan1985). This suggests that the source rocks of the turbidites experienced low to moderate weathering before erosion and deposition. In addition, the PIA values of sandstone/siltstone (34–67, average 51) and slate samples (58–68, average 63) are both lower than that of the PAAS (Fedo et al. Reference Fedo, Nesbitt and Young1995), also indicating a relatively low degree of chemical weathering.
Discrimination diagrams illustrating weathering and sediment recycling. (a) Th/Sc versus Zr/Sc diagram (after McLennan et al. Reference McLennan, Hemming, Mcdaniel and Hanson1993); (b) A-CN-K diagram (after Fedo et al. Reference Fedo, Nesbitt and Young1995), Data for basalt (Ba) tonalite (To), granodiorite (Gd), granite (Gr), and average post-Archean upper crust are from Condie (Reference Condie1993) and Nesbitt and Young (Reference Nesbitt and Young1984); (c) Th/U versus Th (after McLennan et al. Reference McLennan, Hemming, Mcdaniel and Hanson1993); (d) ICV versus CIA diagram (after Nesbitt and Young, Reference Nesbitt and Young1984).
The ICV values are used to reflect the maturity of the source rocks of clastic rocks (Cox & Lowe, Reference Cox and Lowe1995). Sedimentary rocks which were derived from a source dominated by clay minerals generally have low ICV values, while those sourced from an immature source with high percentage of nonclay minerals commonly possess high IVC values (Van de & Leake, Reference Van de and Leake1985). Both the sandstone/siltstone and slate samples display high ICV values of 0.89–1.50 and 0.89–0.93 (Fig. 7b), higher than the PAAS, suggesting a relatively immature source. In the Th/Sc–Zr/Sc plot (Fig. 7c), the samples show a trend of compositional variation instead of sedimentary recycling. The sandstone/siltstone samples have Th/U ratios lower than that of the UCC (3.8; Taylor and McLennan, Reference Taylor and McLennan1985), while the slate samples have higher Th/U ratios than the UCC (Fig. 7d), suggesting that the sandstones/siltstones were derived from source rocks with less weathering than that of the slates.
The SiO2/Al2O3 ratio is also an index for sedimentary maturity because quartz has stronger weathering resistance than feldspars, mafic minerals, and lithic grains (Roser et al. Reference Roser, Cooper, Nathan and Tulloch1996; Zhang, Reference Zhang2004). The average values range from ∼3.0 to ∼5.0 for unaltered basic and felsic igneous rocks, while values >5.0–6.0 in sediments are an indication of progressive maturity (Zhang, Reference Zhang2004). The sandstone/siltstones have SiO2/Al2O3 ratios of 3.3–5.5, indicating low degree and sorting and deposition close to sources. The slates have slightly higher SiO2/Al2O3 ratios of 4.4–5.8, suggesting possibly stronger weathering.
To conclude, the turbidites from the NTAC were first cycle sediments and derived from immature materials with low degree of chemical weathering. The slates might contain hither contents of materials which experience slightly stronger weathering than the sandstone/siltstone samples.
5.b. Source rocks and tectonic setting
The turbidites from the NTAC are characterized by large proportions of angular volcanic lithic fragments and immature geochemical signatures, suggesting that they were derived from a nearby magmatic arc. The sandstone/siltstone samples have more mafic to andesitic volcanic clasts than the slates, indicating a more mafic source. This is consistent with the geochemical data. Geochemical characteristics of sedimentary rocks derived from different source rocks have been studied for a long time (Bhatia and Taylor, Reference Bhatia and Taylor1981; Bhatia and Crook, Reference Bhatia and Crook1986; McLennan and Taylor, Reference McLennan and Taylor1991; Safonova et al. Reference Safonova, Perfilova, Savinskiy, Kotler, Sun and Wang2022a, b). According to the source rock discrimination diagrams using major elements (Fig. 8a), the sandstone/siltstone from the NTAC were mainly derived from intermediate and felsic igneous rocks, while the slates were mainly from felsic igneous rocks and quartz-rich sedimentary rocks. In the La/Th versus Hf diagram (Fig. 8b), most sandstone/siltstone samples plot in the mixed felsic/basic source area, while the slates plot in the acid arc source area. Potassium and Rb are relatively mobile elements during low grade metamorphism and diagenesis, but their covariant relation is generally considered as a standard for determining the source rocks of clastic rocks (Floyd et al. Reference Floyd, Winchester and Park1989). Although the samples from the NTAC show a wide range of K/Rb ratios, their high Rb and K2O contents indicate the derivation from acidic to intermediate igneous rocks (Fig. 8c). Some trace element ratios, such as Th/Sc, Co/Th, La/Sc, and Ti/Zr are also good indicators for provenance study, as silicic rocks trends to have more contents of Th and Zr, whereas basic rocks hold more contents of Sc, Co, and Ti (Cullers, Reference Cullers1994). Th/Sc and Zr/Sc ratios increase with supply of felsic detritus, whereas Ti/Zr and Co/Th ratios decrease. Most of the sandstone/siltstones have higher Ti/Zr ratios (12.4–83.3, av. = 39.5) compared to the UCC (19.9), indicating dominantly basic-intermediate igneous rocks in the source area. In comparation, the slates exhibit relatively low Ti/Zr ratios (15.6–35.1, av. = 20.2), suggesting a dominantly felsic igneous source. In the Co/Th versus La/Sc diagrams (Fig. 8d), the sandstone/siltstone samples plot in a mixing source from basalt to andesite and felsic volcanic rocks. As a contrast, the slates mainly distribute around the felsic rocks (Fig. 8d). These features indicate a provenance of basaltic to felsic rocks for the sandstone/siltstones and mainly felsic rocks for the slates.
Source rock discrimination diagrams. (a) discriminant function diagram using major elements after Roser and Korsch (Reference Roser and Korsch1988); (b) La/Th versus Hf diagram (after Floyd and Leveridge, Reference Floyd and Leveridge1987); (c) K2O versus Rb diagram (after Folyd et al. Reference Floyd, Winchester and Park1989); (d) Co/Th versus La/Sc diagram (after Gu et al. Reference Gu, Liu, Zheng, Tang and Qi2002).
The major and trace element compositions of sediments are also widely used to constrain the tectonic setting of their source rocks (Bhatia and Taylor, Reference Bhatia and Taylor1981; Roser and Korsch, Reference Roser and Korsch1986; Bhatia and Crook, Reference Bhatia and Crook1986). Generally, sandstones sourced from oceanic island arcs the have higher Fe2O3T+MgO, Al2O3, and TiO2, and lower SiO2, K2O/Na2O, and Al2O3/(Na2O+CaO) than those sourced from passive margins (Bhatia, Reference Bhatia1983). In the K2O/Na2O versus Fe2O3T+MgO, Al2O3/(Na2O+CaO) versus Fe2O3T+MgO, TiO2 versus Fe2O3T+MgO, and Al2O3/SiO2 versus Fe2O3T+MgO diagrams (Fig. 9a-d), the sandstone/siltstones all plot within or near the oceanic island arc and continental island arc fields, while the slates mainly distribute within or around the active continental margin field. When plotting on the Df1 versus Df2 diagram which considers the role of multiple variables (Fig. 9e), it shows that the sandstone/siltstones illustrate an oceanic island and active continental margin setting, while the slates reflect an active continental margin setting. In the (Fe2O3T + MgO)/(SiO2 + Na2O + K2O) versus Al2O3/SiO2 diagram (Fig. 9f), most of the sandstone/siltstone samples fall within the immature island arc (IIA) field and evolved island arc (EIA) field, while the slates mainly fall within the mature magmatic arc (MMA) field. Similar results are observed in the tectonic setting discrimination diagrams using trace elements (Fig. 10).
Tectonic setting discrimination diagrams using major elements for turbidites in the NTAC (after Bhatia, Reference Bhatia1983; Kumon and Kiminami, Reference Kumon and Kiminami1994). (a) K2O/Na2O versus (Fe2O3+ MgO) diagram; (b) Al2O3/(Na2O + CaO) versus (Fe2O3+ MgO) diagram; (c) TiO2 versus (Fe2O3+ MgO) diagram; (d) Al2O3/SiO2 versus (Fe2O3+ MgO) diagram; (e) Df1 versus Df2 diagram; (f) Al2O3/SiO2 versus (Fe2O3+ MgO)/(SiO2+ Na2O + K2O) diagram. The discriminant functions are: Df1 = −0.0447 × SiO2 − 0.972 × TiO2 + 0.008 × Al2O3- 0.267 × Fe2O3 + 0.208 × FeO − 3.082 × MnO + 0.14 + MgO + 0.195 × CaO + 0.719 × Na2O − 0.032 × K2O + 7.510 × P2O5 + 0.303; Df2 = −0.421 × SiO2 + 1.988 × TiO2 − 0.526 × Al2O3 − 0.551 × Fe2O3 − 1.61 × FeO + 2.72 × MnO + 0.881 × MgO − 0.907 × CaO − 0.177 × Na2O − 1.84 × K2O + 7.244 × P2O5 + 43.57. Abbreviations for tectonic settings: A, oceanic island arc; B, continental arc; C, active continental margin; D, passive continental margin; IIA, immature island arc; EIA, evolved island arc; MMA, mature magmatic arc.
Tectonic setting discrimination diagrams using trace elements for turbidites in the NTAC (after Bhatia and Crook, Reference Bhatia and Crook1986). (a) Ti/Zr versus La/Sc diagram; (b) La/Y versus Sc/Cr; (c) Th-Sc-Zr/10 diagram; (d) Th-Co-Zr/10 diagram; (e) La-Th-Sc diagram. Abbreviations for tectonic settings are the same as in Fig. 9.
In the northern YB and the NTAC, there are massive Paleozoic volcanic and intrusive rocks which exhibit distinguishable Nd isotope characteristics (Fig. 6). The sandstone/siltstone samples have comparable Sr-Nd isotope compositions with the Late Paleozoic mafic igneous rocks in the northern YB and Carboniferous volcanic rocks in the NTAC (Fig. 6). If they were mainly sourced from the northern YB, their source rocks were supposed to contain the felsic volcanic rocks as these rocks make up a larger proportion in the northern YB (Fig. 1b). This is inconsistent with the above suggestion that the source rocks of the sandstone/siltstones are mainly basic-intermediate igneous rocks. In addition, the U-Pb ages of the turbidites are unimodal with major peaks at 323 Ma (Fig. 11a; Wang et al. Reference Wang, Zhang, Pei, Liu, Zhang and Chen2018a). It is completely different from the detrital zircon age spectrum of Early Carboniferous sandstone just to the south of the North Tianshan Fault, which contain many Early Paleozoic and Precambrian detrital zircons (Wang et al. Reference Wang, Zhang, Pei, Zhang, Chen, Xiao and Zheng2018c). What is more, the sandstone/siltstones from the NTAC have depleted whole-rock Nd isotope and detrital zircon Hf isotope composition (Fig. 6 and Fig. 11b), indicating a juvenile mantle source for their parental rocks, which might be part of an intra-oceanic arc. These features indicate the sandstone/siltstone were sourced from a Carboniferous intra-oceanic arc and deposited in a fore-arc/trench. The source rocks of the slate are more felsic and might be part of a continental arc. Therefore, the turbidites from the NTAC were likely deposited between an intra-oceanic arc (the North Tianshan arc) and a continental arc (the northern YB).
(a) Relative probability plot for detrital zircons from turbidites in the NTAC (after Wang et al. Reference Wang, Zhang, Pei, Liu, Zhang and Chen2018a); (b) Plot of εHf(t) versus U-Pb age for detrital zircons from turbidites in the NTAC (after Wang et al. Reference Wang, Zhang, Pei, Liu, Zhang and Chen2018a).
5.c. Implication for crustal growth in the CAOB
Juvenile continental growth can be recorded by trondhjemite-granodiorite (TTG) and mafic magmatism at supra-subduciton zone and intra-plate settings (Safonova, Reference Safonova2017). However, the oceanic crust was mostly subducted into deep mantle and the intra-oceanic arcs could have been eroded or buried deeply in the lower crust, which might lead to the underestimation of proportion of juvenile crust. Turbidites and greywackes deposited at convergent margin are sourced from upper crust of the adjacent plates and thus recorded the nature of these upper crust rocks. If these sediments were sourced from the juvenile materials, then it represents continental growth (Kröner et al. Reference Kröner, Kovach, Alexeiev, Wang, Wong, Degtyarev and Kozakov2017). Recent geochemical, detrital zircon geochronological and Hf-Nd isotopic studies on sandstones from central and eastern Kazakhstan had implied several pre-existed intra-oceanic arcs (Safonova et al. Reference Safonova, Perfilova, Savinskiy, Kotler, Sun and Wang2022a, b).
Though the outcrop is limited, the occurrence of Carboniferous arc type volcanic rocks in the NTAC, and juvenile source nature of the turbidites allow us to infer that there was a Late Devonian to Carboniferous intra-oceanic arc to the north of the YB, which was partly eroded and/or covered later. Here, we term this intra-oceanic arc as the North Tianshan arc, also mentioned by Bai et al. (Reference Bai, Chen, Song, Xiao, Windley, Ao, Li and Xiang2020). However, the extent of the North Tianshan arc was unknown, maybe equivalently to the NTAC, because most of this arc was overlain by the Carboniferous turbidites. Another evidence for the arc nature of the NTAC basement is the magnetic anomaly data in the northern Xinjiang (He et al. Reference He, Li, Fan and Yang2013). To the east of Urumqi, the Bogda Range contains massive Carboniferous magmatic rocks with geochemical signatures of oceanic island arc and was called the Bogda arc (Xie et al. Reference Xie, Luo, Xu, Chen, Hong, Ma and Ma2016; Memtimin et al. Reference Memtimin, Pe-Piper, Piper, Guo and Zhang2020). According to the magnetic anomaly data, the Bogda arc is characterized by a medium–low magnetic anomaly, and this anomaly extends westward to the North Tianshan area (He et al. Reference He, Li, Fan and Yang2013). This similarity suggests that the basement of the NTAC (the North Tianshan arc) might be linked with the Bogda arc. The Late Carboniferous Arbasay Formation may represent limited outcrop of the North Tianshan arc (Bai et al. Reference Bai, Chen, Song, Xiao, Windley, Ao, Li and Xiang2020).
In summary, our study indicates the possibly pre-existence of a Devonian to Carboniferous intra-oceanic arc, i.e., the North Tianshan arc, to the north or the YB. This arc was then partly eroded and transferred to form the Carboniferous turbidites with overlain sequences. This study further proves the validity of using sedimentary record to reveal neglected juvenile crustal growth in accretionary orogens. The contribution of the eroded or covered intra-oceanic arc should be considered when calculating the net crustal growth in the CAOB.
6. Conclusions
(1) The turbidites in the North Tianshan Accretionary Complex are mainly composed of sandstone, siltstone and slate, and belong to greywackes.
(2) The sandstone and siltstone are characterized by immature geochemical features suggesting derivation from mafic to intermediate volcanic rocks, while the slates from a felsic magmatic source.
(3) The sandstone and siltstone display high positive εNd(t) values (+5.5 to +7.9) with only one exception of +0.8, implying derivation from juvenile crust rocks. Combining with previous studies, we suggest that the basement of the North Tianshan might be a Devonian to Carboniferous intra-oceanic arc.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/S0016756824000207
Acknowledgements
This research was supported by the National Natural Science Foundation of China (42072267, 41602229), the West light Foundation of the Chinese Academy of Sciences (XAB2020YW03), the Natural Science Fountain of Shaan’xi Province, China (2023-JC-YB-237), Opening Foundation of State Key Laboratory of Continental Dynamics, Northwest University (19LCD06), and the Youth Innovation Team of Shaanxi Universities.
Open access
