Multiple sources of the Upper Triassic flysch in the eastern Himalaya Orogen, Tibet, China: Implications to palaeogeography and palaeotectonic evolution

Multiple sources of the Upper Triassic flysch in the eastern Himalaya Orogen, Tibet, China: Implications to palaeogeography and palaeotectonic evolution

Tectonophysics 666 (2016) 12–22 Contents lists available at ScienceDirect Tectonophysics journal homepage: www.elsevier.com/locate/tecto Multiple s...

2MB Sizes 8 Downloads 53 Views

Tectonophysics 666 (2016) 12–22

Contents lists available at ScienceDirect

Tectonophysics journal homepage: www.elsevier.com/locate/tecto

Multiple sources of the Upper Triassic flysch in the eastern Himalaya Orogen, Tibet, China: Implications to palaeogeography and palaeotectonic evolution Xianghui Li a, Frank Mattern b, Chaokai Zhang a, Qinggao Zeng c, Guozheng Mao c a b c

School of Earth Sciences and Engineering, Nanjing University, Nanjing 210046, China College of Science/Earth Science Department, Sultan Qaboos University, PO Box 36, Oman The 1st Geology Survey, Tibet Bureau of Land and Resources, Lhasa 850000, China

a r t i c l e

i n f o

Article history: Received 19 March 2015 Received in revised form 16 August 2015 Accepted 8 October 2015 Available online 21 October 2015 Keywords: Provenance Palaeogeography Palaeotectonics Upper Triassic Langjiexue Group Himalaya Orogen

a b s t r a c t The Upper Triassic flysch—Langjiexue Group (tentatively named the “Shannan Terrane”) of the eastern Himalaya Orogen has been tectonically assigned either to the Tethys Himalaya or the Yarlung Zangbo Suture Zone (YZSZ). In this work, geochronology of detrital zircon U–Pb isotope shows that the Shannan Terrane is characterized by the population of ~ 260–200 Ma (peak ca. 240 Ma), strongly supporting the view of no affinity to the Tethys Himalaya. The detrital zircons dated as ~ 400–290 Ma display relatively positive εHf(t) values of − 5.0 to +15.0 with TCDM ages of 2.6–1.3 Ga for the Shannan Terrane, whereas highly negative are of −20.0 to −5.0 for the Lhasa Terrane, indicating that the two terranes have different Devonian–Carboniferous sources. Numerous Cr-spinels found in the Shannan Terrane but not in the Lhasa Terrane, exhibit contents in Cr2O3 and Cr# of mainly 44–100% and 48–95%, in TiO2 of 0.01–1.0%, and in Al2O3 of 5–257%, respectively, denoting several parent lithologies. These differences suggested that the Shannan Terrane has multiple sources, not only from the Lhasa Terrane, but also from oceanic (island) arc/seamount and mid-ocean ridge areas as well as likely from Greater India and Australia. Considering the Early Cretaceous diabase dykes within the Upper Triassic flysch defined to the Comei–Bunbury Large Igneous Province, we propose that the Langjiexue Group could have been deposited on the ocean between India and Australia, and would have not stopped till the Lhasa Terrane was separated from Australia during the terminal Triassic. According to the Cenozoic deformation and metamorphic history and palaeogeography of the Langjiexue Group, we postulate that the Shannan Terrane could have been loaded onto the Greater India during the middle Early Cretaceous, and subsequently drifted northward to the collision zone of India and Asia, implying that it does not represent an accretionary prism within the YZSZ. © 2015 Elsevier B.V. All rights reserved.

1. Introduction During the Late Triassic–Early Jurassic, the rifting reached its climax within the Pangaea supercontinent (e.g. Stampfli et al., 2013; Vaughan and Storey, 2007). Rifting and subsequent oceanic spreading resulted in the formation of the Neotethys, which separated the Cimmeride tectonic elements from Gondwanaland. The northward drifting of the Cimmeride elements led to the closure of the Palaeotethys (e.g. Ferrari et al., 2008; Metcalfe, 2009; Stampfli et al., 2013). In this stage, lots of huge marine basins were formed in both divergent and convergent settings, for instance, the Hoh Xil–Songpan–Ganzi Fold Zone represents the biggest flysch basin. It had formed during closure of Palaeotethys (Yin and Harrison, 2000). There is a similar Late Triassic flysch basin with over 400 × 600 km2 in size (Zhang et al., 2015), south to the Yarlung Zangbo River in the Shannan Prefecture, southern Tibet, China (Fig. 1). It is composed of

E-mail address: [email protected] (X. Li).

http://dx.doi.org/10.1016/j.tecto.2015.10.005 0040-1951/© 2015 Elsevier B.V. All rights reserved.

flysch sandstones, siltstones, and slates with strong deformation and low-grade metamorphism, namely the Upper Triassic Langjiexue Group (including the Nieru Formation. Li et al., 2011). This group lies tectonically south to the eastern YZSZ within the eastern Himalaya Orogen, and has been thought to be either a part of the northern deep sea Tethys Himalaya (Fig. 1B) in the northern Indian passive margin (e.g. Gansser, 1991; Wang et al., 1996; Yin and Harrison, 2000) or representing the eastern section of the accretionary prism within the YZSZ associated with the mélange of the Xiukang Group (e.g. Wang et al., 2013; Zhou et al., 1984). Obviously, this dispute has been hindering the reconstruction of the Late Triassic palaeogeography and palaeotectonics for the eastern Tethys, including the Cimmeride, Neotethys, and even northern Gondwanaland. However, results of recent geological investigations in two locations challenged the above traditional viewpoints. In the Qonggyai region (Fig. 1C), mainly southward directed palaeocurrents and orogenic sources as indicated by compositional aspects as well as geochemical and Nd isotope evidences revealed that the Langjiexue Group did not

X. Li et al. / Tectonophysics 666 (2016) 12–22

13

Fig. 1. Tectonic and geological sketches of South Tibet with sample locations. A, principal continental blocks, terranes and sutures of eastern Asia (simplified from Metcalfe, 2011), showing the location of the Shannan Terrane. B (inset), tectonic sketch with the Shannan Terrane in rectangle within the eastern Himalaya Orogen in southern Tibet, showing sample locations in neighbouring terranes. C, geological sketch of southern Tibet (simplified after Pan et al., 2004) with sample locations, numbers representing the samples (details refer to the legend and Table 1).

accumulate on the Indian passive margin (Li et al., 2003a,b). This new insight was also supported by data related to palaeocurrents, clastic compositions, heavy minerals, geochemistry and U–Pb isotopic age populations of detrital zircons from Rinbung county (Fig. 1C) in west (Li et al., 2010; Xu et al., 2011; Zeng et al., 2009). A derivation of either the Lhasa Terrane or a possible independent continent in the NeoTethys Ocean was then proposed (Li et al., 2004) and supported (Li et al., 2010, 2014; Webb et al., 2013). For testing new views, three questions need to be answered. 1) Is the proposition applicable for all rocks of the Upper Triassic flysch of southern Tibet? 2) Was the flysch only derived from the Lhasa Terrane, or possible from other source areas? 3) How did the Shannan Terrane arrive at its present place? We carried out widespread geological field investigations, took sandstone samples from the main outcrop region of the Langjiexue Group (Fig. 1C) and from other Triassic units of neighbouring terranes (Fig. 1B), and also took samples of diabase dykes within the flysch (Fig. 1C). We used data on detrital zircon U– Pb ages, Hf isotopes, and Cr-spinel geochemistry to put the new insights for the entire group to test with special consideration of the provenance, and tried to analyze the palaeogeography and palaeotectonics. For an easier description in this article, we preliminarily use the terms ‘Shannan Terrane’ and ‘Shannan Basin’ to relate to the flysch of the Upper Triassic Langjiexue Group when we tectonically discuss relationships with other terranes and sedimentary basins, respectively. 2. Materials and methods Twenty sandstone samples were taken from four terranes for extracting detrital zircon (Table 1), in which fourteen were from the Langjiexue Group in the Shannan Terrane, and two from the shallow marine Upper Triassic in the Tethys Himalaya, the mélange of the

Triassic Xiukang Group in the YZSZ, and the shallow marine Triassic of the southern Lhasa Terrane (Fig. 1C and Table 1). Within the Upper Triassic flysch, six samples of diabase dykes were taken for isolation of single zircon grains (Table 2) to measure the intrusive age, and eleven sandstones were taken for extracting detrital Cr-spinels (Table 3) to have geochemistry analyzed. One to three kg of each sandstone and eight to ten kg of each diabase sample were collected to be crushed in the lab. Heavy mineral grains were obtained by heavy-liquid separation and magnetic methods in the Laboratory of Langfang Geological Survey of Hebei Province, China. Over 200 detrital zircons and spinels of sandstones, and all single zircons of diabase were handpicked from each sample and pasted on resin discs under the microscope, and then polished to expose the grain centres. Grains were randomly selected in size and shape when pasted on disc. In this work, 75–80 detrital zircons from each sample (Table 1) were used for analysis of U–Pb isotope dating in order to include all recognizable minor populations. All zircons from the diabase samples were adopted to measure U–Pb isotopes (Table 2) for the dyke's intrusion age. Hf isotopes were analyzed only for those zircons dated as 400 Ma to 200 Ma. Ten to forty spinels of each sandstone sample were selected for geochemical analyses (Table 3). Measurements of zircon U–Pb and Hf isotopes and spinel geochemistry were carried out at State Key Laboratory for Mineral Deposits Research in Nanjing University, China. Zircon U–Pb isotope measurements were conducted by LA-ICP-MS using Agilent 7500a ICP-MS coupled to New Wave 213 nm laser ablation. For laser ablation we left out grains with obvious cracks or inclusions as much as possible. For details of the analytical procedures and conditions the reader may be referred to Liu et al. (2008). The U–Pb fractionation was corrected using zircon standard GEMOC GJ-1 with 207Pb/206Pb age of 601 ± 12 Ma, and accuracy was controlled using the zircon standard Mud Tank with an age of 735 ± 12 Ma (Black and Gulson, 1978).

14

X. Li et al. / Tectonophysics 666 (2016) 12–22

Table 1 List of sandstone samples of the Upper Triassic from southern Tibet, China and detrital zircons analyzed. No

Sample

Latitude

Tethys Himalaya (shallow sea) 1 Lixh 282,757.7N 2 715-01 290,307.6N

Longitude

Location

Strata

860,941.8E 873,229.6E

Derirong, Nyalam West Lhaze

Derirong Fm. “Nieru Fm.”

87 75

84 67

3 8

Xiukang, Lhaze, ~80 km to Xigaze Duikang, Lhaze South to Xialu, ~20 km to Xigaze

Xiukang Gr Xiukang Gr Xiukang Gr

80 80 62

70 73 58

10 7 4

This work This work Aitchison et al. (2011)

2 km SE to Bailang Aima, Rinbung Dejilin, Rinbung Dejilin, Rinbung Dejilin, Rinbung Nagarze Chegutsu Pass, ~30 km to Qonggyai Dala, Lhunze Gudui, Comei Xoisar, Lhunze San'an Qulin, Lhunze Zara, Lhunze Dejilin, Rinbung Dejilin, Rinbung Kari, Rinbung Gambala pass Bocun of Quxu, ~40 km to Gonggar Jiangxiong, Gonggar 2 km North to Zhela pass, Gonggar Baina, Qonggyai NW of Chegutsu Pass, Qonggyai Yala Xiangbo, Qonggyai

75 80 60 60 60 67 51 47 80 80 80 75 60 60 90 80 80 90 80 80 47 45 56 80 79 92

72 74 60 40 24 63 49 32 73 75 71 72 44 47 85 63 73 86 73 67 36 36 39 70 76 89

3 6 0 20 36 4 2 15 7 5 9 3 16 13 5 17 7 4 7 13 11 9 17 10 3 3

This work This work Li et al. (2010) Li et al. (2010) Li et al. (2010) This work Aikman et al. (2008)

Sewu, Qusum Purulang, Nang Baina, Qonggyai

Nieru Fm. Nieru Fm. Nieru Fm. Nieru Fm. Nieru Fm. Nieru Fm. Nieru Fm. Nieru Fm. Nieru Fm. Nieru Fm. Nieru Fm. Nieru Fm. Songre Fm. Songre Fm. Jiedexiu Fm. Songre Fm. Jiangxiong Fm. Jiangxiong Fm. Jiangxiong Fm. Songre Fm. Jiedexiu Fm. Songre Fm. Jiedexiu Fm. Songre Fm. Zhangcun Fm. Jiangxiong Fm.

Quesang, Duilongdeqing, ~60 km to Lhasa Mailonggang of Dhaze, ~100 km to Lhasa

Chaqupu Fm. Mailonggang Fm.

80 80 100 100 100 100 100 2878

59 76 96 88 94 99 99 2552

21 4 4 12 6 1 1 326

Yarlung Zangbo Suture Zone (Triassic melange) 3 616-01 290,748.1N 875,910.9E 4 616-02 290,747.3N 880,240.7E 5 06T013 290,508.1N 890,074.1E Shannan Terrane (deep sea) 6 719-T3n 290,411.0N 7 NR-06 290,626.2N 8 NR-01 Unavailable 9 NR-02 Unavailable 10 NR-03 Unavailable 11 TL01-001b 290,425.5N 12 410029 Unavailable 13 410031 Unavailable 14 TL12-01 283,423.2N 15 TL11-01 283,602.2N 16 TL10-04 283,244.9N 17 TL14-01 283,809.3N 18 SR-01 Unavailable 19 SR-05 Unavailable 20 817-12 291,415.5N 21 TL01-01 290,701.6N 22 TL01-08 291,711.1N 23 TL03-03 290,718.2N 24 PM026 290,024.1N 25 TL05-05 290,705.4N 26 410016 Unavailable 27 410003 Unavailable 28 410006 Unavailable 29 TL06-01 285,611.9N 30 TL08-01 284,857.2N 31 AY06-29-06-10A Unavailable

891,832.1E 895,349.3E Unavailable Unavailable Unavailable 902,345.5E Unavailable Unavailable 915,216.6E 923,114.5E 925,326.2E 932,309.0E Unavailable Unavailable 902,355.0E 902,556.8E 904,202.4E 910,636.5E 910,424.2E 914,239.9E Unavailable Unavailable Unavailable 921,046.4E 924,000.2E Unavailable

Southern Lhasa Terrane (shallow sea) 32 722-01 295,944.0N 904,520.0E 33 T3m-04 295,629.7N 912,821.1E 34 LZ11-2-5 Close to T3m-04 above 35 LZ11-2-7 Close to T3m-04 above 36 LZ11-2-8 Close to T3m-04 above 37 LZ11-2-9 Close to T3m-04 above 38 LZ11-2-11 Close to T3m-04 above

a

Sum

b

c

Sources This work This work

This work This work This work This work Li et al. (2010) Li et al. (2010) This work This work This work This work This work This work Aikman et al. (2008)

This work This work Webb et al. (2013)

This work This work Li et al. (2014) Li et al. (2014) Li et al. (2014) Li et al. (2014) Li et al. (2014)

Note: 1) a, zircon grains analyzed; b, zircon grains of concordant age; c, zircon grains of age disconcordant. 2) Those in column c from Aikman et al. (2008) are N3% in 1 sigma. 3) Those data from Aikman et al. (2008) are not as good as others by quantity of discordant zircons. 4) Samples from the Tethys Himalaya–Yarlung Zangbo Suture, and Lhasa Terrane are quartz arenite, others are lithic sandstone.

Common lead correction was carried out using the EXCEL programme for Pb correction (Andersen, 2002). Preferred U–Th–Pb isotopic ratios used for 91500 are from Wiedenbeck et al. (1995). 207Pb/206Pb ages were used for those grains older than 1000 Ma, and 206Pb/238U ages were selected for younger zircons. ISOPLOT (version 3.0; Ludwig, 2003) was

used for plotting age spectra and for age calculations. All ages with either N10% discordance or age error 1σ N 3.0% were excluded from data (crossed as in Supplement Tables DR1 and DR2). In-situ Hf isotopes were measured on a Neptune MC-ICP-MS, which is a double focusing multi-collector ICP-MS and has the capability of

Table 2 Zircon data from diabase within the Langjiexue Group. Number

Sample

GPS

a

Youngest age (Ma)

Sources

D1 D2 D3 D4 D5 D6 b D7 b D8 b D9

1307-02zk 2180WT 3036WT PM013-4zk TL01-3zk TL10-2-3zk YM04-1 YM04-4 QG3-2

N28°26′24″, E92°16′28″ N28°51′06″, E91°15′42″ N29°14′05″, E91°27′31″ N29°04′12″, E91°06′28″ N29°11′01″, E90°34′04″ N29°39′18″, E92°55′45″ N29°06′00″, E90°22′44″ N28°56′20″, E90°23′06″° N28°03′40″, E92°21′58″

4-4 (1) 18-14 (3) 13-8 (8) 20-10 (1) 15-13 (1) 10-10 (1) 16-12 (12) 15-13 (13) 14-13 (13)

139 ± 2 133 ± 3 140 ± 3 133 ± 4 114 ± 2 128 ± 4 134.1 ± 2 133.4 ± 1.6 131.1 ± 6.1

This work This work This work This work This work This work Jiang et al. (2006) Jiang et al. (2006) Zhu et al. (2009b)

a b

Zircons

The first number is the total number of zircons measured, the second one is the number of concordant zircons, and that in parenthesis is the number of the youngest zircons. Dating data using SHRIMP technique. For locations of samples refer to Fig. 1.

X. Li et al. / Tectonophysics 666 (2016) 12–22

15

high mass resolution measurements in a multiple collector mode. The measurements were done using LA-ICP-MS with a beam size of 35 μm and a laser pulse frequency of 10 Hz (80%). The calculation of initial 176 Hf/177Hf ratios and εHf(t) values were referred to the chondritic reservoir (CHUR) at the time of zircon growth from magmas. The 176Lu decay constant is adopted as 1.867 × 10−11 year−1 (Söderlund et al., 2004), and chondritic ratios of 176Hf/177Hf and 176Lu/177Hf are 0.282772 and 0.0332 (Blichert-Toft and Albarède, 1997; Griffin et al., 2000), respectively. Depleted mantle model ages (TDM) used for basic to intermediate rocks were calculated using a present-day 176Hf/177Hf ratio of 0.28325 with reference to the depleted mantle. The Hf isotope crustal model age (TCDM) was also calculated for each measured zircon under the assumption that the 176Lu/177Hf of average crust is 0.015 from the depleted mantle source (Griffin et al., 2002). All the Lu–Hf isotopic data of zircons are given in Supplement Table DR3. Spinel geochemistry was analyzed using a JEOL JXA-8100M electron microprobe. More details on methods, techniques, and procedures followed the description by Hu et al. (2010). It is noted that Ferric iron content was calculated following Barnes and Roeder (2001) as shown in Supplement Table DR4. In order to have reliable and valid compositions of detrital Cr-spinels, principles of probe analysis are obeyed: 1) close to grain cores and 2) away from microveins or altered zones after checking in black-scattered light. All geochemical data are attached as Supplement Table DR4.

(peak at ca. 1100 Ma), 600–500 Ma (peak at ca. 550 Ma), and 360– 250 Ma (peak at ca. 300 Ma). From the above ages, four distinct populations are summarized and indicated. The first and oldest one dates as 1220–1140 Ma (late MidMesoproterozoic). This can be matched in both West Australia and Lhasa Terrane (Fig. 2B–D) and indicates a document of the second stage of the Albany–Fraser Orogenesis between the southern margin of the North Australian Craton (e.g. Cawood and Nemchin, 2000; Veevers et al., 2005) and the western margin of the Mawson Craton (Clark et al., 2000). The second population dates as 1000–740 Ma (Mid–Early Neoproterozoic), which can be mainly found in the Tethys Himalaya (Fig. 2F–H) and subordinately in the mélange of the YZSZ. It may indicate rocks of the Pinjarran orogeny (Fitzsimons, 2003) as a source, corresponding to the Genville Orogeny (Boger, 2011). The third population dates as 600–460 Ma (Late Neoproterozoic– Middle Ordovician) and can be commonly observed in terranes with affinities to Gondwanaland (Fig. 2B–H). This population, thus, may broadly represent terranes of West Australia (Cawood and Nemchin, 2000; Veevers et al., 2005), the Qiangtang Terrane (Pullen et al., 2008), the Lhasa Terrane (Leier et al., 2007; Li et al., 2014; and this work), and the Tethys Himalaya (Zhu et al., 2011 and this work) as well as the Shannan Terrane (this work). The interval of ages corresponds to the famous Panafrican orogenesis (Kröner and Stern, 2004). The fourth and youngest population ranges from 300 Ma to 200 Ma (roughly Permian–Triassic), corresponding to the Pangaea supercontinent with regional geological events, such as the subduction–collision (Palaeotethys Suture) of the Cimmeride and Asia, the continental rifting within the northern Indian margin (opening of the Mesotethys Ocean). At the first glance it seems similar in the Upper Triassic sandstones from terranes of Lhasa, Songpan-Ganzi, and Shannan (Fig. 2A–H). However, distinct differences exist when making a detailed comparison (Fig. 2A–H). There are two sub-groups of 310–290 Ma and 275– 250 Ma for the Songpan–Ganzi Fold Zone (e.g. Weislogel et al., 2006), and one group of 310–290 Ma (peak at 300 Ma) for the Lhasa Terrane (Li et al., 2014, and this work), but a different sub-group 260–200 Ma (peak at ca. 240 Ma, Li et al., 2010, and this work) was only observed from the Shannan Terrane. These sub-groups indicate a discrepancy of sources among the three terranes.

3. Results and indications

3.2. U–Pb isotope ages of zircons from diabase

3.1. U–Pb isotope ages of detrital zircons

Ages from seventy zircons of six diabase samples (Supplement Table DR2) range between 2540 Ma and 128 Ma. Of them, some detrital zircon geochronological styles may be attributed to host rock capture from the Upper Triassic flysch, leading to a mixing of detrital and newly-crystalline zircons within the diabase dykes. The youngest zircon ages from the diabase are mainly 140 Ma to 128 Ma old (Table 2) with an average of 131.2 Ma, similar to the published data 131–134 Ma within the Shannan Terrane (Zhu et al., 2009b). Considering the age of ca. 132 Ma for rocks of the Comei–Bunbury Large Igneous Province between Australia and India (Zhu et al., 2009b), the 128–140 Ma diabase dykes within both the Jurassic–Cretaceous of the eastern Tethys Himalaya and the Upper Triassic of the Shannan Terrane may indicate that the intrusive diabase are coeval and cogenetic with the igneous rocks of the Comei–Bunbury Province.

Table 3 Detrital Cr-spinels of sandstones from the Upper Triassic flysch in southern Tibet. Number

Sample

GPS

Grains

C1 C2 a C3 C4 a C5 a C6 a C7 a C8 C9 C10 C11

TL3-02ZK TL4-01ZK TL4-05ZK TL4-06ZK TL5-01ZK TL5-07ZK TL8-01ZK TL11-07ZK TL13-03AZK TL13-03ZK TL13-04ZK

290,422.7N; 910,635.7E 290,050.4N; 911,832.5E 291,155.0N; 911,949.4E 291,420.7N; 912,027.5E 285,205.3N; 913,812.1E 290,857.3N; 914,318.3E 284,857.2N; 924,000.2E 283,920.2N; 923,301.1E 283,913.5N; 930,218.9E 283,913.5N; 930,218.9E 283,647.7N; 930,452.7E

N400 N100 N1000 90 N400 N1000 N200 N500 5 mg 70 mg 34

a

Samples for probe geochemical analysis.

Altogether 1678 zircons of the twenty-one sandstone samples were measured. Of them 1522 with b10% disconcordance and b3% by 1 sigma (1σ) error were selected for geochronological analysis (Table 1, Supplement Table DR1). Rounded to subhedral zircon crystals of 30–150 μm in size were seen from microscope images, and oscillatory zones were also observed in CL images. Th/U ratios range from 0.01 to 18.0, in which 0.01–0.1 are from 34 zircons and 5.0–18.0 from 11 grains (Supplement Table DR1), indicating that the analyzed grains are predominantly of magmatic origin. Concordant U–Pb ages of detrital zircons from sandstones in the Shannan Terrane range from 3942 Ma to 200 Ma (Supplement Table DR1 and Fig. DR1), from which two distinct populations of 600– 460 Ma and 260–200 Ma can be distinguished (Fig. 2E). Ages of detrital zircons from the Upper Triassic quartz arenites differ, 3219–447 Ma in the Tethys Himalaya and 3493–468 Ma in the mélange of the YZSZ (Supplement Table DR1 and Fig. 2F–H). Two populations of 1000– 740 Ma (peak at ca. 910 Ma) and 600–460 Ma (peak at ca. 530 Ma, 510 Ma) can be also distinguished in the Tethys Himalaya (Fig. 2F and G), similar to those from the Permian sandstones (Zhu et al., 2011) (comp. Fig. 2F–H). Those from the shallow marine Triassic sandstones of the southern Lhasa Terrane display ages of 3041–256 Ma (Supplement Table DR1), and three populations of 1300–1000 Ma

3.3. Hf isotopes of detrital zircons dated as 400–200 Ma Zircons with ages from 400 Ma to 200 Ma (Devonian–Triassic) were selected to undertake Hf isotope analysis. 175 zircons are eligible, and 11 were discarded as they are negative in both the TDM and TCDM age models (Supplement Table DR2). Detrital zircons with this population from the Shannan Terrane have εHf(t) values of − 9.6 to + 14.7 and mostly range − 5.0 to + 12.0 with featured TCDM ages ~ 1.3–0.5 Ga (Fig. 3). Those zircons with the same age range from the Lhasa Terrane have

16

X. Li et al. / Tectonophysics 666 (2016) 12–22

Fig. 2. Probability diagram of detrital zircon ages by terranes. Population peaks are shadowed in grey, n = total number of analyzed concordant zircons. Data details are given in Supplement Table DR1. A, n = 485 from Weislogel et al. (2006); B, n = 348 from Cawood and Nemchin (2000), n = 127 from Veevers et al. (2005); C, n = 72 from Leier et al. (2007), n = 475 from Zhu et al. (2011); D, n = 476 from Li et al., 2014, n = 135 from this work; E, n = 192 from Aikman et al. (2008), n = 215 from Li et al. (2010), n = 89 from Webb et al. (2013), n = 1093 from this work; F, n = 58 from Aitchison et al. (2011), n = 143 from this work; G, n = 151 from this work; H, n = 98 from Zhu et al. (2011).

εHf(t) values −20.6 to +13.9 with TCDM ages ~2.6–0.4 Ga, mostly show a similar source with the Shannan Terrane. However, zircons of ~ 400– 290 Ma, εHf(t) values − 20.0 to − 5.0 with TCDM ages ~ 2.5–1.3 Ga were observed (Fig. 3, right inlet), implying that the Lhasa Terrane predominantly had different parent rock sources formed in the Devonian– Carboniferous from the Shannan Terrane. 3.4. Geochemistry of detrital Cr-spinels 141 Cr-spinels from five sandstone beds of the Langjiexue Group were analyzed (Table 3, and Supplement Table DR4). The TiO2 content range is 0–2.0%, but mainly 0.01–1.0% (Fig. 4), and that of Al2O3 lies between 3

and 27%, with the emphasis between 5 and 27%. Ratios of Fe2+/Fe3+ greatly change by about − 40 to 50 (mainly about − 30 to 30). Both TiO2 content and Fe2+/Fe3+ ratios show mantle and volcanic origins of Cr-spinels (e.g. Kamenetsky et al., 2001). And the crossplot of TiO2– Al2O3 contents (Kamenetsky et al., 2001) indicates that the spinels chiefly originated from arc, suprasubduction zone, and mantle peridotites, and partly from oceanic island and mid-ocean ridge basalts (Fig. 4A). The contents of Cr2O3 and Cr# (Cr/(Cr + Al)) are 44–100% and 48– 95%, respectively, and are mainly characterized by 60–80% and 55–90% (Fig. 4B, Supplement Table DR4). The crossplot of TiO2–Cr# (Arai, 1992) indicates that their origin is from island arc basalts and boninites, possibly from mid-ocean ridge basalts.

X. Li et al. / Tectonophysics 666 (2016) 12–22

17

Fig. 3. Plots of εHf(t) vs. U–Pb ages of the detrital zircons (400–200 Ma) from the Shannan and Lhasa terranes. CHUR—chondritic uniform reservoir, n = total number of analyzed zircons. Details of the data and calculations of TCDM ages and depleted-mantle (DM) lines are provided in Supplement Table DR3. Pikang granite within the Lhasa Terrane were extracted from the Table 3 in the reference of Zhu et al., 2009a (Tectonophycs).

In summary, contents of TiO2, Al2O3, Cr2O3, Fe2+/Fe3+ ratio, and Cr# of measured spinels show multiple origins of source rocks, and they are dominated by island arc basalts and peridotites. To some degree, this seems consistent with those from the Carboniferous and Permian basalts within Tethys Himalaya, but there are minor differences in Cr# (see Fig. 4B), leading to the interpretation as deriving from continental rift flood basalts (Sciunnach and Garzanti, 1997). The Cr-spinel geochemistry is either similar or different between that of the Shannan Terrane and that of neighbouring terranes (the Xigaze Forearc Basin, the YZSZ, and the Tethys Himalaya), but they may have causal linkages or not because samples from neighbouring terranes are mainly from the Mesozoic and Early Cenozoic strata (Hu et al., 2014). It is very difficult to decipher their relationships due to dating problem of Cr-spinels. 4. Provenance analyses A number of data in both Qonggyai and Rinbung counties demonstrate that the Shannan Terrane is neither affinitive to the Tethys Himalaya, nor to the northern part of the Indian continent. These data include: 1) invariably southward palaeocurrent flow directions (e.g. Li

et al., 2003b); 2) sediments of dominantly recycled orogenic sources (e.g. Li et al., 2004); and 3) zircon age populations of 260–200 Ma (peak ca. 240 Ma. e.g. Li et al., 2010) that are absent in the pre-Triassic strata of the Tethys Himalaya. Li et al. (2004) proposed a possible source of either north or an independent continent within the Neo-Tethys Ocean. Li et al. (2010, 2014) and Webb et al. (2013) supported this viewpoint and suggested the provenance of the Lhasa Terrane. In this work, new chronological results of detrital zircons from the Langjiexue Group of the entire region (Figs. 1 and 2E) strongly support the idea that the Shannan Terrane has no affinities to the Tethys Himalaya (comp. Fig. 2E–H). Then does that mean the Upper Triassic flysch was only derived from the Lhasa Terrane? There are two distinct chronological populations of detrital zircons from the Upper Triassic flysch of the Shannan Terrane, from the marine Palaeozoic and Mesozoic of the Tethys Himalaya, and from the mélange of the YZSZ (Fig. 2E–H), and three populations are recognized with confidence in West Australia and Lhasa terranes (Fig. 2B–D). As we can see, of them, the chronological population of 460–600 Ma is common in terranes with affinities to Gondwanaland during the Panafrican Orogenesis (e.g. Kröner and Stern, 2004), pointing to possible sources from any of

Fig. 4. Diagram of Cr-spinel geochemistry of detrital zircons from the Upper Triassic flysch of the Shannan Terrane (this work) and from the Carboniferous and Permian (Sciunnach and Garzanti, 1997) of the Tethys Himalaya. A, TiO2–Al2O3 crossplot (after Kamenetsky et al., 2001. Modified by adding a new lower limit for TiO2); B, TiO2–Cr# crossplot (Arai, 1992). For details of sources refer to Supplement Table DR3.

18

X. Li et al. / Tectonophysics 666 (2016) 12–22

these terranes. That means all the terranes mentioned above could be related to the Gondwanaland before the Ordovician. On the other hand, if one considers the discrepancies of the Permian–Triassic chronological populations among the terranes, one may see that neither northern India (Tethys Himalaya), nor Lhasa nor West Australia are not the only sources of the Shannan Basin. Perhaps, it is difficult to explain why the Shannan Terrane lacks the age population of 1220–1140 Ma (Albany–Fraser orogenic stage, Clark et al., 2000) of the Lhasa Terrane and West Australia, and it is short of the population of 1000–740 Ma (Pinjarran and Genville orogeny, e.g. Fitzsimons, 2003) of the Tethys Himalaya. Superficially, both the Lhasa Terrane–West Australia and the Shannan Terrane have sources reflecting the 300 Ma to 200 Ma magmatism, but the discrepancy between them can be differentiated in detail. The Shannan Terrane has much more Triassic (ca. 260 Ma– 200 Ma) sources, and the Lhasa Terrane and West Australia have more Pennsylvanian–Early Permian (ca. 310 Ma–280 Ma) sources (comp. Fig. 2D and E enlargement inset). Those distinct differences suggest that the Lhasa Terrane was not the only source for the Shannan Basin. Hf isotopes of detrital zircons from the Upper Triassic sandstones of both the Shannan and Lhasa terranes can corroborate the new interpretation. Totally, the two terranes have similar Hf isotopes for detrital zircons aged 400–200 Ma (Fig. 3 left). The wide similar εHf(t) values are ~−5.0 to +15.0 with TCDM ages ~ 1.3–0.5 Ga (Fig. 3), indicating that source areas of both terranes are similar in lithology and suggesting they have similar multiple sources. However, a detailed look at the crossplot of εHf(t)-age, reveals that the Lhasa Terrane has highly negative εHf(t) values of − 20.0 to −5.0 with TCDM ages ~2.5–1.3 Ga for 400–290 Ma zircons (Fig. 3, right inlet), indicating its parent source of Devonian–Carboniferous rocks differs from the Shannan Terrane. We postulate that these rocks stacked within the Permian–Triassic orogen in the Lhasa Terrane (e.g. Chen et al., 2008; Dong et al., 2011) could have supplied the shallow sea basin in front of the terrane. This difference of Hf isotopes shows that the two terranes may not have had exactly the same sources for the Devonian–Carboniferous rocks. That means the source area with more negative Hf isotopes of zircons supplying the basin of the Lhasa Terrane could not have provided clastic material to the Shannan Basin during the Late Triassic. Given the Lhasa Terrane was one of the sources, how can one explain that the highly negative εHf(t) source was excluded from reaching the Shannan Basin? We suggest that the Permian–Triassic convergent margin of the Lhasa Terrane developed farther in the east, where the source with highly negative εHf(t) zircons could not reach the Shannan Basin farther in the west (refer to Section 5). The Cr-spinel geochemistry also shows a difference for the two basins. We have found lots of Cr-spinels in our sandstone samples of the Upper Triassic Langjiexue Group from the Shannan Terrane (Table 3), but there are none in the heavy mineral associations of seven sandstone samples of the Upper Triassic Mailonggang Formation from the Lhasa Terrane (see Supplementary data of Li et al., 2014). This may indicate that the Lhasa Terrane is not the unique supplier to the Shannan Basin. As shown in Section 3.4, the geochemistry of Cr-spinels demonstrates that the parent rocks are chiefly basalts, peridotites, and boninites (Fig. 4), implying that the possible source was the Lhasa Terrane, and/or mid-ocean ridges, and/or oceanic islands/seamounts. But the Carboniferous–Permian rift flood basalts of the Panjal Traps, the Gyrong, and Abor Volcanics within the western Tethys Himalaya (Sciunnach and Garzanti, 1997) can be precluded as the source due to their rifting properties. Why Li et al. (2010, 2014) and Webb et al. (2013) preferred an interpretation of only the Lhasa Terrane as the source, is possibly attributed to not only their unawareness of discrepancies related to age populations and εHf(t) value-TCDM ages between the Shannan and Lhasa terranes, but also by possibly paying too much attention to the Permian– Triassic orogen within the Lhasa Terrane (e.g. Chen et al., 2008; Dong et al., 2011). The Permian–Triassic orogen remains problematic because of its small scale (b 200 km) and because it may be a terrane coming far away from the Palaeotethys Suture and orogenic belt in north. Zhu et al.

(2013) interpreted the relevant Sumdo eclogites to be derived from the anatexis of ancient continental crustal material, with or without intraplate basalt-derived contributions. This interpretation is compatible with the εHf(t) values of detrital zircons, but not in harmony with the Cr-spinel geochemistry of this work, implying other sources for the Cr-spinel of the Langjiexue Group. If the Lhasa Terrane was the only source, the Langjiexue Group would be expected to show signs of deformation in the course of the accessionary prism formation at its southern side during the Jurassic and Cretaceous. Evidences from granitoid emplacement, foliation alternation, Kübler Index and vitrinite reflectance, demonstrate that deformation and metamorphism of the Langjiexue Group had chiefly occurred during the Palaeogene and Neogene (Antolín, et al., 2011; Dunkl et al., 2011). This suggests that the Langjiexue Group does not represent the accretionary prism within the YZSZ (e.g. Zhou et al., 1984) related to the Jurassic–Cretaceous subduction of the Neo-Tethys oceanic crust, ruling out that the Shannan Basin could have formed at the southern margin of the Lhasa Terrane. In summary, the Lhasa Terrane could be one of the source areas for the Shannan Basin as both have the same detrital zircon age populations originating from the Panafrican Orogenesis, and they also have partly similar Permian zircons in age range and Hf isotopes. Yet, the Lhasa Terrane is not the only source area because of its special age populations and Hf isotopes as well as its lack of Cr-spinel as a possible source. Thus, there must be additional source regions to have fed the Shannan Basin. 5. Palaeogeography and palaeotectonics Which other more source regions existed besides of the Lhasa Terrane and what are their tectonic relationships? As discussed above, the common age 600–460 Ma population of detrital zircons indicates the possibility for the Panafrican Orogeny within Gondwanaland to be considered as a sediment source to the Shannan Basin. With respect to this feature, there are many terranes to be considered as sources of the Shannan Terrane such as terranes of Lhasa, Australia, India, Indochina, Sibumasu, Qiangtang, etc. Except for Australia and India, others occur along the Palaeotethys Suture Zone in the Late Triassic. The others are optional. In order to narrow down the provenance options we need to go beyond to only consider the Panafrican Orogeny. The zircon age population of 300–200 Ma can provide a clue, in which the population 260–200 Ma is a relative feature that is helpful in tracking down one or more additional sources for the Shannan Basin. Terranes featuring the corresponding age populations (see summaries by Searle et al., 2012) were widespread along the Palaeotethys Suture Zone, including those in the Cimmeride to the south and those on the southern margin of Asia to the north (e.g. Ferrari et al., 2008). If all these terranes together would have fed ancient drainages along the Palaeotethys Orogen (e.g. Roger et al., 2008; Sengör and Natalin, 1996), remnant ocean/sea deltas and related submarine fan systems would have had to occupy both the eastern and western terminal outlets. A remnant basin was recognized in the eastern Bangong–Nujiang Zone between the Qiangtang and Lhasa terranes (Schneider et al., 2003). If this remnant basin likely was linked to the Shannan Basin, there are three questions to explain. 1), What is the remnant Bangong– Nujiang basin? Is it a tectonic relict or a remnant ocean/sea basin at the terminal outlet of an orogen? Unfortunately no data are available. 2), The eastern remnant Bangong–Nujiang basin was geographically far away from the eastern terminal of the Palaeotethys suture/orogen. How did the Shannan Terrane move to its present position? Such an origin of the Shannan Terrane would call for a strike–slip megafault in E–W direction. Schneider et al. (2003) concluded that there could have been a Permian and Triassic sinistral strike–slip zone between the Lhasa Terrane and the Qiangtang/Songpan Ganzi Terrane, but the Shannan Terrane requires Jurassic–Cretaceous strike–slip, and this would have happened in

X. Li et al. / Tectonophysics 666 (2016) 12–22

the Neotethys Ocean, not in Mesotethys Ocean. 3), Clastic composition, chronology and geochemistry of detrital zircons, tectonics and other from the eastern Bangong–Nujiang remnant basin need to match the Shannan Basin. Such data are not available. Therefore, it is difficult to link the Bangong–Nujiang remnant basin to the Shannan Basin in origin till present. Similar problems would be solved if one would link the Shannan Basin to other relevant basins within the Palaeotethys Suture between the Cimmeride and South Asia. Perhaps, the disappeared remnant basin in the eastern terminal outlet of the Palaeotethys suture/orogen between West Sumatra and E Malaya (Indochina) is a candidate. It is extremely challengeable and difficult for its moving to present (Shannan Terrane) location if so. In other words, it has to have westward proceeded by a sinistral strike–slip megafault with an extent of at least 3000 km movement while the Neotethys ocean was convergent to South Asia. According to plate tectonic reconstruction, the Palaeotethys Suture Zone and relevant terranes were unexceptionally reconstructed north to the ancient equator (e.g. Ferrari et al., 2008; Metcalfe, 2009). If the Shannan Basin was related to the Palaeotethys Suture, it should be located in the northern hemisphere during the latest Triassic. This to great degree conflicts the fact that the Shannan Terrane was confined in southern hemisphere, near which the Comei–Bunbury Large Igneous Province (Zhu et al., 2009b) developed in the eastern Gondwanaland by abundant coeval diabase dykes. The existence of the large igneous province was demonstrated in both Australia (e.g. Coffin et al., 2002) and the eastern Tibetan Tethys Himalaya (Zhu et al., 2009b). In southern Tibet, basaltic lavas, mafic sills and dykes, gabbroic intrusions together with subordinate layered ultramafic intrusions and silicic volcanic rocks were found and dated as ca. 132 Ma (Zhu et al., 2009b). Igneous rocks occur within both the marine Jurassic and Cretaceous of the Tethys Himalaya Superterrane and the Upper Triassic flysch of the Shannan Terrane. Diabase dykes of this age were identified within the Langjiexue Group at Qonggyai (sample D9, Fig. 1; Zhu et al., 2009b), Nagarze (samples D7 and D8, Fig. 1; Jiang et al., 2006), and other places (Fig. 1 and Table 2; this work), respectively. The youngest zircons of samples D1–D6 diabase dykes were dated as 130–140 Ma, but some zircons are much older (Table 2) due to the inclusion of detrital zircons from the Upper Triassic sedimentary host rock during intrusion. Thus, the Shannan Terrane was associated with the Comei–Bunbury Igneous Province, in contrast to the Lhasa Terrane. It seems unrealistic to assume that the Shannan Terrane could have moved back to the northern edge of Gondwanaland during the Early Cretaceous, having been in the Palaeotethys Suture Zone during the Latest Triassic. It is possible that microplates could have moved long distances during an interval of ca. 70 Myr. However, the Neotethys had been spreading during the Jurassic–Early Cretaceous, and it is impossible that the Shannan Terrane traversed the spreading mid-ocean ridge within the Neotethys oceanic crust. The above palaeogeographic constraints allows for both eastern India and western Australia to be considered as sources to the Shannan Basin. For this linkage, the palaeogeographic role of the Lhasa Terrane is crucial. In the past, the Lhasa Terrane was mostly assigned a position in the northern hemisphere by the Cimmerian Superterrane after Permian (e.g. Ferrari et al., 2008; Stampfli et al., 2013), and also pointed to the north of India before the Triassic (e.g. Metcalfe, 2009; Yin and Harrison, 2000). The later palaeogeographic postulation led to a rift basin model for the flysch of the Langjiexue Group while the Lhasa Terrane was suggested to be the single source area (e.g. Li et al., 2010). We argue against the rift basin model as there are at least five reasons. Firstly, the Lhasa Terrane could have been detached from western Australia since the Triassic (e.g. Zhu et al., 2011, 2013). Secondly, detrital zircon Hf isotope and detrital Cr-spinel geochemistry support the idea of dissimilar sources for the two terranes. Thirdly, it is difficult to interpret the Shannan Terrane as the accessionary prism (Wang et al., 2013) of the Neotethys Ocean's Jurassic–Cretaceous subduction complex south to the Lhasa Terrane because the Upper Triassic flysch has the younger

19

Cenozoic deformation and metamorphism (e.g. Antolín, et al., 2011; Dunkl et al., 2011; see Section 4). Fourthly, if the Shannan Terrane had been accreted to the south of the Lhasa Terrane (southern Asia), its position in the northern hemisphere would be implied, but it was still a part of Gondwanaland in the southern hemisphere as indicated by the diabase dykes of the Comei–Bunbury Igneous Province (see above). At last, it is highly unlikely that the Shannan Terrane was removed back to the northern Gondwanaland (Indian) margin by southward transpassing the Neotethys mid-ocean ridge in the Early Cretaceous if it is a part of the Lhasa block during the Late Triassic through the Jurassic. Therefore, it is more reasonable to consider that the Lhasa Terrane was detached from western Australia during the (terminal) Triassic (Zhu et al., 2013), at that time the Lhasa Terrane was located northwest of Australia and separated from Australia by the primary (backarc) Neotethys Ocean (Zhu et al., 2011, 2013). With respect to the Lhasa Terrane's location and the extended submarine fan system (over 400 × 600 km2) and the elongate fan shape (Zhang et al., 2015), we propose a new palaeogeographic model for the Shannan Terrane (Fig. 5a and b). According to this model, the Late Triassic of the Shannan Terrane could have accumulated on the oceanic floor between northeastern India and northwestern Australia, where the two subcontinents had been neighbours from the Neoproterozoic to the Early Cretaceous (refer to Clark et al., 2000; Boger, 2011). In this model, several parent areas could have coevally existed in relative proximity (Fig. 5a and b). To the west, some pre-Triassic sediments of Greater India (disappeared parts) may have been somewhat reworked by bottom currents and were sparsely introduced to the eastern Shannan Basin at the same time. Towards the southeast, parts of West Australia may have provided detritus that is related to the Panafrican Orogeny. To the northeast, the western part of the Lhasa Terrane with its northern forearc and southern backarc could have been feeding the southwestern Shannan Basin. To the north, disappeared oceanic structural or volcanic highs (oceanic islands, seamounts and/or a mid-ocean ridge) may have supplied fragments of oceanic basalt and peridotite (Fig. 5a) by special oceanic water mass/oceanic current. All these microcontinent positions are compatible with the observed palaeocurrent data (Li et al., 2003b). Not only does this palaeogeographic model explain for multiple sources of the Shannan Basin by the geochronological and geochemical data, it also drives forward the understanding of the regional palaeotectonic evolution (Fig. 5b–d). During the Late Triassic, the Langjiexue Group had been deposited in the deep Shannan Basin between India and Australia, in which submarine fans dominated sedimentation (Fig. 5a and b). By drifting northward, the Lhasa Terrane and northern oceanic highs became gradually less involved as sediment suppliers to the Shannan Basin. They stopped shedding material into the basin by the end of the Triassic. Since then, the Shannan Terrane could have been upheaved to the eastern edge of Greater India (Fig. 5c) and continued this way to the collision zone of India with Asia at about 60 Ma–50 Ma (e.g. Decelles et al., 2014; Rowley, 1996). It is causal that the Shannan Terrane loading on the eastern Greater India (Fig. 5c and d) has not been altered even on the drifting way to South Asia since the rifting and opening of Indian Ocean at ca. 132 Ma (Zhu et al., 2009b). It would have been deformed while Greater India was subducted beneath Asia, causing ˃70% of shortening (e.g. Patzelt et al., 1996; Robinson et al., 2006) and 20° of clockwise rotation (Antolín, et al., 2011) of the Langjiexue Group. In addition, it could have become the basin basement of the deep sea Jurassic-Cretaceous sediments during the way to South Asia.

6. Summary From this work, we arrived at new results and conclusions regarding the provenance, palaeogeography and palaeotectonics of the Upper Triassic (Langjiexue Group) flysch of the Shannan Terrane in the eastern Himalaya Orogen.

20

X. Li et al. / Tectonophysics 666 (2016) 12–22

Fig. 5. Reconstructed palaeogeography sketch of the Shannan and neighbouring terranes, illustrating the tectonic evolution. a), The submarine fan dominated deep Shannan Basin of the Langjiexue Group, showing the relationship to multiple source areas. b), The Shannan Basin with relationship to other source terranes in eastern Gondwanaland during the Late Triassic (ca. 230–210 Ma), also showing the Palaeotethys Suture Zone and relevant terranes. c), Terranes and continents around the Neotethys Ocean (YZSZ) during the Early Cretaceous (ca. 140– 130 Ma), including the locations of the Shannan Terrane and the Comei–Bunbury Large Igneous Province (LIP). d), Palaeogeography of the Shannan Terrane drifting northward along with Greater India during the Late Cretaceous (80–70 Ma). b)–d). Background map was modified from Zhu et al. (2013).

1) There are two age populations of detrital zircons from the Upper Triassic flysch sandstones of the Shannan Terrane, 600–460 Ma (peak ca. 520 Ma) and 260–200 Ma (peak ca. 240 Ma), respectively. The latter strongly supports the view of disaffinity to the Tethys Himalaya. The former, related to the Panafrican orogen, indicates that the Shannan Terrane's detritus could derive from the Lhasa Terrane, West Australia, the Qiangtang Terrane, and other areas of Gondwanaland, or not. 2) Besides the geochronology of detrital zircons, there are differences in Hf isotope and Cr-spinels between the Shannan and Lhasa terranes. The detrital zircons dated as 400–290 Ma are highly negative −20.0 to −5.0 in εHf(t) value for the Lhasa Terrane, but relatively positive −5.0 to 10 in εHf(t) value for the Shannan Terrane. Whereas there are lots of Cr-spinels in the Shannan Terrane, there are none in the Lhasa Terrane; and the geochemistry of Cr-spinels (Cr2O3 44–100%, Cr# 48– 95%, TiO2 0.01–1.0%, Fe2+/Fe3+ ratio, and Al2O3 of 5–25%) indicates several parent lithologies. 3) The differences in geochronology and Hf isotopes of detrital zircons and in existence and geochemistry of detrital Cr-spinels suggest that the Shannan Terrane could have multiple sources, not only from the Lhasa Terrane, but also from oceanic (island) arcs/seamounts and mid-ocean ridge areas as well as from Greater India and West Australia. 4) The mid-Early Cretaceous (ca. 140–128 Ma) diabase dykes within the Upper Triassic flysch are probably related to the Comei–Bunbury Igneous Province. These diabase dykes would confine the Shannan Terrane in eastern Gondwanaland (South Hemisphere) during the Early

Cretaceous. Together with the multiple sources, we propose a new palaeogeographic and basin model. This entails that the Langjiexue Group of the Shannan Terrane could have formed on a deep oceanic floor between India and Australia, and would have not stopped till the Lhasa Terrane was completely separated from Australia during the terminal Triassic. 5) By the Cenozoic deformation history of the Langjiexue Group and its proposed palaeogeographical model, we postulate that the Shannan Terrane could have been upheaved onto the eastern side of Greater India since the Indian Ocean rifting during the mid-Early Cretaceous. Afterwards it was drifting northward to the collision zone of India and Asia. This palaeogeography and palaeotectonic evolution implies that the Shannan Terrane is not a Jurassic to Cretaceous accretionary prism within the YZSZ. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.tecto.2015.10.005. Acknowledgements We are grateful to Evgueni Burov and Guangwei Li for their helpful comments and constructive suggestions, and to Xiumian Hu for discussions. Yin Wang, Yong Sun, Wenli Xu, Kai Luo, Feng Lu, and others are thanked for partly participating in the field work. We thank Dicheng Zhu for providing the digital map as the background map of Fig. 5. Thanks are acknowledged to the National Natural Sciences Foundation

X. Li et al. / Tectonophysics 666 (2016) 12–22

of China (project 41072075) and to the Chinese Bureau of Geological Survey (South Zhanang 1: 50,000 geological mapping project, 12120113034800) for partly financially supporting this work. References Aikman, A.B., Harrison, T.M., Ding, L., 2008. Evidence for Early (N44 Ma) Himalayan Crustal Thickening, Tethyan Himalaya, southeastern Tibet. Earth Planet. Sci. Lett. 274, 14–23. Aitchison, J.C., Xia, Xiaoping, Baxter, A.T., Ali, J.R., 2011. Detrital zircon U–Pb ages along the Yarlung–Tsangpo suture zone, Tibet: implications for oblique convergence and collision between India and Asia. Gondwana Res. 20, 691–709. Andersen, T., 2002. Correction of common Pb in U–Pb analyses that do not report 204Pb. Chem. Geol. 192, 59–79. Antolín, B., Appel, E., Montomoli, C., Dunkl, I., Ding, L., Gloaguen, R., El Bay, R., 2011. Kinematic evolution of the eastern Tethyan Himalaya: constraints from magnetic fabric and structural properties of the Triassic flysch in SE Tibet. In: Poblet, J., Lisle, R.J. (Eds.), Kinematic Evolution and Structural Styles of Fold-and-Thrust Belts. Geol. Soc. Lond., Spec. Publ. 349, pp. 99–121. http://dx.doi.org/10.1144/SP349.6. Arai, S., 1992. Chemistry of chromian spinel in volcanic rocks as a potential guide to magma chemistry. Mineral. Mag. 56 (383), 173–184. Barnes, S.J., Roeder, P.L., 2001. The range of spinel compositions in terrestrial mafic and ultramafic rocks. J. Petrol. 42 (12), 2279–2302. Black, L.P., Gulson, B.L., 1978. The age of the mud tank carbonatite, strang ways range northern Territory. Bureau of Mineral Resources. J. Aust. Geol. Geophys. 3, 227–232. Blichert-Toft, J., Albarède, F., 1997. The Lu–Hf geochemistry of chondrites and the evolution of the mantle–crust system. Earth Planet. Sci. Lett. 148, 243–258. Boger, S.D., 2011. Antarctica — before and after Gondwana. Gondwana Res. 19, 335–371. http://dx.doi.org/10.1016/j.gr.2010.09.003. Cawood, P.A., Nemchin, A.A., 2000. Provenance record of a rift basin: U/Pb ages of detrital zircons from the Perth Basin, Western Australia. Sediment. Geol. 134, 209–234. Chen, S.Y., Yang, J.S., Xu, X.Z., Li, H.Q., Yang, Y.H., 2008. Study of Lu–Hf geochemical tracing and LA-ICPMS U–Pb isotopic dating of the Sumdo eclogite from the Lhasa block, Tibet. Acta Petrol. Sin. 24 (7), 1528–1538. Clark, D.J., Hensen, B.J., Kinny, P.D., 2000. Geochronological constraints for a two-stage history of the Albany–Fraser Orogen, Western Australia. Precambrian Res. 102, 155–183. Coffin, M.F., Pringle, M.S., Duncan, R.A., Gladczenko, T.P., Storey, M., Müller, R.D., Gahagan, L.A., 2002. Kerguelen hotspot magma output since 130 Ma. J. Petrol. 43, 1121–1139. DeCelles, P.G., Kapp, P., Gehrels, G.E., Ding, L., 2014. Paleocene–Eocene foreland basin evolution in the Himalaya of southern Tibet and Nepal: implications for the age of initial India–Asia collision. Tectonics 33, 824–849. http://dx.doi.org/10.1002/2014TC003522. Dong, X., Zhang, Z.M., Liu, F., Wang, W., Yu, F., Shen, K., 2011. Zircon U–Pb geochronology of the Nyainqentanglha Group from the Lhasa terrane: new constraints on the Triassic orogeny of the south Tibet. J. Asian Earth Sci. 42, 732–739. http://dx.doi.org/10. 1016/j.jseaes.2011.01.014. Dunkl, I., Antolín, B., Wemmer, K., Rantitsch, G., Kienast, M., Montomoli, C., Ding, L., Carosi, R., Appel, E., El Bay, R., Xu, Q., Von Eynatten, H., 2011. Metamorphic evolution of the Tethyan Himalayan flysch in SE Tibet. In: Gloaguen, R., Ratschbacher, L. (Eds.), Growth and Collapse of the Tibetan Plateau. Geol. Soc. Lond., Spec. Publ. 353, pp. 45–69. http://dx.doi.org/10.1144/SP353.4. Ferrari, O.M., Hochard, C., Stampfli, G.M., 2008. An alternative plate tectonic model for the Palaeozoic–Early Mesozoic Palaeotethyan evolution of Southeast Asia (Northern Thailand–Burma). Tectonophysics 451, 346–365. http://dx.doi.org/10.1016/j.tecto. 2007.11.065. Fitzsimons, I.C.W., 2003. Proterozoic basement provinces of southwestern Australia, and their correlation with Antarctica. In: Yoshida, Y., Windley, B.F., Dasgupta, S. (Eds.), Proterozoic East Gondwana: Supercontinent Assembly and Breakup. Geol. Soc. Lond., Spec. Publ. 206, pp. 93–130. Gansser, A., 1991. Facts and theories on the Himalayas. Ecologae Geol. Helv. 84, 33–60. Griffin, W.L., Pearson, N.J., Belousova, E., Jackson, S.E., van Achterbergh, E., O'Reilly, S.Y., Shee, S.R., 2000. The Hf isotope composition of cratonic mantle: LAM-MC-ICPMS analysis of zircon megacrysts in kimberlites. Geochim. Cosmochim. Acta 64, 133–147. Griffin, W.L., Wang, X., Jackson, S.E., Pearson, N.J., O'Reilly, S.Y., Xu, X., Zhou, X., 2002. Zircon chemistry and magma mixing, SE China: in-situ analysis of Hf isotopes, Tonglu and Pingtan igneous complexes. Lithos 61, 237–269. Hu, X.M., Jansa, L., Chen, L., Griffin, W.L., O'Reilly, S.Y., Wang, J.G., 2010. Provenance of Lower Cretaceous W long volcaniclastics in the Tibetan Tethyan Himalaya: implications for the final breakup of eastern Gondwana. Sediment. Geol. 223 (3–4), 193–205. Hu, X.M., An, W., Wang, J.G., Garzanti, E., Guo, R.H., 2014. Himalayan detrital chromian spinels and timing of Indus–Yarlung ophiolite erosion. Tectonophysics 621, 60–68. Jiang, S.H., Nie, F.J., Hu, P., Liu, Y., 2006. Important spreading event of the Neo-Tethys ocean during the Late Jurassic and Early Cretaceous: evidence from Zircon U–Pb SHRIMP dating on diabase in Nagarze, southern Tibet. Acta Geol. Sin. 80, 522–527. Kamenetsky, V.S., Crawford, A.J., Meffre, S., 2001. Factors controlling chemistry of magmatic spinel: an empirical study of associated olivine, Cr-spinel and melt inclusions from primitive rocks. J. Petrol. 42 (4), 655–671. Kröner, A., Stern, R.J., 2004. Encyclopedia of Geology. Elsevier, Amsterdam, pp. 1–12. Leier, A.L., Kapp, P., Gehrels, G.E., DeCelles, P.G., 2007. Detrital zircon geochronology of Carboniferous–Cretaceous strata in the Lhasa terrane, Southern Tibet. Basin Res. 19, 361–378. Li, X.H., Zeng, Q.G., Wang, C.S., 2003a. Sedimentary characteristics of the Upper Triassic Langjiexue Group south to Qiongjie, Tibet. Modern Geol. 17 (1), 52–58 (in Chinese with English abstract).

21

Li, X.H., Zeng, Q.G., Wang, C.S., 2003b. Palaeocurrent data: evidences for source of the Langjiexue Group in southern Tibet. Geol. Rev. 49 (2), 132–137 (in Chinese with English abstract). Li, X.H., Zeng, Q.G., Wang, C.S., 2004. Provenance analysis of the Upper Triassic Langjiexue Group in the Southern Tibet, China. Acta Sedimentol. Sin. 22 (4), 553–559 (in Chinese with English abstract). Li, G.W., Liu, X.H., Pullen, A., Wei, L.J., Liu, X.B., Huang, F.X., Zhou, X.J., 2010. In-situ detrital zircon geochronology and Hf isotopic analyses from Upper Triassic Tethys sequence strata. Earth Planet. Sci. Lett. 297, 461–470. Li, X.H., Wang, Y., Xu, W.L., Sun, Y., Kong, Q.Y., Zeng, Q.G., Xie, Y.W., Mao, G.Z., Nima, C.R., Zhou, Y., Liu, L., 2011. Contrasting the Upper Triassic Flysch Langjiexue Group and Nieru Formation in Southern Tibet. Acta Geol. Sin. 85 (10), 1551–1562 (in Chinese with English abstract). Li, G.W., Sandiford, M., Liu, X.H., Xu, Z.Q., Wei, L.J., Li, H.Q., 2014. Provenance of Late Triassic sediments in central Lhasa terrane, Tibet and its implication. Gondwana Res. 25 (4), 1680–1689. http://dx.doi.org/10.1016/j.gr.2013.06.019. Liu, Y.S., Hu, Z.C., Gao, S., Günther, D., Xu, J., Gao, C.G., Chen, H.H., 2008. In situ analysis of major and trace elements of anhydrous minerals by LA-ICP-MS without applying an internal standard. Chem. Geol. 257, 34–43. Ludwig, K.R., 2003. ISOPLOT 3.00: A Geochronological Toolkit for Microsoft Excel. Spec. Publ. 4. Berkeley Geochronology Center, Berkeley, California, p. 74. Metcalfe, I., 2009. Late Palaeozoic and Mesozoic tectonic and palaeogeographic evolution of SE Asia. In: Buffetaut, E., Cuny, G., Le Loeuff, J., Suteethorn, V. (Eds.), Late Palaeozoic and Mesozoic Ecosystems in SE Asia. Geol. Soc. Lond., Spec. Publ. 315, pp. 7–23. Metcalfe, I., 2011. Tectonic framework and Phanerozoic evolution of Sundaland. Gondwana Research 19, 3–21. http://dx.doi.org/10.1016/j.gr.2010.02.016. Pan, G.T., Ding, J., Yao, D.S., Wang, L.Q., 2004. Geological Map of the Tibetan Plateau and Adjacent Areas (1: 1,500,000). Chengdu Map Publishing House, Chengdu, China (English version). Patzelt, P., Li, H.M., Wang, J.D., 1996. Palaeomagnetism of Cretaceous to Tertiary sediments from southern Tibet: evidence for the extent of the northern margin of India prior to the collision with Eurasia. Tectonics 259, 259–284. Pullen, A., Kapp, P., Gehrels, G.E., Vervoort, J.D., Ding, L., 2008. Triassic continental subduction in central Tibet and Mediterranean-style closure of the Paleo-Tethys Ocean. Geology 36, 351–354. http://dx.doi.org/10.1130/G24435A.1. Robinson, D.M., DeCelles, P.G., Copeland, P., 2006. Tectonic evolution of the Himalayan thrust belt in western Nepal: implications for channel flow models. GSA Bull. 118, 865–885. Roger, F., Jolivet, M., Malavieille, J., 2008. Tectonic evolution of the Triassic fold belts of Tibet. C. R. Geosci. 340, 180–189. http://dx.doi.org/10.1016/j.crte.2007.10.014. Rowley, D.B., 1996. Age of initiation of collision between India and Asia: a review of stratigraphic data. Earth Planet. Sci. Lett. 145, 1–13. Schneider, W., Mattern, F., Wang, P.J., Li, C., 2003. Tectonic and sedimentary basin evolution of the eastern Bangong/Nujiang zone (Tibet): a reading cycle. Int. J. Earth Sci. (Geologische Rundschau) 92, 228–254. Sciunnach, D., Garzanti, E., 1997. Detrital chromian spinels record tectono-magmatic evolution from Carboniferous rifting to Permian spreading in Neotethys (India, Nepal and Tibet). Ofioliti 22 (1), 101–110. Searle, M.P., Whitehouse, M.J., Robb, J., Ghani, A.A., Hutchison, C.S., Sone, M., Ng, S.W.-P., Roselee, M.H., Chung, S.L., Oliver, G.J.H., 2012. Tectonic evolution of the Sibumasu– Indochina terrane collision zone in Thailand and Malaysia: constraints from new U–Pb zircon chronology of SE Asian tin granitoids. J. Geol. Soc. Lond. 169, 489–500. http://dx.doi.org/10.1144/0016-76492011-107. Sengör, A.M.C., Natalin, B.A., 1996. Paleotectonics of Asia, fragments of a synthesis. In: Yin, A., Harrison, T.M. (Eds.), Tectonic Evolution of Asia. Cambridge Univ. Press, Cambridge, pp. 486–640. Söderlund, U., Patchett, P.J., Vervoort, J.D., Isachsen, C.E., 2004. The 176Lu decay constant determined by Lu–Hf and U–Pb isotope systematics of Precambrian mafic intrusions. Earth Planet. Sci. Lett. 219, 311–324. Stampfli, G.M., Hochard, Vérard, C.C., Wilhem, C., von Raumer, J., 2013. The formation of Pangea. Tectonophysics 593, 1–19. Vaughan, A.P.M., Storey, B.C., 2007. A new supercontinent self-destruct mechanism: evidence from the Late Triassic–Early Jurassic. J. Geol. Soc. Lond. 164, 383–392. Veevers, J.J., Saeed, A., Belousov, E.A., Griffin, W.L., 2005. U–Pb ages and source composition by Hf-isotope and trace-element analysis of detrital zircons in Permian sandstone and modern sand from southwestern Australia and a review of the paleogeographical and denudational history of the Yilgarn Craton. Earth-Sci. Rev. 68, 245–279. http://dx.doi.org/10.1016/j.earscirev.2004.05.005. Wang, C.S., Xia, D.X., Zhou, X., Chen, J.P., Lu, Y., Wang, G.H., He, Z.W., Li, X.H., Wan, X.Q., Zeng, Q.G., Pubu, C.R., Liu, Z.H., 1996. Field Trip Guide: T121/T387 Geology between the Indus-YZSZ and the Himalaya Mountains (Xizang), China. Geological Publishing House, Beijing, p. 72. Wang, Liquan, Pan, Guitang, Ding, Jun, and Yao, Dongsheng, 2013, Geological Mapping illumination (1:1500000) of Qinghai–Tibet Plateau and Its adjacent area: Beijing, Geological Publishing House. Webb, A.A.G., Yin, A., Dubey, C.S., 2013. U–Pb zircon geochronology of major lithologic units in the eastern Himalaya: implications for the origin and assembly of Himalayan rocks. GSA Bull. 125 (3/4), 499–522. http://dx.doi.org/10.1130/B30626.1. Weislogel, A.L., Graham, S.A., Chang, E.Z., Wooden, J.L., Gehrels, G.E., Yang, H.S., 2006. Detrital zircon provenance of the Late Triassic Songpan–Ganzi complex: sedimentary record of collision of the North and South China blocks. Geology 34 (2), 97–100. Wiedenbeck, M., Alle, P., Corfu, F., Griffin, W.L., Meier, M., Oberli, F., Quadt, A.V., Roddick, J.C., Spiegel, W., 1995. Three natural zircon standards for U–Th–Pb, Lu–Hf, trace element and REE analyses. Geostand. Geoanal. Res. 19, 1–23. Xu, W.L., Li, X.H., Wang, Y., Zeng, Q.G., Sun, Y., Nima, C.R., 2011. Provenance analysis of the Upper Triassic Flysch in Renbu Area, Southern Tibet. Geol. J. China Univ. 17 (2), 220–230 (in Chinese with English abstract).

22

X. Li et al. / Tectonophysics 666 (2016) 12–22

Yin, A., Harrison, T.M., 2000. Geologic evolution of the Himalayan–Tibetan Orogen. Annu. Rev. Earth Planet. Sci. 28 (1), 211–284. Zeng, Q.G., Li, X.H., Xu, W.L., Nima, C.R., Cui, Y.Q., Li, J., 2009. Heavy mineral assemblages and provenance analysis of the Upper Triassic in Renbu area, southern Tibet, China. Geol. Bull. China 28 (1), 38–44 (In Chinese with English summary). Zhang, C.K., Li, X.H., Mattern, F., Mao, G.Z., Zeng, Q.G., Xu, W.L., 2015. Architectures of lithofacies and deposystems of a submarine fan-dominated deep sea succession in an orogen: a case study from the Upper Triassic of southern Tibet. J. Asi. Ear. Sci. 111, 222–243. http://dx.doi.org/10.1016/j.jseaes.2015.07.013. Zhou, X., Cao, Y.G., Zhu, Y.M., Qian, D.Y., Xia, D.X., 1984. Map of Plate Tectonics in Tibet and Notes. Geological Publishing House, Beijing (in Chinese with English abstract). Zhu, D.C., Mo, X.X., Niu, Y.L., Zhao, Z.D., Wang, L.Q., Pan, G.T., Wu, F.Y., 2009a. Zircon U–Pb dating and in-situ Hf isotopic analysis of Permian peraluminous granite in the Lhasa

terrane, southern Tibet: implications for Permian collisional orogeny and paleogeography. Tectonophysics 469, 48–60. Zhu, D.C., Chung, S.L., Mo, X.X., Zhao, Z.D., Niu, Y.L., Song, B., Yang, Y.H., 2009b. The 132 Ma Comei–Bunbury Large Igneous Province: remnants identified in present-day SE Tibet and SW Australia. Geology 37, 583–586. Zhu, D.C., Zhao, Z.D., Niu, Y.L., Dilek, Y., Mo, X.X., 2011. Lhasa Terrane in southern Tibet came from Australia. Geology 39, 727–730. Zhu, D.C., Zhao, Z.D., Niu, Y.L., Dilek, Y., Hou, Zeng-Qian, Mo, Xuan-Xue, 2013. The origin and pre-Cenozoic evolution of the Tibetan Plateau. Gondwana Res. 23, 1429–1454. http://dx.doi.org/10.1016/j.gr.2012.02.002.