Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Negligible surface uplift following foundering of thickened central Tibetan lower crust
Geology ( IF 5.8 ) Pub Date : 2020-08-25 , DOI: 10.1130/g48142.1 Yunchuan Zeng 1, 2 , Mihai N. Ducea 2, 3 , Jifeng Xu 1, 4 , Jianlin Chen 4 , Yan-Hui Dong 4
Geology ( IF 5.8 ) Pub Date : 2020-08-25 , DOI: 10.1130/g48142.1 Yunchuan Zeng 1, 2 , Mihai N. Ducea 2, 3 , Jifeng Xu 1, 4 , Jianlin Chen 4 , Yan-Hui Dong 4
Affiliation
This study used clinopyroxene (cpx) compositions and zircon Hf-O isotopes of Eocene adakitic rocks (EARs) from the Qiangtang block to resolve the mechanism(s) responsible for the formation of the central Tibetan Plateau. The two leading and opposing hypotheses for the origin of these rocks are (1) partially molten foundered lower crust, and (2) partial melting of continentally subducted upper crust. The consensus is that some crustal sources within the mantle have reached eclogite facies, while evidence remains insufficient. Reverse zonation for cpx in high Mg# andesitic samples shows a low Mg# core with lower Sr and Sr/Y than the high Mg# rim, suggesting derivation of parent magma by interaction between some eclogite-derived felsic melts and mantle peridotite. Overall, the mantle-like zircon δ18O (mean value of ∼5.9‰) and εHf(t) (up to +6.7) values argue for a mafic source rather than buried upper-crustal rocks. Given the EARs were formed within a short time span after the end of crustal shortening, the original felsic melts were most likely derived from the foundered and eclogitized lower crust. The foundering process explains the early Eocene low-relief topography and the intermediate, eclogite-free modern crustal composition of central Tibet. Surface uplift as a response to lithosphere removal, however, was likely negligible, based on various lines of evidence, including sediment provenance, isotope paleoaltimetry, and thermochronology, perhaps because the central Tibetan crust was weak. INTRODUCTION The driving mechanism(s) of thickening and uplift of the Tibetan crust during NeoTethyan subduction termination and the subsequent India-Asia collision is(are) of great significance to our understanding of orogenesis. Studies show that, in the early stages of collision, crustal shortening and thickening were localized in the hinterland, i.e., central Tibet, far from the collision front, to form an early Paleogene protoplateau (Fig. 1; for reviews, see Wang et al. [2014] and Kapp and DeCelles [2019]). The lower modern central Tibetan crust has an intermediate-felsic average composition and is eclogite-free, as inferred from the low seismic velocities (Vp < 6.6 and Vs < 4.25 km/s; Galve et al., 2006; Yang et al., 2012), and limited crustal xenoliths in Pleistocene lavas (Hacker et al., 2000). Extensive Eocene volcanic rocks with minor intrusive equivalents formed abruptly after a magmatic lull since the Early Cretaceous in the Qiangtang block (QB), a key part of the protoplateau (Fig. 1), and they show chemical similarities with adakitic rocks (referring to intermediate-felsic high Sr/Y and La/Yb rocks; Castillo, 2012). These Eocene adakitic rocks (EARs) provide us prime access to the deep processes responsible for formation of the protoplateau. There are two opposing hypotheses for the formation of these distinct rocks in central Tibet. The first hypothesis is that the EARs were partial melts of continentally subducted flysch-rich upper crust at mantle depth, while the southward subduction of Asian lithosphere beneath central Tibet kinetically drove protoplateau formation (Tapponnier et al., 2001; Wang et al., 2008; Replumaz et al., 2016). The opposing hypothesis is that the crust was shortened, thickened, and uplifted by plate stress from India-Asia collision, and then the thickened lower crust (presumably turned into eclogite) foundered to cause further uplift and partially melted to generate the EARs (Chen et al., 2013; Chapman et al., 2018; Kelly et al., 2020). The two hypotheses fundamentally impact our overall knowledge of India-Asia continental tectonics (see Replumaz et al.[2016] and Kelly et al. [2020], respectively). Here, we present majorand trace-element compositions of clinopyroxene (cpx) and zircon Hf-O isotopic data for the EARs to test between the two end-member hypotheses. The compositional variation in reverse cpx and mantle-like zircon δ18O values provide strong evidence for the lower-crustal foundering model, but we contend the surface uplift associated with lithosphere removal was negligible, perhaps because the crust was weak. GEOLOGICAL SETTING As the central part of the Himalayan-Tibetan orogen, the QB is a wide plateau with thickened continental crust of 60–70 km and high but flat topography (average >4800 m; Fig. 1; Owens and Zandt, 1997; Yang et al., 2012). It underwent two pre-Cenozoic collisional events, i.e., with Songpan-Ganzi block (SG) in the Late Triassic and with the Lhasa block in the Early Cretaceous, forming the Jinsha and Bangong sutures, respectively (Fig. 1; Kapp and DeCelles, 2019). At the northern boundary of the protoplateau *E-mail: jifengxu@gig.ac.cn †Current address: Key Laboratory of Submarine Geosciences, Second Institute of Oceanography, Ministry of Natural Resources, Hangzhou 310012, China Downloaded from https://pubs.geoscienceworld.org/gsa/geology/article-pdf/doi/10.1130/G48142.1/5137752/g48142.pdf by Mihai N. Ducea on 27 August 2020 2 www.gsapubs.org | Volume XX | Number XX | GEOLOGY | Geological Society of America (Fig. 1), Tanggula thrust belt movement has shortened the QB crust up to 50% and formed the Hoh Xil foreland basin (Staisch et al., 2014; Li et al., 2017; Kapp and DeCelles, 2019). Recent combined studies of field and structural geology, basin analyses, and stratigraphic analyses have shown that the timing of thrusting likely ended ca. 50 Ma, following a major phase of deformation and uplift of the QB crust at ca. 54 Ma (Li et al., 2017; Jin et al., 2018; Kapp and DeCelles, 2019). Early Paleogene magmatic rocks in the QB chiefly include EARs and minor small Eocene–Oligocene mafic dikes (Chen et al., 2013). The EARs are widely distributed in the QB but are more concentratedly in the mideastern part, where they form a field covering a total area of ∼9000 km2 (Fig. 1B). The strata exposed there are composed of mainly Mesozoic shallow-marine and minor Cretaceous–Paleogene terrestrial rocks. The EARs in this region lie flat and unconformably above the strongly shortened and folded Mesozoic–Lower Paleocene strata and the weakly deformed or undeformed Lower Eocene strata (Fig. S1 in the Supplemental Material1; Kapp and DeCelles, 2019). SAMPLES AND ANALYTICAL RESULTS In total, 20 andesitic, 20 dacitic, and four rhyolitic samples from the mideastern QB (Fig. 1B; Fig. S1) were collected for geochemical analyses (see the Supplemental Material for petrography, analytical methods, and results). These samples are adakitic according to bulkrock geochemical compositions; i.e., high SiO2 (57.25–71.58 wt%), Sr (>415 ppm), Sr/Y (72–215), and La/Yb (18.6–100), and low Y (<15.5 ppm) and Yb (<1.72 ppm) (Fig. 2; Table S3). In addition, 22 samples from Dongyuehu, Luanqinshan, Zhuozishan, and Yuejinla (Fig. S1) present high Mg# (>60) with high Cr (69–314 ppm) and Ni (69–149 ppm) contents, resembling typical primitive andesite (Kelemen et al., 2014; this broad definition includes dacite and rhyodacite with Mg# >60, so hereafter we refer our samples with Mg# >60 as primitive EARs). Eleven samples (andesite to rhyolite) showed relatively uniform Sr-Nd isotopic values, (Sr/Sr)i = 0.7064–0.7078 and εNd(t) = −6.21 to −2.51, which are within the range of literature data (Fig. 2C). The zircon 206Pb/238U ages of five samples were also consistent with previous results and together suggest a short life span for EAR magmatism (ca. 45–36 Ma, with a distinctive peak at ca. 40 Ma; Chen et al., 2013). The age, morphology (e.g., euhedral to subhedral, homogeneous internal structures; Fig. S3), and high Th/U ratios (0.11–1.81) of the analyzed zircon grains are indicative of a magmatic origin, and thus the measured Hf-O isotopic data represent the primary value precipitating from parent magma. These grains showed slightly variable δ18O (4.7‰–7.6‰, with mostly ∼5.9‰, and a mean value of 5.9‰ ± 0.2‰) and slightly negative to positive εHf(t) values (−6.1 to 4.7; Table S2). Reverse zonation (Fe-rich cores and Mg-rich rims) was observed in some cpx grains from primitive EARs (Fig. 3; Fig. S2). The Mg-rich rims had the same compositions as normal (unzoned) grains. Although the Mg# as well as the Ni and Cr contents increased from the core to rim, the low-Mg# core showed lower Sr, Sr/Y, and Dy/Yb, but higher Y and Yb than the rim. Notably, inverse calculation by using the Fe-Mg exchange and partition coefficient between cpx and melt showed that only the melt equilibrated with the high-Mg# cpx has adakitic features (Fig. 3; Table S6). PETROGENESIS OF THE EARS Primitive andesites can be produced by crustderived melts reacting with mantle peridotite or mixing with mantle-derived melts, or by ultralowdegree melting of a shallow hydrous mantle, and those with adakitic affinity require the involvement of garnet ± amphibole during partial melting and/or magmatic evolution (e.g., Xu et al., 2002; Castillo, 2012; Kelemen et al., 2014). The positive correlation between La/Yb and Dy/Yb ratios suggests that the adakitic features of EARs were controlled by garnet, although amphibole fractionation in the magmatic evolution cannot be excluded (Fig. 2E). Additionally, the lower Ba, Sr, Th, and rare earth element contents of the EARs compared with Eocene–Oligocene mafic dikes are in stark contrast with the trend of low-degree mantle melting (Fig. 2F). The presence of reverse-zoned cpx in primitive EARs implies open-system magma evolution with involvement of two end members, and the key to distinguishing the eclogite-derived melt–mantle interaction from magma mixing models is whether the primitive end member represented by the high-Mg# rim cpx originally had adakitic features. As elaborated before, 1Supplemental Material. Experimental methods, Figures S1–S5, and Tables S1–S8. Please visit https:// doi .org/10.1130/GEOL.S.12811868 to access the supplemental material, and contact editing@geosociety. org with any questions. 95°E 90°E 85°E 80°E 75°E 40°N
更新日期：2020-08-25
京公网安备 11010802027423号