Search

The continental crust in the overriding plate of the India-Asia collision zone in southern Tibet is characterized by an overthickened layer of felsic composition with an underlying granulite-eclogite layer. A large data set indicates that this crust experienced magmatism from 245 to 10 Ma, as recorded by the Gangdese Batholith. Magmatism was punctuated by flare-ups at 185−170, 90−75, and 55−45 Ma caused by a combination of external and internal factors. The growth of this crust starts with a period dominated by fractional crystallization and the formation of voluminous (ultra)mafic arc cumulates in the lower crust during subduction, followed by their melting during late-subduction and collision, due to changes in convergence rate. This combined accumulation-melting process resulted in the vertical stratification and density sorting of the Gangdese crust. Comparisons with other similarly thickened collision zones suggests that this is a general process that leads to the stabilization of continental crust. ▪ The Gangdese Batholith records the time-integrated development of the world's thickest crust, reaching greater than 50 km at 55–45 Ma and greater than 70 km after 32 Ma. ▪ The Gangdese Batholith records three magmatic flare-ups in response to distinct drivers; the last one at 55−45 Ma marks the arrival of India. ▪ Magmatism was first dominated by fractional crystallization (accumulation) followed by crustal melting: the accumulation-melting process. ▪ Accumulation-melting in other collision zones provides a general process for vertical stratification and stabilization of continental crust.

The temporal and spatial distribution of enriched source components at sites of continent–continent collision provides critical insights into mantle dynamic processes associated with subduction- and collision-related events. However, determining the origin of such enriched components remains a significant challenge. We report a comprehensive dataset of the Linzizong volcanic succession (LVS) from four locations with varying distance across-strike to the Indus–Yarlung suture in southern Tibet, which marks the exposed surface expression of the India–Asia collision zone. The LVS rocks in this study can be divided into two groups: a calc-alkaline Group 1 (69–55 Ma), mainly including basaltic–andesitic varieties, and a shoshonitic Group 2 (52–50 Ma), consisting predominantly of silicic rocks with minor mafic compositions. Group 1 samples are likely derived from the fractional crystallization of primitive basaltic melts as a result of the partial melting of a metasomatized mantle wedge. These samples are decoupled in Nd–Hf isotopic compositions, suggesting an incorporation of subducting sediment-derived melts into the mantle wedge. The influence of sediment-derived melt on the mantle source increases away from the suture zone toward Asia (i.e., from the south to the north) as indicated by the more enriched Sr, Nd, Pb, and Hf isotopic compositions, as well as elevated Th/La and La/Sm ratios. The heavy δ26Mg values, and high Ba/Th and Sr/Th ratios of samples close to the suture coincide with the dehydration of the subducting Neo-Tethyan slab. Group 2 mafic samples have depleted and coupled εNd–εHf isotopic compositions and are characterized by elevated (La/Yb)N and Dy/Yb ratios as well as low Zr/Nb ratios, indicating an origin of enriched garnet-bearing lithospheric mantle with contributions from asthenosphere-derived materials. Group 2 silicic samples are isotopically enriched both near and far away from the suture, which can be attributed to the involvement of ancient lower crust-derived melt from Tethyan Himalaya and central Lhasa subterrane basement, respectively. Our results show that the LVS are the magmatic response to late subduction (Group 1), and to initial India–Asia collision and slab breakoff (Group 2). Negative trends in the whole-rock Nd and zircon Hf isotopic compositions at ~51 Ma should be treated with caution for geodynamic interpretations, depending on the distance between the samples and the India–Asia suture.

This research was financially supported by the MOST of China (No. 2016YFC0600304 and No. 2016YFC0600407), the Strategic Priority Research Program (B) of the Chinese Academy of Sciences (XDB03010301), the National Science Foundation of China (41225006 and 41472061), and the MOST Special Fund from the State Key Laboratory of Geological Processes and Mineral Resources (China University of Geosciences). The surface uplift of mountain belts is in large part controlled by the effects of crustal thickening and mantle dynamic processes (e.g., lithospheric delamination or slab breakoff). Understanding the history and driving mechanism of uplift of the southern Tibetan Plateau requires accurate knowledge on crustal thickening over time. Here we determine spatial and temporal variations in crustal thickness using whole-rock La/Yb ratios of intermediate intrusive rocks from the Gangdese arc. Our results show that the crust was likely of normal thickness prior to ca. 70 Ma (~37 km) but began to thicken locally at ca. 70 − 60 Ma. The crust reached (58 − 50) ± 10 km at 55 − 45 Ma extending over 400 km along the strike of the arc. This thickening was likely due to magmatic underplating as a consequence of rollback and then breakoff of the subducting Neo-Tethyan slab. The crust attained a thickness of 68 ± 12 km at ca. 20 − 10 Ma, as a consequence of underthrusting of India and associated thrust faulting. The Gangdese Mountains in southern Tibet broadly attained an elevation of >4000 m at ca. 55 − 45 Ma as a result of isostatic surface uplift driven by crustal thickening and slab breakoff, and reached their present-day elevation by 20 − 10 Ma. Our paleoelevation estimates are consistent not only with the C − O isotope-based paleoaltimetry but also with the carbonate clumped isotope paleothermometer, exemplifying the promise of reconstructing paleoelevation in time and space for ancient orogens through a combination of magmatic composition and Airy isostatic compensation. Publisher PDF

Two types of K-rich magma of Eocene to Early Oligocene (ca. 40–30) and Plio-Pleistocene (ca. 5–0.1 Ma) age were emplaced prior to and following left-lateral slip on the Ailao Shan-Red River (ASRR) fault, a regional shear zone extending between southwest China and the Tonkin Gulf (South China Sea) that accommodated ‘escape’ of the Indochina block. The first type is exposed in the Dali-Lijiang and adjacent regions of western Yunnan and Sichuan and comprises ultramafic potassic to ultrapotassic ‘absarokites’ and their shoshonite, banakite, and SiO 2 -rich derivatives which were emplaced immediately prior to activation of the ASRR fault. They are characterized by high Mg.-nos, and low contents of fusible oxides (FeO*, CaO, Al 2 O 3 ), for equivalent MgO content, and pronounced primitive mantle-normalized high-field strength element (HFSE) depletions. In contrast, ‘post-escape’ K-rich magmas were erupted in the Puer, Maguan-Pingbian regions of south and southeast Yunnan. Apart from their relative enrichments in potassium they show typical HFSE-rich intra-plate compositional affinity. Geological and geomorphic evidence, and thermochronologic age dating of metamorphisc events, suggest that left-lateral shearing occurred between ca. 30 and 17 Ma; thereby accommodating the southeastward ‘escape’ of Indochina and (possibly) two episodes of spreading in the South China Sea. The southwestern part of Dali-Lijiang magmatic products was detached and offset by ca. 600 km and are now located in Phan Xi Pang in northern Viet Nam. The same is true for the Permo-Triassic Emeishan flood basalts, whose western exposures were likewise displaced by the same amount and are now represented by the Song Da complex, also in northern Viet Nam. Here, we report geochemical, isotopic, and 40 Ar/ 39 Ar age data for samples from both the ‘pre-escape’ Dali-Lijiang magmas and the ‘post-escape’ K-rich Puer, Maguan-Pingbian basalts and basanites, with a view to comparing and contrasting their interpolated source compositions, estimated conditions of upper mantle melt segregation and, by inference, their mantle dynamic and contamination histories insofar as these were conditioned by the India-Asia collision. Our interpretations yielded two complementary conclusions. The first contends that the pre-escape magmas result from adiabatic melting of crust-contaminated asthenosphere comprising a ‘melange’ of continental lithospheric mantle (CLM) (hydrated by sab-derived hysdrous fluids released at 0.2–0.5 GPa) and lower crust, delaminated from the overriding plate during mantle wedge corner flow and further enriched by metasomatic melts of subducted continental crust. We postulate that incipient H 2 O-saturated melting of the ‘melange’ occurs at depths of between ca. 100 and 200 km after being ‘dragged’ down by relict oceanic slab fragments, in response to the dehydration of supra-subduction amphibole- and phlogopite. The ensuing viscosity ‘crisis’ and buoyancy relative to ambient ‘fertile’ convecting mantle of such asthenospheric ‘pockets’, and the collision-related change from lithospheric compression to extension, almost certainly predisposes such a refractory yet crust-contaminated ‘pockets’ to rapid adiabatic melting. The second conclusion concerns the post-escape K-rich basalts and basanites and is based on the contention that decompression melting of thermally anomalous K-rich asthenospheric occurred in response to regional post-escape transtension, concomitant with the cessation Indochina escape and contiguous seafloor spreading. However, although these magmas share the HFSE-rich fertile source character of other, widely dispersed, post-escape Cenozoic basalts they more specifically resemble relatively rare examples of intra-plate, K-rich activity observed in northeast China, central Spain, and elsewhere in Asia and Europe, arguably (indirectly) reflecting mantle perturbations caused by major continental collisions.

article i nfo Ophiolite plays a key role in identifying paleo-ocean and paleo-plate and rebuilding the evolutionary history of ancient orogen. Mafic-ultramafic rocks are distributed in a broadly E-W direction in the middle of the Qiangtang terrane, northern Tibetan plateau. However, interpretation of these rocks as ophiolite and subsequent tectonic implications have been much disputed, and they were regarded by some to have an origin in a continental rift set- ting. A detailed zircon dating and geochemical and Sr-Nd-Hf isotopic study of this suite of rocks has been under- taken in order to clarify this important issue. The mafic-ultramafi cs uite was collected from the Gangma Co and Guoganjianian localities and is composed of cumulate and isotropic gabbro, basalt, actinolitite and plagiogranite. All basaltic rocks are tholeiitic and have low rare earth element (REE) abundances with variable REE patterns and slight negative Nb and Ti anomalies. These features are comparable with those of normal mid-ocean ridge ba- salts (N-MORB) and/or enriched mid-ocean ridge basalts (E-MORB). The positive whole-rock eNd(t) and zircon eHf (t) values indicate that these rocks were derivedfrom a long-term depleted mantle source. Zircon U-Pb dating using a sensitive high-resolution ion microprobe (SHRIMP) on two cumulate gabbros and two plagiogranites sam- ples yielded Carboniferous ages of 357±2.5 Ma, 356.1±3.0 Ma, 354.7±4.7 Ma and 345.4±4.6 Ma. The ophiolite marks a Paleo-Tethys Ocean basin in middle of the Qiangtang terrane, and it is interpreted as the western exten- sion of the Changning-Menglian Paleo-Tethys ophiolite in the eastern margin of the Tibetan plateau.

Felsic magmatic rocks of Triassic ages occur on the north side of the Qiangtang metamorphic belt. Zircon dating using the SHRIMP and LA-ICP-MS techniques reveals that these rocks were emplaced between 223 and 205 Ma. The emplacement time is similar to the Ar–Ar ages of phengites and sodic amphiboles of the eclogites and blueschist from the Qiangtang metamorphic belt. The felsic magmatic rocks can be divided into two groups with distinct geochemical and Sr–Nd–Hf isotopic compositions. The first and older group (223–219 Ma) includes diorite from the Baohu area and some volcanic rocks from the Juhuashan area. These rocks have high Sr (291–1367 ppm) and low Y (5.5–13.9 ppm) contents with high Sr/Y ratios (28–117). They are characterized by slightly negative eNd(t) values (−1.9 to −5.7), positive eHf(t) values (+3.2 to +5.4) and young zircon Hf model ages (644–729 Ma). The second and younger group (215–210 Ma) comprises volcanic rocks from the Jiangai and Guoganjianian areas. They show high initial 87Sr/86Sr ratios (0.708–0.714), distinct negative eNd(t) (−9.6 to −7.9) and eHf(t) values (−15.3 to −10.1) and much older bulk-rock Nd model ages (1658–2178 Ma) and zircon Hf model ages (1251–2012 Ma). Geochemical arguments led us to suggest that the rocks of the first group were generated by the melting of subducted oceanic crust and associated sediments, and the second group rocks were derived from anatexis of the Proterozoic continent crust. In conclusion, the Late Triassic magmatic rocks from the northern Qiangtang have recorded the process from subduction to break-off of an oceanic slab. The Qiangtang metamorphic belt to the south represents a northward oceanic subduction zone.

New geochronological and geochemical data on magmatic activity from the India-Asia collision zone enables recognition of a distinct magmatic flare-up event that we ascribe to slab breakoff. This tie-point in the collisional record can be used to back-date to the time of initial impingement of the Indian continent with the Asian margin. Continental arc magmatism in southern Tibet during 80–40 Ma migrated from south to north and then back to south with significant mantle input at 70–43 Ma. A pronounced flare up in magmatic intensity (including ignimbrite and mafic rock) at ca. 52–51 Ma corresponds to a sudden decrease in the India-Asia convergence rate. Geological and geochemical data are consistent with mantle input controlled by slab rollback from ca. 70 Ma and slab breakoff at ca. 53 Ma. We propose that the slowdown of the Indian plate at ca. 51 Ma is largely the consequence of slab breakoff of the subducting Neo-Tethyan oceanic lithosphere, rather than the onset of the India-Asia collision as traditionally interpreted, implying that the initial India-Asia collision commenced earlier, likely at ca. 55 Ma.

Raman spectra of aluminosilicate minerals, namely kyanite, andalusite, and sillimanite and K2O-Al2O3-SiO2 glasses were recorded. Four alumino-silicon tetrahedral model clusters were calculated by self-consistent (SCF) molecular orbital ab-ini-tio calculation of the quantum chem (QC) method. The result shows a decrease tendency in Raman frequencies in the 800-1200 cm(-1) frequency region with increase in four-coordinated Al content, which is assigned to the Si--Onb symmetry stretching vibrations. The Raman spectra in the 700-800 cm(-1) frequency region is attributed to Al-Onb symmetry stretching vibrations.