Abstract

The 1600km-long Gangdese magmatic belt features extensive Paleocene–Eocene I-type intrusive rocks and coeval volcanic successions, abundant but more localized Oligo-Miocene calc-alkaline to alkaline plutons, and Miocene potassic to ultrapotassic volcanic rocks. These Cenozoic igneous rocks record geodynamic changes related to the India–Asia collision which began at ~55–50Ma. New and published lithogeochemical and multiple isotopic (Os–Sr–Nd–O–Hf) analyses of these Cenozoic igneous rocks reveal that the Paleocene–Eocene magmas have similar compositions to continental arc rocks throughout the belt, but later Miocene magmas show sharp longitudinal contrasts in geochemical and isotopic compositions, which are also correlated with the occurrence of porphyry-type mineralization.Sparse Miocene high-K calc-alkaline to shoshonitic volcanic rocks in the eastern Gangdese belt have low to moderate (87Sr/86Sr)i ratios (0.7057–0.7121), moderately negative εNdi values (−9.4 to −3.4), low (187Os/188Os)i ratios (0.154–0.210), highly variable εHfi values (−5.9 to +10.1), and low zircon δ18O values (+5.0–+6.7‰), which are interpreted to reflect derivation by partial melting of subduction-modified Tibetan sub-continental lithospheric mantle (SCLM). In contrast, Miocene high-K calc-alkaline to shoshonitic volcanic rocks in the western Gangdese belt have higher (87Sr/86Sr)i ratios (0.7069–0.7263), more negative εNdi values (−17.5 to −6.0) and εHfi values (−15.2 to +0.7), and crust-like zircon δ18O values (+6.2–+8.8‰), but mantle-like (187Os/188Os)i values (0.156–0.182), and high Ni and Cr contents. These features suggest that potassic to ultrapotassic magmas in the western Gangdese belt were also derived from partial melting of Tibetan SCLM but with ~3–25% input of melts±fluids from the underthrust Indian plate (87Sr/86Sr=0.74–0.76, εNd=−18 to −10, δ18O=+10−+14‰). In contrast, Miocene alkaline magmas to the east were unaffected by this source.Oligo-Miocene calc-alkaline to high-K calc-alkaline granitoids related to large porphyry Cu–Mo deposits in the eastern Gangdese belt (east of ∼89° E) are geochemically broadly similar to the early Paleocene–Eocene rocks. They are thought to be derived from partial melting of subduction-modified lower crust with mixing of alkaline melts from partial melting of SCLM, and have relatively low (87Sr/86Sr)i ratios (0.7047–0.7076), high εNdi values (−6.1 to +5.5) and εHfi values (1.4–8.7), moderate (187Os/188Os)i ratios (0.224–0.835), and low zircon δ18OVSMOW values (+5.5–+6.6‰). These magmas also had high water contents (weak Dy/Yb enrichment, characterized with amphibole fractionation) and oxidation states (ΔFMQ 0.8–2.9), which explain their unique association with porphyry Cu–Mo mineralization. In contrast, Miocene high-K calc-alkaline to shoshonitic granitoids in the western Gangdese belt (west of ∼89° E) show differences in geochemical and isotopic compositions to the earlier Paleocene–Eocene magmatism, and are characterized by crust-like zircon δ18O values (+6.2–+8.8‰), high (87Sr/86Sr)i ratios (0.7147–0.7165), negative εNdi values (−11.3 to −7.9), crust-like (187Os/188Os)i values (0.550–1.035), and low εHfi values (−13.0 to 3.9). These magmas are interpreted to reflect involvement of melts±fluids from the underthrust Indian plate and high degrees of crustal contamination upon emplacement. Only one small porphyry Cu–Mo deposit is known to be associated with these western granitoids.We suggest that this difference reflects the variable extent of underthrusting of the Indian plate continental lithosphere beneath Tibet in the Oligo-Miocene, and diachronous breakoff of the Greater India slab. In the absence of underthrust Indian lithosphere to the east of ~89° E in the Oligo-Miocene, slab breakoff triggered asthenospheric upwelling and partial melting of previously subduction-modified Tibetan lithosphere, generating hydrous, oxidized calc-alkaline magmas with the potential to generate porphyry Cu–Mo deposits. In contrast, underthrusting of the Indian plate to the west at this time limited the involvement of asthenospheric melts and the extent of partial melting of subduction-modified lithosphere, with the result that melts±fluids derived from the underthrust lithosphere were infertile.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call