Abstract

The possible role of continental sediments in the generation of potassium-enriched lavas of the Alpine-Himalayan belt depends on their melting behaviour either during subduction or during post-collisional relaxation. Although usually classed as orogenic lavas, these volcanic rocks may result from re-melting of newly formed mantle lithosphere 30–40millionyears after collision ends, and can thus be considered as the first stage of intraplate volcanism. The potassic component in these volcanics is characterized by a high Th/La signature for which there are two competing explanations: melting of subducted continental clastic sediments, and the involvement of lawsonite blueschists in the protoliths to the melting assemblages. Here, we report on a series of high-pressure experiments at 1–3GPa and 900 to 975°C on the melting behaviour of natural phyllite from Serbia, which serves as a proxy for Balkan upper continental crust. Hydrous granitic melts are present in all runs (68wt% SiO2, ~4–5wt% K2O, Mg#<54 and ~5wt% H2O). Garnet, quartz/coesite, plagioclase, K-feldspar, biotite/phengite, clinopyroxene and sillimanite/kyanite, and accessory phases including zircon, rutile, ilmenite, apatite and monazite occur in the charges. LA-ICP-MS analyses establish that the melts are extremely enriched in LILE (except for Sr), Th and U, but depleted in Nb and Zr, with LREE higher than HREE. Accessory phases accommodate several trace elements, especially HFSE and REE. Partition coefficients for some trace elements between residue and crustal melts are close to 1, contrasting strongly with melts of peridotite. Our dataset indicates that the direct melting of upper continental crust alone would generate siliceous, high-K magmas with enriched LILE, Th and U, but cannot explain the high Th/La fingerprint of K-rich lavas of the Alpine-Himalayan orogenic belt. We demonstrate that the Alpine-Himalayan orogenic volcanics attribute their unusual trace element geochemistry to the involvement of lawsonite blueschists that are imbricated together with extremely depleted fore-arc peridotites to form new lithosphere in the source region. There is no need or evidence for deep subduction in which a succession of additional reactions would only serve to modify and dilute the high Th/La signature.

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