Revisiting the fluid regime during anatexis of metasedimentary rocks: New constraints from an integrated approach on the Cenozoic Himalayan granites

Abstract

Water plays a critical role in the formation of granitic magmas and continental crust, but distinguishing water-present and water-absent anatectic scenarios using the geochemistry of granites is controversial. In this study we use an integrated approach that combines whole-rock major and trace element geochemistry, Ti-in-zircon thermometry, phase equilibrium modeling and trace element modeling to study the water regime that produced the Miocene granites from the Malashan-Gyirong area in central Himalaya to solve this controversy. The Gyirong granites have relatively low CaO, Sr and Ba and high Rb and (87Sr/86Sr)i and the Malashan granites have relatively high CaO, Sr and Ba and low Rb and (87Sr/86Sr)i, and were classified as group A and group B granites, respectively. Following previous interpretation, group A and group B granites are consistent with the products of water-absent and water-present melting of metasedimentary rocks, respectively. Ti-in-zircon thermometry yielded maximum values of 761–796 °C for the Gyirong granites and 730–764 °C for the Malashan granites. Distinct variation trends in zircon trace element compositions indicate that these two groups of granites were not linked by crystallization differentiation. Using average compositions of Proterozoic pelite as starting materials, phase equilibrium modeling was carried out at a variety of P–T–H2O conditions typical of the Himalayan orogen, P = 5, 10 and 15 kbar, T = 600–800 °C, H2O = 0–10 wt%. Pelite can produce melts with coupled CaO–Na2O contents for both groups at 10 kbar. Specifically, constraints from compositions and temperatures require that the bulk H2O content is ca. 1–2 wt% for group A granites and > ca. 4 wt% for group B granites. Compared with the maximum structural water content of the pelite at 10 kbar (1.77 wt%), this study testifies that group A granites formed under water-absent conditions and group B granites under water-present melting conditions. Modeling shows that water-present melting can produce melts with high Sr–Ba and low Rb contents resembling group B granites, while water-absent melting can produce melts with low Sr–Ba and high Rb contents resembling group A granites. This study highlights that water can indeed cause differences in granite geochemistry but a comprehensive investigation is required to better determine the role of water during crustal anatexis.