Geochemistry of Eocene-Early Oligocene low-temperature crustal melts from Greater Himalayan Sequence (Nepal): a nanogranitoid perspective

Abstract

Despite melt inclusions in migmatites and granulites provide a wealth of information on crustal anatexis in different geological scenarios, a complete compositional study (including trace elements and HO) is yet to be made for the Himalayan rocks. In this contribution, we focus on nanogranitoids occurring in peritectic garnet of migmatites from Kali Gandaki valley in central Nepal (Greater Himalayan Sequence, GHS). The microstructural position of the nanogranitoids proves that these melts were produced at 650–720 °C and 1.0–1.1 GPa, during the Eohimalayan prograde metamorphism (41–36 Ma) associated with crustal thickening. Nanogranitoid compositions (mostly granodiorites, tonalities and trondhjemites) resemble those of experimental melts produced during HO-present melting of meta-sedimentary rocks. They have variable HO concentrations (6.5–14.4 wt%), which are similar to the expected minimum and maximum values for crustal melts produced at the inferred P–T conditions. These compositional signatures suggest that melt formation occurred in the proximity of the HO-saturated solidus, in a rock-buffered system. The low-to-very low contents of Zr (3–8 ppm), Th (0.1–1.2 ppm) and LREE (4–11 ppm) along with the weak-to-moderate positive Eu anomalies (Eu/Eu* = 1.2–3.3), the high B concentrations (200–3400 ppm) and the high U/Th ratio (up to 21) point to the lack of equilibration between melt and accessory minerals and are consistent with melting of plagioclase at low temperature. Kali Gandaki nanogranitoids record the beginning of melting in a HO-present system that, in other GHS localities, may have produced voluminous crustal melts. We demonstrate that compositional comparison with nanogranitoids may be useful to reconstruct the petrogenesis of Eohimalayan granitoids.