Post-collisional Ultrapotassic Mafic Magmatism in South Tibet: Products of Partial Melting of Pyroxenite in the Mantle Wedge Induced by Roll-back and Delamination of the Subducted Indian Continental Lithosphere Slab

Post-collisional (25–8 Ma) ultrapotassic mafic magmatic rocks occur to the north of the India–Asia collision zone within the Lhasa terrane of the southern Tibetan Plateau, forming a near 1000 km long semi-continuous igneous belt. They include both extrusive and intrusive facies, although lava flows dominate. To understand their petrogenesis, the mineral chemistry of olivine phenocrysts and xenocrysts and whole-rock major and trace element and Sr–Nd–Pb isotope data are presented for the most primitive mafic magmatic rocks (MgO > 6 wt %) from west to east. The studied samples are characterized by high MgO (6·28–15·75 wt %), K2O (4·76–8·89 wt %), SiO2 (46·44–59·74 wt %), Ba (1368–14076 ppm), Th (69–336 ppm) and Ni (106–527 ppm) contents. Chondrite-normalized rare earth element (REE) patterns show enrichment in light rare earth elements (LREE), flat heavy REE (HREE) patterns and negative Eu anomalies. These REE patterns have a very distinctive inverted ‘spoon shape’, which appears to be a common characteristic of collision-related ultrapotassic magmas. Primitive mantle-normalized incompatible trace element patterns exhibit strong enrichments in large ion lithophile elements (LILE) relative to high field strength elements (HFSE) and strong negative Ta–Nb–Ti anomalies, which are typical of subduction-related magmas. The ultrapotassic magmatic rocks studied have extremely radiogenic initial Sr isotopic compositions (0·712379–0·737616) and low (143Nd/144Nd)i (0·511662–0·511984). Combined with their Pb isotope compositions [(206Pb/204Pb)i = 18·30–18·92; (207Pb/204Pb)i = 15·65–15·87; (208Pb/204Pb)i = 39·02–39·76] these data are consistent with the involvement of a subducted continental crustal component in their petrogenesis. The Sr–Nd–Pb isotope compositions exhibit linear trends between depleted mid-ocean ridge basalt (MORB)-source mantle (DMM) and Indian continental crust. The extreme enrichment of the upper mantle below south Tibet is considered to result from the addition of components derived from subducted Indian continental crust to the overlying mantle wedge during northward underthrusting of Indian continental lithosphere beneath the Lhasa terrane since India–Asia collision at ∼55 Ma. The post-collisional K-rich mafic magmas in south Tibet were generated by partial melting of pyroxenite in a mantle source region that was created by reaction of hydrous fluids and siliceous melts from subducted granulite–eclogite-facies Indian continental crustal rocks with the surrounding peridotitic mantle. A continuous process from slab roll-back, through break-off, to detachment of the slab may have induced partial melting of the pyroxenites. Cessation of the post-collisional ultrapotassic magmatism at ∼8 Ma may be linked to the onset of flat slab subduction beneath southern Tibet and the elimination of the wedge of Tibetan subcontinental lithospheric mantle and underlying asthenosphere; geophysical data indicate that at the present day eclogite-facies Indian continental crust directly underthrusts the crust of the Lhasa terrane with no intervening mantle wedge. The proportion of the Indian continental crustal component in the mantle source of the ultrapotassic mafic magmas decreases eastward, as do the ages and volumes of the magmatic rocks. There are no outcrops of post-collisional K-rich mafic magmatic rocks (MgO > 6 wt %) to the east of 87°E in the Lhasa terrane, which may indicate a change in subduction geometry at this longitude.