Post-collision, Shoshonitic Volcanism on the Tibetan Plateau: Implications for Convective Thinning of the Lithosphere and the Source of Ocean Island Basalts

Potassic volcanism has been widespread and semi-continuous onthe Tibetan plateau since 13 Ma, post-dating the orogenic thickeningof the India-Asia collision. Volcanism may have commenced slightlyearlier (16–20 Ma) in the southern portion of the plateauand then ceased around 10 Ma. The dominant lavas are pyroxen-and plagioclase-phyric shoshonites with subordinate occurrencesof dacites and rhyolites. Their mineralogy reflects crystallizationfrom high-temperature (1100C) magmas which had elevated oxygenand water fugacities. Geochemically, they are characterizedby relatively low TiO², Al²O³ and Fe²O³, and high Na²O, coupledwith variable abundances of compatible trace elements and veryhigh contents of incompatible trace elements. Normalized incompatibleelement patterns have marked negative Nb, Ta and Ti anomalieswhereas K²O appears to be buffered at 4% over a wide range ofSiO². Isotope data show a relatively broad and enriched rangeof ⁸⁷Sr/⁸⁶ Sr (0.7076–0.7106) at more restricted ENd (–5.2to –8.1). Pb isotopes are characterized by a range of²⁰⁷Pb/²⁰⁴ Pb (15.51–15.72) and ²⁰⁸ Pb/²⁰⁴Pb (38.67–39.30) at very uniform ²⁰⁶Pb/²⁰⁴ Pb (18.39–18.83), leadingto vertical arrays. Volcanics from the southern parts of theplateau have more primitive isotopic compositions: ⁸⁷Sr/⁸⁶ Sr0.7048–0.7080, _Nd 1.4 to –3.3, ²⁰⁶Pb/²⁰⁴ Pb 18.48–18.67,²⁰⁷Pb/²⁰⁴ Pb 15.59–15.68 and ²⁰⁸Pb/²⁰⁴ Pb 38. 73–38.98. In general, the geochemical and isotopic data most closely approximatepartial melting arrays, although fractionation processes haveclearly operated. The isotopic ratios and the enrichment ofincompatible elements and LREE/HREE cannot be derived from adepleted mantle source via a single-stage melting process. Instead,a metasomatized, garnet peridotite source containing 6% phlogopiteis required and this is inferred to lie within the lithosphericmantle. The enrichment in incompatible elements in this sourcemust have been sufficiently ancient to generate the observedisotopic ratios, and Nd depleted mantle model ages suggest thiswas Proterozoic in age (1.2 Ga), whereas Pb model ages recordan Archaean event, suggesting the source had a multi-stage enrichmenthistory. The negative Ta, Nb and Ti anomalies and low Rb/Basuggest that metasomatism may have occurred during an ancientsubduction episode. The high ²⁰⁸Pb/²⁰⁴Pb, ²⁰⁷Pb/²⁰⁴ Pb and lowNb/U, Ce/Pb of the Tibetan shoshonites are distinct from oceanisland basalts. Thus, if convectively removed lithospheric mantleprovides a source for ocean island basalts, it must differ significantlyfrom the source of the Tibetan shoshonites. A lithospheric mantle source for the volcanism places importantconstraints on geodynamic models for the evolution of the Tibetanplateau and the India-Asia collision. For likely thermal structuresbeneath the plateau, the temperatures required to trigger meltingwithin the lithospheric mantle can only be plausibly obtainedif the lower parts of the lithospheric mantle were removed byconvective thinning. This is consistent with recent models whichinvoke the same process to explain the current elevation andextensional deformation of the plateau. The age data suggestthis occurred at 13 Ma and the duration of volcanism may beexplained by continued conductive heating since that time. Poorlysampled but slightly older volcanics from the southern portionsof the plateau may indicate that convective thinning began inthe south and migrated northwards. Rapid uplift of the plateaumay have resulted in increased rates of chemical weathering,which led to global cooling, as indicated by oxygen isotopedata from Atlantic sediments. KEY WORDS: Climate; lithospheric mantle; OIB; Tibet; volcanism ^*Corresponding author.