Million-year melt–presence in monotonous intermediate magma for a volcanic–plutonic assemblage in the Central Andes: Contrasting histories of crystal-rich and crystal-poor super-sized silicic magmas

Abstract

The melt–present lifetime of super-sized monotonous intermediate magmas that feed supereruptions and end life as granodioritic plutons is investigated using zircon chronochemistry. These data add to the ongoing discussion on magma assembly rates and have implications for how continental batholiths are built. Herein, we estimate ∼1.1 Ma of continuous melt presence before and after the climactic caldera-forming 2.89±0.01 Ma (2σ error) Pastos Grandes Ignimbrite (PGI) supereruption (∼1500 km3 of magma) in the Andes of southwest Bolivia. Zircon crystallization in PGI pumice and lava from the faulted Southern Postcaldera Dome span ∼0.7 Ma prior to the climactic eruption and formation of the eponymous caldera, whereas younger, unfaulted Postcaldera Dome lavas (termed Northern and Middle) and a granodioritic plutonic clast within the products of a Pleistocene eruption indicate a further ∼0.4 Ma of post-climactic zircon crystallization. Bulk-rock compositions as well as zircon thermometry and geochemistry indicate the presence of homogeneous dacitic magma before and after the climactic eruption, but a trend to zircon crystallization at higher temperatures and from less evolved melts is seen for post-climactic zircon. We propose a model in which a large volume of crystal-rich dacite magma was maintained above solidus temperatures by periodic andesitic recharge that is chemically invisible in the erupted components. The climactic caldera-forming eruption vented the upper portions of the magma system zircon was saturated. Zircon in postcaldera lavas indicate that residual magma from this system remained locally viable for eruption at least for some time after the caldera-forming event. Subsequently, deeper “remnant” dacite magma previously outside the zone of zircon saturation rose to shallower levels to re-establish hydraulic and isostatic equilibrium where zircon crystallization commenced anew, and drove more resurgent volcanism and uplift. The same magma crystallized as a granodiorite pluton which was sampled as xenoliths in much later volcanic events. Over the ∼1.1 Ma zircon crystallization history for the PGI, postcaldera lavas and xenoliths, the melt remained in an ∼100–150 °C temperature window as indicated by Ti-in-zircon thermometry. Although chemical trends are consistent with zircon crystallization at variable temperatures, there is no secular cooling, but rather a thermal rejuvenation following the 2.89 Ma PGI eruption. As such these data provide a “low and slow” temporal constraint for models for the pre-eruptive lifetimes of mushy magma in contrast to the “rapid” mobilization of crystal-poor silicic magmas, consistent with a model where the latter are incubated within the former and extracted rapidly prior to eruption. The thermal and chemical monotony of crystal-rich dacites throughout a caldera cycle connotes conditions where near-eutectic melt can be maintained in near-surface magma reservoirs for an extended period of time if the subvolcanic magma reservoir is sufficiently large so that hotter and initially zircon-undersaturated magma can replenish shallow magma vented in a supereruption.