Buoyant replenishment in silicic magma reservoirs: Experimental approach and implications for magma dynamics, crystal mush remobilization, and eruption

Abstract

We present new experiments on replenishment of rhyolite magma chambers by rhyolite magma using corn syrup-water solutions. We emphasize small density contrasts and show that buoyancy is the key controlling factor for whether injections will rise to the top (if buoyant) or pond at the base (if denser). During emplacement, we observe little or no mixing of the injected liquid with the reservoir liquid, as predicted by the fact that our injections have low Reynolds numbers (<10, typically). At later stages, the low-buoyancy (≤ 1 kg m-3) injected liquid, which has accumulated at the top of the reservoir, undergoes mixing with the reservoir liquid, which may originate by the gravitational destabilization of a thin layer of denser resident liquid trapped above the injected liquid layer. The presence of a basal crystal mush, modeled by acrylic beads in a corn syrup-water solution matrix is also considered. Slightly buoyant injections entrain a small fraction of mush particles to the top of the overlying liquid layer. Entrainment efficiency increases dramatically for high-buoyancy injections. We hypothesize that the injected liquid can entrain a maximum quantity of mush particles, which corresponds to the amount required for the injected liquid/mush particle suspension to attain neutral buoyancy in the resident liquid. Hence for silicic systems, a replenishing melt can entrain up to 12.5% crystals during its ascent through the mush. Our results have implications for rhyolites bearing crystals with disequilibrium features, as they may represent mush crystals remobilized by a replenishing silicic magma. Copyright 2009 by the American Geophysical Union.