Caldera faults can be sintered volcanic conduits without necessarily requiring catastrophic rapid slip events

Caldera subsidence is accommodated by the eruption of large volumes of fragmented magma removed from the crust, commonly transported along caldera ring-faults or between foundering crustal blocks. The largest types of silicic caldera-forming eruptions have not been witnessed, and so the fragmental magma flux and associated caldera subsidence rates are poorly constrained. Based on the presence of aphanitic fault-filling rocks in some calderas, caldera fault slip rates have been inferred to be sufficiently high to cause frictional melting, leading to a ‘super-fault’ model for caldera slip. We revisit this model at the classic Glencoe caldera volcano locality, and find textural evidence that the caldera fault fill may instead be the sintered remnants of the fragmented eruptive products, and that frictional melting may not be required. We support this with geochemical mass balance calculations that lead us to conclude that (1) the fault fill is a sintered remnant associated with the eruption, (2) partially resorbed quartz grains in the glassy groundmass could originate from the country rock quartz and quartzite lithics that resorb when incorporated into quartz-undersaturated sintered mass, and (3) that the late-stage fault-intrusions represent the squeeze-up of the mush from which the eruptive melts are derived. In the context of caldera-forming eruptions, our model is consistent with the proposal that caldera faults may be active throughout much of a sustained, relatively long-lived eruption, rather than moving only rapidly and briefly.