The influence of magmatic volatiles in priming a magma system for eruption.

A variety of mechanisms have been cited for priming a magmatic system for eruption, from injection of mafic magma to primary and secondary boiling. Volatile saturation and the presence of fluids are often invoked as a prerequisite for priming a magma system for large explosive eruptions. The presence of an exsolved volatile phase increases the compressibility of the system, allowing continued growth of the magma reservoir without generating significant overpressures that could trigger an eruption, and helping to sustain eruptions. However, recent work unlocking volatile budgets using apatite suggests that water saturation is not a systematic prerequisite for caldera forming eruption.

We present results on the apatite volatile record from a series of eruptions from two caldera systems, Santorini (Greece) and Mt Mazama (USA). Both have undergone multiple large explosive eruptions and smaller sub-Plinian eruptions. At Santorini (Greece), each eruption exhibits an extended? range of apatite volatile chemistry with a distinct volatile signature, indicating extensive fractionation. The inferred volatile saturation state remains constant throughout the evolution of each magma, regardless of the scale of the eruption. Water-saturated magmas usually feed Plinian to caldera-forming eruptions, whilst water-undersaturated magmas feed both sub-Plinian and caldera-forming eruptions. In contrast, apatite from the caldera-forming dacitic eruption of Mt. Mazama and its two-preceding explosive eruptions display a very restricted compositional range for each eruption. Apatite from co-erupted mafic scoria in the deposits of the caldera-forming eruption are consistent with water-undersaturated conditions. This suggests that melt volatile evolution in the dacite was initially affected by efficient hybridisation within the magma reservoir, maintaining the quantity of volatiles in the melt at a near-consistent concentration, but the eruption was triggered by interaction with water-undersaturated mafic magma, immediately prior to eruption.

These case studies highlight the complex role of water saturation in priming a magma reservoir for eruption and the multiple pathways by which a system is conditioned for eruption. There is no evidence that magmas reached H₂O saturation immediately prior to eruption. We therefore infer that, for most magmas, an external trigger such as mafic magma injection or fluid migration within the storage system is typically required to initiate an eruption. This work demonstrates that whilst the presence of exsolved fluid is critical to syn-eruption processes, a magma body is not necessarily required to be saturated with volatiles to trigger the wholesale mobilisation and eruption of a magmatic mush.