Pre-eruptive conditions, crustal processes, and magmatic timescales recorded in products of Calbuco and Osorno volcanoes, Southern Andes

The Southern Volcanic Zone (SVZ) of the Andes is an active volcanic chain generated by the subduction of the Nazca Plate below the South American Plate. The SVZ has been subdivided into four segments, based on tectonic setting. One of those four subdivisions is the Central Southern Volcanic Zone (CSVZ), where the most active volcanoes of the SVZ are located: Villarrica, Llaima, Calbuco, Puyehue-Cordón Caulle, Osorno, and Mocho-Choshuenco. In recent years, some of these volcanoes have erupted (e.g. Villarrica in 2015, Llaima in 2008, Puyehue-Cordón Caulle in 2011‒2012) and those eruptions have been captured by volcano monitoring. By contrast, other active volcanoes have erupted for the last time before monitoring instruments were installed in the neighbouring areas (e.g. Osorno in 1835 and Mocho-Choshuenco in 1864). Petrological studies are useful to assess previous eruptions and, as a consequence, can be used in volcano monitoring and forecasting. The present study is focused on the most recent eruptions of Calbuco and Osorno stratovolcanoes. In April 2015, an unpredicted rapid-onset eruption occurred at Calbuco Volcano in the Southern Andes of Chile. This event consisted of two, sub-Plinian eruptions separated by a few hours. I identify an upper-crustal reservoir using available geophysical data combined with amphibole geobarometry. Based on textural features, I conclude the presence of a mush zone within this reservoir. From the collected samples, whole-rock chemistry and an array of geothermometers of silicate phases (amphibole, amphibole-plagioclase, two-pyroxenes) gave similar conditions for all samples possessing ~40 crystal volume, with the exception of the sample Cal-160 (~60 % crystal volume), which is slightly more evolved and yields lower temperatures for all silicate-phase geothermometers. In addition, I performed Fe-Ti oxides thermometry, which gave lower temperatures than those silicate-phase based. Comparing temperatures calculated by pairs of ilmenite-titanomagnetite core compositions with those calculated using rim compositions, I observe a late-stage temperature increase of between 70ºC and 200ºC in sample Cal-160, which represents the bottom of the crystal mush and the magma reservoir. I suggest this recognised heating triggered the eruption requiring the involvement of a hot, presumably mafic magma injection at the base of a shallow, crystal-rich reservoir, though the mafic magma was not itself erupted. I determine timescales of pre-eruptive heating at the partially-solidified chamber base (represented by sample Cal-160) and constrain the magma residence time for the bulk of the carrier magma (represented by the rest of the collected samples). The ilmenite-titanomagnetite pairs from sample Cal-160 yielded magmatic timescales of < 4 days, interpreted as time between the triggering and the eruption. These short magmatic timescales explain why the April 2015 Calbuco eruption showed precursors of only a few hours. By contrast, the ilmenite-titanomagnetite pairs of other samples (distinct from Cal-160) yielded magmatic timescales of > one year, which represent resident, eruptible magmas in the middle of the reservoir at stable temperatures and oxygen fugacity. This portion of the magma reservoir did not interact with any magma recharge immediately prior to or during the eruption, indicating the eruption was triggered before the thermal homogenisation of the magma reservoir beneath Calbuco volcano. During 1835, two eruptive events occurred; the first during January-February and the second during November-December. The erupted products of both events are lavas and tephra fall deposits of basaltic-andesite composition (52.4‒52.9 SiO2 wt%) with of crystallinity ranging from 23 to 45 vol% comprising: olivine, plagioclase, clinopyroxene, and Cr-spinel. Using the erupted products, I calculated pre-eruptive conditions: temperature of ~1,140ºC (via olivine-augite and Ca-in-olivine thermometry), oxygen fugacity of ΔQFM +0.3 (via Cr-spinel-melt oxybarometry and numerical modelling), dissolved water content up to ~1.5 wt% (via numerical modelling), and depths of up to 4.5 km (numerical modelling). Based on textural features, I conclude the presence of a mush zone within this reservoir, which underwent partial disaggregation prior both eruptive events. The studied eruption presents the same eruptive features, mineral assemblage, and composition as the historical products from Osorno volcano, in consequence, it can be used to assess other previous eruptive events and future behaviour. In addition, comparison of products from the neighbour stratovolcanoes Calbuco, Osorno and La Picada was performed using trace elements (fluid mobile/immobile and rare earth elements) and 87Sr/86Sr and 143Nd/144Nd isotope ratios. These analyses display higher both slab-derived fluid input and partial melting degree in the mantle beneath stratovolcanoes (Osorno, Calbuco, and La Picada) than in small eruptive centres (represented by La Viguería cones). Although the most significant signature comes from the mantle source, these results suggest also the minor influence of crustal assimilation. Finally, implications of this study are argued, comparing with other volcanic system from the CSVZ at 39º30’S latitude, which exhibit some similarities.