Probabilistic volcanic hazard assessments require (1) an identification of the hazardous volcanic source; (2) estimation of the magnitude-frequency relationship for the volcanic process; (3) quantification of the dependence of hazard intensity on magnitude and external conditions; and (4) estimation of hazard exceedance from the magnitude-frequency and hazard intensity relationship. For volcanic mass flows, quantification of the hazard intensity is typically undertaken through the use of computationally expensive mass flow simulators. However, this computational expense restricts the number of samples that can be used to produce a probabilistic assessment and limits the ability to rapidly update hazard assessments in response to (e.g.) changing source probabilities. We develop an alternate approach to defining hazard intensity through a surrogate model that provides a continuous estimate of simulation outputs at negligible computational expense, demonstrated through a probabilistic hazard assessment of dome collapse (block-and-ash) flows at Taranaki volcano, New Zealand. A Gaussian Process emulator trained on a database of simulations is used as the surrogate model of hazard intensity across the input space of possible dome collapse volumes and configurations, which is then sampled using a volume-frequency relationship of dome collapse flows. The demonstrated technique is a tractable solution to the problem of probabilistic volcanic hazard assessment, with the surrogates providing a good approximation of the simulator at very limited computational expense, and is generally applicable to volcanic hazard and geo-hazard assessments that are limited by the demands of numerical simulations.

Prolonged volcanic activity can induce surface weathering and hydrothermal alteration that is a primary control on edifice instability, posing a complex hazard with its challenges to accurately forecast and mitigate. This study uses a frequently active composite volcano, Mt Ruapehu, New Zealand, to develop a conceptual model of surface weathering and hydrothermal alteration applicable to long-lived composite volcanoes. The rock samples were classified as non-altered, supergene argillic alteration, intermediate argillic alteration, and advanced argillic alteration. The first two classes have a paragenesis that is consistent with surficial infiltration and circulation of the low-temperature (40 degree C) neutral to mildly acidic fluids, inducing chemical weathering and formation of weathering rims on rock surfaces. The intermediate and advanced argillic alterations are formed from hotter (100 degree C) hydrothermal fluids with lower pH, interacting with the andesitic to dacitic host rocks. The distribution of weathering and hydrothermal alteration has been mapped with airborne hyperspectral imaging through image classification, while aeromagnetic data inversion was used to map alteration to several hundred meters depth. The joint use of hyperspectral imaging complements the geophysical methods since it can numerically identify hydrothermal alteration style. This study established a conceptual model of hydrothermal alteration history of Mt Ruapehu, exemplifying a long-lived and nested active and ancient hydrothermal system. This study highlights the need to combine mineralogical information, geophysical techniques and remote sensing to distinguish between current and ancient hydrothermal and supergene alteration systems, to indicate the most likely areas of future debris avalanche initiation.

Crystals within erupted volcanic rocks record geochemical and textural signatures during magmatic evolution prior to the onset of eruptions. Growth times of microlites can be derived through Crystal Size Distribution (CSD) analysis combined with well-constrained microlite growth rates, yielding petrologically-determined magma ascent timescales. Our newly developed, machine learning image processing scheme allows for the rapid generation of CSD, saving many hours of processing time, which previously involved hand-drawing the outer margins of crystals. For the present study, we examined a range of andesitic tephras from the Tongariro Volcanic Centre, New Zealand. A total of 228 plagioclase and pyroxene microlites CSDs were generated from individual tephra shards. All combined pyroxene and plagioclase microlite CSDs exhibit concave-up shapes, and similar intercepts and slopes at the smallest sizes. This implies similar growth durations of the smallest microlites of 15±9 to 28±15 (2σ) hours, regardless of the eruptive style or source, using an orthopyroxene microlite growth rate constrained from one of the samples. The orthopyroxene thermometer and the plagioclase hygrometer reveal the magmas were erupted at ~ 1079 to 1149 (±39 SEE), and H2O contents ranging from 0-0.4 to 0-1.7 wt.% (95% confidence maxima). In the absence of CO2, these results indicate shallow H2O exsolution pressures of < 240 bars, using a recent H2O-CO2 solubility model. Given the microlite residence times, shallow H2O exsolution driving microlite growth is inconsistent with the explosivity of the eruptions. Instead, our data suggest that the melts either carried large amounts of CO2, triggering earlier degassing of volatiles including H2O, or that microlite crystallisation began prior to degassing. Ongoing work investigates the H2O and CO2 contents hosted by melt inclusions in phenocrysts and microphenocrysts in these tephras to provide constraints on magma ascent rates, with implications for hazard characterization and mitigation.