Aluto volcano, in the Main Ethiopian Rift, is a peralkaline caldera system, which comprises conglomerations of rhyolite (obsidian) lavas and enigmatic pumice cones. Recent work at Aluto has found that pumice cone eruptions are highly unsteady, and form convective eruption plumes that frequently collapse to generate pyroclastic density currents (PDCs). We develop a methodology and present results for the first probabilistic volcanic hazard assessment (PVHA) for PDCs at a pumice cone volcano. By doing so, we estimate the conditional probability of inundation by PDCs around Aluto volcano, incorporating the aleatory uncertainty in PDC hazard. We employ a Monte Carlo energy cone modeling approach, which benefits from parameterization informed by field investigations and volcanic plume modeling. We find that despite the relatively modest eruptions that are likely to occur, the wide distribution of past vent locations (and thus the high uncertainty of where future vents might open), results in a broad area being potentially at risk of inundation by PDCs. However, the aleatory uncertainty in vent opening means that the conditional probabilities are lower (≤ 0.12), and more homogeneous, over the hazard domain compared to central-vent volcanoes (where conditional probabilities are often ≤ 1 close to the vent). Despite this, numerous settlements, amenities, and economically valuable geothermal infrastructure, lie within the most hazardous (P(PDC|eruption) ≥ 0.05) regions of Aluto caldera. The Monte Carlo energy cone modeling approach provides a quantitative, accountable and defendable background and long-term PVHA for PDCs from Aluto. These results could be combined in the future with hazard assessments relating to tephra fall and/or lava to develop a comprehensive volcanic hazard map for the caldera. Following appropriate parameterization, the approach developed here can also be used to compute a PDC PVHA at other volcanoes where vent location is uncertain.
Full Text Finder