Probabilistic, Multi-sensor Eruption Forecasting DOI
Yannik Behr, Annemarie Christophersen, Craig Miller

et al.

Authorea (Authorea), Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 29, 2024

We developed an eruption forecasting model using data from multiple sensors or streams with the Bayesian Network method. The generates probabilistic forecasts that are interpretable and resilient against sensor outage. applied at Whakaari/White Island, andesite island volcano off coast of New Zealand, seismic tremor recordings, earthquake rate, CO2, SO2, H2S emission rates. At our shows increases in probability months to weeks prior three explosive eruptions were recorded since 2012. also observe consistently occurred after cumulative exceeded least 80%. Although for can be easily adapted other volcanoes, complementing existing methods rely on single streams.

Language: Английский

Petrological Evidence for Magma Mobilization Years Before the 2020/2021 Eruption of La Soufrière Volcano, St. Vincent DOI Creative Commons
Michal Camejo‐Harry, Jon Blundy, Euan Mutch

et al.

Geochemistry Geophysics Geosystems, Journal Year: 2025, Volume and Issue: 26(4)

Published: April 1, 2025

Abstract Anticipating the onset of 2020/2021 effusive‐explosive eruptive sequence at La Soufrière volcano, St. Vincent was challenging despite established monitoring networks in operation. Here, we integrate petrological data to decipher retrospectively signs imminent eruption from available pre‐eruptive data. Using diffusion chronometry, estimated timescales over which magmas transported surface. We examined olivine crystals hosted basaltic andesite scoria, categorizing them into four groups based on their textures (euhedral anhedral) and core compositions (Fo 73–89 ). Multiply zoned populations are tracked through a multi‐stage journey depth surface corresponding periods magma ascent accumulation years before eventual eruption. This correlates temporally with two phases unrest data: (a) protracted priming phase (lasting more than decade) manifesting low‐level seismicity, small crater transformations (rockfalls new fumaroles) an elevated CO 2 degassing signal; (b) subsequent transition phase, initiating just year geophysical form discrete episodes seismicity volcano inflation. Our findings provide insight dynamics mobilization Soufrière. demonstrate that magmatic roots sub‐volcanic system precedes precursors by years, drawing connections between individually ambiguous signals long timescales. Monitoring strategies optimized detect early stages unrest, such as identifying locating rarer deep routine sampling plume, could improve future responses volcanic crises Vincent.

Language: Английский

Citations

0
Tracking seismicity in an underfunded institution: The case of La Soufrière St Vincent volcanic eruption 2020–2021 DOI
Rodrigo Contreras‐Arratia, Bernard Chouet, Corentin Caudron

et al.

Journal of Volcanology and Geothermal Research, Journal Year: 2024, Volume and Issue: 446, P. 107990 - 107990

Published: Jan. 1, 2024

Language: Английский

Citations

2
Probabilistic, Multi-sensor Eruption Forecasting DOI
Yannik Behr, Annemarie Christophersen, Craig Miller

et al.

Authorea (Authorea), Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 29, 2024

We developed an eruption forecasting model using data from multiple sensors or streams with the Bayesian Network method. The generates probabilistic forecasts that are interpretable and resilient against sensor outage. applied at Whakaari/White Island, andesite island volcano off coast of New Zealand, seismic tremor recordings, earthquake rate, CO2, SO2, H2S emission rates. At our shows increases in probability months to weeks prior three explosive eruptions were recorded since 2012. also observe consistently occurred after cumulative exceeded least 80%. Although for can be easily adapted other volcanoes, complementing existing methods rely on single streams.

Language: Английский

Citations

0