Probing the Seismicity and Magmatic Plumbing System of Erebus Volcano Using Machine Learning Techniques and a Dense Near Summit Seismic Array
Description:
Nodal-style deployments have great utility for studies of active volcanoes. Mount Erebus, Antarctica, the southernmost active volcano in the world, has been continuously active since its discovery in 1841. For the past decades the volcano has produced frequent Strombolian explosions originating from a lava lake within its summit crater. The seismic activity at Mt. Erebus has been monitored by the Mount Erebus Volcano Observatory (MEVO). Eruptions from the lava lake and icequakes from the highly glaciated volcano dominate seismicity catalogues from Erebus. This contrasts with other active volcanoes which typically exhibit a wide range of internal seismicity, including volcano-tectonic earthquakes, (deep) long-period events, and hybrid events.
In this study, we focus on detecting and interpreting multiple types of seismicity at Erebus volcano using artificial intelligence techniques. To accomplish this, we apply multiple machine learning phase pickers (EQTransformer and PhaseNet), trained with different data sets that include waveforms from tectonic and volcanic observatories, to approximately two months of seismic data from an array of over 100 three component short-period stations, augmented by a small number of broadband stations, that were deployed during the austral summer of 2008. We associate picks into events using PyOcto, a high-throughput seismic phase associator. This procedure detects thousands of events within the two months of data, with the greatest number of events being detected by EQTransformer trained on a volcanic dataset. We obtain hypocenter locations using a solver that employs the fast marching method and accommodates topography. We characterize and cluster the seismic events based on dynamic time warping distances, frequency indices, and hypocentral locations. Our objective is to develop a concise but high-resolution seismic event catalog for Mount Erebus to explore its internal and near-surface seismicity for further insight into its magmatic plumbing system.
Session: Innovative Applications of Seismic Nodal Technology for Hazard Mitigation and Earth System Monitoring - I
Type: Oral
Date: 4/15/2025
Presentation Time: 02:00 PM (local time)
Presenting Author: Andres
Student Presenter: No
Invited Presentation:
Poster Number:
Authors
Andres Pena Castro Presenting Author Corresponding Author andresfpenacastro@gmail.com University of New Mexico |
Brandon Schmandt bschmandt@unm.edu University of New Mexico |
Ricardo Garza Giron r.garza_giron@colostate.edu Colorado State University |
Richard Aster rick.aster@colostate.edu Colorado State University |
|
|
|
|
|
Probing the Seismicity and Magmatic Plumbing System of Erebus Volcano Using Machine Learning Techniques and a Dense Near Summit Seismic Array
Category
Innovative Applications of Seismic Nodal Technology for Hazard Mitigation and Earth System Monitoring