A generalized deep learning model to detect and classify volcano seismicity

Volcano seismicity is often detected and classified based on its spectral properties. However, the wide variety of volcano seismic signals and increasing amounts of data make accurate, consistent, and efficient detection and classification challenging. Machine learning (ML) has proven very effective...

Full description

Saved in:
Bibliographic Details
Main Authors: David Fee, Darren Tan, John Lyons, Mariangela Sciotto, Andrea Cannata, Alicia Hotovec-Ellis, Társilo Girona, Aaron Wech, Diana Roman, Matthew Haney, Silvio De Angelis
Format: Article
Language:English
Published: Volcanica 2025-06-01
Series:Volcanica
Subjects:
Online Access:https://jvolcanica.org/ojs/index.php/volcanica/article/view/349
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849336919847075840
author David Fee
Darren Tan
John Lyons
Mariangela Sciotto
Andrea Cannata
Alicia Hotovec-Ellis
Társilo Girona
Aaron Wech
Diana Roman
Matthew Haney
Silvio De Angelis
author_facet David Fee
Darren Tan
John Lyons
Mariangela Sciotto
Andrea Cannata
Alicia Hotovec-Ellis
Társilo Girona
Aaron Wech
Diana Roman
Matthew Haney
Silvio De Angelis
author_sort David Fee
collection DOAJ
description Volcano seismicity is often detected and classified based on its spectral properties. However, the wide variety of volcano seismic signals and increasing amounts of data make accurate, consistent, and efficient detection and classification challenging. Machine learning (ML) has proven very effective at detecting and classifying tectonic seismicity, particularly using Convolutional Neural Networks (CNNs) and leveraging labeled datasets from regional seismic networks. Progress has been made applying ML to volcano seismicity, but efforts have typically been focused on a single volcano and are often hampered by the limited availability of training data. We build on the method of Tan et al. [2024] (10.1029/2024JB029194) to generalize a spectrogram-based CNN termed the VOlcano Infrasound and Seismic Spectrogram Neural Network (VOISS-Net) to detect and classify volcano seismicity at any volcano. We use a diverse training dataset of over 270,000 spectrograms from multiple volcanoes: Pavlof, Semisopochnoi, Tanaga, Takawangha, and Redoubt volcanoes\replaced (Alaska, USA); Mt. Etna (Italy); and Kīlauea, Hawai`i (USA). These volcanoes present a wide range of volcano seismic signals, source-receiver distances, and eruption styles. Our generalized VOISS-Net model achieves an accuracy of 87 % on the test set. We apply this model to continuous data from several volcanoes and eruptions included within and outside our training set, and find that multiple types of tremor, explosions, earthquakes, long-period events, and noise are successfully detected and classified. The model occasionally confuses transient signals such as earthquakes and explosions and misclassifies seismicity not included in the training dataset (e.g. teleseismic earthquakes). We envision the generalized VOISS-Net model to be applicable in both research and operational volcano monitoring settings.
format Article
id doaj-art-1bdabea9f17c4ec8b3ffb7070853019c
institution Kabale University
issn 2610-3540
language English
publishDate 2025-06-01
publisher Volcanica
record_format Article
series Volcanica
spelling doaj-art-1bdabea9f17c4ec8b3ffb7070853019c2025-08-20T03:44:51ZengVolcanicaVolcanica2610-35402025-06-018130532310.30909/vol/rjss1878334A generalized deep learning model to detect and classify volcano seismicityDavid Fee0https://orcid.org/0000-0002-0936-9977Darren Tan1https://orcid.org/0000-0001-8210-6041John Lyons2https://orcid.org/0000-0001-5409-1698Mariangela Sciotto3https://orcid.org/0000-0001-8711-1392Andrea Cannata4https://orcid.org/0000-0002-0028-5822Alicia Hotovec-Ellis5https://orcid.org/0000-0003-1917-0205Társilo Girona6https://orcid.org/0000-0001-6422-0422Aaron Wech7https://orcid.org/0000-0003-4983-1991Diana Roman8https://orcid.org/0000-0003-1282-5803Matthew Haney9Silvio De Angelis10https://orcid.org/0000-0003-2636-3056Alaska Volcano Observatory, Geophysical Institute, University of Alaska Fairbanks, Fairbanks, AK, USA.Alaska Volcano Observatory, Geophysical Institute, University of Alaska Fairbanks, Fairbanks, AK, USA.U.S. Geological Survey, Volcano Science Center, Alaska Volcano Observatory, Anchorage, AK, USA.Istituto Nazionale di Geofisica e Vulcanologia, Osservatorio Etneo – Sezione di Catania, Catania, Italy.Istituto Nazionale di Geofisica e Vulcanologia, Osservatorio Etneo – Sezione di Catania, Catania, Italy.U.S. Geological Survey, California Volcano Observatory, Moffett Field, CA, USA.Alaska Volcano Observatory, Geophysical Institute, University of Alaska Fairbanks, Fairbanks, AK, USA.U.S. Geological Survey, Volcano Science Center, Alaska Volcano Observatory, Anchorage, AK, USA.Earth and Planets Laboratory, Carnegie Institution for Science, Washington DC 20015, USA.U.S. Geological Survey, Volcano Science Center, Alaska Volcano Observatory, Anchorage, AK, USA.School of Environmental Sciences, University of Liverpool, Liverpool, England, UK.Volcano seismicity is often detected and classified based on its spectral properties. However, the wide variety of volcano seismic signals and increasing amounts of data make accurate, consistent, and efficient detection and classification challenging. Machine learning (ML) has proven very effective at detecting and classifying tectonic seismicity, particularly using Convolutional Neural Networks (CNNs) and leveraging labeled datasets from regional seismic networks. Progress has been made applying ML to volcano seismicity, but efforts have typically been focused on a single volcano and are often hampered by the limited availability of training data. We build on the method of Tan et al. [2024] (10.1029/2024JB029194) to generalize a spectrogram-based CNN termed the VOlcano Infrasound and Seismic Spectrogram Neural Network (VOISS-Net) to detect and classify volcano seismicity at any volcano. We use a diverse training dataset of over 270,000 spectrograms from multiple volcanoes: Pavlof, Semisopochnoi, Tanaga, Takawangha, and Redoubt volcanoes\replaced (Alaska, USA); Mt. Etna (Italy); and Kīlauea, Hawai`i (USA). These volcanoes present a wide range of volcano seismic signals, source-receiver distances, and eruption styles. Our generalized VOISS-Net model achieves an accuracy of 87 % on the test set. We apply this model to continuous data from several volcanoes and eruptions included within and outside our training set, and find that multiple types of tremor, explosions, earthquakes, long-period events, and noise are successfully detected and classified. The model occasionally confuses transient signals such as earthquakes and explosions and misclassifies seismicity not included in the training dataset (e.g. teleseismic earthquakes). We envision the generalized VOISS-Net model to be applicable in both research and operational volcano monitoring settings.https://jvolcanica.org/ojs/index.php/volcanica/article/view/349volcano seismicitymachine learningtremorexplosion
spellingShingle David Fee
Darren Tan
John Lyons
Mariangela Sciotto
Andrea Cannata
Alicia Hotovec-Ellis
Társilo Girona
Aaron Wech
Diana Roman
Matthew Haney
Silvio De Angelis
A generalized deep learning model to detect and classify volcano seismicity
Volcanica
volcano seismicity
machine learning
tremor
explosion
title A generalized deep learning model to detect and classify volcano seismicity
title_full A generalized deep learning model to detect and classify volcano seismicity
title_fullStr A generalized deep learning model to detect and classify volcano seismicity
title_full_unstemmed A generalized deep learning model to detect and classify volcano seismicity
title_short A generalized deep learning model to detect and classify volcano seismicity
title_sort generalized deep learning model to detect and classify volcano seismicity
topic volcano seismicity
machine learning
tremor
explosion
url https://jvolcanica.org/ojs/index.php/volcanica/article/view/349
work_keys_str_mv AT davidfee ageneralizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT darrentan ageneralizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT johnlyons ageneralizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT mariangelasciotto ageneralizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT andreacannata ageneralizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT aliciahotovecellis ageneralizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT tarsilogirona ageneralizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT aaronwech ageneralizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT dianaroman ageneralizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT matthewhaney ageneralizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT silviodeangelis ageneralizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT davidfee generalizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT darrentan generalizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT johnlyons generalizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT mariangelasciotto generalizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT andreacannata generalizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT aliciahotovecellis generalizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT tarsilogirona generalizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT aaronwech generalizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT dianaroman generalizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT matthewhaney generalizeddeeplearningmodeltodetectandclassifyvolcanoseismicity
AT silviodeangelis generalizeddeeplearningmodeltodetectandclassifyvolcanoseismicity