Ergodic seismic precursors and transfer learning for short term eruption forecasting at data scarce volcanoes

datacite.alternateIdentifier.citationNature Communications, 16 (1), 2025
datacite.alternateIdentifier.doi10.1038/s41467-025-56689-x
datacite.alternateIdentifier.issn2041-1723
datacite.creatorArdid, Alberto
datacite.creatorDempsey, David
datacite.creatorCaudron, Corentin
datacite.creatorCronin, Shane J.
datacite.creatorKennedy, Ben
datacite.creatorGirona, Társilo
datacite.creatorRoman, Diana C.
datacite.creatorMiller, Craig
datacite.creatorPotter, Sally H.
datacite.creatorLamb, Oliver D.
datacite.creatorMartanto, Anto
datacite.creatorÇubuk Sabuncu, Ye?im
datacite.creatorCabrera, Leoncio
datacite.creatorRuiz, Sergio
datacite.creatorContreras-Arratia, Rodrigo
datacite.creatorPacheco, J. F.
datacite.creatorMora, Mauricio M.
datacite.creatorDe Angelis, Silvio
datacite.date2025
datacite.rightsAcceso Abierto
datacite.subjectBiodiversidad
datacite.subjectEcología de ecosistemas
datacite.subjectEcología de aguas dulces
datacite.subject.englishBiodiversity
datacite.subject.englishEcosystem ecology
datacite.subject.englishFreshwater ecology
datacite.titleErgodic seismic precursors and transfer learning for short term eruption forecasting at data scarce volcanoes
dc.date.accessioned2025-08-06T18:22:41Z
dc.date.available2025-08-06T18:22:41Z
dc.description.abstractAbstract Seismic data recorded before volcanic eruptions provides important clues for forecasting. However, limited monitoring histories and infrequent eruptions restrict the data available for training forecasting models. We propose a transfer machine learning approach that identifies eruption precursors signals that consistently change before eruptions across multiple volcanoes. Using seismic data from 41 eruptions at 24 volcanoes over 73 years, our approach forecasts eruptions at unobserved (out-of-sample) volcanoes. Tested without data from the target volcano, the model demonstrated accuracy comparable to direct training on the target and exceeded benchmarks based on seismic amplitude. These results indicate that eruption precursors exhibit ergodicity, sharing common patterns that allow observations from one group of volcanoes to approximate the behavior of others. This approach addresses data limitations at individual sites and provides a useful tool to support monitoring efforts at volcano observatories, improving the ability to forecast eruptions and mitigate volcanic risks.
dc.formatPDF
dc.identifier.issn2041-1723
dc.identifier.urihttps://repositoriodigital.uct.cl/handle/10925/6505
dc.language.isoen
dc.publisherNature Publishing Group
dc.rightsObra bajo licencia Creative Commons Atribución 4.0 Internacional
dc.rights.licensehttps://creativecommons.org/licenses/by/4.0/
dc.sourceNature Communications
dspace.entity.typePublication
oaire.citationEdition2025
oaire.citationIssue1
oaire.citationTitleNature Communications
oaire.citationVolume16
oaire.fundingReferenceMinistry of Business, Innovation and Employment - MBIE
oaire.resourceTypeArtículo
oaire.resourceType.enArticle
uct.catalogadorjvu
uct.comunidadRecursos Naturalesen_US
uct.departamentoDepartamento de Ciencias Ambientales
uct.facultadFacultad de Recursos Naturales
uct.indizacionScience Citation Index Expanded - SCIE
uct.indizacionSCOPUS
uct.indizacionWOS
uct.indizacionDOAJ
uct.indizacionPubMed
Files
Collections