Machine Learning in Large and Small Earthquakes - From Rapid Large Earthquake Characterization to Slow Fault Zone Processes - Dissertation

Aug 5, 2026

Upload by Patrick

Jiun-Ting Lin’s dissertation integrates machine-learning with traditional seismic analysis to address challenges in large- and small-earthquake research. For large events, it introduces M-LARGE to rapidly predict magnitude using rupture simulations and GNSS data, achieving about 99% accuracy and avoiding saturation limitations. It also uses M-LARGE to infer finite-fault parameters and their evolution for fast ground-motion forecasting. For smaller events, it detects low-frequency earthquakes from noisy waveforms, increasing event counts and improving temporal resolution. The work links fast and slow slip behaviors in the 2018 M7.1 Hawaii earthquake and validates rupture simulations using effective stress to support more accurate hazard assessment.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages145

Reading Time6 h 5 min

AccessFree to download

Dissertation Abstract

Large-earthquake magnitude prediction with M-LARGE

Rapid finite-fault parameter forecasting

Small-earthquake detection of LFEs

Linking fast/slow earthquakes and hazard implications

MACHINE LEARNING IN LARGE AND SMALL EARTHQUAKES: FROM RAPID LARGE EARTHQUAKE CHARACTERIZATION TO SLOW FAULT ZONE PROCESSES by JIUN-TING LIN A DISSERTATION Presented to the Department of Earth Sciences and the Division of Graduate Studies of the University of Oregon in partial fulfillment of the requirements for the degree of Doctor of Philosophy December 2022 DISSERTATION APPROVAL PAGE Student: Jiun-Ting Lin Title: Machine Learning in Large and Small Earthquakes: from Rapid Large Earthquake Characterization to Slow Fault Zone Processes This dissertation has been accepted and approved in partial fulfillment of the requirements for the Doctor of Philosophy degree in the Department of Earth Sciences by: Amanda M. Thomas Diego Melgar Valerie J. Sahakian Jake Searcy Thien Nguyen Chairperson and Advisor Core Member and Advisor Core Member Core Member Institutional Representative and Krista Chronister Vice Provost for Graduate Studies Original approval signatures are on file with the University of Oregon Division of Graduate Studies. Degree awarded December 2022 © 2022 Jiun-Ting Lin This work is licensed under a Creative Commons Attribution License DISSERTATION ABSTRACT Jiun-Ting Lin Doctor of Philosophy Department of Earth Sciences December 2022 Title: Machine Learning in Large and Small Earthquakes: from Rapid Large Earthquake Characterization to Slow Fault Zone Processes This dissertation summarizes the work of integrating machine-learning and traditional seismic analysis techniques into large and small earthquake problems. Earthquake early warning for large magnitude earthquakes is one of the most challenging problems in seismology. Here I develop an algorithm, called M-LARGE, that harnesses machine-learning, rupture simulations, and GNSS data to rapidly predict magnitude without saturation issue with an accuracy of 99%, outperforming other similar methods. I then show how M-LARGE can predict finite fault parameters and their evolution when rupture unfolds for fast and accurate ground motion forecasting. This dissertation will demonstrate how machine-learning can be used as a data mining tool to detect small magnitude seismicity buried in noisy waveforms. I will show its application to detect LFEs, a special class of small earthquakes typically occur downdip of the seismogenic zone. The model detects more than five times the number of events than the original catalog in Vancouver Island and can apply to unseen stations, which provides a more flexible way to refine the temporal resolution of subduction zone processes. Finally, I will show how do small and slow earthquakes link to large and fast events and their implication on earthquake hazard assessment. With jointly inverted GNSS, strong motion, and tsunami data of the 2018 M7.1 Hawaii earthquake, I find that fast slip ruptures into the area previously hosts slow slip. The result is further validated by rupture simulations, where we find that the effective stress can be a factor that exerts a dominant control on the rupture extent. This reinforces the idea that an individual section of fault can host a variety of distinct slip behaviors, and slow slip should be considered as rupture extent for a more accurate hazard assessment. This dissertation includes previously published and unpublished co-authored material. CURRICULUM VITAE NAME OF AUTHOR: Jiun-Ting Lin GRADUATE AND UNDERGRADUATE SCHOOLS ATTENDED: University of Oregon, Eugene National Central University, Taiwan National Central University, Taiwan DEGREES AWARDED: Doctor of Philosophy, Earth Sciences, 2022, University of Oregon Master of Science, Geophysics, 2014, National Central University Bachelor of Science, Earth Sciences, 2012, National Central University AREAS OF SPECIAL INTEREST: Seismology Machine Learning Data Science Inverse Problem PROFESSIONAL EXPERIENCE: Graduate Employee, University of Oregon, 2017–2022 Intern, La

FAQs

What does M-LARGE do in the dissertation?

M-LARGE uses machine learning with rupture simulations and GNSS data to rapidly predict earthquake magnitude, avoiding saturation issues, and supports accurate forecasting.

How is machine learning used for small earthquakes?

The dissertation shows a data-mining approach to detect low-frequency earthquakes buried in noisy waveforms, improving event detection compared with an original catalog.

How do small and slow earthquakes relate to large and fast events in the findings?

Jointly inverted GNSS, strong-motion, and tsunami data for the 2018 M7.1 Hawaii earthquake indicate that fast slip can rupture into areas previously hosting slow slip, with rupture simulations suggesting effective stress governs rupture extent.

Aug 5, 2026

Upload by Patrick