← Back to Sessions

Enhancing Eruption Forecasting at Axial Seamount With Real-Time, Machine Learning-Based Seismic Monitoring

Description: 

Axial Seamount, an extensively instrumented submarine volcano, lies at the intersection of the Cobb-Eickelberg hot spot and the Juan de Fuca Ridge. Since late 2014, the Ocean Observatories Initiative (OOI) has operated a seven-station cabled Ocean bottom seismometers (OBS) array that captured Axial’s last eruption in April 2015. This network streams data in real-time, facilitating seismic monitoring and analysis for volcanic unrest detection and eruption forecasting. In this study, we introduce a machine learning (ML) based real-time seismic monitoring framework for Axial Seamount. Utilizing both supervised and unsupervised ML techniques, we constructed a comprehensive, high-resolution earthquake catalog, and effectively discriminated between various seismic and acoustic events. These signals include earthquakes generated by different physical processes, acoustic signals of lava-water interaction, and oceanic sources such as whale calls. Notably, our unsupervised ML analysis revealed two subgroups of earthquakes that have different spectral features and spatiotemporal behavior before and during the 2015 eruption. We interpret them as being driven by two distinct mechanisms: earthquakes on the caldera ring faults triggered by tidal stress changes, and mixed frequency earthquakes (MFEs) generated by brittle crack opening followed by influx of magma or volatile. Our system integrates ML- and double-difference based catalog construction and semi-supervised event classification in real time. Operational since 2022, it enables the discrimination and tracking of different seismic events as they occur, including precursory MFEs that potentially indicate the preparation of an eruption and seafloor impulsive events that can be used to track magma outflows during an eruption. The high-resolution earthquake catalog and real-time analysis capability complement the current deformation-based long-term forecasting methods, providing valuable short-term constraints that may enhance eruption forecasting at Axial Seamount and potentially other volcanoes.

Session: Multidisciplinary Approaches for Volcanic Eruption Forecasting - II

Type: Oral

Date: 5/2/2024

Presentation Time: 05:30 PM (local time)

Presenting Author: Kaiwen

Student Presenter: No

Invited Presentation: 

Authors