Oral Presentation

Near-field observations of the dynamic and quasi-static processes before, during, and after earthquakes are essential for understanding earthquake physics. In particular, the nucleation process, characterized by foreshocks, aseismic slip, or both as widely observed in laboratory experiments, is critical to earthquake forecasting and hazard mitigation but remains an observational challenge in the field. Fiber-optic sensing technologies, particularly Distributed Acoustic Sensing (DAS), are revolutionizing seismology through long-range, continuous, and cost-effective sensing along fiber-optic cables across diverse environments. The rapidly growing field of “fiber-optic seismology” has enabled high-resolution imaging of earthquake sequences and fault structures. Our recent work demonstrates that dense spatial sampling of DAS significantly improves the resolution of earthquake focal mechanisms and rupture processes. Beyond seismic applications, DAS’s broadband sensitivity has extended into geodetic applications, heralding the emergence of “fiber-optic geodesy”. Notably, our study in Iceland shows that DAS can capture quasi-static deformation from magma intrusions, providing precursory signals that enabled early warnings tens of minutes before eruptions. This dual capability of seismic and geodetic sensing allows for a more holistic monitoring of seismogenic zones across multiple spatiotemporal scales. Dense spatial sampling along hundred-kilometer-long telecom cables or borehole-installed fibers may provide critical near-source observations of earthquake nucleation processes in the field. Integrating these multi-parameter datasets, including seismic, geodetic, and thermal measurements, with machine learning, has the potential to greatly advance earthquake forecasting.