[S-CG45] Science of slow-to-fast earthquakes

[SCG45-39] Illuminating Earthquake Sources with Borehole Fiber Optics: Insights from the 2022 M_L6.8 Chihshang, Taiwan, Earthquake

*Jolan Liao¹, Yen-Yu Lin¹, Kuo-Fong Ma² (1.National Central University, Taiwan , 2.Institute of Earth Sciences, Academia Sinica, Taiwan)

Keywords:Distributed Acoustic Sensing (DAS) , Seismic Source Characteristics , Borehole Seismic Analysis

Recent advancements in distributed acoustic sensing (DAS) offer a transformative approach for seismological research. A DAS system has a broadband instrument response, measuring continuous strain or strain rate along fiber-optic cables, effectively functioning as a spatially dense seismic array. With its high spatial resolution, integrating DAS technology into existing seismic observation frameworks has the potential to enhance the ability to capture detailed seismic information. Despite these promising applications, DAS remains an emerging tool for seismic source analysis. In this study, we explore the utility of DAS from the Milun fault Drilling and All-inclusive Sensing (MiDAS) project for analyzing source characteristics of the 18 September 2022 M_L6.8 Chihshang, Taiwan, earthquake. Established in 2022, MiDAS integrates borehole drilling, DAS, and borehole seismometers (BHS) to facilitate unprecedented seismic observations across the Milun fault in Hualien, Taiwan. MiDAS recorded over 10 seismic events with magnitudes exceeding 6 in Taiwan, including the 2022 Chihshang earthquake sequence. Notably, the 18 September 2022 M_L6.8 Chihshang mainshock occurred within a complex fault system, where the west-dipping Central Range fault system (CRFs) meets the east-dipping Longitudinal Valley fault system (LVFs), forming a head-to-head conjugate fault system. It is a unique opportunity to assess the capabilities of DAS in resolving rupture characteristics of large earthquakes in complex tectonic environments. Our analysis utilizes the borehole fiber-optic data, along with seismic records at frequencies of 0.05-10 Hz from downhole seismometers and nearby broadband and strong-motion stations at the surface, to evaluate the relative source time function (RSTF) using the empirical Green’s function (EGF) deconvolution method. Due to complications associated with later arrivals in near-field seismic recordings for large earthquakes, we focus on the initial 20 seconds of the rupture process rather than attempting to resolve the entire rupture behavior through RSTFs. A key advantage of using borehole-based system with DAS and BHS is their high signal-to-noise ratio, which enables the detection of subtle variations in rupture behavior. Additionally, borehole DAS records, logging data, and drilling core samples allow for high-resolution site response analysis and identification of rock site locations, helping to determine whether signals in RSTFs originate directly from the source. The results reveal that RSTFs exhibit a bumpy-shaped pattern of energy release with three significant energy bursts. Moreover, the spectrum of RSTFs derived from both DAS and seismometers exhibit consistent high-frequency (> 1 Hz) source characteristics. These findings highlight the potential of DAS technology to complement traditional seismic networks and capture small-scale rupture details, opening new avenues for understanding earthquake rupture process in greater detail.