Academic Journal

Bayesian Source Characterization of Oceanic Transform Fault Earthquakes Using Relative Teleseismic Rayleigh Wave Spectra.

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Bayesian Source Characterization of Oceanic Transform Fault Earthquakes Using Relative Teleseismic Rayleigh Wave Spectra.
Συγγραφείς: Zang, Chong1 (AUTHOR), Wu, Wenbo2 (AUTHOR) wenbo.wu@berkeley.edu, Xu, Min1 (AUTHOR), Ni, Sidao3 (AUTHOR)
Πηγή: Journal of Geophysical Research. Solid Earth. Aug2026, Vol. 131 Issue 8, p1-24. 24p.
Θεματικοί όροι: *Rayleigh waves, *Parameter estimation, *Seismic wave studies, *Frequency spectra, *Surface fault ruptures, *Strike-slip faults (Geology), *Earthquakes
Περίληψη: Oceanic transform faults frequently generate moderate‐sized earthquakes in quasi‐periodic cycles, offering attractive cases for studying earthquake physics. However, characterizing these remote events remains challenging due to the scarcity of local seismic data. Teleseismic body waves can provide some constraints, but they are most effective for larger events with strong signals and suffer from significant uncertainties caused by 3‐D Earth structures along the traveling path. To improve source characterization, we develop a Bayesian inversion method based on relative spectra of teleseismic Rayleigh waves—defined as the amplitude ratios and phase shifts between target and reference events. This approach effectively mitigates the 3‐D structure effects and significantly improves the accuracy of source parameters such as horizontal locations and centroid depths. In contrast to conventional methods that assume fixed data errors, our approach infers them directly from observations—crucial for evaluating uncertainty of model parameters by the Bayesian inversion. Applying the method to five Mw ≥ 5.8 earthquakes on the Blanco transform fault yields results consistent with prior 3‐D teleseismic body wave inversions, validating the method's accuracy. Extending the analysis to 23 Mw ≥ 4.8 earthquakes indicates that coseismic slips are predominantly confined above the 600°C isotherm. Furthermore, we resolve more complex rupture processes in two Mw ≥ 6.2 events, demonstrating the method's capability to capture complex rupture behavior beyond the single point source approximation. This new method offers a promising tool for analyzing moderate‐sized earthquakes on oceanic transform faults globally using teleseismic surface waves. Plain Language Summary: Oceanic transform faults, where tectonic plates slide horizontally past each other, are active earthquake zones and serve as natural laboratories for understanding earthquake behavior. A major challenge in studying these faults is the lack of nearby seismic instruments, which limits our ability to precisely locate and characterize earthquakes. While distant seismic stations can detect earthquake signals, the signals are often weak and distorted by Earth's complex internal structure. To overcome this, we use Rayleigh surface waves—seismic waves that travel efficiently over long distances and typically have stronger amplitudes than body waves. Because earthquakes on the same fault tend to generate seismic waves that follow similar paths to distant stations, comparing their waveforms cancels out much of the distortion caused by Earth's structure and isolates differences arising from the earthquakes themselves. Applying this technique to 23 moderate‐sized earthquakes (Mw ≥ 4.8) on the Blanco transform fault, we find that most ruptures occurred above the 600°C isotherm. This supports the hypothesis that earthquake activity on these faults is limited by thermal conditions. We also show that our method can reveal detailed rupture patterns of larger events (Mw ≥ 6.0). Key Points: Bayesian method is developed for earthquake source parameter estimation using relative spectra of teleseismic Rayleigh wavesA Gaussian‐windowed narrowband approach is used to estimate frequency‐dependent measurement errors based on station‐specific spectraRelative horizontal locations and centroid depths between earthquakes can be determined with an accuracy of 1 km [ABSTRACT FROM AUTHOR]
Βάση Δεδομένων: Academic Search Index
Περιγραφή
ISSN:21699313
DOI:10.1029/2026JB034688