Japan Geoscience Union Meeting 2024

Presentation information

[E] Poster

S (Solid Earth Sciences ) » S-CG Complex & General

[S-CG40] Science of slow-to-fast earthquakes

Tue. May 28, 2024 5:15 PM - 6:45 PM Poster Hall (Exhibition Hall 6, Makuhari Messe)

convener:Aitaro Kato(Earthquake Research Institute, the University of Tokyo), Asuka Yamaguchi(Atomosphere and Ocean Research Institute, The University of Tokyo), Yohei Hamada(Japan Agency for Marine-Earth Science and Technology), Akemi Noda(Meteorological Research Institute, Japan Meteorological Agency)

5:15 PM - 6:45 PM

[SCG40-P05] Insights into the Mechanics of the Lower Crust from Focal Mechanisms of Low-Frequency Earthquakes on the San Andreas Fault

*Naofumi Aso1, Miki Aso2, David R Shelly3, Satoshi Ide4 (1.Tokyo University of Science, 2.OYO RMS, 3.USGS, 4.University of Tokyo)

The structure of the deep extension of faults, where ordinary earthquakes rarely occur, is not well understood. However, the mechanics of the deeper parts of faults can influence the potential for large earthquakes at shallower depths.
Although ordinary crustal earthquakes are usually limited to shallow depths (above about 15 km depth), low-frequency earthquakes (LFEs) occur at deeper depths of some tectonic faults. LFEs provide direct evidence of ongoing deformation at these depths and strong constraints on crustal stress. We focus on LFEs near Parkfield on the San Andreas Fault, which have been studied in considerable detail except for their focal mechanisms.
To determine focal mechanisms, we utilize the catalog of more than a million LFEs since 2001 (Shelly, 2017). Since they are classified into 88 families based on waveform correlations, events in the same family are considered to share the same location and mechanism. After waveform stacking for all events in the same family, we inverted the absolute values of P- and S-wave energies to determine focal mechanisms. For efficient optimization, we first grid search for the common focal mechanism for all families and then carry out an iterative approach to determine family-specific focal mechanisms and site amplification terms together. We also carried out the same analysis for each year's data to evaluate the potential temporal change of seismometer response.
As a result, we find that most of the LFEs are consistent with strike-slip plate motion on the San Andreas Fault (N40ºW), with moderate variations. Families of LFEs located to the northwest of Parkfield have clockwise rotation (N30ºW), consistent with the local alignment of LFE hypocenters. Together with shallow seismicity distribution, we find a left-handed twisting structure, which would be related to the transitional mechanical coupling.
Meanwhile, southeastern families have focal mechanisms consistent with shallow seismicity and surface trace, but their distribution has significant across-fault offsets. We speculate a sub-parallel branching structure in the southeast region, consistent with the mechanical separation for off-fault families suggested by migration patterns (Shelly, 2015).
We also find variations in rake angle (slip inclination) up to ~20º. We investigate the correlation between their spatial pattern and shallow seismicity or surface creep rate to understand the implications for the ductile deformation field at deeper depths