← Back to Sessions

Constraining 3D Fault Geometry With a Data-Driven Approach at the San Andreas – Calaveras Fault Junction

Description: 

Accurately characterizing three-dimensional (3D) fault geometry is crucial for understanding earthquake behavior and refining seismic hazard models. However, detailed subsurface fault structure is often unknown, particularly in complex fault networks. This study presents a data-driven approach to reconstruct 3D fault geometries by leveraging mapped surface traces, double-difference relocated earthquake hypocenters, and newly computed and improved focal mechanisms (Shelly et al., 2022; Skoumal et al., 2022). Our method integrates established machine learning techniques for event clustering to associate nearby events together and assigning events to mapped faults, offering improvements in resolving intricate fault networks.

Preliminary synthetic tests demonstrate that incorporating focal mechanism information significantly improves the accuracy of hypocenter clustering in anastomosing fault systems. Applying this methodology to the well-studied San Andreas-Calaveras fault junction reveals an unexpected finding that continuous quasi-linear trends of seismicity are erroneously subdivided by the clustering algorithm. We propose a refinement that combines adjacent clusters with similar orientations, resulting in a more accurate depiction of the fault network.

This research provides a low-user-input workflow for reconstructing multiple fault geometries at depth from readily available data. By analyzing the reconstructed fault network, we shed light on the complex structures involved in accommodating the partitioning of slip at San Andreas-Calaveras fault junction and highlight the potential differences in fault network morphology between locked and creeping sections of the San Andreas Fault. Our findings will contribute to a more comprehensive understanding of earthquake dynamics and improve constrains on fault connectivity for rupture forecasts.

Session: Characteristics and Mechanics of Fault Zone Rupture Processes, from Micro to Macro Scales [Poster Session]

Type: Poster

Date: 5/2/2024

Presentation Time: 08:00 AM (local time)

Presenting Author: Travis

Student Presenter: No

Invited Presentation: 

Authors