← Back to Sessions

Earthquake Rupture Propagation and Arrest in a Highly Variable Stochastic Stress Field

Description: 

Earthquakes occur in populations with few large events and many small ones, yet many earthquake rupture models employ a relatively smooth stress field and other conditions that prohibit the co-occurrence of both large M7 earthquakes and smaller M1 to M3 foreshocks and aftershocks. We present a model where the multi-scale geometrical complexity of rough faults and the heterogeneous stresses that develop within fault zones are represented with a stochastic stress field Dt_pot(x). Sometimes referred to as effective stress, Dt_pot(x) is related to the strain energy that can provide fuel for an earthquake and is defined as the initial shear stress level relative to the fault’s residual sliding strength. We study populations of earthquakes that propagate and arrest in the stochastic stress field based on a highly simplified 1-D fracture mechanics framework (e.g. Ampuero, Ripperger, Mai, 2006). Specifically, we explore how properties of the earthquake populations—such as the magnitude dependence of stress drop and the Gutenberg-Richter frequency magnitude distribution—vary as a function of statistical properties of the stress field such as the coefficient of variation and the scaling exponent m (Beeler, 2023) which describes how Dt_pot(x) changes as a function of wavelength. We find that realistic populations of earthquakes, with more small ones than large ones and magnitude independent stress drops, are produced for stochastic stress fields with m < 0.5 (m = 1.5 is self-similar, m = 1.0 is Brownian, m = 0 is white noise). This suggests that Dt_pot(x) must be highly variable even at short wavelengths, a property that is absent from most state-of-the-art dynamic rupture models and unresolvable with kinematic finite fault inversions. The highly variable stress fields produce earthquake ruptures that nearly arrest and rapidly reaccelerate. These characteristics are consistent with observations of earthquake source time functions with multiple sub-events, and they provide a mechanism to generate high frequency ground motion that is lacking in many rupture models.

Session: Numerical Modeling in Seismology: Theory, Algorithms and Applications - II

Type: Oral

Date: 4/17/2025

Presentation Time: 11:30 AM (local time)

Presenting Author: Gregory

Student Presenter: No

Invited Presentation: 

Poster Number:

Authors