Characteristic scales of earthquake rupture from numerical models

Abstract

Abstract. Numerical models of earthquake rupture are used to investigate characteristic length scales and size distributions of repeated earthquakes on vertical, planar fault segments. The models are based on exact solutions of static three-dimensional (3-D) elasticity. Dynamical rupture is approximated by allowing the static stress field to expand from slip motions at a single velocity. To show how the vertical fault width affects earthquake size distributions for a broad range of fault behaviors, two different fault strength models are used; a smooth model and a heterogeneous asperity model. The smooth model is a simplified version of the Dieterich-Ruina rate and state dependent friction law. The heterogeneous asperity model uses a slip-dependent random powerlaw strength distribution. It is shown that the characteristic scale of fault segmentation is proportional to the vertical width of a seismogenic fault. This conclusion holds for both the smooth and the heterogeneous models. For the smooth models characteristic quake distributions result, with populations of large events that are obviously distinct from smaller events. The distributions of large events have well-defined mean lengths and moments. The heterogeneous models result in Gutenberg-Richter (GR) powerlaw distributions of event sizes up to a characteristic quake size. Quakes larger than the characteristic size fall off the GR distribution such that the powerlaw would greatly overestimate the probability of occurrence of the larger events.

The Dynamics of Elongated Earthquake Ruptures

“…proposed that continental strike-slip earthquakes have a characteristic length of segmentation related to the seismogenic width (Klinger, 2010), a feature found in 3-D earthquake models (Heimpel, 2003). Our results can also help anticipate the range of rupture scenarios possible in faults where the seismogenic width changes substantially along strike, that is, the San Andreas Fault (Smith-Konter et al, 2011) and the Alpine Fault in New Zealand (Michailos et al, 2019).…”

“…Fracture mechanics theory and dynamic rupture simulations indicate that the seismogenic width controls the maximum fault‐step over distance that a rupture can jump (Bai & Ampuero, ) and the maximum thickness of fault damage zones (Ampuero & Mao, ). It has also been proposed that continental strike‐slip earthquakes have a characteristic length of segmentation related to the seismogenic width (Klinger, ), a feature found in 3‐D earthquake models (Heimpel, ). Our results can also help anticipate the range of rupture scenarios possible in faults where the seismogenic width changes substantially along strike, that is, the San Andreas Fault (Smith‐Konter et al, ) and the Alpine Fault in New Zealand (Michailos et al, ).…”

Seismicity in a model governed by competing frictional weakening and healing mechanisms

“…Carlson 1991;Rice 1993;Ben-Zion 1996;Langer et al 1996;Fisher et al 1997;Dahmen et al 1998;Lapusta et al 2000;Shaw & Rice 2000;Weatherley et al 2002;Heimpel 2003;Zöller et al 2005;Mehta et al 2006;Hillers et al 2007). Properties such as the energy dissipation and stress interaction distance, dimension, the degree of heterogeneity, and the range of size scales have been used as control parameters.…”

Self-Organized Criticality in Asymmetric Warfare

“…This is consistent with other areas, such as seismology, when the weak events are typically behaving more like a random system obeying Poisson distribution rather than the power law. 15 This observation is supported by the fact that omitting the incidence of a single casualty from the distribution improves the goodness of fit to R 2 ∼ 0.98 . Since the breakdown of fatalities based on the cause of death was not available, it is impossible to say if there is a difference in the scaling of fatalities based on the cause of death (like in Afghanistan).…”