Abstract
AbstractDamage zones are ubiquitous components of faults that may affect earthquake rupture. Simulations show that pulse‐like rupture can be induced by the dynamic effect of waves reflected by sharp fault zone boundaries. Here we show that pulses can appear in a highly damaged fault zone even in the absence of reflected waves. We use quasi‐static scaling arguments and quasi‐dynamic earthquake cycle simulations to show that a crack turns into a pulse after the rupture has grown larger than the fault zone thickness. Accompanying the pulses, we find complex rupture patterns involving back‐propagating fronts that emerge from the primary rupture front. Our model provides a mechanism for back‐propagating fronts recently observed during large earthquakes. Moreover, we find that slow‐slip simulations in a highly compliant fault zone also produce back‐propagating fronts, suggesting a new mechanism for the rapid tremor reversals observed in Cascadia and Japan.
Highlights
Pulse-like rupture is a common mode of earthquake propagation in which the duration of slip at each point of the fault, known as the rise time, is short compared to the total rupture duration (Heaton, 1990)
Further research is warranted to investigate whether the effects observed in our idealized fault zone model remain after releasing some of the simplifying assumptions, in particular the quasi-dynamic approximation and the 2-D tabular low-velocity fault zones (LVFZs) geometry
We develop a formal analogy between a fault zone model and a nearest-neighbor (BK) model that explains the emergence of pulses
Summary
Pulse-like rupture (hereafter referred to as pulses) is a common mode of earthquake propagation in which the duration of slip at each point of the fault, known as the rise time, is short compared to the total rupture duration (Heaton, 1990). Pulses play a prominent role in the theory of earthquake mechanics: They can radically affect the earthquake energy balance (Nielsen & Madariaga, 2003), reduce the apparent strength of faults (Noda et al, 2009), enhance the spatial heterogeneity of earthquake slip and stress (Aagaard & Heaton, 2008), and promote complexity of seismicity manifested by a broad range of event magnitudes (Cochard & Madariaga, 1996). Our simulations reveal that the quasi-static effects of a highly damaged LVFZ are sufficient to generate back-propagating fronts
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
