Aftershocks

Aftershocks region have been extensively reported to expand logarithmically with time. The associated migration velocity is typically of the order of several km/decade but can be much larger, especially when observing early aftershock sequences, seismic swarms, or tremors. We present here a model for the expansion of aftershock zones based on the idea that aftershocks are triggered as afterslip grows with time along the fault. One of the model assumptions is that aftershocks are triggered when a critical level of afterslip is reached. We predict that the migration velocity V p at time t following the mainshock is given by V p = í µí¼ A ′ Δí µí¼ × c t , where A ′ is a frictional parameter, Δí µí¼ a characteristic value for the stress perturbation, c the radius of the stress perturbation, and í µí¼ a constant of order unity. The scaling V p ∝1∕t implies a logarithmic expansion of the aftershock zone with time. The migration velocities predicted by our model are in quantitative agreement with the observations reported following aftershock sequence of small and large earthquakes in various tectonic settings, seismic swarms, and tremor sequences. Plain Language Summary Aftershocks are shown to migrate with time away from the rupture area of the mainshock. This migration typically occurs as the logarithm of time. We present here a model based on the idea that afterslip drives aftershocks. The model is able to predict the migration as the logarithm of time, predicting apparent propagation velocities consistent with the observations. The propagation velocity is simply related to the physical parameters of the model.

The postseismic deformation following a large (M w > 7) earthquake is expressed both seismically and aseismically. Recent studies have appealed to a model that suggests that the aseismic slip on the plate interface following the mainshock can be the driving factor in aftershock sequences, reproducing both the geodetic (afterslip) and seismic (aftershocks) observables of postseismic deformation. Exploiting this model, we demonstrate how a dense catalog of aftershocks following the 2015 M w 8.3 Illapel earthquake in Central Chile can constrain the frictional and rheological properties of the creeping regions of the subduction interface. We first expand the aftershock catalog via a 19 month continuous matched-filter search and highlight the log-time expansion of seismicity following the mainshock, suggestive of afterslip as the main driver of aftershock activity. We then show how the time history of aftershocks can constrain the temporal evolution of afterslip. Finally, we use our dense aftershock catalog to estimate the rate and state rheological parameter (a − b) as a function of depth and demonstrate that this low value is compatible either with a nearly velocity-neutral friction (a ≈ b) in the regions of the megathrust that host afterslip, or an elevated pore fluid pressure (low effective normal stress) along the plate interface. Our results present the first snapshot of rheology in depth together with the evolution of the tectonic stressing rate along a plate boundary. The framework described here can be generalized to any tectonic context and provides a novel way to constrain the frictional properties and loading conditions of active faults. Plain Language Summary The slow postseismic deformation, or afterslip, that lasts several years following a major earthquake can be as strong as the earthquake itself and is therefore a key component in understanding the seismic hazard along tectonic plate boundaries. Afterslip is typically studied with GPS that measures the deformation at the surface, but with a low spatial precision. It is therefore difficult, if not impossible, to tease out the finer details of how the plate interface responds to a large earthquake and to identify which parts of the plate interface could rupture next. Here we develop a new framework to study what happens after a major earthquake using the precise evolution in time and space of aftershocks. We leverage the high-resolution aftershock distribution to determine how the friction of the plate boundary varies in depth, which allows us to better understand which parts of the plate interface are susceptible to afterslip. Our methods described here can be generalized to any tectonic plate boundary and provide a novel way to constrain how active faults are influenced by major earthquakes.