Recent Global Positioning System observations of major earthquakes such as the 2014 Chile megathrust show a slow preslip phase releasing a significant portion of the total moment (Ruiz et al., 2014, https://doi.org/10.1126/science.1256074). Despite advances from theoretical stability analysis (Rubin & Ampuero, 2005, https://doi.org/10.1029/2005JB003686; Ruina, 1983, https://doi.org/10.1029/jb088ib12p10359) and modeling (Kaneko et al., 2017, https://doi.org/10.1002/2016GL071569), it is not fully understood what controls the prevalence and the amount of slip in the nucleation process. Here we present laboratory observations of slow slip preceding dynamic rupture, where we observe a dependence of nucleation size and position on the loading rate (laboratory equivalent of tectonic loading rate). The setup is composed of two polycarbonate plates under direct shear with a 30‐cm long slip interface. The results of our laboratory experiments are in agreement with the preslip model outlined by Ellsworth and Beroza (1995, https://doi.org/10.1126/science.268.5212.851) and observed in laboratory experiments (Latour et al., 2013, https://doi.org/10.1002/grl.50974; Nielsen et al., 2010, https://doi.org/10.1111/j.1365-246x.2009.04444.x; Ohnaka & Kuwahara, 1990, https://doi.org/10.1016/0040-1951(90)90138-X), which show a slow slip followed by an acceleration up to dynamic rupture velocity. However, further complexity arises from the effect of (1) rate of shear loading and (2) inhomogeneities on the fault surface. In particular, we show that when the loading rate is increased from 10−2 to 6 MPa/s, the nucleation length can shrink by a factor of 3, and the rupture nucleates consistently on higher shear stress areas. The nucleation lengths measured fall within the range of the theoretical limits Lb and L∞ derived by Rubin and Ampuero (2005, https://doi.org/10.1029/2005JB003686) for rate‐and‐state friction laws.

The transpressional Alpine Fault in New Zealand has created a thick shear zone with associated highly anisotropic rocks. Low seismic velocity zones and high seismic reflectivity are recorded in the Alpine Fault Zone, but no study has explored the underlying physical rock parameters of the shallow crust that control these observations. Protomylonites are the volumetrically dominant lithology of the fault zone. Here we combine experimental measurements of P‐wave speeds with numerical models of elastic wave anisotropy of protomylonite samples to explore how the fault zone can be seismically imaged. Numerical models that account for the porosity‐free real samples' fabric elastic tensors from electron backscatter diffraction (EBSD) are calculated by MTEX and a finite element model (FEM), while microfractures are modeled with differential effective medium (DEM) theory. At effective pressures representative of the Alpine Fault brittle zone, experimental wave speeds are lower than those predicted by MTEX/FEM. A possible DEM model suggests that a combination of random and aligned microfractures with aspect ratios increasing with pressure can explain the experimental wave speeds for pressures <70 MPa. Such microporosity in the form of foliation‐ and mica basal plane‐parallel microfractures and grain boundaries is validated with synchrotron X‐ray microtomography and transmission electron microscopy (TEM) images. Finally, by modeling anisotropy of seismic reflection coefficients with angle of incidence, we demonstrate that the high reflectivity and low‐velocity zone (LVZ) observed at the Alpine Fault can only be explained if this microporosity is accounted for throughout the brittle fault zone, even at depths of 7–10 km.

No abstract

&lt;p&gt;Natural faults are heterogeneous features, with complex geometries and material properties. Understanding how the geometrical complexities of a fault affects the dynamics and preparatory phase of earthquakes is of crucial importance for seismic hazard assessment. In laboratory samples, frictional sliding along prefabricated faults may produce so called stick-slips comparable to dynamic ruptures observed during earthquakes. While the effect of roughness has been shown to influence significantly the frictional behavior of laboratory faults, there are only a few studies investigating more complex types of fault heterogeneities. In this study, we conduct friction experiments on granite with inclined sawcut faults, under a constant confining pressure of 35MPa. Samples are loaded using an axial displacement rate of 0.5 &amp;#181;m/s. &amp;#160;At &amp;#160;similar boundary conditions we compare the slip behavior of (1) a smooth fault, (2) a smooth fault with a single asperity, a 7 mm diameter vertical pin traversing the contact interface, and (3) a rough fault prepared by sandblasting the surface with silicon carbide. A key result of this study is that slip behavior depends on fault roughness and is influenced in a non-trivial way by asperities. The smooth fault displays unstable stick-slip as opposed to the rough fault showing predominantly creep. The smooth fault with the pin exhibits a slip behavior in-between, with very regular stress oscillations that seem to be attenuated by the presence of the pin (asperity). Only after failure of the pin, we observe the stress drop during instabilities to increase regularly with cumulative slip. We also show that in the case of a fault with a single asperity, the slip velocity is less than an order of magnitude lower compared to a similar smooth fault without this asperity.&lt;/p&gt;

No abstract

Aseismic slip may occur during a long preparatory phase preceding earthquakes, and what controls it remains poorly understood. In this study, we explored the role of load point velocity and surface roughness on slow slip during the preparatory stage prior to stick‐slip events. To that end, we conducted displacement‐rate controlled friction experiments by imposing varying load point velocities on sawcut granite samples with different surface roughness at a confining pressure of 35 MPa. We measured the average slip along the fault with the recorded far‐field displacements and strain changes, while acoustic emission sensors and local strain gages were used to capture local slip variations. We found that the average amount of aseismic slip during the preparatory stage increases with roughness, whereas precursory slip duration decreases with increased load point velocity. These results reveal a complex slip pattern on rough faults which leads to dynamic ruptures at high load point velocities.

No abstract

No abstract