Insights into Fault Roughness Throughout the Seismic Cycle of Laboratory Earthquakes

Earthquakes are produced by slip events along faults driven by the accumulation and release of elastic energy in the Earth’s crust. They are cyclic, and a broad spectrum of slip behaviors is observed along seismogenic faults. The irregular and chaotic nature of earthquakes makes it difficult to establish a predictive law. Furthermore, the constitutive behavior of active fault zones remains a subject of debate, particularly regarding the relative roles of static and dynamic energy controls during seismic events. Elastic energy accumulation in the crust leads to rupture nucleation, while rupture propagation and arrest are likely governed by the physical properties of the fault zone, which vary between events and evolve during slip. While numerous parameters have been proposed to influence these processes, we emphasize that fault geometry and fault–fault interactions represent fundamental controls on rupture behavior and the evolution of the seismic cycle. To better understand how the size and distribution of asperities along faults control the earthquake cycle, we conducted laboratory experiments on analog material samples with root-mean-square (RMS) roughness values ranging from 0.5 to 30 micrometers. We used the Energy-Controlled Rotary Shear (ECoR) apparatus to replicate the earthquake cycle in the laboratory. The ECOR allows for spontaneous nucleation of laboratory earthquakes at velocities, accelerations, displacements, and magnitudes comparable to those observed in natural earthquakes. In these experiments, we used a loading spring with an effective elastic constant and varied the sample-averaged normal stress. Across experiments, we observe a range of slip behaviors, from stick–slip to steady creep, over the lifetime of the laboratory fault. We hypothesize that the size and distribution of asperities along the fault control the style of fault slip. Furthermore, over the course of the seismic cycle and in the presence of frictional weakening, we propose that the power density, another aspect that we will explore in the future, and the critical nucleation size control the magnitude of earthquakes.