2018 Volume 124 Issue 9 Pages 725-739
Earthquakes occur when shear stress exceeds the static strength of a fault; thus, knowledge of fault strength is important. Various estimates of fault strength have been made over the last half century; however, there is still disagreement on whether faults are weak (~10 MPa), which would account for observed low stress drops and low heat flow anomalies, or strong (~100 MPa), as estimated from laboratory experiments using Byerlee’s law. We review methods to estimate fault strength and propose that a cause of the ongoing debate is the definition of strength used in different methods. High-speed friction experiments have played a significant role in resolving the debate. The experiments have shown that faults are only weak during an earthquake. Detailed seismic analysis, additionally, has revealed a heterogeneous distribution of fault strength. These findings require that fault models combine the deformation mechanisms of both strong and weak fault patches, which result in non-uniform strength and stress distributions. We propose that a multidisciplinary approach is required. Geological fieldwork should be used to test fault models that are developed based on seismology and rock mechanics.
