An investigation into the role of time-dependent cohesion in interseismic fault restrengthening

M P A Van Den Ende,A R Niemeijer

doi:10.1038/s41598-019-46241-5

Abstract

Earthquakes typically exhibit recurrence times that far exceed time-scales attainable in a laboratory setting. To traverse the temporal gap between the laboratory and nature, the slide-hold-slide test is commonly employed as a laboratory analogue for the seismic cycle, from which the time-dependence of fault strength may be assessed. In many studies it is implicitly assumed that all fault restrengthening emanates from an increase in the internal friction coefficient, neglecting contributions from cohesion. By doing so, important information is lost that is relevant for numerical simulations of seismicity on natural faults, as well as for induced seismicity. We conduct slide-hold-slide experiments on granular halite gouge at various normal stresses to assess the time-dependence of the internal coefficient of friction, and of the cohesion, independently of one another. These experiments reveal that both the internal friction coefficient and cohesion increase over time, but that these quantities do not share a common evolution, suggesting different underlying mechanisms.

Highlights

Earthquakes are among the most disruptive of natural hazards known
After the hold phase, when sliding is re-initiated, a clear peak stress is observed even for the shortest hold duration of 1 s, after which the stress evolves towards a new steady-state
We report a suite of slide-hold-slide laboratory experiments conducted on granular halite, from which we inferred the time-dependence of the internal coefficient of friction and of the sample cohesion independently

Summary

Introduction

Earthquakes are among the most disruptive of natural hazards known. Owing to their destructive potential and poor predictability, earthquakes and unstable frictional sliding in general receive considerable attention in laboratory, field, and modelling studies. To bridge the temporal gap between natural and laboratory interseismic periods, the time-dependent strength recovery of faults is commonly studied by conducting slide-hold-slide (SHS) experiments[3], in which the transient shear strength of the sample is monitored following a predetermined hold period. We conducted a suite of experiments in which fluid-rock interactions are significant at the time-scale of the experiment, and interseismic restrengthening is simulated. From these experiments, the time-dependence of the sample cohesion is estimated independently of the coefficient of friction.

Methods

Results

Discussion

Conclusion