Inferring earthquake physics and chemistry using an integrated field and laboratory approach

Earthquakes are the result of a combination of (1) physico-chemical processes operating in fault zones, which allow ruptures to nucleate and rock friction to decrease with increasing slip or slip rate, and (2) of the geometrical complexity of fault zones. In this review paper, we summarize recent experimental findings from high velocity (conducted at about 1 m/s slip rate, or typical seismic slip rates) rock friction experiments with an emphasis on potential dynamic weakening mechanisms (melt lubrication, nanopowder lubrication, etc.) and how these mechanisms might be recognized by means of microstructural and mineralogical studies in exhumed fault zones. We discuss how earthquake source parameters (coseismic fault strength, weakening distances, energy budgets, etc.) might be derived from the field and laboratory experiments. Additionally, we discuss what needs to be considered in terms of fault zone geometry and morphology (focusing on fault surface roughness) in order to develop models of realistic fault surfaces and present theoretical considerations for microphysical modeling of laboratory data at seismic slip rates, with an emphasis on the case of melt lubrication. All experimental data and, in the case of melt lubrication, microphysical models indicate that faults must be very weak (μ < 0.1) during coseismic slip. Moreover, experiments have shown that the slip weakening distance during coseismic slip is on the order of a few tens of centimeters at most under natural conditions, consistent with inferences from field observations. Finally, we discuss open questions, future challenges and opportunities in the field of earthquake mechanics.