Title: The strength of subduction megathrusts and implications to earthquake processes

Abstract:
The strength of subduction megathrusts and how it varies during slip are important to earthquake physics. Actual megathrusts feature multi-scale heterogeneity, but fundamental understanding can be gained by focusing on the big picture and representative end members. This seminar provides a big-picture sketch of what has been learned over the past three decades. An ideal creeping fault has only one strength, but an ideal stick-slip fault has three strengths: the true strength, apparent strength, and dynamic strength. At a regional scale, there are only two useful methods to determine megathrust strength: the study of force balance in which gravity provides a reference and the study of thermal fields in which frictional heat dissipation reflects fault strength. Global applications of these methods yield effective friction coefficients of megathrusts in the range 0.03 – 0.13, much lower than the value of about 0.4 predicted by Byerlee’s law with hydrostatic fluid pressure. Despite this remarkable weakness, stress drop in megathrust earthquakes generally is only 10 – 30% of the absolute fault strength. Seismogenesis depends not on the absolute strength but how the fault weakens or strengthens during slip. Thirty years ago, updip and downdip limits of the megathrust seismogenic zone were proposed on the basis of thermal controls on weakening or strengthening behaviour. Today, new understanding of fault mechanics and fault gouge petrology and rheology requires the development of a new conceptual framework. I will explain why and how the previously proposed limits should be changed to “soft barriers” to seismic slip.