Grand Séminaire d'Institut

Vendredi 28 Septembre 2018 à 11h00.

Nucleation process of earthquakes: friction theory and sensitivity to perturbations

''Earthquake is a serious natural disaster in some certain areas in the world such as the Pacific-rim, Italy, Greece, and Turkey. One of the remarkable properties of earthquakes is their abruptness: Almost all the earthquakes appear to occur without any precursors. Partly due to this abruptness, we cannot predict earthquakes based on a firm scientific basis.
On the other hand, earthquake should be described by classical mechanics, which is deterministic. Thus, if we understand how earthquakes start, we can say something about the possibility (or impossibility) of earthquake prediction in a scientific manner. As an earthquake is the unstable slip of a fault in the Earth's crust, the most important physics regarding earthquakes is the frictional properties of rocks, which we do not fully understand to this date.
In this talk, I will briefly review what has been known about the rock friction in geophysics. This is formulated as an empirical law referred to as the rate and state dependent friction law. Based on this friction law, we can show how the creeplike (quasistatic) slip is accelerated to the fast slip, which is eventually recognized as an earthquake. I explain that this acceleration process should occur in a small area on an earthquake fault, and therefore the area of accelerating slip is fundamental to the predictability of earthquake. I will also explain how small/large this "nuclei" can be, and discuss their response to some small stress perturbations such as tides or snowfalls.''