What Do Foreshock Swarms Tell Us About Subsurface Conditions and an Impending Large Earthquake?

Abstract:

I describe the mechanics of foreshocks and the information they might provide about an impending larger earthquake. The work is informed both by 3-meter laboratory rock experiments and natural observations of foreshocks. A diversity of earthquake nucleation conditions is expected in the Earth's crust. I present two plausible scenarios which lead to vastly different interpretations for foreshocks.

In the first scenario—likely representative of highly coupled crustal faults—the critical nucleation length scale h* is small (<1 m) so dynamic rupture initiates rapidly in space and time, and M -2 events are earthquakes that met unfavorable conditions and stopped soon after initiating, while larger earthquakes continued to rupture and grow. In this scenario, if the regional stress approaches a critical state, earthquakes will begin to initiate at higher rates, but during this increase in seismicity rate, most of the fault remains essentially locked. Smaller events (foreshocks) may redistribute stress, and make the ratio of stress/strength more uniform in space which will allow a large event to eventually propagate without being stopped. In this case, foreshocks herald a larger earthquake.

In a second scenario—likely representative of many subduction zones, oceanic transform faults, and some mature faults such as sections of the San Andreas Fault—slow slip can occur without immediately initiating an earthquake. When considering the large-scale spatial average of fault properties, the effective nucleation length scale h* is large (~1-10 km), but heterogeneity of fault properties causes higher coupling on some fault patches which can rupture to produce “sub-h*” earthquakes or repeating earthquake sequences. These “sub-h*” foreshocks are driven by aseismic slip and can occur in swarms that may or may not eventually ignite a large earthquake in a “cascade up” process.