Extension of Aseismic Slip Propagation Theory to Slow Earthquake Migration

Natural faults host various types of migrating slow earthquake phenomena, with migration speeds much lower than seismic wave speeds and different moment‐duration scaling from regular earthquakes. To advance the obtained quantitative understanding of the migration process and long duration of slow earthquakes, I study a chain reaction model in a population of brittle asperities based on a rate‐ and state‐dependent friction on a 3‐D subduction plate boundary. Simulation results show that the migration speed is quantitatively related to frictional properties by an analytical relation derived here. By assuming that local pore water in front of the migration drives rapid tremor reversal and is so local as to hold a constant stress drop, the application of the analytical solution to observational results suggests that (a) the temporal changes of observed migration speeds for the rapid tremor reversal could be explained by about 70% reduction of the effective normal stress; (b) effective normal stress for the deeper extension of seismogenic segment in the western part of Shikoku is about 1.5 times greater than that in the central part. Applying rupture time delay between slow earthquake asperities for the duration longer than regular earthquake, I also conclude that (c) the characteristic slip distance of rate‐and‐state friction for low‐frequency earthquakes is roughly between 30 µm and 30 mm; (d) the stress and strength drops of very low‐frequency earthquakes is much smaller than 1 MPa. Plain Language Summary Previous computer simulations and a few observational studies suggest that large subduction earthquakes may be preceded by slow earthquakes (not felt by humans) whose migration speed increases as the occurrence of the large earthquake approaches. So far, this precursory process has only been discussed qualitatively. In this study, I consider a chain reaction model on a heterogeneous fault made of small brittle asperities embedded in a viscous matrix: when a small asperity breaks rapidly, it generates a wave of slow slip around it, which in turn triggers the rupture of neighboring asperities, and so on. I develop a theoretical relation between slow earthquake migration speed and frictional properties. The model helps explain why slow earthquakes are slow, and provides a basis for precursory slow earthquake migration phenomena. The theoretical relation also provides an estimate of the characteristic slip distance of slow earthquakes, a rock property that is d