Magnitude dependent seismic quiescence has been observed for a number of major events including the 1982 Urakawa-Oki earthquake, 1994 Hokkaido-Toho-Oki earthquake, 1994 Northridge earthquake, 1995 Kobe earthquake, and 2011 Tohoku earthquake. For example, in Urakawa-Oki, the region surrounding the future rupture became significantly depleted in Mw 3.0 earthquakes a few years prior to the event. Magnitude dependent quiescence tends to occur for high stress drop events.  Two recent California earthquakes, one high stress drop (1995 Northridge) and one low stress drop (2010 Laguna Salada), are compared to the Urakawa-Oki earthquake.  Magnitude dependent quiescence is found for Northridge but not Laguna Salada.

Dilatant and hydrological effects are added to the fault simulator of Sacks and Rydelek (1995) to include physics consistent with observations of seismic quiescence.  These modifications generate precursory seismic quiescence and associated changes in b value.

In dilatancy theory, as the fault is loaded toward failure and stress increases, the rock can begin to dilate.  As dilation occurs, the pore pressure decreases, the effective normal stress increases, and the material stiffens.  The fault is now both stronger and stiffer, so it can support more of the stress, causing the surrounding region to deform less. This decreased loading and deformation of the surrounding region could explain the observed quiescence.  Over time (~2-20 years) the water will percolate back into the fault core.  Eventually, the pore pressure increases, dilatancy collapses, and failure is encouraged.  The excess water expected from dilatancy collapse could explain the anomalous water measurements found for the Kobe earthquake, the 1999 Chi-Chi earthquake, and the 1975 Haicheng earthquake (which was successfully predicted).

Understanding precursory phenomena via dilatancy theory may help progress the science of earthquake prediction, which has been largely unsuccessful to date.