Can Precursory Moment Release Scale With Earthquake Magnitude? A View From the Laboratory

Today, earthquake precursors remain debated. While precursory slow slip is an important feature of earthquake nucleation, foreshock sequences are not always observed, and their temporal evolution remains poorly constrained. We report on laboratory earthquakes conducted under upper-crustal stress and fluid pressure conditions. The dynamics of precursors (slip, seismicity, and fault coupling) prior to the mainshock are dramatically affected by slight changes in fault conditions (fluid pressure and slip history). A relationship between precursory moment release and mainshock magnitude is systematically observed, independent of fault conditions. Based on nucleation theory, we derive a semiempirical scaling relationship which explains this trend for laboratory earthquakes. Several natural observations of earthquakes ranging from similar to M-w 6.0-9.0, where precursory moment release could be estimated, seem to follow the extrapolation of the laboratory-derived scaling law. Notwithstanding spatiotemporal complexity in natural seismicity, some moderate to large earthquake magnitudes might be estimated through integrated seismological and geodetic measurements of both seismic and aseismic slips during nucleation. Plain Language Summary Understanding the preparation phase that precedes earthquake ruptures (nucleation) is crucial for seismic hazard assessment because it might provide information on the impending earthquake. Here, we show that the temporal evolution of laboratory earthquake precursors (precursory slow slip, precursory seismicity, and fault coupling) is of little use in assessing an impending earthquake's size. Nevertheless, independent of fault slip behavior (seismic or aseismic) and environmental conditions (stress state and fluid pressure and slip history), the amount of moment released during the preparation phase scales with the earthquake's magnitude. This property is demonstrated by laboratory observations and earthquake nucleation theory and seems compatible with several natural observations of earthquakes ranging from M-w 6.0 to M-w 9.0. As a consequence, if earthquakes exhibit a preparation phase, it could be possible that this phase is larger or longer for higher magnitude earthquakes and consequently, more likely to be detectable.