Tectonic faults fail in a continuum of modes from slow earthquakes to elastodynamic rupture. Precursory variations in elastic wavespeed and amplitude, interpreted as indicators of imminent failure, have been observed in limited experimental and natural settings for this spectrum of slip modes. Such variations are thought to arise from microcracking within and around the fault zone. However, the physical mechanisms and connections to fault creep are not well understood. Here, we vary loading stiffness to generate a range of slip modes and measure fault zone properties using elastic waves transmitted through the fault. We find that elastic wave amplitudes show clear changes before failure. The temporal onset of amplitude reduction scales with lab earthquake magnitude and the magnitude of this reduction varies with fault slip. Our data suggest that continuous seismic monitoring in proximity to natural faults could be useful for assessing fault state and seismic hazard potential.

Machine learning (ML) techniques have become increasingly important in seismology and earthquake science. Lab-based studies have used acoustic emission data to predict time-to-failure and stress state, and in a few cases the same approach has been used for field data. However, the underlying physical mechanisms that allow lab earthquake prediction and seismic forecasting remain poorly resolved. Here, we address this knowledge gap by coupling active-source seismic data, which probe asperity-scale processes, with ML methods. We show that elastic waves passing through the lab fault zone contain information that can predict the full spectrum of labquakes from slow slip instabilities to highly aperiodic events. The ML methods utilize systematic changes in p-wave amplitude and velocity to accurately predict the timing and shear stress during labquakes. The ML predictions improve in accuracy closer to fault failure, demonstrating that the predictive power of the ultrasonic signals improves as the fault approaches failure. Our results demonstrate that the relationship between the ultrasonic parameters and fault slip rate, and in turn, the systematically evolving real area of contact and asperity stiffness allow the gradient boosting algorithm to ‘learn’ about the state of the fault and its proximity to failure. Broadly, our results demonstrate the utility of physics-informed machine learning in forecasting the imminence of fault slip at the laboratory scale, which may have important implications for earthquake mechanics in nature.