Supershear earthquakes, whose rupture speeds are faster than that of the seismic shear waves, can be highly destructive owing to the associated strong ground shaking. These events are rare, and understanding why they occur may provide insights into fault mechanics. However, identified samples of supershear earthquakes are rare, which hinders systematic analysis of their mechanism. In this work, I optimized observational methods and applied them to find more supershear earthquakes. Eventually, a global survey of earthquake rupture speeds reveals that supershear earthquakes occur much more frequently than previously thought. Our findings provide a much greater basis for evaluating the conditions favourable for driving supershear rupture. We observe that supershear rupture is more common along faults that separate oceanic and continental plates. These oceanic–continental supershear events also generate rupture in the same relative direction as the stiffer fault block moves. We find that fault zone width is thicker in regions where supershear earthquakes occur, and that supershear ruptures occur relatively frequently in mature oceanic transform faults. Our works confirm observations made from laboratory and numerical experiments and will be able to guide future research into these destructive earthquakes.

Earthquake source imaging is an efficient tool to reveal the kinematic rupture process of earthquakes. The accuracy and robustness of source imaging is of vital importance to con- strain source properties and help discriminate among potential physical mechanisms. In this work, I develop array-based approach: enhanced back-projection and coherence method to provide better constraints on kinematic source imaging.

Slowness correction is proposed to improve the source imaging of complicated rupture in conjugate fault systems. The correction is capable of mitigating 3D structural effects that commonly exist for M8 earthquakes.

For deep earthquakes, I introduce 3D back-projection which provides depth resolution of the rupture front. Resolving ruptures in 3D space helps us to understand the rupture process and physical mechanisms of deep earthquakes. Furthermore, coherence model is established to relate earthquake source size with inter-station cross-correlation. This method creates independent measures to validate and constrain the source sizes of deep earthquakes.

Aforementioned array-based approaches have been successfully applied to recently oc- curred earthquakes, and I will take this chance to revisit the rupture mechanisms of the events revealed by enhanced source imaging

Benefiting from the recent deployments of dense seismic arrays, seismologists have the opportunity to detect earthquakes of small magnitudes (M < 2), referring to the occurrence of these events as microseismicity. The characterization of microseismicity is essential in solving many geophysical problems, such as probing fault structures, investigating the process of hydraulic fracturing, and imaging the preseismic, coseismic, and postseismic slip of megathrust earthquakes. Traditionally, earthquakes are identified by picking seismic phases on continuous waveforms using the short-time average to long-time average ratio, kurtosis and skewness, waveform polarization, Akaike information criterion, and discrete wavelet transforms. These single-station methods, however, may fail to detect microseismicity with weak phase arrivals hidden in the noise.

This thesis introduces three new approaches to detecting microseismicity with multistation data: backprojection, matched-filter, and deep learning. The backprojection approach tracks and back-projects the coherent seismic pulses to the target region to determine the timing and location of any microseismic earthquakes. The matched-filter approach searches for similar patterns of existing template earthquakes in the continuous recordings as suggestive of a new event. We characterize the aftershock sequence of the 2011 M 9 Tohoku earthquake with the backprojection and matched-filter methods. Based on the spatial consistency between aftershock-depleted zones and large coseismic slip, we identify a possible large coseismic slip zone in the near-trench region offshore Fukushima. The deep learning approach is data-driven and it learns to characterize microseismicity with a neural network based on a large number of microseismicity recordings. We trained a Deep Learning Phase-picking model named EdgePhase with a Southern California dataset and applied it to detect the early aftershocks following the 2020 M 7 Samos, Greece earthquake. Compared to a local earthquake catalog, EdgePhase showed 190% more detections with an event distribution that is more conformative to a planar fault interface, suggesting higher fidelity in event locations. In addition to characterizing microseismicity, backprojection can also improve the prediction of earthquake ground motions. We propose a high-frequency distance metric based on backprojection, which outperforms traditional distance metrics in predicting the ground shaking intensity of megathrust earthquakes in Japan and Chile.

Investigating the characteristic of earthquake source more accurately and faster is required to mitigate earthquake hazard and improve the understanding of the mechanism of earthquake source. In this work, I develop a new method: multi-array local back-projection (MLBP). Local back-projection potentially resolve the rupture process of large earthquakes faster and more accurately. Multiple arrays are combined to increase the stability and accuracy of back-projection results. The method are successfully applied to large earthquakes from M 6 to M 9 using the existing seismic stations in California and Japan. We also propose a tsunami early warning approach based on the muti-array local back-projection method.