Research
Our research interest lies in seismic source processes and fault rupture dynamics of natural and induced earthquakes. We explore a range of seismological problems such as earthquake triggering and interactions, seismic source and rupture properties, effects of fluid on fault systems, hazard mitigation, etc.
Both data- and model-driven approaches are applied in our work:
Fully dynamic earthquake-cycle simulations
Seismic waveform modeling
Statistical analysis of seismicity occurrence
Research highlights
Rupture processes of induced seismicity
A clear understanding of the source physics of induced seismicity is the key to effective seismic hazard mitigation. In particular, resolving their rupture processes can shed lights on the stress state prior to the main shock, as well as ground motion response. We utilize the rich pool of seismic data in the central US to constrain the rupture processes of major induced earthquakes and characterize the causative relationships between the injection operations and the induced rupture process.
Figure: Rupture directivity of each analyzed earthquakes relative to their respective fluid injection site locations.
The effect of induced stress perturbation on fault motion
The key to effect induced seismic hazard mitigation is to obtain a comprehensive understanding of the source physics and the triggering mechanisms of induced seismicity. Through numerical simulation, we model the behavior of faults experiencing fluid-induced shear stress perturbations. In particular, we study two aspects of the perturbation: timing and magnitude. We compare the difference in both seismic and aseismic slip on the fault with the case where the stress perturbation is absent. Results show that induced stress perturbations can cause a wide range of responses, including the advancement of the next earthquake, instantaneous triggering of seismic events and aseismic transients, as well as advancing and delaying subsequent earthquakes in the long term.
Interaction of repeating micro-earthquakes
Studying small repeating earthquakes enables better understanding of fault physics and characterization of fault friction properties. We find that postseismic creep dominates the interaction, with earthquake triggering occurring at distances much larger than typically assumed. Our results open a possibility of using interaction of repeating sequences to constrain friction properties of creeping segments.
Movie: The time-dependent evolution of the slip rate of two interacting repeating earthquake sequences in a rate-and-state fault model.
MovieS1_final.movRupture dynamics with solid-fluid interaction
We find that additional mechanisms for the RS framework are necessary to model the source properties and interaction of the repeating earthquake sequences near Parkfield, CA. Enhanced coseismic weakening (e.g. thermal pressurization) and elevated normal stress on the seismogenic patches are possible causes for the high stress drops. Also, the occurrence of aseismic slip is the key to explaining the variability in the repeating source properties.
Rapid seismic source assessment / Earthquake early warning
We develop a systematic method to accurately estimate moment magnitude and focal mechanism within 3–6 s after the first P arrival. Focal depth can also be constrained within ∼10s upon the arrival of S waves. To determine the direction of fault rupture, we employ the empirical Green’s function (eGf) approach to simulate the rupture direction of the beginning motion generated by larger events.
Figures: (top) Multiple source properties can be constrained with less than 10s of incoming waveform (bottom) Modeling ground motion using the eGf approach.
