Research Projects

I have worked on both earth structure and earthquake process related projects.

Earth Structure

The following projects are related to earth structure.

Western United States Structure

The EarthScope USArray Transportable Array (TA) significantly expanded seismic data in the United States. The massive high quality seismic records provide us a great opportunity to improve the seismic imaging. As the western U.S. data were fully collected, our focus was technique development. We developed a receiver function smoothing/interpolation technique to reduce scattering effects on receiver function waveforms. The receiver function smoothing/interpolation also equalizes the lateral sampling distances between receiver functions and surface wave dispersion. The interpolated receiver functions were simultaneously inverted with Rayleigh wave group velocities and gravity observations. We also used clustering analysis to regionalize the 1D shear velocity profiles. The results show the crust and upper mantle structures correlate with geological provinces.

Eastern United States Structure

We are working on the seismic imaging of the eastern part of the country, as the TA deployment was completed in the eastern U.S. in the winter of 2015. Over half a million of receiver functions were computed and interpolated. The interpolated receiver functions are simpler than single station averaged receiver functions as the scattering effects are reduced. Developed from the study of the eastern US, we added Rayleigh wave phase velocities in the simultaneous inversion.

Structure Beneath Antarctica Stations

Slow speed shallow structures (e.g. ice, sediments) cause large amplitude reverberations in P-wave receiver function waveforms and mask useful information from deeper speed variations. We are applying wavefield continuation and decomposition technique to derive subsurface signals by removing the shallow reverberations. The process utilizes the better constrained shallow structure to back-propagate the teleseismic wavefield. Features in the subsurface waveforms provide additional information of the velocity just beneath the shallow structure. The resulting subsurface signals can be used to tightly constrain seismic speed changes at depth. We are using Monte Carlo Markov Chain simulations to estimate the speed variations beneath several Antarctica stations. The process can also be applied to sediment-covered regions.

Earthquake Process

Detect & Locate Remote Earthquakes

Remote earthquakes are not easy to detect and locate due to poor station coverage. Surface waves are the dominating signal recorded for remote earthquakes. We use surface wave cross-correlations to detect and locate remote earthquakes. Using several known earthquakes as templates, the remote events were detected by comparing the similarity of the template waveforms with the candidate signals. The earthquake locations were obtained by searching a grid around template events. The candidate location was determined by maximizing the cross-correlation values with corresponding time shifts. The magnitude of the detected events was also computed from the cross-correlation values. We found the oceanic ridge has a low seismicity.