Research

  • What happens when background noise is louder than an earthquake? 

In addition, I’m interested in utilizing signal processing techniques to work on seismic data. We presented our work on developing a denoising algorithm to remove persistent culture noise at the recent AGU conference (2021). Also, check out the news story, “What happens when background noise is louder than an earthquake?“, which summarizes our work. Similarly, we developed a routine seismic array processing tool to detect and analyze seismic events in and around Alaska. Check out our AGU poster (2020).

Fig3_metrics_web

  • Related publications:
  • How do achieve bias-free earthquake locations, and why is this important? 

As an extension of my Ph.D. research, I continue to work on investigating source characteristics of earthquakes; by utilizing earthquake location and waveform modeling techniques, and complementing with geodesy. I specifically work on achieving minimally biased (calibrated), absolute earthquake locations. These calibrated locations typically have epicentral errors less than 5 km, and for recent events with available near-source data, focal depths can be constrained within 5 km of errors. A more in-depth description of the technique can be found on the MLOC website, and the dataset of calibrated earthquake locations can be downloaded from the Global Catalog of Calibrated Earthquake Locations (GCCEL).

Uncertainties in standardized earthquake locations, which run into tens-of kilometers in many regions of the world, are a serious limitation to seismotectonic studies and nuclear treaty monitoring. For example, they preclude individual events from being confidently attributed with mapped surface faults, complicate the study of mainshock-aftershock sequences and triggering behavior, and errors in depth are problematic for establishing the mechanical and rheological properties of the crust. These errors mostly arise from unknown Earth velocity structure. Well-established relative multiple-earthquake relocation techniques help eliminate these errors, but it remains challenging to achieve bias-free absolute (‘calibrated’) locations.

In my research, I use an advanced multiple-earthquake relocation technique (mloc Engdahl & Bergman, 2000, Ritzwoller et al., 2003, Walker et al., 2011*) that utilizes calibration of discrete clusters of earthquakes by exploiting near source data, aftershock deployments and InSAR observations. This technique focuses on achieving calibrated, absolute earthquake locations that have reduced epicenter uncertainties (< 5 km), improved focal depth resolution, and well-characterized hypocentral uncertainty.

The map below shows the calibrated earthquake locations done so far in Iran and Turkey. Please visit the GCCEL website for a complete and up-to-date Global Catalog of Calibrated Earthquake Locations.

fig_cross_section_web

plot_main_web

  • Related publications: