← Back to Sessions

WITHDRAWN Inter-station Approaches to Identify Repeated Seismic Sources

Description: 

WITHDRAWN Many seismic sources on the Earth are small and complex, but repetitive. For instance, the small earthquakes in tremor repeatedly rupture fault patches. Trucks repeatedly drive on the same road. And water repeatedly flows over the same riverbed.

These repeated sources represent promising datasets to learn more about the shallow and deep subsurface, but they can be difficult to analyze. As a first step, we would like to obtain the Green’s functions, or path effects, of sources recurring at particular locations. But we do not know the timing or source time functions of those sources, and we must somehow disentangle the source time functions from the data in order to retrieve the Green’s functions.

Here I tackle this blind source separation problem by building on approaches in speech processing. I attempt to directly extract the Green’s functions from long records of complex shaking, with an approach that effectively ignores the source time functions. Specifically, this approach searches for the set of Green’s functions that promotes the maximum coherence between stations. Since the approach maximizes the coherence between stations, it has little sensitivity to the complex and unknown source time functions. And since the approach is a maximization, it can use many of our standard optimization tools.

This optimization problem is not without challenges. It is many dimensional, and in some ways underdetermined. I present initial results in addressing these challenges. I show that the analysis is effective in quickly identifying first-order features the Green’s functions in synthetics and in tectonic tremor given some regularization. However, the method does not yet work to identify subtle features of the synthetics or without significant regularization.

Session: New Directions in Environmental, Seismic Hazard and Mineral Resource Exploration Studies [Poster]

Type: Poster

Date: 4/17/2025

Presentation Time: 08:00 AM (local time)

Presenting Author: Jessica

Student Presenter: No

Invited Presentation: 

Poster Number: 113

Authors