"Big data for small earthquakes: Shaking up seismology with data science"

Speaker: Dr. Karianne Bergen

Abstract: Earthquake detection, the problem of extracting weak earthquake signals from continuous waveform data recorded by sensors in a seismic network, is a critical and challenging task in seismology. New algorithmic advances in “big data” and artificial intelligence have created opportunities to advance the state-of-the-art in earthquake detection algorithms.

In this talk, I will present a new earthquake detection algorithm called Fingerprint and Similarity Thresholding (FAST; Yoon et al, 2015). FAST approaches earthquake detection as a data mining problem and adapts technology for rapid audio identification (similar to the Shazam mobile app) to seismic waveform data. In particular, FAST leverages locality sensitive hashing (LSH), a technique for efficiently identifying similar items in large data sets, to detect new candidate earthquakes without template waveforms (“training data”). Recent work has extended FAST to multiple station to enable event detection in data recorded over a seismic network (Bergen & Beroza, 2018), and optimized the FAST software for use on long-duration continuous data sets (Rong et al, 2018). FAST is now capable of discovering new events with unknown sources in 10 years of continuous data recorded on dozens of channels.

Lunch will be provided. As always, please plan to bring reusable plates and cutlery to reduce waste.