Mining Seismic Wavefields (Stanford)
During my Ph.D., I worked with Greg Beroza in the Department of Geophysics on applications of data mining and machine learning techniques in earthquake seismology.
Ph.D. thesis: Big data for small earthquakes: Detecting earthquakes over a seismic network with waveform similarity search.
FAST Earthquake Detector
Fingerprint and Similarity Thresholding (FAST), introduced in Yoon et al. (2015) is a novel method for large-scale earthquake detection. FAST draws on techniques used by content-based audio recognition systems (like the Shazam app, or Google's Waveprint algorithm), and adapts these methods for the unique characteristics of seismic waveform data.
FAST is an unsupervised detector -- it does not require any examples of known event waveforms or waveform characteristics for detection. This allows FAST to discover new earthquake sources, even if template waveforms (training data) is not available.
FAST was developed as part of a multidisciplinary collaboration at Stanford, involving researchers from the Department of Geophysics, the Institute for Computational and Mathematical Engineering (ICME), and the Department of Computer Science.
The FAST code is available on Github!
Related Publications
FAST Overview: Earthquake detection through computationally efficient similarity search
Fingerprint Extraction
Network Detection: Detecting Earthquakes over a Seismic Network using Single-Station Similarity Measures
Implementation and Performance: Locality-Sensitive Hashing for Earthquake Detection: A Case Study Scaling Data-Driven Science (also see Stanford DAWN blog)
FAST at scale (Diablo Canyon case study): Unsupervised Large-Scale Search for Similar Earthquake Signals
Other FAST applications:
Broader context: Machine learning for data-driven discovery in solid Earth geoscience
Press
Stanford Scientists develop "Shazam for Earthquakes" (Dec 4, 2015)
also featured in: IEEE Spectrum (Dec.9, 2015), Smithsonian Magazine (Dec 11, 2015),
local news coverage: NBC Bay Area (video) (Dec. 10, 2015), KRON4 News (video) (Dec 11, 2015)
Algorithms spot millions of California’s tiniest quakes in historical data (Apr 18, 2019)
Video
Earthquake monitoring in the age of "big data:" challenges and opportunities (video), UTIG Seminar, Jackson School of Geosciences, UT Austin (Sept 2019)
Big data for small earthquakes: a data mining approach to earthquake detection (video), FISH Seminar, Earth Resources Laboratory, MIT (October 2018)
Earthquake Detection Through Computationally Efficient Similarity Search (video), USGS Earthquake Science Center Seminar (August 2015, with Clara Yoon)
Awards
Editorial Board of Geophysical Journal International: 2018 GJI Student Author Award (November 2018)
Seismological Society of America (SSA) 2016 Annual Meeting: Student Presentation Award (April 2017)
Seismological Society of America (SSA) 2016 Annual Meeting: Student Presentation Award (April 2016)
American Geophysical Union (AGU) 2015 Annual Meeting: Outstanding Student Paper Award (December 2015)
Seismic data sets for Machine Learning
STanford EArthquake Dataset (STEAD): A Global Data Set of Seismic Signals for AI. Compiled by Mousavi et al. (2019) [ref].
SCEDC Deep Learning Datasets. Compiled by Ross et al. (2018) [ref] [ref].
INSTANCE: The Italian Seismic Dataset for Machine Learning. Compiled by Michelini et al. (2021) [ref].
LEN-DB: Local earthquakes detection: A benchmark dataset of 3-component seismograms build on a global scale. Compiled by Magrini et al. (2020) [ref].
LANL Earthquake Prediction Dataset (hosted by Kaggle). Reference: Johnson et al. (2021) [ref].
Collaborators
Stanford Geophysics: Clara Yoon (now at USGS), Ossian O'Reilly (now at USC), Greg Beroza
Stanford Computer Science: Kexin Rong, Hashem Elezabi, Peter Bailis, Phil Levis
Visitor in research group of Professor Satoshi Ide at the University of Tokyo (October 2016)