Making phase-picking neural networks more consistent and interpretable

The Seismic Record
By: , and 

Metrics

3
Crossref references
Web analytics dashboard Metrics definitions

Links

Abstract

Improving the interpretability of phase‐picking neural networks remains an important task to facilitate their deployment to routine, real‐time seismic monitoring. The popular phase‐picking neural networks published in the literature lack interpretability because their output prediction scores do not necessarily correspond with the reliability of phase picks and can even be highly inconsistent depending on how we window the waveform data. Here, we show that systematically shifting the waveforms during training and using an antialiasing filter within the neural network architecture can substantially improve the consistency of the output prediction scores and can even make them scale with the signal‐to‐noise ratios of the waveforms. We demonstrate the improvements by applying these approaches to a commonly used phase‐picking neural network architecture and using waveform data from the 2019 Ridgecrest earthquake sequence.

Publication type Article
Publication Subtype Journal Article
Title Making phase-picking neural networks more consistent and interpretable
Series title The Seismic Record
DOI 10.1785/0320230054
Volume 4
Issue 1
Publication Date March 06, 2024
Year Published 2024
Language English
Publisher Seismological Society of America
Contributing office(s) Geologic Hazards Science Center - Seismology / Geomagnetism
Description 9 p.
First page 72
Last page 80
Additional publication details