Reliable Real‐Time Seismic Signal/Noise Discrimination With Machine Learning

Надёжное распознавание сейсмических сигналов и шумов в реальном времени с помощью машинного обучения
Zefeng Li, Suraj Nair, Zachary E. Ross, Egill Hauksson, Yisong Yue, Egill Hauksson, Men‐Andrin Meier, Jennifer R. Andrews, Anshul Ramachandran, Ashwin Balakrishna, Peter Kundzicz
2018-12-31

convolutional neural networksearthquake early warninggenerative adversarial networkrandom forestseismic signal/noise discrimination
Abstract In earthquake early warning (EEW), every sufficiently impulsive signal is potentially the first evidence for an unfolding large earthquake. More often than not, however, impulsive signals are mere nuisance signals. One of the most fundamental—and difficult—tasks in EEW is to rapidly and reliably discriminate real local earthquake signals from all other signals. This discrimination is necessarily based on very little information, typically a few seconds worth of seismic waveforms from a small number of stations. As a result, current EEW systems struggle to avoid discrimination errors and suffer from false and missed alerts. In this study we show how modern machine learning classifiers can strongly improve real‐time signal/noise discrimination. We develop and compare a series of nonlinear classifiers with variable architecture depths, including fully connected, convolutional and recurrent neural networks, and a model that combines a generative adversarial network with a random forest. We train all classifiers on the same data set, which includes 374 k local earthquake records (M3.0–9.1) and 946 k impulsive noise signals. We find that all classifiers outperform existing simple linear classifiers and that complex models trained directly on the raw signals yield the greatest degree of improvement. Using 3‐s‐long waveform snippets, the convolutional neural network and the generative adversarial network with a random forest classifiers both reach 99.5% precision and 99.3% recall on an independent validation data set. Most misclassifications stem from impulsive teleseismic records, and from incorrectly labeled records in the data set. Our results suggest that machine learning classifiers can strongly improve the reliability and speed of EEW alerts.
1
All nonlinear classifiers outperform existing simple linear classifiers, while complex models trained directly on raw waveforms provide the greatest improvement.
2
Modern machine-learning classifiers substantially improve real-time discrimination between local earthquake signals and impulsive noise in earthquake early-warning systems.
3
Most remaining errors involve impulsive teleseismic records and incorrectly labeled training records, indicating key sources of classification failure.
4
The study compares fully connected, convolutional, recurrent, and generative-adversarial-network–random-forest classifiers trained on 374,000 local earthquake records and 946,000 impulsive noise signals.
5
Using only 3-second waveform snippets, convolutional and generative-adversarial-network–random-forest models achieve 99.5% precision and 99.3% recall on an independent validation set.

Real-time seismic signals, including local earthquake waveforms and impulsive noise signals, used in earthquake early warning

Reliable rapid discrimination of local earthquake signals from impulsive noise and its effect on earthquake early warning alert performance

Publication Details
Publication Date
2018-12-31
Journal
Publisher
ISSN
Cited by
175
Access Type
Author Information
Authors
Zefeng Li
Suraj Nair
Zachary E. Ross
Egill Hauksson
Yisong Yue
Egill Hauksson
Men‐Andrin Meier
Jennifer R. Andrews
Anshul Ramachandran
Ashwin Balakrishna
Peter Kundzicz
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%