Ensemble Learning for Earthquake Detection and Phase Picking: Methodology and Application
Description:
Seismic detection and picking is the first step toward earthquake catalog building, earthquake monitoring, and seismic hazard management. Recent advances in deep learning have leveraged the amount of labeled seismic data to improve the capability of detecting and picking earthquake signals. While these deep learning methods have shown great promise, they remain challenged by low generalizability and sensitivity to low signal-to-noise ratios (SNRs). Here, we propose a new processing workflow, which integrates conventional deep learning models, multi-frequency band predictions, and ensemble estimations including coherence-based and learning-based ensemble algorithms, in order to enhance the prediction performance. Our primary test on INSTANCE, STEAD, and PNW datasets to demonstrate our multiband and ensemble estimation strategies can significantly improve detection and picking rate and accuracy, especially for low SNR signals. We further show the effectiveness of this workflow for characterizing volcanic seismicity in Mount St. Helens region. Available detection or picking models can be easily assembled into our workflow without further training costs.
Session: Opportunities and Challenges for Machine Learning Applications in Seismology [Poster]
Type: Poster
Date: 4/19/2023
Presentation Time: 08:00 AM (local time)
Presenting Author: Congcong Yuan
Student Presenter: Yes
Invited Presentation:
Authors
Congcong Yuan Presenting Author Corresponding Author cyuan@g.harvard.edu Harvard University |
Yiyu Ni niyiyu@uw.edu University of Washington |
Youzuo Lin ylin@lanl.gov Los Alamos National Laboratory |
Marine Denolle mdenolle@uw.edu University of Washington |
|
|
|
|
|
Ensemble Learning for Earthquake Detection and Phase Picking: Methodology and Application
Category
Opportunities and Challenges for Machine Learning Applications in Seismology