Signal-based Bayesian Seismic Monitoring

David Moore, Stuart Russell
; Proceedings of the 20th International Conference on Artificial Intelligence and Statistics, PMLR 54:1293-1301, 2017.

Abstract

Detecting weak seismic events from noisy sensors is a difficult perceptual task. We formulate this task as Bayesian inference and propose a generative model of seismic events and signals across a network of spatially distributed stations. Our system, SIGVISA, is the first to directly model seismic waveforms, allowing it to incorporate a rich representation of the physics underlying the signal generation process. We use Gaussian processes over wavelet parameters to predict detailed waveform fluctuations based on historical events, while degrading smoothly to simple parametric envelopes in regions with no historical seismicity. Evaluating on data from the western US, we recover three times as many events as previous work, and reduce mean location errors by a factor of four while greatly increasing sensitivity to low-magnitude events.

Cite this Paper


BibTeX
@InProceedings{pmlr-v54-moore17a, title = {{Signal-based Bayesian Seismic Monitoring}}, author = {David Moore and Stuart Russell}, booktitle = {Proceedings of the 20th International Conference on Artificial Intelligence and Statistics}, pages = {1293--1301}, year = {2017}, editor = {Aarti Singh and Jerry Zhu}, volume = {54}, series = {Proceedings of Machine Learning Research}, address = {Fort Lauderdale, FL, USA}, month = {20--22 Apr}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v54/moore17a/moore17a.pdf}, url = {http://proceedings.mlr.press/v54/moore17a.html}, abstract = {Detecting weak seismic events from noisy sensors is a difficult perceptual task. We formulate this task as Bayesian inference and propose a generative model of seismic events and signals across a network of spatially distributed stations. Our system, SIGVISA, is the first to directly model seismic waveforms, allowing it to incorporate a rich representation of the physics underlying the signal generation process. We use Gaussian processes over wavelet parameters to predict detailed waveform fluctuations based on historical events, while degrading smoothly to simple parametric envelopes in regions with no historical seismicity. Evaluating on data from the western US, we recover three times as many events as previous work, and reduce mean location errors by a factor of four while greatly increasing sensitivity to low-magnitude events.} }
Endnote
%0 Conference Paper %T Signal-based Bayesian Seismic Monitoring %A David Moore %A Stuart Russell %B Proceedings of the 20th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2017 %E Aarti Singh %E Jerry Zhu %F pmlr-v54-moore17a %I PMLR %J Proceedings of Machine Learning Research %P 1293--1301 %U http://proceedings.mlr.press %V 54 %W PMLR %X Detecting weak seismic events from noisy sensors is a difficult perceptual task. We formulate this task as Bayesian inference and propose a generative model of seismic events and signals across a network of spatially distributed stations. Our system, SIGVISA, is the first to directly model seismic waveforms, allowing it to incorporate a rich representation of the physics underlying the signal generation process. We use Gaussian processes over wavelet parameters to predict detailed waveform fluctuations based on historical events, while degrading smoothly to simple parametric envelopes in regions with no historical seismicity. Evaluating on data from the western US, we recover three times as many events as previous work, and reduce mean location errors by a factor of four while greatly increasing sensitivity to low-magnitude events.
APA
Moore, D. & Russell, S.. (2017). Signal-based Bayesian Seismic Monitoring. Proceedings of the 20th International Conference on Artificial Intelligence and Statistics, in PMLR 54:1293-1301

Related Material