Identification of Low‐Frequency Earthquakes on the San Andreas Fault With Deep Learning

Thomas, Amanda M.; Inbal, Asaf; Searcy, J.; Shelly, D. R.; Bürgmann, Roland

doi:10.1029/2021gl093157

Cited by 12 publications

(13 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The target of such classifiers can be expanded to various phenomena by preparing appropriate training data. For example, it can be used to classify/identify surface waves (Chai et al 2022), volcanic earthquakes (Bueno Rodriguez et al 2022Canário et al 2020;Lara et al 2021), moonquakes (Civilini et al 2021), low-frequency earthquakes (Nakano et al 2019;Rouet-Leduc et al 2020;Thomas et al 2021;Takahashi et al 2021;Chen et al 2023), and mining-induced earthquakes/bastings (Linville et al 2019;Tibi et al 2019;Peng et al 2021;Wang et al 2023b). Applications in distributed acoustic sensing, which require efficient processing of large amounts of data, have also been reported (Hernandez et al 2022).…”

Section: Event Detection/classification and Arrival Time Pickingmentioning

confidence: 99%

Recent advances in earthquake seismology using machine learning

Kubo,

Naoi,

Kano

2024

Earth Planets Space

View full text Add to dashboard Cite

Given the recent developments in machine-learning technology, its application has rapidly progressed in various fields of earthquake seismology, achieving great success. Here, we review the recent advances, focusing on catalog development, seismicity analysis, ground-motion prediction, and crustal deformation analysis. First, we explore studies on the development of earthquake catalogs, including their elemental processes such as event detection/classification, arrival time picking, similar waveform searching, focal mechanism analysis, and paleoseismic record analysis. We then introduce studies related to earthquake risk evaluation and seismicity analysis. Additionally, we review studies on ground-motion prediction, which are categorized into four groups depending on whether the output is ground-motion intensity or ground-motion time series and the input is features (individual measurable properties) or time series. We discuss the effect of imbalanced ground-motion data on machine-learning models and the approaches taken to address the problem. Finally, we summarize the analysis of geodetic data related to crustal deformation, focusing on clustering analysis and detection of geodetic signals caused by seismic/aseismic phenomena. Graphical Abstract

show abstract

Section: Event Detection/classification and Arrival Time Pickingmentioning

confidence: 99%

Recent advances in earthquake seismology using machine learning

Kubo,

Naoi,

Kano

2024

Earth Planets Space

View full text Add to dashboard Cite

show abstract

“…We use several selection criteria to improve confidence in a tremor's location: i) We only locate the source of waveforms that the neural network model classifies as tremor above a fixed 'confidence' threshold (softmax ≥ 0.7, on a minimum of two stations). For comparison, decision thresholds used for tremor or LFE detection tend to be included within the [0.5, 0.75] interval [29], [45].…”

Section: Location Proceduresmentioning

confidence: 99%

Tremor Waveform Extraction and Automatic Location With Neural Network Interpretation

Hulbert

Jolivet²,

Gardonio³

et al. 2022

IEEE Trans. Geosci. Remote Sensing

View full text Add to dashboard Cite

Active faults release tectonic stress imposed by plate motion through a spectrum of slip modes, from slow, aseismic slip, to dynamic, seismic events. Slow earthquakes are often associated with tectonic tremor, non-impulsive signals that can easily be buried in seismic noise and go undetected.We present a new methodology aimed at improving the detection and location of tremors hidden within seismic noise. After identifying tremors with a classic convolutional neural network, we rely on neural network attribution to extract core tremor signatures. We observe that the signals resulting from the neural network attribution analysis correspond to a waveform traveling in the Earth's crust and mantle at wavespeeds consistent with seismological estimates. We then use these waveforms signatures to locate the source of tremors with standard arraybased techniques.We apply this method to the Cascadia subduction zone, where we identify tremor patches consistent with existing catalogs. This approach allows us to extract small signals hidden within the noise, and to locate more tremors than in existing catalogs.

show abstract

“…Tectonic tremor was discovered using seismic arrays (Obara, 2002) and can be located using the coherence of the envelope of the signals (Ide, 2010a, 2012; Wech & Creager, 2008). It is detected worldwide, mainly in subduction zones (Bartlow, 2020; Brown et al., 2009, 2013; Gallego et al., 2013; Husker et al., 2012; Ide, 2012; Obara, 2020), but also on strike‐slip faults (Shelly, 2017; Thomas et al., 2021; Wech et al., 2012), and other plate boundaries (Chamberlain et al., 2014; Tang et al., 2010).…”

Section: Introductionmentioning

confidence: 99%

“…, but also on strike-slip faults (Shelly, 2017;Thomas et al, 2021;Wech et al, 2012), and other plate boundaries (Chamberlain et al, 2014;Tang et al, 2010).…”

mentioning

confidence: 99%

Laboratory Hydrofractures as Analogs to Tectonic Tremors

Yuan,

Cochard,

Denolle

et al. 2024

AGU Advances

View full text Add to dashboard Cite

The fracture of Earth materials occurs over a wide range of time and length scales. Physical conditions, particularly the stress field and Earth material properties, may condition rupture in a specific fracture regime. In nature, fast and slow fractures occur concurrently: tectonic tremor events are fast enough to emit seismic waves and frequently accompany slow earthquakes, which are too slow to emit seismic waves and are referred to as aseismic slip events. In this study, we generate simultaneous seismic and aseismic processes in a laboratory setting by driving a penny‐shaped crack in a transparent sample with pressurized fluid. We leverage synchronized high‐speed imaging and high‐frequency acoustic emission (AE) sensing to visualize and listen to the various sequences of propagation (breaks) and arrest (sticks) of a fracture undergoing stick‐break instabilities. Slow radial crack propagation is facilitated by fast tangential fractures. Fluid viscosity and pressure regulate the fracture dynamics of slow and fast events, and control the inter‐event time and the energy released during individual fast events. These AE signals share behaviors with observations of episodic tremors in Cascadia, United States; these include: (a) bursty or intermittent slow propagation, and (b) nearly linear scaling of radiated energy with area. Our laboratory experiments provide a plausible model of tectonic tremor as an indicative of hydraulic fracturing facilitating shear slip during slow earthquakes.

show abstract

Identification of Low‐Frequency Earthquakes on the San Andreas Fault With Deep Learning

Cited by 12 publications

References 43 publications

Recent advances in earthquake seismology using machine learning

Recent advances in earthquake seismology using machine learning

Tremor Waveform Extraction and Automatic Location With Neural Network Interpretation

Laboratory Hydrofractures as Analogs to Tectonic Tremors

Contact Info

Product

Resources

About