scholarly journals Tremor Waveform Denoising and Automatic Location with Neural Network Interpretation

2021 ◽  
Author(s):  
Claudia Hulbert ◽  
Romain Jolivet ◽  
Blandine Gardonio ◽  
Paul Johnson ◽  
Christopher Ren ◽  
...  
Keyword(s):  
Neural Network ◽  
Seismic Noise ◽  
Active Faults ◽  
Tectonic Stress ◽  
Plate Motion ◽  
Cascadia Subduction Zone ◽  
Aseismic Slip ◽  
Tectonic Tremor ◽  
Slow Earthquakes ◽  
Slip Event

<p>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. </p><p>We present a new methodology aimed at improving the detection and location of tremors hidden within seismic noise. After detecting tremors with a classic convolutional neural network, we rely on neural network attribution to extract core tremor signatures. By identifying and extracting tremor characteristics, in particular in the frequency domain, the attribution analysis allows us to uncover structure in the data and denoise input waveforms. In particular, we show that these cleaned signals correspond to a waveform traveling in the Earth's crust and mantle at wavespeeds consistent with local estimates. We then use these cleaned waveforms to locate tremors with standard array-based techniques. </p><p>We apply this method to the Cascadia subduction zone. We analyze a slow slip event that occurred in 2018 below the southern end of the Vancouver Island, Canada, where we identify tremor patches consistent with existing catalogs. Having validated our new methodology in a well-studied area, we further apply it to various tectonic contexts and discuss the implications of tremor occurrences in the scope of exploring the interplay between seismic and aseismic slip.</p>

Complexity of low-frequency earthquakes activity in western Shikoku

2020 ◽  
Author(s):  
Natalia Poiata ◽  
Jean-Pierre Vilotte ◽  
Nikolai Shapiro ◽  
Mariano Supino ◽  
Kazushige Obara
Keyword(s):  
Subduction Zones ◽  
Large Scale ◽  
Active Faults ◽  
Low Frequency ◽  
Space Time ◽  
Interaction Patterns ◽  
Slow Slip Events ◽  
Time Activity ◽  
Tectonic Tremor ◽  
Slow Earthquakes

<p>Short-duration transient seismic events known as low-frequency earthquakes (LFEs) are a component of the slow earthquakes family observed in the transition zone, at the root of seismogenic regions of the subduction zones or active faults. LFEs are the signature of impulse seismic energy radiation associated to and often mixed within complex tectonic tremor signal. Detailed analysis and characterization of LFE space-time activity in relation to other slow earthquake phenomena can provide important information about the state and the processes of fault interface.</p><p>We derive a catalog of LFEs in western Shikoku (Japan) by applying a full waveform coherency-based detection and location method to the 4-year continuous data covering the period of 2013-2016 and recorded at Hi-net seismic stations of NIED. The obtained catalog of over 150,000 detected events allows looking into the details of LFE space-time activity during the tectonic tremor sequences and inter-sequence periods.</p><p>We use this catalogue of LFEs to perform a systematic statistical analysis of the event occurrence patterns by applying correlation and clustering analysis to infer the large-scale (long temporal ~ 1-2 day duration) space-time characteristics and interaction patterns of activity and its potential relation to the structural complexity of the subducting plate. We also analyze the correlation between the migration of clustered LFE activity during energetic tremor sequences and short-term slow slip events occurring in the area during the analyzed period.</p>

scholarly journals Tectonic stress regime in the 2003–2004 and 2012–2015 earthquake swarms in the Ubaye Valley, French Alps

2018 ◽  
Vol 175 (6) ◽  
pp. 1997-2008 ◽  
Author(s):  
Lucia Fojtíková ◽  
Václav Vavryčuk
Keyword(s):  
Seismic Activity ◽  
Joint Inversion ◽  
Earthquake Swarm ◽  
Active Faults ◽  
Tectonic Stress ◽  
Geological Mapping ◽  
Earthquake Swarms ◽  
Stress Regime ◽  
French Alps ◽  
Principal Axes

Abstract We study two earthquake swarms that occurred in the Ubaye Valley, French Alps within the past decade: the 2003–2004 earthquake swarm with the strongest shock of magnitude ML = 2.7, and the 2012–2015 earthquake swarm with the strongest shock of magnitude ML = 4.8. The 2003–2004 seismic activity clustered along a 9-km-long rupture zone at depth between 3 and 8 km. The 2012–2015 activity occurred a few kilometres to the northwest from the previous one. We applied the iterative joint inversion for stress and fault orientations developed by Vavryčuk (2014) to focal mechanisms of 74 events of the 2003–2004 swarm and of 13 strongest events of the 2012–2015 swarm. The retrieved stress regime is consistent for both seismic activities. The σ 3 principal axis is nearly horizontal with azimuth of ~ 103°. The σ 1 and σ 2 principal axes are inclined and their stress magnitudes are similar. The active faults are optimally oriented for shear faulting with respect to tectonic stress and differ from major fault systems known from geological mapping in the region. The estimated low value of friction coefficient at the faults 0.2–0.3 supports an idea of seismic activity triggered or strongly affected by presence of fluids.

Observing seismic signatures of slow slip events with unsupervised learning

2021 ◽  
Author(s):  
Leonard Seydoux ◽  
Michel Campillo ◽  
René Steinmann ◽  
Randall Balestriero ◽  
Maarten de Hoop
Keyword(s):  
Clustering Algorithm ◽  
Low Frequency ◽  
Seismic Signal ◽  
Geodetic Data ◽  
Slow Slip ◽  
Slow Slip Events ◽  
Slow Slip Event ◽  
Intermittent Behavior ◽  
Slow Earthquakes ◽  
Slip Event

<p>Slow slip events are observed in geodetic data, and are occasionally associated with seismic signatures such as slow earthquakes (low-frequency earthquakes, tectonic tremors). In particular, it was shown that swarms of slow earthquake can correlate with slow slip events occurrence, and allowed to reveal the intermittent behavior of several slow slip events. This observation was possible thanks to detailed analysis of slow earthquakes catalogs and continuous geodetic data, but in every case, was limited to particular classes of seismic signatures. In the present study, we propose to infer the classes of seismic signals that best correlate with the observed geodetic data, including the slow slip event. We use a scattering network (a neural network with wavelet filters) in order to find meaningful signal features, and apply a hierarchical clustering algorithm in order to infer classes of seismic signal. We then apply a regression algorithm in order to predict the geodetic data, including slow slip events, from the occurrence of inferred seismic classes. This allow to (1) identify seismic signatures associated with the slow slip events as well as (2) infer the the contribution of each classes to the overall displacement observed in the geodetic data. We illustrate our strategy by revisiting the slow-slip event of 2006 that occurred beneath Guerrero, Mexico.</p>

Afterslip and slow slip events in the postseismic deformation of the 2016 Pedernales earthquake, Ecuador

2020 ◽  
Author(s):  
Frederique Rolandone ◽  
Jean-Mathieu nocquet ◽  
Patricia Mothes ◽  
Paul Jarrin ◽  
Mathilde Vergnolle
Keyword(s):  
Subduction Zones ◽  
Postseismic Deformation ◽  
Slow Slip ◽  
Aseismic Slip ◽  
Plate Interface ◽  
Slow Slip Events ◽  
North East ◽  
Rupture Area ◽  
Slip Event ◽  
Seismic Swarms

<p>In subduction zones, slip along the plate interface occurs in various modes including earthquakes, steady slip, and transient accelerated aseismic slip during either Slow Slip Events (SSE) or afterslip. We analyze continuous GPS measurements along the central Ecuador subduction segment to illuminate how the different slip modes are organized in space and time in the zone of the 2016 Mw 7.8 Pedernales earthquake. The early post-seismic period (1 month after the earthquake) shows large and rapid afterslip developing at discrete areas of the megathrust and a slow slip event remotely triggered (∼100 km) south of the rupture of the Pedernales earthquake. We find that areas of large and rapid early afterslip correlate with areas of the subduction interface that had hosted SSEs in years prior to the 2016 earthquake. Areas along the Ecuadorian margin hosting regular SSEs and large afterslip had a dominant aseismic slip mode that persisted throughout the earthquake cycle during several years and decades: they regularly experienced SSEs during the interseismic phase, they did not rupture during the 2016 Pedernales earthquake, they had large aseismic slip after it. Four years after the Pedernales earthquake, postseismic deformation is still on-going. Afterslip and SSEs are both involved in the postseimsic deformation. Two large aftershocks (Mw 6.7 & 6.8) occurred after the first month of postseismic deformation in May 18, and later in July 7 2016 two other large aftershocks (Mw 5.9 & 6.3) occurred, all were located north east of the rupture. They may have triggered their own postseismic deformation. Several seismic swarms were identified south and north of the rupture area by a dense network of seismic stations installed during one year after the Pedernales earthquakes, suggesting the occurrence of SSEs. Geodetically, several SSEs were detected during the postseismic deformation either in areas where no SSEs were detected previously, or in areas where regular seismic swarms and repeating earthquakes were identified. The SSEs may have been triggered by the stress increment due to aftershocks or due to afterslip.</p>

scholarly journals The San Andreas Fault Paleoseismic Record at Elizabeth Lake: Why are There Fewer Surface-Rupturing Earthquakes on the Mojave Section?

2021 ◽  
Author(s):  
Sean P. Bemis ◽  
Kate Scharer ◽  
James F. Dolan
Keyword(s):  
San Andreas Fault ◽  
Active Faults ◽  
Structural Complexity ◽  
Plate Motion ◽  
Fault System ◽  
San Andreas ◽  
Time Period ◽  
Earthquake Ruptures ◽  
Major Fault ◽  
Restraining Bend

ABSTRACT The structural complexity of active faults and the stress release history along the fault system may exert control on the locus and extent of individual earthquake ruptures. Fault bends, in particular, are often invoked as a possible mechanism for terminating earthquake ruptures. However, there are few records available to examine how these factors may influence the along-fault recurrence of earthquakes. We present a new paleoearthquake chronology for the southern San Andreas fault at Elizabeth Lake and integrate this record with existing paleoearthquake records to examine how the timing and frequency of earthquakes vary through a major restraining bend. This restraining bend features a mature, throughgoing right-lateral strike-slip fault, two major fault intersections, proposed subsurface fault dip changes, and a >200  km long section of fault misaligned with the regional plate motion. The Frazier Mountain, Elizabeth Lake, Pallett Creek, Wrightwood, and Pitman Canyon paleoseismic sites are located on this relatively linear surface trace of the San Andreas fault between fault bends. Our paleoseismic investigations at Elizabeth Lake document 4–5 earthquakes, since ∼1100  C.E., similar to the number of earthquakes recorded at Pallett Creek. In contrast, the Frazier Mountain and Wrightwood sites each record 8–9 earthquakes during this same time period. Differences in earthquake frequency demonstrate that fewer earthquakes rupture the central portion of the restraining bend than occur near the fault bends and intersections. Furthermore, the similarity of earthquake records from the Bidart Fan paleoseismic site northwest of the restraining bend and the Frazier Mountain paleoseismic site suggests that the broad, 30° curve of the Big Bend section of the San Andreas fault exerts less influence on fault rupture behavior than the 3D geometry of the Mojave sections of the fault.

scholarly journals Comparative study of tectonic tremor locations: Characterization of slow earthquakes in Guerrero, Mexico

2016 ◽  
Vol 121 (7) ◽  
pp. 5136-5151 ◽  
Author(s):  
J. Maury ◽  
S. Ide ◽  
V. M. Cruz-Atienza ◽  
V. Kostoglodov ◽  
G. González-Molina ◽  
...  
Keyword(s):  
Comparative Study ◽  
Tectonic Tremor ◽  
Slow Earthquakes

Effect of plate motion on plume-induced subduction initiation

2021 ◽  
Author(s):  
Marzieh Baes ◽  
Stephan Sobolev ◽  
Taras Gerya ◽  
Robert Stern ◽  
Sascha Brune
Keyword(s):  
Subduction Zone ◽  
Late Cretaceous ◽  
Subduction Zones ◽  
Plate Tectonics ◽  
Plate Motion ◽  
Cascadia Subduction Zone ◽  
Caribbean Plate ◽  
Subduction Initiation ◽  
Southern Margin ◽  
The Caribbean

<p>Subduction zones are key components of plate tectonics and plate tectonics could not begin until the first subduction zone formed. Plume-induced subduction initiation, which has been proposed as triggering the beginning of plate tectonics (Gerya et al., 2015), is one of the few scenarios that can break the lithosphere and recycle a stagnant lid without requiring any pre-existing weak zones. So far, two natural examples of plume-induced subduction initiation have been recognized. The first was found in southern and western margins of the Caribbean Plate (Whattam and Stern 2014). Initiation of the Cascadia subduction zone in Eocene times has been proposed to be the second example of plume-induced subduction initiation (Stern and Dumitru, 2019).</p><p>The focus of previous studies was to inspect plume-lithosphere interaction either for the case of stationary lithosphere (e.g., Gerya et al., 2015) or for moving lithosphere without considering the effect of lithospheric magmatic weakening above the plume head (e.g., Moore et al., 1998). In present study we investigate the response of moving oceanic lithosphere to the arrival of a rising mantle plume head including the effect of magmatic lithospheric weakening. We used 3D numerical thermo-mechanical modeling. Using I3ELVIS code, which is based on finite difference staggered grid and marker-in-cell with an efficient OpenMP multigrid solver (Gerya, 2010), we show that plate motion may affect the plume-induced subduction initiation only if a moderate size plume head (with a radius of 140 km in our experiments) impinges on a young but subductable lithosphere (with the age of 20 Myr). Outcomes indicate that lithospheric strength and plume buoyancy are key parameters in penetration of the plume and subduction initiation and that plate speed has a minor effect. We propose that eastward motion of the Farallon plate in Late Cretaceous time could play a key role in forming new subduction zones along the western and southern margin of the Caribbean plate.</p><p> </p><p>References:</p><p>Gerya, T., 2010, Introduction to Numerical Geodynamic Modelling.. Cambridge University Press.</p><p>Gerya, T.V., Stern, R.J., Baes, M., Sobolev, S.V. and Whattam, S.A., 2015. Plume-induced subduction initiation triggered Plate Tectonics on Earth. Nature, 527, 221–225.</p><p>Moore, W. B., Schubert, G. and Tackley, P., 1998, Three-dimensional simulations of plume-lithosphere interaction at the Hawaiian swell. Science, 279, 1008-1011.</p><p>Stern, R.J., and Dumitru, T.A., 2019, Eocene initiation of the Cascadia subduction zone: A second example of plume-induced subduction initiation? Geosphere, v. 15, 659-681.</p><p>Whattam, S.A. and Stern, R.J., 2014. Late Cretaceous plume-induced subduction initiation along the southern margin of the Caribbean and NW South America: The first documented example with implications for the onset of plate tectonics. Gondwana Research, 27, doi: 10.1016/j.gr.2014.07.011.</p>

Mapping mantle flows underneath the North American and Caribbean Plates

2021 ◽  
Author(s):  
Hejun Zhu
Keyword(s):  
North American ◽  
Subduction Zones ◽  
Current Model ◽  
Waveform Inversion ◽  
Body Waves ◽  
Plate Motion ◽  
Cascadia Subduction Zone ◽  
Fast Axis ◽  
The North ◽  
Short Period

<p>In this talk, I will present a new 3-D azimuthally anisotropic tomographic model, namely US32, for the North American and Caribbean Plates. This model is constrained by using seismic data from USArray and full waveform inversion. The inversion uses data from 180 regional earthquakes recorded by 4,516 seismographic stations, resulting in 586,185 frequency-dependent phase measurements. Three-component short-period body waves and long-period surface waves are combined to simultaneously constrain deep and shallow structures. The current azimuthally anisotropic model US32 is the result of 32 pre-conditioned conjugate-gradient iterations. In the current model, I observe a complex depth-dependent pattern for fast axis directions across the North American and Caribbean Plates.<span>  </span>At shallow depths, these fast axis directions delineate local geological provinces, such as the Snake River Plain, Cascadia subduction zone, Rio Grand Rift, etc. At greater depths, the fast axis directions follow the absolute plate motion trajectories at most places. At depths around 700 km, the fast axis directions are perpendicular to the strikes of the mapped Farallon slab, suggesting the presence of 2-D corner flows induced by this ancient subduction underneath the mantle transition zone. In addition, underneath the Cascadia and Cocos subduction zones at depths from 250 to 500 km, the fast axis directions suggest the presence of toroid-mode mantle flows, following the geometry of fast downwelling materials.</p>

Sign in / Sign up

Export Citation Format

Share Document