Physical and Statistical Behavior of Multifault Earthquakes: Darfield Earthquake Case Study, New Zealand

Quigley, Mark; Jiménez, Antonia Aránega; Duffy, Brendan; King, Tamarah R

doi:10.1029/2019jb017508

Cited by 21 publications

(14 citation statements)

References 110 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This type of 3D space-time feature can be handled by three-dimensional convolutional layers, rather than the two-dimensional convolutional layers we use in this study. Earthquakes in nature sometimes rupture through complex fault networks (Cesca et al, 2017;Quigley et al, 2019;Wang et al, 2018). While it is unknown whether these types of multifault ruptures emit precursors that warn of upcoming event size and timing, a 3D neural network approach, as described above, could be used as a tool to assess this question.…”

Section: Discussionmentioning

confidence: 99%

Can Deep Learning Predict Complete Ruptures in Numerical Megathrust Faults?

Blank

Morgan

2021

Geophysical Research Letters

View full text Add to dashboard Cite

Promising results have been attained in the laboratory, where time to failure of simulated gouge in a double-direct shear apparatus was predicted using shallow machine learning methodology, with acoustic emission data as inputs (Rouet-Leduc et al., 2017). Furthermore, earthquake catalog data derived from a double-direct shear apparatus was used to predict timing of upcoming events, shear stress, and friction (Lubbers et al., 2018), showing that simplified representations of raw continuous data sets might hold predictive signals. Machine learning methodology has been applied with great success to the engineering fields, where deep learning algorithms have been used to predict the remaining useful life (time to failure) of mechanical equipment such as roller bearings (Deutsch & He, 2017;Li et al., 2018;Ren et al., 2017). With regard to earthquake prediction, however, time to failure alone is not enough. As a society, we care about predicting the one or two large events that take place in a given year, rather than the tens of thousands of small events that cause little damage. Corbi et al. (2019Corbi et al. ( , 2020) used a subduction zone analog model to address the problem of size and timing prediction. The megathrust interface in their model consisted of two velocity-weakening asperities that ruptured quasi-periodically in response to constant loading boundary conditions. The model experienced a range of earthquake magnitudes, as the fault underwent either complete or partial interface failure. They developed multiple machine learning models, using trench parallel and perpendicular surface displacements as inputs, to predict time to failure (TTF regression;Corbi et al., 2019) and failure imminence (binary classification; Corbi et al., 2020) along nine discrete along-strike zones. Their model was apparently able to recognize precursory displacement associated with the asperities. When both asperities were emitting

show abstract

Section: Discussionmentioning

confidence: 99%

Can Deep Learning Predict Complete Ruptures in Numerical Megathrust Faults?

Blank

Morgan

2021

Geophysical Research Letters

View full text Add to dashboard Cite

show abstract

“…However, in the absence of a confirmed through‐going rupture pathway identified in the well data, here we use the results of the static Coulomb stress modeling to investigate how proposed large‐magnitude earthquakes on the Pitas Point and Ventura faults may influence subsequent seismicity on the San Cayetano and Southern San Cayetano faults, via static Coulomb stress transfer. Models of static Coulomb stress change have previously been utilized to understand the kinematics for the multifault rupture sequence during the 2010 M w 7.1 Darfield earthquake, New Zealand (Quigley et al, 2019) and to test the feasibility of various multifault rupture scenarios between the Fox Creek and Fox Peak faults on the South Island of New Zealand (Stahl et al, 2016).…”

Section: Discussionmentioning

confidence: 99%

“…Moreover, the importance of having an accurate understanding of the three-dimensional fault network geometry to improve assessment of seismic hazard has been brought into focus by several recent large-magnitude earthquakes that propagated along multiple faults with complex geometry and kinematics. For example, the 2016 M w 7.8 Kaikōura earthquake, New Zealand (Hamling et al, 2017), the 2010 M w 7.2 El Mayor-Cucapah earthquake, Mexico (Fletcher et al, 2014(Fletcher et al, , 2016, and the 2010 M w 7.1 Darfield earthquake, New Zealand (Beavan et al, 2012;Quigley et al, 2019) all involved slip on multiple faults with various orientations and senses of slip. During the Kaikōura earthquake, small faults that may have been previously interpreted to represent low seismic hazard played a key role in enhancing both fault connectivity and stress transfer during the event (Clark et al, 2017;Hamling et al, 2017).…”

Section: Introductionmentioning

confidence: 99%

Three‐Dimensional Structure, Ground Rupture Hazards, and Static Stress Models for Complex Nonplanar Thrust Faults in the Ventura Basin, Southern California

et al. 2020

View full text Add to dashboard Cite

To investigate the subsurface geometry of a recently discovered, seismically active fault in the Ventura basin, southern California, USA, we present a series of cross sections and a new three-dimensional fault model across the Southern San Cayetano fault (SSCF) based on integration of surface data with petroleum industry well log data. Additionally, the fault model for the SSCF, along with models of other regional faults extracted from the Southern California Earthquake Center three-dimensional Community Fault Model, are incorporated in static Coulomb stress modeling to investigate static Coulomb stress transfer between thrust faults with complex geometry and to further our understanding of stress transfer in the Ventura basin. The results of the subsurface well investigation provide evidence for a low-angle SSCF that dips~15°north and connects with the western section of the San Cayetano fault around 1.5-3.5 km depth. We interpret the results of static Coulomb stress models to partly explain contrasting geomorphic expression between different sections of the San Cayetano fault and a potential mismatch in timings between large-magnitude uplift events suggested by paleoseismic studies on the Pitas Point, Ventura, and San Cayetano faults. In addition to new insights into the structure and potential rupture hazard of a recently discovered active reverse fault in a highly populated area of southern California, this study provides a simple method to model static Coulomb stress transfer on complex geometry faults in fold and thrust belts.

show abstract

“…Generally, the lack of fault data prevents such kind of analysis in these regions, but as seen, it can imply very important hazard (PGA) changes respect to classical approaches. Moreover, considering multi-fault ruptures is a step forward towards a more realistic seismic hazard assessment, as the occurrence of multi-fault ruptures in nature is becoming an increasingly identified phenomena (e.g., Langridge et al, 2018;Quigley et al, 2019). We therefore aim that our study could serve as a case example for other studies focused in low-to-moderate seismicity regions, such as the EBSZ.…”

Section: Overview: Limitations and Perspectivesmentioning

confidence: 99%

Fault System-Based Probabilistic Seismic Hazard Assessment of a Moderate Seismicity Region: The Eastern Betics Shear Zone (SE Spain)

et al. 2020

View full text Add to dashboard Cite

Including faults as seismogenic sources in probabilistic seismic hazard assessments (PSHA) has turned into a common practice as knowledge of active faults is improving. Moreover, the occurrence of earthquakes in multi-fault ruptures has evidenced the need to understand faults as interacting systems rather than independent sources. We present a PSHA for the Southeastern Spain obtained by including the faults of a moderate seismicity region, the Eastern Betics Shear Zone (EBSZ) in SE Spain, as the main seismogenic sources in two separate source models, one considering background seismicity. In contrast with previous studies in Spain, earthquake occurrence of the EBSZ system is modeled considering different hypotheses of multi-fault ruptures at the whole fault system scale and weighted in a logic tree. We compare the hazard levels with those from an area source PSHA and a previous fault-based approach. The results show a clear control of the EBSZ faults in the seismic hazard for all return periods, increasing drastically the hazard levels in the regions close to the fault traces and influencing up to 20 km farther with respect to the area source PSHA. The seismic hazard is dependent on the fault slip rates as peak ground accelerations and territorial extension of the fault influence appear higher around the Alhama de Murcia and Carboneras faults, while lower slip rate faults (Palomares Fault) show minor contribution to the hazard. For the return period of 475 years and near-fault locations, our models are more consistent with the ground motion values reached in the 2011 Mw 5.2 Lorca event than the building code or national seismic hazard map, which suggest that our fault system-based model performs more accurate estimations for this return period. Fault data, mainly slip rates, and its uncertainties have a clear impact on the seismic hazard and, for some faults, the lack of detailed paleoseismic studies can compromise the reliability of the hazard estimations. This, together with epistemic uncertainties concerning the background seismicity, are key discussion points in the present study, having an impact on further research and aiming to serve as a case example for other low-to-moderate seismicity regions worldwide.

show abstract

Physical and Statistical Behavior of Multifault Earthquakes: Darfield Earthquake Case Study, New Zealand

Cited by 21 publications

References 110 publications

Can Deep Learning Predict Complete Ruptures in Numerical Megathrust Faults?

Can Deep Learning Predict Complete Ruptures in Numerical Megathrust Faults?

Three‐Dimensional Structure, Ground Rupture Hazards, and Static Stress Models for Complex Nonplanar Thrust Faults in the Ventura Basin, Southern California

Fault System-Based Probabilistic Seismic Hazard Assessment of a Moderate Seismicity Region: The Eastern Betics Shear Zone (SE Spain)

Contact Info

Product

Resources

About