Spatio-temporal foreshock evolution of the 2019 M 6.4 and M 7.1 Ridgecrest, California earthquakes

https://doi.org/10.1016/j.epsl.2020.116582Get rights and content

Highlights

  • High-resolution foreshock sequences are obtained for both M 6.4 and M 7.1 events.

  • The NW-SE foreshock alignment implies M 6.4 rupture begins on an NW trending fault.

  • The M 7.1 nucleation zone is featured with foreshock concentration and acceleration.

  • The M 7.1 nucleation zone has a very low b value, implying a critical stress state.

Abstract

The 2019 M 6.4 and M 7.1 Ridgecrest, California earthquake sequence provides an ideal opportunity to study the seismicity evolution and interaction among multiple complex fault structures. Here, we apply the matched-filter detection method to obtain a relatively complete (magnitude of completeness ≈ 0.9) and precisely relocated earthquake catalog. The results show a short-duration (∼ 31 minutes) foreshock sequence with 28 events, before the M 6.4 earthquake. The foreshock sequence started with a M 4.0 event and was aligned along the NW-SE direction. This implies that the M 6.4 rupture initiated on a NW trending fault segment, before rupturing the primary SW trending fault. Repeating earthquakes before and after the M 7.1 event are separated in space and bound the areas of large coseismic slip in the M 6.4 and M 7.1 events. This might reflect local slow slip acceleration near the edges of coseismic rupture asperities. The NW-striking fault zones illuminated by seismicity are separated into several sub-regions with distinct pre-M7.1 seismicity rate evolutions. The M 7.1 event nucleated in a region of local seismicity concentration which intensified ∼ 3 hr before the M 7.1 mainshock. The M 7.1 nucleation zone is characterized by a significantly low b value of events that occurred since the M 6.4 event, which might indicate local failure conditions approaching a critical state.

Introduction

On 5 July 2019, the Mw 7.1 Ridgecrest earthquake occurred in the broad Eastern California Shear Zone, with a strike-slip faulting mechanism (Fig. 1a). The M 7.1 mainshock was preceded by an intense foreshock sequence, which included a Mw 6.4 event that occurred ∼ 34 hours earlier. The M 6.4 event was followed by abundant aftershocks, both along SW- and NW-striking faults (Fig. 1a). The M 6.4 event ruptured both SW- and NW- trending fault planes, with the primary slip along the SW-trending fault (Figs. 1b and 1c). In contrast, aftershocks of the M 7.1 event were mainly distributed along a primarily NW striking fault zone, and along at least 20 orthogonal faults cutting across the main fault (Ross et al., 2019). The multiple fault segments involved in the M 6.4 and M 7.1 events provide a rare opportunity to investigate how earthquake sequences evolve in complex fault systems, which is important to improve our understanding of fault interactions, earthquake triggering, and evolving earthquake hazard during an ongoing earthquake sequence.

Small earthquakes are important to illuminate the fault structure as well as the evolution of an earthquake sequence. In addition, small events are also key to understanding the preparation and nucleation process of large earthquakes (e.g., Bouchon et al., 2011, Kato et al., 2012). However, due to the low signal-to-noise ratio and/or overlapping waveforms, the routine earthquake catalogs usually miss a significant portion of small events. The matched-filter detection method is based on exploiting the waveform similarity of events occurring at similar locations (Gibbons and Ringdal, 2006). It has been widely used to detect low-frequency earthquakes (e.g., Shelly et al., 2007) as well as to obtain more complete records of foreshock or aftershock sequences down to low magnitudes (e.g., Peng and Zhao, 2009, Kato et al., 2012, Huang et al., 2017).

Foreshock sequences are inferred to be manifestations of mainshock nucleation processes and have been observed to precede a number of recent large earthquakes (e.g., Bouchon et al., 2013, Ellsworth and Bulut, 2018, Kato et al., 2012, Socquet et al., 2017). The foreshock sequences may be driven by static stress transfer between consecutive events (cascade model) or background aseismic slip (preslip model). Various physical mechanisms can contribute to advancing the evolution towards rupture nucleation along fault systems. Static stress changes induced by previous earthquake ruptures can permanently change the Coulomb stress in nearby regions and promote the occurrence of earthquakes at locations of positive stress change. On the other hand, dynamic stresses can induce temporary stress perturbations during the passage of body and/or surface waves (Hill and Prejean, 2015). In addition, other physical processes such as aseismic slip or fluid diffusion can also progressively alter the stress field and/or frictional properties of the fault to trigger seismicity (e.g., Shelly et al., 2016). It is an ongoing debate whether distinctive features exist and allow foreshocks recognizable as precursory phenomena prior to eventual mainshocks (e.g., Shearer, 2012; Ogata and Katsura, 2014; Seif et al., 2018).

From an earthquake catalog built from matched-filter detections, Ross et al. (2019) document a prominent foreshock sequence before the M 6.4 event. In addition, they suggest that seismic activity on the part of the NW-striking fault separating the M 6.4 rupture intersection from the M 7.1 hypocenter, eroded away barriers to slip during the 34 hour delay between the M 6.4 and M 7.1 events. In this study, we utilize the matched-filter method to detect and relocate events (including mainshock hypocenters) in a similar time period but focus on the detailed evolution of small events relative to the hypocenters of the M 6.4 and M 7.1 events. We identify repeating earthquakes from the waveform analysis, which are possible indicators of aseismic slip. In addition, we explore the b-value variations from the magnitude-frequency statistics, which may indicate variations in faulting style, fault zone complexity or stress conditions along the fault zone (e.g., Petruccelli et al., 2019 and references therein). Our results shed new light on the nucleation processes of the M 6.4 and M 7.1 events.

Section snippets

Data and methods

We collect continuous seismograms from 1 January to 1 August 2019, recorded by 35 local broadband and short-period stations (Figure S1). We select 25,227 template events from 1 January to 8 September 2019 listed in the Southern California Earthquake Data Center (SCEDC) catalog, with the requirement that both horizontal and vertical uncertainties are smaller than 1 km. The data processing steps generally follow previous studies (e.g., Peng and Zhao, 2009, Huang and Meng, 2018). Template

Results

We detect a total of 84,873 new events (Table S1) from 1 January to 1 August 2019, ∼ 4.9 times the number of events listed in the SCEDC catalog. In the new catalog which combines the SCEDC catalog with the newly detected events, 67,834 events (∼ 66%) are relocated. In addition, both the M 6.4 and M 7.1 events are successfully relocated, because of the sufficiently high similarity of initial P waveforms with nearby events. Among the new catalog events, we detect 513 repeating earthquakes (249

Discussion and conclusion

In this study, we apply the matched-filter detection to obtain a complete (Mc = 0.9) and accurate earthquake catalog during the 2019 M 6.4 and M 7.1 Ridgecrest earthquake sequence. Our results reveal aligned foreshock sequences concentrated near the hypocenters of both the M 6.4 and M 7.1 events, which are inferred to be related to the nucleation process of the respective mainshocks. We discuss the spatio-temporal evolution of the foreshock sequences and their implications for the nucleation

CRediT authorship contribution statement

Hui Huang: Conceptualization, Formal analysis, Methodology, Software, Writing - original draft, Writing - review & editing. Lingsen Meng: Conceptualization, Funding acquisition, Writing - review & editing. Roland Bürgmann: Conceptualization, Supervision, Writing - review & editing. Wei Wang: Formal analysis, Software, Writing - review & editing. Kang Wang: Formal analysis, Writing - review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

We thank Jean-Philippe Avouac and two anonymous reviewers for their constructive reviews, which significantly improved this paper. We thank David Shelly and Zachary Ross for valuable discussions. The Southern California Earthquake Data Center (scedc.caltech.edu, SCEDC (2013)) provided access to seismograms used in this study. Quaternary fault traces in California are retrieved from USGS. This work relies on the computational and storage services associated with the Hoffman2 Shared Cluster

References (60)

  • M. Bouchon et al.

    The long precursory phase of most large interplate earthquakes

    Nat. Geosci.

    (2013)
  • E.E. Brodsky

    Determining whether the worst earthquake has passed

    Nature

    (2019)
  • K. Chen et al.

    Cascading and pulse-like ruptures during the 2019 Ridgecrest earthquakes in the Eastern California Shear Zone

    Nat. Commun.

    (2020)
  • X. Chen et al.

    Analysis of foreshock sequences in California and implications for earthquake triggering

    Pure Appl. Geophys.

    (2016)
  • K. Dascher-Cousineau et al.

    Two foreshock sequences Post Gulia and Wiemer (2019)

    Seismol. Res. Lett.

    (2020)
  • D.A. Dodge et al.

    Detailed observations of California foreshock sequences: implications for the earthquake initiation process

    J. Geophys. Res.

    (1996)
  • P. Dublanchet

    The dynamics of earthquake precursors controlled by effective friction

    Geophys. J. Int.

    (2018)
  • W.L. Ellsworth et al.

    Seismic evidence for an earthquake nucleation phase

    Science

    (1995)
  • W.L. Ellsworth et al.

    Nucleation of the 1999 Izmit earthquake by a triggered cascade of foreshocks

    Nat. Geosci.

    (2018)
  • S.J. Gibbons et al.

    The detection of low magnitude seismic events using array-based waveform correlation

    Geophys. J. Int.

    (2006)
  • L. Gulia et al.

    Real-time discrimination of earthquake foreshocks and aftershocks

    Nature

    (2019)
  • L. Gulia et al.

    Pseudoprospective evaluation of the foreshock traffic-light system in Ridgecrest and implications for aftershock hazard assessment

    Seismol. Res. Lett.

    (2020)
  • H. Huang et al.

    Slow unlocking processes preceding the 2015 Mw 8.4 Illapel, Chile, earthquake

    Geophys. Res. Lett.

    (2018)
  • K. Hutton et al.

    Earthquake monitoring in southern California for seventy-seven years (1932–2008)

    Bull. Seismol. Soc. Am.

    (2010)
  • T. Igarashi et al.

    Repeating earthquakes and interplate aseismic slip in the northeastern Japan subduction zone

    J. Geophys. Res., Solid Earth

    (2003)
  • A. Kato et al.

    Propagation of slow slip leading up to the 2011 Mw 9.0 Tohoku-Oki earthquake

    Science

    (2012)
  • O. Lengliné et al.

    Inferring the coseismic and postseismic stress changes caused by the 2004 Mw = 6 Parkfield earthquake from variations of recurrence times of microearthquakes

    J. Geophys. Res.

    (2009)
  • G. Lin

    Waveform cross-correlation relocation and focal mechanisms for the 2019 Ridgecrest earthquake sequence

    Seismol. Res. Lett.

    (2020)
  • C. Liu et al.

    Coseismic rupture process of the large 2019 Ridgecrest earthquakes from joint inversion of geodetic and seismological observations

    Geophys. Res. Lett.

    (2019)
  • A. Lomax

    Absolute location of 2019 Ridgecrest seismicity reveals a shallow Mw 7.1 hypocenter, migrating and pulsing Mw 7.1 foreshocks, and duplex Mw 6.4 ruptures

    Bull. Seismol. Soc. Am.

    (2020)
  • Cited by (38)

    • Investigation of the 2013 Eryuan, Yunnan, China M<inf>S</inf> 5.5 Earthquake Sequence: Aftershock Migration, Seismogenic Structure and Hazard Implication

      2022, Tectonophysics
      Citation Excerpt :

      In our study, the earthquakes following the 2013 Eryuan MS 5.5 mainshock show bilateral expansion along their predominant distribution (Figs. 3), which is a typical behavior of afterslip-driven aftershock sequences (Huang et al., 2017; Liu et al., 2019; Peng and Zhao, 2009; Yao et al., 2017). Besides, we use strict criteria to identify six repeating earthquake pairs (Fig. 5), implying small-scale local aseismic transients (Huang et al., 2020). These repeating earthquakes have the same expansion behavior as other aftershocks (Fig. 6), which suggests that the aftershock expansion of the 2013 Eryuan MS 5.5 earthquake may be driven by afterslip (Yao et al., 2017).

    • The cascading foreshock sequence of the Ms 6.4 Yangbi Earthquake in Yunnan, China

      2022, Earth and Planetary Science Letters
      Citation Excerpt :

      Foreshocks have been considered promising precursors to examining how earthquakes nucleate and grow (Bouchon et al., 2013; Ellsworth and Bulut, 2018; Huang et al., 2020; Kato and Ben-Zion, 2020; Kato et al., 2012).

    View all citing articles on Scopus
    View full text