Elsevier

Tectonophysics

Volume 670, 22 February 2016, Pages 1-15
Tectonophysics

Review Article
Teleseismic constraints on the geological environment of deep episodic slow earthquakes in subduction zone forearcs: A review

https://doi.org/10.1016/j.tecto.2016.01.005Get rights and content

Abstract

More than a decade after the discovery of deep episodic slow slip and tremor, or slow earthquakes, at subduction zones, much research has been carried out to investigate the structural and seismic properties of the environment in which they occur. Slow earthquakes generally occur on the megathrust fault some distance downdip of the great earthquake seismogenic zone in the vicinity of the mantle wedge corner, where three major structural elements are in contact: the subducting oceanic crust, the overriding forearc crust and the continental mantle. In this region, thermo-petrological models predict significant fluid production from the dehydrating oceanic crust and mantle due to prograde metamorphic reactions, and their consumption by hydrating the mantle wedge. These fluids are expected to affect the dynamic stability of the megathrust fault and enable slow slip by increasing pore-fluid pressure and/or reducing friction in fault gouges. Resolving the fine-scale structure of the deep megathrust fault and the in situ distribution of fluids where slow earthquakes occur is challenging, and most advances have been made using teleseismic scattering techniques (e.g., receiver functions). In this paper we review the teleseismic structure of six well-studied subduction zones (three hot, i.e., Cascadia, southwest Japan, central Mexico, and three cool, i.e., Costa Rica, Alaska, and Hikurangi) that exhibit slow earthquake processes and discuss the evidence of structural and geological controls on the slow earthquake behavior. We conclude that changes in the mechanical properties of geological materials downdip of the seismogenic zone play a dominant role in controlling slow earthquake behavior, and that near-lithostatic pore-fluid pressures near the megathrust fault may be a necessary but insufficient condition for their occurrence.

Introduction

Subduction zone great earthquakes (with moment magnitude Mw > 8) are generated by rupture of the megathrust fault within the so-called “seismogenic zone”. There are many ways of determining the seismogenic zone based on a variety of indicators. In a static association with physical properties, the seismogenic zone is the portion of the fault that behaves in a brittle fashion. The depth to a specific geotherm commonly dictates the extent of the zone beyond which rocks deform by thermally activated creep processes (e.g., Handy et al., 2007). The strength of the brittle zone depends on the coefficient of friction on the fault and the pore-fluid pressure around the fault zone. Dynamically, the seismogenic zone is associated with the “locked” portion of the plate boundary fault as constrained by GPS and other geodetic data and coincides with the unstable slip region of fault rupture under rate and state friction (e.g., Lay et al., 2012). The extent of the seismogenic zone and the transition from unstable to stable slip depends on material properties and pore-fluid pressure around the fault surface (e.g., Scholz, 1998). Knowledge of the extent, geometry and material properties of the megathrust fault zone is therefore crucial to constrain the static and dynamic conditions under which large potentially damaging subduction zone earthquakes occur.

The megathrust fault remains largely inaccessible to direct sampling, except for exhumed ancient subduction thrust faults (e.g., Meneghini, F., et al., 2010, Angiboust, S., et al., 2015) and rare instances of direct drilling into the upper part of the subduction thrust (e.g., Chester et al., 2013). We therefore generally rely on indirect measurements to constrain its structure and physical properties via remote geophysical methods. Recent large-scale, high-density seismic array data have been very successfully in imaging the seismic structure of the subduction zone forearc, such as the megathrust fault zone, oceanic Moho, and upper plate Moho using receiver functions. Detailed images from three-dimensional regional seismic tomography and interface reflection structure models also provide additional constraints on subduction zone properties.

In recent years our view of the seismogenic zone has been improved by the discovery of widespread episodic slow slip events that occur downdip of the seismogenic zone. Slow slip events (sometimes referred to as slow earthquakes) had been recognized for some time in the literature using strainmeters (Linde et al., 1996) and seismometers (e.g., Kanamori and Stewart, 1979), although their importance and locations only became apparent when networks of GPS stations started recording and documenting their widespread occurrence. Associated with slow slip, a new source of seismic energy (called non-volcanic tremor) in the forearc was discovered by Obara (2002) off southwest Japan, which appeared as coherent noise propagating across arrays of seismograph stations. Rogers and Dragert (2003) then found similar signals in the forearc of the Cascadia subduction zone that occurred concurrently with slow slip events and both phenomena recurred episodically, which led to the term “Episodic Tremor and Slip”, or ETS. Although initial studies focused on subduction zones where young and warm plates are being subducted beneath a continental margin, episodic slow slip and/or tremor events were later recognized in various other subduction zones, in the shallow part of the megathrust fault near the trench (Saffer and Wallace, 2015, and references therein) as well as in the deeper parts of strike-slip faults (e.g., Nadeau and Dolenc, 2005). In this paper we refer to episodic slow fault rupture events (with either or both documented slow slip and tremor) as slow earthquakes, and focus on the main deep subduction zone slow earthquakes, i.e. those that happen at depths of 30 to 45 km usually along a large extent of the subduction zone.

These slow earthquakes display important characteristics that provide a deeper understanding of fault zone properties and dynamics. Slow slip events and tremor can be correlated both temporally and spatially (e.g., Cascadia and Japan, Bartlow, N.M., et al., 2011, Hirose and Obara, 2010), or temporally only (e.g., Hikurangi, Yabe et al., 2014). It is possible to have observed slow slip without detectable tremor, and perhaps observed tremor without detectable slow slip (however, see Frank et al., 2015). Both deep slow slip and tremor generally occur near but downdip of the brittle–ductile transition, approximately coinciding with the transition from unstable to stable sliding in a region of conditional stability (e.g., Scholz, 1998; Fig. 1), although there may be a gap between the slow earthquake source region and the seismogenic zone in some cases (Hyndman et al., 2015). Their occurrence may be induced by fluid flow and fluid processes at the plate interface and within the overlying plate (Rubinstein et al., 2010, and references therein). Low frequency earthquakes (LFEs) have also been observed in coincidence with the slow slip and tremor in subduction zones, with focal mechanism and location consistent with interplate slip (Shelly, D.R., et al., 2006, Shelly, D.R., et al., 2007). Recent seismic and numerical modeling results point to the contribution of elevated fluid pressure near the plate interface (Kodaira, S., et al., 2004, Liu, Y. and Rice, J.R., 2007, Audet, P., et al., 2009, Song, T.A., et al., 2009). These findings are consistent with thermo-petrological models that predict significant fluid production in the vicinity of the slow earthquake source region from dehydration of the subducting oceanic crust (Hyndman and Peacock, 2003). Nevertheless, the variety of thermal and petrologic conditions across different subduction zones precludes a simple relation between slow earthquakes and the thermal state or dehydration stages in the subducting plate in which they occur (Peacock, 2009).

Several reviews on the observations of slow earthquakes have been published recently (Beroza, G.C. and Ide, S., 2011, Gomberg, J., et al., 2010, Rubinstein, J.L., et al., 2010, Schwartz and Rokowski, 2007). In this review we focus on the structural and geological environment in which slow earthquakes occur as inferred primarily from teleseismic studies. In particular we examine structural properties of the slow earthquake source region at six well-studied subduction zones: Cascadia, Nankai, Mexico, Costa Rica, Alaska, and Hikurangi (Fig. 2). The first three are considered as “hot” subduction zones, i.e., for which the seismogenic zone is thermally controlled downdip before the mantle wedge corner, and the other three are cooler. We first define the forearc structural elements and describe their seismic properties based on thermo-petrological models of metamorphism. We then focus on subduction zone structure inferred from teleseismic scattering techniques and high-resolution tomographic studies for the six subduction zones. Lastly we discuss the controls that structure may have on slow earthquake behavior.

Section snippets

Seismic structure of the subduction zone forearc

Slow earthquakes in subduction zones generally occur at and near the interface between the subducting slab and the overlying crust and mantle wedge (Fig. 1). The slab-mantle interface downdip of the forearc mantle corner is composed of some combination of heterogeneously deformed oceanic crust and sedimentary cover juxtaposed with hydrated and metasomatized materials. Recent receiver function studies of the slab-mantle interface have identified zones of low seismic velocity and high P-to-S

Cascadia

The Cascadia subduction zone extends from northern California in the south to northern Vancouver Island in the north where the Juan de Fuca plate subducts beneath the North American plate; here we mainly focus on its northern extent (Fig. 5a). The forearc of the Cascadia subduction zone has been extensively studied from active and passive source surveys (e.g., Green, A.G., et al., 1986, Clowes, R.M., et al., 1987, Soyer, W. and Unsworth, M., 2006). Bostock (2013) provides a recent review of the

Structural controls on deep slow earthquake occurrence

There are at least two important features relating deep slow earthquake occurrence and the geologic structure across all subduction zones examined above. First, tremor and LFEs appear to occur near the top of the LVZ, whereas regular Wadati–Benioff seismicity is mostly restricted to the lower oceanic crust and upper mantle. Although tremor was originally thought to reflect hydrofracturing within the overlying continental crust (Kao et al., 2005), source mechanisms of repeating LFEs during

Conclusions and future directions

Teleseismic scattering images of the various structural elements in subduction zone forearcs obtained from teleseismic scattering techniques provide constraints on the thermo-petrological environment of the deep extension of the megathrust fault. These images identify the distribution of inferred pore-fluid pressure and hydrated mineral assemblages and their potential effects on the stability of the megathrust fault in the slow earthquake source region. Fluids appear to facilitate slow slip

Acknowledgments

P. Audet is funded by the Natural Sciences and Engineering Research Council of Canada through grant RGPIN-418288-2012. Y. Kim is funded by the National Research Foundation of Korea Grant funded by the Korean Government (NRF-2014S1A2A2027609), and by the Korea Meteorological Administration Research and Development Program under Grant KMIPA2015-7020. The authors thank S. Bannister, M. Bostock, H. DeShon, W. Frank, A. Kato, and S. Ide for the data used in figures, and two anonymous reviewers for

References (159)

  • C.E. Manning

    The solubility of quartz in H2O in the lower crust and upper mantle

    Geochim. Cosmochim. Acta

    (1994)
  • G.A. Abers et al.

    Imaging the source region of Cascadia tremor and intermediate depth earthquakes

    Geology

    (2009)
  • S. Angiboust et al.

    Probing the transition between seismically coupled and decoupled segments along an ancient subduction interface

    Geochem. Geophys. Geosyst.

    (2015)
  • J.G. Armbruster et al.

    Accurate tremor locations from coherent S and P waves

    J. Geophys. Res.

    (2014)
  • P. Audet

    Receiver functions using OBS data: Promises and limitations from numerical modelling and examples from the Cascadia Initiative

    Geophys. J. Int.

    (2015)
  • P. Audet et al.

    Possible control of subduction zone slow earthquake periodicity by silica enrichment

    Nature

    (2014)
  • P. Audet et al.

    Hydrologic control on forearc strength and seismicity in the Costa Rican subduction zone

    Nat. Geosci.

    (2013)
  • P. Audet et al.

    Slab morphology in the Cascadia forearc and its relation to episodic tremor and slip

    J. Geophys. Res.

    (2010)
  • P. Audet et al.

    Seismic evidence for overpressured subducted oceanic crust and megathrust fault sealing

    Nature

    (2009)
  • S. Bannister et al.

    Imaging the Hikurangi subduction zone, New Zealand, using teleseismic receiver functions: crustal fluids above the forearc mantle wedge

    Geophys. J. Int.

    (2007)
  • N.M. Bartlow et al.

    Space–time correlation of slip and tremor during the 2009 Cascadia slow slip event

    Geophys. Res. Lett.

    (2011)
  • G.C. Beroza et al.

    Slow earthquakes and nonvolcanic tremor

    Annu. Rev. Earth Planet. Sci.

    (2011)
  • P. Bird

    An updated digital model of plate boundaries

    Geochem. Geophys. Geosyst.

    (2003)
  • M.G. Bostock et al.

    An inverted continental Moho and serpentinization of the forearc mantle

    Nature

    (2002)
  • M.G. Bostock et al.

    Low frequency earthquakes below southern Vancouver Island

    Geochem. Geophys. Geosyst.

    (2012)
  • N. Brantut et al.

    Effect of dehydration reactions on earthquake nucleation: stable sliding, slow transients, and unstable slip

    J. Geophys. Res.

    (2011)
  • J.R. Brown et al.

    Deep low frequency earthquakes in tremor localize to the plate interface in multiple subduction zones

    Geophys. Res. Lett.

    (2009)
  • J.R. Brown et al.

    Deep low-frequency earthquakes in tectonic tremor along the Alaska–Aleutian subduction zone

    J. Geophys. Res.

    (2013)
  • M.R. Brudzinski et al.

    Segmentation in episodic tremor and slip all along Cascadia

    Geology

    (2007)
  • J.F. Cassidy et al.

    S-wave velocity structure of the northern Cascadia subduction zone

    J. Geophys. Res.

    (1993)
  • F.M. Chester et al.

    Structure and composition of the plate-boundary slip-zone for the 2011 Tohoku-oki earthquake

    Science

    (2013)
  • N.I. Christensen

    Pore pressure and oceanic crustal seismic structure

    Geophys. J. R. Astron. Soc.

    (1984)
  • N.I. Christensen

    Poisson's ratio and crustal seismology

    J. Geophys. Res.

    (1996)
  • R.M. Clowes et al.

    LITHOPROBE southern Vancouver Island: Cenozoic subduction complex imaged by deep seismic reflections

    Can. J. Earth Sci.

    (1987)
  • H.R. DeShon et al.

    Evidence for serpentinization of the forearc mantle wedge along the Nicoya Peninsula, Costa Rica

    Geophys. Res. Lett.

    (2004)
  • H.R. DeShon et al.

    Seismogenic zone structure beneath the Nicoya Peninsula, Costa Rica, from three-dimensional local earthquake P- and S-wave tomography

    Geophys. J. Int.

    (2006)
  • H. Dragert et al.

    A silent slip event on the deeper Cascadia subduction interface

    Science

    (2001)
  • D. Eberhart-Phillips et al.

    Imaging the Hikurangi Plate interface region, with improved local-earthquake tomography

    Geophys. J. Int.

    (2012)
  • M. Faccenda et al.

    Fault-induced seismic anisotropy by hydration in subducting oceanic plates

    Nature

    (2008)
  • A. Fagereng et al.

    Non-volcanic tremor and discontinuous slab dehydration

    Geophys. Res. Lett.

    (2011)
  • A. Fagereng et al.

    Shear veins observed within anisotropic fabric at high angles to the maximum compressive stress

    Nat. Geosci.

    (2010)
  • W.B. Frank et al.

    Uncovering the geodetic signature of silent slip through repeating earthquakes

    Geophys. Res. Lett.

    (2015)
  • W.B. Frank et al.

    Using systematically characterized low-frequency earthquakes as a fault probe in Guerrero, Mexico

    J. Geophys. Res.

    (2014)
  • W.B. Frank et al.

    Low-frequency earthquakes in the Mexican sweet spot

    Geophys. Res. Lett.

    (2013)
  • B. Fry et al.

    Deep tremor in New Zealand triggered by the 2010 Mw 8.8 Chile earthquake

    Geophys. Res. Lett.

    (2011)
  • S.B. Giger et al.

    Permeability evolution in quartz gouges under hydrothermal conditions

    J. Geophys. Res.

    (2007)
  • J. Gomberg et al.

    Slow-slip phenomena in Cascadia from 2007 and beyond: a review

    Geol. Soc. Am. Bull.

    (2010)
  • A.G. Green et al.

    Seismic reflection imaging of the Juan de Fuca plate

    Nature

    (1986)
  • R.N. Harris et al.

    Thermal regime of the Costa Rican convergent margin: 2. Thermal models of the shallow Middle America subduction zone offshore Costa Rica

    Geochem. Geophys. Geosyst.

    (2010)
  • Cited by (86)

    View all citing articles on Scopus
    View full text