Subcretionary tectonics: Linking variability in the expression of subduction along the Cascadia forearc

https://doi.org/10.1016/j.epsl.2020.116724Get rights and content

Highlights

  • Creation of 3D velocity-corrected discontinuity model for the Cascadia margin.

  • Forearc crustal thickness and seismic structure highlights zones of subcretion.

  • Slab morphology, fluid release, and subcretion contribute to significant margin variations.

  • Subcretion is the primary control on Cascadia forearc differential topography.

Abstract

A number of seismic and other geophysical phenomena exhibit significant lateral heterogeneity along the strike of the Cascadia forearc. Both the overriding and downgoing plate have been invoked to play the dominant role in controlling along-strike correlations between seismogenic behavior, potential field measurements, morphological/tectonic characteristics, and seismic structure; however, significant feedbacks likely exist between the two. In this study, we apply a 3D velocity correction to receiver function data and interpret the resulting discontinuity model alongside a recently published shear-wave velocity model to understand the possible causative relationships between correlative along-strike variations. Our discontinuity model indicates that the forearc crust thickens as it approaches the mantle wedge corner before progressively thinning toward the magmatic arc, likely due to the basal accretion of material from the downgoing plate to the overriding plate. In the northern and southern portions of the forearc, this “subcreted” material is characterized by thick (∼10 km) anomalously low shear-wave velocity zones. The thickness, high internal reflectivity, and low Bouguer gravity signatures associated with the low velocity zones likely indicate that this subcreted material is composed of dominantly (meta)sedimentary material that has been emplaced through successive subcretion events over geologic timescales. Furthermore, the anomalously low velocities and spatial correlation with high non-volcanic tremor (NVT) density and short slow slip recurrence intervals indicate that these regions are fluid-rich. While first-order variations in the fluids that control NVT and slow slip likely result from differences in the permeability of the downgoing slab as inferred from its stress state and the distribution of intraslab seismicity, these subcreted packages likely represent thick, vertically-impermeable regions in the lower crust that further accentuate this correlation. Variability in the amount of subcretion explains patterns of exhumation and uplift along the Cascadia margin and the resulting forearc topography over geologic timescales, and is likely controlled by some combination plate interface geometry/rheology and overriding plate architecture.

Introduction

The modern Cascadia margin is a ∼1200 km long subduction zone along the northwestern coast of North America that has been operating since the accretion of the Siletzia terrane at ∼50 Ma (Wells et al., 2014). This margin is characterized by the subduction of the young oceanic lithosphere of the Juan de Fuca plate (5–10 Myr at the trench) at slow-to-intermediate convergence rates (between 30 and 42 mm/yr) in a northeasterly direction (Fig. 1) (Wilson, 2002). The combination of the young oceanic lithosphere and low variations in convergence rates makes Cascadia one of the endmember examples of a warm subduction system (van Keken et al., 2011). This results in a significant amount of fluid release from the subducting lithosphere starting at relatively shallow depths (∼40 km; Condit et al., 2020), which will affect the seismogenic and rheological characteristics of the plate interface. While significant variations in fluid-mediated subduction processes along the Cascadia forearc are observed, as seen in variable slow-slip recurrence intervals (Brudzinski and Allen, 2007), non-volcanic tremor (NVT) density, and intraslab seismicity (McCrory et al., 2012), the predicted depth of dehydration appears relatively invariant along-strike based on first-order characteristics of the subducting slab (Condit et al., 2020), making this an unlikely cause of the variation in seismogenic behavior. Other observations also spatially correlate with the seismogenic behavior of the margin, which may lend insight into controls on this variation. These include crustal seismicity (Bostock et al., 2019), gravity and magnetic fields (Blakely et al., 2005), uplift and erosion rates (Balco et al., 2013; Burgette et al., 2009), topography, slab morphology (McCrory et al., 2012), intraslab deviatoric stresses (Wada et al., 2010), and interface locking (Li et al., 2018; Schmalzle et al., 2014) (Fig. 1), and are also reflected in the seismic properties of the margin, such as forearc Rayleigh wave phase and crustal shear-wave velocities (Delph et al., 2018; Janiszewski et al., 2019; Porritt et al., 2011), seismic attenuation (Littel et al., 2018), and subslab mantle structure (Bodmer et al., 2018). The significant correlations between these different observations have proven difficult to link in a mechanistic way due to their different temporal and spatial scales. However, understanding these linkages is necessary to decipher the governing factors that lead to along-strike variations in subduction characteristics not only along the Cascadia forearc, but along other subduction margins as well.

Recent studies posit the dominant controlling mechanism behind the distribution of NVT in Cascadia is related to either the properties of the overriding or downgoing plate. For instance, NVT distribution has been related to overriding plate structure through variations in either: 1) the permeability of the overriding crust due to the presence of vertically extensive forearc faulting (Wells et al., 2017), or 2) the strength of the overriding plate implied from surface geology (Fig. 1A; Brudzinski and Allen, 2007). While these explanations likely contribute to distribution of NVT and could plausibly be related to variations in topography and Bouguer gravity along the margin, it is unclear how it would contribute to other phenomena that correlate with NVT, such as intraslab seismicity (Fig. 1B) and upper mantle velocity structure (Bodmer et al., 2018). Alternatively, buoyancy variations in the upper mantle below the subducting oceanic lithosphere have been invoked as the primary control on plate locking, and perhaps also uplift, and exhumation, by controlling slab curvature and thereby modulating changes in both along-strike and down-dip friction of the interface (Bodmer et al., 2018, 2020). This hypothesis could also be linked with intraslab seismicity, but its connection with variations in NVT and overriding crustal characteristics, such as Bouguer gravity variations, crustal seismic properties, or surface geology, are difficult to explain. Delph et al. (2018) attempted to link slab structure with NVT distribution by inferring that intraslab seismicity reflected permeability changes within the downgoing slab, leading to variable fluid release from the downgoing slab. The regions of high NVT density also coincided with thick low velocity zones interpreted as underplated sedimentary packages below the forearc, perhaps resulting from intrinsic vertical impermeability of the material. However, no controlling mechanism for the variable amounts of underplating was presented.

In this study, we provide plausible mechanistic linkages between the seismic structure of the Cascadia margin and other margin characteristics. We leverage the existing dense station distribution along with a recently published 3D shear-wave velocity model (Delph et al., 2018) and a new variable Vp/Vs ratio crustal model derived from local seismicity to convert P-wave receiver functions to depth. The resulting 3D velocity-corrected discontinuity model better delineates the seismic discontinuity structure of the Cascadia margin and allows for the systematic estimation of crustal thickness, which has proven difficult in previous studies due to the complexity of the crust-mantle transition in subduction zones (e.g., Bostock, 2013). This new 3D discontinuity model in combination with our previously computed 3D model of shear-wave velocities allows us to investigate how the seismic structure of the margin relates to lateral variations in subduction characteristics.

Section snippets

Methods

Receiver functions are commonly used to locate the crust-mantle transition and delineate the discontinuity structure beneath a region, as their peaks and troughs directly correspond to velocity increases and decreases in the Earth. However, delineating the crust-mantle transition in Cascadia has been challenging due to: 1) the difficulty of differentiating between the base of the overriding crust and the downgoing plate's crust in the forearc where they are in contact, as they are both

Seismic model comparisons and crustal thickness

The 3D corrected ACCP model highlights the discontinuity structure of the forearc to arc (Fig. 5B–D), and all depths in the model are relative to sea level. The main features include: 1) a strong, positive east-dipping arrival along the majority of the margin starting at about 20 km depth near the coast and extending to the base of the model, 2) a generally weak positive arrival between the forearc high topography and the volcanic arc, and 3) a relatively shallow (∼30–35 km depth), strong,

Forearc subcretion

The thickest overriding plate crust is generally found near the forearc mantle corner along the Cascadia margin, reaching up to ∼50 km and thinning in both the seaward and landward directions (Fig. 5A). This observation is consistent with the findings of smaller-scale studies (Ramachandran et al., 2006; Preston et al., 2003; Stanley et al., 1999; Calvert et al., 2006), and has been explained by basal accretion and structural duplexing within crustal material near the base of the overriding

Conclusion

Significant lateral variations in seismicity and other geophysical observations along the Cascadia forearc cannot be explained by variations in only the overriding or downgoing plate. To better understand how these variations are linked to the seismic structure of the margin, we applied a 3D variable Vp/Vs ratio velocity model correction to P-wave radial receiver functions to create a model of discontinuity structure for the Cascadia margin. When interpreted alongside a recently published

CRediT authorship contribution statement

Jonathan R. Delph: Conceptualization, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing. Amanda M. Thomas: Funding acquisition, Investigation, Methodology, Supervision, Writing – review & editing. Alan Levander: Writing – review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that influenced this research.

Acknowledgments

This research was funded by NSF #1848302. This manuscript benefitted from constructive reviews by Andrew Calvert and Rob Porritt as well as an informal review by Doug Toomey. The authors want to thank Tyler Newton for sharing tide gauge and GPS decadal uplift rate measurements, Miles Bodmer for sharing his tomographic model, Armel Menant for sharing his thermomechanical model, Shaoyang Li and Gina Schmalzle for the locking model predictions, and Michael Bostock and Alexandre Plourde for LFE P

References (86)

  • K. Liu et al.

    Asthenospheric flow and lithospheric evolution near the mendocino triple junction

    Earth Planet. Sci. Lett.

    (2012)
  • R.W. Porritt et al.

    Investigation of Cascadia segmentation with ambient noise tomography

    Earth Planet. Sci. Lett.

    (2011)
  • A.A. Royer et al.

    A comparative study of low frequency earthquake templates in northern Cascadia

    Earth Planet. Sci. Lett.

    (2014)
  • M. Saillard et al.

    Non-steady long-term uplift rates and Pleistocene marine terrace development along the Andean margin of Chile (31°S) inferred from 10Be dating

    Earth Planet. Sci. Lett.

    (2009)
  • G.A. Abers et al.

    The cold and relatively dry nature of mantle forearcs in subduction zones

    Nat. Geosci.

    (2017)
  • J.C. Arkle et al.

    Focused exhumation in the syntaxis of the western chugach mountains and prince William sound, Alaska

    Bull. Geol. Soc. Am.

    (2013)
  • P. Audet et al.

    Fluid pressure and shear zone development over the locked to slow slip region in Cascadia

    Sci. Adv.

    (2018)
  • P. Audet et al.

    Seismic evidence for overpressured subducted oceanic crust and megathrust fault sealing

    Nature

    (2009)
  • P. Audet et al.

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

    J. Geophys. Res., Solid Earth

    (2010)
  • G. Balco et al.

    Erosional response to northward-propagating crustal thickening in the coastal ranges of the U.S. Pacific northwest

    Am. J. Sci.

    (2013)
  • B.C. Beaudoin et al.

    Transition from slab to slabless: results from the 1993 Mendocino triple junction seismic experiment

    Geology

    (1996)
  • B.C. Beaudoin et al.

    Location of the southern edge of the Gorda slab and evidence for an adjacent asthenospheric window: results from seismic profiling and gravity

    J. Geophys. Res., Solid Earth

    (1998)
  • R.J. Blakely et al.

    Subduction-zone magnetic anomalies and implications for hydrated forearc mantle

    Geology

    (2005)
  • M. Bodmer et al.

    Seismic anisotropy beneath the Juan de Fuca plate system: evidence for heterogeneous mantle flow

    Geology

    (2015)
  • M. Bodmer et al.

    Buoyant asthenosphere Beneath Cascadia influences megathrust segmentation

    Geophys. Res. Lett.

    (2018)
  • M.G. Bostock

    Kirchhoff-approximate inversion of teleseismic wavefields

    Geophys. J. Int.

    (2002)
  • M.T. Brandon et al.

    Late Cenozoic exhumation of the Cascadia accretionary wedge in the Olympic Mountains, northwest Washington state

    Bull. Geol. Soc. Am.

    (1998)
  • T.M. Brocher et al.

    Seismic evidence for widespread serpentinized forearc upper mantle along the Cascadia margin

    Geology

    (2003)
  • M.R. Brudzinski et al.

    Segmentation in episodic tremor and slip all along Cascadia

    Geology

    (2007)
  • R.J. Burgette et al.

    Interseismic uplift rates for western Oregon and along-strike variation in locking on the Cascadia subduction zone

    J. Geophys. Res., Solid Earth

    (2009)
  • J.A. Calkins et al.

    Shallow structure of the Cascadia subduction zone beneath western Washington from spectral ambient noise correlation

    J. Geophys. Res., Solid Earth

    (2011)
  • A.J. Calvert et al.

    Sedimentary underplating at the Cascadia mantle-wedge corner revealed by seismic imaging

    Nat. Geosci.

    (2011)
  • A.J. Calvert et al.

    Cascadia low frequency earthquakes at the base of an overpressured subduction shear zone

    Nat. Commun.

    (2020)
  • J.P. Canales et al.

    Dry Juan de Fuca slab revealed by quantification of water entering Cascadia subduction zone

    Nat. Geosci.

    (2017)
  • N.I. Christensen

    Poisson's ratio and crustal seismology

    J. Geophys. Res.

    (1996)
  • N.I. Christensen

    Serpentinites, peridotites, and seismology

    Int. Geol. Rev.

    (2004)
  • R.M. Clowes et al.

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

    Can. J. Earth Sci.

    (1987)
  • K. Creager

    Collaborative Research: Illuminating the architecture of the greater Mount St. Helens magmatic systems from slab to surface

    International Federation of Digital Seismograph Networks

    (2014)
  • K.J. Crosbie et al.

    Shear velocity structure from ambient noise and teleseismic surface wave tomography in the Cascades around Mount St. Helens

    J. Geophys. Res., Solid Earth

    (2019)
  • J.R. Delph et al.

    A new approach to obtaining a 3D shear wave velocity model of the crust and upper mantle: an application to eastern Turkey

    Tectonophysics

    (2015)
  • J.R. Delph et al.

    Fluid controls on the heterogeneous seismic characteristics of the Cascadia margin

    Geophys. Res. Lett.

    (2018)
  • K.G. Dueker et al.

    Mantle discontinuity structure from midpoint stacks of converted P to S waves across the Yellowstone hotspot track

    J. Geophys. Res.

    (1997)
  • G.S. Fuis et al.

    Geologic interpretation of seismic-refraction results in Northeastern California

    Bull. Geol. Soc. Am.

    (1987)
  • Cited by (24)

    • Differential coastal uplift quantified by luminescence dating of marine terraces, central Cascadia forearc, Oregon

      2022, Quaternary Science Reviews
      Citation Excerpt :

      In comparison coastal geodetic uplift rates further south (∼41.5 °N – 43.5 °N) are on average ∼3 m/kyr (Fig. 8) (Burgette et al., 2009). Elevated geodetic uplift rates in southern Cascadia have been attributed to a variety of tectonic processes, including transitions in upper-plate strength (McKenzie et al., 2022), sediment underplating (Delph et al., 2021), and aestheospheric buoyancy (Bodmer et al., 2020). Similarly, long-term uplift rates from marine terrace platforms are slightly higher between 41°N and 43.5 °N (∼0.7 m/kyr) than father north from ∼43.5 °N to 45 °N (∼0.4 m/kyr) (Kelsey et al., 1994) (Fig. 8).

    • Subduction earthquake cycles controlled by episodic fluid pressure cycling

      2022, Lithos
      Citation Excerpt :

      Except for the left and right side of the model, the velocity boundary conditions resemble those of an analog sandbox model (Dal Zilio et al., 2020b): The top, left, and right boundaries have a free-slip boundary condition (zero shear stress), whereas the lower boundary has an imposed loading rate of 4 cm/yr. Several studies suggest that the megathrust permeability decreases with depth for a number of reasons: (1) volumetric expansion due to serpentinization (Wada et al., 2008), which reduces grain boundary connectivity and, in turn, permeability; (2) large deposits of silica (SiO2) from slab-derived fluids (Audet and Bürgmann, 2014); and (3) stacking of relict shear zones atop the active plate boundary (Delph et al., 2021). To account for such variability, in this study we test constant hydraulic properties and a depth-varying permeability on the fault zone (Fig. 1c).

    View all citing articles on Scopus
    View full text