• Media type: E-Article
  • Title: Influence of Subduction Zone Dynamics on Interface Shear Stress and Potential Relationship With Seismogenic Behavior
  • Contributor: Beall, Adam; Fagereng, Åke; Davies, J. Huw; Garel, Fanny; Davies, D. Rhodri
  • imprint: American Geophysical Union (AGU), 2021
  • Published in: Geochemistry, Geophysics, Geosystems
  • Language: English
  • DOI: 10.1029/2020gc009267
  • ISSN: 1525-2027
  • Keywords: Geochemistry and Petrology ; Geophysics
  • Origination:
  • Footnote:
  • Description: <jats:title>Abstract</jats:title><jats:p>Whether tectonic convergence at subduction zones is accommodated predominantly through seismic or aseismic deformation, the former potentially generating large earthquakes, varies considerably between subduction margins. This margin‐scale variability has previously been linked to overriding plate deformation, trench migration, and their influence on the plate interface stress state. While these processes are linked to mantle‐scale dynamics, it is unclear how such dynamics influence interface stress. We systematically analyze the interface stress state in a suite of 2‐D thermo‐mechanical subduction models, where slabs display a range of morphologies that arise from diverse multiscale interactions with adjacent mantle and the overriding plate. We demonstrate that the thickness of the interface layer varies dynamically, in response to Poiseuille flow induced by slab bending or unbending, leading to associated effects on interface shear stress at typical seismogenic depth. Lower shear stress occurs when slab unbending is significant, which is commonly associated with trench retreat and draping of the slab as it impinges on the higher‐viscosity lower‐mantle. Conversely, higher shear stress is associated with limited slab unbending, which is promoted by negligible trench migration and vertically subducting slabs. We conclude that the diversity of slab dynamics may cause large variations in interface stress state between and maybe within margins. This is an additional variable that potentially controls seismogenic behavior, and we compare broad stress estimates for Circum‐Pacific margins to previous studies. Although predicted shear stress varies with observed seismogenic behavior, more detailed constraints on stress state are needed to test for correlation.</jats:p>