The direction of tectonic plate motion at the Earth's surface and the flow field of the mantle inferred from seismic anisotropy are well correlated globally, suggesting large-scale coupling between the mantle and the surface plates. The fit is typically poor at subduction zones, however, where regional observations of seismic anisotropy suggest that the direction of mantle flow is not parallel to and may be several times faster than plate motions. Here we present three-dimensional numerical models of buoyancy-driven deformation with realistic slab geometry for the Alaska subduction-transform system and use them to determine the origin of this regional decoupling of flow. We find that near a subduction zone edge, mantle flow velocities can have magnitudes of more than ten times the surface plate motions, whereas surface plate velocities are consistent with plate motions and the complex mantle flow field is consistent with observations from seismic anisotropy. The seismic anisotropy observations constrain the shape of the eastern slab edge and require non-Newtonian mantle rheology. The incorporation of the non-Newtonian viscosity results in mantle viscosities of 10(17) to 10(18) Pa s in regions of high strain rate (10(-12) s(-1)), and this low viscosity enables the mantle flow field to decouple partially from the motion of the surface plates. These results imply local rapid transport of geochemical signatures through subduction zones and that the internal deformation of slabs decreases the slab-pull force available to drive subducting plates.

[1] Several models have been proposed to relate slab geometry to parameters such as plate velocity or plate age. However, studies on the observed relationships between slab geometry and a wide range of subduction parameters show that there is not a simple global relationship between slab geometry and any one of these other subduction parameters for all subduction zones. Numerical and laboratory models of subduction provide a method to explore the relative importance of different physical processes in determining subduction dynamics. Employing 2-D numerical models with a viscosity structure constrained by laboratory experiments for the deformation of olivine, we show that the observed range in slab dip and the observed trends between slab dip and convergence velocity, subducting plate age, and subduction duration can be reproduced without trench motion (i.e., slab roll-back) for locations away from slab edges. Successful models include a stiff slab that is 100-1000 times more viscous than previous estimates from models of plate bending, the geoid, and global plate motions. We find that slab dip in the upper mantle depends primarily on slab strength and plate boundary coupling, with a small dependence on subducting plate age. Once the slab sinks into the lower mantle the primary processes controlling slab evolution are (1) the ability of the stiff slab to transmit stresses up dip, (2) resistance to slab descent into the higherviscosity lower mantle, and (3) subduction-induced flow in the mantle-wedge corner.

No abstract

SUMMARY Our understanding of mantle convection and the motion of plates depends intimately on our understanding of the viscosity structure of the mantle. While geoid and gravity observations have provided fundamental constraints on the radial viscosity structure of the mantle, the influence of short‐wavelength variations in viscosity is still poorly understood. We present 2‐D and 3‐D finite‐element models of mantle flow, including strong lateral viscosity variations and local sources of buoyancy, owing to both thermal and compositional effects. We first use generic 2‐D models of a subduction zone to investigate how different observations depend on various aspects of the viscosity structure, in particular, the slab and lower‐mantle viscosity and the presence of a low‐viscosity region in the mantle wedge above the slab. We find that: (1) the strain rate provides a strong constraint on the absolute viscosity of the slab (1023 Pa s); (2) stress orientation within the slab is sensitive to the relative viscosity of the slab, lower mantle and the wedge; and (3) the stress state and topography of the overriding plate depend on the wedge viscosity and local sources of buoyancy. In particular, the state of stress in the overriding plate changes from compression to extension with the addition of a low‐viscosity wedge. We then use observations of strain rate, stress orientation, dynamic topography and the geoid for the Tonga–Kermadec subduction zone as simultaneous constraints on the viscosity and buoyancy in a 3‐D regional dynamic model. Together these observations are used to develop a self‐consistent model of the viscosity and buoyancy by taking advantage of the sensitivity of each observation to different aspects of the dynamics, over a broad range of length‐scales. The presence of a low‐viscosity wedge makes it possible to match observations of shallow dynamic topography and horizontal extension within the backarc, and down‐dip compression in the shallow portion of the slab. These results suggest that a low‐viscosity wedge plays an important role in controlling the presence of backarc spreading. However, for a model with a low‐viscosity and low‐density region that provides a good fit to the observed topography, we find that a reduction of the slab density by a factor of 1.3 relative to the reference density model, is required to match the observed geoid. These results suggest that compensation of the slab by dynamic topography may be a much smaller effect at short to intermediate wavelengths than predicted by long‐wavelength modelling of the geoid.

Cold, dense subducting lithosphere provides the primary force driving tectonic plates at Earth's surface. The force available to drive the plates depends on a balance between the buoyancy forces driving subduction and the mechanical and buoyancy forces resisting subduction. Because both the buoyancy and rheology of the slab and mantle depend on temperature, composition, grain size, water content, and melt fraction, unraveling which of these processes exert a first-order control on slab dynamics and under what circumstances other processes become first-order effects can be challenging. Laboratory and numerical models of slab dynamics provide a powerful method for testing the combined effects of buoyancy and strength changes that accompany the slab evolution in the upper mantle, transition zone, and lower mantle. Recent studies have focused on understanding how rheologic variations (Newtonian versus non-Newtonian viscosity or water content), geometry (2D versus 3D), and plate motions (trench roll-back or advance) influence the evolution of slabs in the upper mantle and how they sink into the lower mantle. These models suggest that spatial and temporal variations in slab strength and the history of subduction determine whether slabs sink directly into the lower mantle or are trapped in the transition zone. 325 Click here for quick links to Annual Reviews content online, including: • Other articles in this volume • Top cited articles • Top downloaded articles • Our comprehensive search Further ANNUAL REVIEWS Subduction: the term subduction was introduced into the literature in 1970 as an alternative to a large range of descriptive terms to describe the complete underthrusting of one plate beneath another Slab: 3D section of subducted lithosphere in the mantle 326 Billen