Slab dynamics in the transition zone

https://doi.org/10.1016/j.pepi.2010.05.005Get rights and content

Abstract

Seismic images of horizontal slab segments in, across or below the transition zone invoke scenarios in which slabs are laid down in the mantle through progressive trench retreat. However, observations of subduction characteristics do not exhibit a clear correlation between trench retreat and slab dip in the transition zone. Instead analysis of a range of subduction characteristics demonstrates that while transition-zone slabs include slabs of many ages and subduction velocities, with both retreating and advancing trench motion, they also have several enigmatic characteristics such as faster sinking rates and a larger range in slab dip than slabs that extend to either shallower or deeper depths. Comparison of subduction characteristics with several analytical and numerical models of subduction dynamics suggests that many of the possible mechanisms for trapping slabs in the transition zone (e.g., positive buoyancy sources, viscous resistance, slab weakening) are only viable if the slab is already shallow-dipping. Two scenarios for the formation of stagnant slabs are proposed: (1) trench retreat prior to slabs entering the transition zone or caused by the negative buoyancy forces associated with the wadsleyite and ringwoodite phase transitions and (2) slow, lateral migration of slabs in stable subduction zones.

Introduction

The Japanese subduction zone is characterized by long-term subduction of old lithosphere (125•132 Ma) at rates of 86•93 mm/yr (Lallemand et al., 2005), suggesting that the sinking slab has more than enough negative buoyancy to pull the subducting plate and to sink deep into the mantle. Yet seismological observations show that the slab is subducting at a shallow dip of 19•26° extending more than 800 km laterally before reaching the transition zone and lies horizontally within the transition zone for another 600 km (Niu et al., 2005). While trench retreat is often suggested as a cause of stagnant slabs in the transition zone, the present-day trench retreat rate in Japan is only 20 mm/yr contributing less than 3° of slab shallowing per million years.

The Japan slab is also characterized by scattered seismicity to depths of 670 km with down-dip compression axis at all depths (Isacks and Molnar, 1969) indicating that the slab remains stiff enough to transmit stresses up-dip. Seismic data indicate that the 410-km phase change is elevated in the slab contributing to the negative buoyancy of the slab, while the 660-km phase change is depressed over a broad region below the horizontal slab, which appears to rest on this boundary (Niu et al., 2005). However, there is also new evidence for a metastable olivine wedge extending from 400 to 560 km at the core of the slab (Jiang et al., 2008).

So why has the Japan slab failed to sink deeper into the lower mantle? Is the positive buoyancy of a metastable olivine wedge or delayed transformation to perovskite at 660 km to blame? Has intense deformation locally weakened the slab and prevented it from pushing into the higher viscosity lower mantle? Did transient trench roll-back in the past cause shallowing of the slab and therefore setting the stage later trapping of the slab in the transition zone?

While the Japan slab is a particularly well-studied stagnant slab (Fukao et al., 2010), these questions can be asked of any of the present-day slabs that appear to be stagnant in the transition zone (e.g., Tonga, Java, Hebrides). Is the stagnation due to a lack of sufficient negative buoyancy, changes in slab strength, the background mantle viscosity structure, geometric effects, or all of these, possibly with each subduction zone exhibiting a unique combination? Are stagnant slabs a transient feature of all subduction zones (Fukao et al., 2001) or does a particular combination of conditions need to occur?

Isolating the first order cause of stagnant slabs is a difficult task given the limited number of observations, and in particular the limited constraints on the evolution of many of these subduction zones. Instead, here I attempt to eliminate possible causes by combining analysis of subduction characteristics with comparisons to predictions from analytical models and the results of recent analog and numerical simulations of subduction.

Section snippets

Enigmatic subduction characteristics

The simplest models consider first order effects of the driving forces of subduction in combination with the effects of coupling a sinking slab with mantle flow, and provide basic intuition on the expected behavior of slabs. For example, we expect a slab with more negative buoyancy (e.g., older and/or longer) to sink more quickly, or a stiffer slab (e.g., colder, dryer, and/or less deformed) to have a more shallow dip. However, as many studies have shown before subduction zone characteristics

Dynamics of slab descent

The evolution of a slab as it descends through the mantle can be thought of as occurring through perturbations of the slab's vertical descent by the evolution of the rheologic structure of the slab. In other words, without the strength of the slab, dense things sink vertically down through the mantle: changes in buoyancy (phase changes) or changes to the background viscosity can only slow down or speed up the descent. A stiff slab, however, can redirect the sinking motion into lateral motion of

Two scenarios for stagnant slabs

The analysis of the slab buoyancy forces, viscous resistance to sinking and slab strength demonstrate that in the absence of trench motion it is difficult for these processes to modify the behavior of an initially steeply dipping slab. In addition, the observations of slab dip show that slabs tend to steepen as they lengthen in the upper mantle. Therefore, in order to trap slabs in the transition zone, it is necessary to shallow the slab by other means either before or after it reaches the

Conclusions

Observations of subduction zone characteristics provide many important insights on slab dynamics both outside and inside the transition zone, by illustrating the ways in which slab behavior deviates from simple model predictions, and by making it possible to isolate combinations of parameters contributing to observed behavior. By comparing these observations to a range of model predictions, two scenarios for the formation of stagnant slabs are proposed: episodic trench retreat of young slabs

References (80)

  • S. Lebedev et al.

    The 660-km discontinuity within the subducting NW-Pacific lithospheric slab

    Earth and Planetary Science Letters

    (2002)
  • J.L. Mosenfelder et al.

    Experimental constraints on the depth of olivine metastability in subducting lithosphere

    Physics of Earth and Planetary Interiors

    (2001)
  • Y. Nishihara et al.

    Grain growth kinetics in wadsleyite: effects of chemical environment

    Physics of the Earth and Planetary Interiors

    (2006)
  • F. Niu et al.

    Mapping the subducting Pacific slab beneath southwest Japan with Hi-net receiver functions

    Earth and Planetary Science Letters

    (2005)
  • W.P. Schellart

    Influence of the subducting plate velocity on the geometry of the slab and migration of the subduction hinge

    Earth and Planetary Science Letters

    (2005)
  • H. Schmeling et al.

    The influence of olivine metastability on the dynamics of subduction

    Earth and Planetary Science Letters

    (1999)
  • A. Shimojuku et al.

    Si and O diffusion in (Mg,FE)2SiO4 wadsleyite and ringwoodite and its implications for the rheology of the mantle transition zone

    Earth and Planetary Science Letters

    (2009)
  • Y. Torii et al.

    Physical conditions producing slab stagnation: constraints of the Clapeyron slope, mantle viscosity, trench retreat, and dip angles

    Tectonophysics

    (2007)
  • D.J. Weidner et al.

    Subduction zone rheology

    Physics of the Earth and Planetary Interiors

    (2001)
  • D. Yamazaki et al.

    Grain growth kinetics of ringwoodite and its implication for rheology of the subducting slab

    Earth and Planetary Science Letters

    (2005)
  • D. Yamazaki et al.

    Texture of (Mg, Fe)SiO3 perovskite and ferro-periclase aggregrate: implications for rheology of the lower mantle

    Physics of Earth and Planetary Interiors

    (2009)
  • G.K. Batchelor

    An Introduction to Fluid Dynamics

    (1967)
  • T.W. Becker et al.

    The developmant of slabs in the upper mantle: insights from numerical and laboratory experiments

    Journal of Geophysical Research

    (1999)
  • N. Bellahsen et al.

    Dynamics of subduction and plate motion in laboratory experiments: insights into the “plate tectonics” behavior of the Earth

    Journal of Geophysical Research

    (2005)
  • D. Bercovici et al.

    Whole-mantle convection and the transition-zone water filter

    Nature

    (2005)
  • M.I. Billen et al.

    Constraints on subducting plate strength within the Kermadec Trench

    Journal of Geophysical Research

    (2005)
  • M.I. Billen et al.

    Rheologic controls on slab dynamics

    Geochemistry, Geophysics and Geosystems

    (2007)
  • C.R. Bina et al.

    Phase transition Clapeyron slopes and transition zone seismic discontinuity topography

    Journal of Geophysical Research

    (1994)
  • M.R. Brudzinski et al.

    Earthquakes and strain in subhorizontal slabs

    Journal of Geophysical Research

    (2005)
  • R. Bñ/4rgmann et al.

    Rheology of the lower crust and upper mantle: evidence from rock mechanics, geodesy and field observations

    Annual Reviews of Earth and Planetary Science

    (2008)
  • R.L. Carlson et al.

    The driving mechanism of plate tectonics: relation to age of the lithosphere at trench

    Geophysical Research Letters

    (1983)
  • U.R. Christensen et al.

    The interaction of a subducting lithospheric slab with a chemical or phase boundary

    Journal of Geophysical Research

    (1984)
  • H. Cíų/4ková et al.

    The influence of rheological weakening and yield stress on the interaction of slabs with the 670 km discontinuity

    Earth and Planetary Science Letters

    (2002)
  • C.P. Conrad et al.

    Mantle convection with strong subduction zones

    Geophysical Journal International

    (2001)
  • C.P. Conrad et al.

    How mantle slabs drive plate tectonics

    Science

    (2002)
  • E. Di Giuseppe et al.

    On the relation between trench migration, seafloor spreading and the strength of the subducting lithosphere

    Lithosphere

    (2009)
  • E. Di Giuseppe et al.

    Slab stiffness control of trench migration: insights from numerical models

    Geochemistry, Geophysics and Geosystems

    (2008)
  • A. Enns et al.

    The dynamics of subduction and trench migration for viscosity stratification

    Geophysical Journal International

    (2005)
  • Y. Fei et al.

    Experimentally determined postspinel transformation boundary in Mg2SiO4 using MgO as an internal pressure standard and its geophysical implications

    Journal of Geophysical Research

    (2004)
  • A.M. Forte et al.

    Deep-mantle high viscosity flow and thermochemical structure inferred from seismic and geodynamic data

    Nature

    (2001)
  • Cited by (0)

    View full text