Influence of the subducting plate velocity on the geometry of the slab and migration of the subduction hinge
Introduction
Geological data indicate that in two subduction zones, overriding plate extension (i.e. back-arc extension) occurred coeval with a reduction in convergence velocity or absolute velocity of the subducting plate [1], [2]. Northrup et al. [1] argued that convergence between Eurasia and the Pacific plate slowed down from ∼13 cm/yr in the Late Cretaceous to ∼4–9 cm/yr during the latest Cretaceous – middle Miocene and increased again to over 10 cm/yr during the late Miocene – Present. As the absolute motion of Eurasia during this time interval was relatively slow (∼1 cm/yr [2]) this change in convergence velocity can be mainly regarded to have resulted from a change in absolute velocity of the subducting Pacific plate. According to Northrup et al. [1] the relatively low convergence rate coincided with major phases of extension along the East Asian active margin (e.g. Kuril Basin, Japan Sea, South China Sea, East China rift system) during the latest Cretaceous to middle Miocene. It was argued that such a reduction in convergence velocity would result in a reduction of compressive stresses at the plate boundary, which could result in extension along the East Asian margin [1]. Nevertheless, with a mere reduction in convergence velocity one would still obtain a net compressive stress across the subduction boundary, which would inhibit extension to occur. However, a reduction in absolute velocity of the subducting plate could promote sinking and hinge-retreat (rollback) of the subducting lithosphere. Such rollback would then enable the overriding plate to extend (e.g. [3], [4], [5]). Similar ideas have been suggested for the relative timing of slowdown of absolute motion of the African plate and the opening of several back-arc basins in the Mediterranean region (Alboran Basin, Liguro–Provencal Basin, Pannonian Basin, Aegean Sea) at ∼30 Ma [2]. It was suggested that a slowdown in absolute motion of the African plate facilitated initiation of rollback of the subducting slabs, which induced back-arc extension in the overriding plate.
In this paper, results of fluid dynamical experiments are described to obtain quantitative insight into the influence of the horizontal trench-perpendicular velocity of the subducting plate (vsp) on the migration velocity of the subduction hinge (vh) and the kinematics of the slab during subduction (Fig. 1). In each experiment a different velocity has been applied to the subducting plate to investigate the dependence of vh on vsp. Physical experiments investigating subduction and rollback have been done before, but vsp was in most cases set to zero [6], [7], [8], [9]. In some cases, vsp was not applied but only resulted from slab pull forces [6], [8], [9], [10]. It was found that in experiments with no applied force or velocity at the trailing edge of the plate, subduction hinges always retreat [6], [8]. In other analogue experiments, vsp was non-zero in the initial stage of the experiments to create a subduction instability and was later set to zero. In these experiments, rollback was mainly observed after setting vsp to zero [11]. With some exceptions (e.g. [11], [12], [13]), most of the numerical models simulating subduction and rollback imposed an external velocity boundary condition to the hinge to enforce hinge-migration (e.g. [14], [15], [16], [17], [18]) and therefore the self-consistent dynamical behaviour of hinge-migration could not be investigated. In this paper, the relation between vh and vsp will be investigated quantitatively under an applied constant vsp. Such an applied velocity boundary condition simulates far field forces such as arising from the mid oceanic ridge and from slab segments attached to the same plate but located at some distance from the particular subduction zone under investigation. In the experiments vh is not externally imposed as a boundary condition but results only from interaction between the negative buoyancy force of the slab, the viscous resistive forces and the applied vsp.
Section snippets
Fluid dynamic model
The models consist of a layered system with a high-viscosity layer (1.3 cm thick) overlying a low-viscosity layer (12 cm thick) confined in a box (Fig. 2). The sidewalls and bottom of the box have no-slip boundary conditions and the top surface has a free-slip boundary condition. Similar designs have been adopted before to investigate subduction processes [6], [7], [8], [9], [10]. The upper layer is made of a Newtonian silicon putty and simulates a ∼65-km-thick oceanic lithosphere. The putty
Slab geometry during sinking
The evolution of the slab has been plotted in Fig. 3 for five experiments with a different applied horizontal velocity. For experiment 9 with vsp=0 cm/hr (Fig. 3a), the evolution of the slab geometry is relatively simple. The slab initially sinks and rolls back with an increasing slab dip angle until the bottom part of the slab is approximately vertical. When the slab tip hits the upper–lower mantle discontinuity (the bottom of the box), the slab folds backward and is subsequently draped over
vsp versus vh
Northrup et al. [1] argued that the convergence velocity between the Pacific and Eurasian plates decreased from 13.0 cm/yr in the Late Cretaceous to 7.8 cm/yr at 68.5–53.0 Ma to a minimum of 3.8 cm/yr in the Eocene (53.0–39.5 Ma) to 6.9–9.0 cm/yr at 39.5–10.5 Ma and finally to 10.6 cm/yr since 10.5 Ma. It was suggested that the timing of slow convergence from 68.5 Ma to 10.5 Ma roughly coincided with periods of back-arc extension in the overriding plate. For the following argumentation it will
Conclusions
In this paper, fluid dynamical experiments have been presented to quantify the influence of the subducting plate velocity on the slab geometry, the mode of subduction and the hinge-migration velocity. Results show that a low subducting plate velocity promotes hinge-retreat. For relatively high plate velocities, phases of hinge-retreat alternate with phases of hinge-stability or hinge-advance due to interaction of the slab with the upper–lower mantle discontinuity. Such slab behaviour could
Acknowledgements
Stimulating discussions with Louis Moresi, Gordon Lister, Chris Kincaid and Ross Griffiths are acknowledged. Comments from Roberto Weinberg on an early version of the manuscript improved its contents. Constructive comments and suggestions from the journal reviewers Mike Gurnis and Jim Buttles, as well as from the editor Rob van der Hilst are greatly appreciated.
References (60)
The influence of trench migration on slab penetration into the lower mantle
Earth Planet. Sci. Lett.
(1996)- et al.
The influence of rheological weakening and yield stress on the interaction of slabs with the 670 km discontinuity
Earth Planet. Sci. Lett.
(2002) - et al.
Episodic back-arc extension during restricted mantle convection in the central Mediterranean
Earth Planet. Sci. Lett.
(2001) - et al.
Timing, kinematics and cause of Aegean extension; a scenario based on a comparison with simple analogue experiments
Tectonophysics
(1999) - et al.
Travel-time tomography of the European–Mediterranean mantle down to 1400 km
Phys. Earth Planet. Inter.
(1993) - et al.
Seismic tomography with P and S data reveals lateral variations in the rigidity of deep slabs
Earth Planet. Sci. Lett.
(1999) - et al.
Laboratory experiments on the structure of subducted lithosphere
Earth Planet. Sci. Lett.
(1995) - et al.
Tethyan subducted slabs under India
Earth Planet. Sci. Lett.
(1999) - et al.
Effects of relative plate motion on the deep structure and penetration depth of slabs below the Izu–Bonin and Mariana island arcs
Earth Planet. Sci. Lett.
(1993) Pre-spread rifting and heterogeneous volcanism in the southern Havre Trough back-arc basin
Mar. Geol.
(1993)
A laboratory investigation of effects of trench migration on the descent of subducted slabs
Earth Planet. Sci. Lett.
On subduction dip angles
Tectonophysics
On the role of subducting oceanic plateaus in the development of shallow flat subduction
Tectonophysics
Narrow subducting slabs and the origin of backarc basins
Tectonophysics
Subducted slabs beneath the eastern Indonesia–Tonga region; insights from tomography
Earth Planet. Sci. Lett.
4-D evolution of SE Asia's mantle from geological reconstructions and seismic tomography
Earth Planet. Sci. Lett.
Motion of the Pacific Plate relative to Eurasia and its potential relation to Cenozoic extension along the eastern margin of Eurasia
Geology
Mediterranean extension and the Africa–Eurasia collision
Tectonics
Sea-floor spreading as thermal convection
J. Geophys. Res.
Origin of the Betic-Rif mountain belt
Tectonics
Asymmetric deformation in the back-arc region of the Kuril arc, northwest Pacific: new insights from analogue modeling
Tectonics
An experimental study of subduction and slab migration
J. Geophys. Res.
Dynamics of retreating slabs: 2. Insights from three-dimensional laboratory experiments
J. Geophys. Res.
Kinematics of subduction and subduction-induced flow in the upper mantle
J. Geophys. Res.
Quantifying the net slab pull force as a driving mechanism for plate tectonics
Geophys. Res. Lett.
Model experiment of plate movements
Nature (Phys. Sci.)
The development of slabs in the upper mantle; insights from numerical and laboratory experiments
J. Geophys. Res.
Controls of the structure of subducted slabs
Nature
Mantle convection with plates and mobile, faulted plate margins
Science
Mantle circulation and the lateral migration of subducted slabs
J. Geophys. Res.
Cited by (136)
Back-arc basins: A global view from geophysical synthesis and analysis
2023, Earth-Science ReviewsCitation Excerpt :Not all subduction systems form back-arc basins. The reasons why they do or do not form BABs are not yet understood despite numerous studies which addressed magmatic, thermal, mechanical, and kinematic processes associated with various mechanisms proposed for back-arc extension (Karig, 1971; Sleep and Toksöz, 1971; Molnar and Atwater, 1978; Uyeda and Kanamori, 1979; Shemenda, 1993; Faccenna et al., 1996, 2012; Scholz and Campos, 1995; Stern, 2002; Heuret and Lallemand, 2005; Schellart, 2005; Schellart et al., 2007; Doglioni et al., 2009). With an intuitive notion that old and cold oceanic plates are less buoyant and therefore should subduct easier than young and buoyant plates, a number of studies attempted to find correlations between subduction dynamics and age of subducting plates (Uyeda and Kanamori, 1979; Furlong et al., 1982; Grellet and Dubois, 1982; Carlson et al., 1983; Jarrard, 1986; Sdrolias and Mueller, 2006).
Role of convergence obliquity and inheritance on sliver tectonics: Insights from 3-D subduction experiments
2022, TectonophysicsCitation Excerpt :Interestingly, periods of slower trench-parallel displacements of the sliver forearc correlate with periods of faster trench-orthogonal relative motion between the forearc and backarc, i.e., faster shortening in this area (Fig. 9). Subduction models have been performed to give insights into the interplays of forces acting on the subduction system and their control on the slab geometry, plate and trench motions and in fine the strain regime within the overriding plate (e.g., Funiciello et al., 2003, 2004; Bellahsen et al., 2005; Martinod et al., 2005, 2013; Schellart, 2005; Heuret et al., 2007; Espurt et al., 2008; Guillaume et al., 2009; Chen et al., 2016). Previous experiments have shown a straightforward relationship between the slab dip, the trench motion, and the strain regime of the overriding plate (e.g., Heuret et al., 2007; Guillaume et al., 2009).
Subduction initiation in the Scotia Sea region and opening of the Drake Passage: When and why?
2021, Earth-Science ReviewsCitation Excerpt :The mantle-stationary trench continued until 30–25 Ma, during which the dragging of the subducted slab (extended by then to ~500 km) to the west-northwest was largely compensated by east-southeast rollback. Such interaction between advancing plate motion and rollback has been observed in laboratory experiments (Schellart, 2005). Finally, accelerating ESE-ward slab roll-back started around 30–25 Ma when spreading ridges had formed across the upper plate (Eagles et al., 2005).