Subduction style is controlled by a variety of physical parameters. Here we investigate the effect of subducting plate length on subduction style using laboratory experiments of time-evolving buoyancy-driven subduction in 3-D space. The investigation includes two experimental sets, one with a lower (~740) and one with a higher (~1,680) subducting plate-to-mantle viscosity ratio (η SP /η M). Each set involves five models with a free-trailing-edge subducting plate and variable plate length (20-60 cm, scaling to 1,600-4,800 km), and one model with a fixed-trailing-edge subducting plate representing an infinitely long plate. Through determining the contact area between subducting plate and underlying mantle, plate length affects the resistance to trenchward motion of the subducting plate and thus controls the partitioning of the subduction velocity (v S) into the subducting plate velocity (v SP) and trench velocity (v T). This subduction partitioning thereby determines the subduction style by controlling the dip angle of the slab tip once it first touches the 660 km discontinuity. The low η SP /η M models display two subduction styles. Short plates (≤40 cm) induce a higher subduction partitioning ratio (v SP /v S), promoting trench advance and slab rollover geometries, whereas longer plates (≥50 cm) lead to a lower v SP /v S , producing continuous trench retreat and backward slab draping geometries. In contrast, the high η SP /η M models exclusively show trench retreat with draping geometries, as the high η SP /η M enables less slab bending before its tip touches the 660 km discontinuity. Our study indicates that future modeling work should consider the effects of plate length on the style and evolution of subduction.

Some subduction zones in nature show mainly overriding plate (OP) extension and low topography, and others show mainly shortening and elevated topography. Here we investigate how end‐member subduction modes (trench retreat with slab rollback and trench advance with slab rollover) affect overriding plate deformation (OPD), topography, and mantle flow with time‐evolving three‐dimensional fully‐dynamic analog models using particle image velocimetry. We conduct two sets of experiments, one of which is characterized by trench retreat, and the other characterized by trench advance. Experiments showing continuous trench retreat experience overall OP extension, while experiments dominated by trench advance experience overall shortening. Both subduction modes present fore‐arc shortening and intra‐arc extension. Our experiments indicate that the overall OPD is mainly driven by the horizontal mantle flow at the base of the OP inducing a viscous drag force (FD), and is determined by the horizontal gradient of the horizontal mantle shear rate dγ˙/dx $d\dot{\gamma }/dx$, which controls the horizontal trench‐normal gradient in FD. Furthermore, a large‐scale trenchward OP tilting and overall subsidence are observed in the experiments showing continuous trench retreat, while a landward OP tilting and an overall uplift are observed during long‐term trench advance. The two types of topography during the two different subduction modes can be ascribed to the downward component of the large‐scale trenchward mantle flow and the upward component of the landward mantle flow, respectively, and thus represent forms of dynamic topography. Our models showing trench advance provide a possible mechanism for OPD and topography at the Makran subduction zone.

<p>Overriding plate deformation (OPD) and topography vary at different subduction zones, with some subduction zones showing mainly overriding plate extension and low topography (e.g. Mariana, Tonga, Izu-Bonin subduction zones), while some showing mainly shortening and elevated topography (e.g. Makran, southern Manila subduction zones). Here we investigate how different subduction modes, namely trench retreat and trench advance, affect OPD and generate corresponding topography with fully dynamic analogue models of time-evolving subduction in three-dimensional space. We conduct two sets of experiments, one of which is characterized by trench retreat and slab rollback, and the other characterized by trench advance and slab rollover. We compute the mantle flow, the overriding plate strain and topography during subduction using the particle image velocimetry technique (PIV). The overriding plate in the experiments showing continuous trench retreat experiences overall extension, while in the experiments with trench advance following trench retreat it experiences overall shortening. The overriding plate in both trench retreat and trench advance subduction modes present fore-arc shortening and intra-arc extension. Our experiments indicate that the overall OPD except in the fore-arc region is mainly driven by the horizontal mantle flow at the base of the OP inducing a viscous drag force (F<sub>D</sub>), and is determined by the gradient of the horizontal mantle flow velocity (dv<sub>x</sub>/dx). Furthermore, a large-scale trenchward overriding plate tilting and an overall subsidence of the overriding plate were observed in the experiments showing continuous trench retreat, while a landward tilting and an overall uplift of the overriding plate were observed during long-term trench advance. The two types of topography during the two different subduction modes can be ascribed to the large-scale trenchward and landward mantle flow, respectively, and thus represent forms of dynamic topography. Our models showing trench advance provide a possible mechanism for OPD in the Makran subduction zone, which has experienced overall trench-normal tectonic shortening in the overriding plate, but shows extension in a local region of the coastal Makran that is spatially comparable to that in our experiments.  In addition, these models might also provide an explanation for the regional topography at the Makran subduction zone, which shows a long-wavelength topographic high in the overriding plate near the trench that decreases northward.</p>

<p>India-Eurasia convergence velocities have dropped significantly from ~18 cm/yr in the Late Cretaceous-earliest Eocene to ~4-5 cm/yr since ~50 Ma. The mechanisms of convergence deceleration, continued convergence since ~50 Ma, long-term continental subduction and long-term Indian indentation into Eurasia still remain controversial. Many previous studies consider an external driving force for the long-term convergence, continental subduction and Indian indentation, and the initial India-Eurasia collision as the trigger for the deceleration. In this study, we investigate the mechanism(s) of the abrupt deceleration, the continued convergence, the long-term continental subduction and long-term Indian indentation using buoyancy-driven analog experiments. We conduct three large-scale experiments to simulate the subduction and collision process at the convergent boundary with different boundary conditions at the 660-km discontinuity, including an infinite viscosity step (the lower-upper-mantle viscosity ratio (η<em><sub>LM</sub></em><em>/η</em><em><sub>UM</sub></em>) is infinitely high), no viscosity step (η<em><sub>M</sub></em><em>/η</em><em><sub>UM</sub></em> =1) and an intermediate viscosity step. The experiment with infinite η<em><sub>LM</sub></em><em>/η</em><em><sub>UM</sub></em> shows a deceleration when the slab tip reaches the 660-km discontinuity, while the other two experiments show a deceleration at the onset of continental subduction. Our experiments show that a higher η<em><sub>LM</sub></em><em>/η</em><em><sub>UM</sub></em> favors a lower velocity drop at the onset of continental subduction, lower convergence velocities, reduced continental subduction and a higher indentation amount, and vice versa. Furthermore, our models suggest that in nature, with an intermediate-high <em>η<sub>LM</sub>/η<sub>UM</sub></em>, the negative buoyancy force of both upper and lower mantle slab segments is the main driver of long-term convergence, continental subduction and Indian indentation.</p>