Back-arc basins behind subduction zones often abandon an old spreading centre and open a new one closer to the trench, but what controls this episodicity has been contested. Schliffke et al. (2022) showed with 3D visco-plastic models that jumps occur spontaneously when the resistance of the lengthening transform faults that decouple the back-arc from neighbouring plates exceeds the resistance to rupture the overriding plate near the trench, yielding a near-linear threshold in subducting-plate width: jumps appeared only in models 600–1000 km wide, while 400 km and 1500 km systems did not jump. The source itself left open how this strength-ratio criterion survives more realistic physics: whether brittle failure, strain weakening and reactivation of pre-existing faults lower the STEP-fault tear resistance enough to permit jumps below the ~600 km lower width limit; whether the linear threshold extrapolates to the widest subduction zones (≳2000 km), where along-strike trench curvature and rollback-rate variations driven by toroidal mantle flow may dominate; whether the widest jump-less back-arc basins (e.g., New Hebrides) require external triggers; and how robust the scaling is to 3D mantle flow and lithospheric heterogeneity. No study to date tests the strength-ratio criterion or its width limits. The problem remains open.

Background

A back-arc basin is a small oceanic basin that opens behind a volcanic arc because the subducting slab sinks faster than it converges, so the trench migrates oceanward ("rollback") and stretches the overriding plate. Many back-arc basins do not spread continuously at one locus: spreading at a given centre typically lasts only 5–15 Myr before it ceases, and in several basins the old spreading centre is then abandoned and a new one forms closer to the retreating trench — an episodic spreading-centre jump, recorded in the Tyrrhenian Sea (Vavilov and Marsili basins), the Scotia Sea and the Lesser Antilles. Conventional explanations invoke triggers external to the spreading system: plate-velocity changes, pre-existing weaknesses, punctuated events such as slab-window opening, or, in recent synthesis, off-axis mantle upwelling and plate-motion changes (Guan et al. 2025; Magni et al. 2021). A key geometric ingredient in retreating narrow subduction systems is the STEP fault (subduction-transform-edge-propagator): the tear at a slab edge that propagates through the lithosphere and, together with long transform faults, lets the retreating slab and overriding plate decouple from the neighbouring plates. Schliffke, van Hunen, Allen, Magni and Gueydan (2022, Nature Communications 13:582) presented 3D numerical models of subduction of an oceanic plate 400–1500 km wide (along-strike width) between continental neighbours, with a simple visco-plastic rheology (viscous creep plus a plastic yield criterion, without true brittle failure, strain weakening or fault reactivation) and homogeneous plates. In their models jumps arise with no trigger: as rollback lengthens the transform faults, frictional resistance on them eventually exceeds the resistance to rupture the overriding plate near the trench, the old spreading centre locks, and a new one jumps trenchward. The jump timing and trench-to-spreading-centre distance therefore scale with the ratio of transform-fault to overriding-plate strength, and both correlate linearly with plate width. Jumps occurred only in 600–1000 km wide models: a 400 km slab could not overcome STEP-fault tear resistance and ceased subducting, and a 1500 km slab retreated more than 4000 km without jumping. Natural systems with jumps (Calabria/Tyrrhenian, Scotia, Lesser Antilles) plot close to the model regression line while jump-less basins plot below it, and the New Hebrides back-arc (~1600 km), the widest with a back-arc basin, is predicted not to jump within a realistic plate lifetime. The strength ratio is not directly measurable in nature; it is inferred from this force-balance argument and from model–data comparison of trench–spreading-centre distances and spreading durations (constrained by dated sedimentary rocks and magnetic seafloor lineations). The authors themselves flagged the limits of these models: the simplified rheology, the homogeneity of the plates, the extrapolation of the linear threshold to the widest systems, where along-strike variations in trench curvature and rollback velocity driven by toroidal mantle flow (the flow circulating around slab edges) may dominate the force balance, and the possibility that the widest jump-less basins need external factors such as plate reorganisation, lower-mantle slab penetration or nearby collision.

Problem Statement

Determine the domain of validity of the transform-fault-to-overriding-plate strength-ratio control on episodic back-arc spreading-centre jumps — the single overarching objective of establishing whether, and under what conditions, Schliffke et al.'s (2022) internally driven jump criterion holds across the spectrum of natural subduction systems. Concretely, and as jointly serving this objective, all four of the following parts are required for its resolution, and no single part completes the task: (1) quantify how realistic rheologies — true brittle failure, strain weakening, reactivation of pre-existing faults, and weaker plate boundaries — modify STEP-fault tear resistance and the strength-ratio criterion, in particular whether they enable spreading jumps in subduction systems narrower than the model-predicted lower limit of about 600 km; (2) test whether the linear jump threshold in plate width extrapolates to the widest subduction zones (≳2000 km), where along-strike variations in trench curvature and rollback velocity driven by toroidal mantle flow may dominate the force balance (a specific instance of the general robustness question in (4)); (3) establish whether the widest back-arc basins that lack spreading jumps, such as the New Hebrides (~1600 km), require external factors — plate reorganisation, slab penetration into the lower mantle, nearby collision — to trigger a jump, or whether the internal transform-versus-overriding-plate feedback can become effective under more realistic rheologies and boundary conditions; and (4) assess how robust the strength-ratio scaling and its linear correlation with plate width are to 3D mantle flow and lithospheric heterogeneities.

References

  1. Schliffke, N., van Hunen, J., Allen, M. B., Magni, V., & Gueydan, F. (2022). Episodic back-arc spreading centre jumps controlled by transform fault to overriding plate strength ratio. Nature Communications, 13, 582. DOI: 10.1038/s41467-022-28228-5, https://doi.org/10.1038/s41467-022-28228-5
  2. Magni, V., Naliboff, J., Prada, M., & Gaina, C. (2021). Ridge Jumps and Mantle Exhumation in Back-Arc Basins. Geosciences, 11(11), 475. DOI: 10.3390/geosciences11110475, https://doi.org/10.3390/geosciences11110475
  3. Dong, Y., Xiong, S., Wang, F., Ji, Z., Li, Y., Yamamoto, S., Niida, K., & Xu, W. (2023). Triggering of episodic back-arc extensions in the northeast Asian continental margin by deep mantle flow. Geology, 51(2), 193–198. DOI: 10.1130/G50724.1, https://doi.org/10.1130/G50724.1
  4. Gea, P. J., Mancilla, F. de L., Negredo, A. M., & van Hunen, J. (2024). Overriding Plate Thickness as a Controlling Factor for Trench Retreat Rates in Narrow Subduction Zones. Geochemistry, Geophysics, Geosystems, 25(2), e2023GC011345. DOI: 10.1029/2023GC011345, https://doi.org/10.1029/2023GC011345
  5. Zhang, D., & Davies, J. H. (2024). How a volcanic arc influences back-arc extension: insight from 2D numerical models. Solid Earth, 15(9), 1113–1132. DOI: 10.5194/se-15-1113-2024, https://doi.org/10.5194/se-15-1113-2024
  6. Guan, H., Xu, M., Zhao, M., Xie, W., Quan, Y., Mittelstaedt, E. L., Li, S., & Morgan, J. P. (2025). Mid-ocean ridge jumps: Impacts, classification, and mechanisms. Earth-Science Reviews, 267, 105167. DOI: 10.1016/j.earscirev.2025.105167, https://doi.org/10.1016/j.earscirev.2025.105167
  7. Gomes, A. C., Balázs, A., & Gerya, T. (2025). Arc Splitting and Back-Arc Spreading Evolution: The Control of Slab Dehydration and Melting Processes. Geochemistry, Geophysics, Geosystems, 26(10), e2025GC012255. DOI: 10.1029/2025GC012255, https://doi.org/10.1029/2025GC012255
  8. Su, H., Leng, W., Liao, J., & Brune, S. (2026). Weak mantle wedge causes mantle exhumation punctuated with discrete oceanic crust in the Tyrrhenian basin. Nature Communications, 17, 1291. DOI: 10.1038/s41467-025-68052-1, https://doi.org/10.1038/s41467-025-68052-1
  9. Liao, J., Liao, L., Zhuo, M., Qing, J., Wang, G., Shen, Y., Fan, Z., & Gao, R. (2025). Review of structures and geodynamics of mid-ocean ridges. Science China Earth Sciences, 68(12), 3912–3927. DOI: 10.1007/s11430-025-1689-y, https://doi.org/10.1007/s11430-025-1689-y