Subduction zones are pivotal for the recycling of Earth’s outer layer into its interior. However, the conditions under which new subduction zones initiate are enigmatic. Here, we constructed a transdisciplinary database featuring detailed analysis of more than a dozen documented subduction zone initiation events from the last hundred million years. Our initial findings reveal that horizontally forced subduction zone initiation is dominant over the last 100 Ma, and that most initiation events are proximal to pre-existing subduction zones. The SZI Database is expandable to facilitate access to the most current understanding of subduction zone initiation as research progresses, providing a community platform that establishes a common language to sharpen discussion across the Earth Science community.

Earth's surface topography is a direct physical expression of our planet's dynamics. Most is isostatic, controlled by thickness and density variations within the crust and lithosphere, but a significant proportion arises from forces exerted by underlying mantle convection. This dynamic topography directly connects the evolution of surface environments to Earth's deep interior, but predictions from mantle flow simulations are often inconsistent with inferences from the geological record, with little consensus about its spatial pattern, wavelength and amplitude. Here, we demonstrate that previous comparisons between predictive models and observational constraints have been biased by subjective choices. Using measurements of residual topography beneath the oceans, and a hierarchical Bayesian approach to performing spherical harmonic analyses, we generate a robust estimate of Earth's oceanic residual topography power spectrum. This indicates power of 0.5 ± 0.35 km 2 and peak amplitudes of ∼0.8 ± 0.1 km at long-wavelength (∼10 4 km), decreasing by roughly one order of magnitude at shorter wavelengths (∼10 3 km). We show that geodynamical simulations can only be reconciled with observational constraints if they incorporate lithospheric structure and its impact on mantle flow. This demonstrates that both deep (long-) and shallow (shorter-wavelength) processes are crucial, and implies that dynamic topography is intimately connected to the structure and evolution of Earth's lithosphere. Between Earth's crust and core lies the mantle, a 2,900 km-thick layer of hot rock that constitutes greater than 80% of Earth's volume. Carrying heat to the surface, the convecting mantle is the 'engine' that drives our dynamic planet: it is directly or indirectly responsible for almost all large-scale tectonic and geological activity [1]. As the mantle flows, it transmits normal stresses to the lithosphere-Earth's rigid outermost shell-that are balanced by gravitational stresses arising through topographic deflections of Earth's surface [2, 3, 4, 5, 6, 7, 8, 9]. This so-called dynamic topography is transient, varying both spatially and temporally in response to underlying mantle flow. As a result, it is more challenging to isolate than isostatic topography. The relative importance of dynamic versus isostatic topography varies according to setting: for example, the elevation of the Himalaya is principally isostatic, due to the presence of Earth's thickest continental crust; but the broad excess elevation of the stable South African craton has been

No abstract

Convection in the Earth's mantle is mainly driven by cold, dense subducting slabs, but 15 relatively little is known about how 3D variations in slab morphology and buoyancy 16 affect mantle flow or how the surface above deforms in response (i.e. dynamic 17 topography). We investigate this problem by studying the dynamics of an active region of 18 flat-slab subduction located in Peru in South America. Here the slab geometry is well 19 known, based on the regional seismicity, and we have observations from the local 20 geological record to validate our models. Of particular interest is the widespread 21 subsidence and deposition of the Solimões Formation across western Amazonia that This formation covers an extensive area from the foredeep to the Purus Arch located 24 ~2 000 km away from the trench. Close to the Andes the preservation of several 25 kilometers of sedimentary thicknesses can be easily accounted for by flexure. Based on 26 an estimate of the Andean loading we predict 2.8 to 3.6 km of accommodation space that 27 spans 100 km. The spatial and temporal history of the Solimões Formation however, 28 particularly the thick distal foreland accumulations up to 1.2 km deep, can only be 29 matched with the addition of a longer-wavelength dynamic source of topography. 30 Following the transition from normal to flat subduction, we predict over 1 km of dynamic 31 subsidence (~1500 km wide) that propagates over 1000 km away from the trench, 32 tracking the subduction leading edge. This is followed by a pulse of dynamic uplift over 33 the flat segment behind it. We therefore propose that a combination of uplift, flexure and 34 dynamic topography during slab flattening in Peru is responsible for the sedimentation 35 history and landscape evolution of western Amazonia that eventually led to the 36 configuration of the Amazon Drainage Basin we know today.

Within oceanic lithosphere a fossilised fabric is often preserved originating 15 from the time of plate formation. Such fabric is thought to form at the mid-ocean 16 ridge when olivine crystals align with the direction of plate spreading 1,2 . It is 17 unclear, however, whether this fossil fabric is preserved within slabs during 18 subduction or over-printed by subduction-induced deformation. The alignment of 19 olivine crystals, such as within fossil fabrics, can generate anisotropy that is sensed 20 by passing seismic waves. Seismic anisotropy is therefore a useful tool for 21 investigating the dynamics of subduction zones, but it has so far proven difficult to 22 observe the anisotropic properties of the subducted slab itself. Here we analyse 23

[1] Flat or shallow subduction is a relatively widespread global occurrence, but the dynamics remain poorly understood. In particular, the interaction between flat slabs and the surrounding mantle flow has yet to be studied in detail. Here we present measurements of seismic anisotropy to investigate mantle flow beneath the Peruvian flat-slab segment, the largest present-day region of flat subduction. We conduct a detailed shear wave splitting analysis at a long-running seismic station (NNA) located near Lima, Peru. We present measurements of apparent splitting parameters (fast direction ϕ and delay time δt) for SKS, ScS, and local S phases from 80 events. We observe well-defined frequency dependence and backazimuthal variability, indicating the likely presence of complex anisotropy. Forward modeling the observations with two or three layers of anisotropy reveals a likely layer with a trench-normal fast direction underlying a layer with a more trench-oblique (to trench-subparallel) fast direction. In order to further constrain the anisotropic geometry, we analyzed the source-side splitting from events originating within the slab measured at distant stations. Beneath the flat-slab segment, we found trench-normal fast splitting directions in the subslab mantle, while within the dipping portion of the slab further to the east, likely trench-subparallel anisotropy within the slab itself. This subslab pattern contradicts observations from elsewhere in South America for "normal" (i.e., more steeply dipping) slab conditions. It is similar, however, to inferences from other shallowly dipping subduction zones around the world. While there is an apparent link between slab dip and the surrounding mantle flow, at least beneath Peru, the precise nature of the relationship remains to be clarified.

No abstract

The dynamics of flat subduction, particularly the interaction between a flat slab and the overriding plate, are poorly understood. Here we study the (seismically) anisotropic properties and deformational regime of the mantle directly above the Peruvian flat slab. We analyze shear wave splitting from 370 local S events at 49 stations across southern Peru. We find that the mantle above the flat slab appears to be anisotropic, with modest average delay times (~0.28 s) that are consistent with~4% anisotropy in a~30 km thick mantle layer. The most likely mechanism is the lattice-preferred orientation of olivine, which suggests that the observed splitting pattern preserves information about the mantle deformation. We observe a pronounced change in anisotropy along strike, with predominately trench-parallel fast directions in the north and more variable orientations in the south, which we attribute to the ongoing migration of the Nazca Ridge through the flat slab system.