The Earth's mantle embeds the engine for much of its dynamics. Its structure is notoriously diffiult to probe with geophysical techniques such as resolution limited seismology.. Its chemical and mineralogical composition can be inferred from volcanic rocks and meteorites. It is currently thought possible to delineate the style and rate of mantle convection by dynamic modelling governed by continuum mechanical and experimental constraints based on mantle composition and thermal structure concomittant with a wide range of boundary conditions and convection styles. Here I shall address a new field: mantle kinematics. Mantle images from seismic travel time tomography have identified elongated bodies of seismically faster regions connected to the surface lithosphere correlatable with subducting slabs. Correlations between plate reconstructions and tomography have provided a handle from geological records of mountain belts, to date when a particular slab started and finished subduction. Systematic correlation between younger (older) records of subduction and shallower (deeper) seismic anomalies, interpreted as slab remnants, led to an "Atlas of the Underworld" in which 100 subducted lithospheric remnants are correlated with their geological record. Current anomalies in the mantle represent a history of subduction of ~ 250 million years. They suggest that whereas upper mantle slab sinking rates are as variable a plate motion rates (up to 20 cm/yr), lower mantle sinking rates are uniform between 1-2 cm/yr, and in situ rates even lower, down to 0 5 cm/yr. No noticeable lateral motion relative to one another is found between slabs: they appear to sink vertically as lead balloons, oposite to but reminiscent of the apparent vertical rise of mantle plumes and hotspots which appear to be mantle stationary. The next frontier are the discovery of New Guinea and the Gakkel Ridge remnants of subduction-modified mantle wedges that formed during subduction, 30 to 140 Ma ago respectively. They show no movement relative to their corresponding slabs in the deep mantle. These corelations, if globally reproducible, may provide a novel way to reconstruct 3D motions of the Earth's mantle on a timescale relevant for plate tectonics independent from and as boundary conditions to dynamic interpretations. The current state of the art suggests that these motions in the mantle are much slower than those of plates and plumes that seem to move over and through an otherwise sluggish mantle.

Mantle kinematics as constraint of the plate tectonic engineDouwe van Hinsbergen, Douwe van der Meer, Wim SpakmarDepartment of Earth Sciences, Utrecht University, the Netherlands

ACS Spring 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.