Superplume, Supercontinent, and Post-Perovskite: Mantle Dynamics and Anti-Plate Tectonics on the Core–Mantle Boundary
S. Maruyama,M. Santosh,D. Zhao
DOI: https://doi.org/10.1016/j.gr.2006.06.003
IF: 6.1
2007-01-01
Gondwana Research
Abstract:The Western Pacific Triangular Zone (WPTZ) is the frontier of a future supercontinent to be formed at 250 Ma after present. The WPTZ is characterized by double-sided subduction zones to the east and south, and is a region dominated by extensive refrigeration and water supply into the mantle wedge since at least 200 Ma. Long stagnant slabs extending over 1200 km are present in the mid-Mantle Boundary Layer (MBL, 410-660 km) under the WPTZ, whereas on the Core-Mantle Boundary (CMB, 2700-2900 km depth), there is a thick high-V anomaly, presumably representing a slab graveyard. To explain the D" layer cold anomaly, catastrophic collapse of once stagnant slabs in MBL is necessary, which could have occurred at 30-20 Ma, acting as a trigger to open a series of back-arc basins, hot regions, small ocean basins, and presumably formation of a series of microplates in both ocean and continent. These events were the result of replacement of upper mantle by hotter and more fertile materials from the lower mantle.The thermal structure of the solid Earth was estimated by the phase diagrams of Mid Oceanic Ridge Basalt (MORB) and pyrolite combined with seismic discontinuity planes at 410-660 km, thickness of the D" layers, and distribution of the ultra-low velocity zone (ULVZ). The result clearly shows the presence of two major superplumes and one downwelling. Thermal structure of the Earth seems to be controlled by the subduction history back to 180 Ma, except in the D" layer. The thermal structure of the D" layer seems to be controlled by older slab-graveyards, as expected by paleogeographic reconstructions for Laurasia, Gondwana and Rodinia back to 700 Ma.Comparison of mantle tomography between the Pacific super-plume and underneath the WPTZ suggests the transformation of a cold slab graveyard to a large-scale mantle upwelling with time. The Pacific superplume was born from the coldest CMB underneath the 1.0-0.75 Ga supercontinent Rodinia where huge amounts of cold slabs had accumulated through collision-amalgamation of more than 12 continents. A high velocity P-wave anomaly on a whole-mantle scale shows stagnant slabs restricted to the MBL of circum-Pacific and Tethyan regions. The high velocity zones can be clearly identified within the Pacific domain, suggesting the presence of slab graveyards formed at geological periods much older than the breakup of Rodinia.We speculate that the predominant subduction occurred through the formation period of Gondwana, presumably very active during 600 to 540 Ma period, and again from 400 to 300 Ma during the formation of the northern half of Pangea (Laurasia). We correlate the three dominant slab graveyards with three major orogenies in earth history, with the emerging picture suggesting that the present-day Pacific superplume is located at the center of the Rodinian slab graveyard. We speculate the mechanism of superplume formation through a comparison of the thermal structure of the mantle combined with seismic tomography under the Western Pacific Triangular Zone (WPTZ), Laurasia (Asia), Gondwana (Africa), and Rodinia (Pacific). The coldest mantle formed by extensive subduction to generate a supercontinent, changes with time of the order of several hundreds of million years to the hottest mantle underneath the supercontinent. The Pacific superplume is tightly defined by a steep velocity gradient on the margin, particularly well documented by S-wave velocity. The outermost region of the superplume is characterized by the Rodinia slab graveyard forming a donut-shape. We develop a petrologic model for the Pacific superplume and show how larger plumes are generated at shallower depths in the mantle. We link the mechanism of formation of the superplume to the presence of the mineral post-perovskite, the phase transformation of which to perovskite is exothermic, and thus aids in transporting core heat to mantle, and finally to planetary space by plumes.We summarize the characteristics of tectonic processes operating at the CMB to propose the existence of an "anti-crust" generated through "anti-plate tectonics" at the bottom of the mantle. The chemistry of the anti-crust markedly contrasts with that of the continental crust overlying the mantle. Both the crust and the anti-crust must have increased in volume through geologic time, in close relation with the geochemical reservoirs of the Earth. The process of formation of a new superplume closely accompanies the process of development of anti-crust at the bottom of mantle, through the production of dense melt from the partial melting of recycled MORB, observed now as the ULUZ. When CMB temperature is recovered to near 4000 K through phase transformation, the recycled MORB is partially melted imparting chemical buoyancy of the andesitic residual solid which rises up from CMB, leaving behind the dense melt to sink to CMB and thus increase the mass of anti-crust. These small-scale plumes develop to a large-scale superplume through collision and amalgamation with time. When all recycled MORBs are consumed, it is the time of demise of superplume. Immediately above the CMB, anti-plate tectonics operates to develop anti-crust through the horizontal movement of accumulated slab and their partial melting. Thus, we speculate that another continent, or even a supercontinent, has developed through geologic time at the bottom of the mantle.We also evaluate the heating vs. cooling models in relation to mantle dynamics. Rising plumes control not only the rifting of supercontinents and continents, but also the Atlantic stage as seen by anchored ridge by hotspots in the last 200 Main the Atlantic. Therefore, we propose that the major driving force for the mantle dynamics is the heat supplied from the high-T core, and not the slab pull force by cooling. The best analogy for this is the atmospheric circulation driven by the energy from Sun. (C) 2006 International Association for Gondwana Research. Published by Elsevier B.V All rights reserved.