Avalanches at the Core-Mantle Boundary: Possible Role in Geomagnetic Reversals, Mantle Plumes, and Superchrons

Avalanches at the core-mantle boundary have not been directly observed, but if they exist they could affect many geophysical phenomena. Avalanches occur in ?sediment? accumulating on the inner surface of the mantle (according to the theory of Buffett et al.). Because the sediment is not evenly deposited, avalanches could provide the primary mechanism to redistribute sedimentary material evenly over the core-mantle boundary. Core-mantle avalanches, like turbidity flows in the ocean, consist of both solid material and entrained liquid. Such flows can occur at shallow angles (less than a few degrees) and could continue for many kilometers or hundreds of kilometers, depending on the topography. However, these avalanches are upside-down: they flow upward, propelled by buoyancy, into inverted valleys on the mantle surface. The avalanches mix relatively cool sediment with hot liquid iron, creating a redistribution of heat near the boundary. If the avalanche is sufficiently thick (100 m) then the cold pulse will create a downward plume in the core which can disrupt the convective cells that maintain the Earth?s dipole field. When the cells reestablish, the result is a geomagnetic reversal or excursion. We predict a reversal pattern different from that of the chaotic reversals seen in simulations by Glatzmeier. Avalanche-triggered reversals begin with a rapid drop in the dipole moment (but with higher order moments increasing), followed by a period with low dipole moment lasting from hundreds to thousands of years, followed by a rapid build-up of the reversed dipole field. Studies of the detailed time structure of reversals can test the model. As with turbidity flows, we expect a spectrum of avalanche sizes. The largest avalanches are the least probable. The sudden removal of a sediment blanket exposes the lower mantle to a pulse of heat, and for sufficiently large avalanches (>> 100 meters thick) this can contribute to the conditions needed for a mantle plume. A large avalanche could trigger sympathetic avalanches at other locations, wiping clean the topography of much of the boundary. No further avalanches could occur until the slopes rebuild, tens of millions of years later. The Cretaceous superchron (no reversals from 120-85 Ma) could have been initiated by such a super avalanche. In this interpretation, the Ontong-Java Plateau is identified as the result of the mantle plume triggered from the same 120 Ma event. A similar quiet period, the Kiaman reversed superchron, occurred from ca. 320-250 Ma. Our interpretation predicts a large mantle plume at the beginning of this superchron. Oblique extraterrestrial impacts impart high shear to the boundary, and could trigger one or more simultaneous avalanches. Such events could account for reported coincidences of reversals with impact craters and tektite fields. Triggered avalanches could also provide a mechanism for the reported coincidences between large flood basalts and extinctions. In our picture, it is the impact that caused both the extinctions and the flood basalts. Small avalanches, occurring every few years, might be detectable by synthetic aperture seismic analysis. Phase lags can be introduced at many seismic detectors that focus on the core-mantle boundary. The resulting map can then be compared to maps just above and below the boundary. It might even be possible to detect the motion of the avalanche. Avalanches could also change the Earth?s dynamic oblateness and be detectable as changes in the Earth?s gravitational field moments, J2 and higher. Such avalanches might contribute to the variations recently reported in J2.