Detecting melt in the deep mantle with seismic anisotropy and attenuation

点击翻译按钮获取中文摘要

Melting is a critically important component of the Earth system. The chemistry of the crust and upper mantle is controlled by melting and crystallisation at mid-ocean ridges, hotspots and island arcs. Plate tectonics, volcanos and heat flow, and many other Earth processes are all affected by shallow melting. However, melting may be equally important at the other 'end' of the mantle - the core-mantle boundary (CMB). This boundary is, objectively speaking, the most significant in the Earth system as it represents a huge contrast in most physical properties (e.g., temperature, density, chemistry, viscosity).The presence of a melt phase in the lowermost mantle would significantly affect whole mantle thermodynamics and chemistry. The presence of melt would alter the viscosity at the base of the mantle and could affect the generation of plumes. It could provide a persistent hidden reservoir for primordial chemical components 'missing' from the upper mantle. These include incompatible radiogenic elements which might raise the temperature at the base of the mantle, altering the heat-flow out of the core. This would have consequences for the energy available to power the geodynamo: the rapid convection of liquid iron in the Earth's core which generates its magnetic field. It could also potentially sequester volatile phases like water and CO2, altering our picture of the evolution of the abundance of these near the surface through deep time. Experiments have provided plausible candidates for such a melt phase. These include the melting of basalts that have formed at mid-ocean ridges (MORB) which have descended to the core mantle boundary in subducting slabs. Another possibility is that melt is left over from the time when the entire Earth was molten (billions of years ago). While these have been shown experimentally to be possible, we would like to be able to observe their presence directly - a long standing challenge in the Earth Sciences. Seismology provides the only direct probe of the deepest parts of the Earth. The lowermost mantle shows a range of interesting seismic features, including a strong signature of seismic anisotropy (the variation of seismic wavespeed with direction). This is generally ascribed to the deformation of lower mantle minerals but can also be caused by the preferred alignment of an included melt phase. In order to distinguish between these two mechanisms, we propose a new technique which includes measurements of another parameter: seismic attenuation. We have a large dataset of seismic waveforms which image the lowermost mantle across the world, to which we will apply our new methodology. This will allow us to test for the presence of melt across a broad swath of D''. The map of melt we aim to generate will allow us to assess its effect on the broader Earth system, provide insights into the dynamics and structure of the base of the mantle, and probe the origins of deep melting.