Thermo-chemical structures at the core-mantle boundary: A joint dynamical and seismological perspective

The boundary between the core and mantle is a primary interface within the deep interior, and seismologists have revealed that the mantle side of this boundary (D") is extraordinarily complex with a myriad of fine structure. Thermal and chemical heterogeneity, phase transitions, anisotropy, and melting within the lower mantle may all be required to explain the observed fine structure. This strongly suggests that the lower boundary of the mantle is as complex as the mantle's top boundary. We use seismology in combination with dynamic models to better understand the nature of D". In this talk, we will focus on putative slabs at the core mantle boundary, large-scale low shear velocity structures which could be upwellings, and the boundary between the two. It is unlikely that D", manifest as a ~2% jump in S and P velocity ~200 km above the CMB is a simple, dense chemical layer; seismic body waveforms in association with tomography show that these jumps in seismic velocity are associated with putative slabs, opposite to predictions from dynamic models. Even for mantle models with a high-viscosity lower mantle, a phase transition at 660 km depth, depth-dependent thermal expansivity, and depth-dependent thermal diffusivity, ancient slabs could be associated with lateral temperature anomalies ~500C cooler than ambient mantle. Plausible increases of thermal conductivity with depth will not cause slabs to diffuse away in the lower mantle. However, regional spherical models with actual plate evolution, show that slabs are unlikely to be continuous from the upper mantle to D", even for radially simple mantle models. The observation from tomography showing only a few continuous slab-like features from the surface to the CMB may be a result of complex plate kinematics, not mantle layering. There are important consequences of deeply penetrating slabs. Plumes preferentially develop on the edge of slabs, consistent with the boundary between tomographically inferred high/low seismic velocities and (a) hot spots and (b) ULVZ. Dynamic models show that beneath slabs, a substantial amount of hot mantle could become trapped over long periods of time, leading to 'mega-plume' formation. We predict there may be large regions of low seismic velocity directly beneath large-scale high seismic velocity structures at the core-mantle boundary, structures interpreted as ancient slabs. This could be the explanation for low shear velocities found with ScS-S differential times under tomographically imaged fast regions. The so-called 'African superplume' could be associated with sharp discontinuities with a broader low velocity (high temperature) thermal halo. However, the uplift of Africa suggests that the African lower mantle anomaly is not strongly buoyant. SKS waveforms from South American events traveling up the eastern boundary of the African lower mantle anomaly suggest that the lateral boundary in shear velocity is relative sharp. We try to reconcile this observation with dynamic models.