Seismic tomography reveals two antipodal LLSVPs in the Earth’s mantle, each extending from the core-mantle boundary (CMB) up to ∼1000 km depth. The LLSVPs are thought to host primordial mantle materials that bear witness of early-Earth processes, and/or subducted basalt that has accumulated in the mantle over billions of years. A compositional distinction between the LLSVPs and the ambient mantle is supported by anti-correlation of bulk-sound and shear-wave velocity (Vs) anomalies as well as abrupt lateral gradients in Vs along LLSVP margins. Both of these observations, however, are mainly restricted to the LLSVP bottom domains (2300∼2900 km depth), or hereinafter referred to as “deep distinct domains” (DDD). Seismic sensitivity calculations suggest that DDDs are more likely to be composed of primordial mantle material than of basaltic material. On the other hand, the seismic signature of LLSVP shallow domains (1000∼2300 km depth) is consistent with a basaltic composition, though a purely thermal origin cannot be ruled out.

A mysterious mid-mantle slowdown The viscosity of Earth's deep interior plays a key role in mediating plate tectonics. Rudolph et al. combined several geophysical data sets to model the viscosity of the mantle. Mantle viscosity abruptly increases below 1000 km. The increase could explain the stalling of subducting slabs and the deflections of hot upwelling plumes around this depth. Although the viscosity increase explains some recent unexpected observations, the origin of the jump itself remains a mystery. Science, this issue p. 1349 Geodynamic modeling reveals a large viscosity increase in Earth’s mid-mantle. The viscosity structure of Earth’s deep mantle affects the thermal evolution of Earth, the ascent of mantle plumes, settling of subducted oceanic lithosphere, and the mixing of compositional heterogeneities in the mantle. Based on a reanalysis of the long-wavelength nonhydrostatic geoid, we infer viscous layering of the mantle using a method that allows us to avoid a priori assumptions about its variation with depth. We detect an increase in viscosity at 800- to 1200-kilometers depth, far greater than the depth of the mineral phase transformations that define the mantle transition zone. The viscosity increase is coincident in depth with regions where seismic tomography has imaged slab stagnation, plume deflection, and changes in large-scale structure and offers a simple explanation of these phenomena.