Effects of chemical, mineralogical, and temperature variations on the density and bulk sound velocity of the bottom lower mantle

Abstract

In this study, we have modeled the density (ρ) and bulk sound velocity (V Φ) profiles of the bottom lower mantle using the experimental thermal equation of state (EoS) parameters of lower-mantle minerals, including bridgmanite, ferropericlase, CaSiO3-perovskite, and post-perovskite. We re-evaluated the literature pressure-volume-temperature relationships of these minerals using a self-consistent pressure scale in order to avoid the long-standing pressure scale problem and to provide more reliable constraints on the thermal EoS parameters. With the obtained thermal EoS parameters, we have constructed the ρ and V Φ profiles of the bottom lower mantle in different composition, mineralogy, and temperature models. Our modelling results show that the variations of chemistry, mineralogy, and temperature have different seismic signatures from each other. The Fe and Al enrichment at the bottom lower mantle can cause an increase in ρ but greatly lower V Φ. A change in mineralogy needs to be considered with the lateral variation in temperature. The cold slabs will be shown as denser regions compared to the normal mantle because of the combined effect of a lower temperature and the presence of a denser post-perovskite at a shallower depth, whereas the hot regions will have a 1–2% lower ρ than the normal mantle. V Φ of both cold slabs and hot regions will be lower than the normal mantle when bridgmanite is the dominant phase in the normal mantle, yet they will be greater once bridgmanite transforms into post-perovskite in the normal mantle. Our modeling also shows that the presence of a (Fe, Al)-enriched bridgmanite thermal pile above the core-mantle boundary will exhibit a seismic signature of enhanced ρ and V Φ, but a reduced V S, which is consistent with the observed seismic anomalies in the large-low-shear-velocity-provinces (LLSVPs). The existence of such a (Fe, Al)-enriched bridgmanite thermal pile thus can help to understand the origin of the LLSVPs. These results provide new insights for the chemical and structure of the deepest lower mantle.