An evolving view of transition zone and midmantle viscosity

In a new study, Liu and Zhong (this issue) investigate the impact of thermochemical anomalies in the lower mantle on the viscosity structure inferred from the geoid, finding that both thermochemical and thermal only mantle models can be reconciled with the geoid with nearly the same viscosity structure. They also find a preference for radial viscosity models where the lowest viscosity layer occurs in the transition zone as opposed to the upper mantle. While their weak transition zone viscosity model is not unprecedented, it differs from the most widely used viscosity models, which assume the smallest viscosity occurs in the upper mantle. Liu and Zhong's investigation does not support the midmantle viscosity increase recently suggested by Rudolph et al. [2015] who use a nearly identical methodology but different tomographic model. Key Points: Both weak transition zone and weak upper mantle radial viscosity models are consistent with the geoid A viscosity increase at 670 or 1000 km are both compatible with the geoid The best fitting viscosity model may depend on the choice of seismic tomography model