Observation of Subducted Oceanic Crust in the Upper Mantle Transition Zone?

Recycling of oceanic crust has large consequences for the chemical evolution of the Earth. Subducted oceanic crust or its high-pressure assembly (garnetite) is denser than peridotite in the upper mantle, but is less dense than perovskite in the lower mantle. It has been suggested that subducted oceanic crust may accumulate above the 660-km discontinuity as a consequence of neutral buoyancy. Observational evidence for this scenario, however, remains elusive. In this study, we synthesize observations from several regional studies to search for evidence for or against subducted oceanic crust in the transition zone. Teleseismic body waves recorded at the southern East Pacific Rise reveal a prominent P-to-S conversion between the converted phases from the 410- and 660-km discontinuities (P410s and P660s, respectively). This arrival has a polarity opposite to those of P410s and P660s, indicating a decrease in velocity with depth at a discontinuity near 600 km depth. This negative-polarity phase is observed intermittently beneath Iceland, but consistently beneath the Hawaiian Islands and South Africa. Stacking along reverberation moveout curves greatly reduces the amplitude of this negative-polarity phase as well as those of P410s and P660s, confirming that this negative-polarity phase is not a reverberation from shallower mantle. Because subducted oceanic crust has lower seismic velocities than pyrolitic mantle in the transition zone, we conclude that the negative-polarity phase is consistent with a layer of subducted oceanic crust accumulated above the 660-km discontinuity in various regions.