DI017-08
A Panoptic View of Scattering in the Core-Mantle Boundary Region of the Pacific

Friday, 11 December 2020: 17:58
Virtual
Doyeon Kim1, Vedran Lekic2, Brice Menard3, Andrea Mundl-Petermeier4, Val Finlayson5 and Richard J Walker1, (1)University of Maryland College Park, Department of Geology, College Park, MD, United States, (2)University of Maryland, Department of Geology, College Park, MD, United States, (3)Johns Hopkins University, Department of Physics and Astronomy, Baltimore, MD, United States, (4)University of Maryland College Park, Department of Geology, College Park, United States, (5)University of Hawaii at Manoa, College Park, MD, United States
Abstract:
The core-mantle boundary (CMB) region hosts structures tied to the origin of plumes, fate of slabs, and reservoirs of primitive and recycled material. Scattering of seismic waves by structures near the CMB has traditionally been analyzed in a targeted way, leveraging geometric arrangements to contextualize and understand waveform characteristics. Here, we systematically analyze thousands of seismograms of waves diffracting along the core-mantle boundary (Sdiff) and obtain a panoptic view of scattering across the Pacific. We do so using a new manifold learning algorithm called the Sequencer which is capable of arranging waveforms in an optimal order that can reveal underlying trends in the data. In nearly half of the diffracting waveforms from deep events of Mw > 6.5 during 1990-2018, we detect seismic waves scattered by 3D structures near the core-mantle boundary. The prevalence of these scattered arrivals shows that the CMB region hosts pervasive lateral heterogeneity. The relative amplitude of the scattered signals is constant with lag-time, suggesting that they are produced by oblique interaction with laterally-abrupt edges of large low-shear-velocity provinces, or by widespread ultralow-velocity zones (ULVZs). We also detect uniquely loud signals near Hawaii, which are best explained by either a mega-ULVZ or seismically slow plume root beneath Hawaii. Finally, we discover a 25 km tall, 1000 km diameter, mega-ULVZ beneath the Marquesas Islands, with a Vs reduction of ~20%. This discovery presents an opportunity to test the association between 182W/184W anomalies in ocean island basalts and mega-ULVZs proposed by Mundl et al. (2017).