Collaborative Research: NSFGEO-NERC: Advancing capabilities to model ultra-low velocity zone properties through full waveform Bayesian inversion and geodynamic modeling

This proposal will improve the description of features called Ultra-Low Velocity Zones (ULVZs). ULVZs are thin patches of material that sit on the Earth’s core-mantle boundary (CMB), which is at a depth of 2891 km (1796 miles). As their name implies, ULVZs slow passing seismic waves by as much as 50%.

This makes which makes them some of the most anomalous features of the Earth’s interior. Several ULVZs are beneath some of Earth’s largest hot spot volcanoes, such as Hawaii, Iceland, and Samoa. Imaging of the Earth’s interior shows that hot rising materials, or mantle plumes, connect these volcanoes from ULVZs to the surface.

But the origin, lifetime, size, and composition of these ULVZs are not known, nor is their role in the formation of the mantle plumes. This project will provide greater details as to where these features exist, their formation and composition. This work analyzes two types of seismic waves sensitive to thin ULVZ layering that interact with and bounce off these ULVZ features.

The study focuses on an area beneath the western Pacific Ocean because of extensive seismic data coverage and known ULVZs in the region. The project will create a new seismic database built from public records and networks in South Korea, Taiwan, and Japan. This database will then enable imaging of ULVZ structures at the CMB using modern statistical capabilities.

Computer simulations of mantle flow will help interpret the formation, motion, and role in mantle processes of the ULVZs. This project includes an annual outreach STEM event for K-8 students for the local Utah Pacific Island community. This event will include activities to teach about the mantle plumes that form Pacific Islands through hot spot volcanism.

Ultra-Low Velocity Zones (ULVZs) are structures with strongly reduced seismic velocities at Earth’s core-mantle boundary (CMB) which have been associated with or linked to hot-spot volcanism, Large Igneous Provinces, core-mantle interaction, downwelling subduction and Large-Low Velocity Provinces (LLVPs) and thus are a critical component of global mantle dynamics. ULVZs are typically studied using waveform analysis of ULVZ-sensitive seismic probes (e.g., SPdKS, ScP, ScS), but previous studies suffer from large uncertainties in ULVZ parameters due to modeling trade-offs and lack a geophysical inference of ULVZs through rigorous parameter estimation.

This project will develop a transformative waveform inversion approach and apply it to characterize ULVZ properties including their seismic velocities, density, size, and shape. Specifically, the project will (1) collect a new database of ULVZ-sensitive ScP and PcP waveforms utilizing publicly available seismic array data as well as new data from South Korea, Taiwan, Japan, and the International Monitoring System (IMS) arrays, (2) advance capabilities for full waveform Bayesian inversion for ULVZ properties to quantitatively distinguish regions with and without ULVZ structures (i.e., Bayesian model selection), and to perform joint inversion of ScP and PcP waveforms, (3) quantify the waveform effects from 2-D/3-D ULVZ structures, (4) search for additional, previously unlooked-for arrivals in the ScP wavefield consistent with 3-D ULVZ structure using array processing approaches, and (5) relate observed ULVZ localities and properties to lowermost mantle flow and dynamics through 3D geodynamic models.

This is a collaborative proposal between scientists at the University of Utah, Arizona State University, and Leeds University, and is therefore co-funded by the National Science Foundation (NSF) and the United Kingdom’s Natural Environment Research Council (NERC).

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.