PhD Thesis:
Using Seismic Methods to Constrain Mantle Convection Processes

Mei Xue
Doctor of Philosophy
Earth and Planetary Science
University of California, Berkeley

Advisor: Richard M. Allen

January 2008

Download the thesis: XuePhDThesis2008.pdf

Abstract
We use seismic anisotropy and traveltime tomography to investigate mantle convection processes with a focus on upwelling at Iceland, Newberry, and Yellowstone, and subduction in Cascadia. Beneath Iceland the observed seismic anisotropy pattern as revealed by SKS splitting analysis shows the outward horizontal flow of upwelling material first toward and then down the North Atlantic Ridge. In contrast, beneath the Newberry hotspot track, SKS splits are not parallel to the track instead showing the average fast directions of ENE-WSW to the northwest and of more E-W to the east. This suggests that the age-progressive volcanism along the Newberry track is unlikely a direct product of asthenospheric flow along the track and is most likely the product of lithosphere-controlled processes. West of the Newberry track is the Juan de Fuca subduction system. We image the subducted slab in the mantle east of the Cascades beneath Oregon to a depth of 400 km but no deeper. We propose that the absence of the slab below 400 km today is due to the arrival of the Yellowstone plume head around 17 Ma which destroyed the Juan de Fuca slab at depths greater than the thickness of the continental lithosphere. A low-velocity layer also imaged beneath the slab is perhaps remnant plume material that has been pulled down by the subduction. This tectonic model is supported by our recent high resolution velocity models for the entire western USA. The models show that the mantle structure beneath the western USA is very heterogeneous but has a strong correspondence with the complicated tectonics of the region. The observation of the Juan de Fuca subduction system confirms it stops at # 400 km, and is disrupted in Oregon, likely due to interaction with the Yellowstone plume head. Beneath Yellowstone, we do not detect a low velocity conduit reaching greater that 500 km depth which implies that either (1) any plume was short-lived; or (2) the conduit is beyond our resolution; or (3) there was no deep mantle plume. We prefer the short- lived plume model as it best explains many of the imaged features the Pacific Northwest.

© Richard M Allen