Seismic anisotropy is one of the most widely utilized tools for testing models of mantle deformation and flow. Its widespread use results from the apparently simple correspondence between the orientation of mantle flow and seismic fast-direction, due to the coherent fabric that is produced in olivine crystals in mantle peridotites when they deform. However, recent laboratory studies have cast doubt on this relationship between flow and fast-wavespeed directions when mantle rocks are strained in the presence of a small amount of water or melt. Deformation studies of olivine rocks in the presence of water and melt suggest that the seismic fast direction can rotate to 90 degrees from the shear direction, rather than parallel to the shear direction as is commonly assumed. These observations potentially turn on their heads our models of mantle flow in melt-rich regions, but to date the observations are only from laboratory experiments, and it has not been demonstrated that they are relevant to mantle scales and/or conditions. We will test the degree to which partial melt and/or water may control the mantle fabric in a melt-rich region, specifically the East African Rift in Ethiopia. This location is ideal for this experiment, because the apparently melt-rich region is very well defined, and abundant seismic data exists. However, it is unclear whether the existing data have sufficient spatial resolution of anisotropy to differentiate between competing models. In this one-year proof-of-concept study, we will perform the seismic surface-wave analysis and test several simple models of mantle fabric and flow. Our primary aim is to determine wither the anisotropy beneath the rift is indicative of the "standard" fabric found in dry, melt-free samples, or whether the observed anisotropy is more consistent with deformation fabrics observed under melt-rich or wet conditions. In the process, we will greatly improve our understanding of the mantle dynamic processes that are producing extension and volcanism in this region. Among the broader impacts are that the work will provide training in seismology for LDEO Post-Doctoral Fellow Ben Holtzman, who is serving as a co-PI on the project. In addition, we will be collaborating with Prof. Michael Kendall of the University of Bristol, and Dr. Graham Stuart of University of Leeds.
