This project is applying a recently developed, three-dimensional, direct wavefield imaging method to Earthscope Transportable Array (TA) data. The approach applies a series of algorithms that implement a form of three-dimensional, migration inversion. Three-component waveforms of body waves recorded by the TA are manipulated in a manner similar to seismic reflection processing to produce a three-dimensional image volume of P to S scattering strength beneath the continent. The method is being used to produce high-resolution images of the subducting lithosphere of the Juan de Fuca/Farallon system and to develop a 3D model of the geometry of that system. As the array moves eastward the method will be used to image lithospheric scale boundaries that can be applied to test models of how the Precambrian core of the continent was assembled. Technical problems being attacked to improve the capabilities of this new method are: (1) extension of the method to image S to P conversions, (2) new methods for estimating the impulse response of the medium (deconvolution), (3) improved handling of irregular data that are the norm for passive arrays, (4) improved error appraisal metrics, and (5) multiple removal. Broader impacts include support for a PhD student's research and plans to make 3D image volumes available via the web.
This project saw the maturity of a new technology for imaging the earth that had been under development by the PI for more than a decade. The project has produced what are arguably the highest resolution images ever produced of the Earth’s mantle. The figure included with this report demonstrates the current capability with one section from a three-dimension image we have produced from Earthscope Transportable Array (TA) data. The image illustrated is comparable to modern 3D seismic reflection data that are the standard tool used for oil and gas exploration. There is, however, a huge difference in scale. Oil and gas seismic images image upper crust to depths of a few kilometers. The example shown extends to a depth of 1000 km. This is possible because the technique uses large earthquakes from around the globe to illuminate the subsurface. We processed raw ground motion data from the Earthscope TA to produce the 3D image illustrated. In the project we used these data to produce new insights on two large-scale geologic problems. The geology of the Pacific Northwest is controlled by what we call subduction. The oceanic crust off Washington and Oregon is the last remnant of a huge oceanic plate commonly called the Farallon Plate. Plate tectonic models show that the Farallon Plate has been consumed by westward motion of North America for nearly 300 million years. The debris from that process fills the mantle under North America and one of the key problems the USArray was deployed to address was to better define the 3D geometry of the Farallon Plate under the continent. This project contributed to understanding of this issue in two ways: (a) our images provide fundamentally new constraints on the geometry of the system, and (b) we combined our results with "seismic tomography" models produced by other investigators in the first comprehensive review article on TA imaging results. The results from different tomography models as well as our results point to new insights on the mantle flow processes under North America that the boarder community is still working to digest. Our results have produced a new view of the detailed geometry of two global-scale discontinuities seismologists commonly refer to as the 410 and 660 km discontinuities (see figure linked to this report). The prevailing model in mineral physics is that these two discontinuities are linked to phase changes of the mineral olivine that is a major component of the mantle. Detailed mapping of the depth to these horizons has been suggested as a means to remotely sense variations in temperature of the mantle. Our new imaging results suggest these discontinuities have richness of texture very different from the prevailing model today of a pure interface separating two different phases those depth is controlled by temperature. There are hints these discontinuities have a previously unresolved roughness that is beginning to be resolved with our new plane wave migration method.
