This is a collaborative study between seismologists, mineral physicists and geodynamicists at UC Berkeley and at the Univ. of Michigan, the goal of which is to combine observational constraints and theoretical tools from their respective fields and investigate large scale 3D variations in temperature and major element chemistry in the earth's upper mantle directly from the inversion of seismic waveforms. At the heart of this proposed study is the conjecture, supported by theoretical considerations, that the one step inversion of seismic waveforms for physical parameters such as composition C and temperature T, should yield more stable results than the standard approach of first inverting for velocity and attenuation structure separately, and then interpreting the obtained maps in terms of physical parameters. The latter is currently inhibited by the poor resolution in 3D upper mantle attenuation, Q, due to contamination of amplitudes by largely unmodelled, but significant, effects of focusing and scattering. The coupled character of the equations in the one step approach, combined with the strong sensitivity of amplitudes to T (through anelastic terms) should offset the approximate character of the available mineral physics partial derivatives as well as the errors due to inadequate theoretical treatment of focusing.
The investigators will: 1) invert a global collection of long period seismic waveforms for T and radial anisotropy, using recent experimental contraints on dispersion and attenuation in the uppermost mantle; 2) starting from the model obtained in (1), they will test their ability to invert for lateral variations in composition (as expressed in terms of basalt depletion). Partial derivatives with respect to C and their uncertainties will be evaluated and provided by the mineral physicists in the team using a self-consistent formalism which incorporates results from experiments and first principles calculations. At this point, they will also incorporate geodynamical constraints to increase resolution of the compositional component of the inversion. An iterative procedure will be required at each step in the inversion process, because of the dependence of various partial derivatives on temperature, due to the presence of strong attenuation and phase changes, as well as the large uncertainties remaining in the estimation of these partial derivatives. The project includes funding for two students and a post doc, who will be trained using a cross-disciplinary approach.
