HR: 16:30h
AN: S54B-03    [Abstracts]
TI: Statistical Analysis of Common Image Point Gathers at and near the Core-Mantle Boundary
AU: * Ma, P
EM: pingma@uiuc.edu
AF: Dept. Statistics, University of Illinois 725 S. Wright St, Champaign, IL 61820, United States
AU: Wang, P
EM: wangp@mit.edu
AF: MIT Earth Atmos Planet Sci, 77 Mass Ave Rm 54-522, Cambridge, MA 02139, United States
AU: Tenorio, L
EM: ltenorio@mines.edu
AF: Center for Wave Phenomena, Colorado School of Mines, Golden, CO 80401, United States
AU: de Hoop, M V
EM: mdehoop@math.purdue.edu
AF: Center for Computational and Applied Mathematics, Purdue University 150 N University St, West Lafayette, IN 47907, United States
AU: van der Hilst, R
EM: hilst@mit.edu
AF: MIT Earth Atmos Planet Sci, 77 Mass Ave Rm 54-522, Cambridge, MA 02139, United States
AB: Remote sensing of the core-mantle boundary region is at the cutting edge of seismological research. We developed statistical method to produce the best estimates of the variations in Earth deep interior from the common image point gathers (Wang et al 2006, van der Hilst et al 2007) using a data-driven approach, in which the image gathers are modeled nonparametrically using mixed effects models. In this framework, the random noise in the signal is allowed to have white and coherent components, and the latter are estimated from the data through generalized cross validation. This methodology, a flexible type of Tikhonov regularization, can be used with different types of correlated noise and with the typically sparsely and unevenly sampled image gathers owing to the geographic distribution of sources and receivers. With synthetic data we show that conventional images deteriorate substantially, in some cases to the point at which weak reflectors can no longer be detected, due to effects of uneven sampling, wave phenomena that are not accounted for in the underlying single scattering approximation, or errors in the assumed background wave speed model. We demonstrate that even in these circumstances, our method can yield adequate estimates of the true model. Our method produces robust images of the core-mantle boundary beneath Central America and suggests the presence of several significant structures in the D double prime region, in particular between 100 and 200 and between 270 and 320 km above the CMB proper.
DE: 0545 Modeling (4255)
DE: 3260 Inverse theory
DE: 7207 Core (1212, 1213, 8124)
DE: 7208 Mantle (1212, 1213, 8124)
DE: 7290 Computational seismology
SC: Seismology [S]
MN: 2007 Fall Meeting