HR: 17:00h
AN: S32E-05    [PDF]
TI: Tomographic modeling of the North American upper mantle
AU: * Nettles, M
EM: nettles@eps.harvard.edu
AF: Earth and Planetary Sciences, Harvard University, 20 Oxford St, Cambridge, MA 02138
AU: Dziewo\'nski, A
EM: dziewons@eps.harvard.edu
AF: Earth and Planetary Sciences, Harvard University, 20 Oxford St, Cambridge, MA 02138
AB: A large amount of digital seismic data recorded at stations in North America is now available, and the Earthscope project will, over the next decade, increase the amount of data available dramatically. This large amount of high-quality data makes it possible to constrain the three-dimensional velocity structure of the North American upper mantle with a resolving wavelength of several hundred kilometers; improvements in resolution can be expected in the near future. We have collected Love and Rayleigh wave dispersion measurements ($T$=35--150~s) at stations of the national networks of the U.S. and Canada, as well as at the stations of several temporary networks, using the method of Ekstr\"om et al.\ (1997). These measurements are combined with a similar dataset of global dispersion measurements ($T$=35--350~s) in an inversion for 3-D, radially anisotropic, S-velocity structure under North America. This modeling follows the approach of Boschi (2002), in which a variable-resolution, global parameterization is used to allow for the simultaneous retrieval of long-wavelength structure globally and shorter-wavelength structure locally. The model we retrieve shows good correspondence with major geological features, with a narrow transition from fast to slow velocities near the edge of the North American craton and a thick ($\sim$275~km) continental root under the Canadian shield. No strong regional variations in radial anisotropy are observed. A second approach to the modeling of seismic velocity structure under the North American continent is to invert for a regional perturbation to a pre-existing global model, with appropriate corrections made for structure encountered by seismic waves along their travel paths outside of North America. Preliminary results suggest that this approach provides an effective and efficient means by which to produce and update a regional or continental-scale model. Such an approach may prove particularly useful as data from USArray begin to arrive during the coming years. In addition to a comparison of results obtained using the two approaches described here, we make a quantitative comparison of our models and previous models of the North American upper mantle (van der Lee, 2002; Godey, 2002) obtained using different datasets and modeling methods.
DE: 7200 SEISMOLOGY
SC: Seismology [S]
MN: 2003 Fall Meeting