S. Ozalaybey, A. F. Sheehan*, and M. K. Savage
Seismological Laboratory, University of Nevada, Reno, NV 89557
*University of Colorado at Boulder, CIRES, Boulder, CO 80309
The University of Nevada, Reno network of ten broadband digital seismic stations has been collecting continuous teleseismic recordings of events since 1988. This data base was used to study crustal shear velocity structure at 6 sites located in the Basin and Range Province. We analyze the variation of source-equalized teleseismic P to S converted waveforms known as receiver functions, and use a linearized inversion technique to obtain 1 D crustal shear velocities beneath each site. Preliminary inversion results, using initial models taken from previous seismic reflection and surface wave dispersion data where locally available for each site, show that upper mantle shear velocities (~4.5 km/s) are reached near a depth of 30 km either as a sharp or gradational boundary over a 5 km thickness. The velocity models do not show evidence for major midcrustal discontinuities though variations are observed in the uppermost crust. However, using a different initial model (found by a parallel genetic search algorithm) we obtain another velocity model, which has a low velocity channel in the middle to nearly the base of the crust (from 14 to 28 km). This is typical for receiver function inversions, which have been shown to be non-unique because receiver functions are sensitive to relative velocity discontinuity and depth-velocity ratio, rather than to absolute velocity. To overcome the non-uniqueness problem we use local two station surface wave phase velocity measurements. The fundamental mode Rayleigh wave phase velocities for the two models differ by 0.35 km/s within the period range 10 to 20 seconds. Initial phase velocity measurements from two local stations show low phase velocity at 10 seconds (2.85 km/s) favoring the low velocity channel crust model.