Comparison of crustal velocity profiles determined by seismic refraction and teleseismic methods

Recently, two diverse seismic techniques were applied independently to the study of the crustal structure of the Cumberland Plateau, eastern Tennessee. One involved a reinterpretation of a refraction experiment performed in 1965 by the U.S. Geological Survey, consisting of two 400 km long, reversed refraction lines. The other entailed the inversion of broadband teleseismic P waveforms recorded at a single three-component broadband station, RSCP, located at the intersection of the two refraction profiles. A comparison of the two sets of velocity profiles revealed many similarities and some significant differences. Both sets of velocity models consist of three major crustal layers: (1) an upper crust (V_p = 6.1-6.4 km/s) down to about 17 km, (2) a mid-crust (V_p = 6.7-6.9 km/s) between 17 and 40 km depth, (3) a lower crust (V_p = 7.2-7.4 km/s) from 40 to 51 km depth. The refraction models have linear transition zones up to 11 km thick at the base of each layer, whereas the teleseismic models have more irregular transition zones at the base of the mid- and lower crust. The differences in the results of these studies are attributed to the differing frequency bandwidths of the data sets; the predominant sensitivity of the teleseismic data to shear velocities, compared to compressional velocities for the refraction data; and the different analysis procedures involved in each method. Nevertheless, the similarities indicate that the teleseismic waveform method with broadband data is capable of retreiving comparable crustal information as the Cumberland Plateau refraction survey. In addition, it provides the kind of complementary information required to constrain the composition of the continental lower crust and uppermost mantle.