Broad-Band Seismic Studies of Body Waves

Citation

Burdick, Lawrence James (1977) Broad-Band Seismic Studies of Body Waves. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/zt3a-vh80. https://resolver.caltech.edu/CaltechTHESIS:12042020-183920934

Abstract

Analytic information from both long period and short period seis­mometers is used in three different studies of teleseismic body waves. The composite broad-band information is first used in a source study of the 9/12/66 Truckee, 8/1/75 Oroville and 4/9/68 Borrego Mountain earthquakes in California. The purpose of the study is to determine the fault area and displacement from the body wave pulse shape and to compare this information with the postseismic data. The pulse shapes are determined by a simultaneous short period-long period deconvolution procedure and matched with theoretical pulses from fault models. The results indicate that the area which radiated the body waves was smaller than the area of the aftershock zone and that the displacement in this area was larger than the offsets observed at the surface. The purpose of the second study is to find the value of t*β for teleseismic S waves with a raypath under the continental United States. The data set consists of long and short period body waves from the Borrego Mountain earthquake as observed in the northeastern U. S. The P waveforms are dominated by the sP phase and the SH waveforms by the sS. It is assumed that there are no losses in pure compression so that the relative attenuation rate of P and S waves is known. The initial source radiation is determined from the sP phase and the value of t*β from the spectral content of the S wave. The results indicate that it is 5.2 ± .7 seconds along this raypath. Long and short period body waves from some deep South American events are used to test for lateral asymmetry of the Q distribution under the U. S. The results indicate that the attenuation rate of teleseismic body waves is roughly constant across the North American continent. The t*β value for a 600 km deep earthquake appears to be about 3. seconds. The purpose of the final study is to find an upper mantle compressional velocity profile which explains both the short period and long period waveform data. The region of study is the western or tectonically active portion of the United States. The short period waveforms are from NTS bombs and the long period waveforms are from shallow California earthquakes with known source mechanisms. Travel time and apparent velocity data are also used to constrain the model. The new velocity profile is called T7. It accurately predicts the long and short period body waveshapes from 10° to 30°. The new model is substantially different than the previous one for the region. The first discontinuity is at 400 km depth which is shallower than before and the second is at 670 km which is deeper. The velocity jumps have been reduced in size to 5% and 4% respectively. The velocities through much of the profile have been reduced slightly to improve the fit to the travel time data.

Thesis Files