Earthquake Characteristics and Earthquake-Explosion Discrimination

Abstract

Abstract : Waveform modeling at distances less than 40 degrees can predict the large effects that both rapid velocity variation and anelasticity have on the waveforms of seismic body waves. Because the effects of non-planar layers, lateral heterogeneity, and the frequency dependence of the radiation pattern of the source all become more difficult to predict as the wavelength of seismic radiation decreases; waveform modeling has generally been more successful at longer periods of seismic radiation (100-10 sec.) than shorter periods (5- .1 sec). Earth models derived from the study of body waves of long period are nonetheless valuable in providing simple starting models for the study of short period data valuable to the seismic discrimination problem. One such long period model, T7, for P wave velocity in western North America has been given recently by Burdick and Helmberger (1978). In extending the full wave theory summarized in previous technical reports (Cormier, 1978, 1977) to practical earth models for P waves, we have begun by examining the theory's waveform predictions for the T7 model compared to the predictions given by the Cagniard-de Hoop method of synthesis used by Burdick and Helmberger. In completing this comparison we hope to (1) resolve whether the assumption of primary ray multiples in the Cagniard- de Hoop method, which has been shown to be a problem for body waves interacting with a strong velocity discontinuity such as the core-mantle boundary (Cormier and Ruff, 1978), is inadequate near weak velocity discontinuities postulated for the earth's mantle and to demonstrate the applicability and efficiency of the full wave theory to regions of complicated ray interactions.