Spallation and the generation of surface waves by an underground explosion

Abstract

Underground nuclear explosions, even though they are thoroughly contained, invariably produce spallation. This mass of earth and rock, kicked up by the initial shock wave, falls back within a second or two. It is possible that the surface waves observed from explosions originate from the impact of this material. This hypothesis has been investigated, first, by estimating the spall impulse from accelerometer data, computing the surface wave amplitudes consistent with this impulse, and comparing the results with observed amplitudes and, second, by performing computer explosion simulation experiments with and without spallation and comparing the amplitudes of the resulting surface waves. The comparison of the amplitudes consistent with the spall momentum with the observed Rayleigh amplitudes at ranges of 50-500 km indicates that the spall has sufficient impulse to account for these waves. The computer simulation experiments show that the Rayleigh amplitudes corresponding to realistic inelastic spallation have several times the amplitude 'of Rayleigh waves computed from purely elastic behavior. The surface wave for the spallation case is also delayed relative to the wave from the elastic source. Many studies [Brune et al., 1963; Press et al., 1963; Liebermann and Pomeroy, 1969; Evernden, 1969] have confirmed the development of an effective discriminant between earthquakes and explosions based on a comparison of the excitation of long-period surface waves Me and short-period body waves mo. However, the process by which an underground explosion generates surface waves is not well understood. Theoretical solutions for approximate explosionlike sources have been worked out for the case of linear elastic behavior [Lamb, 1904; Harkrider, 1964; ToksSz et al., 1964; Thiruvenkatachar and Visw,anathan, 1969; Tanyi, 1966; Alterman and Aboudi, 1969]. The actual material behavior in the neighborhood of the explosion is of course inelastic and highly nonlinear tRodcan, 1971]. If one could assume that the inelastic region of a contained underground explosion remained well below the surface, one would have some justification for using the linear approximation in the study of surface wave generation. In the case of a nuclear explosion it is unlikely that this assumption is ever valid, because of the phenomenon of surface spallation [Rinehart, 1960; Eider and Chilton, 1964; Eisler Copyright ¸ 1973 by the American Geophysical Union. et al., 1966; Chilton et al., 1966; Perret, 1971]. Spallation occurs when the compression shock from the explosion strikes the surface and refleets as a tension pulse traveling back into the rock. At some depth the tensile stress in the downward traveling wave exceeds the sum of the tensile strength of the rock, the lithostatic stress, and the compressive stress remaining in the tail end of the upward traveling compression wave. At that depth the rock parts and traps a substantial part of the shock wave momentum in a layer of rock, which flies up from the surface. The parting of the rock forms new surfaces,...

The Slapdown Phase in High-acceleration Records of Large Earthquakes

“…There is a substantial body of work on spall that comes from explosion seismology beginning in the 1960s Day et al 1983;Viecelli 1973;Springer 1974;Day and McLaughlin 1991). In the process of nuclear explosions, an upper soil layer separates (spalls) and is flung upward due to large tensile stress from extremely large accelerations caused by the explosion (on the order of several tens of g).…”

Source Amplitudes of NTS Explosions Inferred from Rayleigh Waves at Albuquerque and Tucson

“…A strong contribution to the Rayleigh waves may be present due to some form of tectonic release (Toks5z and Kehrer, 1972;Lambert, et al, 1972) or spall slapdown (Viecelli, 1973).…”

“…Using data from surface accelerometers, some estimates can be made for the extent of the spall region (Eisler and Chilton, 1964;Viecelli, 1973;Sobel, 1978). Viecelli (1973) …”

“…The impulse delivered by the impact of the large mass of spalled material returning to the free surface has been supposed to have some significant effect on the Rayleigh wave signature of explosions (e.g., Viecelli, 1973). Using estimates for the spall impulse given by Viecelli (1973) and somewhat larger estimates given by Sobel (1978), we compute expected Rayleigh waves at ALQ and TUC to determine its effect. We find it to be quite small.…”

“…These data can be used to estimate the dynamics of the spall zone and its extent (e.g., Chilton, et al, 1966;Viecelli, 1973;Sobel, 1978). Viecelli (1973) suggested that the spall closure makes a large contribution to the recorded Rayleigh waves in the far-field. If this were true, it would strongly compromise the results of our analysis in previous sections.…”