In‐situ Investigation of Seismic Body Wave Attenuation in Heterogeneous Media*

Abstract

NEWMAN, P.J. and WORTHINGTON, M.H. 1982, In-situ Investigation of Seismic Body Wave Attenuation in heterogeneous Media, Geophysical Prospecting 30, 377-400.Field experiments have been carried out to study the nature and magnitude of seismic wave attenuation for a variety of lithologies. In each survey two three-component sets of geophones with wall clamping mechanisms were lowered down boreholes and signals originating from surface compressional and shear-wave sources were recorded. The data collected were corrected for spherical divergence and analyzed to determine intrinsic attenuation. For situations of anomalous wavefront expansion and in cases where multiple reflection losses may be significant, corrections were supplied by a synthetic seismogram programme to improve the estimates of intrinsic attenuation.Values of attenuation were obtained for pure sandstone, sandstone-mar1 sequences and fissured-unfissured chalk sequences. These formations were all near surface and relatively porous. Significant differences in the relative values of compressional and shear-wave attenuation in the various lithologies were noted. In particular compressional absorption in the fissured zones of chalk was twice the absorption associated with the seismic velocities and the absorption of shear waves fluctuated much less. A large contrast in P-wave attenuation was also observed between pure Bunter Sandstone and the sandstone-marl sequence (absorption was over three times as small in the former). A smaller contrast was noted for shear attenuation.The results obtained suggest that, for the relatively homogeneous formations such as pure sandstone and " unfissured " chalk, " shear " absorption was dominant over " bulk " absorption. In contrast bulk absorption was larger than shear absorption for other formations, e.g. the ' I fissured " chalk sequences and a "partially saturated" chalk section. Attenuation was found to be approximately proportional to frequency in all experiments.

Attenuation of P‐ and S‐waves in limestones

“…For example, chalk is a limestone, with a well-developed dual porosity system consisting of fractures and microscale interparticle porosity. Newman and Worthington (1982) measured the compressional-wave quality factor and the shear-wave quality factor in two near-surface fissured chalks to be 4 and 5.2, and 3.5 and 5.9, respectively, at seismic frequencies. Interpretation of the propagation of low-frequency seismic waves must take account of the potential for anelastic attenuation arising from a similar squirt-flow mechanism based on a large-scale dual porosity system existing in limestones.…”

Influence of welded boundaries in anelastic media on energy flow, and characteristics of P, S‐I, and S‐II waves: Observational evidence for inhomogeneous body waves in low‐loss solids

“…For this example, the media were chosen such that the low-loss approximation is not valid in order to illustrate several characteristics of anelastic waves that in some cases would not otherwise be apparent. With the media characterized by the phase velocity vn and Qn-• for homogeneous S waves, the material parameters chosen for this first example (see Table 1) correspond to those which might exist at interfaces between water-saturated sediments [Hamilton et al, 1970] or at interfaces between fractured near-surface materials [Newman and Worthington, 1982]. Notation for the problem of incident homogeneous linear S waves is illustrated in Figure 1.…”

A Field Study of Seismic Attenuation In Layered Sedimentary Rocks-I.VspData

“…A great deal of evidence exists for frequencydependent intrinsic and apparent Q from theoretical studies and laboratory experiments and more limited evidence from in situ field experiments (see Paper I1 for review). However, the assumption of linear dependence of attenuation with frequency (constant Q) has been shown to be consistent with observations of many VSP data sets (Ganley & Kanasewich 1980;Hauge 1981;Balch et al 1982;Newman & Worthington 1982;Stainsby & Worthington 1985). In general, a constant Q model can only be assumed with confidence over a relatively narrow band of fequencies.…”