SYSTEMATIC CLASSIFICATION OF LAYER‐INDUCED TRANSVERSE ISOTROPY*

Abstract

AbstractWaves propagating through a sequence of layers that are thin compared with the wavelength show effects of anisotropy: velocity and displacement direction depend on the angle between the plane of layering and the wave normal, and shear waves split up into two distinct types of different velocity. The layered medium can thus be replaced by a transversely isotrophic medium the parameters of which depend on the parameters of the individual constituent layers.A survey of the anisotropy effects possible in such a medium is generally done by varying the layer parameters in order to obtain different replacement media. This approach guarantees that the replacement medium is realistic, but it does not guarantee adequate sampling of the set of replacement media. To this end one has to begin by selecting the replacement media and then check whether the chosen media possess stable (and eventually realistic) representations by layer sequences. In general, there is an infinite number of layer representations for any transversely isotropic medium that can at all be represented. However, if one restricts the solutions to those requiring the minimal number of layers and the minimum number of different layer parameters, the set of solutions has only one free parameter (i.e., it is a one‐dimensional manifold), and an important subset even has a unique solution. A simple algorithm exists for the determination of these “simplest representations”.Aside from sampling the set of representable transversely isotropic media for survey purposes, the method can be applied to the problem of determining the cause of observed anisotropy effects (or lateral changes in such effects). If this method can be applied to real data, it would for instance allow to determine changes in relative thickness or lithology on a scale smaller than the limit of resolution of the seismic method.