Search

Analysing S‐wave splitting has become a routine step in processing multicomponent data. Typically, this analysis leads to determining the principal directions of a transversely isotropic medium with a horizontal symmetry axis, which is assumed to be responsible for azimuthal anisotropy, and to the time delays between the fast and slow S‐waves. These parameters are commonly estimated layer‐by‐layer from the top. Errors in layer stripping occurring in shallow layers might propagate to deeper layers. We propose a method for S‐wave splitting analysis and compensation that consists of inverting interval values of splitting intensity to obtain a model of anisotropic parameters that vary with time and/or depth. Splitting intensity is a robust attribute with respect to structural variations and is commutative, which means that it can be summed along a ray (or throughout a sensitivity kernel volume) and can be linearly related to anisotropic perturbations at depth. Therefore, it is possible to estimate anisotropic properties within a geological formation (e.g. the reservoir) by analysing the differences of splitting intensity measured at the top and at the bottom of the layer. This allows us to avoid layer stripping, in particular, for shallow layers where anisotropic parameters are difficult to estimate due to poor coverage, and it makes S‐wave splitting analysis simpler to apply. We demonstrate this method on synthetic and real data. Because the splitting intensity attribute shows usefulness in S‐wave splitting analysis in transversely isotropic media, we extend the splitting intensity theory to lower symmetry classes. It enables the characterization of tilted transversely isotropic and tilted orthorhombic media, opening new opportunities for anisotropic model building. [ABSTRACT FROM AUTHOR]

The paper reports on the use of surface-wave testing for the construction of a three-dimensional (3D) shear-wave velocity model of an alluvial basin. The town of Tarcento (Italy) is located in a region with a high level of seismic hazard and was strongly affected by the two Friuli earthquakes in 1976. The seismic surveys were performed with surface-wave multistation methods using a combination of active-source and passive-source experimental setup. Experimental data were collected at 16 sites, and inversion of the data is based on an innovative procedure for spatially constrained inversion with a single objective function in which a priori information is included. The method provides an improvement in the robustness of the solution, mitigating solution nonuniqueness. Available borehole logs at different locations are integrated into the data set in terms of a priori bedrock information. Three independent cross-hole tests are used for a posteriori comparison of the inverted one-dimensional (1D) profiles. Three-dimensional interpolation of the obtained profiles leads to a shear-wave velocity model that is internally consistent and complies with a priori information, cross-hole results, and suitable boundary profiles. The model is intended for numerical simulations of the seismic response of the basin. [ABSTRACT FROM AUTHOR]