2D CO CRS Stack for Multi-Component Seismic Reflection Data

Abstract

Summary.The Common-Reflection-Surface (CRS)stack as an alternative to conventional stacking methods has so far mainly been applied to single-component data. We introduce an approach that allows to gener- ate separate stacks of compressional and transversal waves from multi-component seismic reflection data. Based on the traveltime approximation for finite offset, the polarization is analyzed during the search for the optimum orientation and curvature of the CRS stacking operator. We apply this approach to a simple synthetic data set and obtain stacked sections and kinematic wavefield attribute sections separately for PP and PS reflection events. Introduction. The CRS stack was originally developed to stack single-component prestack data acquired along one straight line into a 2D zero-offset (ZO) section. This technique is referred to as 2D ZO CRS stack, see for instance Mann et al. (1999) and Muller (1999). Zhang et al. (2001) extended the CRS stack in order to stack 2D prestack data into a selected finite-offset (FO) gather rather than into a ZO section. If this FO gather is a common-offset (CO) gather this method is referred to as 2D CO CRS stack (Bergler, 2001). In this case the moveout surfaces are described by five kinematic wavefield attributes rather than by three in the ZO case. The search for the wavefield attributes is performed by means of coherence analyses in the prestack data volume. Thus, a velocity model of the subsurface is not required to perform the stack except from the near-surface velocities at the sources and receivers in order to calculate wavefield attributes with a geometrical meaning and to compute the search ranges for the attributes. Bergler (2001) showed that the 2D CO CRS stacking operator can be used to describe traveltimes of PS con- verted waves by choosing a P-wave velocity at the sources and a S-wave velocity at the receivers. Moreover, Bergler et al. (2002) discussed the application of the 2D CO CRS stack to data that were acquired with two components (vertical and horizontal). However, in this approach the CRS stack was performed with both components separately and the distinction between PP and PS reflection events was achieved after the CRS stack. The objective of this paper is to show that the 2D CO CRS stack is able to distinguish between both wave types during the CRS stack to obtain a PP and a PS CO CRS stacked section and five kinematic wavefield attribute sections for each of the both wave types. This is demonstrated with a simple synthetic 2D land data set where vertical and horizontal components have been simulated.