A MODEL-BASED APPROACH TO THE COMMON-DIFFRACTION-SURFACE STACK: THEORY AND SYNTHETIC CASE STUDY

摘要

The Common-Reflection-Surface stack method parameterizes and stacks seismic reflection events in a generalized stacking velocity analysis. The common 2D implementation of the Common-Reflection-Surface stack is able to consider a discrete number of events contributing to a given stack sample such that conflicting dip situations can be handled. However, the reliable detection of such conflicting dip situations is difficult and missed contributions to the stacked section might cause artifacts in a subsequent poststack migration, just as unwanted spurious events that might be introduced by this approach. This is deleterious for complex data where prestack migration is no viable option due to its requirements concerning the accuracy of the velocity model. There, we might have to rely on poststack migration, at least for the first structural image in the depth domain. In addition to the approach which considers a small number of discrete dips, the conflicting dip problem has been addressed by explicitly considering a virtually continuous range of dips with a simplified Common-Reflection-Surface stack operator. Due to its relation to diffraction events, this process was termed Common-Diffraction-Surface stack. In analogy to the Common-Reflection-Surface stack, the Common-Diffraction-Surface stack has been implemented and successfully applied in a data-driven manner. The conflicting dip problem has been fully resolved in this way, but the approach comes along with significant computational costs. To overcome this drawback we now present a much more efficient model-based approach to the Common-Diffraction-Surface stack which is designed to generate complete stack sections optimized for poststack migration. Being a time-domain stacking process, this approach only requires a smooth macro-velocity model of minor accuracy. We present the result for the Sigsbee 2A data set and compare its poststack-migrated result to its counterparts obtained with the data-driven Common-Diffraction-Surface approach or the Common-Reflection-Surface stack, respectively. Compared to the data-driven approach, the computational effort is dramatically reduced with even improved results very close to the results of a prestack depth migration.