Recycling stacking velocities for better seismic processing
A.F. McIntosh and J.P. Oden
Exploration Geophysics
24(4) 671 - 678
Published: 1993
Abstract
In processing seismic data, interpretation and quality control of the primary velocity field is one of the most laborious tasks undertaken by both the processing contractor and the client. The required level of effort is even greater in areas with poor signal to noise ratio and a high level of multiple contamination. In order to consistently achieve optimal stack quality, a simple automated velocity conditioning technique is used to allow for the continuous recycling of a 3-D modelled stacking velocity database into subsequent processing projects. This approach reduces time spent determining the primary velocity field and improves accuracy of the results, especially in difficult seismic areas. The modelled primary velocity field is equally suited for use in both reprocessing and new infill acquisition projects. The velocity conditioning technique creates a horizon-consistent stacking velocity database by cross-interpolation between the interpreted seismic time and raw stacking velocity databases. The top down conditioning approach commences in the shallow part of the section where velocity determination has both far less uncertainty and increased sensitivity. Stacking velocities for the initial stratigraphic interval are converted to the interval velocity domain, midpoint time-normalised to remove compaction effects, heavily smoothed using knowledge of the spatial distribution of known lithologic variations, then reconstructed back to a new lower layer stacking velocity after inverse time-normalisation. This conditioned lower layer becomes the upper layer for the next stratigraphic interval and the process is repeated. The final three-dimensional stacking velocity model is strictly seismic-based, horizon-consistent, tied in both the strike and dip directions, and contains only geologically plausible interval velocities. The horizon-consistent aspect of this technique integrates geophysical interpretation with seismic processing. A preliminary horizon interpretation for all major chronostratigraphic boundaries is available to the velocity interpreter as a set of inflection points in the supplied central function velocity curve. The velocity model also provides an accurate velocity field for process testing during the early stages of seismic processing projects. The same conditioning technique can be used to spatially check the final primary velocity field prior to committing to stack and migration. A case history from the Timor Sea demonstrates the application of this technique. This historically difficult seismic area has used two and sometimes three passes of velocity analyses. The conditioned stacking velocity database allows the processing sequence to be reduced to a single post-demultiple/post-DMO quality control velocity analysis step for both reprocessing and new infill acquisition programs ? which yeilds a 50% reduction in the time spent picking and generating velocity analyses and eliminates the need to produce iso-velocity plots. The results show that proper conditioning of a stacking velocity database allows velocities to be recycled for improved quality and efficiency in pre-stack seismic processing, migration, and stacking velocity assisted depth conversion techniques.https://doi.org/10.1071/EG993671
© ASEG 1993