EAGE/SEG Summer Research Workshop - Processing and imaging of seismic data

Extracting information about the subsurface velocity from multi-offset data may require a two-fold strategy because the acquisition is sensitive to two different characteristic wavelengths of the velocity structure. The initial step is the reconstruction of the smooth part of the velocity structure by using first-arrival time tomography in contrast with the classical velocity analysis requiring strong subhorizontal interface contrasts

Multiples can be suppressed in the angle-domain image space after migration. Primaries and multiples have different angle-domain moveout, therefore they can be separated using techniques similar to the ones employed in the data space prior to migration. We use Radon transforms in the image space to discriminate between primaries and multiples. This method has the advantage of working with 3-D data and complex geology. It offers an alternative to the more expensive Delft approach.

Prestack reflection tomography is gradually being promoted from a method adjunct to the vertical velocity updating scheme, to become the main model building tool, used routinely on all imaging projects.

The use of the differential semblance misfit function on common image point gathers in the angle domain, lends itself to an automated tomographic approach through a gradient based search of the model space. The common image point gathers are created by the AVA-corrected or incomplete generalized Radon transform. We present a real data example of PP scattering events and an approach to regularize the optimization by respecting geological features.

Picking of the information in prestack data cube is certainly one of the most difficult tasks in velocity estimation methods. This difficulty is enhanced when dealing with converted waves since PP and PS events have usually to be paired in order to obtain two cinematically consistent velocity models.

Summary not available

New developments in seismic data integration ask more and more for tomographic inversion method able to handle data from alternative acquisition geometries.

For optimal imaging results, velocity model updates must be based on image gathers derived from the same migration engine that is used in the final imaging.

Over the years, travel-time inversion (TTI) has established itself within Shell as the preferred method to construct 3-D velocity models in depth for subsequent imaging purposes. Vast experience has been built up with layer based TTI and more recently its use in gridded or hybrid models has taken off. TT1 in gridded and hybrid models has been developed for geologies for which a layer-based description is less appropriate. In this presentation we present a number of case studies demonstrating the value of TII.