Tomography for Near-Surface and Imaging Velocity Determination -- Examples from the Western Desert in Egypt

By

 John Bedingfield¹, August Lau¹, Chris Koeninger², Nabil El Kady², Maurice Nessim²

(1) Apache Egypt Company, Maadi, Egypt (2) WesternGeco, Cairo, Egypt

 The accurate determination of the velocity distribution in the subsurface is probably the single most important part in the long sequence of processing steps. In land surveys velocities impact the noise patterns, the amplitude behaviour, the static and dynamic distortions. In general processing attempts to correct for these effects with velocities that are individually tuned.

Noise attenuation and amplitude correction velocities are spatially smooth or even constant. Near-Surface velocities for static corrections require much more detail if their effects on the timing of structures are to be reliably resolved. The velocity model that is used for depth imaging generally undergoes various iterations and is the most difficult yet also the most important part of the velocity puzzle.

This paper will concentrate mainly on the near-surface and depth imaging velocity models and the tomographic approach to derive them. For the near-surface model first-arrival times are raytraced and inverted for slowness updates in a cell-based model. Similarly for the depth imaging case a cell-based model is used to compute slowness updates derived from user-supplied velocity information. This velocity information is much easier to obtain and more readily available in routine processing than reflection time picks from prestack data which requires good data quality and is very time consuming.

Examples of the techniques are shown on data from the Western Desert in Egypt with a discussion of specific issues that were encountered during processing.