The effectiveness of processing seismic data and the accuracy of its interpretation depend very strongly on how well the velocities of propagation in the area are known. Fortunately, good digital analysis techniques which can be used to derive velocity information from the reflection data itself are available from a number of sources. As velocity data have become more readily available, other use s outside of seismic processing have developed. Engineers are using it to predict locations of high pressure zones that might be encountered in drilling. Geologists have sometimes been able to estimate lithology with it.
The procedure generally used in making a velocity analysis is to measure the moveout of all consistent events on a set of common depth point traces and then !o as sign a velocity value to each one of these which would have produced the observed moveout (assuming simple straight-ray path reflections). Such a complete picture of all the energy on the record within the velocity range of the analysis provides a wealth of valuable information but at the same time requires careful interpretation.
In order to examine the more common situations encountered in typical seismic surveys, a number of sets of synthetic traces have been constructed and analyzed for velocity based on well data from offshore Louisiana. Pure, simple primary reflections were considered first. Next, errors in reference datum and static corrections were introduced. Finally, the model was generalized in successive steps to include dip, diffractions, water bottom multiples, reverberations, and surface multiples. The analysis plots show the types of effects to be expected in each of these cases.
During the last few years velocity analysis of seismic reflection data has become an important part of most seismic data processing systems. For offshore work, this use has been especially great because adequate well control is almost never available.
Velocity information is needed first of all to carry out the seismic data processing job itself. Since raw reflection data corresponding to depths of current exploration interest are frequently contaminated with or even dominated by various types of extraneous energy, it has been found helpful literally to focus upon the desired primary reflection signals.1 This focusing operation is achieved by means of common depth point stacking. Normal moveout corrections based on average velocity determine how the stack will be made.
Finally after all processing has been completed, the seismic data are converted from travel times to depths through the use of average velocity information.
Other applications of velocity analysis developed when it became known that a new kind of measurement of an actual physical property of earth materials was available. Geologists are now predicting lithology at depth,2 and engineers are locating high pressure zones that may prove troublesome when a well is drilled.3, 4
It has been our experience that the step between acquiring a velocity analysis and making a meaningful application may not always be a simple one. There are several reasons why this is so.