Prestack Imaging

The ability to perform Prestack Time Migration with relative independence from a velocity model has been a key component of our success in processing structurally complex data. The workflow involves a velocity independent DMO followed by Prestack Imaging, which is an implementation of prestack migration that allows the analysis and construction of the velocity model after the computational burden of imaging. This creates a true migration velocity, at every trace and sample, which is used either to complete the migration with Prestack Imaging or to drive a Kirchhoff or other prestack migration process.

Automatic Imaging is the latest implementation of this technology; software that automatically, and in detail creates both the velocity model and the prestack migrated image of the 3D volume. Cluster computing allows this to happen quickly with greater accuracy and repeatability than conventional human interactive analysis can perform.

If you would like us to send you a power point presentation of this technology, please contact us.

For more information, please visit http://www.data-modeling.com.

Comments are closed.

To compare results, roll over one of the data images below.
To see an enlargement of the results, click on one of the images below.

To demonstrate the reduction in solution space, we can look at the repeatability of the migration velocity solution when different starting velocities are used. The data set used is the Benjamin data set, an overthrust data from Canada. On the right, the middle image is the usual starting migration velocity field, which is the migrated DMO velocities. This velocity changes laterally and in time. The other starting migration velocity, shown on the top right, was built from a single velocity profile (and was not the same as any velocity profile in the first velocity field). The difference in velocity between these two starting velocities is significant (as high as 10%), as shown in the bottom section.

V(t)

V(x,t)

Velocity Difference

After 9 iterations, we compare the resulting velocities from the two starting points. Visually, the two velocitites appear to be very similar. Let's look at this more carefully with velocity differences, shown on the bottom section. On the corners, up shallow, where we have very little data, the differences are as high as 8%. However, throughout the rest of the section, the differences are very small, generally on the order of 1% or less.

V(t)

V(x,t)

Velocity Difference