Depth migration of reflected and refracted energy

Tomography and forward modelling based on ray-tracing tend to smooth the first order discontinuities. Major tectonic elements and structures can be derived by this process and positions of the major faults delineated. Small scale tectonic features, however, are difficult to delineate and cannot be resolved. Velocity values are very accurately mapped (usually within 5 %) and depth to the first order discontinuities can in most cases be located within ±100 m. However, reflectivity of boundaries is occasionally poorly mapped, when lateral inhomogeneities are intense and structures can easily be overseen. This shortcoming is compensated by applying the "pre-stack" migration.

The computation includes the complete wavefield considering also the refracted energy, usualy muted in commercialy available software packages. In this way detailed structures of the interfaces between different layers can be resolved also in areas with poor normal incidence reflectivity. The final product therefore, is obtained when the migrated sections and the velocity-depth models derived by forward modelling are superimposed with the depth migrated sections and compared.

In order to have a better understanding and control of the depth migration we used three different results of the velocity models and compared the migrated sections. That is we used the firstbreak migration tomography velocities, the non linear tomography velocities and the final velocity model and migrated the CSGs three times. Since the reliability of the prestack migration is a function of the data quality and the correct velocity model we wanted to see what is the difference if we migrate using the three different velocity solutions and how this influences the migration.