Migration from Topography

The traditional approach to handling data taken in areas of extreme (or even modest) topography is to statically correct the data to some datum, and then migrate as if the source and receivers were on the usually absolutely flat plane. It is quite easy to perform the migration directly from the topography. The major difficulty is not in making the algorithm handle the topography, but, instead, the difficulty arises in having accurate topographic data available and in being able to estimate an accurate near surface velocity field. What this section shows is that migrating from topography is not an issue.

Migration from topography requires the exact specification of the location and elevation of each and every source and receiver in the acquisition. Even though this information is routinely available, it is frequently not stored concomitant with the data, or it is lost after static corrections. Implementation of migration from topography requires only that wavefields be generated at the source or back-propagated from the receiver locations and elevations. Part (a) of Figure  9 shows a relatively simple model with relatively complex topography. Parts (b) and (c) show Kirchhoff and two-way images of these data. Visual comparison suggests that, again, the two-way approach is vastly superior to the Kirchhoff method.



Figure 9: Migration from topography.

(a) The BP (AMOCO) Canadian foothills topographic model PIC

(b) Kirchhoff migration of the BP tomographic data PIC (c) Full wavefield migration of the BP tomographic data PIC


 
Introduction
Seismic Modeling
History
Zero Offset Migration Algorithms
Exploding Reflector Examples
Prestack Migration
Prestack Migration Examples
Data Acquisition
Migration Summary
Isotropic Velocity Analysis
Anisotropic Velocity Analysis
Case Studies
Course Summary