next up previous print clean
Next: Acknowledgment Up: Examples Previous: Discussion

Conclusions

Subsalt imaging is one of the most challenging problems of modern seismic imaging because the sharp and irregular salt-sediments interface causes multipathing and uneven illumination. Wavefield-continuation migration methods produce high-quality images under salt, but the estimation of the migration velocity function in the subsalt is an unresolved problem. Conventional MVA methods based on traveltimes computed by ray tracing often fail to provide reliable velocity estimates because ray tracing is unstable and sensitive to the fine details of the salt-sediment interface.

In this chapter, I demonstrate that the wave-equation migration velocity analysis (WEMVA) method presented in wemva overcomes many of the problems encountered by ray-based MVA methods when estimating velocity under salt. I use a complex and realistic subsalt datasets to test this methodology. I also illustrate with numerical examples that wavepaths computed by wavefield extrapolation are robust with respect to shadow zones, and that they model the finite-frequency wave propagation that occurs in such environments better than rays do. I demonstrate that velocity errors can be effectively measured by residual migration scans. These scans provide useful velocity information almost in all the subsalt areas, although the reliability of these measurements decreases where poor illumination drastically deteriorates the quality of the angle-domain common image gathers.

I demonstrate that WEMVA is capable of overcoming the limitations of the first-order Born approximation, by testing the convergence of WEMVA in presence of large velocity anomalies. The magnitude and spatial extents of the anomalies are such that reflectors in the migrated images shift by several wavelengths. Notwithstanding these large shifts, WEMVA converges to an accurate approximation of the true velocity function.

I also demonstrate that WEMVA is applicable to velocity analysis using focusing of un-collapsed diffractions. This application makes use of information that is commonly overlooked in traditional migration velocity analysis. This application shows potential for imaging subsalt, but also for imaging datasets where moveout information is not readily available, e.g. typical Ground-Penetrating Radar data.


next up previous print clean
Next: Acknowledgment Up: Examples Previous: Discussion
Stanford Exploration Project
11/4/2004