next up previous print clean
Next: Acknowledgments Up: Subsalt reservoir monitoring: Ayeni Previous: Discussion

Conclusions

Sub-salt reservoir monitoring of hydrocarbon reservoirs is a major challenge. We propose the computation of the time-lapse image either as a difference between inverse images or as a solution to a least-squares inverse problem. By solving for the time-lapse response through inversion (rather than a difference between two adjoint operations) it may be possible to obtain more reliable amplitude information and to correct for geometry and other effects which have heretofore limited seismic monitoring of sub-salt reservoirs. Suitable regularization and/or preconditioning may be used to better constrain the inversion result and to potentially reduce the required number of iterations in the inversion scheme. It may also be possible to structure the regularization operator to account for geomechamical and other changes outside the reservoir.

 
GboyeImg
GboyeImg
Figure 3
Migrated Images for velocity models in Figure [*]. Survey direction: right to left. Data migrated using the correct (smoothed) velocity models. (a) and (b) Baseline; (c) and (d) Monitor. Notice the poor illumination of the reflectors close to the salt flank (region marked 'X' in (b) and (d).
view burn build edit restore

 
GboyeImg2
GboyeImg2
Figure 4
Migrated Images for velocity models in Figure [*]. Survey direction: left to right. Data migrated using the correct (smoothed) velocity models. (a) and (b) Baseline; (c) and (d) Monitor. Note the poor illumination of the reflectors close to the salt flank (region marked 'X' in (b) and (d).
view burn build edit restore

 
GboyeFB
GboyeFB
Figure 5
Raw difference Images (with and without the salt wedge) for same (a and b) and opposite (c and d) acquisition directions. Data migrated with the correct velocity models. Note the weak time-lapse response close to the reservoir (a result of the poor illumination) in the region marked 'A'.
view burn build edit restore

 
GboyeSFB
GboyeSFB
Figure 6
Raw difference Images (with and without the salt wedge) for same (a and b) and opposite (c and d) acquisition directions. Data migrated with the smoothed baseline velocity models. Note the differences in migration artifacts between Figures [*]b and [*]b in the regions marked 'B' and 'C'.
view burn build edit restore


next up previous print clean
Next: Acknowledgments Up: Subsalt reservoir monitoring: Ayeni Previous: Discussion
Stanford Exploration Project
5/6/2007