Re-picking Velocities - Conflicting NMO due to backscattering

As mentioned before, one issue that we observed in tape 20 is the conflicting NMO directions  in CMP gathers 2400 to 4000 due to back scattering coming from adjacent structures . We tried to improve the stacking velocities along these CMP gathers. The goal is that reflections coming  directly from the 2D line are stacked constructively while backscattering coming from adjacent structures stack destructively. For that, we re-picked velocities following the NMO with increasing offset. Fig. \ref{967683} shows the comparison between  the NMO corrected CMP gathers with the initial velocities (left) and re-picked velocities (right) . Offset increases to the right. Notice that we improved  specially the flattening of the first reflector. Notice also  the crossing of reflectors with opposite moveouts. Fig. \ref{853220} shows the comparison between Kirchhoff migrated images with the initial stacking velocities (left)  and with the re-picked stacking velocities (right). Notice that the image has improved specially at the flanks of the structural high (around stacks 2600 and 3100). However there are still artifacts. Later  in the report we suggest as future work to apply pre-stack migration to overcome this issue.