Shallow-origin FEAVO effects are visible in the image domain

Figure b shows a pattern of slanted symmetrical linear patterns (Kjartansson ``V''s) that are the signature of FEAVO in the midpoint-offset domain. Figure illustrates the way the patterns formed. The Kjartansson ``V''s are a good way to discriminate the FEAVO anomalies in an inversion process. If we decide that the inversion discrimination will be done in the image domain, we need to make sure that the anomalies will not be destroyed by downward continuation. This was already shown on a synthetic in Fig. 5 of (). In order to show that the V-shaped patterns are visible as well in angle domain images produced from a real dataset, we migrated the Grand Isle prestack dataset then transformed to angle domain. The migration velocity depends only on depth (Figure ), but not on midpoint (to avoid focusing by migration with an inappropriate velocity model). This assumption is close to the truth - Figure shows that the geology is quite flat in the area. An examination of the midpoint-angle slices (Figure ) reveals V-shaped patterns at the same locations as the data domain ones.

 

vint Figure 4 Interval velocity used in the migration of the Grand Isle dataset. The velocity does not depend on midpoint in order to not inadverdently focus the energy (the purpose of the migration is seeing whether the focusing anomalies are preserved). The v(z) assumption is close to geological reality in that area too.

 

kimag
Figure 5 Migration (stack) of the Grand Isle dataset. The structure is quite flat.

 

angpanel
Figure 6 FEAVO anomalies in midpoint-angle depth slices. Each vertical panel shows the angle range between 0 and 20 degrees. The first panel (z=1km), the last (z=3.35km) and to a certain extent the second (z=2km) do not contain Kjartansson ``V''s