** Next:** Conclusions
** Up:** SYNTHETIC TESTS
** Previous:** SYNTHETIC TESTS

As mentioned earlier, least-squares Kirchhoff migration
involves using only a single slice out of a much larger
transform domain cube. If the rest of the cube contained
no interesting information, then the extra computations
of the least-squares method would not be justified. However,
the rest of the transform domain cube does serve a purpose.
Figure shows a single slice taken from two
such cubes. This is for a non least-squares example.
The *x* coordinate is fixed at the location
of the single point scatterer; the axes are *t* and *t*_{0}.
The cube has been shifted so that the single slice that
corresponds to the migrated image lies along the top of
the cube. On the left is the result of migrating with the
correct constant velocity; we see energy focusing near the
top of the plot. On the right is the result of migrating
with too high a velocity (3.5 km/sec instead of 3 km/sec).
The focusing of energy away from the *t*=*t*_{0} plane tells
us that an incorrect migration velocity was used. The
cube contains important information, similar to a
migration velocity analysis, away from the image plane.
This gives us reason to think that the added cost of the least-squares
method might be worthwhile.
**focus
**

Figure 9 Left, migration with correct velocity.
Slice for a single *x* coordinate, at the scatterer location.
Focus is on the *t*=*t*_{0} line, which has been shifted to lie along
the top of the plot.
Right, migration with incorrect velocity gives focus that is
away from the *t*=*t*_{0} line.

** Next:** Conclusions
** Up:** SYNTHETIC TESTS
** Previous:** SYNTHETIC TESTS
Stanford Exploration Project

11/17/1997