PERMFILE50081

• Permit No: M1999002
• IBM Index Class Name: Permit File
• Doc Date: 6/19/1999
• Doc Name: OBJECTORS EXHIBITS
• Media Type: D
• Archive: No

,::
.~
hip'sto'tk -
n chalk,
using (or
d lowei -
~werPe
~ dogleg -
the ta~-
verPek -
c upper,
cks into '
peraloi '
than `3'
'm fgUr ;
-. ~.
chalk
ne
ud
3utla
Mown
e hole,
of the
igle bi
motor
ifican
t0 IfIE
to sur
m, call
.-well
'ion is
ftware
-- ..a. p. ,-. n
s N N
41 •' N ~ ~ ~' O N
., N .f.d. m o a m
~'tiI ~•4 (~!-iii '~~~:!! ~c
& . ~ - I}t_
=• ..~ p
- y {_ Yi{ f:
.. ,_.. ~-) •.,`` --"'ate,.. r_.. ~~ ~ry ...
~.,,~... .=su.:

. G~•.
Al.:, ='~
tivih• (Fig. 76).
Ne+t Conte 3D Dh10 and band pass
fifterng.
Tte final stack from these sups,
applied to each line across the sun•ev
at 500 m increments, produces an
'usage with further improvements in
the base of salt and subsalt refectiuns
(Fig, 7c).
The interpreter now has a velocity
field in time with which to stack all the
data in the sun~ev.
Aiigration of all the data in time
folloics.
Cut a problem arises at this point:
Fig 13
N
N
~,~1
Jt_. {;~
t
.~
~~ 1 ~_ti~.) ~(
_~ ;.
_. -p- ..

3D time migration algorithms assume
a velocih~ field that is fairly consistent,
irhich the earlier steps have shown
not to ba the case within and below
salt.
The solution is to edit out the base-
of-soli and out-of-plane picks on the
3D D\t0 yelocih~ field. Those picks
were needed to stack the data at the
base of salt but result in velocities that
change too abruptly to be used in 3D
time migration.
In Fig. Sa, velocities on a 3D D~[O
velocity field at '_5 sec show wide
variation because of the salt sill. Fig.
86 is n cross section from the data
sample.
Fig. Sc is the same velocity field
^fter hvo editing passes that removed
the base-of-salt and out-of-plane ve-
IUIIIIC'S.
The related cross section in Fig. Sd
shows how the editing made celodty
changes across the field much more
gradual and therefore useful in time
migration.
A final step smooths the data even
more, producing velocities that are
close enough to earth velocities to
work as a first guess at the velocity
field for 3D PSD,bt. Sir or seven itera-
tions will be needed at that point.
Fig. 9 sho+vs final3D time migration
sections from trvo lines in the Culf of
plexiccis Ship Shoal South Addition
area: one over the tvtahogany prospect
and the other over the Alexandrite
prospect on Block 337.
The images are good, with excellent
tops and bases of salt; moreover, they
presen•e reFledivity below salt. But
events are not located properly be-
cause the data have not been migrated
in depth.
lime vs. depth migration
Fig. 10 shows how the ~-lahogany
salt sill image changes when process-
ing moves from ?D time migration
(Fig. ]Oa) to 2D prestack depth migra-
tion (Fig. 106).
Imaging of the base of salt and
subsalt reflections is much better on
the depth-migrated section than it is
on the time-migrated section.
,also, the direction of structural dip
changes belo+y salt behceen the incor-
rect time image and the correct depth
image.
Oi course, accurate salt and subsalt
Fig Ia
.~- st