NAME
vmodvavg - Velocity Perturbation Model Generation
SYNOPSIS
vmodvavg [-Nntap] [-Ootap] [-Ccardfile] [-V]
DESCRIPTION
vmodvavg generates a V(t) suite which can be stretched to
depth by program vttd and used for prestack depth migration
in program cubevz or poststack V(z) migration on the Connec-
tion Machine. In the typical vmodvavg flow a V(z) trace is
extracted from an input V(x,z) dataset using program editt
and converted to time using program vttd. vmodvavg reads
this V(t) trace and converts it to average velocity vs.
time, perturbs in average velocity as directed by user input
cards, converts the suite back to V(t) and outputs it.
Another vttd converts the V(t) suite to a V(z) suite. To
setup a canned vmodvavg flow type:
catpat vmodvavg > vmodvavg.flow
Edit the flow with a text editor, filling in the input
V(x,z) dataset name, the trace number to extract and an out-
put dataset name, save it, and type:
startjob vmodvavg.flow &
When using the canned flow, all command line arguments will
be substituted into the vmodvavg step for you, but you do
need to fill out the card deck.
Command line arguments
-N ntap
Enter the name of the input V(t) dataset after typing
-N.
-O otap
Enter the name of the output V(t) suite dataset after
typing -O.
-V Enter -V for a verbose printout.
-Ccardfile
If the parameters are in a separate card file enter the
name of that file after typing -C, or by using the cat-
pat command a script file will be created which will
contain the command line arguments and the other neces-
sary parameters for executing vmodvavg. The script
file may then be editted to enter the correct values.
The script file may be executed using the "startjob"
command.
Script file format:
vmodvavg -Nntap -Ootap \
-vmodvavg.crd"\
VEL MODEL # FUNCTS VBLOCK T/D LOW/HI
nfunc vblock timdep lowhi
VMULT TIME/DPTH VMULT
VMULT 0 1.05
VMULT 3000 1.1
VMULT 20000 1.2
"
# FUNCTS nfunc
Number of V(t) functions to produce. Assume that
LOW/HI flag is 3 for a symmetrical suite. Actual out-
put functions can be limited by setting LOW/HI to 1 or
2.
VBLOCK vblock
Allowable error (in velocity units). The velocity
functions will be blocked into constant velocity layers
such that the error in the velocity values is no
greater than VBLOCK. Velocity blocking is used for
greater efficiency in the migrations.
T/D timdep
1 -- input velocities are a function of time
2 -- input velocities are a function of depth
LOW/HI lowhi
A flag that selects functions to be output.
1 -- input function and the lower half of the suite
2 -- input function and the higher half of the suite
3 -- (default) symmetrical suite about input function
NOTE: do NOT adjust #FUNCTS to compensate for 1 or 2.
TIME/DPTH timedpth
Time value for velocity multipliers int the VMULT card.
(if T/D is 2, then this is a depth value)
VMULT vmult
Ratio of maximum average velocity to the input average
velocity at the current time/depth. A symmetric func-
tion is also produced that is slower than the input by
this ratio. The intervening functions are produced by
linear interpolation in average velocity vs. time
between the minimum, input and maximum functions.
SEE ALSO
cubevz, vmod, fsi, cubevz.flow, apkr.flow, apkr, p2iv
BUGS
None known.
AUTHOR
David Nelson
COPYRIGHT
copyright 2001, Amoco Production Company
All Rights Reserved
an affiliate of BP America Inc.
Man(1) output converted with
man2html