NAME
fxsmig - 2D V(x,z) Prestack Shot Migration from Topography
using an FX Algorithm
SYNOPSIS
[-Nntap] [-Ootap] [-VELvtap] [-GEOMgeom] [-CTABctab] [-
BRirec1] [-ERirec2] [-IRincrec] [-BSismp1] [-ESismp2] [-
PSnspad] [-NXPADnxpad] [-NXWINnxwin] [-NXOUTnxout] [-
NSYMnsym] [-DTidt] [-ff1] [-fcf2] [-FCf3] [-Ff4] [-DXdx] [-
DZdz] [-ZMAXzmax] [-ZMULzmul] [-NAPnap] [-TMUTEtmute] [-
VMUTEvmute] [-MAXAPnap_max] [-LAMBDAxlam] [-V] [-?] [-h]
DESCRIPTION
fxsmig performs a 2D V(x,z) prestack shot migration from
topography using an FX algorithm. fxsmig reads shot
records, a velocity tape and a geometry file created with
program prepmig. Program fxprep is run before fxsmig to
prepare the shot records for migration. Refer to the manual
page on fxprep. Program fxsns is run immediately after the
migration to create the migrated stacked section.
Command line arguments
-Nntap
After typing -N enter the input dataset name of the
seismic data which is the output from fxprep. This
input file should include the complete path name if the
file resides in a different directory. Example
-N/b/vsp/dummy tells the program to look for file
'dummy' in directory 'vsp' stored on the 'b' disk. A
stdin pipe is assumed when this flag is missing from
the command line.
-Ootap
Enter the output dataset name after typing -O. A stdout
pipe is assumed when this flag is missing from the com-
mand line.
-VELvtap
Enter the input dataset name of the velocity model
after typing -VEL. This cannot be a pipe when the
seismic data is a pipe.
-GEOMgeom
Enter the input dataset name of the geometry data after
typing -GEOM. This dataset is the geometry output from
program fxprep. Refer to the manual page on fxprep.
-CTABctab
If this command line parameter is present fxsmig will
calculate the extrapolator table and write it into a
file using the name given. If a table by this name
already exists fxsmig will read this table. If this
command line parameter is not present fxsmig will cal-
culate the extrapolator table internally and will NOT
write it into a file. This option is useful because the
table may be calculated once and used in subsequent
runs permitting the overhead for calculating the table
to be paid only once. A header is written in the table
to keep track of certain parameters that were used when
building the table. When reusing the table, the pro-
gram will abort if these key parameters have been
changed since building the table.
-BRirec1
This value is the first record to process. Default is
1.
-ERirec2
This value is the last record to process. Default is
all records.
-IRincrec
This value is the record increment. Default is 1. When
piping from fxprep into fxsmig, the three flags -BR,
-ER, and -IR for skipping records should be used ONLY
in fxprep. These parameters are ignored by fxsmig when
piping from fxprep.
-BSismp1
This value is the first sample to process. Default is
1.
-ESismp2
This value is the last sample to process. Default is
all samples.
-PSnspad
This value is the number of zero samples to add at the
end of each trace. The default is 0.
-NXPADnxpad
This value is the number of zero traces to add at the
beginning and at the end of each record. A total of
2*nxpad zero traces will be added. The default is 0.
-NXWINnxwin
This value gives the window in x for which the migra-
tion is performed. The program rounds this down to the
nearest integer. For example, if nxwin = 241, there
will be 120 stations each to the left and right of the
source, plus the 1 station at the source. If nsym=0,
only stations which have actual recorded data are used
and the actual window is truncated if the source is
near the left or right edge of the data. The default
(and maximum) value is nxwin=ntrc, the total number of
traces in the input dataset record. The default is
number of traces per record + 1.
-NXOUTnxout
This parameter gives the window in x for which the
results of the migration are output. The program rounds
this down to the nearest integer. For example, if nxout
= 241, there will be 120 stations each to the left and
right of the source, plus the 1 station at the source.
Only stations which have actual recorded data are out-
put and the actual window is truncated if the source is
near the left or right edge of the data. The default
(and maximum) value is nxout=nxwin. Note the output
data set contains all the data packed to the left of
the record.
-NSYMnsym
This controls whether or not an attempt is made to
always put a symmetric window in x about the source
during the migration. A value of nsym=0 (the default)
makes no attempt; a value of nsym=1 turns this on.
This makes a difference only for records in which the
shot is close to the left or right edges of the data.
On output the data is packed to the left of the output
record, and the extra zero-traces padded to perform the
migration are not output.
-DTidt
The value represents the time sample rate. If a value
is given it overrides the value given in the input
dataset line header. Default is input line header.
-ff1 The value is the low cutoff frequency in hertz for the
trapezoidal filter. The default is 0.
-fcf2
This value is the low corner frequency in hertz for the
trapezoidal filter. The default is f1.
-FCf3
This value is the high corner frequency in hertz for
the trapezoidal filter. The default is f4.
-Ff4 This value is the high cutoff frequency in hertz for
the trapezoidal filter. The default is nyquist -
delta_omega.
-DXdx
This value is the trace spacing. The default is the
Dx1000 value in the velocity field line header divided
by 1000.
-DZdz
This value is the output depth sample spacing. The
default is the Dz1000 value in the velocity field line
header divided by 1000. This parameter must agree with
the actual delta-z spacing of the velocity input model.
-ZMAXzmax
This value is the total depth for the migration.
Default is determined by the depth of the input velo-
city dataset.
-ZMULzmul
This value is a multiple of the depth spacing and it
controls the multi-gridding in the z-direction during
the migration. The default is 1, which maintains the
z-gridding of the input velocity dataset. If ZMUL is 0,
the maximum possible value for delta-z is computed by:
dz_maximum = (1/4) * minimum_velocity / maximum_frequency
Then the multi-gridding value is obtained as the ratio
of dz_maximum to the original delta-z of the input
velocity dataset. If ZMUL is 2 the program uses a
delta-z that is 2 times the delta-z of the input velo-
city dataset. If this value exceeds the maximum
theoretically possible value (dz_maximum) the program
issues a warning but lets the user violate the con-
straint. fxsmig resamples the data before output to
restore the original delta-z spacing.
-MAXAPmaxap
This value determines the size of the half-aperture
when building the extrapolator table. It is recommended
to use a larger number for building the table and use a
much smaller value for -NAP which is the half-aperture
size to use for the migration. See description below.
-TMUTEtmute
This value represents time and is used with the vmute
parameter below to perform a mute on the input data.
The default is 0.0 (no mute).
-VMUTEvmute
This value represents velocity and is used with the
tmute parameter above to perform a mute on the input
data. The default is 0.0 (no mute).
-NAPnap
This value gives the half-aperture width to use for
each frequency and may be much less than the value
given for MAXAP. Larger values of NAP give sharper
resolution, but the CPU cost is proportional to the
value of NAP. By using values of LAMBDA greater than
zero the half-aperture width can be extended which will
result in seeing more features at larger angles with
respect to the horizontal and at distances farther from
the shot. However, as LAMBDA increases for a fixed NAP
and more features become visible, the resolution drops.
Larger NAP values mean higher resolution and higher
cost. Larger LAMBDA values mean seeing further and at
bigger angles with lower resolution and lower cost.
-LAMBDAxlam
This value gives the minimum number of wavelengths per
half-aperture to use. The default is 0.0.
-V Enter -V for a verbose printout.
-h Enter -h for online help. The program will terminate
after printing this.
-? Enter -? for online help. The program will terminate
after printing this.
SEE ALSO
mbs, fxprep, prepmig, fxsns
BUGS
None known.
AUTHOR
N. D. Whitmore, Jr.; Ron D. Coleman, CETech; J. Cooperstein,
CETech
COPYRIGHT
copyright 2001, Amoco Production Company
All Rights Reserved
an affiliate of BP America Inc.
Man(1) output converted with
man2html