NAME

     avgain - apply temporally and  spatially  averaged  gain  to
     trace data


SYNOPSIS

     avgain [ -Nntap ] [ -Ootap ] [ -rsirs ] [ -reire ] [ -sscale
     factor  ]  [  -wwindow  ]  [  -tsteptstep  ]  [  -mdmode ] [
     -llsmooth ] [ -ststacked data ] [ -agc ] [ -V ] [ -? ] [  -H
     ]


DESCRIPTION

     AVGAIN attempts to balance amplitudes  temporally  and  spa-
     tially by computing and applying gain control to trace data,
     on a record-by-record basis, based on  the  temporally-  and
     spatially-smoothed   arithmetic  average  (AAA),  root  mean
     square (RMS), geometric mean (GMAA), or median of the  abso-
     lute values of the trace amplitudes within a sliding window.
     The window is fixed in length by the user and moves down the
     trace  in  user-specified  steps.  Mean or median values are
     computed for the center sample of each window position,  and
     these values are linearly interpolated between temporal con-
     trol points for temporal  smoothing.  Spatial  smoothing  is
     optional, but if desired, it is accomplished before temporal
     smoothing by isotime convolution of the  average  amplitudes
     at  each  time  window  center  with either a smoothed (sine
     squared) window or with a simple boxcar operator,  which  is
     equivalent  to  a  simple  running  sum.  The  lenght of the
     smoothing operator is controlled by  the  spatial  smoothing
     length  parameter.   Scale  factors  are  computed  from the
     smoothed mean or median values.

     AVGAIN was created to handle data  in  an  unstacked  domain
     (common  shot,  common  receiver,  common  offset, or common
     depth point) and expects data in one of these  domains  when
     applying  spatial  smoothing to the gain functions. That is,
     if spatial smoothing is desired, the data is expected to  be
     in  the  format of one or more N-trace records.  However, it
     is frequently desirable to apply smoothed scaling to stacked
     data,  so  the  program  will  allow  the  user, via the -st
     option, to tell the program to treat a stacked data set as a
     single N-trace record rather than N 1-trace records, as gen-
     erally output by program STACK, so the smoothing option will
     work as anticipated.

     Every attempt is made to ignore mute zones and dead  traces,
     and  muting  before  execution  of  the  program is strongly
     recommended.

     Avgain gets both its data and its  parameters  from  command
     line  arguments.  These arguments specify the input, output,
     the amplitude scale factor ,  temporal  and  spatial  window
     parameters, and smoothing parameters.

  Command line arguments
     -Nntap
          Enter the input data set name or file immediately after
          typing  -N;  entering  nothing  specifies a stdin pipe.
          This input put file should include  the  complete  path
          name  if the file resides in a directory other than the
          one  from  which  you  start  the  program.    Example:
          -N/work_area/dummy  tells  the program to look for file
          "dummy" in directory "work_area".

     -Ootap
          Enter the output data  set  name  or  file  immediately
          after typing -O.  This output file is not required when
          piping the output to another process.  The output  data
          set also requires the full path name (see above).

     -rs irs
          (integer) First (sequential) record  to  process.   All
          records  prior  to this one are skipped and not written
          to the output data set. Default = 1 (the first).

     -re ire
          (integer) Last (sequential)  record  to  process.   All
          records  after  this are ignored and not written to the
          output data set. Default = process all.

     -sscale factor
          Enter the scale factor that specifies  what  percentage
          of  2047  you  want  the  average  absolute  amplitude.
          (default = 15%)

     -wwindow
          Enter the sliding window length, in ms.  Default = 1000
          ms.

     -tsteptstep
          Enter the step size, as % of window  length,  by  which
          the  window  is  moved  for  each  gain  control point.
          Default = 50%.

     -llsmooth
          Enter the length, in traces, of the running sum used to
          smooth  the  gain spatially.  Default = 0 or no spatial
          smoothing.

     -mdmode
          Enter the value which specifies the type of gain compu-
          tation  desired.  Valid values are "0" for AAA, "1" for
          RMS, "2" for GMAA, and "3" for median.  Default = 0.

     -ststacked data
          Enter the command  line  argument  '-st'  to  tell  the
          program to treat the input a one N-trace record (when N
          is the lineheader entry for number of  records).   This
          option  will  allow the smoothing option to function as
          anticipated for stacked data.  Default = No.

     -agc Enter the command line argument '-agc'  to  output  the
          gain  curves only.  These curves will be in the form of
          seismic traces which have the exact same headers as the
          input traces from which they were computed.  The curves
          may be applied to the data with program vmult.

     -box Enter the command line argument  '-box'  to  cause  the
          program  to use a boxcar, or simple running sum, opera-
          tor for spatial smoothing.  Otherwise, a smoothed (sine
          squared) operator is used.  Default = No.

     -V   Enter the command line argument '-V' to get  additional
          printout.

     -?   Enter this (or '-?' if you are running in  c-shell)  to
          get  online  help.   The  program  will terminate after
          printing this.

     -H   Enter this to get online help.  The program  will  ter-
          minate after printing this.


BUGS

     Unknown


AUTHOR

     Richard Crider, Houston x7070


COPYRIGHT

     copyright 2001, Amoco Production Company
               All Rights Reserved
          an affiliate of BP America Inc.

















Man(1) output converted with man2html