NAME

     tvd  - module to compute [time, velocity, dip]  tweak  func-
     tions  using an input XSD pickfile model of structure and an
     input TDFN velocity model.  These functions are used by pro-
     gram  tvdin  and tvdnmo to apply the calculated tweak to the
     data in the CEP Demult process.


SYNOPSIS

     tvd  [  -Mdipmodel  ]  [  -vvelocitymodel  ]  [  -v0v0  ]  [
     -msosmoothop  ] [ -mssmodelshotstart ] [ -mnsmodelnumshots ]
     [   -msimodelshotint   ]   [   -assanalysisstartshot   ]   [
     -aseanalysisendshot ] [ -asianalysisincshot ] [ -tmaxmaxtime
     ] [ -gapgap ] [ -gigroupinterval ]  [  -sishotinterval  ]  [
     -nchannumchannels   ]   [  -dirboatdirection  ]  [  -domend-
     pointdomain ] [ -twtweak ] [ -opcoffsetpercent ] [ -V ]


DESCRIPTION

     tvd given an input dip  and  velocity  model,  computes  the
     velocity and dip that will nmo shift the model events into a
     given quadrant of fk space. The time  amount  of  the  shift
     (ms)  at  about  the half offset point is referred to as the
     (tweak).  An event that is  tweaked  up  and  away  from  an
     observer  standing  at the source is termed a (green) tweak.
     An event tweaked down and away from  the  same  observer  is
     termed  a  (red) tweak.  The dip model is supplied as an XSD
     pickfile customarily picked from a neartrace display of  the
     data  being  processed.  Indexing in this file is treated as
     sequential.  The velocity model is supplied as a TDFN format
     function  using  the  same sequential indexing as assumed in
     the dip model.  This was done to  make  life  easy  for  the
     user.   One  must simply plot in XSD the near traces of your
     data, pick the control dip model for  the  events  that  you
     wish to tweak, and supply a simple velocity file to go along
     with it.  This can be readily created from preexisting  data
     or  by  running  a  velspec  analysis on the data to be pro-
     cessed.  Since the process is completely run from a model no
     actual  USP dataset need be attached.  Of course some infor-
     mation regarding that dataset is required such as  the  shot
     interval,   group  interval  and  gap  associated  with  the
     acquisition.  Also required is the maximum time of the  data
     being  processed.   This will ensure that the functions gen-
     erated span the entire dataset for application through tvdin
     and tvdnmo.

  Command line arguments
     -M dipmodel
          Enter the name of  the  XSD  pickfile  containing  your
          input  dip  model.   This would be picked from a simple
          near trace display of your time data.   The  model  may
          contain as many or as few interfaces as you wish.  Hor-
          izons need not extend across the entire model.

     -v vvelocity model
          Enter the name of the TDFN format velocity model.  This
          model  should span the entire dataset.  If not a static
          extrapolation will be done [the outermost value will be
          extrapolated to the edges.

     -v0 v0   [default = 1500]
          Enter the velocity at time zero in ft or m/sec  if  not
          already present on your input velocity model.

     -mso smoothop   [default = not used]
          Enter the smoothing order to use if you wish to  smooth
          the horizons prior to extraction of dip information.

     -mss modelshotstart   [default = 1]
          Enter the numeric shot for the first location  in  your
          model.   This  is usually unity as most sequential sets
          will start from one.

     -mns modelnumshots   [required]
          Enter the number of shots present in  your  model.   If
          you have picked your model from a near trace display of
          your endpoint data this will be the number  of  records
          in the dataset to be processed.  What we want here is a
          model that traverses your entire dataset.

     -msi modelshotint   [default = 1]
          Enter the shot interval used in your model.  Unless you
          have displayed every other record or some such thing in
          XSD this will generally  be  unity  and  hence  may  be
          defaulted.

     -ass analysisstartshot   [default = 1]
          Enter the shot location of the first  desired  analysis
          location at which an output [t,v,d] tweak function will
          be calculated.

     -ase analysisendshot   [default = all in model]
          Enter the shot location of the  last  desired  analysis
          location at which an output [t,v,d] tweak function will
          be calculated.

     -asi analysisendshot   [default = 1]
          Enter the shot interval at which tweak  functions  will
          be calculated.

     -tmax maxtime   [required]
          Enter  the  maximum  time  [in  milliseconds]  of  your
          dataset for which the functions are being generagted.

     -gap gap   [default = 0]
          Enter the distance from the source to  near  offset  in
          the units of the dataset.

     -gi groupinterval   [default = 12.5]
          Enter the group interval of the data to be processed in
          the units of the data.

     -si shotinterval   [default = 25.0]
          Enter the shot interval of the data to be processed  in
          the units of the data.

     -nchan numchannels   [default = 240]
          Enter the number of channels in spread of the  data  to
          be processed.  This is usually the number of traces per
          record of the dataset from which the model was derived.

     -dir dir   [default = rl]
          Enter rl if the shooting direction  is  from  right  to
          left  on  the  model or lr if the acquisition direction
          was left to right on the model.  This parameter  [along
          with  -dom below] allows the routine to determine which
          side of the [T-x] diagram to be basing the calculations
          on.

     -dom dom   [default = shot]
          Enter the common endpoint domain  that  the  associated
          dataset  is  sorted  to.   Valid  entries  are shot and
          receiver. This parameter [along with -dir above] allows
          the  routine  to  determine  which  side  of  the [T-x]
          diagram to be basing the calculations on.

     -tw tweak   [default = 25ms]
          Enter the desired tweak in milliseconds.  This  is  the
          approximate  time  [at  the  half offset distance] that
          the model events will be moved above [or below] flat.

     -opc offsetpercent   [default = not used]
          Enter percent of the  spread  to  use  in  calculation.
          Note: This parameter is only usefull when experimenting
          with tvd in areas of  severe  structural  complexity  (
          where  CEP  Demult  is of limited use ) and should nor-
          mally be defaulted.  If you are coming up on a  syncli-
          nal  axis  you may want to make this a small percentage
          of the spread so that the dip  used  at  each  analysis
          location is more representative of the near offset sec-
          tion of the record.  In this routine there is no facil-
          ity  to handle variations in dip within a single record
          along a single event.

     -V   Include on command line to get verbose printout.


HOW TO



     An outline of procedure would look like this:

      - Plot in XSD a near trace display of the data to  be  pro-
     cessed.

      - Pick a control dip model from this display.  This may  be
     a model of primaries or a model of multiples.

      - Generate a TDFN velocity function set over the extent  of
     the model.

      - Run tvd to generate as many tweak functions  as  required
     for your purposes.

      - The output of tvd is converted to  velocity,  dip  traces
     using  the  USP  routine  tvdin.  This is very much the same
     approach as using velin to associate a velocity dataset with
     a  USP  seismic  datset.The difference here is that an addi-
     tional dip trace is also created.

      - Use the USP routine, tvdnmo, to apply the tweak functions
     to the endpoint data.

      - Filter the tweaked data with you choice  of  filters  [to
     attack either primary or multiple]

      - Remove the tweak function through the inverse application
     of tvdnmo.

      - Apply nmo to the data using tvdnmo with  the  dip  multi-
     plier set to zero.

      - Examine the result and determine in a time varient  sense
     the set of tweaks to be used to process your data.



SEE ALSO

     xsdtweak sunlay sunform sunhybrd tvdin tvdnmo qdchop bdimage
     headerscan fetch usparg


AUTHOR

      P.G.A.Garossino, Dec,1997
      (original code by Garossino & Bush, Cairo, 1988)
      (original USP implimentation by Paul Gutowski, 1989)


COPYRIGHT

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



Man(1) output converted with man2html