Swell filter of Edgetech GeoStar data

Plot of raw data.         Plot with swell filter.         Plot with filter and gains.

Step 1:    Create a SEG-Y file
gstar2xstar 23_18_01.10

The program may error on the last trace if the file is incomplete.
File yr2003day274-1338z.gstar was created.

Step 2:   Convert the data to "normal" amplitude data.
sioseis << eof
procs diskin xstar diskoa end
diskin
   ipath yr2003day274-1338z.gstar end
end
xstar
   type 0 end
end
diskoa
   opath yr2003day274-1338z.segy end
end
end


Step 3:   Plot the data.
sioseis << eof
procs diskin plot end
diskin
   ipath yr2003day274-1338z.segy end
end
plot
   dptr 1 colors .001 gray3 .002 gray4 .003 gray5 .004 gray6 .005 gray7
   opath siofil nsecs .01 wiggle 0 ann gmtint anninc 1
  nibs 2859 vscale 300 tlines .01 .05 .1
  def .005 clip .01 trpin 300
  end
end
end
eof
sio2sun siofil sunfil.ras
xloadimage -r 90 sunfil.ras &

Step4:   Plot a small potion of the data to determine the swell period.
sioseis << eof
procs diskin plot end
diskin
   fgmt 2344 lgmt 2348 allno no
   ipath yr2003day274-1338z.segy end
end
diskoa
    opath short.segy end
end
plot
   scalar 100
   dptr 1 colors .001 gray3 .002 gray4 .003 gray5 .004 gray6 .005 gray7
   opath siofil nsecs .005 wiggle 0 ann gmtint anninc 1
  nibs 2859 vscale 500 tlines .001 .005 .01
  def .01 clip .005 trpin 300
  end
end
end
eof
sio2sun siofil sunfil.ras
xloadimage -r 90 sunfil.ras &


Step 5:  Find the amplitude of the water bottom.
sioseis << eof
procs diskin prout end
diskin
   fno 9663 lno 9664 ipath short.segy allno no end
end
prout
   sets 0 .002 fno 0 lno 99999 end
end
end



Step 6:   Based on a different web example, my first try was:
sioseis << eof
procs diskin wbt header mix header2 shift plot end
diskin
   fno 9663 lno 9750 allno no
   ipath short.segy end
end
wbt
  thres 0.1E-03 track .005
  end
end
header
    fno 0 lno 999999 r59 = r50 end
end
mix
    type 4 hdr 59 weight 1 1 1 1 1 1 1 1 1 1 end
end
header2
    fno 0 lno 999999 ftr 0 ltr 999 r60 = r59 - r50 end
end
shift
   fno 0 lno 999999 indices r60 end
end
prout
   fno 0 lno 999999 ftr 0 ltr 999
   indices l3 l4 r50 r59 r60 end
end
plot
   dptr 1 colors .001 gray3 .002 gray4 .003 gray5 .004 gray6 .005 gray7
   opath siofil nsecs .01 wiggle 0 ann gmtint anninc 1
  nibs 2859 vscale 300 tlines .001 .005 .01
  def .005 clip .01 trpin 300
  end
end
end
eof
sio2sun siofil sunfil.ras
xloadimage -r 90 sunfil.ras &

 thres 0.1E-03 track .005   led to same mispicks and bad shifts.
Narrow the search window by using track .0002.
Decrease the pick threshold because it was picking noise.
Use:
   thres 0.5E-04 track .0002

 Some of the swell remains, so increse the period a little.
Use:
    type 4 hdr 59 weight 1 1 1 1 1 1 1 1 1 1 1 1 end



Step 6:  Play with bandpass filter and gain.
sioseis << eof
procs diskin wbt header mix header2 shift filter gains plot end
diskin
   ipath short.segy end
end
wbt
  thres 0.5E-04 track .0002
  end
end
header
    fno 0 lno 999999 r59 = r50 end
end
mix
    type 4 hdr 59 weight 1 1 1 1 1 1 1 1 1 1 1 1 end
end
header2
    fno 0 lno 999999 ftr 0 ltr 999 r60 = r59 - r50 end
end
shift
   fno 0 lno 999999 indices r60 end
end
filter
   ftype 0 dbdrop 48 pass 1000 10000 end
end
gains
   type 3 alpha 1.1 end
end
plot
   dptr 1 colors .001 gray3 .002 gray4 .003 gray5 .004 gray6 .005 gray7
   opath siofil nsecs .005 wiggle 0 ann gmtint anninc 1
  nibs 2859 vscale 300 tlines .001 .005 .01
  def .04 clip .01 trpin 300
  end
end
end
eof
sio2sun siofil sunfil.ras
xloadimage -r 90 sunfil.ras &