GEPH316 Field Data: Stacking

GEPH316 Seismic Data Processing

Field Data Workshop 2000

19 Stacking


Previous: Residual Statics	Up: Contents	Next: AGC

Stacking NMO corrected gathers

In the previous section, the line was corrected for NMO normal moveout. The effects of offset distance have been removed.
The traces in each of the CDP point gathers should now show horizontal events for reflections, so that each trace now corresponds to the single trace model.
The traces in each of the CDP gathers should now be identical samplings of the reflecting interfaces, but with different sets of random and geological noise, corresponding to the different shot and geophone positions, and the different ray paths for the increasing offsets.
The traces for each of the CDP gathers can be combined to form the desired single trace, with the signal to noise ratio improved by n^0.5, where n is the number of traces in the gather and thus the fold of the data.

Stacking the line

Seismic unix provides the sustack command for stacking traces which share a particular parameter. Check the operation of the command using:

sustack

The sustack command performs a crude summation stack of the data, without any corrections unless requested.
The main trace selection parameter is the value of the cdp keyword and the line has already been sorted according to this parameter.
Because the fold of the data varies between the early, middle and later gathers, the stacked traces will not all have similar amplitudes. Therefore, it is useful to normalise the power in the traces using the normpow=1.0 parameter.
Now stack the appropriate data traces in the line and do a wiggle plot of the line. Because the number of traces is greatly reduced, a wiggle plot should be satisfactory for showing the variations along the line.

sustack < line.nmo > line.stack key=cdp normpow=1.0
suxwigb < line.stack title='Final stacked line' windowtitle='Final stacked line' label1='Time (s)' label2='Trace = CDP position - 6 (m)' &

Note that the fold of the plot will vary along the line, so the quality of the data will vary, with the 11 traces at each of the line being of reduced quality.
The first event may have been produced either from the first reflection or from the refraction event. The undulations should show the depth behaviour of the main boundary between the first and second layers.
Try to interpret the physical behaviour of the boundary.
The subleties of the reflecting interfaces are not emphasised by the wiggle plot. More features can be seen on the xmovie plot.

suxmovie < line.stack title='Final stacked line' windowtitle='Final stacked line' label1='Time (s)' label2='Trace = CDP position - 6 (m)' &

It appears possible to seem some evidence of later reflection events on this xmovie display.
The two versions of the line may be printed using the following commands.

supswigp  < line.stack title='Final stacked line'  label1='Time (s)' label2='Trace = CDP position - 6 (m)' | lp -d mono
supsmovie < line.stack title='Final stacked line'  label1='Time (s)' label2='Trace = CDP position - 6 (m)' | lp -d mono

Remove the unwanted graphical windows and files.

zap xmovie
zap xwigb
ls


Previous: Residual Statics	Up: Contents	Next: AGC

Go to <--->