Key #4: Data Processing

We must make sense of the recorded seismic 'squiggles' to produce the truest possible image of the Earth's sub-surface geologic structure. Reflected seismic response is a mixture of our output pulse, the effect of the Earth upon that pulse, and background noise, all convolved together.

We must remove the output pulse and the noise to leave just the 'Earth model'. This is the role of seismic data processing, which requires accuracy, reliability, speed and substantial computing power. The advanced mathematical algorithms and complex geophysical processes applied to 3D seismic data require enormous computing resources.

Not to mention the massive volumes of data involved.

For example, the amount of seismic data recorded by CGGVeritas during just ONE medium-sized marine 3D survey would fill more than 20,000 compact disks, forming a stack over 650 feet high!

Processing: Deconvolution

Ideal seismic response would be a single sharp reflection for each sub-surface rock layer boundary. Actual seismic response is less than ideal because our output pulse is not perfectly sharp and changes its shape while passing through the Earth.

Deconvolution 'deconvolves' our output pulse from the seismic response and converts it into a cleaner, sharper, less confusing pulse.

Can you determine the number of rock layers here by examining the actual seismic response (before deconvolution)?

Data Processing: Stacking

Seismic traces from the same reflecting point are gathered together (CRP gather) and summed, or 'stacked'.

The six seismic traces on the left are from the same reflecting point. As the traces are merged into one (right), background noise cancels itself out while the seismic signals add together, producing a stronger signal-to-noise ratio. (The output trace on the right is shown here six times only to provide a better comparison.)

The more of these seismic traces we can stack together into one output trace, the clearer the seismic image.

This first image shows a seismic section produced after the seismic traces have been sorted, adjusted for varying path lengths and signal strength, and stacked.

Here, each trace is the summation of 48 individual 'shot' traces.

Note the water bottom 'multiple' reflection (arrowed) -- a seismic 'echo' of the seafloor caused by energy bouncing back-and-forth within the water layer to produce a 'false' reflection obscuring the real data.

This second image shows the result of suppressing the water bottom multiple.

The seismic image is enhanced by a process that suppresses the multiple without harming real reflections.

This third image is further enhanced by 'focusing' energy for both flat and steep reflectors.

Any missing traces are 'filled in' by interpolation.

This fourth image most closely resembles the true sub-surface geology.

A process called 'migration' moves reflected energy to its true sub-surface position of origin.

Advanced Data Processing

More advanced processing techniques, such as Prestack Depth Migration (PSDM), can significantly improve seismic imaging, especially in areas of complex geology.

In this example from the Gulf of Mexico, see how PSDM has improved the imaging of a) a massive salt body, and b) sedimentary layers beneath the salt.

Processes such as PSDM take more time, expertise and resources to apply, but accurate 3D seismic images can mean the difference between success or an expensive dry hole.

In-Field Data Processing

Our customers usually need the data delivered as fast as possible!

In fact, today's industry demands for ever-faster turnaround of seismic projects necessitates that data now be processed, at least to a preliminary stage, in the field immediately after recording.

This requires equipment and personnel in the field to be almost as sophisticated as those onshore.