Designature

DP START Oslo
2004
 At the end of this Session, you will be able to:

• Define the equation for the convolutional model

• Define the term deterministic deconvolution.
• Describe how a signature can be affected by changes in each of the
following : gun depth, cable depth, gun drop outs, air pressure leaks.
• Define the term 'zero phase conversion'
• Define the term 'minimum phase conversion
• Define the term CMS
• Produce a schematic diagram that shows the processing steps involved in
derivation, application and QC of designature on a shot by shot basis.

DP START Oslo
2004
 The Role of Seismic Source

•A seismic reflector can only reflect back to the surface an image of the energy pulse
it receives.

• If we send a complex pulse into the ground, that pulse will be superimposed on
every reflector we record.

DP START Oslo
2004
The principle of Superposition

For any linear system (such as earth) and for any input waveform, the output waveform is
the superposition of all the impulse-responses obtained by regarding the input waveform
as a succession of spikes.

In other words, what we record as a seismic trace, is achieved computationally by

convolving the basic wavelet (the signal we send into the earth) with the reflectivity
series (layer boundaries) of subsurface earth.
DP START Oslo
2004
Spectrum of a spike

Amplitude

Phase
0

0
Time Frequency Frequency

Seismic signal generated by an explosive charge is very nearly a spike

DP START Oslo
2004
Ideal vs Reality
The recorded seismic trace may be modelled as a series of interactions
between the source signature (a finite, band-limited wavelet) and the earth.

Actual
seismogram

A plot of the reflection

coefficients of the layers
as a function of two way
travel time.

Sedimentary Reflectivity
Column Function
DP START Oslo
2004
The Convolutional Model
A seismic trace X(t) is given by the convolution of the basic seismic
wavelet W(t) with the reflectivity series r(t) plus random noise N(t).

X(t) = W(t) * R(t) + N(t)

Basic seismic
wavelet

Seismic trace Reflectivity series Random Noise

DP START Oslo
2004
Components of the basic seismic wavelet
The wavelet is the combination of all the individual components that
affect the seismic wave

For example :

ω (t) = S(t) * SG(t) * R(t) * RG(t) * I(t) * E(t) * M(t) * A(t) * P(t) * ….Etc

But, what are all of these individual components?

DP START Oslo
2004
Components of the basic seismic wavelet

ω (t) = S(t) * SG(t) * R(t) * RG(t) * I(t) * E(t) * M(t) * A(t) * P(t) * ….Etc

S(t) ; Source

ω (t) = S(t) * SG(t) * R(t) * RG(t) * I(t) * E(t) * M(t) * A(t) * P(t) * ….Etc

SG(t) ; Source Ghost

ω (t) = S(t) * SG(t) * R(t) * RG(t) * I(t) * E(t) * M(t) * A(t) * P(t) * ….Etc

DP START Oslo
R(t) ; Receiver
2004
Components of the basic seismic wavelet

ω (t) = S(t) * SG(t) * R(t) * RG(t) * I(t) * E(t) * M(t) * A(t) * P(t) * ….Etc

RG(t) ; Receiver Ghost

ω (t) = S(t) * SG(t) * R(t) * RG(t) * I(t) * E(t) * M(t) * A(t) * P(t) * ….Etc

I(t) ; Recording Instrument

ω (t) = S(t) * SG(t) * R(t) * RG(t) * I(t) * E(t) * M(t) * A(t) * P(t) * ….Etc

DP START Oslo
E(t) ; Earth Filter (Q)
2004
Components of the basic seismic wavelet

ω (t) = S(t) * SG(t) * R(t) * RG(t) * I(t) * E(t) * M(t) * A(t) * P(t) * ….Etc

M(t) ; Multiples

ω (t) = S(t) * SG(t) * R(t) * RG(t) * I(t) * E(t) * M(t) * A(t) * P(t) * ….Etc

A(t) ; Array Effects

ω (t) = S(t) * SG(t) * R(t) * RG(t) * I(t) * E(t) * M(t) * A(t) * P(t) * ….Etc

P(t) ; Processing Effects

DP START Oslo
2004
Deterministic Deconvolution
Deconvolution where part of the seismic system is known. No random
elements are involved. For example, where the source wavelet is
accurately known we can do source signature deconvolution.

X(t) = S(t) * E(t) * R(t) + N(t)

source system random noise
seismic trace reflectivity
signature wavelet
assumed
to be zero

If we have the recorded source signature then we can do deterministic

deconvolution (‘de-signature’) to remove S(t) from the equation

DP START Oslo
2004
Source Signature
If we are to remove the effect of our source signature then we must know
exactly what our source signature is. Typical practice is for the
manufacturers to measure the gun signature and then supply this signature
to the acquisition companies. Can you see any problem with this?
Yes
Will our source signature be exactly the same every time the guns fire?
No

You can have air pressure leakage, gun dropouts, cable ghosts, gun ghosts.

DP START Oslo
2004
Gun and Cable Notch Frequencies
Remembering our fourier transform pairs.
Depending on the depth of your gun and cable
you may see a notch (blip) in your amplitude
spectrum. We can calculate the frequency at
which this notch will appear using the
following equation.

750 / Shot Depth Hz 750 / Receiver Depth Hz

6m depth gives notch at 125Hz 5m depth gives notch at 150Hz

We try to set the depth of our guns and cables so that their notch
frequency will not occur in the frequency range of our seismic data.
DP START Oslo
2004
Typical Marine Signature

DP START Oslo
2004
Derivation and Application of Designature Operator

Input Seismic Data Input farfield Signature

(shot ordered)

Input Desired output

Derive Matching operator

Apply operator to Seismic

DP START Oslo
2004
Zero and Minimum Phase Conversion
A Zero Phase wavelet is symmetric about time 0. It has energy before time 0
and so is not physically realisable.

A Minimum Phase wavelet has its energy concentrated at the front end of the
pulse, but has no energy before time 0.

DP START Oslo
2004
 CMS Processing
CMS (Calibrated Marine Source)
•CMS allows source-by-source designature to a chosen output wavelet. For
each airgun in each array, a near field hydrophone measures the pressure
field when the gun goes off. A notional signature is derived for each airgun, it
can be thought of as the signature of an individual gun as if it had been fired
in isolation, removing the interactions of neighbouring guns. The notional
signatures for each airgun in the array are combined to generate the far-field
response for each shot. This farfield signature is recorded on an auxiliary
trace in the SEG-D record for that shot.

•The CMS Processing involves three stages

•QC of Far field Signatures
•Generation of Desired Output
•Generation & Application of source-by-source designature operator

DP START Oslo
2004
QC of Far Field Signatures

•Several source attributes are extracted from each far-field signature: array
volume, primary peak and ghost amplitudes and times, primary bubble ratio,
central frequency.
•These are graphically displayed together with the far-field signatures to allow a
visual QC for each sail line
•“Bad” Signatures can be replaced by a neighbour or an average signature
DP START Oslo
2004
Generation of Desired Output

Far Field signature after removal of ghosts and debubble Desired Zero-phase output

•Using the Omega Application SWP (Seismic Wavelet Processor)

•SWP can be used to remove cables ghosts and bubble effects, it can also be
used to create a minimum or zero phase output as desired
•Finally it generates the actual designature operator which is applied to each
shot to give the desired output

DP START Oslo
2004
Derivation and Application of Operator for CMS

Input Seismic Data Input CMS Signatures

(shot ordered) (shot ordered)

Input Desired output

Derive Matching operator

for each shot

Apply operator to Seismic

(based on exact shotpoint match)

DP START Oslo
2004
 At the end of this Session, you will be able to:

• Define the equation for the convolutional model

• Define the term deterministic deconvolution.
• Describe how a signature can be affected by changes in each of the
following : gun depth, cable depth, gun drop outs, air pressure leaks.
• Define the term 'zero phase conversion'
• Define the term 'minimum phase conversion
• Define the term CMS
• Produce a schematic diagram that shows the processing steps involved in
derivation, application and QC of designature on a shot by shot basis.

DP START Oslo
2004