Scribd
Upload
72 views3 pages

Inverse Q

The document discusses applying inverse Q filtering to land seismic data to improve signal quality where there is no VSP control data. Inverse Q filtering can remove the effects of dispersion and attenuation on seismic data by restoring higher frequencies, improving resolution. The authors designed a Q filter structure for real data using deterministic processing where signal improvement was needed. Applying the inverse Q filter increased signal bandwidth and amplitudes, enhancing data reliability.

Uploaded by

solomon
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
72 views3 pages

Inverse Q

The document discusses applying inverse Q filtering to land seismic data to improve signal quality where there is no VSP control data. Inverse Q filtering can remove the effects of dispersion and attenuation on seismic data by restoring higher frequencies, improving resolution. The authors designed a Q filter structure for real data using deterministic processing where signal improvement was needed. Applying the inverse Q filter increased signal bandwidth and amplitudes, enhancing data reliability.

Uploaded by

solomon
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

6th International Conference & Exposition on Petroleum Geophysics “Kolkata 2006”

Application of Inverse Q Filtering to Land Seismic Data to


Improve Signal where there is no VSP Data Control
R. L. Basak*, S. Shanmugasamy, R. V. S. Murthy, and A. K. Roy
ONGC, Geophysical Services, Chennai

Summary
The effect of dispersion and attenuation on land seismic data can be removed by applying inverse Q filtering thus
improving the seismic resolution. In this case study, the Q filter structure was designed by deterministic processing sequences
on real data where there is a need to improve signal to compensate for attenuation losses. This method is useful in areas
where there well data is available.

Introduction property of subsurface media (Kjartonson E., 1979).


Deducing lithological information like porosity,
An increasing loss of amplitude and phase permeability and viscosity from seismic data, as is heavily
distortion of the wavelet is normally seen in recorded seismic dependant on the inversion schemes, compensating for
data. The loss in amplitude is due to frequency dependent seismic attenuation assumes paramount importance.
attenuation while the phase distortion is due to dispersion.
The high frequency components of the spectrum of the To remove the attenuation effect, an inverse Q
wavelet travel with a higher velocity are more quickly filtering is designed and applied to seismic records (Bickel
attenuated and phase shifted than the low frequency ones. and Natarajan 1985; Hargreaves and Calvert 1991). The
Therefore, it is necessary to invert this attenuation by appropriate amplitude and phase deabsorption was achieved
restoring the higher frequencies to increase the resolution by establishing an empirical relation of Q values with the
of seismic data. An inverse Q filter when applied to data stacking velocities (Varela et al., 1993). Integration of zero
will increase the signal bandwidth and correct the signal offset well data with sonic log data provides the basis for
amplitudes thus improving the reliability of data. Q values determination ( Pramanik et al., 2000).

The earth filter, considered as a minimum phase Q values are traditionally estimated by measuring
process, could be represented by a Q model where Q denotes spectral ratios between two receivers straddling a constant
the quality factor of the medium. A forward Q filtering is Q through the direct downgoing wavefield of VSP data
used to test attenuation and dispersion effects generally on (Leaney, 1999). This method determines spectral ratios
synthetic data, and an inverse Q filtering is used to remove between all possible receiver pairs in the VSP resulting in a
them from seismic data .The application of this inverse plot of many Q values versus depth. These estimated values
filtering can not recover lost information, but can effectively of Q can be used in designing the inverse Q filter for
restore the natural balance of the remaining information. application to the seismic data to effectively compensate
attenuation losses.
The degree of frequency loss and phase distortion
Harris et al. (1997) found that Q was independent
are inversely related to the Q values of the medium. Large
of frequency over the seismic bandwidth for a North Sea
Q values imply low absorption, where as small Q values
VSP. In this presented study, we estimated Q structure by
imply large absorption. The Q factor depends on the velocity,
calculating the spectral ratio of a reference signal’s
the density and the bulk modulus of the medium. It can
amplitude spectrum with that of a data window in
vary from lithology to lithology.
overlapping time windows moving down the trace. To obtain
a good compromise between stability and resolution, we
Theoretical background used the multi-taper method (Thomson, 1982).

Seismic attenuation impacts the amplitude and Inverse Q filters aiming to compensate for both
wave shape of recorded seismic data. It is a fundamental amplitude and phase losses are not stable as they entail a

(713)
large amplification of high frequencies. To improve the
stability of inverse Q filters, constant gain limited operators
and high cut filters are used. By limiting the maximum gains
of the inverse filters, these operators are active only on the
spectrum in which signal dominates the noise.

In fact, it is never possible to completely correct


for the loss of high frequencies because of limitation in
dynamic range and the risk of over amplifying high
frequency noise.

A stable inverse filter was designed based on the


theory of wavefield downward continuation used in a layered
manner, assuming a depth-dependent, layered earth Q model
and the results were evaluated post application of inverse
Q filtering (Wang, 2002).
Figure (1-b): Amplitude spectrum after inverse Q filtering.

Application of inverse Q filtering on real data pass filtered with (5-10-80-95) bandwidth. Flattened spectrum
indicates compensation of loss of high frequencies, which
The 3D land data pertaining to Cauvery Basin was in turn results in wavelet compression and the phase
taken for this study where there were no spectral readings stability.
available from VSP data. Conditioning of raw data was done
by removal of high amplitude noise, application of statics Figures (2-a) and (2-b) show the time-variant am-
and surface consistent amplitude correction. plitude strength for different frequency components both
before and after inverse Q filtering indicating that ampli-
Random noise attenuation was done both before tudes have been sufficiently recovered after inverse Q filter
and after the application of inverse Q filtering followed by application.
gapped deconvolution. The prior made volume stack was
used to arrive at the Q structure for application of inverse Figure (3-a) shows the stack section before inverse
Q filter on pre-stack gathers. Q filtering and Figure (3-b) the same section after the
filtering. The inverse Q filtered section shows higher
Figures (1-a) and (1-b) testify that inverse Q dominant frequency than the other. The lateral coherence
filtering has flattened the amplitude spectrum within the of the events also is better in the filtered section.
signal band. The amplitude spectra were subsequently band-

Fig. (2-a): Amplitude curves before inverse Q filtering. ( Time win-


Figure (1-a): Amplitude spectrum before inverse Q filtering. dow 0 - 4000 msec., Amplitudes in centibel ).

(714)
6th International Conference & Exposition on Petroleum Geophysics “Kolkata 2006”

Fig. (2-b): Amplitude curves after inverse Q filtering. ( Time window 0


- 4000 msec., Amplitudes in centibel ).

Figure (3-b): Stack section after inverse Q filtering.

computation of Q structure. This method is suitable


especially when the signal to noise ratio is low.

Acknowledgements

The authors sincerely thank Shri D. Dutta, GM-


HGS and Shri Chaman Singh, DGM-SAOM for their help
and constant encouragement to carry out this work and
permitting to publish the same.

References

Bickel S.H. and Natarajan R.R., 1985. Plane-wave Q-


deconvolution: Geophysics 50,1426-1439.
Hargreaves N.D. and Calvert A.J., 1991. Inverse Q filtering by
Fourier transform: Geophysics 56, 519-527.
Harris,P.E.,et al.,1997. Multichannel estimation of frequency-
Figure (3-a): Stack section before inverse Q filtering. dependent Q from VSP data:Geophysical Prospecting,
45, 87-109.
Kjartonson E., 1979. Constant Q wave propagation and
Conclusions attenuation: J. Geophys. Res. 84, 4737-4748.
Thomson, D.J., 1982. Spectrum estimation and harmonic
The 3D land data processed with inverse Q filtering analysis. Proc IEEE, 70, 1055-1096.
not using any well data has shown significant improvement Pramanik et al., 2000. 70th Annual meeting and International
Exposition, Expanded Abstract, pp 2013-2016.
over the data processed without using inverse Q filter. The
Carlos Lopo Varela et al., 1993. Modeling of attenuation and
filtering has improved the amplitude spectrum, strengthened dispersion: Geophysics 58, No.8, 1167-1173.
the time-variant amplitudes of respective frequencies in the Wang Y., 2002. A stable and efficient approach of inverse Q
signal band, improved the S/N ratio, and thus increasing filtering: Geophysics 67, No.2.
the resolution in the stack section. Further, improvements Wang Y., 2002. The effectiveness of stable inverse Q filtering to
are possible with improved velocities that are necessary for land seismic. EAGE 64th Conf. & Exh - Florence.

(715)

You might also like