White convolutional model

Digital Imaging and Deconvolution: The ABCs of Seismic Exploration and Processing

Series	Geophysical References Series
Title	Digital Imaging and Deconvolution: The ABCs of Seismic Exploration and Processing
Author	Enders A. Robinson and Sven Treitel
Chapter	9
DOI	http://dx.doi.org/10.1190/1.9781560801610
ISBN	9781560801481
Store	SEG Online Store

What is the white convolutional model? The white convolutional model for the seismic trace x is

${\begin{aligned}x=w*\varepsilon .\end{aligned}}$

(28)

In this model, w is the field wavelet and $\varepsilon$ is a white-reflection-coefficient series. The field wavelet can be represented by the triple convolution

${\begin{aligned}w=s*i*a*m,\end{aligned}}$

(29)

where s is the source wavelet, a is the absorption filter, m is the overall signal representing the multiples, and i is the response of the receiving instruments. These instruments include the geophones or hydrophones, the amplifiers, and the modulation equipment. Often the actual source wavelet, the instrumentation response, and certain near-surface effects all can be lumped together under the rubric of signature wavelet. The field wavelet w is necessarily causal, but it will not necessarily be minimum phase unless all its components are minimum phase.

As seismic waves propagate through the earth, part of the wave energy is transformed into heat. This inelastic absorption phenomenon has been an active research topic for many years. Experimental results suggest that as a seismic wavelet travels through homogeneous rock, it systematically loses high frequencies to heat according to a process that can be represented by the action of a linear convolutional filter, the so-called absorption filter (see above). This absorption filter a simulates the real earth’s absorption response and has a magnitude spectrum approximately given by ${\rm {\ exp\ }}\left(-\alpha d\right)$, where $\alpha$ is the distance attenuation coefficient in units of reciprocal distance and d is the distance that the pulse has traveled. For most dry rocks, $\alpha$ is roughly proportional to the first power of frequency at frequencies used in the field and is given by $\alpha =\omega {\rm {/2}}vQ$, where $\omega$ is angular frequency, v is velocity, and Q is the seismic quality factor, which for most dry rocks is independent of $\omega$ in the seismic frequency band (Aki and Richards, 1980). In addition, it can be shown that the phase spectrum of the earth filter is minimum phase. This absorption response is approximately removable by a process we can call absorption deconvolution. It consists of the convolution of a given trace x with the inverse $a^{-1}$ of the minimum-phase absorption filter a.

Continue reading

Previous section	Next section
Spiking filter	Wavelet processing
Previous chapter	Next chapter
Synthetics	Deconvolution
Table of Contents (book)

Also in this chapter

• Wavelets
• The shaping filter
• Spiking filter
• Wavelet processing
• All-pass filter
• Convolutional model
• Nonminimum-delay wavelet
• Signature deconvolution
• Vibroseis
• Dual-sensor wavelet estimation
• Deconvolution: Einstein or predictive?
• Summary
• Appendix I: Exercises

External links

find literature about White convolutional model/en