Rock-physics Foundation for AVO
Analysis & Zoeppritz Equations
and their Approximations
Lecturer: Msc. Truong Quoc Thanh
Student: Vo Hoang Duy 1410637
Tran Ngoc Tu - 1450
ZOEPPRITZ EQUATIONS & THEIR
APPROXIMATIONS
Zoeppritz Equations
RP : reflected P
RS : reflected S
TP : transmitted P
TS : transmitted S
SOME APPROXIMATIONS
Bortfelt (1961)
Principle of this method:
1. Divide the major interface into
many smaller ones with small
changes in VP, VS and .
2. Use the approximated equation
to compute reflections and
transmission.
3. Integrate the results, including
all mode conversions and
multiples, setting time delays to
zero (delta z = 0).
SOME APPROXIMATIONS
Bortfelt (1961)
After subsitution in to Zoeppritz equations, we have the following
differential coefficients:
SOME APPROXIMATIONS
Bortfelt (1961)
For the reflected compressional wave:
Where,
The equation can also be rearranged as follows:
SOME APPROXIMATIONS
Bortfelt (1961)
This form of Bortfelds approximation shows a singularity when the
shear-wave velocities are the same.
According to Bortfeld, this approximation can be applied for all angles
and contrasts of the elastic parameters that occur normally with
reflection seismic data, with a small differential values
SOME APPROXIMATIONS
The Richards & Frasier approximation
Richards & Frasier (1976) and Aki & Richards (1980) derive the equation
for the reflection compressional wave in form of 3 terms: density, P-wave
velocity, shear-wave velocity.
SOME APPROXIMATIONS
The Richards & Frasier approximation
This figure compares
the exact solution of
Zoeppritz equation
(thick solid line) with
the Bortfeld
approximation (small
circles) and the Aki &
Richards
approximation (thin
line) for model shown
SOME APPROXIMATIONS
The Richards & Frasier approximation
Which is of the form A: the AVO intercept, the zero-offset
stack
B: the AVO slope or gradient (rate of
change of amplitude with time
C: curvature
SOME APPROXIMATIONS
The Richards & Frasier approximation
Note that
Chen et al. (2001b) showed that the curvature C deviates dramatically
from the true Zoeppritz result at the wide angles. Thus, they found that
increased accuracy can be achieved at large angles if
SOME APPROXIMATIONS
The Richards & Frasier approximation
By setting VS/VP=1/2 and ignore third term, Wiggins et al. (1983, 1985)
suggest that
SOME APPROXIMATIONS
The Shuey approximation
Shuey (1985a,b) just rearranges the Aki and Richards equation (in that the
terms are arranged according to angle-range contribution to the RC)
The first term is the contribution of the normal-incidence RC
The second term is related to its influences in the middle range of angles of incidence
(15-30)
The last term is the large vertical variations in velocity
SOME APPROXIMATIONS
The Shuey approximation
Re-arrange
SOME APPROXIMATIONS
The Shuey approximation
For a restricted angle range, we can linearize the equation by dropping the
third term:
If we neglect the effect of density on AVO:
SOME APPROXIMATIONS
The Hilterman approximation
Hilterman (1990) simplifies Shueys equation with some assumptions:
Assume that VP/VS = 2, and thus set v=1/3 which gives A0= -1.
Then, limit the angles: <30o, thus tan2 sin2
Simplified equation:
SOME APPROXIMATIONS
The Smith & Gidlow approximation
By using Gardners velocity-density empirical relationship to replace the
density change with the velocity change, Smith & Gidlow (1987) provide
another approximation
SOME APPROXIMATIONS
The Smith & Gidlow approximation
Gardners equation:
Take derivatives both sides
Substitute into Aki & Richards equation
SOME APPROXIMATIONS
The Smith & Gidlow approximation
When it is applied to NMO-corrected CMP gathers, using a least-squares
method which is achieved by a weighted stacking of the traces in the
gather
The offset and time-dependent weights applied to these traces are derived
from the ratio VP/VS, the angle of incidence and to fold of the data
The NMO-corrected traces in the gather are multiplied by the weights and
then summed.
SOME APPROXIMATIONS
The Smith & Gidlow approximation
SOME APPROXIMATIONS
The Smith & Gidlow approximation
For an anomalous hydrocarbon application, Smith & Gidlow define a residual as
the fluid factor:
Note that Smith & Gidlow only use mud-rock equation to justify the form of the
fluid-factor equation.
In fact, the coefficient 1.16 can be replaced by a time-varying variable to
minimize for background reflection. That process has been called geostack,
deviation from background, and universal hydrocarbon indicator
SOME APPROXIMATIONS
The Fatti et al. approximation
Fatti at al. Approximation can be considered in case Gardners relationship
is not valid for particular case.
Fatti et al. Approximation rearrange the Aki & Richards equation as
SOME APPROXIMATIONS
The Fatti et al. approximation
Note that this equation solves for P- and S-impedance reflectivities. For
small angles, because , so the third term is much smaller and
can be neglected
This approximation makes no assumption for VP/VS or density, and is
good out to large precritical angles.
SUMMARY
