AVO Workshop – Part 1

HRS10
Table of Contents - Part 1

Overview of the AVO process

Exercise 1 : The Colony Gas Sand – Setting up the project

Rock Physics & Fluid Replacement Modeling

Exercise 2 : The Colony Gas Sand – Biot-Gassmann analysis and RPT

AVO Theory and Zoeppritz Modeling

Exercise 3 : The Colony Gas Sand – Creating Zoeppritz Synthetics

Elastic Waves and Anisotropy

Exercise 4 : The Colony Gas Sand – Elastic wave synthetics

AVO Analysis on Seismic Data

Exercise 5 : The Colony Gas Sand – Calculating AVO Attributes

Cross Plotting AVO Attributes and the Third Order Term

Exercise 6 : The Colony Gas Sand – Cross Plotting AVO Attributes

Appendices

2
Table of Contents – Part 2
AVO Inversion - Elastic Impedance
Exercise 7 : The Colony Gas Sand – Extended elastic impedance
AVO Inversion - Independent AVO Inversion
AVO Inversion - Simultaneous Inversion
Exercise 8 : The Colony Gas Sand – Simultaneous Inversion
AVO Inversion - Lambda Mu Rho Theory
Exercise 9 : The Colony Gas Sand – LMR application
Processing Issues in AVO
Exercise 10 : Gulf Coast – Data Preparation
AVO Modeling Summary
Exercise 11: Gulf Coast – AVO Modeling
AVO Inversion - Some Practical Issues
Exercise 12: Gulf Coast – AVO Inversion
Summary
References
Appendices
3
Overview of the AVO Process

 This tutorial is a brief introduction to the Amplitude Variations

with Offset, or Amplitude Versus Offset (AVO) method.

 We will briefly review how the interpretation of seismic data has

changed through the years.

 We will then look at why AVO was an important step forward for
the interpretation of hydrocarbon anomalies.

 Finally, we will show why the AVO response is closely linked to

the rock physics of the reservoir.

4
A Seismic Section

The figure above shows a stacked seismic section recorded over the shallow
Cretaceous in Alberta. How would you interpret this section?

5
 Structural Interpretation

Your eye may first go to an anticlinal seismic event between 630 and 640 ms. Here, it
has been picked and called H1. A seismic interpreter prior to 1970 would have looked
only at structure and perhaps have located a well at CDP 330.
6
 Gas Well Location

And, in this case, he or she would have been right! A successful gas well was drilled
at that location. The figure above shows the sonic log, integrated to time, spliced on
the section. The gas sand top and base are shown as black lines on the log.
7
 “Bright Spots”

But this would have been a lucky guess, since structure alone does not tell you that a
gas sand is present. A geophysicist in the 1970’s would have based the well on the
fact that there is a “bright spot” visible on the seismic section, as indicated above.
8
 What is a “Bright Spot”?

Geology Seismic
Surface

Seismic
raypath

Shale r1 V1 r 2V2  r1V1

R0 
r2 V2
r 2V2  r1V1
Gas Sand

Interface at
Reflection at time Seismic
depth = d
t = 2d/V1 Wavelet

To understand “bright spots”, recall the definition of the zero-offset reflection

coefficient, shown in the figure above. R0 , the reflection coefficient, is the amplitude
of the seismic trough shown. Note also that the product of density, r, and P-wave
velocity, V, is called acoustic impedance.
9
 Gardner’s results for GOM

This figure, from Difference between shale and gas

Gardner et al. (1974), sand velocity at shallow depth.
shows a big difference
between shale and gas
sand velocity at
shallow depths in the
Gulf of Mexico. The
paper also derived the
“Gardner” equation,
which states that
density and velocity are
related by the equation
r = 0.23 V 0.25
Thus, we would expect
a large reflection
coefficient, or “bright
spot”, for shallow gas
sands.
10
The AVO Method

“Bright spots” can

be caused by
lithologic variations
as well as gas
sands.
Geophysicists in
the 1980’s looked at
pre-stack seismic
data and found that
amplitude change
with offset could be
used to explain gas
sands (Ostrander,
1984). This example
is a Class 3 gas
sand, which we will
discuss later.

11
What causes the AVO Effect?
Surface

q2 q1
q3

r1 VP1 VS1 Reflector

r2 VP2 VS2

The traces in a seismic gather reflect from the subsurface at increasing

angles of incidence q. The first order approximation to the reflection
coefficients as a function of angle is given by adding a second term to the
zero-offset reflection coefficient:
R(q )  R0  B sin2 q
B is a gradient term which produces the AVO effect. It is dependent on
changes in density, r, P-wave velocity, VP, and S-wave velocity, VS.
12
Why is S-wave Velocity Important?
As just shown, the
gradient term is
dependent on density, P
and S-wave velocity. The
plot on the left shows P
and S-wave velocity plot
as a function of gas
saturation (100% gas
saturation = 0% Water
Saturation), computed
with the Biot-Gassmann
equations. Note that P-
wave velocity drops
dramatically, but S-wave
velocity only increases
slightly (why?). This will
be discussed in the next
chapter.
13
AVO Modeling
Poisson’s
P-wave Density S-wave Synthetic Offset Stack
ratio

Based on AVO theory and the rock physics of the reservoir, we can perform AVO
modeling, as shown above. Note that the model result is a fairly good match to the
offset stack. Poisson’s ratio is a function of Vp/Vs ratio and will be discussed in the
next chapter.
14
 AVO Attributes
Intercept: A

Gradient: B
AVO Attributes are
used to analyze
large volumes of
seismic data,
looking for
hydrocarbon
anomalies.
15
 Cross-Plotting of Attributes

Gradient (B) One of the AVO methods that we will be

discussing later in the course involves
cross-plotting the zero-offset reflection
coefficient (R0, usually called A), versus the
gradient (B), as shown on the left.

As seen in the figure below, the highlighted

zones correspond to the top of gas sand
(pink), base of gas sand (yellow), and a hard
Intercept (A)
streak below the gas sand (blue).

16
 AVO Inversion
A new tool combines
inversion with AVO
Analysis to enhance the
reservoir discrimination.
Here, we have inverted for
P-impedance and Vp/Vs
ratio, cross-plotted and
identified a gas sand.

Gas
Sand
17
 Summary of AVO Methodology
Input NMO-corrected Gathers

Perform optimum processing sequence

Modeling Recon Methods Inversion

Rock Physics
Modeling Partial Intercept
Stacks Gradient Elastic Simultaneous
Impedance Inversion

Wave Eq. Zoeppritz Cross

LMR
Synthetics Synthetics Plot

18
Conclusions

 Seismic interpretation has evolved over the years, from strictly

structural interpretation, through “bright spot” identification, to
direct hydrocarbon detection using AVO.

 In this course we will elaborate on the ideas that have been

presented in this short introduction.

 As a starting point, the next chapter will discuss the principles of

rock physics in more detail.

 We will then move to AVO modeling.

 Finally, we will look at AVO analysis on real seismic data.

 In each case, we will first look at the theory and then perform a
workstation example.

19
Exercise 1
The Colony Gas Sand
Setting up the project
 Exercise 1

Our first set of exercises comes from the Colony sand formation, a Cretaceous sand
from Western Canada.
Poisson’s
P-wave Density S-wave Ratio
The target is a thin, 8 meter
thick, gas sand.

We have a single well with

measured sonic and
density logs.

The sand is known to

contain 50% water, and 50%
gas.

In this exercise, we will set

up the project and read in
the data.

21
Exercise 1

Start the HRS9 Geoview program by clicking

Geoview icon on your desktop:

When you launch

Geoview, the first
window that you see
contains a list of any
projects previously
opened in Geoview.
Your list will be blank
if this is the first time
you are running
Geoview.

22
 Exercise 1
For this exercise, we will start a new
project. Before doing that, it will be
helpful to set all the data paths to point
to the location where we have stored
the workshop data. To do that, click the
Settings tab:

To change all of the directories to the same location, click on

the option Set all default directories and then click the button
to the right:

Then, in the File

Selection Dialog,
select the folder which
contains the workshop
data:
23
Exercise 1

After setting all three paths, the Geoview window will now show
the selected directories (note that yours may be different):

When you
have finished
setting all the
paths, click
Apply to store
these paths:

24
Exercise 1

Now click the Projects tab and choose the option to create a
New Project:

25
Exercise 1

A dialog appears,
where we set the
project name. We
will call it colony, as
shown. Enter the
project name and
click OK on that
menu:

26
Exercise 1

Now a dialog appears, asking you the name of the database to use for
this project:

The database is used to store all the

wells used in this project. By default,
Geoview creates a new database, with
the same name as the project and
located in the same directory. For
example, this project is called
colony.prj, so the default database will
be called colony.wdb. This is desirable
since we were starting a new project,
intending to read in well logs from
external files. Occasionally, we may
want to use an existing database, which
has wells already stored. Then we
would click on the option Specify
database. For this exercise, click OK to
accept the default database name.
27
Exercise 1

The Geoview Window now looks like this:

28
Exercise 1: Loading the well log data

One part of the Geoview window

(called the Project Manager)
shows all the project data so far.
The tabs along the left side select
the type of project data. Right
now, the Well tab is selected. It is
empty, since we have not yet
loaded any data

To load a well, click the button

Import Well, and select Logs, Check
Shots, Tops, …:

29
Exercise 1

You need to select the file avo_well.las. Highlight the file name in the
list of available files on the left and then click the Select option:

Note that Geoview assumes this

file has an LAS format because of Now click OK to load this file in
the name extension. LAS format:

30
Exercise 1

The Geoview window now looks like this:

By default,
the program
has opened
and displayed
all of the
available log
curves and
tops in the
avo_well.las
file.

31
Exercise 1

One part of the window (called the

Project Manager) shows all the project
data so far. The tabs along the left side
select the type of project data. Right
now, the Well tab is selected and we can
see the well (AVO_WELL) which has
been loaded into the project. Click the
arrow sign near the well name to see a
list of curves in that well:

To see more details about the

wells, click the Data Explorer
tab to the right:

32
Exercise 1

The window now

changes as shown:

Click the arrow next

to the well name to
get more information
about the curves in
that well:

33
Exercise 1

Note that all the well logs we saw in

the display are listed, as well as the
Depth-time_P-wave log, which was
created from the sonic log and will
be used for depth-to-time
conversion.

Note that we can edit items in this

table. For example, if the Density
units were wrong, we could
change them here. Also, we can
click the arrow next to any of the
curve names to see the numerical
values in that curve:

34
Exercise 1

Click on the arrow that is

pointing to the left to go
back to the previous
menu since we will not
be editing the density
values in this tutorial.

To the right of the workspace, we can

see a series of buttons. Click the button
Show Map, the map view will appear
showing the location of the wells:

35
Exercise 1

Click the Single Well Display

button. This shows the
curves for the selected well:

36
Exercise 1

Finally, to see the most complete view

of the log curves within a well, double-
click the icon for that well within the
Project Data window:

This creates a new

tab within the main
Geoview window,
called the Wells tab,
which displays the
well curves:

37
Exercise 1

You can adjust the well plotting parameters

by clicking the “eyeball” icon, to bring up a
dialog for that purpose:

You can also apply various log

processing options, like Log Editing, by
going to the Processes list:

For this workshop, we will assume that

the logs have been properly edited.

38
Exercise 1: Loading the seismic data

We have now loaded the well which will

be used in the AVO Modeling process.
The next step is to load the seismic
volume, which we will compare with the
calculated synthetic.

On the far left side of the Geoview

window, click the Seismic tab:

The window to the right of this tab shows all

seismic data loaded so far. This is empty. Go
to the bottom of the window and click the
Import Seismic button:

On the pull-down menu, select From

SEG-Y File:

39
Exercise 1

On the dialog that

appears, select the file
gathers.sgy:

The selected file

appears on the right of
the menu, as shown.
Click Next at the base of
the dialog:

Set the Geometry Type

to 2D and click Next:

40
 Exercise 1

On the third page, we are telling the

program what information it can use from
the trace headers. In fact, in this data set,
there are X and Y coordinates. That is
why we answer Yes to this question:

Click Next to see the SEG-Y Format page:

By default, this page assumes that the

seismic data is a SEG-Y file with all
header values filled in as per the
standard SEG-Y convention. For
example, it expects to find the Inline
and Xline numbers at the byte
locations shown. If you are not sure
that is true, you can click Header Editor
to see what is in the trace headers.

41
Exercise 1

In our case, we believe the format

information is correct, so click Next to
move to the next page. Now the
following warning message appears
because the program is about to scan
the SEG-Y file:

Click Yes to begin the scanning

process. When the scanning has
finished, the Geometry Grid page
appears:

Because we have not read X-Y

information from the headers, the
program assumes this is a single
straight line, which is correct.

Click OK.
42
Exercise 1

After building the geometry files, a new window appears, showing how
the well is mapped into this seismic volume.
In this case, the mapping is correct because we supplied the X-Y
location of the well, and there were X-Y coordinates in the seismic trace
headers.

If this information
were not correct, we
could manually
locate the well at the
known CDP location
(330).

Click OK to accept the well location shown on this window.

43
Exercise 1

Now the seismic data

appears within the
workspace:

The workspace
currently shows the
single line, positioned
at the left, from this
dataset. To see other
parts of the line, slide
the scroll bar at the
base of the display.

44
Exercise 1

To see the display positioned at the well

location, go to the Well icon and click
the down arrow:

The drop-down menu shows

the one well in the project.
Select the well and the
Geoview window shows the
seismic data in the vicinity of
that well location:

We can also modify other plotting parameters

by using the Seismic View Parameters
window. To make that window appear, click
the “eyeball” icon:
45
Exercise 1

The Seismic View Parameters window

contains a series of pages which
control various aspects of the plotting.

To see the parameters for a specific

item, select that item from the list at the
left side. For example, here we have
selected the Inserted Wells item:

Let us (temporarily) insert the

density log by selecting that item as
shown from the pull-down tab of the
Inserted Curve Log option:

46
Exercise 1

Now click Apply on the Seismic View

Parameters window. The display is
modified accordingly:

We will continue the workshop with

the sonic log inserted. To do this,
click Cancel on the View Parameters
window. This redraws the Geoview
window as before.

(End of Exercise 1)
47
Rock Physics & Fluid
Replacement Modeling
Basic Rock Physics

The AVO response is dependent on the properties of P-wave velocity (VP),

S-wave velocity (VS), and density (r) in a porous reservoir rock. As shown
below, this involves the matrix material, the porosity, and the fluids filling
the pores:

Rock Matrix Pores / Fluid

49
Density

Density effects can be modeled with the following equation:

ρsat  ρm (1   )  ρw S w  ρhc (1  S w )

where : ρ  density,
  porosity,
S w  water saturation,
sat,m,hc, w  saturated, matrix,
hydrocarbon, water subscripts.

This is illustrated in the next graph.

50
 Density versus Water Saturation

Here is a plot of density Density vs Water Saturation

vs water saturation for a Sandstone with Porosity = 33%
porous sand with the Densities (g/cc): Matrix = 2.65, Water = 1.0,
parameters shown, Oil = 0.8, Gas = 0.001
where we have filled the 2.2
pores with either oil or
gas. 2.1

2
In the section on AVO
Density

we will model both the 1.9

wet sand and the 50%
saturated gas sand. 1.8
Note that these density 1.7
values can be read off
the plot and are: 1.6
rwet = 2.11 g/cc 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
rgas = 1.95 g/cc Oil Gas Water Saturation

51
 P and S-Wave Velocities

Unlike density, seismic velocity involves the deformation of a rock as a

function of time. As shown below, a cube of rock can be compressed, which
changes its volume and shape or sheared, which changes its shape but not
its volume.

52
P and S-Wave Velocities
The leads to two different types of velocities:
P-wave, or compressional wave velocity, in which the direction of
particle motion is in the same direction as the wave movement.
S-wave, or shear wave velocity, in which the direction of particle
motion is at right angles to the wave movement.

P-waves S-waves

53
Velocity Equations using  and 

The simplest forms of the P and S-wave velocities are derived for
non-porous, isotropic rocks. Here are the equations for velocity
written using the Lamé coefficients:

  2 
VP  VS 
r r

where:  = the first Lamé constant,

 = the second Lamé constant,
and r = density.

54
Velocity Equations using K and 

Another common way of writing the velocity equations is with

bulk and shear modulus:

4
K  
VP  3 VS 
r r

where: K = the bulk modulus, or the reciprocal of compressibility.

=  + 2/3 
 = the shear modulus, or the second Lamé constant,
and r = density.

55
Poisson’s Ratio from strains
F
 If we apply a compressional R
force to a cylindrical piece of
rock, as shown on the right, we R+R
change its shape.

L+L L
 The longitudindal strain is given
by L/L and the transverse strain
is given by R/R.

F (Force)
 The Poisson’s ratio, , is defined as the negative of the ratio
between the transverse and longitudinal strains:
  (R / R) /(L / L)
(In the typical case shown above, L is negative, so  is positive)
56
Poisson’s Ratio from velocity

A second way of looking at Poisson’s ratio is to use the ratio of VP to VS,

and this definition is given by:

 2 2
 2
2  2
VP
where :  
VS
This formula is more useful in our calculations than the formula given
by the ratio of the strains. The inverse to the above formula, allowing
us to derive VP or VS from , is given by:

2  2
 
2

2  1
57
Poisson’s Ratio vs VP/VS ratio

Vp/Vs vs Poisson's Ratio

0.5
0.4
0.3
Poisson's Ratio

0.2
0.1
0
-0.1
-0.2
0 1 2 3 4 5 6 7 8 9 10
Gas Case Wet Case Vp/Vs

58
Poisson’s Ratio

From the previous figure, note that there are several values of
Poisson’s ratio and VP/VS ratio that are important to remember.

If VP/VS = 2, then  = 0

If VP/VS = 1.5, then  = 0.1 (Gas Case)

If VP/VS = 2, then  = 1/3 (Wet Case)

If VP/VS = , then  = 0.5 (VS = 0)

Note also from the previous figure that Poisson’s ratio can
theoretically be negative, but this has only been observed for
materials created in the lab (e.g. Goretex and polymer foams).

59
Velocity in Porous Rocks
Velocity effects can be modeled by the volume average equation:

t sat  t m( 1  )  t w S w  t hc( 1  S w ) , where t  1 / V

Velocity vs Water Saturation
A plot of velocity versus Wyllie's Equation
water saturation using Porosity = 33%
the above equation. We Vmatrix = 5700 m/s, Vw = 1600 m/s,
Voil = 1300 m/s, Vgas = 300 m/s.
used a porous sand with
3500
the parameters shown
and have filled the pores 3000

with either oil or gas. Velocity (m/sec) 2500

2000
This equation does not
hold for gas sands, and 1500
this lead to the
1000
development of the Biot-
Gassmann equations. 500
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Oil Gas Water Saturation

60
The Biot-Gassmann Equations
It has been found that the volume average equation gives incorrect
results for gas sands. Independently, Biot (1941) and Gassmann (1951),
developed a more correct theory of wave propagation in fluid saturated
rocks, especially gas sands, by deriving expressions for the saturated
bulk and shear moduli and substituting into the regular equations for P
and S-wave velocity:

4
K sat  sat sat
VP _ sat  3 VS _ sat 
r sat r sat

Note that rsat is found using the volume average equation discussed
earlier, or:

ρsat  ρm (1   )  ρw S w  ρhc (1  S w )
61
The Biot-Gassmann Equations
To understand the Biot-Gassmann equations, let us update the figure we
saw earlier to include the concepts of the “saturated rock” (which includes
the in-situ fluid) and the “dry rock” (in which the fluid has been drained.)

Saturated
Rock
Dry rock (pores full)
frame, or
skeleton
(pores
empty)

Rock Matrix Pores and fluid

62
Biot-Gassmann - Shear Modulus
In the Hampson-Russell AVO program, Biot-Gassmann analysis is done
using the FRM (Fluid Replacement Modeling) option. Let us first look at
some theory and then consider several practical considerations when
using the FRM option.

In the Biot-Gassmann equations, the shear modulus does not change for
varying saturation at constant porosity. In equations:

sat  dry

where sat  the shear modulus of the saturated rock,

and dry  the shear modulus of the dry rock.

63
Biot-Gassmann – Saturated Bulk Modulus
The Biot-Gassmann bulk modulus equation is as follows:
2
 K dry 
1  
(1) K sat  K dry   Km 
 1   K dry
  2
K fl Km Km
Mavko et al, in The Rock Physics Handbook, re-arranged the above
equation to give a more intuitive form:

K sat K dry K fl
(2)  
K m  K sat K m  K dry  ( K m  K fl )
where sat = saturated rock, dry = dry frame, m = mineral, fl = fluid,
and  = porosity.

64
Biot’s Formulation
Biot defines b (the Biot coefficient) and M (the fluid modulus) as:

K dry 1  b 
b  1 , and   ,
Km M K fl Km

Equation (1) then can be written as: K sat  K dry  b 2 M

If b = 0 (or Kdry = Km) this equation simplifies to: K sat  K dry

1  1
If b = 1 (or Kdry= 0), this equation simplifies to:  
K sat K fl Km

Physically, b = 0 implies we have a non-porous rock, and b = 1 implies we

have particles in suspension (and the formula given is called Wood’s
formula). These are the two end members of a porous rock.

65
The Rock Matrix Bulk Modulus

We will now look at how to get estimates of the various bulk modulus
terms in the Biot-Gassmann equations, starting with the bulk modulus of
the solid rock matrix. Values will be given in gigaPascals (GPa), which
are equivalent to 1010 dynes/cm2.

The bulk modulus of the solid rock matrix, Km is usually taken from
published data that involved measurements on drill core samples.
Typical values are:

Ksandstone = 40 GPa,

Klimestone = 60 GPa.

66
The Fluid Bulk Modulus
The fluid bulk modulus can be modeled using the following equation:

1 S 1  Sw
 w 
K fl K w K hc

where K fl  the bulk modulus of the fluid,

K w  the bulk modulus of the water,
and K hc  the bulk modulus of the hydrocarbon.

Equations for estimating the values of brine, gas, and oil bulk modulii are
given in Batzle and Wang, 1992, Seismic Properties of Pore Fluids,
Geophysics, 57, 1396-1408. Typical values are:

Kgas = 0.021 GPa, Koil = 0.79 GPa, Kw = 2.38 GPa

67
 Estimating Kdry

The key step in FRM is calculating a value of Kdry. This can be done in
several ways:

(1) For known VS and VP, Kdry can be calculated by first calculating Ksat
and then using Mavko’s equation (equation (2)), given earlier.

(2) For known VP, but unknown VS, Kdry can be estimated by:

(a) Assuming a known dry rock Poisson’s ratio dry. Equation (1) can
then be rewritten as a quadratic equation in which we solve for Kdry.

(b) Using the Greenberg-Castagna method, described later.

68
 Data Examples

 In the next few slides, we will look at the computed responses for
both a gas-saturated sand and an oil-saturated sand using the
Biot-Gassmann equation.

 We will look at the effect of saturation on both velocity (VP and VS)
and Poisson’s Ratio.

 Keep in mind that this model assumes that the gas is uniformly
distributed in the fluid. Patchy saturation provides a different
function. (See Mavko et al: The Rock Physics Handbook.)

69
Velocity vs Saturation of Gas
A plot of velocity vs water Velocity vs Water Saturation - Gas Case
saturation for a porous gas Sandstone with Phi = 33%, Density as previous figure for gas,
Kmatrix = 40 Gpa, Kdry = 3.25 GPa, Kw = 2.38 Gpa,
sand using the Biot-Gassmann Kgas = 0.021 Gpa, Shear Modulus = 3.3. Gpa.

equations with the parameters 2600

shown.
2400

In the section on AVO we will 2200

model both the wet sand and 2000

Velocity (m/s)
the 50% saturated gas sand.
1800
Note that the velocity values
can be read off the plot and 1600

are: 1400
VPwet = 2500 m/s
1200
VPgas = 2000 m/s
1000
VSwet = 1250 m/s 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Sw
VSgas = 1305 m/s
Vp Vs

70
Poisson’s Ratio vs Saturation of Gas

A plot of Poisson’s ratio vs Poisson's Ratio vs Water Saturation - Gas Case

water saturation for a porous Sandstone with Phi = 33%, Density as previous figure for gas,
Kmatrix = 40 Gpa, Kdry = 3.25 GPa, Kw = 2.38 Gpa,
gas sand using the Biot- Kgas = 0.021 Gpa, Shear Modulus = 3.3. Gpa.

Gassmann equations with the

0.5
parameters shown.

In the section on AVO we will 0.4

model both the wet sand and

Poisson's Ratio
the 50% saturated gas sand. 0.3

Note that the Poisson’s ratio

values can be read off the plot 0.2
and are:
wet = 0.33 0.1

gas = 0.12
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Sw

71
Velocity vs Saturation of Oil

Velocity vs Water Saturation - Oil Case

A plot of velocity vs water Sandstone with Phi = 33%, Density as previous figure for oil,
Kmatrix = 40 Gpa, Kdry = 3.25 GPa, Kw = 2.38 Gpa,
saturation for a porous oil Koil = 1.0 Gpa, Shear Modulus = 3.3. Gpa.
sand using the Biot-
2600
Gassmann equations with
the parameters shown. 2400

2200
Note that there is not much
of a velocity change. 2000

Velocity (m/s)
However, this is for “dead” 1800
oil, with no dissolved gas
1600
bubbles, and most oil
reservoirs have some 1400

percentage of dissolved 1200

gas.
1000
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Sw

Vp Vs

72
Poisson’s Ratio vs Saturation of Oil

A plot of Poisson’s ratio vs Poisson's Ratio vs Water Saturation - Oil Case

Sandstone with Phi = 33%, Density as previous figure for oil,
water saturation for a porous Kmatrix = 40 Gpa, Kdry = 3.25 GPa, Kw = 2.38 Gpa,
oil sand using the Biot- Koil = 1.0 Gpa, Shear Modulus = 3.3. Gpa.

Gassmann equations with the 0.5

parameters shown.
0.4
Note that there is not much of
a Poisson’s ratio change.

Poisson's Ratio
However, again this is for 0.3

“dead” oil, with no dissolved

gas bubbles, and most oil 0.2
reservoirs have some
percentage of dissolved gas.
0.1

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Sw

73
Fluid substitution in carbonates
In general carbonates are thought to have a smaller fluid sensitivity than
clastics. This is a consequence of the fact that they are typically stiffer (i.e.
have larger values of Km and Kdry ) implying a smaller Biot coefficient b and
hence fluid response.

This general observation is complicated by the fact that carbonates often

contain irregular pore shapes and geometries.
 High aspect ratio pores make the rock more compliant and thus more
sensitive to fluid changes.
 Aligned cracks require the use of the anisotropic Gassmann equation,
resulting in the saturated bulk modulus being directionally dependent.
 Gassmann assumed that pore pressure remains constant during wave
propagation. If the geometry of the pores and cracks restrict the fluid
flow at seismic frequencies then the rock will appear stiffer.

All these factors make the application of the Biot-Gassmann fluid

substitution in carbonates more complex.

74
Patchy Saturation
When multiple pore fluids are present, Kfl is usually calculated by a Reuss
averaging technique (see Appendix 2):

1 S w So S g
  
K fl K w Ko K g

This method heavily biases compressibility of the combined fluid to

the most compressible phase.
Kfl vs Sw and Sg
This averaging 3

Bulk modulus (Gpa)

technique assumes 2.5
uniform fluid 2
distribution! 1.5
1
0.5
-Gas and liquid must
0
be evenly distributed
0 0.25 0.5 0.75 1
in every pore.
Water saturation (fraction)

75
Patchy Saturation
When fluids are not uniformly mixed, effective modulus values cannot be
estimated from Reuss averaging. Uniform averaging of fluids does not
apply.

When patch sizes are large with respect to the seismic wavelength, Voigt
averaging (see Appendix 2) gives the best estimate of Kfl (Domenico, 1976):

K fl  S w K w  So Ko  S g K g

When patch sizes are of intermediate size, Gassmann substitution should

be performed for each patch area and a volume average should be made.
This can be approximated by using a power-law averaging technique,
which we will not discuss here.

76
Patchy Saturation
Gassmann predicted velocities
Unconsolidated sand matrix
Porosity = 30%
100% Gas to 100% Brine saturation

2.5
2.3
Vp (km/s)

2.1 Patchy
Voigt
1.9 Reuss
1.7
1.5
0 0.25 0.5 0.75 1
Water Saturation (fraction)
77
The Mudrock Line

The mudrock line is a linear relationship between VP and VS

derived by Castagna et al (1985):

VP  1.16 VS  1360 m / s

Note that for a constant Poisson’s ratio, the intercept is zero:

2  2
VP  VS
2  1

This will be illustrated in the next few slides.

78
The Mudrock Line

ARCO’s original mudrock derivation

(Castagna et al, Geophysics, 1985)
79
 The Mudrock Line

6000

5000
Mudrock Line
4000

3000
Gas Sand
VP (m/s)
2000

1000

0
0 1000 2000 3000 4000
VS(m/s)
80
 The Mudrock Line
6000

5000
 = 1/3 Mudrock Line
or
4000 VP/VS = 2

3000
Gas Sand
VP (m/s)
2000

1000

0
0 1000 2000 3000 4000
VS(m/s)
81
 The Mudrock Line

6000

5000
 = 1/3 or Mudrock Line
VP/VS = 2
4000

3000
Gas Sand
VP
(m/s)
2000
 = 0.1 or
VP/VS = 1.5
1000

0
0 1000 2000 3000 4000
VS(m/s)
82
The Greenberg-Castagna method
Greenberg and Castagna (1992) extended the previous mud-rock
line to different mineralogies as follows, where we have now
inverted the equation for VS as a function of VP:

Sandstone : VS  0.856 km / s  0.804 VP

Limestone : VS  1.031 km / s  1.017 VP  0.055VP2
Dolomite : VS  0.078 km / s  0.583 VP
Shale : VS  0.867 km / s  0.770 VP

Using the regression coefficients given above, Greenberg and Castagna

(1992) first propose that the shear-wave velocity for a brine-saturated rock
with mixed mineral components can be given as a Voigt-Reuss-Hill
average (see Appendix 2) of the volume components of each mineral.

83
The Greenberg-Castagna method
To compute the shear-wave velocity of a rock with multiple
minerals and a known hydrocarbon component (i.e. SW < 1),
Greenberg and Castagna (1992) then propose the following
iterative scheme:
1. Estimate the brine-filled P-wave velocity. This is nothing more
than an initial guess.
2. Compute the S-wave velocity from the regression just given.
3. Perform Gassmann fluid substitution with the values from
steps 1 and 2 to compute the P-wave velocity for the SW < 1
case. This requires estimates of the moduli and density of
each component.
4. Based on the error between the measured and computed P-
wave velocities (for SW < 1), go back to step 1 and perturb the
estimate of the brine-filled P-wave velocity.
5. Iterate until the brine-saturated P-wave velocities agree.

The Greenberg-Castagna method will be used in the next exercise.

84
 The rock physics template (RPT)

Ødegaard and Avseth

(2003) proposed a
technique they called the
rock physics template
(RPT), in which the fluid
and mineralogical
content of a reservoir
could be estimated on a
crossplot of Vp/Vs ratio
against acoustic
impedance, as shown
here.

from Ødegaard and Avseth (2003)

85
 The rock physics template (RPT)

 Ødegaard and Avseth (2003) compute Kdry and dry as a

function of porosity  using Hertz-Mindlin (HM) contact
theory and the lower Hashin-Shtrikman bound.
 Hertz-Mindlin contact theory assumes that the porous rock
can be modeled as a packing of identical spheres, and the
effective bulk and shear moduli are computed from:
1 1
 n 2 (1  c ) 2 m2 
3
5  4 m  3n 2 (1  c ) 2 m2 
3
K eff   P  , eff   P ,
18 (1   m )  5( 2   m )  2 (1   m )
2 2 2 2

where : P  confining pressure, m  mineral shear modulus,
n  contacts per grain,  m  mineral Poisson' s ratio,
and c  high porosity end - member.
86
 The rock physics template (RPT)

 The lower Hashin-Shtrikman bound is then used to compute

the dry rock bulk and shear moduli as a function of porosity
with the following equations:
1
  / c 1   / c  4
K dry     eff
 K eff  ( 4 / 3) eff K m  ( 4 / 3) eff  3
1
  / c 1   / c  4
dry     z, where :
 eff  z m  z  3
eff  9 K eff  8eff 
z   and K m  mineral bulk modulus.
6  K eff  2 eff 

 Standard Gassmann theory is then used for the fluid
replacement process.
87
The rock physics template (RPT)

Here is the RPT for a range of porosities and water saturations, in a

clean sand case. We will build this template in the next exercise.
88
Using the Biot-Gassmann Equations
(1) Fluid Substitution

The basic use of the Biot-Gassmann equations is to “substitute” or replace the fluids
in a set of target layers with another set of fluids.

In this case, VP, VS, and ρ must all be known for the input logs, along with the fluid
content (SW). Generally all three logs are changed within the target zone.

Input Logs Output Logs

VP ρ VS VP ρ VS
SW = 50% SW = 100%
89
Using the Biot-Gassmann Equations
(2) Calculating Vs

The second use of the Biot-Gassmann equations is to calculate a VS curve, which

has not been measured in the well. Either KDRY is assumed known or the Greenberg-
Castagna method is used.

In this case, VP and ρ must both be known, along with the fluid content (SW). The VP
and ρ logs are unchanged, and a new VS log is created.

Input Logs Output Logs

ρ VP ρ VS
VP
SW = 50% SW = 50% 90
Conclusions
 An understanding of rock physics is crucial for the interpretation of AVO
anomalies.

 The volume average equation can be used to model density in a water

sand, but this equation does not match observations for velocities in a
gas sand.

 The Biot-Gassmann equations match observations well for

unconsolidated gas sands.

 When dealing with more complex porous media with patchy saturation,
or fracture type porosity (e.g. carbonates), the Biot-Gassmann equations
do not hold.

 The ARCO mudrock line is a good empirical tool for the wet sands and
shales.

91
Exercise 2:
The Colony Gas Sand
Biot-Gassmann analysis
Exercise 2
Now that we have read in all the data
necessary for the AVO Modeling, we are ready
to start the process.
First, look at the tabs to the left of the
Geoview window. You will see that one of
those tabs is called Processes. Click on that
tab to see a list of all the operations which are
available in Geoview. Each of the processes
can be expanded. For example, if you click
on the AVO Modeling option, the following
expanded list is seen:

As we can see, the list includes Log Processing

options, Seismic Processing options, AVO
Modeling and Analysis tools, Inversion options,
etc. One way to do the modeling would be to
apply each of the desired options in turn. That
would be the traditional approach.

93
Exercise 2

We will use an alternate procedure in this

tutorial. We will use the pre-defined
Workflows. Click the Workflows tab. The
window changes like this:

Each of the items on this window contains

a complete workflow for the specified
process. Click on the item called AVO
Modeling. The window changes like this:

We now see the suggested series of steps

to be followed for AVO Modeling. The
steps are colored red to indicate that the
parameters have not yet been supplied.
These are the “default” steps, but the list
can be edited and customized, as we will
see later.

94
Exercise 2

Double-click on the first item Select

Well. An arrow will appear in front of
the item, as shown here:

Now a dialog appears on

the right with a list of all
wells in the project:

Since we have only loaded one well,

that well is selected. Note that at the
lower right corner of the dialog, there is
a button for importing more wells:

We do not need to import another well, so click Select on this dialog.

95
 Exercise 2

Now, double-click the second item on

the Workflow list, Select Logs:

The dialog on the right shows that for

AVO Modeling, we need three log
curves:
Two of them, the P-wave velocity
(sonic log) and the Density curves are
available in the well. The third, S-wave
velocity, is not present in the well and
will be computed in the next step.

Click Select on this menu to use the

latest curves for the calculation:

96
Exercise 2: Calculating the Shear Wave Log

Now double-click the third step on the

workflow, Shear Wave Estimation:
This step is necessary because a shear
wave velocity log is required to do AVO
offset modeling. However, like many
wells, this well does not include a
measured shear wave log. Therefore,
we must compute it now from the
existing sonic and density logs. If the
well had contained a measured shear
wave log, we could skip this step in the
workflow.

The dialog on the right shows

a series of tabs, which must
be completed in order:

97
 Exercise 2

The first tab specifies the location

of the Reservoir within the well:

This page is necessary because there

are two types of equations used,
depending on whether we are within
the hydrocarbon zone or outside in
the wet background. This page
specifies the Reservoir or
hydrocarbon zone. The actual
equations are described on the next
page. A convenient way of specifying
the zone is to use the formation tops,
which have been imported from the
LAS file. These tops have been called
TOP_GAS and BASE_GAS for this
well. Alternatively, we could specify
the depth range directly.

98
 Exercise 2

Now click the Velocity tab:

Here we see that the P-wave

velocity is coming from the
sonic log in the well:

For the S-wave velocity, two

calculations will be used, one
for the samples within the
reservoir, and the other for
samples outside the reservoir:

These calculations are

explained in greater detail in
the tabs below:

99
Exercise 2

Now click the Density tab:

This tab appears because the density and porosity of the reservoir are
related by the Volume Average Equation:

ρsat  ρm( 1  )  ρw S w  ρhc( 1  S w )

By default, with a measured density log, the porosity will be

calculated. Alternatively, with a measured porosity log, the density
can be calculated. Finally, the option exists to use both measured
density and porosity logs, and that requires a calculation of the
matrix density.

100
Exercise 2

Now click the In-Situ Fluid tab:

On this page, we specify the

fluid properties within the
reservoir. The first thing we
specify is the relative
concentrations of fluid:

101
Exercise 2

In this case, we have told the program that the reservoir is 50% brine
and 50% gas. This is assumed known about the reservoir.
Alternatively, we could use a water saturation log, if available. Change
the saturations as shown above.

Lower on the menu page we have the

parameters for the density and bulk
modulus of the fluids:

102
Exercise 2

We can also calculate the fluid

properties using empirical
relationships:

but for this tutorial, we will skip

this calculation, so unselect that
check box:

103
 Exercise 2

Now click the Matrix tab:

This page specifies the matrix

properties within the reservoir.
The top part is grayed out – i.e.,
disabled – because we have
chosen to use the Greenberg-
Castagna equations within the
reservoir. Those equations
require a mixture of minerals,
which we specify below:

Note that, since we do not have any volumetric logs in the well, our
only option is to specify constant percentages for the whole reservoir
zone. By default, the specification is 100% sand for the reservoir,
which we will accept for this tutorial.

104
 Exercise 2

Finally, click the Output tab:

Here we specify the name of the output

S-Wave velocity curve we will create.
We also can click the QC Display button
to see some of the internal calculations
within the reservoir:

This plot can help show any major

problems with the model.

Now click Run to create the S-Wave

velocity log:
105
Exercise 2

The Geoview window

now includes the
calculated shear
wave log:

106
Exercise 2 The rock physics template (RPT)

Next, we will build a rock physics

template for this well, which will allow
us to see the effect of the porosity as
well as the fluid.

The standard RPT uses a plot of Vp/Vs

ratio versus P-impedance, so first we
will create these logs.

For P-impedance, click the Processes

tab in the project manager, select Log
Processing>Log Transform>
Impedances>P-impedance:

107
Exercise 2

Fill in the menu as shown

and click OK.

108
 Exercise 2

To create the Vp/Vs ratio log, select

Log Processing>Log Transform>Other
>Vp/Vs Ratio in the project manager,
fill in the menu as shown and click OK.

109
Exercise 2

You will see that the two log curves are created and displayed as
new tracks at the right end of the AVO_WELL display.

110
Exercise 2

On the log display you will notice that

the SP curve has not been displayed,
due to having the wrong default limits.
An easy way to fix this problem is to
left-click on the area around the SP
heading to get the menu shown below:

Then click the third icon from the left

on the bottom of this menu, set
display range to data range, and click
OK to set the new values.

The new display looks like this, and

the SP curve is properly displayed.

111
Exercise 2

Next, we will cross plot the newly

created logs P-impedance and
Vp/Vs ratio. To do that, double
click Cross Plot Logs in the
project manager:

On the top of the menu, select

the well and the logs as shown
below:

112
Exercise 2

At the bottom of the menu,

we will define the Zone of
Interest by constant depths
[of 600-700m].

To initiate the correct

menus, click on All Data
first; select All Data
Between Targets.
Then fill in the menu as
shown and click OK:

113
 Exercise 2

On the crossplot window, click on the box to the right of RPT Selection.
This brings up the RPT menu. Note that the menu is all blank, and if you
know your parameters you can fill them in.

Note also that the

small cluster of
yellow points (with
depths between
630 and 650 m)
shows the gas
sand and the other
trends are from
shales and wet
sands.

114
 Exercise 2

If you do not know what values to put in the menu,

you can create a default RPT by clicking on the RP
Template menu and selecting New:

On the crossplot
window, the default
parameters have been
filled on the RPT menu.

We also notice that

some of the cross plot
points are not visible
due to the data range of
Y-axis.

115
Exercise 2

To change the data range,

select Plot Options on View
menu:

On the Plot Options menu,

change Max to 3.0. Click
Apply and close this menu.

(Note that this could also

be done using View > Reset
Display Range or by right
clicking on the vertical
scale itself and selecting
Set axis scale..)

116
 Exercise 2

We will change the Kdry calculation type to User input, and toggle the Bulk
dry modulus up to 2.65 and the Shear dry modulus to 2.0 to get a very
good fit to the gas sand at 35%. Select Apply. Note the interactive curve
changes.

117
 Exercise 2
Next, add a second template by selecting RP
Template>New and change the RPT type to Shale line.
The new template shown below fits the shales quite
well. Experiment with the parameters to get a better fit.

118
 Exercise 2

Click on one of the points on the

original Standard template to make it
live (the points will turn yellow).

Now, click the Fluid tab on the RPT

Menu Option. Check on Advanced
fluid property analysis.

Experiment with various options (e.g.

starting fluid, which can be done
interactively on the ternary diagram
for Brine, Oil and Gas; Temperature,
Bubble Point Calculation, etc) and
see what effect they have on the
template.

119
 Exercise 2

Finally, click the Matrix tab, check on Advanced matrix property analysis
and select Rock Type of Matrix Property Calculator. Create a matrix of
50% Quartz and 50% clay and click Apply. Note the effect on the RPT.

120
Exercise 2

Obviously, you can go on experimenting with the RPT menu to get

as good a fit to your points as you want.

Note that as many templates can be added as you want, and that
options exist for Limestone, Dolomite and Quartz lines as well.

The Rock Physics Template option allows you to obtain empirical

porosity, lithology and fluid fits to your well log data.

Later in the course we will cover pre-stack inversion and it should

be noted that the RPT can also be applied to the inverted seismic
data crossplot.

(End of Exercise 2)

121
AVO Theory & Zoeppritz Modeling
 P and S-Waves

(a) (b) (c)

The above diagram shows a schematic diagram of (a) P, or compressional,
waves, (b) SH, or horizontal shear-waves, and (c) SV, or vertical shear-waves,
where the S-waves have been generated using a shear wave source. (Ensley,
1984)
123
From P and S-Waves to AVO

 In the previous slide, the P and SH-waves were generated at the

surface by P and S-wave sources. We could use the differences
between the recorded P and S reflections to discriminate gas-filled
sands from wet sands, using the properties discussed in the last
section.

 However, most seismic surveys record P-wave data only, and S-wave
data is not available.

 As shown in the next slide, if we record P-wave data at an angle greater

than zero, we produce mode conversion from P to S-wave data.

 The objective of this course will be to utilize mode conversion both

quantitatively and qualitatively.

124
Mode Conversion of an Incident P-Wave
If q > 0°, an incident P-wave will produce both P and SV reflected and
transmitted waves. This is called mode conversion.
Reflected
Incident SV-wave = RS(q1)
P-wave
Reflected
1 P-wave = RP(q1)

q1
q1
VP1 , VS1 , r1
VP2 , VS2 , r2 q2
2 Transmitted
P-wave = TP(q1)
Transmitted
SV-wave = TS(q1)
125
Utilizing Mode Conversion
But how do we utilize mode conversion? There are actually two ways:

(1) Record the converted S-waves using three-component receivers

(in the X, Y and Z directions). Note that when we analyze the
converted waves, we need to be very careful in their processing and
interpretation.

(2) Interpret the amplitudes of the P-waves as a function of offset, or

angle, which contain implied information about the S-waves. This is
called the AVO (Amplitude versus Offset) method.

In the AVO method, we can make use of the Zoeppritz equations, or some
approximation to these equations, to extract S-wave type information
from P-wave reflections at different offsets. Before discussing these
equations, the next figures shows a typical set of gathers over a gas sand
and intuitively explain the relationship between offset and angle.

126
A Data Example

The box in the top

figure shows the
“stacked” traces
corresponding to the
five P-wave gathers
shown in the bottom
figure.

Notice that the

amplitudes in the
gathers over the
highlighted region
show an increase in
amplitude as a
function of offset.
This is called an AVO
(Amplitude Variation
with Offset) effect.

127
 Angle and Offset

Source Offset Receiver

Surface

Angles q2 q1
q3

r1 VP1 VS1 Reflector

r2 VP2 VS2
Common mid-point

Seismic data is recorded as a function of offset. However, as shown above,

there is a direct relationship between angle and offset, which depends on
velocity. We can model these amplitude changes using either the full
Zoeppritz equations or the linearized Aki-Richards approximation.

128
 The Zoeppritz Equations (1919)

Karl Zoeppritz derived the amplitudes of the

reflected and transmitted waves using the
conservation of stress and displacement
across the layer boundary, which gives four
equations with four unknowns.

1
  sin q1  cos 1 sin q 2 cos 2 
 RP (q1 )  cos q  sin 1 cos q 2  sin 2   sin q1 
 R (q )   1   cos q 
 S 1    sin 2q r 2VS 2VP1 r 2VS 2VP1
cos 22 
2
VP1  1 
cos 21 sin 2q 2
 TP (q1 )   1
VS 1 r1VS12VP 2 r1VS1 2
 sin 2q1 
   r 2VP 2 r 2VS 2
  
T (q
 S 1   cos 21
) VS 1
sin 21 cos 22  sin 22  cos 21 
 VP1 r1VP1 r1VP1 

129
The Zoeppritz Equations at 0 degrees

Although the Zoeppritz equations look intimidating, in the case of normal

incidence (0 degrees) the equations give us the following simple values for
the reflection and transmission coefficients (see Appendix 3 for the
mathematical details):

RS (0o )  RS 0  0, TS (0o )  TS 0  0,
r 2VP 2  r1VP1
RP (0 )  RP 0 
o
,
r 2VP 2  r1VP1
2 r1VP1
TP (0 )  TP 0 
o
 1  RP 0 .
r 2VP 2  r1VP1
These equations tell us that there is no S-wave component at zero angle,
and the reflection and transmission coefficients are related to changes in
the acoustic impedance (P-velocity x density).

130
The Zero Angle Trace

We can think of the zero angle trace on a P-

wave gather as a set of reflection coefficients
corresponding to the changes in acoustic
impedance (density x P-velocity) at each
interface. The equation is below, and an Zi
illustration is on the right. This is not the
complete story, as the reflectivity is Ri
convolved with a wavelet. Zi+1
Z Pi 1  Z Pi
RP 0i = ,
Z Pi 1  Z Pi
where :
Z Pi  r iVPi  impedance,
r  density.

131
 Convolution

Convolution with the seismic wavelet, which can be written mathematically

as S = W*R, is illustrated pictorially below:

* = + + + + =>

W = Wavelet

R = Reflection S = Seismic
Coefficients Trace
132
The A-B-C equation

To model the amplitude changes as a function of angle for a single layer, we

can either use the full Zoeppritz equations given earlier, or the linearized
Aki-Richards equation, which will be discussed later in more detail. We will
show that there are several equivalent versions of this equation, but the
most common is written in the following form:

RP (q )  A  B sin 2 q  C tan 2 q sin 2 q , where :

1   VP r 
2 2
1  VP VS   VS VS  r
A  RP (0 )  
o
 , B   4   2  ,
2  V p r  2 Vp VP  VS VP  r
1  VP
and C  .
2 Vp
A is the linearized zero-offset reflection coefficient and (see Appendix 4)
is called the intercept, B is the gradient, and C the curvature. This
equation tells us that as the angle increases, so does the effect of S-wave
velocity.
133
A two-layer model

We can use the previous equation to model the top and base of a simple
sand. The figure on the left below shows the wet case and the one on the
right shows the gas case, using values computed in our rock physics section.
Notice the difference between using two terms and three terms in the
modeling.

134
 AVO Class 3

The model curves just shown for the gas case were for a Class 3 AVO
anomaly, of which the Colony sand we are considering is an example.

Here is a set of modeled well logs for a Class 3 sand, with the computed
synthetic (using all three terms in the A-B-C equation) on the right. Note that
the P-wave velocity and density (and thus the P-impedance) decrease in the
gas sand, the S-wave velocity increases, and the VP/VS ratio decreases. The
synthetic shows increasing amplitude versus offset for both the overlying
trough and underlying peak. The far angle is 45o.

135
AVO Class 2

As will be discussed later, there are several other AVO classes, of which
Class 1 and 2 are the most often seen.

Here is a Class 2 example well log, where the P-impedance change is very
small and the amplitude change on the synthetic is very large. Note that the
VP/VS ratio is still decreasing to 1.5, as expected in a clean gas sand (recall
the discussion in the rock physics section).

136
AVO Class 1

Here is a Class 1 well log example, where the P-impedance change is now
an increase and the amplitudes on the synthetic are seen to change
polarity. Again, the VP/VS ratio is still decreasing to 1.5, as expected in a
clean gas sand.

The figure on the next slide compares all three classes and also shows the
picked amplitudes.

137
 The three AVO Classes
A comparison of the
synthetic seismic
gathers from the three
classes, where the top
and base of the gas
sand have been picked. Class 1 Class 2 Class 3
The picks are shown at time (ms)
the bottom of the
display and clearly
show the AVO effects.

These synthetics were

created at the same
time, but in practice
amplitude

class 1 sands are deep,

class 2 sands are at
medium depths and
class 3 sands are at
shallow depths.
138
Multi-Layer AVO Modeling
We are usually interested in modeling a lot more than one or two layers.
Multi-layer modeling in the AVO program consists first of creating a stack
of N layers, generally using well logs, and defining the thickness, P-wave
velocity, S-wave velocity, and density for each layer, as shown below:

139
 Multi-Layer AVO Modeling

You must then decide what effects are to be included in the model: primaries
only, converted waves, multiples, or some combination of these.

140
AVO Modeling Options

There are three main options for the modeling process:

Zoeppritz – Primaries only, using the Zoeppritz equations for

calculation.

Aki-Richards – Primaries only, using the Aki-Richards equations

for calculation.

Elastic Wave – Computation of the full elastic wave solution (with

optional anelastic effects), which includes primaries, converted
waves, and multiples.

141
 AVO Modeling
Poisson’s
P-wave Density S-wave Synthetic Offset Stack
ratio

Based on AVO theory and the rock physics of the reservoir, we can perform AVO
modeling, as shown above. In this case, we have used the Aki-Richards equation in the
modeling. Note that the model result is a fairly good match to the offset stack.
Let us now do an exercise where we will perform this modeling.
142
Exercise 3:
The Colony Gas Sand
Creating Zoeppritz Synthetics
Exercise 3
Now double-click the fourth step on the
workflow, Select Seismic:
We need the seismic data for these steps
in the AVO Modeling workflow:
• To extract a wavelet.
• To correlate the well, i.e., to optimize the
depth-time relationship between well and
seismic.
• To compare with the resulting synthetic.

On the dialog on the right, we see a

list of the seismic volumes which
have been loaded into the project.
Since there is only one volume, that
has been selected:

Click Select to accept the volume

gathers:

144
Exercise 3
Now the seismic data
appears inserted within
the Wells tab:

The next step is to apply the process

Extract Statistical Wavelet, by double-
clicking that option:

145
Exercise 3
There are two basic methods for extracting the wavelet. One method uses
the wells, and can give a good estimate of both amplitude and phase
spectra of the wavelet. However, that method cannot be used until the
well is correlated, i.e., until the proper depth-time relationship has been
determined.

The second method – called “statistical” – uses the seismic data

alone to extract the wavelet. This method will estimate the amplitude
spectrum from the seismic data, but we must make an assumption
about the phase – typically we assume the data are zero phase. In
this step, we are extracting a statistical wavelet.

The dialog on the right shows the

range of data to analyze:

146
 Exercise 3

By default, the program will examine

the entire data volume, but this is
rarely appropriate. In particular, we
want to set a time window around the
zone of interest. Change the dialog
to extract just using the limited time
window shown below:

When you have changed the dialog as

shown above, click Run to extract the
wavelet.
The extracted wavelet appears in its
own pop-up window:

Note that the time domain response is

in the upper window, while the
amplitude and phase spectra are in the
lower window.
147
 Exercise 3
Note also this small button at the lower
right of the wavelet window:

If you click that button, the

wavelet window will be “docked”
within its own Wavelets tab:

This is a handy way to keep track of any window

created within Geoview. To release the wavelet
window from its tab, click on the “airplane” at the
lower right of the wavelet window:
148
 Exercise 3

All the windows created within Geoview can

be docked or floated in this way. Finally,
send the wavelet window back to the
wavelets tab by clicking the Wavelets button
once again:

The next step in the workflow is to

Correlate the Well, so double-click that
item:

The process of log correlation examines the current depth-time conversion

and optimizes it so that the derived synthetic matches the seismic optimally.
149
 Exercise 3
A dialog appears, specifying which seismic volume will be used for the
correlation process, and how the composite trace will be extracted from
that volume:

The composite trace is

created by averaging the
traces around the borehole
location we are using for
the seismic correlation. For
a vertical well, such as this
one, that means averaging
a selected set of
neighboring traces around
the borehole. By default
this is plus or minus 1
inline or crossline.

For this tutorial, we will accept the defaults. Click OK on this dialog.
150
 Exercise 3
The Log Correlation Window now
appears:

The blue traces on this display are

copies of the synthetic trace. This
trace was calculated from the sonic
and density logs in this well, the
depth-time curve currently stored in
the database, and the wavelet we
have previously extracted.

The red traces are repeated copies of

the average (or composite) trace
extracted from the seismic data.
The plot at the upper right shows the
cross correlation between the
synthetic trace and the composite
trace:
151
 Exercise 3
That correlation result depends on the
analysis window. We can improve the
calculation by changing this window. The
cross correlation window defaults to be the
largest possible window containing both
the synthetic and real trace. We should
narrow the analysis to the region where the
log tie is best:

Set the start time to 500, as shown above

and click on Apply. The correlation plot
now shows a maximum correlation of 62%.

It also suggests that the synthetic

should be shifted down by 50 ms. That
information is also displayed on the
menu bar at the base of the window:

Click Apply Shift to apply the suggested 50 ms shift.

152
Exercise 3

The correlation plot now looks like this:

The Cross Correlation window now

shows a roughly symmetrical peak at
zero Lag Time, with a maximum
correlation of 61%

153
Exercise 3

From this we can conclude that we have

a good correlation for this well. Click
OK to accept this correlation.

A dialog pops up suggesting a name for

the new sonic log we have created by
the log correlation process. Even
though this process, by default, has
only changed the depth-time curve and
not the actual sonic log, Geoview
calculates a new sonic log (identical to
the previous) as a place holder for
identifying the new depth-time curve.
Click OK to accept that new name:

154
Exercise 3 Identifying scenarios and creating synthetics

The next step in the Workflow is Extract wavelet using wells. We very
often perform that step within the Log Correlation Window. In this
case, we will assume the zero-phase statistical wavelet is adequate, so
we will skip that step here.

Double-click the step after, Identify Scenarios:

155
 Exercise 3
By “scenarios”, we mean the geologic
conditions which we wish to model.
Each scenario is a different fluid
combination within the target
reservoir.

At least one scenario must be chosen

– the in situ scenario which is present
in the logs. In addition, we can model
up to 4 other scenarios. In the figure
above, we have chosen Pure Oil and
Pure Brine, as well. Note that in
addition to specifying pure
hydrocarbons, we can specify any
combination of the 3 components
using the Ternary diagram. Set the
dialog as shown above, and click Run.

156
 Exercise 3

The display is now modified to show

the 3 scenarios for each of the P-
wave, S-wave, Density, and Poisson’s
Ratio curves:

Now double-click the next step on

the workflow, Create Synthetics:

157
 Exercise 3
The dialog on the right shows that, by
default, Zoeppritz ray-tracing will be used
to calculate the synthetics:

The range and distribution of offsets has

been automatically set to be consistent
with the real data being used:

The wavelet is the most recently extracted:

Also, note at the bottom of the dialog a

button for viewing all the Advanced
Parameters:

For this tutorial, click Run to create

synthetics using the default parameters.

158
Exercise 3

The three synthetics look like this:

The last step of the AVO Modeling

Workflow is AVO What Ifs:

159
 Exercise 3
This step is used after the synthetics have been created. Here we can
interactively modify various parameters and see their effects on the
calculated synthetics.

The dialog on the right contains all the

parameters used to generate the in-situ
synthetic.

For example, the in-situ fluid

saturation is 50% brine and 50% gas:

That can be modified to contain any % combination of the 3 components.

Another interesting change is to modify the

thickness of the target layer. The dialog
currently shows the reservoir thickness as
7 meters:

160
Exercise 3
To see the effect of thickness
change, change the thickness
to 20 meters, as shown here:

The display does not

immediately change. To
produce the new synthetic
click the Preview button:

Now the display is modified to

produce new log curves (with
a thicker reservoir) and a new
synthetic:

NOTE: A more rigorous tool

for assessing the impact of
thickness changes is to use
the Wedge Modeling process.
161
Exercise 3

By default, the model changes are not calculated until the Preview button
is clicked. This is because some calculations may take a while. However,
by selecting the Interactive Preview option, you can force the model to be
updated immediately after every change.

Finally, the model changes are normally temporary and disappear as soon
as the dialog is closed. You can save the current model by clicking the
Save Results button.

For this tutorial, just Close the menu, without saving any results:

162
Modifying and saving the workflow

We have now completed the AVO Modeling Workflow. The last

process we will examine in this tutorial is customizing and saving a
new workflow.

Currently, the AVO Modeling Workflow

looks like this:

One of the items, Extract wavelet using

wells, is still colored red, because we
did not explicitly perform that step. We
might wish to create a new workflow,
with that item removed.

163
Exercise 3

To do that, select the item, Extract

wavelet using wells, and right-click:

On the pop-up, select the item

Remove process. Click OK on this
dialog which appears:

164
Exercise 3

The new workflow now appears under

the User tab:

Note that the original workflow is still

saved under the Default tab.

Sometimes we might wish to

add a process. For example,
select the item Extract statistical
wavelet on the User tab, and
right click as shown:

Select Insert New Processes

Below and double-click Check
Shot Correction.

165
Exercise 3

Now we have added the process of

applying a Check Shot Correction
before doing Log Correlation:

As it stands, the new customized

workflow is only available within this
project. To make it available to other
projects and other users, we need to
export the workflow.

166
Exercise 3

To do that, right click anywhere

on the workflow, and click Export
Workflow and Parameters:

On the dialog which appears,

give the new workflow a name,
like Test , and click OK:

167
 Exercise 3
We have now saved the new workflow, and the
parameters used in this project, to two separate
files. To import the saved workflow and
parameters into a new project, click on the Import
Workflow button at the top of the Workflow
menu:
On the dialog which appears, we see the two
files which have been created:
The file with the shorter name,
Test_workflow.xml, is the list of process
names in the new workflow. This is the file
we need to import if we wish to use the
chosen steps in a new project.
The other file, TestAVOModelingParameter_parameter.xml, is the complete
list of parameters used in this current project. If we import this second file,
as well as the first, the dialogs which are created will have exactly the same
parameters as used previously. Thus, the combination of both files together
will be a reproducible history of the project.
168
 Exercise 3

Double-click each of the files named

above. The right side of the dialog now
changes to this:

If we now click OK, we will import both

the list of processes and their
parameters. For this tutorial, click
Cancel on this dialog.

(End of Exercise 3)

169
Elastic Waves and Anisotropy
 Multi-Layer AVO Modeling

As we saw at the end of the previous theory section, multi-layer modeling

in the AVO program consists first of creating a stack of N layers, generally
using well logs, and defining the thickness, P-wave velocity, S-wave
velocity, and density for each layer.
171
Elastic Wave Modeling

 Elastic Wave Modeling is the exact solution for a plane wave

propagating through a series of layers.

 The calculation is done in the frequency domain, so the user must

specify a frequency range, which affects the run-time.

 Ideally, the modeling should include all effects, such as multiples

and converted waves. In principle, these can be turned off, but
that may produce instability.

 Elastic Wave Modeling can model critical and post-critical events.

172
Zoeppritz – Elastic Wave Comparison

Zoeppritz Elastic wave

Modeling
Primaries x x
Multiples x
Converted Waves x
Refractions x
Anisotropy (VTI) x x
Frequency dependent absorption x

Higher order move-out x

Post-critical events x
Fast computation x

The following example, taken from a paper by Simmons and Backus (1994),
illustrates the difference between Zoeppritz and Elastic modeling.

173
The Oil Sand Model

Simmons and Backus used the thin bed oil sand model shown above.

174
The Possible Modeled Events

Simmons and Backus (1994)

175
Responses to Various Algorithms
(A) Primaries-only Zoeppritz, (B) + single leg shear, (C) + double-leg shear,
(D) + multiples, (E) Wave equation solution, (F) Linearized approximation.

Primaries only Zoeppritz

+ single leg shear

+ double leg shear

+ multiples

Wave equation

Aki-Richards
Simmons and Backus (1994)
176
Zoeppritz vs Elastic Wave Summary

 EW Modeling models a wider range of effects, including

multiples and converted waves.

 Zoeppritz Modeling is much faster than EW Modeling.

 For thin layer models, the results may be significantly different.

 EW Modeling should always be more accurate than Zoeppritz

Modeling, if the input parameters are correct.

 However, EW Modeling may be more sensitive to log editing

problems.

177
Anisotropic AVO
In an isotropic earth P and S-wave velocities are independent of angle.

However, in an anisotropic earth, velocities and other parameters are

dependent on direction, as shown below.

VP(90o)

VP(45o)
VP(0o)

In particular, we will consider the cases of Transverse Isotropy with a

vertical symmetry axis, or VTI, and Transverse Isotropy with a Horizontal
symmetry axis, or HTI . In particular, the HTI model gives us a way to model
azimuthal AVO, or AVAZ.

178
 Anisotropic AVO
The figure below, from Ruger, illustrates the difference between the VTI and
HTI models of anisotropy.

From “Reflection Coefficients and Azimuthal AVO Analysis in Anisotropic

Media” by Andreas Ruger, SEG Geophysical Monograph No. 10, 2002
The VTI model consists of horizontal layers and can be extrinsic, caused
by fine layering of the earth, or intrinsic, caused by particle alignment as in
a shale. The HTI model consists of vertical layers and is caused by parallel
vertical fractures or steeply dipping shales (see Appendix 5).
179
Velocities for the VTI case
Although the equations for full anisotropy are quite complex, Thomsen
(1986) showed that for weakly anisotropic materials the velocities in VTI
media are dependent on the parameters , , and  , called Thomsen’s
parameters.

Here is Thomsen’s original definition of the velocities as a function of his

parameters:

VP (q )  VP (0 ) 1   sin q cos q   sin q 

o 2 2 4

 V 2
(0 o
) 2 
VSV (q )  VSV (0 ) 1  2 o (   ) sin q cos q 
o P 2

 VSV (0 ) 
VSH (q )  VSH (0o ) 1   sin 2 q 

180
 Thomsen’s Parameters
Thomsen’s parameters are simply combinations of the differences between
the P and S velocities at 0, 45, and 90 degrees. The following relationships
can be derived quite easily using the velocities in the previous slide:

VP (90o )  VP (0o ) VSH ( 90 o )  VSH ( 0 o )

 
VP (0o ) VSH ( 0 o )

VP ( 45 o )  VP ( 0 o )  VP ( 45 o )  VP ( 0 o ) 
  4 o         4 o 
 V P ( 0 )   V P ( 0 ) 

In the next slides, we will look at the AVO response as a function of

angle for different values of  and . (As mentioned, VSH will not be
used in AVO).

181
AVO and VTI
Thomsen (1993) showed that VTI terms could be added to the Aki-Richards
equation using his weak anisotropic parameters  and , where Ran(q ) is the
anisotropic AVO response and Ris(q ) is the isotropic AVO response.

Ruger (2002) gave the following form of Thomsen’s original equation:

 
Ran (q )  Ris (q )  sin q 
2
sin 2 q tan 2 q ,
2 2
where :    2  1 , and    2   1.
   2    2
or : Ran (q )  A   B   sin q  C   sin q tan 2
q
 2   2 

182
Typical Values for Delta, Epsilon and Gamma

Typical values for , , and  were given by Thomsen (1986). Here are some
representative values from his table:

Lithology VP(m/s) VS(m/s) rho(g/cc) epsilon delta gamma

sandstone_1 3368 1829 2.50 0.110 -0.035 0.255

sandstone_2 4869 2911 2.50 0.033 0.040 -0.019

calcareous sandstone 5460 3219 2.69 0.000 -0.264 -0.007

immature sandstone 4099 2346 2.45 0.077 0.010 0.066

shale_1 3383 2438 2.35 0.065 0.059 0.071

shale_2 3901 2682 2.64 0.137 -0.012 0.026

mudshale 4529 2703 2.52 0.034 0.211 0.046

clayshale 3794 2074 2.56 0.189 0.204 0.175

silty limestone 4972 2899 2.63 0.056 -0.003 0.067

laminated siltstone 4449 2585 2.57 0.091 0.565 0.046

183
AVO and VTI
Blangy (1997) computed the effect of anisotropy on VTI models of the three
Rutherford-Williams type. Blangy’s models are shown below, but since he
used Thomsen’s formulation for the linearized approximation, his figures
have been recomputed in the next slide for the wet and gas cases using
Ruger’s formulation. The slide after that shows our example.

184
 VTI – AVO Effects

Class 1
Class 1
 = -0.15
 = -0.3
Class 2
Class 2

Class 3

Class 3

Isotropic
--- Anisotropic
(a) Gas sandstone case: Note (b) Wet sandstone case:
that the effect of  and  is Note that the effect of  and
to increase the AVO effects.  is to create apparent AVO
decreases.
185
VTI Applied to Colony Example

Isotropic vs Anisotropic AVO

Gas Sand Top,  = -0.15,  = -0.3

0.000
Amplitude

-0.100
-0.200
-0.300
-0.400
0 5 10 15 20 25 30 35 40 45
Angle (degrees)

R (Isotropic) R (Anisotropic)

186
 VTI AVO Model Example

In the above display, we have added simple  and  logs to the sonic
and density logs from the Colony gas sandstone play in Alberta. Notice
that only the gas sand is isotropic.
187
Anisotropic AVO Synthetics

(a) Isotropic (b) Anisotropic (a) – (b)

In this display, the synthetic responses for the logs shown in the
previous slide are shown. Note the difference due to anisotropy.

188
Exercise 4:
The Colony Gas Sand
Elastic Wave Modeling
 Exercise 4
In the previous exercise, we used Zoeppritz
ray-tracing to create synthetics corresponding
to a series of lithologic scenarios. In this
exercise, we will use elastic wave modeling.

We could do that by changing the AVO

Modeling workflow. Instead, we will use the
Processes list. To see that list, click on the
Processes tab:

Then, from the list, expand the AVO Modeling

item and double click Create AVO Synthetics:

190
 Exercise 4

On the menu which appears, we

see that all the parameters have
been saved from our previous
Zoeppritz synthetic.

As shown, change only the two

items – the Algorithm is Elastic
Wave and the Output Name is
modified to include the word
“synthetic_elastic”.

When you have changed these

items, click on OK to create the
new synthetic.

191
 Exercise 4
The new elastic wave synthetic is
plotted in the Geoview window.

By default, the original synthetics

are no longer visible. We would
like to see both the new elastic
wave synthetic and the original
Zoeppritz synthetic.

An easy way to add components to

the window is to drag-and-drop
from the Project Data list. Click on
the Project Data tab and the
Seismic side-tab, as shown. Then
select the Zoeppritz synthetic,
which was called In-
situ_AVO_WELL.
192
 Exercise 4
Holding the left mouse button down, drag the name (In-situ_AVO_WELL) over
between the elastic wave synthetic and the real data display. You will know
where you are dropping it, because of the green vertical line which appears.
Release the mouse button and the original synthetic is added to the display.

193
 Exercise 4

The new elastic wave

synthetic is plotted along with
the previous Zoeppritz
synthetics. We can see
significant differences in
character due to the elastic
wave modeling. In particular,
the elastic wave model looks
more noisy and does not
display the AVO anomaly as
strongly.

194
 Exercise 4
A very convenient tool for analyzing the
amplitudes of individual events is AVO Gradient
Analysis.

To start that, select the Processes tab and

double-click the item on the Processes list:

On the menu which appears, first

set the Input as the in-situ
synthetic which was originally
created. Then click the Specify
Velocity button:

195
 Exercise 4
The dialog which appears is
used to set the velocity field
for the AVO attribute
calculation. In this case, we
used a single P-wave log. So,
select Single Well: P-wave
Curve from the pull down
menu:

Click on P-wave_corr.

Click OK on the bottom of the

Velocity Field Model dialog:

196
 Exercise 4
Then, click OK on the bottom of
the gradient analysis menu:

The display which

appears shows the
in-situ synthetic
along with a series of
picked amplitudes:

The picks are not

quite correct, since
we have not
specified the location
of the target event.

197
Exercise 4
First, improve the synthetic
display by clicking on Fit to
View:

We can create more space for the

Gradient Analysis display by
temporarily removing the Project
Manager. To do that, click the “x” as
shown:

Now, click on the trough

corresponding to the top of the gas
sand at around 630 ms:

198
Exercise 4
The display should change to this, showing the picked events from the
synthetic along with an AVO curve. If your display looks quite
different, try clicking close to the event again.

199
 Exercise 4

There are many useful options associated with the

AVO Curve display, which we will study in a later
exercise. For now, we will use the display to
compare the picked amplitudes between the two
synthetics. To bring in the second synthetic, click
on View 2, as shown:

On the menu which appears,

set the Input for View 2 to be
the Elastic Wave synthetic
we just created. Then click
Apply at the base of the
menu.

200
 Exercise 4
Now the two synthetics appear, with the two sets of picks:

As we saw on the
synthetic displays,
the curve for the
Elastic Wave
synthetic does not
show as much AVO
variation as the
Zoeppritz synthetic.

201
Exercise 4
We can see this even better by
normalizing the amplitudes of the near
traces.

To do that, click on Pick Normalization:

Check on Tie Intercepts and

click Ok:

202
 Exercise 4

Now we see clearly the

difference in AVO behavior:

Of course, we want to compare

these curves with the real data
event. To see that, click to turn on
View 3:

203
 Exercise 4

On the menu which appears, we

must select the input (gathers),
and also specify which CDP to
show. Click the down arrow and
select the well location:

When you are done, the menu will

look like this. Click Apply at the
base of the window.

204
 Exercise 4
Once again, click Pick Normalization. Then
click Ok on the dialog that appears to force
all the intercepts together.

Now we see that the real data event

and the Zoeppritz model lie
practically on top of each other.

This confirms that the Zoeppritz

model is a very good representation
of the real response.

(End of Exercise 4)
205
AVO Analysis on Seismic Data
Introduction

 Although the Zoeppritz equations give us the exact amplitudes as a

function of angle, the equations themselves do not lend themselves to an
intuitive understanding of the AVO process for angles greater than zero
degrees.

 For that reason, although modeling should be done with the Zoeppritz
equations, most AVO theory for analyzing real data is based on a
linearized approximation to the Zoeppritz equations initially derived by
Bortfeld (1961) and then refined by Richards and Frasier (1976) and Aki
and Richards (1980).

 The equations on the next few slides will show various equivalent
formulations of the Aki-Richards equations.

 We will then develop an intuitive understanding of these equations using

the rock physics model developed in the first section.

207
The Zoeppritz Equations

Zoeppritz Equations
1919

1
  sin q1  cos 1 sin q 2 cos 2 
 RP (q1 )  cos q  sin 1 cos q 2  sin 2   sin q1 
 R (q )   1   cos q 
 S 1    sin 2q r 2VS 2VP1 r 2VS 2VP1
cos 22 
2
VP1  1 
cos 21 sin 2q 2
 TP (q1 )   1
VS 1 r1VS12VP 2 r1VS1 2
 sin 2q1 
   r 2VP 2 rV
  
 TS (q1 )   cos 21 cos 21 
VS 1
sin 21 cos 22  2 S 2 sin 22 
 VP1 r1VP1 r1VP1 

208
Approximations to Zoeppritz Equations

Zoeppritz Equations 1919

Ostrander Aki - Richards Equation 1980 Connolly

1980 Vp, Vs and Density 1999 EI

Shuey 1985 Wiggins 1983 Smith and Gidlow 1987 FF

A, B, C
The AVO equation
Fatti 1994
Rp, Rs, Rd

HRS Intercept,
Gradient, and
combinations HRS Sim Inv HRS Rp,Rs

209
The Aki-Richards Equation
The Aki-Richards equation is a linearized approximation to the Zoeppritz
equations. The initial form of this equation separated the velocity and
density terms.

 VP  VS r
RP (q )  a b c , where :
2VP 2VS 2r

r 2  r1
a  1  tan q ,
2 r , r  r 2  r1 ,
2
 VS 
2
VP 2  VP1
b  8  sin 2 q , VP  , VP  VP 2  VP1 ,
2
 VP 
VS 2  VS 1
2 VS  , VS  VS 2  VS 1 ,
V  2
c  1  4 S  sin 2 q ,
 VP  q q
and q  1 2 .
2
210
The Aki-Richards Equation
The Aki-Richards equation is a linearized approximation to the Zoeppritz
equations. The initial form of this equation separated the velocity and
density terms (note that sometimes this is written without the factor 2 in
the denominator and with a, b and c scaled by 2):
 VP  VS r
RP (q )  a b c , where :
2VP 2VS 2r
r 2  r1
a  1  tan q ,
2 r , r  r 2  r1 ,
2
 VS 
2
VP 2  VP1
b  8  sin 2 q , VP  , VP  VP 2  VP1 ,
2
 VP 
V  VS 1
V 
2 VS  S 2 , VS  VS 2  VS 1 ,
c  1  4 S  sin 2 q , 2
 VP  q q
and q  1 2 .
2
As we will see when we get to the section on Elastic Impedance (EI), this
is the form of the equation that was used in the derivation of EI.

211
 Understanding Aki-Richards

To understand the Aki-Richards equation, let us look at a picked event

at a given time on the 3 trace angle gather shown below:

Constant Angle
o o o Each pick at time t and angle q is equal to
0 15 30 the Aki-Richards reflectivity at that point
600 ms
(after convolution with an angle-dependent
t Picks wavelet) given by the sum of the three
weighted reflectivities. If we assume that at
time t, (VS/VP)2= 0.25, we see that:
700 ms
 VP r
RP (0 o ) 
2VP
0
2r

Note : sin 0 o  tan 0 o  0 
 VP  VS r
RP (30 o )  1.333  0.500  0.750
2VP 2VS 2r
Note : sin 2

30 o  0.25 and tan 2 30 o  0.333
212
 The A, B, C Equation

Wiggins et al. (1983) derived an equivalent, but algebraically reformulated,

form of the Aki-Richards equation. They separated the equation into three
reflection terms, each weaker than the previous term:

RP (q )  A  B sin 2 q  C tan 2 q sin 2 q , where :

1   VP r 
2 2
1  VP VS   VS VS  r 1  VP
A   , B   4   2  ,C  .
2  V p r  2 Vp VP  VS VP  r 2 Vp

This is often considered to be “the” AVO equation. A is called the intercept,

B the gradient, and C the curvature. Note that A is identical to the
linearized zero-angle reflection coefficient, which we called RP(0o) in the
previous slide.

Also, the physical interpretation of this equation is the same as the

previous slide except that the weights are now 1, sin2q, tan2q *sin2q, and
the physical parameters are A, B and C.
213
Fatti et al’s Formulation of the Aki-Richards Equation
An third equivalent form of the Aki-Richards equation was formulated by
Fatti et al. (Geophysics, September, 1994) and is written:

RP (q )  c1RP (0o )  c2 RS (0o )  c3 RD , where :

c1  1  tan 2 q , c2  8(VS / VP ) 2 sin 2 q , c3  4(VS / VP ) 2 sin 2 q  tan 2 q ,
1   VP r  1   VS r  r
RP (0 )  
o
  , RS (0 )  
o
  , and RD  .
2  VP r  2  VS r  2r
Note that the RP(0o) term given above is identical to the A term in the
previous equation. Also, the first two scaling terms are identical to those in
the original Aki-Richards equation. This equation will be used later in the
course as the basis for independent and simultaneous pre-stack inversion.

The physical interpretation of this equation is the same as for the original
Aki-Richards equation except that the weights are now c1, c2, c3, and the
physical parameters are RP(0o), RS(0o) and RD.
214
 A Summary of the Aki-Richards Equation

All three forms of the Aki-Richards equation consist of the sum of three
terms, each term consisting of a weight multiplied by an elastic parameter
(i.e. a function of VP , VS or r). Here is a summary:

Equation Weights Elastic Parameters

 VP  VS r
Aki-Richards a, b, c , ,
2VP 2VS 2 r

Wiggins et al. 1, sin 2 q , tan 2 q sin 2 q A, B, C

Fatti et al. c1 , c2 , c3 RP (0o ), RS (0o ), RD

Note that the weighting terms b, c and c2, c3 contain the squared VS/VP ratio as
well as trigonometric functions of q. However, in the Wiggins et al.
formulation, this term is in the elastic parameter B.

215
Physical Interpretation
A physical interpretation of the three equations is as follows:

(1) Since the seismic trace consists of changes in impedance rather than
velocity or density independently, the original form of the Aki-Richards
equation is rarely used.
(2) The A, B, C formulation of the Aki-Richards equation is very useful for
extracting empirical information about the AVO effect (i.e. A, which is
called the intercept, B, called the gradient, and C, called the curvature)
which can then be displayed or cross-plotted. As pointed out in the
previous slide, explicit information about the VP/VS ratio is not needed
in the weights.
(3) The Fatti et al. formulation gives us a way to extract quantitative
information about the P and S reflectivity which can then be used for
pre-stack inversion. As shown in Appendix 1, the terms RP0 and RS0
are the linearized zero-angle P and S-wave reflection coefficients.

216
Wet and Gas Models

Let us now see how to get from the geology to the seismic using the
second two forms of the Aki-Richards equation. We will do this by using
the two models shown below. Model A consists of a wet, or brine, sand,
and Model B consists of a gas-saturated sand.

VP1,VS1, r1 VP1,VS1, r1

VP2,VS2, r2 VP2,VS2, r2

(a) Wet model (b) Gas model

217
Model Values
In the section on rock physics, we computed values for wet and
gas sands using the Biot-Gassmann equations. Recall that the
computed values were:

Wet: VP2 = 2500 m/s, VS2= 1250 m/s, r2 = 2.11 g/cc, 2 = 0.33

Gas: VP2 = 2000 m/s, VS2 = 1310 m/s, r2 = 1.95 g/cc, 2 = 0.12

Values for a typical shale are:

Shale: VP1 = 2250 m/s, VS1 = 1125 m/s, r1 = 2.0 g/cc, 1 = 0.33

The next four figures will show the results of modeling with the
ABC and Fatti equations. On these four figures, the curves have
been calculated as a function of incident angle and scaled to
average angle.

218
 Zoeppritz vs the ABC Method – Gas Sand

This figure on the right

shows the AVO curves AB method
computed using the
Zoeppritz equations
and the two and three
term ABC equation, for
the gas sand model. ABC method
Notice the strong
deviation for the two
term versus three term Zoeppritz
sum.

219
 Zoeppritz vs the ABC Method – Wet Sand

This figure on the

right shows the AVO
curves computed
using the Zoeppritz
equations and the ABC method
two and three term
ABC equation, for
the wet sand model. Zoeppritz

Again, notice the

strong deviation for
the two term versus
three term sum. AB method

220
Zoeppritz vs the Fatti Method – Gas Sand

This figure on the

right shows the AVO
curves computed
using the Zoeppritz
Zoeppritz
equations and the
two and three term
Fatti equation, for Fatti method,
the gas sand model.
two term
Notice there is less
deviation between
Fatti method,
the two term and three term
three term sum than
with the ABC
approach.

221
 Zoeppritz vs the Fatti Method – Wet Sand

This figure on the

right shows the AVO
curves computed
using the Zoeppritz
equations and the two Zoeppritz
and three term Fatti
equation, for the wet Fatti method,
sand model. two term
As in the gas sand
case, there is less
deviation between the
two term and three Fatti method,
term sum than with three term
the ABC approach.

222
 The Two-Term Aki-Richards Equation

Intercept / gradient analysis is done with the two-term Aki-Richards equation.

Recall that:

R( q )  A  B sin 2 q
where we have dropped the C term and define A and B as:

1   VP r 
2 2
1  VP VS   VS VS  r
A   , B  4   2  ,
2  V p r  2 Vp VP  VS VP  r

or, using Shuey’s approximation (Appendix 6):

 1  2    VP / VP
B  A D  2( 1  D )   , D  .
 1  ( 1 ) 2
 VP / VP  r / r
223
The Two-Term Aki-Richards Equation
It is common practice to use only 2 terms because:

(1) It simplifies the analysis considerably

(2) For angles less than about 40 degrees, the third term is not significant,
as shown previously:

224
Estimating the Intercept and Gradient

Our objective is to determine values for A and B by analyzing real seismic

data.

Step 1 involves converting from offset to angle.

Step 2 involves fitting a regression line to the amplitude picks as a function

of the sine of the angle squared.

The next two slides will illustrate this in simple fashion.

225
 Converting from Offset to Angle
450 Offset (m) 6000 0 Angle (degrees) 90

The offset
domain is the The angle domain
conventional represents a
CDP stack with theoretical
each trace at a acquisition
different geometry in which
offset. The each trace
acquisition corresponds to a
geometry is constant incidence
shown below. angle.

226
Converting from Offset to Angle

Conversion from offset to angle can be done using one of these options:
(1) Straight ray assumption (constant velocity)
(2) Ray Parameter approximation (variable velocity approximation)
(Reference: Walden, 1991, Making AVO sections more robust: Geophysical
Prospecting, 39 , no. 7, 915-942.)
(3) Ray-tracing (variable velocity)

(1) Straight Ray :

(2) Ray Parameter :
X X
tan q   , XVINT
2d Vt0 sin q  2 ,
tVRMS
where X  offset ,
where VINT  Interval velocity,
VRMS t0
d  depth  , t  total traveltime.
2
t0  2  way time,
VRMS  RMS velocity.

227
 Converting from Offset to Angle
Ray Ray
Tracing Parameter

The Ray Parameter method is

usually used when analyzing
seismic data.

This is because it is much faster

than Ray Tracing.

Because of the approximations,

it begins to degrade slightly at
angles greater than 50 degrees.

The example on the right shows 65 65

angle contours calculated using
Ray Tracing and Ray Parameter.

35 43 50 35 43 50
228
 Common Offset Picks as Function of sin2q
The pick amplitudes are extracted at all
times, two of which are shown.

Offset
+A
+B

sin2q

Time -B
-A
The Aki-Richards equation predicts a
linear relationship between these
amplitudes and sin2θ.

Regression curves are then

calculated, to give A and B values for
each time sample.
229
 Intercept: A
The result of this
calculation is to
produce 2 basic
attribute volumes

Gradient: B

230
Derived Attributes

The raw A and B attribute volumes are rarely used in that form. Instead,
other AVO attributes are usually calculated from them.

The most popular AVO attributes are:

(1) AVO Product : A*B

(2) Scaled Poisson’s Ratio Change : A+B
(3) Shear Reflectivity : A-B
(4) Fluid factor

Note that the full mathematical way of extracting attributes is given in

Appendix 7.

231
Derived Attributes AVO Product : A*B
Many AVO anomalies have the form
shown at the right.

In this case, both the intercept (A) and

the gradient (B) are large numbers or
“bright”. Also, they have the same
sign.
+A
+B
This is an example of a Class 3
anomaly.
sin2q
Forming the product of A and B, we
get:
-B
-A
Top of sand : (-A)*(-B) = +AB
Base of sand : (+A)*(+B) = +AB

This gives a positive “bright” response

at both top and base.
232
Derived Attributes: AVO product

The AVO product shows a positive response at the top and base of the
reservoir:

Top

Base

233
Derived Attributes
Scaled Poisson’s Ratio Change : A+B
The second combination is derived from Shuey’s equation:
RP (q )  A  B sin 2 q , where :
1   VP r   1  2  
A   , B  A D  2(1  D)   ,
2  V p r   1    (1   ) 2

VP / VP   1
D ,  2 , and    2   1.
VP / VP  r / r 2

If we assume the background Poisson’s Ratio = σ = 1/3, then:

 1 
B  A D  2(1  D)    2.25  A
 2  (2 / 3) 2

or: A  B  2.25 The sum A+B is proportional to the

change in Poisson’s Ratio.

234
Derived Attributes
Scaled Poisson’s Ratio Change : A+B

The AVO sum (A+B) shows a negative response at the top of the reservoir
(decrease in σ) and a positive response at the base (increase in σ):

Top

Base

235
Derived Attributes
Shear Reflectivity : A-B
The third combination is derived from the Aki_Richards equation:

RP (q )  A  B sin 2 q  C tan 2 q sin 2 q , where :

  VP r 
2 2
 VP  VS   VS  VS  r
A   and B   4   2  .
 2V p 2 r  2V p VP  VS VP  r
If we assume the background VS /VP = 1/2, then:

 VP  VS r   VP r    VS r 
B         RP 0  2 RS 0 ,
2V p VS 2 r  2V p 2 r   VS r 
  VP r    VS r 
where : RP 0  A     and RS 0    
 p
2 V 2 r   SV r 

or: RS  0.5 A  B  The difference A-B is proportional to

the Shear Reflectivity.
236
A-B Difference Attribute

The AVO difference (A-B) shows an increase in Shear Impedance at the top
of the reservoir. This calculation is usually done with the more accurate
Fatti equation, which we will see next.

Top

Base

237
 RP and RS Attributes

An alternate form of the Aki-Richards equation was formulated by Fatti et al.

(Geophysics, September, 1994) which can be written (for 2 terms) as:

RP (q )  c1 RP (0o )  c2 RS (0o ), where :

c1  1  tan 2 q , c2  8(VS / VP ) 2 sin 2 q ,
1   VP r  1   VS r 
RP (0 )  
o
  and RS (0 )  
o
 .
2  VP r  2  VS r 
This allows us to calculate RP0 and RS0 volumes from seismic data in exactly
the same way as A and B volumes.

Again, note that the full mathematical way of extracting attributes is given in
Appendix 7.

238
RP0 and RS0 Attributes
RP0

RS0

239
RP0 and RS0 Attributes

The RP0 and RS0 attributes are usually transformed into one of 2 new
attributes:

(1) Fluid Factor, which will be discussed next.

(2) AVO Inversion or Lambda-Mu-Rho, which will be discussed in

a later section.

240
Derived Fluid Factor Attribute
The Fluid Factor attribute (Smith and Gidlow, 1987, Fatti et al., 1994) is
based on Castagna’s mudrock equation, which is assumed to be true for
non-hydrocarbon filled layers:
VP  1.16 VS  1360 m / s
Using calculus, we can derive the following equivalent equation:
VP  VS  VS
VP  1.16VS  Divide by Vp   1.16 
VP  VP  VS
The Fluid Factor is defined to highlight layers where Castagna’s equation
does not hold, i.e., potential hydrocarbon zones:

VP  VS  VS  VS 
F   1.16  , or : F  RP  1.16  RS
VP  VP  VS  VP 
 VS 
Note that the factor 1.16  is often customized to fit the local data.

 VP 
241
 Mudrock Line

- Castagna et al (1985)
cross-plotted Vp vs. Vs

S-wave velocity (km/s)

for different types of
sedimentary rocks 3 gas
sandstones
carbonates
- ‘mudrock’ line for fluid 2 water
sandstones
saturated sandstones
dry
sandstones
1
- deviations from the
Mudrock line
mudrock line indicate
other lithologies and 1 2 3 4 5
pore fluids
P-wave velocity (km/s)

242
Fluid Factor attribute
The AVO Fluid Factor shows a strong deviation from the mudrock trend at
both the top and base of the 15 m Colony reservoir. Fluid Factor anomalies
also appear for two stacked gas sands above the main Colony reservoir.
The carbonate under the Paleozoic unconformity also deviates from the
from the mudrock trend. Note the different polarity in this case (red over
blue instead of blue over red).

Top

Base Top Colony

Base Colony

Top Carbonate

Downton, Russell & Lines, 2001

243
Approximations to Zoeppritz Equations

Zoeppritz Equations 1919

Ostrander Aki - Richards Equation 1980 Connolly

1980 Vp, Vs and Density 1999 EI

Shuey Wiggins 1983 Smith and Gidlow 1987 FF

1985 A, B, C
The AVO equation
Fatti 1994
Rp, Rs, Rd

HRS Intercept,
Gradient, and
HRS Sim Inv HRS Rp,Rs
combinations

244
Exercise 5
The Colony Gas Sand
Calculating AVO Attributes
Exercise 5
So far, in our analysis, we have used the AVO
Modeling Workflow to create AVO synthetics.
We will now turn to the analysis of the real
seismic data.

One way to do that is to use the AVO Attribute

Analysis workflow, shown in the list of
standard Workflows.

We will do that in the 3-D AVO exercise later

in the course.

For this exercise, we will use the standard

Processes list.

246
Exercise 5

To start that, click the tab called

Processes. You will see a list of all the
operations which are available in
Geoview. Each of the processes can be
expanded. For example, if you click on
both the Seismic Processing and AVO
Analysis options, the following
expanded list is seen:

247
Exercise 5
To start, we will create a CDP stack.
Click next to the Stack option in the
Seismic Processing submenu to see the
two types of stack available, and
double-click CDP Stack:

Now the parameters for this

process appear on the right:

248
 Exercise 5
There are some features of this
dialog which are common to all
Process Parameter dialogs. For
example, there is a location to
specify the input and output files
names:

There is a section to specify the data

range to process. By default, it is
the entire volume:

For example, we could choose to

stack only a limited range of offsets:

249
Exercise 5
By default, only the most critical parameters for this process are
specified on this page. To see the more advanced option, click the
button at the base of the menu:

This produces a series of extra tabs, which control in detail the

process to be performed. Note that these will differ from one
process to the next.

At the top of the menu, there is a small

icon showing an “airplane”:

Click that icon and the Parameter Dialog

detaches from the Geoview window to
allow it to be moved aside, making the
data more visible. Clicking the
“airplane” again re-attaches the dialog.:

250
Exercise 5
At the base of the Parameter Dialog, we see a series of buttons:

If we click the Run Batch button, that will create a batch file which
could run the process later. That is often helpful for long, computer-
intensive processes. For now, click OK to start the CDP stack process
as usual.

When the process

finishes, the Geoview
window looks like this
– a split window
showing both the input
and output volumes:

251
Exercise 5
The default split-screen display is very
useful for looking at the results, but
there are many modifications possible.

For example, you can increase the

available plot space by clicking the “x”
on the Project Manager window, as
shown, to temporarily hide that window:

To restore the Project Manager window,

click its name to the left:

252
Exercise 5
You can also temporarily hide one of
the views. For example, click on the
first icon shown below to temporarily
hide View 1, which shows the input
data:

To restore View 1, click it again:

Click the fourth button to set the

orientation horizontally:

Click the fourth button again to restore

the vertical orientation:

253
 Exercise 5
Finally, to see the most complete
control of the seismic display, right-
click on either of the seismic windows.
A pop-up menu appears:

One of the items is View>Seismic View

Parameters:

If you click this item, a dialog

appears, allowing complete
control of the display:

To continue this tutorial,

click Cancel on this dialog.

254
Exercise 5
The next process we will apply is Super Gather. Super Gather is the
process of forming average CDPs to enhance the signal-to-noise ratio.
We do the averaging by collecting similar offset traces within adjacent
CDPs and adding them together. This process reduces random noise,
while maintaining amplitude versus offset relationships.

We could find the Super Gather on

Process list as we did before, but now
we will use a little trick to speed up
that search.

Notice that at the top of the Processes

tab, there is a box called Filter. This is
used to quickly find a process in the
list.

Type in the word “gather”, as shown.

Now, the list is reduced to five entries:

255
Exercise 5

Double-click Super Gather

On the Super Gather Parameter dialog,

the only changes we will make are to set
the Input name to gathers and the output
name to super_gather ; change the Size
of Rolling Window to 5:

This means that five adjacent CDP’s will

be summed to give each output CDP.
Notice that the program has defaulted to
create output bins with 11 offsets each.
This was chosen because that is the
average fold of the input gathers. When
you have changed these parameters,
click OK to run the process.

256
Exercise 5
The result looks like this:

Notice that the gathers look

cleaner and more
consistent, with a
pronounced AVO anomaly at
around 630 ms.

Now we will display the

range of incident angles as
a color display.

On the window showing the

Super Gather, right-click and
select Color Data Volume and
Incident Angle:

257
Exercise 5
We can see from this display that the maximum incident angle
at the zone of interest (630 ms) is around 30 degrees. That
information will be used in a later step.

258
Exercise 5
In this step, we will pick an event at the zone of interest and display those
picks to observe the AVO anomaly.

We will pick the Super Gather volume.

First, turn off the color display by right-
clicking on that display and selecting
Color Data Volume > none:

Then, hide View 1, so that the Super

Gather display fills the window:

259
 Exercise 5
Now select Horizon > Pick Horizons:

On the dialog which appears, we must

specify which data set we are picking.
We are picking the Super Gather in View
2, so this field must be modified:

The final dialog looks like this. Click

OK to start the picking process:

260
 Exercise 5
A series of controls appears at the base of the seismic window. These are
used for the picking process:

One item, called Mode, determines how

picks will be created using mouse
clicks:

The Rubber Band Mode means that if you click somewhere, then
hold the left mouse button down, move the mouse along the
section and release the button, picks will be created in the
region of the “rubber band” which appears between the mouse
clicks. That is very useful for detailed picking.

261
 Exercise 5
For a very clean data set like this one, a convenient mode is Left & Right
Repeat. In this case, you would click a point that you interpret as being part
of the horizon. This becomes the seed point. Picks will be created
throughout the entire line based on this point.

We wish to pick the Trough which

shows the AVO anomaly at around 630
ms. Change both the Mode and Snap
parameters as shown:

Then position the mouse cursor anywhere near the trough at 630
ms and click once:

262
Exercise 5
The entire event should be picked like this:

If your display looks different, check your Mode and Snap parameters
and click again. There is no need to delete the original picks. They
will be automatically replaced.

If we wished to pick a second event, we would select Horizon > New

Horizon from the picking dialog.

In this case, we are happy with the

single event, so click OK to complete
the picking process:
263
Exercise 5
Now that we have picked the event,
we would like to see a display of the
picked amplitudes. To do that, right-
click and select View > Show Pick
Analysis:

264
 Exercise 5
On the Pick Attribute Option dialog
which appears, choose the option to
Show Pre-stack Picks With Gradient
Analysis. This option is based on the
two term Aki-Richards equation.

Click Next several times to accept all

the defaults for this analysis. The
second last page confirms that we
are using the Two Term Aki-Richards
equation and the velocity field we
have defined previously:

Finally, click OK to get the Pick

Analysis display.

265
 Exercise 5
This display shows the
original pick values (in
blue) and the calculated
Aki-Richards curves (in
red). By scrolling through
the data volume, we can
see that the AVO behavior
is most pronounced in the
vicinity of the well and
flattens out as we move
away.

To remove the Pick Analysis display,

right click on the display window and
toggle OFF the display:

266
 Exercise 5
In this step, we transform from the offset to
angle domain. From the Processes tab,
double-click Angle Gather:

On the Angle Gather Parameters dialog, we

see that we are transforming the volume
super_gather into the new volume
angle_gather:

We know from a previous display that the

maximum angle is about 30 degrees, so we
will change the maximum Angle To to 30,
as shown. Also, note that we are using the
velocity field set up in a previous step
(Note: yours may be called velocity field 1,
rather than velfield1 as shown here.)

When you have modified the Angle Gather

Parameters dialog as shown, click OK to
run the Angle Gather process. 267
Exercise 5
When the process has completed, the Geoview window
shows the calculated Angle Gathers:

268
 Exercise 5
The next step we will perform is AVO
Gradient Analysis. The purpose of this
process is to analyze the AVO behavior of
one or more events at a particular CDP. To
start that process, double click AVO Gradient
Analysis:

On the dialog which appears, we specify

the Input Volume as the super_gather. We
also tell the program that we are analyzing
the CDP near the well, which is CDP 330:

Default all the remaining parameters by

clicking OK at the base of the dialog:

269
 Exercise 5
The display which appears
shows the seismic gather at CDP
330, along with AVO pick values
for the default initial time, which
is at the centre of the gather time
scale:

The first thing we can do is improve the time

scale of the gather data. One quick way to do
that is to select the Fit to View check box:

To zoom in more, click the Zoom In button one

or more times:

270
 Exercise 5

Right now, the analysis is being performed at the

arbitrary time of 550 ms:

We would like to analyze the AVO

anomaly at around 630 ms. So, position
the mouse cursor near the trough at 630
ms and click the left mouse key:

The display should now look like this:

271
 Exercise 5
The red line on the seismic display shows the
time location at which the amplitudes have been
extracted. Those amplitudes are plotted as red
squares on the right-hand graph. The curve
which has been fit through the picks is a plot of
the Aki-Richards two-term equation. We can
confirm this by the information at the top of the
graph:
By clicking various time locations on the gather,
we could see the equivalent picks and curve for
any other event on the gather. Actually, it can
often be helpful to see two events at the same
time. To do this click the Two Events toggle ON:

Now, to make the second event appear,

click near the strong peak below the
target trough:

272
 Exercise 5
Now the display should look like this:

Notice that we are seeing a classic

class 3 AVO anomaly with amplitudes
increasing for both the trough at the
top of the sand (red) and the peak at
the base of the sand (green). Notice
also that the fit of the AVO curves is
extremely good. Mathematically, this
is expressed by the normalized
correlation between the picked
amplitudes and the curves, printed at
the top of the graph:

If we wish to see the same information

at other neighboring CDP’s we can
modify this selection item:
273
 Exercise 5
Notice, also, that the AVO curves are plotted as a
function of Offset, because we have used the
super_gather as input. We can see the same plot
as a function of angle this way: Go to this
selection box at the top of the graph and change
to Angle:

Now we see that the maximum angle for

this event is about 30 degrees, as we
observed when creating the angle
gather:

At the base of the graph we see a series

of tabs. One of them, for example,
allows us to access the Parameters,
which control the calculation of the Aki-
Richards curves:
274
 Exercise 5
Another interesting display is the Cross
Plot of calculated Gradient against
Intercept. This is accessed by clicking
the Cross Plot tab:

The red and green squares correspond

to the calculated Intercept/Gradient
values for the selected events. Note
that the locations of these squares are
consistent with the interpretation of this
anomaly as a class 3 AVO anomaly.

The other colored squares correspond

to intercept/Gradient values from other
times on this CDP, and represent an
approximate wet trend.

275
 Exercise 5
We can control the display of the
background wet trend, by turning on Plot
Background and clicking the Background
parameters button:

Then, set up the menu, as shown; you

will need to click on All Data first.

This causes the background to be

created from a window of data, covering
5 CDPs near the zone of interest.
276
 Exercise 5
Now that we have examined the AVO
anomaly using AVO Gradient analysis, we
will apply the calculation to the entire
volume to see the distribution of AVO
anomalies. To start that, double-click
AVO Analysis > AVO Attribute Volume:

The AVO Attribute process uses the two or

three term Aki-Richards equation to
extract AVO attributes from the seismic
data. The attributes are based on
combinations of intercept, gradient and
curvature, as defined by the Aki-Richards
equation.

On the Parameter dialog on the right,

we see the input and output volumes for
this process:
277
 Exercise 5
Because we have created an angle gather volume, this will be used as input.
Note that the original offset gathers or super gathers could also be used, but
then a velocity field would be needed to convert from offset to angle during
this calculation. As output, the program will create several volumes,
depending on the Type of Analysis. For the default case of two-term Aki-
Richards analysis, the volumes will be called avo_a and avo_b,
corresponding to the intercept and gradient.

Looking further down the Parameters

dialog, we see that the default Type of
Analysis is the Two Term Aki-Richards:

That is appropriate for this case because we only have incident

angles less than 30 degrees. In order to reliably extract three terms
we need high angle data, usually exceeding 45 degrees. Click OK to
extract the AVO Attributes using the default parameters.

278
 Exercise 5
When the process
completes, the
calculated attributes
appear in a split screen:

The window showing the AVO Attributes actually

contains two separate volumes. The annotation
at the top of the window shows what is currently
plotted:

The wiggle trace data is the calculated Intercept (A). The color data is
currently the product of intercept and gradient (A*B). Since this is a class 3
AVO anomaly, we can see a strong positive response at the top and base of
the reservoir at 630 ms.
279
 Exercise 5
Actually, the response is currently
obscured a little by the horizon which is
drawn over it. Temporarily remove that
horizon from the display by right clicking
and selecting View > Seismic View
Parameters:

On the Seismic View Attributes dialog,

select Horizons and No Horizons, as
shown.

Then click OK. The seismic display now

clearly shows the positive AVO response
at the top and base of the reservoir.

280
 Exercise 5
To see another combination of attributes
in color, right-click in that window as
shown:

Select Scaled Poisson’s Ratio Change.

This is the sum (A+B), which is roughly
proportional to the change in Poisson’s
Ratio. This produces this attribute plot:

At the top of the reservoir, we

can see a strong negative
response (orange), indicating a
drop in Poisson’s Ratio, while
at the base of the reservoir we
see a positive response
(yellow), indicating an increase
in Poisson’s Ratio.

(End of Exercise 5) 281

Cross-Plotting AVO
Attributes and the Third
Order Term
 AVO Cross-Plotting
AVO cross-plotting involves plotting the intercept against the gradient and identifying
anomalies. The theory of cross-plotting was developed by Castagna et al (TLE, 1997,
Geophysics, 1998) and Verm and Hilterman (TLE, 1995) and is based on two ideas:

(1) The Rutherford / Williams Classification Scheme.

(2) The Mudrock Line.

Rutherford/Williams Classification
Rutherford and Williams (1989) derived the following classification scheme for AVO
anomalies, with further modifications by Ross and Kinman (1995) and Castagna (1997):
Class 1: High impedance sand with decreasing AVO
Class 2: Near-zero impedance contrast
Class 2p: Same as 2, with polarity change
Class 3: Low impedance sand with increasing AVO
Class 4: Low impedance sand with decreasing AVO

283
Rutherford/Williams Classification

The classification depends only on the

contrast in Acoustic Impedance between
the target sand and the surrounding Acoustic Impedance =
shales:
ρVP

Shale

Sand

Shale

Class 1 Class 2 Class 3

284
Rutherford/Williams Classification
These are the generic AVO curves at the top of the gas sand:

285
An Example of a Class 1 Anomaly

(a) Data example.

(b) Model
example.

Rutherford and
Williams (1989)

286
 Angle Stacks over Class 2 & 3 Sands
(a) Class 2 sand. (b) Class 3 sand.

Rutherford and
Williams (1989)

287
Class 2 & 3 Sands

Class 2 sand. Class 3 sand.

288
 Class 4 Anomalies
Castagna (1995) suggested that for a very large value of A, and a small
change in Poisson’s ratio, we may see a reversal of the standard Class 3
anomaly, as shown below. Castagna termed this a Class 4 anomaly. Here is
an example using Shuey’s approximation:

Letting B  2.25  A :
(1) A  0.3,   0.1
 B  0.575 (Class 3)
( 2) A  0.1,   0.3
 B  0.075 (Class 4)

The figure on the right, from

Castagna et al (1998), illustrates
the concept of the Class 4
anomaly in more detail.

289
Intercept vs Gradient Cross-plot
 The second key in understanding AVO cross-plotting is to derive a
linear relationship between intercept, A, and gradient, B.
 This has been done in two different ways by Castagna et al.
(Framework for AVO gradient and intercept interpretation,
Geophysics, May-June, 1998) and Foster et al. (Interpretation of AVO
anomalies, Geophysics, September-October 2010).
 Although both approaches have merit, the Foster et al. (2010)
approach lends itself to the interpretation of both the wet trend line
and the Rutherford-Williams anomalies, so will be used here.
 Foster et al. (2010) start with the two term AVO expression that we
have looked at often in this course:
RPP (q )  A  B sin 2 q , where :
VP r VP VS r V
A  ,B  8 2  4 2 , and   S .
2V p 2 r 2V p 2VS 2r VP
 You will also recall that VP, VS, r and  are averages of the parameter
across a layer boundary and the  terms are differences.
290
The Foster relationship
 Foster et al. (2010) derive their equation by noting that if we neglect
second order terms, we find that:

 VS VP  
  , where    2   1 , and   2 1 .
 VS VP 2
 Substituting the above equation into the gradient on the previous
slide, and re-arranging terms (see the Appendix), gives:

B  (1  8 2 ) A  4  ( 4 2  1) r / 2 r

 Since  is on average close to 0.5, the third term is much smaller

than the first two terms and can often be dropped, giving:

B  (1  8 2 ) A  4
 The first term of this equation defines the slope of the line in cross-
plot space and the second term is an intercept that is non-zero for a
change in VS / VP ratio.

291
Simple wet and gas sand models
 To illustrate the Foster relationship, we will consider the two simple
models shown below, a wet sand and a gas sand, which are slightly
different than the models we considered earlier.
 The difference in these new models is that we will keep the average 
value equal to 0.5 (that is: VP/VS = 2, and 2 = 0.25).
 This means that the third term in the full Foster expression is
identically equal to zero and for the other two terms we get:
B   A  2

VP1, r1, 1 = 0.5 VP1, r1, 1 = 0.333

VP2, r2, 2 = 0.5 VP2, r2, 2 = 0.667

(a) Wet model (b) Gas model

292
Simple wet and gas sand models
 The linear trends for the brine sand ( = 0) and top ( = +1/3) and
base ( = -1/3) of gas sand are shown below on an A-B cross-plot.
 Note that this is an extremely large value for , but the key point is
that as the absolute value of  increases from zero, the hydrocarbon
lines will move further away from the “wet” line.

 Therefore, an increase in Brine

 moves us to the Sand
Base of
Gas Sand
negative quadrant of A-B
space and a decrease in 
moves us to the positive
quadrant.
 For changes in VP/VS ratio,
the inverse of , a decrease
moves us to the negative Top of
quadrant and an increase Gas Sand
to the positive quadrant.

293
Simple wet and gas sand models

 By considering the positions of the different Rutherford-Williams

classes along the A axis on the cross-plot, and noting that all have a
deviation in VP/VS ratio away from the wet trend, we can position
them on the previous cross-plot as shown below:

 Note that the classes show

a roughly elliptical trend Base 1 Base of
Gas Sand
around the wet trend. Base 2p
Base 2
 Also note that the class 3
sand is the easiest to Top 4
Base 3
identify on the cross-plot
and the class 1 sand is the Top 3 Base 4

hardest to identify, since it

Top 2
might fall on the wet trend.
Top of Top 2p
 The next figure shows a Gas Sand Top 1
seismic data interpretation
of a class 3 sand.

294
Intercept / Gradient Cross-Plots

(a) Uninterpreted gas zone (b) Interpreted gas zone

The “scatter” on the seismically derived data can be attributed to:

(1) VP/VS Ratio variability.
(2) Wavelet interference.
295
Seismic Display from A/B Cross-Plots

(a) Before interpretation

(b) After interpretation

296
 Three Term AVO

As mentioned previously, one problem in AVO and crossplot analysis is that

only the first two terms of the Aki-Richards equation are usually extracted
from the CDP gathers. Recall that the full Aki-Richards equation, as shown
below, has three terms:

R(q )  A  B sin 2 q  C tan 2 q sin 2 q

1   VP r 
where: A  RP 0   
2  VP r 
 VP  VS r
B  4  2
2Vp VS r
2
 VP VS 
C , and     .
2Vp VP 
297
Three Term AVO
An alternate form of the Aki-Richards equation was formulated by Fatti et
al. (Geophysics, September, 1994) which can be written:

RP (q )  c1 RP (0o )  c2 RS (0o )  c3 RD ,
where : c1  1  tan 2 q , c2  8(VS / VP ) 2 sin 2 q ,
c3  4(VS / VP ) 2 sin 2 q  tan 2 q ,
1   VP r  1   VS r 
RP (0 )  
o
 , RS (0 )  
o
 ,
2  VP r  2  VS r 
r
and RD  .
r
Either the A, B, C or the RP, RS, RD terms can be extracted from the seismic
gathers using a least-squares fitting technique with different weighting
coefficients.

298
Density Term

Both forms of the Aki-Richards equation can allow us to estimate density

variations.
Using the original A,B,C form, we see that:

1   V r   VP  r
AC   P    
2  VP r  2V p 2r

This means that if we can estimate all three coefficients, we can generate a
density attribute volume.

That can be very valuable since density is a direct measure of hydrocarbon

saturation. This could solve the “fizz water” problem.

However, the third coefficient can be very noisy since it depends on the far
angle data (>45 degrees), and is very sensitive to noise.

299
 Gulf of Mexico Example

Top

Base

These are angle gathers from the Gulf of Mexico, showing a strong Class II
AVO anomaly. Angles range from 0 to 60 degrees. The target layer is
annotated at right.
300
 3 Term Gradient Analysis

These displays show the results of fitting the Aki-Richards equation, using 2
and 3 terms, to the event highlighted on the previous slide.

Note that the equation for 2 terms begins to deviate from the seismic picks
after about 45 degrees.

2 Term 3 Term

Base

Top

301
Conclusions

 This section discussed the AVO intercept and gradient method.

 First, we looked at the theory behind the intercept and gradient.

 We then looked at an example from a shallow gas sand.

 We then discussed the AVO cross-plotting technique.

 Finally, we showed how to estimate three terms from the Aki-

Richards equation and showed several examples of this approach.

302
Exercise 6:
The Colony Gas Sand
Cross Plotting AVO Attributes
Exercise 6 Cross Plotting AVO Attributes

The final step we will perform on this 2-D AVO example is to create a
cross plot of the derived attributes. The purpose of the cross plot is to
further investigate the type of AVO anomaly and to delineate cross plot
zones which can be mapped to the volume. In HRS-9, there are two
separate approaches to cross plotting seismic data, both of which will
be used in this exercise. These two approaches are as follows:

(1) Launch the Cross Plotting > Cross plot seismic option under
Processes. This allows the user to select a single cross plot zone
by typing a time and trace range in a menu and until recently was
our only option. This will be done first in this exercise.
(2) Launch the View > Create Section Zones … option, interactively
pick a number of zones on the seismic volume, and then launch the
cross plot. This allows the picking of multiple zones as well as
interactive views of the changes as each zone is moved. The
launched cross plot has the same features as in option 1. This will
be done second in this exercise.
304
 Exercise 6 Cross Plotting AVO Attributes

To start the first and oldest approach to cross

plotting, double-click Cross Plotting > Cross
plot seismic:

The parameter dialog which appears has a

number of items which need to be filled in.

We are specifying the Cross Plot Type as AVO attributes and the input
volume is the avo volume just created in the previous step:

305
Exercise 6 Cross Plotting AVO Attributes
Give this cross plot a unique name
crossplot_AVO.

We will analyze a range of CDP’s

from 300 to 360:

We will set the analysis window

around the picked horizon, with a
window size of 100 ms:

When you have filled in the dialog as shown, click OK:

306
 Exercise 6 Cross Plotting AVO Attributes
The cross plot which appears shows the We can improve this plot by
expected background trend through the focusing attention on only the
origin, with anomalous events in peaks and troughs. To do that,
quadrants 1 and 3, consistent with class right-click in the plot area and
3 AVO anomalies. select Set data sample filter:

307
 Exercise 6 Cross Plotting AVO Attributes

Change the Filter Type to Peaks and

Troughs as shown, and click OK:

The new cross plot shows a much

simpler character, with anomalies
clearly separated from the
background trend:

308
 Exercise 6 Cross Plotting AVO Attributes
Now we will highlight the two anomalous zones and project those
zones onto the seismic section.

To draw a zone, first click on the Polygon

icon:

Then move to the cross plot and draw the

shape roughly as shown below, using a
series of left-mouse clicks at each of the
corners of the polygon and double-click on
the last corner to finish the polygon. When
you are done, the screen should look
similar to this:

Note that the polygon can be modified by

grabbing the “handles” and dragging
them.

309
 Exercise 6 Cross Plotting AVO Attributes
You will be asked if you want to
create a seismic plot of the zones.
click on Yes.

Note also that the data area

inscribed by the polygon has been
highlighted on the seismic section
which is now visible in the Seismic
tab. If the red zones appear too
small, expand your zone by
dragging the handles.

This nicely delineates the top of the

sand reservoir (you may want to remove
the horizon to see it better). We can
name this zone by going back to the
cross plot window and typing in a new
name (you may need to click Enter on the
keyboard for this change to apply):
310
Exercise 6 Cross Plotting AVO Attributes

Now repeat this process for the base of

the sand. Click on the polygon icon.
Then draw a polygon around the
anomalous points in quadrant 1:

Call this zone Base of sand:

The seismic cross section now

shows both the top and base of the
sand reservoir delineated:

311
Exercise 6 Cross Plotting AVO Attributes

The cross plot window is now floating over

the Geoview window. We can dock it into its
tab by clicking the Cross Plots button at the
lower right.

We can also dock the seismic window into

the seismic tab by clicking the “Airplane”
button at the lower right.

312
 Exercise 6 Cross Plotting AVO Attributes
Now we will move to the second cross plot option, which allows us to pick
multiple seismic zones. Although one or more areas within the seismic may
be selected, this option does not include the ability to use a horizon to
guide the data selection, so does not fully replace the other option.

Go to Seismic
tab under Project
Data. Double-
click avo(A,B) to
display it on
seismic section
window.

313
 Exercise 6 Cross Plotting AVO Attributes

Under the View menu, select

Create Section Zones. A series of
controls appears at the base of
the seismic window. These are
section zone controls.

To create a new zone, select New

under Section Zone Set and click on
the rectangle icon:

314
 Exercise 6 Cross Plotting AVO Attributes
The menu should now look like
this, allowing you to select
Zone_1. Select the color red by
clicking on the color icon.

Using the mouse, draw

a rectangular zone
around the gas sand
(roughly between
traces 320 and 344
around a time centered
at 630 ms).

315
 Exercise 6 Cross Plotting AVO Attributes
Next, check on Show Cross Plot Control to bring up a set of options.
Check on Feature Group Only and then click Launch Cross Plot:

This will bring up an intercept/gradient

cross plot of the points outlined in Zone
1. Note from our previous theory
section that this is a typical gas zone
signature cross plot.

Also note that your crossplot will

depend on exactly where you drew the
Zone_1 box on the seismic section.
316
 Exercise 6 Cross Plotting AVO Attributes

Now, interactively move

zone_1 by positioning the
mouse over the zone on the
seismic scene (you will see a
hand), by left-clicking and
dragging the mouse.

You should see the points

change to cross plots more
typical of wet zones, as seen
below:

You can also resize the zone by moving to an edge, waiting for an arrow
to appear, and repeat the same sequence. Move Zone 1 back to the time
zone around 630 ms.

317
 Exercise 6 Cross Plotting AVO Attributes

Next, click on the rectangle

icon again to create Zone_2.
Change the color to blue by
clicking on the color icon.

Interactively select the

zone roughly as shown
on the right.

318
 Exercise 6 Cross Plotting AVO Attributes

The crossplot will be

updated with the new points,
colored blue.

Move the zone around

interactively until you define
a wet trend as shown on the
right.

Now we have superimposed two separate seismic zones on the

same crossplot.

Note that this is different than what we did previously, in which we

defined separate cross plot zones on a single seismic zone.
319
 Exercise 6 Cross Plotting AVO Attributes
Next, click on the polygon icon to create
Zone_3, and change the color to green by
clicking on the color icon.

select the zone roughly as shown above

(remember to double-click on the last
point),

The crossplot will be updated with the

new points, colored green, and should
look somewhat as seen on the right.

Now we have superimposed three

separate seismic zones on the same
crossplot.

We could keep adding as many zones as

we want, but will stop at three for this
exercise.
320
 Exercise 6 Cross Plotting AVO Attributes
Right-click inside the crossplot and
select the Show histogram>X
Histogram. We can also turn on the
Y Histogram by repeating the same
process.

The resulting display shows the X

and Y histograms for all three
zones. Note that the distributions
overlap in X and Y but show better
separation in two dimensions.

321
 Exercise 6 Cross Plotting AVO Attributes
Next, remove Zone_3 by clicking on the x
icon at the bottom of the seismic window:

Right-click inside the crossplot and

unselect the Show histogram>X
Histogram and Show Histogram>Y
Histogram successively.

The cross plot will revert to

having only two zones, a red gas
zone and a blue wet zone, as
shown here.

Next, let us select cross plot

zones from these seismic zones.

322
 Exercise 6 Cross Plotting AVO Attributes
Previously, we used the polygonal zone
selection. Now we will use elliptical
zone by selecting the ellipse icon to
create Zone1. We will first find the top
of the gas sand in the negative
intercept/gradient region.

To create this elliptical region, click

once to pin the start of the ellipse, a
second time to pin the end, move the
mouse to open the ellipse up and then
click a third time to end the process,
as shown here.

323
 Exercise 6 Cross Plotting AVO Attributes
You will be asked if you want
to create a seismic plot of the
zones. click on Yes.

The resulting plot shows that

you have picked the top of
the gas sand. Notice that the
first cross plot zone roughly
coincides with the first
seismic zone. They will never
be exact because the
crossplot zone extends to
other values in the seismic
data volume.

324
 Exercise 6 Cross Plotting AVO Attributes

Now, let’s create a zone for

the base of the gas sand in
the positive intercept/gradient
region.

Click again on the ellipse icon

to create Zone2 and fit an
ellipse as shown here.

On the floating seismic

display window, the base of
the gas sand should appear.

325
 Exercise 6 Cross Plotting AVO Attributes

Finally, create an elliptical

cross section Zone3 by
using the blue points from
the second seismic zones,
the wet zone.

The final result should look

like this. Now, the zone
name will depend on which
zone you highlight.
Highlighting a zone will
allow you to move, resize,
or rotate the ellipse, which
you can make your zones
look close to the ones
shown here.

326
 Exercise 6 Cross Plotting AVO Attributes

To display the section

zones, click the “eyeball”
button on the floating
seismic window:

On the seismic view

attributes dialog, select
Cross Section Zone on the
left. Then select Set_1 and
click Ok:

327
 Exercise 6 Cross Plotting AVO Attributes

On the seismic section, the zones selected on the cross plot should look like
this, where the red is top gas, the blue is base gas and the green are shales
and wet sands. The cross plot zones were picked from the seismic zones.

328
 Exercise 6 Cross Plotting AVO Attributes
This completes our initial exercise on the two cross plot options in the
Hampson-Russell suite, which are:

1. Define a single seismic section zone using a parameter menu and then
select cross plot zones on the resulting cross plot.
2. Define multiple seismic section zones interactively and then launch the
cross plot to see each zone in a different color. The cross plot zones
can then be selected from the multiple seismic zones.

Note that both options have advantages, but that the second one is
certainly both more powerful and “fun” to use. When the zone window is
moved, it is almost like looking at the data with a microscope! The main
advantage of the first option is that the seismic zone can be more precisely
selected based on a CDP or inline and cross line range, a time range, or on
previously picked seismic events (e.g. all samples from Horizon 1 to
Horizon 2 between Inlines 5 and 10 and Crosslines 40 and 50, etc.)

Before finishing this exercise we will briefly look at the Scenes option.
329
 Exercise 6 Cross Plotting AVO Attributes

There is a very convenient way to

access any of the displays created so
far. Click on the Scenes tab.

The side tabs indicate which set of displays

we are looking at. For example, click on the
Seismic side tab to see all seismic displays
created in the project.

The check box indicates whether the display

is currently visible in the tab. To turn one on,
click this box.

330
 Exercise 6 Cross Plotting AVO Attributes

We have reached the end of the Workshop

for AVO Day 1. To close down the Geoview
program, click File -> Exit.

Click Yes on the Confirmation box. Note

there is no need to save the project, as it
is constantly being saved.

(End of Exercise 6)
&
End of AVO 1

331
Appendices
Appendix 1: Calculation of VS using Castagna’s Assumption
Appendix 2: Averaging Multiple Minerals
Appendix 3: The Zoeppritz Equations
Appendix 4: The Linearized Approximation
Appendix 5: Foster’s approximation
Appendix 6: HTI anisotropy
Appendix 7: Shuey’s Equation
Appendix 8: Extracting Attributes
Appendix 9: Polarization and the AVO Hodogram
Appendix 10: AVO Case Study: Onshore Texas Example
Appendix 11: AVO Fluid Inversion: Analyzing uncertainty in AVO

332
 Appendix 1: Calculation of VS using Castagna’s
Assumption

1) Calculate density for 100% brine 5) Calculate Vp_wet

saturation:
Mwet
ρwet = ρbr φ + ρm ( 1 - φ) Vpwet 
r wet
2) Calculate input P wave modulus: 6) Calculate Vs_wet from Vp_wet
M = Vp2 ρ Vswet =Ac Vpwet + Bc
3) Calculate matrix P wave
modulus:
4 7) Calculate Vs_input from Vs_wet
Mm  Km   m
3
r wet
Vs  Vswet *
4) Adjust P wave modulus to 100% r
water:
M Kfl Kbr
d  
Mm  M  *( Mm  Kfl )  *( Mm  Kbr )
Mm
Mwet  d *
1 d 333
 Appendix 1: Calculation of VS using Castagna
Assumption

8) Calculate K and m from input data: 10) Calculate Ksat with new fluid:

  Vs * r ;
2
a
Kdry out
 out
Kfl out
Km  Kdry out
 *( Km  Kfl out )
4
K  r *Vp  * 2
a
3 K  Km *
out

1 a
9) Obtain K_dry:
11) Get new density:
K Kfl r out  r fl out * out  r m *(1   out )
a 
Km  K  * ( Km  Kfl )
a 12) Finally – the new velocities!
Kdry  Km *
1 a 4 out
K out    out
Vp out  3 ; Vs out 
r out
r out

334
 Appendix 2

Averaging Multiple Minerals

In this appendix, we will discuss the various ways in

which we can average multiple minerals. Note that
these averages also apply to multiple fluids, etc. The
techniques we will discuss are:
1. Voigt averaging
2. Reuss averaging
3. The Voigt-Reuss-Hill average.
4. The Hashin-Shtrikman Bounds

335
 Appendix 2: Voigt, Ruess and Hill

If we let f1 be the fraction of mineral 1, f2 be the fraction of mineral 2

(where f1 + f2 = 1), M1 be the modulus of mineral 1 (bulk or shear) and
M2 be the modulus of mineral 2, then the Voigt average is the
arithmetic average given by:

M V  f1 M 1  f 2 M 2
The Reuss average is the harmonic average given by:
1 f1 f2 M 1M 2
   MR 
M R M1 M 2 f1 M 2  f 2 M 1
Finally, the Hill average is the average of the Voigt and Reuss
averages:
M H  (M V  M R ) / 2
These averages can be easily extended to N components.
336
 Appendix 2: Hashin-Shtrikman Bounds

The Voigt and Reuss bounds give extreme values. Another approach is to
use Hashin-Shtrikman bounds, which are different for the bulk and shear
modulus components. If mineral 1 is stiffer than mineral 2, then the upper
bound is given by (Mavko et al.):

f2
K HS  K1 
( K 2  K1 ) 1  f1 ( K1  ( 4 / 3) 1 ) 1

f2
HS  1 
2 f1 ( K1  2 1 )
( 2  1 ) 
1

51 ( K1  ( 4 / 3) 1 )

The lower bounds are given by reversing the order of the two minerals in the
equations given above. An example is shown on the next page.

337
 Appendix 2: Comparing the Bounds

The figures above show the effect of Voigt, Reuss and Hashin-Shtrikman
upper and lower bounds for materials with K1 = 60 GPa, K2 = 40 GPa, 1 = 45
GPa, and 2 = 15 GPa. Note that the H-S bounds are between the Voigt and
Reuss bounds. In the software, we use the average of the H-S bounds.
338
Appendix 3: The Zoeppritz Equations

Zoeppritz derived the amplitudes of the reflected and transmitted waves

using the conservation of stress and displacement across the layer
boundary, which gives four equations with four unknowns. Inverting the
matrix form of the Zoeppritz equations gives us the exact amplitudes as a
function of angle:

339
Appendix 3: The Zoeppritz Equations at 0 degrees
Although the Zoeppritz equations look intimidating, in the case of normal
incidence the equations reduce to the following simple form:

1
 0 1 0 1 
 RP (0o )  RP 0     0
  1 0 1 0
o    r 2VS 2VP1  1
R
 S ( 0 )    S 0    0 VP1
R
0   
 TP (0o )   TP 0   VS 1 r1VS 1 
2
 0
 o     r 2VP 2   
 TS (0 )   TS 0   1 0 0  1
 r1VP1 
By performing the above matrix inversion, we will see some interesting
features about the zero angle case.

340
 Appendix 3: The Zoeppritz Equations at 0 degrees

The matrix inversion can be done by hand when there are so many zeros
(but great care must be taken with the signs!), and we get:

 r 2VP 2  r1VP1 
 0 0
r 2VP 2  r1VP1 r 2VP 2  r1VP1 
 RP 0    r 2VS 2  r1VS12  0
 
R   0 0 1
r V 
    2 S 2 1 S1
S 0 r V V r V  r V  


P1 2 S 2 1 S1
 TP 0   r1VP1 r1VP1  0
  
0 0 
r VP 2  r1VP1 r VP 2  r1VP1  1
 S0 
T 2 2 
 r1VS1 r1VS1
2

 0 0 
r  r
 2 S 2 1 S1
V V V r V
P1 2 S 2  r V
1 S1  
The zero angle reflection and transmission coefficients are therefore:

r 2VP 2  r1VP1 2 r1VP1

RS 0  TS 0  0, RP 0  , TP 0   1  RP 0
r 2VP 2  r1VP1 r 2VP 2  r1VP1
341
 Appendix 4: The Linearized Approximation
A useful approximation to RP0 can be derived by noting that:

r 2VP 2  r1VP1 Z P 2  Z P1 Z P Z  ZP2

RP 0=   , where Z P  P1 .
r 2VP 2  r1VP1 Z P 2  Z P1 2 Z P 2

From calculus, we know that:

d ln(Z (t )) 1 dZ (t ) dZ (t )
  d ln(Z (t )) 
dt Z (t ) dt Z (t )
Replacing the derivative d with the difference operator  gives:
 ln Z P  ln VP   ln r 1  VP r 
    
r 
RP 0
2 2 2  VP
Notice that the above equation is the linearized A or RP0 term in the Aki-
Richards equation and its various reformulations.

342
Appendix 5: Deriving the Foster relationship
 To derive the Foster et al. (2010) relationship, first recall the
definitions of A and B:
VP r VP VS r V
A  , and B   8 2  4 2 , where   S .
2V p 2 r 2V p 2VS 2r VP
 VS VP
 Next, note that we can write:  
 VS VP
 To prove the above relationship, we first find from calculus:
d d (VSVP1 ) dVP1 1 dVS VS dVP 1 dVS
  VS   2 
dt dt dt VP dt VP dt VP dt
 Transforming from d to , cancelling the t terms and multiplying
both sides by VP,/VS gives:

 VP   V  V V   VS VP
    P VS 2P  S    
 VS   VS  VP VP   VS VP
343
Appendix 5: Deriving the Foster relationship
 Since A is a function of r and VP, but B is a function of r, VP, VS and
, let us transform B into a function of only r, VP, and , by noting:
VS  VP
 
2VS 2 2VP
 Substitution of this expression into B gives:
VP 
2  VP  r V r
B  8    4 2  1  8 2  P  4  4 2
2V p  2 2V  2r 2V p 2r
 p 

 Adding and subtracting density terms then gives:

VP r  r r r 
B  1  8 2   4  4 2  1  8 2    8 2
2V p 2r  2r 2r 2 r 
 Grouping density and VP terms and simplifying gives the final form:
r
B  (1  8 2 ) A  4  (4 2  1)
2r
344
Appendix 6: HTI anisotropy
In this appendix, we will discuss AVO and HTI anisotropy, and AVAZ
(Amplitude versus Azimuth). Let us first define our geometry. As shown
below, the symmetry-axis plane is at right angles to the fractures and the
isotropy plane is parallel to the fractures.

From Ruger (1998) 345

Appendix 6: Azimuth angle
In addition to the raypath angle q, we now introduce an azimuth angle ,
which is defined with respect to the symmetry-axis plane:

Note that the azimuth angle  is equal to 0 degrees along the symmetry-
axis plane and 90 degrees along the isotropy plane.

From Ruger, Geophysics, May-June 1998

346
 Appendix 6: AVO and HTI
With this definition of azimuth angle, we can derive the following
linearized modeling equation for AVO in HTI media:

Ran (q ,  )  Aiso  ( Biso  Bani cos 2  ) sin 2 q  (Ciso  Cani cos 2  ) sin 2 q tan 2 q ,

where Aiso , Biso , and Ciso are the isotropic AVO terms,
 VS 
2

1
2
(V ) 1

Bani    8    and Cani   (V ) sin 2    (V ) cos 2  
 VP   2
are the AVO HTI anisotropy terms, with :
 (V )  Thomsen' s  parameter defined with respect to vertical,
 (V )  Thomsen' s  parameter defined with respect to vertical,
q  incidence angle, and   azimuth angle.
347
 Appendix 6: Ruger’s B term

In the VTI and HTI AVO expressions given by Ruger

(2002), he rewrites the gradient term B in the AVO
equation in terms of VP and , as shown below:
Standard form of B :

1  VP  
2
VS  VS  
2

VS r 1   VP  VS  2 VS r 

2

B  4   2     4     
2 Vp VP  VS VP  r 2  Vp
 VP   VS r 

Ruger' s form of B :

 2
  2
1

1  V  V   
2 VS r  1   VP  V    2 
B   P  4  S        4  S   2 ln    ln r 
2  Vp VP   VS r  2  V p VP   r 
    

1   VP VS 
2
 
 1   VP VS    
2

   4    ln      4  
2  Vp  P
V  2  pV  P
V  

348
 Appendix 6: AVO and HTI
To show the effects of HTI, Ruger (2002) created the following four models:

Model Vp/Vp A /  (V)  (V) 

A 0.1 0.1 0.2 0 0 0.1

B 0.1 0.1 0.2 -0.1 0 0

C 0.1 0.1 0.2 0 -0.1 0

D 0.1 0.1 0.2 -0.05 -0.05 0.15

 2 VS r
Note :  
 VS r
The results of these four models will be shown on the next two slides.
349
 Appendix 6: Models A and B

The reflection coefficients for The reflection coefficients for Model

Model A (change in ) as a B (change in  ) as a function of
function of incidence angle for 0, incidence angle for 0, 30, 60 and 90
30, 60 and 90 degrees azimuth. degrees azimuth.
350
 Appendix 6: Models C and D

The reflection coefficients for The reflection coefficients for Model

Model C (change in ) as a function D (change in , , and ) as a
of incidence angle for 0, 30, 60 and function of incidence angle for 0,
90 degrees azimuth. 30, 60 and 90 degrees azimuth.
351
 Appendix 6: AVAZ

 The HTI approach just described gives us a way to model amplitude

versus offset and azimuth (AVAZ) effects.
 To observe AVAZ effects, AVO analysis is done on seismic data that
has been binned into different sets of azimuths.
 The first step is then to extract an estimate of Bani using the inverse of
the two-term modeling equation.
 As shown on the next slide, Bani will give us an estimate of fracture
density.
 Next, we can estimate the fracture orientation, as discussed in the
slides following the next slide.
 As shown on the final five slides, AVAZ analysis can thus be quantified
to give us an interpretation over a fractured reservoir, both in map and
cross-section view.

352
 Appendix 6: Fracture density

0.1
gas
0.09
As shown in this hudson wet
Gassmann wet
figure, using 0.08
several different 0.07
rock physics
0.06
modeling schemes,
Bani
the value of Bani is a 0.05
good indicator of 0.04
the crack, or
0.03
fracture, density in
a fractured 0.02
reservoir. 0.01

0
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.1
crack density

353
 Appendix 6: Fracture orientation
In our modeling slides, we
assumed that the direction of
the fractures was known.
However, this is often
unknown, and needs to be
determined. Let us first define
sym to be the azimuth angle sym
along the symmetry-axis plane,
and iso to be the azimuth angle
along the isotropy plane, as
shown on the right:

We can then write the near offset HTI AVO equation as either:
Ran (q ,  )  Aiso  [ Biso  Bani cos 2 (  sym )] sin 2 q
or, since sym is orthogonal to iso, as:

Ran (q ,  )  Aiso  [ Biso  Bani sin 2 (  iso )] sin 2 q

354
 Appendix 6: Fracture orientation
Regardless of which convention we choose, we can then plot the reflectivity
as a function of azimuth, as shown below, and determine the symmetry-plane
and isotropy-plane angles from the minimum and maximum values of the
curve. AVAZ Effect

The interpretation of this plot will 2500.00

depend on the sign of Bani.

Assuming Bani is positive, we find 2000.00

that sym = 130o and iso = 40o.

Relative Amplitude
Note that iso gives us the
de 1500.00

fracture orientation.
Amplitu 1000.00

500.00
However, Bani can also be
negative. Later is the appendix 0.00

350
330
we see how this leads to a 90

310
290
270
250
230
degree ambiguity in the

210
5

190
170
20

150
symmetry axis.
130
Azimuth

110
Angle 35
o
90

130
70
50
30

40o
10

355
 Appendix 6: Variations in AVAZ

Amplitude
This figure shows
offset gathers at two
different azimuths
over a fractured
reservoir. As seen
in the next two
slides, 1 = iso and
2 = sym.

Notice the dramatic

change in the AVO
responses.

1 2
Courtesy: Dave Gray, CGGVeritas
356
Appendix 6: AVO parallel to fractures

source

q receiver

iso
fractured medium

azimuth= 1 =iso
Courtesy: Dave Gray, CGGVeritas
357
 Appendix 6: AVO across fractures

receiver

source

isoiso   iso=90o fractured

medium

Azimuth= 2 = iso+90 Courtesy: Dave Gray, CGGVeritas

358
 Appendix 6: Fracture Interpretation

Using the previous ideas,

AVO Fracture Analysis Orientation
measures fracture volume of Fault
from differences in AVO
response with Azimuth.
Fracture strike is
determined where this
difference is a maximum.
Oil Well

Edge
Direction of Line is Effects
estimated fault strike,
length of line and color Fractures curling
is estimated crack into the fault
Fractures abutting
density Interpreted Faults the fault

Courtesy: Dave Gray, CGGVeritas359

Appendix 5: Outcrop compared to AVAZ

Base of
Dunes Fracture Strike

Fractures

Photo courtesy of Golder Associates

NW-SE E-W
360
Appendix 6: Linearized near offset Ruger equation

The near offset Rüger equation may be solved by linear squares by

reparameterizing the problem
Ran (q ,  )  Aiso  [ Biso  Bani sin 2 (  iso )] sin 2 q
1
B  Biso  Bani
2
2
Bani  C 2  D2

tan 2iso  D / C
Ran (q ,  )  A  [ B  C cos 2  D sin 2 ] sin 2 q

Can express this as linear problem Gm  d

which may be solved by least squares 
m  G G GT d
T
1

361
Appendix 6: Fracture orientation
The near offset Ruger equation
Ran (q ,  )  Aiso  [ Biso  Bani sin 2 (  iso )] sin 2 q
is nonlinear and multi-modal. Two sets of parameters fit the data equally
well. The nonlinear inversion solves for the magnitude Bani and azimuth iso.

The magnitude may be either positive or negative. From a rock physics

perspective the expectation is that Bani should generally be positive so the
typical convention is to chose Bani to be positive.

However, Bani is actually an interface parameter so it changes sign whether

the wave is moving from an isotropic to an anisotropic layer or vice versa.

If Bani is negative this introduces a 90 degree phase rotation to the azimuth

estimate compared to a positive Bani

362
 Appendix 6: Fracture orientation
For a particular angle of incidence the
Azimuthal Reflectivity is Azimuthal reflectivity for a
• circular for an isotropic media
constant angle of incidence
• elliptical for an HTI media
0.04
•the anisotropic gradient specifies the perturbation from the
isotropic (circular) solution

There are two possible

0.03
+Bani
parameterizations which describe
0.02 -Bani
the elliptical reflectivity 0.01 iso
iso
• Solution 1 (positive Bani): 0

•isotropic media is characterized by the red circle

-0.01
•Bani is positive forming Blue ellipse
•Isotropy-plane azimuth is defined by red line segment -0.02

• Solution 2: (negative Bani): -0.03

•isotropic media is characterized by the black circle
•Bani is negative forming Blue ellipse -0.04
-0.04 -0.03 -0.02 -0.01 0 0.01 0.02 0.03 0.04
•Isotropy-plane azimuth is defined by black line segment

363
 Appendix 6: Fracture orientation

Fsym (from near offset Rüger) Fsym (nonlinear inversion)

180° 180°

0° 0°

• Downton (2011) shows that is possible to resolve the azimuth ambiguity by

performing a nonlinear inversion on multiple azimuthal Fourier coefficients
• The Fracture strike is consistent with maximum horizontal stress (135 degrees) in
the area
364
 Appendix 7: Shuey’s Equation

Shuey (1985) rewrote the ABC equation using VP, r, and . Only the gradient
is different than in the ABC expression:

 1  2  
B  A D  2(1  D)   ,
 1    (1   ) 2

VP / VP   1
where : D  ,  2 , and    2   1.
VP / VP  r / r 2
The above equation is quite complicated but can be greatly simplified by
assuming that  = 1/3 (the same as Vp/Vs=2). This gives:

 1  9
B  A D  2(1  D )    2.25 Δσ  A
 2 4
This leads to a very intuitive version of the two-term AVO equation:

RP (q )  A  ( 2.25 Δσ  A) sin 2 q

365
 Appendix 7: Shuey’s Equation

Gas Sand Model

Aki-Richards vs Shuey
This figure shows a
comparison between the 0.250
two forms of the 0.200
Aki-Richards equation for 0.150
the gas sand considered 0.100
earlier.
Amplitude
0.050
0.000
Note that the values are -0.050
close but, unlike the -0.100
previous three forms of the -0.150
equation, Shuey’s version
-0.200
does not give exactly the
-0.250
same values. 0 5 10 15 20 25 30 35 40 45
Angle (degrees)

A-R Top Shuey Top

A-R Base Shuey Base
366
 Appendix 7: Hilterman’s Approximation

Hilterman re-arranges Shuey’s equation in a slightly different way:

R(q )  A  2.25  Asin 2 q

 A (1  sin 2 q )  2.25 sin 2 q
 A cos 2 q  2.25 sin 2 q
Notice that this equation is very intuitive, since it shows that, as the
angle increases, so does the dependence on . Keep in mind that this
equation is strictly correct only for  = 1/3 and that the C term has been
dropped. Note also that another way of writing this equation is as
follows, which shows the dependence on A and B:

R(q )  A cos 2 q  ( A  B) sin 2 q

367
Appendix 8: Extracting Attributes

In the course we have often discussed the need to extract attributes from
the pre-stack seismic gathers. To see how this is done, note that all the
linearized equations we have looked at so far can be written as:

RP (q )  f1 p1  f 2 p2  f 3 p3 ,
where f1 , f 2 , and f 3 are functions of q and sometimes VS2 / VP2 ,
and p1, p2 , and p3 are functions of VP ,VS , and r .

For example, the ABC equation is:

RP (q )  A  B sin 2 q  C tan 2 q sin 2 q , where : f1  1, f 2  sin 2 q , f 3  tan 2 q sin 2 q ,

2 2
 VP r  VP V   VS V  r  VP
p1  A   , p2  B   4 S   2 S  , p3  C  .
2V p 2 r 2V p VP  VS VP  r 2V p

368
 Appendix 8: Extracting Attributes

For N traces, where we know the angles, we can write:

RP (q1 )  f1 (q1 ) p1  f 2 (q1 ) p2  f 3 (q1 ) p3

RP (q 2 )  f1 (q 2 ) p1  f 2 (q 2 ) p2  f 3 (q 2 ) p3
   
RP (q N )  f1 (q N ) p1  f 2 (q N ) p2  f 3 (q N ) p3
This can be written in matrix form as:

 RP (q1 )   f1 (q1 ) f 2 (q1 ) f 3 (q1 ) 

 R (q )   f (q )   p1 
f 2 (q 2 ) f 3 (q 2 )  
 P 2  1 2  p2
        
     p2 
 P N   f1 (q N )
R ( q ) f 2 (q 2 ) f 3 (q N )
369
 Appendix 8: Extracting Attributes

The previous equation can be written more simply as:

R  MP,

where R is a known vector of N picked reflection coefficients at a constant

time, M is an N x 3 vector of computed values, and P is the unknown vector
containing the parameters to be estimated.

This is an over-constrained problem which has the following solution:

 p1  1 0 0
P   p2   ( M T M   I ) 1 M T R, where I  0 1 0,
   
 p3  0 0 1
and  is a pre - whitening factor.

370
 Appendix 8: Extracting ABC Attributes
Let us take the specific case of extracting ABC attributes, for which the
forward problem is:

 RP (q1 )  1 sin 2 q1 tan 2 q1 sin 2 q1 

 R (q )     A
 P 2  1 sin 2
q2 tan q 2 sin q 2   
2 2
B
       
    C 
 RP (q N )  1 sin 2
qN tan q N sin q N 
2 2

Since we know that sin qi  X iVINT / tVRMS

2
, we can write :

 RP (q1 )  1 X V / tVRMS 
2 2
X V / tV 2
2
/ 
 X V / tV 2
2

1 
 R (q )  
 
  A
1 INT 1 INT RMS 1 INT RMS

 P 2   1 X V
2 INT / tV 2
RMS 2
X V
2 INT / tVRMS  / X 2VINT / tVRMS   1   
2 2 2 2

B
        
  
 RP (q N ) 1 X NVINT / tVRMS 
2 2 2 2

X NVINT / tVRMS  / X NVINT / tVRMS   1 
2 2

 C 

371
 Appendix 8: Extracting ABC Attributes

Simplifying the notation we get the following solution:

1
 1 b1 c1    RP1 
 A   1 1  1    1 1  1 
 B    b b  b  1 b2 c2   b b

 bN  
R
 P 2 
   1 2 N
    1 2   
C   c c  cN   c1 c2  cN  
  1 2  
1 bN cN    PN 
R
1
 N N
  N

 N

b 
i 1
i
i 1
ci 



 i 1
RPi 

 N N N
  N

    bi RPi , bi  X iVINT / tVRMS  , ci  1
2 bi
 bi bi2 bi ci   2

 i 1 i 1 i 1   i 1  bi  1
 N N N   N 
  b c 
 i 1
ci
i 1
i i
i 1
ci 
2





i 1
ci RPi 


372
 Appendix 8: Extracting RP0, RS0 and RD Attributes

Next, let us take the case of extracting RP0, RS0 and RD attributes, for which
the forward problem is:

 RP (q1 )   d (q1 ) e(q1 ) f (q1 ) 

 R (q )   d (q ) e(q )   RP 0 
f (q 2 )  
 P 2  2 2  RS 0 , where :
        
     RD 
 P N  d (q N ) e(q N )
R ( q ) f (q N )
VS2 2 VS2 2 1
d (qi )  1  tan qi , e(qi )  8 2 sin qi , f (qi )  2 2 sin qi  tan 2 qi ,
2

VP VP 2
1   VP r  1   VS r  r X iVINT
RP 0     , RS 0     , RD  , and sin qi  .
2  VP r  2  VS r  r 2
tVRMS

373
 Appendix 8: Extracting RP0, RS0 and RD Attributes

Simplifying the notation we get the following solution:

1
  d1 e1 f1    RP1 
 RP 0   d1 d 2  d N     d1 d 2  d N 
 R    e e  e   d 2 f2  e e

 eN  
e2 R
 P 2 
 S 0   1 2 N 
      1 2   
 RD    f f 2  cN    f1 f 2  f N  
  1  
d N eN f N    PN 
R
1
 N N N
  N



d d e 
i 1
i
2

i 1
i i
i 1
di fi 




i 1
d i RPi 

 N N N
  N




i 1
d i ei 
i 1
ei2 
i 1
ei f i 




i 1
ei RPi 

 N N N   N 


d f e f 
i 1
i i
i 1
i i
i 1
fi 
2





i 1
f i RPi 


374
Appendix 9: Polarization and the AVO Hodogram

 We have seen that one of the major causes of “scatter” on

seismically derived cross-plots is wavelet interference.

 One approach to reducing wavelet interference is to use the

AVO hodogram (Keho et al: The AVO hodogram: Using
polarization to identify anomalies, TLE, November, 2001 and
Mahob and Castagna: AVO hodograms and polarization
attributes, TLE, January, 2002).

 A second approach is to perform inversion of the AVO

attribute volumes. We will see this in a later section.

375
Appendix 9: Polarization and the AVO Hodogram

Up to now, we have
calculated cross plots of A
and B, using fairly large
analysis windows.

This gives a broad cluster of

points, containing both the
background trend and the AVO
anomalies:
376
 Appendix 9: Polarization and the AVO Hodogram

In Polarization Analysis, we calculate

cross plots over small sliding windows
on a single trace.

377
 Appendix 9: Polarization and the AVO Hodogram

For each cross plot, we

can calculate the
polarization vector.

This measures the q

direction of the
dominant energy for
this cluster.

The length of the

vector measures the
average energy in the
cluster.

378
 Appendix 9: Polarization and the AVO Hodogram

Theoretically, we can expect wet trend

-45o
points to fall around the -45o trend,
while class 3 AVO anomalies fall
around +45o.

+45o

379
 Appendix 9: Polarization and the AVO Hodogram

time

One way to display this result is to plot the calculated polarization vector on
a 3-D display with time as the third axis. This is called a Hodogram.

380
Appendix 9: Polarization and the AVO Hodogram

A more conventional display shows the

calculated polarization angle for a single
trace as a function of time.

Note that this result depends on the size of

the sliding window.

Also note the anomalous positive

Polarization Angle around 630 ms
indicating the Class 3 anomaly.

381
Appendix 9: Polarization and the AVO Hodogram
In addition to the Polarization Angle itself, a very useful attribute is the
Polarization Product, which is Polarization Angle multiplied by the length of
the Polarization Vector. This is expected to highlight bright spots which
have high hydrocarbon potential:

382
Appendix 10:
AVO Case Study
Onshore Texas Example
 AVO Case Study, Onshore Texas Example

This case study comes from a paper by Mark Gregg and Charles Bukowski
(Leading Edge, November, 2000).

This paper shows a very practical example of the application of AVO to a

mature basin.

384
AVO Case Study, Onshore Texas Example

The exploration objective was

the clastic Oligocene Vicksburg
formation in South Texas.

This has produced more than 3

trillion ft3 of gas since the
1920’s, but not much AVO work
has been reported.

The authors believe the lack of

AVO application comes because
“the Vicksburg trend is not a
typical amplitude-supported
play”.

385
AVO Case Study, Onshore Texas Example

The motivation for using AVO

came from results like those
shown on the left.

Using the conventional post-

stack data, it is difficult to
distinguish Gas from Wet sand
before drilling.

Prior to AVO analysis, the

authors had drilled one
commercial gas well, one non-
commercial gas well, and three
dry holes.

386
AVO Case Study, Onshore Texas Example

These curves from the gas

discovery well show both a
Gas and a Wet zone.

The change in acoustic

impedance is small but the
change in Poisson’s ratio is
large.

This suggests a class 2

AVO anomaly.

387
AVO Case Study, Onshore Texas Example

Synthetic modeling
confirmed the expected
class 2 response.

388
AVO Case Study, Onshore Texas Example

The data was

reprocessed to include
nonhyperbolic
moveout. This turned
out to be critical, as the
figure shows.

389
AVO Case Study, Onshore Texas Example

A very useful indicator is the

Near and Far Angle Stack.

Note that the Gas sand shows

its brightest response on the
Far Angle stack, as expected
for the class 2 behavior.

The authors used the Far

Angle Stack as the main tool
for searching for new
anomalies.

390
AVO Case Study, Onshore Texas Example

The authors studied the existing wells and came to these conclusions:

(1) There were about 100 gas wells in the area with cumulative production
> 1 billion ft3.
(2) About ½ of these were associated with class 2 AVO anomalies.
(3) About 65% of the ~70 drilled anomalies were commercial gas
accumulations.
(4) Thicker, better-developed reservoirs produced the most distinctive
anomalies.
(5) Threshold gross reservoir thickness required to produce an anomaly
was about 30-60ft.
(6) Most productive anomalies were at depths of 5,000-10,000 ft.

391
AVO Case Study, Onshore Texas Example

This is the first drilled

anomaly.

100 ft gross interval

with 72 ft of net pay,
producing initially 3
million ft3 of gas per
day.

Since the anomaly is

not visible on the
conventional stack, this
would not have been
drilled without the AVO
analysis.

392
AVO Case Study, Onshore Texas Example

A second anomaly was

identified by interpreting the
far-angle stack using
Landmark’s Earthcube
software.

This had not been identified

before AVO, because of the
poor quality of the
conventional stack. This was
presumed to be because of
the small acoustic impedance
contrast.

Note that there appear to be

multiple anomalies at the
prospective level.

393
AVO Case Study, Onshore Texas Example

The drilled well

encountered 2 pay
zones.

Upper zone: gross

thickness of 54 ft,
with 28 ft net pay.

Lower zone: gross

thickness of 214 ft
with 69 ft net pay.

Initial production
rate was 5.3 million
ft3 with estimated
ultimate recovery of
14 billion ft3.

394
AVO Case Study, Onshore Texas Example

Two more
successful wells
are shown here.

395
 AVO Case Study, Onshore Texas Example

This is an unsuccessful result. The drilling encountered 105 ft of clean,

low-gas-saturated sand at the anomaly.

396
AVO Case Study, Onshore Texas Example

Results:

(1) Six commercial discoveries.

(2) Two dry holes, caused by low gas saturation.
(3) This is a 75% success rate, dramatically improved from the original
20% success rate.

Authors’ conclusions:

(1) Know your rocks. Do the modeling.

(2) Look beyond conventional seismic techniques, e.g. AVO.
(3) Low gas saturation remains a pitfall of the AVO method.

397
Appendix 11:

AVO Fluid Inversion : Analyzing

uncertainty in AVO
Overview

AVO Analysis is now routinely used for exploration and

development.

But: all AVO attributes contain a great deal of “uncertainty” –

there is a wide range of lithologies which could account for
any AVO response.

In this talk we present a procedure for analyzing and

quantifying AVO uncertainty.

As a result, we will calculate probability maps for

hydrocarbon detection.

399
AVO Uncertainty Analysis: The Basic Process

G STOCHASTIC
AVO
CALIBRATED: MODEL
I
 GRADIENT
 INTERCEPT FLUID
 BURIAL DEPTH PROBABILITY
MAPS
AVO ATTRIBUTE
MAPS  PBRI
ISOCHRON
MAPS  POIL
 PGAS

400
“Conventional” AVO Modeling:
Creating 2 Pre-Stack Synthetics
IN SITU = OIL

IO GO

FRM = BRINE

IB GB

401
Monte Carlo Simulation:
Creating Many Synthetics

I-G DENSITY FUNCTIONS

BRINE OIL GAS

75

50

25

402
The Basic Model

We assume a 3-layer
Shale model with shale
enclosing a sand (with
various fluids).
Sand

Shale

403
 The Shales are
Vp1, Vs1, ρ1 characterized by:

P-wave velocity
S-wave velocity
Density
Vp2, Vs2, ρ2

404
 Each parameter has a
Vp1, Vs1, ρ1 probability distribution:

Vp2, Vs2, ρ2

405
 The Sand is characterized by:

Brine Modulus
Brine Density
Shale Gas Modulus
Gas Density
Oil Modulus
Sand Oil Density
Matrix Modulus
Matrix density
Shale Porosity
Shale Volume
Water Saturation
Thickness

Each of these has a probability distribution.

406
Trend Analysis
Some of the statistical distributions are determined
from well log trend analyses:
5000
4500
4000
3500
3000
2500
2000
1500
1000
500
0
0.4 0.9 1.4 1.9 2.4 2.9 3.4
DBSB (Km)
407
Determining Distributions at Selected Locations
Assume a Normal distribution. Get the Mean and Standard
Deviation from the trend curves for each depth:

5000
4500
4000
3500
3000
2500
2000
1500
1000
500
0
0.4 0.9 1.4 1.9 2.4 2.9 3.4
DBSB (Km)
408
 Trend Analysis: Other Distributions
5000
Shale Velocity
4500
3.0
4000 Sand Density
3500 2.8
3000 2.6 3.0 Shale Density
2.8
2500 2.4 40%
2.6 Sand Porosity
2000 2.2
2.4 35%
1500 2.0
2.2 30%
1000 1.8
2.0 25%
500 1.6
1.8
0 1.4 20%
0.41.2
1.6 0.9 1.4 1.9 2.4 2.9 3.4
15%
1.4 DBSB (Km)
1.0 10%
1.2
0.4 0.9 1.4 1.9 2.4 2.9 3.4
1.0 5%
DBSB (Km)
0.4 0.9 1.4 1.9 2.4 2.9 3.4
0% DBSB (Km)
0.4 0.9 1.4 1.9 2.4 2.9 3.4
DBSB (Km) 409
Practically, this is how we set up the distributions:
Shale:
Vp Trend Analysis
Vs Castagna’s Relationship with % error
Density Trend Analysis

Sand:
Brine Modulus
Brine Density
Gas Modulus
Gas Density
Oil Modulus Constants for the area
Oil Density
Matrix Modulus
Matrix density
Dry Rock Modulus Calculated from sand trend analysis
Porosity Trend Analysis
Shale Volume Uniform Distribution from petrophysics
Water Saturation Uniform Distribution from petrophysics
Thickness Uniform Distribution
410
 Calculating a Single Model Response

From a particular model instance, Note that a wavelet is

calculate two synthetic traces at assumed known.
different angles. 0o 45o

Top Shale

Sand

Base Shale

411
 Note that these amplitudes include
On the synthetic traces, pick the
interference from the second interface.
event corresponding to the top of o o
the sand layer: 0 45

Top Shale
P2
P1
Sand

Base Shale

412
Using these picks, calculate the Intercept and Gradient for this
model:
0o 45o
I = P1
G = (P2-P1)/sin2(45)

Top Shale P2
P1

Sand

Base Shale

413
 Using Biot-Gassmann Substitution

Starting from the Brine Sand case, the corresponding Oil and Gas Sand models are
generated using Biot-Gassmann substitution. This creates 3 points on the I-G cross plot:
BRINE
GAS OIL

KGAS KOIL
rGAS rOIL

G G G
I I I

414
Monte-Carlo Analysis
By repeating this process many times, we get a probability distribution for
each of the 3 sand fluids:

Brine
I Oil
Gas

415
The distributions are depth-dependent

416
The Depth-dependence can often be understood using
Rutherford-Williams classification
2 4 6

5
3
1

Sand
Impedance

4
3
Shale
2
5 6
1

Class 1
Class 2

Class 3 Burial Depth 417

 Bayes’ Theorem

Bayes’ Theorem is used to calculate the probability that any new (I,G) point
belongs to each of the classes (brine, oil, gas):


~
P F I,G  
 ~
 ~
p I , G F * P( F )
 k
p I , G Fk * PFk 

where:
P(Fk) represent a priori probabilities and Fk is either brine, oil, gas;
p(I,G|Fk) are suitable distribution densities (eg. Gaussian) estimated
from the stochastic simulation output.

418
Example Probability Calculations

Gas Oil Brine

419
 Real Data Calibration

In order to apply Bayes’ Theorem to (I,G) points from a real seismic data set,
we need to “calibrate” the real data points.

This means that we need to determine a scaling from the real data amplitudes
to the model amplitudes.

We define two scalers, Sglobal and Sgradient, this way:

Iscaled = Sglobal *Ireal

Gscaled = Sglobal * Sgradient * Greal

One way to determine these scalers is by manually fitting multiple

known regions to the model data.

420
Fitting 6 Known Zones to the Model

4 5 4 5
6 6

3 1 3
1

2 2

1 2 3

4 5 6

421
Real Data Example – West Africa

This example shows a real project from West Africa,

performed by one of the authors (Cardamone).

There are 7 productive oil wells which produce from a

shallow formation.

The seismic data consists of 2 common angle stacks.

The object is to perform Monte Carlo analysis using

trends from the productive wells, calibrate to the known
data points, and evaluate potential drilling locations on a
second deeper formation.

422
One Line from the 3D Volume

Near Angle Stack

0-20 degrees

Far Angle Stack

20-40 degrees

423
Near Angle Stack
0-20 degrees

Shallow producing zone

Deeper target zone

Far Angle Stack

20-40 degrees

424
AVO Anomaly

Near Angle Stack

0-20 degrees

Far Angle Stack

20-40 degrees

425
Amplitude Slices Extracted from Shallow Producing Zone

Near Angle Stack

0-20 degrees

+189

-3500

Far Angle Stack

20-40 degrees

426
 Trend Analysis : Sand and Shale Trends
5000
3.00

4500
2.75
4000
Sand velocity Sand density

DENSITY
VELOCITY

2.50
3500

3000
2.25

2500
2.00
2000

1.75
1500

1000 1.50
500 700 900 1100 1300 1500 1700 1900 500 700 900 1100 1300 1500 1700 1900

4000
3.00

3500 Shale velocity 2.75 Shale density

VELOCITY

DENSITY
3000
2.50

2500
2.25

2000
2.00

1500 1.75

1000 1.50
500 700 900 1100 1300 1500
BURIAL DEPTH (m)
1700 1900 2100 2300 2500 500 700
BURIAL DEPTH (m)
900 1100 1300 1500 1700 1900

427
Monte Carlo Simulations at 6 Burial Depths

-1400 -1600 -1800

-2000 -2200 -2400

428
Near Angle Amplitude Map Showing Defined Zones

Wet Zone 1
Well 6

Well 3 Well 5
Well 7 Well 1

Well 2

Well 4

Wet Zone 2

429
Calibration Results at Defined Locations

Wet Zone 1 Well 2

Wet Zone 2 Well 5

430
Well 3 Well 6

Well 4 Well 1

431
Using Bayes’ Theorem at Producing Zone: OIL

Near Angle Amplitudes

1.0

.80
Probability of Oil
.60

.30
432
Using Bayes’ Theorem at Producing Zone: GAS

Near Angle Amplitudes

1.0

.80
Probability of Gas
.60

.30
433
Using Bayes’ Theorem at Target Horizon

Near angle amplitudes of second event

1.0

Probability of oil on second event .80

.60

.30
434
Verifying Selected Locations at Target Horizon

435
Summary

By representing lithologic parameters as probability distributions we

can calculate the range of expected AVO responses.

This allows us to investigate the uncertainty in AVO predictions.

Using Bayes’ theorem we can produce probability maps for different

potential pore fluids.

But: The results depend critically on calibration between the real

and model data.

And: The calculated probabilities depend on the reliability of all the

underlying probability distributions.

436