Top Banner
8/10/2019 Avo_talk http://slidepdf.com/reader/full/avotalk 1/44  New Trends in AVO Brian Russell and Dan Hampson Hampson-Russell Software Calgary, Alberta.
44

Avo_talk

Jun 02, 2018

Download

Documents

Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 1/44

 New Trends in AVO

Brian Russell and Dan Hampson

Hampson-Russell Software

Calgary, Alberta.

Page 2: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 2/44

Outline of Talk

Review of AVO principles

 AVO attributes

 AVO cross-plotting

3D AVO

 AVO and Anisotropy

Page 3: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 3/44

Summary of AVO Methodology

Input Raw Gathers

Optimum Processing

Recon Methods InversionModelling

Gradient/

Intercept

Partial

StacksPrimaries

only

Wave

Equation

Page 4: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 4/44

AVO Example

• We will illustrate AVO with a Cretaceous

gas sand example from Alberta.

• Traditionally, wells were drilled in this

area based on “bright-spot” anomalies.

• Many dry holes were encountered due to

false “bright-spots” caused by coals. 

• Drilling success was been enhanced

through the use of AVO.

Page 5: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 5/44

Basic AVO Analysis

• We will start our AVO analysis by looking

at some simple displays of the gas sand

example:

• The CMP stack

• Near and far trace stacks

• The common offset stack

• Amplitude envelope displays

Page 6: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 6/44

The full stack shows a bright spot at 640 ms.

600- 

700-

Time

(ms)

Page 7: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 7/44

Note increase in amplitude from (a) Near to (b) Far trace

stack. 

(b)

(a)

Page 8: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 8/44

(a) Near and (b) far trace stacks with color

envelope 

(a)

(b)

Page 9: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 9/44

Input gathers showing an amplitude increase

with offset. 

Page 10: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 10/44Gathers with color amplitude envelope

Page 11: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 11/44

More Advanced AVO Analysis

• We will continue our AVO analysis by

looking at the picked top and base of the

common offset stack of the gas sand

example. This will lead to severalconclusions:

• The amplitudes change as a function of

offset or angle.

• These changes can be quantified using

the Zoeppritz or Aki-Richards equations.

Page 12: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 12/44

Picking the common offset stack

(a) Common offset stack 

(b) Picks

from the

trough.

(c) Picks

from the

peak.

Page 13: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 13/44

Reflected

P-wave = R(q

Reflected

S-wave

Transmitted

P-wave

Incident

P-wave

Transmitted

S-wave

Mode Conversion of an Incident P-wave

VP1 , VS1 ,r

VP2 , VS2 , r2

If q > 0o, incident P-waves produce P and S reflections

and transmissions.

Page 14: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 14/44

The Aki-Richards Approximation

• Using the linearized approximationand keeping only second order terms:

R(q) = RP + G sin2q 

where: RP=1/2(DVP/VP+Dr/r) 

= zero-offset P-wave refl.coeff.

and: G = gradient.

Page 15: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 15/44

Common Offset Picks as function of sin2q 

+RP 

-RP

+G 

- G

Offset 

sin2q 

Time

(a) Small part of commonoffset stack.

(b) Peak/trough picks vs

sin2q 

Page 16: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 16/44

Wiggens’ Approximation 

•  Assuming that VP/VS = 2, in Aki-Richards eq:

G = RP - 2*RS 

where: RS = 1/2(DVS/VS+Dr/r)

= zero-offset S-wave refl. coeff.

• This can be rewritten:

RS = (RP - G) / 2

Page 17: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 17/44

Shuey’s Approximation 

•  Assuming thats

av= 1/3, we get theapproximation:

G = 9/4Ds

 - RP 

where: Ds= Change in Poisson’s Ratio 

• This can be rewritten:

Ds

= (RP + G)*4/9

Page 18: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 18/44(a) Intercept (P-wave) and (b) Gradient Stacks

(a)

(b)

Page 19: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 19/44(a) (P + G) and (b) Rs (P - G) StacksDs  

(a)

(b)

Page 20: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 20/44

AVO Modeling and Inversion 

•Finally, AVO effects can be quantified using

modeling and inversion:

• Modeling involves building a blocked log

model and then creating a synthetic byray-tracing and Zoeppritz amplitude

calculation.

• Inversion involves updating the model to

create a better fit between synthetic and

observed common offset stack.

Page 21: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 21/44

Modelling / Inversion Flow

Input Well

Logs

Input CDP

Gathers

Forward

Model

Create

Coffstack

Difference

Update

Model

Finish

Good

Fit?

No

Yes

Page 22: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 22/44

Well Logs and Synthetic/Seismic Tie

Page 23: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 23/44

Page 24: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 24/44

Well Logs and Synthetic After Inversion

Black = Before Red = After

Page 25: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 25/44

Data Comparison after Inversion

(a) Synthetic (b) Real Coffstack

Page 26: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 26/44

AVO Cross-plotting 

 AVO cross-plotting involves plotting theintercept against the gradient and identifying

anomalies. The theory of cross-plotting was

developed by Castagna el al (TLE, 1997,Geophysics, 1998) and Verm and Hilterman

(TLE, 1995) and is based on two ideas:

(1) The Mudrock line

(2) The Rutherford/Williams

classification scheme.

Page 27: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 27/44

The Mudrock Line 

The mudrock line is a linear relationshipbetween VP and VS derived by Castagna et al

(1985):

VP

 = 1.16 VS

 + 1360 m/sec

Smith and Gidlow (1987) derived the “Fluid

Factor” by combining the mudrock line with

 Aki-Richards:DF = RP - 1.16 (VP/VS) RS 

Page 28: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 28/44

 ARCO’s original mudrock derivation

(Castagna et al, Geophysics, 1985.)

Page 29: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 29/44

Rutherford/Williams Classification 

Rutherford and Williams (1989) derived the

following classification scheme for AVOanomalies, with further modifications by Ross

and Kinman (1995) and Castagna (1997):

Class 1: High acoustic impedance contrast

Class 2: Near-zero impedance contrast

Class 2p: Same as 2, with polarity change

Class 3: Low impedance contrast sands

Class 4: Very low impedance contrast

Page 30: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 30/44

The Rutherford and Williams classification

scheme as modified by Ross and Kinman.

Page 31: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 31/44

Theory of Cross-plotting 

Castagna and Swan (1988) start by assuming

both the mudrock line and Gardner’sequation:

r = a VP1/4 

They then show that the linear relationship

can be written:

G = RP [4/5 -32/5c(VS/VP)-1/2(VS/VP)2]

Page 32: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 32/44

Mudrock lines on a crossplot for various

Vp/Vs ratios (Castagna and Swan, 1998)

Page 33: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 33/44

Intercept / Gradient Crossplots

(b) Interpreted gas

zone

(a) Uninterpreted gas

zone

Page 34: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 34/44

Seismic Display from Int/Grad Xplots

(a) Before interpretation

(b) After interpretation

Page 35: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 35/44

3D AVO

3D AVO is an simply an extension of 2D

 AVO using gradient/intercept analysis.

Using 3D allows us to map spatial

variations in AVO effects. We must be careful to get good offset

coverage in the 3D design stage.

It may be possible to detect azimuthal

anisotropy by restricting azimuths in the

attribute calculation.

Lines from a 3D Channel Sand

Page 36: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 36/44

Lines from a 3D Channel Sand

Example 

(a) Inline 10, channel

at Xline 9, 650 msec. 

(b) Inline 20, channel

at Xline 24, 650 msec.

Page 37: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 37/44

Map view of seismic amplitude

from 3D channel sand. 

Pseudo Poisson’s ratio over 3D channel

Page 38: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 38/44

Pseudo-Poisson’s ratio over 3D channel

sand

(a) Inline 10, channel

at xline 9, 650 msec. 

(b) Inline 20,

channel

at xline 24, 650

Page 39: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 39/44

Map view of pseudo-Poisson’s

Ratio over channel sand. 

Page 40: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 40/44

AVO and Anisotropy

Two types of anisotropy most common:

Transverse isotropy - caused by horizontal

layering

 Azimuthal anisotropy - caused by fractures

Transverse isotropy  can be modelled using

Thomsen parameters.

 Azimuthal anisotropy  may be observed by

restricting azimuths when performing

intercept/gradient analysis.

Page 41: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 41/44

Transverse Isotropy 

Blangy (Geophysics, 1997) showed that atransversely isotropic term could be added

to the Aki-Richards’ equation using the

Thomsen weak anisotropic parameters d and e :

Ran(q) = Ris(q) + Dd/2 sin2

(q)- 1/2(Dd - De) sin2(q)tan2(q) 

Page 42: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 42/44

Transverse Isotropy - Gas Case

Note that the effect of Dd and De is to increase the

AVO effects. (Blangy, 1997)

Page 43: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 43/44

Transverse Isotropy - Wet Case

Note that the effect of Dd and De is to create

apparent AVO decreases. (Blangy, 1997) 

Page 44: Avo_talk

8/10/2019 Avo_talk

http://slidepdf.com/reader/full/avotalk 44/44

CONCLUSIONS

This talk was intended to give an

overview of the AVO method. The various techniques used in AVO

were illustrated using a gas sand.

Traditional AVO methods consist ofcomputing intercept/gradient attributes.

Newer techniques include:

  - cross-plotting of attributes

  - extension to 3D

  - analysis of anisotropic effects.