TARGET-ORIENTED LEAST SQUARES MIGRATION - PowerPoint PPT Presentation

1 / 19
About This Presentation
Title:

TARGET-ORIENTED LEAST SQUARES MIGRATION

Description:

... iterations,where each iteration requires several forward modeling and several migration steps ... in suppressing the migration artifacts, but expensive ... – PowerPoint PPT presentation

Number of Views:46
Avg rating:3.0/5.0
Slides: 20
Provided by: hsun
Category:

less

Transcript and Presenter's Notes

Title: TARGET-ORIENTED LEAST SQUARES MIGRATION


1
TARGET-ORIENTED LEAST SQUARES MIGRATION
Zhiyong Jiang
Geology and Geophysics Department University of
Utah
2
Outline
  • Least Squares Migration
  • Target-Oriented LSM
  • Numerical Examples
  • Conclusions

3
Standard Migration
LSM
dLm
dLm
mmigLTd
mmigLTL-1LTd
d observed data m reflectivity model
L forward operator mmig migration image
4
A Simple Model
Distance (km)
1
0
0
Depth (Km)
1.2
True Reflectivity Model
5
Migration Results
Distance (km)
Distance (km)
1
0
1
0
0
0
Depth (Km)
Depth (Km)
1.2
1.2
LSM-20
Standard Migration
6
Limitation of LSM
  • Requires 10 or more iterations,where each
    iteration requires several forward modeling and
    several migration steps
  • Impractical to implement for current 3-D
    prestack imaging of large data sets

7
Outline
  • Least Squares Migration
  • Target-Oriented LSM
  • Numerical Examples
  • Conclusions

8
Target-Oriented LSM
LSM
Data
Model
9
Outline
  • Least Squares Migration
  • Target-Oriented LSM
  • Numerical Results
  • Conclusions

10
Unocal Poststack Data
Distance (km)
14.4
0
0
Depth (Km)
1.5
Standard Migration
11
Unocal Poststack Data
Distance (km)
14.4
0
0
Depth (Km)
1.5
LSM-40
12
Unocal Poststack Data
Distance (km)
14.4
0
0
Depth (Km)
1.5
LSM-80
13
Unocal Poststack Data
Distance (km)
14.4
0
0
Depth (Km)
1.5
MD
14
Target-Oriented LSM
Distance (km)
0
14.4
a
0.8
D (Km)
1.5
b
0.8
D (Km)
1.5
(a) Standard Migration (b) LSM-5
15
Target-Oriented LSM
Distance (km)
0
14.4
a
0.8
D (Km)
1.5
b
0.8
D (Km)
1.5
(a) Standard Migration (b) LSM-10
16
Outline
  • Least Squares Migration
  • Target-Oriented LSM
  • Numerical Results
  • Conclusions

17
Conclusion
  • LSM is effective in suppressing the migration
    artifacts, but expensive
  • Target-oriented LSM improves the image quality
    in a few iterations
  • Target-oriented LSM greatly reduces CPU time

18
Future Work
  • How to better define the target data box
  • Test the target-oriented LSM on prestack data
  • Implement MD filter as a preconditioner to LSM

19
Acknowledgements
We thank Utah Tomography and Modeling/Migration
Consortium sponsors for their financial support
Write a Comment
User Comments (0)
About PowerShow.com