Wavepath Migration versus Kirchhoff Migration: 3-D Prestack Examples

1
Wavepath Migration versus Kirchhoff Migration
3-D Prestack Examples
2
Outline

• Problems in Kirchhoff Migration
• Wavepath Migration
• Implementation of WM
• Numerical Results
• Conclusions

3
Forward Modeling
( Xg, 0 )
( Xs, 0 )
4
3D Kirchhoff Migration
( Xg, 0 )
( Xs, 0 )
5
3-D KM of a Single Trace
C
B
A
R
S
6
Problems in Kirchhoff Migration
Traveltime Information
Where Was Wave Reflected ?
The Whole Fat Ellipsoid !
Problem 1
Problem 2
Strong Far-Field Migration Artifact
Slow for 3-D Iterative Velocity Analysis
7
Outline

• Problems in Kirchhoff Migration
• Wavepath Migration
• Implementation of WM
• Numerical Results
• Conclusions

8
3D Wavepath Migration
( Xg, 0 )
9
3-D WM of a Single Trace
C
B
A
R
S
10
Wavepath Migration
Traveltime Ray Direction
11
Key Goals of 3-D WM

• To Achieve Higher CPU Efficiency
• Compared to 3-D KM
• To Generate Comparable or Better
• Image Quality than 3-D KM

12
Related References

• Time-Map Migration
• Sherrif Geldhart (1985)
• Wave Equation Tomography
• Woodward Rocca (1988)
• Gaussian Beam Migration
• Ross Hill (1990)
• Kirchhoff Beam Migration
• Yonghe Sun et al., (1999)

13
Outline

• Problems in Kirchhoff Migration
• Wavepath Migration
• Implementation of WM
• Numerical Results
• Conclusions

14
Key Steps in WM
R
S
Quasi-ellipsoid
15
Outline

• Problems in Kirchhoff Migration
• Wavepath Migration
• Implementation of WM
• Numerical Results
• 3-D Prestack Point Scatterer Data
• 3-D Prestack SEG/EAGE Salt Data
• 3-D Prestack West Texas Field Data
• Conclusions

16
3-D Prestack KM Point Scatterer Response
0.1
0.4
Reflectivity
Reflectivity
Z0-9
Z0-1
-0.05
-0.2
1
1
1
1
Y Offset (km)
Y Offset (km)
X Offset (km)
X Offset (km)
0
0
1
0.02
Reflectivity
Reflectivity
Z0
Z08
-0.5
-0.01
1
1
1
1
Y Offset (km)
Y Offset (km)
X Offset (km)
X Offset (km)
0
0
17
3-D Prestack WM Point Scatterer Response
0.1
0.4
Reflectivity
Reflectivity
Z0-9
Z0-1
-0.05
-0.2
1
1
1
1
Y Offset (km)
Y Offset (km)
X Offset (km)
X Offset (km)
0
0
1
0.02
Reflectivity
Reflectivity
Z0
Z08
-0.5
-0.01
1
1
1
1
Y Offset (km)
Y Offset (km)
X Offset (km)
X Offset (km)
0
0
18
Outline

• Problems in Kirchhoff Migration
• Wavepath Migration
• Implementation of WM
• Numerical Results
• 3-D Prestack Point Scatterer Data
• 3-D Prestack SEG/EAGE Salt Data
• 3-D Prestack West Texas Field Data
• Conclusions

19
A Common Shot Gather
Trace Number
1
390
0
Time (sec)
5.0
20
Inline Velocity Model
Offset (km)
0
9.2
0
Depth (km)
3.8
21
Inline KM (CPU1)
Inline WM (CPU1/33)
Offset (km)
Offset (km)
0
9.2
0
9.2
0
Depth (km)
3.8
22
Inline KM (CPU1)
Inline WM (CPU1/170)
(subsample)
Offset (km)
Offset (km)
0
9.2
0
9.2
0
Depth (km)
3.8
23
Zoom Views of Inline Sections
WM
KM
Sub WM
Model
Offset 36.5 km, Depth 0.31.8 km
24
Zoom Views of Crossline Sections
WM
KM
Sub WM
Model
Offset 1.84 km, Depth 0.62.1 km
25
Horizontal Slices (Depth1.4 km)
WM
KM
Sub WM
Model
Inline 1.87.2 km, Crossline 04 km
26
Outline

• Problems in Kirchhoff Migration
• Wavepath Migration
• Implementation of WM
• Numerical Results
• 3-D Prestack Point Scatterer Data
• 3-D Prestack SEG/EAGE Salt Data
• 3-D Prestack West Texas Field Data
• Conclusions

27
A Common Shot Gather
Trace Number
54
193
0
Time (sec)
3.4
28
Inline KM (CPU1)
Inline WM (CPU1/14)
Offset (km)
Offset (km)
0.4
4.5
0.4
4.5
0.8
Depth (km)
3.8
29
Inline KM (CPU1)
Inline WM (CPU1/50)
(subsample)
Offset (km)
Offset (km)
0.4
4.5
0.4
4.5
0.8
Depth (km)
3.8
30
Crossline KM (CPU1)
Crossline WM (CPU1/14)
Offset (km)
Offset (km)
0.3
3.5
0.3
3.5
0.8
Depth (km)
3.3
31
Crossline KM (CPU1)
Crossline WM (CPU1/50)
(subsample)
Offset (km)
Offset (km)
0.3
3.5
0.3
3.5
0.8
Depth (km)
3.3
32
Horizontal Slices (Depth2.5 km)
WM (Sub, CPU1/50)
KM (CPU1)
WM (CPU1/14)
Inline 04.6 km, Crossline 03.8
33
Outline

• Problems in Kirchhoff Migration
• Wavepath Migration
• Implementation of WM
• Numerical Results
• Conclusions

34
Conclusions

• SEG/EAGE Salt Data
• Fewer Migration Artifacts
• Better for Complex Salt Boundary
• Higher Computational Efficiency
• CPU
• KM 1 WM 1/33
• Subsampled WM 1/170

35
Conclusions

• West Texas Field Data
• Fewer Migration Artifacts
• Similar Image Quality
• Higher Computational Efficiency
• CPU
• KM 1 WM 1/14
• Subsampled WM 1/50

36
Acknowledgements
We thank UTAM sponsors for their financial support