P-S migration using EO method
Thais Guirigay, John C. Bancroft
November, 2012
1
Outline
• Objective • Examples Hussar Synthetic model NEBC
• Summary • Acknowledgments
Objective
After finding an estimation of the converted wave velocity Vc and shear wave velocity Vs, from P-P and P-S wave data, show the results using EOM.
Hussar
Geophone 3C (SM7 10 Hz) Vibroseis low-dwell sweep
Geometry specification
First breaks picks
Refraction Statics analysis
Geometry Spreading compensation
Residual Statics
Stack
Trace edit
Datum/Elevation correction
NMO correction
Bin definition/ Header Update
Velocity analysis
Gabor Deconvolution
Migration
EOM
Processing flow for P-P data
Processing flow for P-S data Rotation into radial component
Import P-wave shot static and applied
Calculate and applied receiver elevation S statics
Geometry Spreading compensation
Repeat S-wave receiver statics
Check the CRS for additional significant receiver statics
Edit bad traces
Common Receiver Brute Stack (CRS)
Estimate S-wave receiver statics from CRS
Asymptotic CCP binning
Velocity analysis, Initial Vc
Gabor Deconvolution
Noise attenuation: Radial filter
PS wave Stack EOM
Surface wave noise attenuation
Migration
EOM processing flow
Generation of CSP gathers
Stack
NMO correction
Estimation of velocities
• Initial Vc1 with Vp and • Limited range EO gathers • Pick new Vc2
• Estimate Vs
• Full EO gathers with Vp and Vs • Pick new Vc3
• Moveout correction with Vc3 • Stack to complete the prestack migration
2 rms p rms sc
rms p rms s
V VV
V V− −
− −
=+
γ
Estimation of the velocities
Initial estimate of Vc1 2γ = 2.5γ =
0 1000 2000 0 1000 2000
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
Tim
e t
(sec
.)
Initial estimate of Vc1 x =
x =
Stacked line (Supergather)
Stacked line (EO method)
LCCSP gather x = 100 m
Second estimation of Vc2
0 1 2 3 4 5 60
1000
2000
3000
4000
5000
Time T0 (sec)
Vel
ocity
(m
/s)
Vp-rms, Vc-rms(G) for γ = 2 and picked Vc-rms(P)
Vp-rms
Vc-rms(G)
Vc-rms(P)
0 2000 4000 6000 8000 100000
1000
2000
3000
4000
5000
6000
Depth (m)
Vel
ocity
(m/s
)
Comparison of interval velocities in depth
Vp-int(z)Vc-int(Gz)Vc-int(Pz)
0 0.5 1 1.5 2 2.5 3 3.5 40
1000
2000
3000
4000
5000
6000
Time T0 (sec)
Vel
ocity
(m
/s)
Comparison of interval velocities
Vp-intVc-int(G)Vc-int(P)
Estimating the S velocities
0 1 2 3 4 50
1000
2000
3000
4000
5000
Time T0 (sec)
Vel
ocity
(m
/s)
Comparison of RMS velocities, with Vs-rms in Tp time
Vp-rmsVc-rms(P)Vs-rms(P)Vs-rms(P) in Tp time
0 1 2 3 4 50
1000
2000
3000
4000
5000
6000
Time T0 (sec)
Velo
city
(m
/s)
Comparison of interval velocities in time
Vp-intVc-int(P)Vs-int(P)
0 2000 4000 6000 8000 100000
1000
2000
3000
4000
5000
6000
Depth (m)
Velo
city
(m
/s)
Comparison of interval velocities in depth
Vp-int(z)Vc-int(Pz)Vs-int(Pz)
Full EOM
With the Vp and the estimated S velocity, new Vc3 is calculated picked from the new CCSP gathers
2 22 20 0
0 0
( ) ( )4 ( ) 4 ( )
p sc s r
rms p p rms s p
t tx h x ht t tV t V t− −
+ −= + = + + +
2 20
1 1 ˆrms srm p
es
tV
zV
h−−
= + +
Full EO gathers with Vp and Vs
Interval velocities
0 2000 4000 6000 8000 100002000
3000
4000
5000
6000
7000
Depth (m)
Vel
ocity
(m/s
)
Comparison of P interval velocities
Vp-int-Log(z)Vp-int-Picked(z)
0 1000 2000 3000 4000 5000 6000 7000500
1000
1500
2000
2500
3000
3500
4000
Depth (m)
Vel
ocity
(m/s
)
Comparison of Shear interval velocities
Vs-int-Log(z)Vs-int-Estimated(z)
Gamma functions
0 2000 4000 6000 8000 100001
1.5
2
2.5
3
3.5
4
4.5
Depth (m)
Gam
ma
Comparison of interval Gamma functions
Gamma from Vp-rms(Gz)Gamma from picked Vc-rms(Pz)Gamma from well-log
Final stack after Full EOM
0.0
2.0
1.0
2.0
1.0
0.0
3.0
1.0
0.0 P-P section P-S section
Final stack after post-stack migration
0.0
2.0
1.0
2.0
3.0
1.0
0.0 P-P section P-S section
Final stack after post-stack migration Final stack after Full EOM P-S section
Synthetic seismogram for PP
Belly River
Basal Belly River
Viking
Medicine River Coal
Mannville
Colorado
Synthetic seismogram for PS
Belly River
Basal Belly River
Viking
Medicine River Coal
Mannville
Colorado
Final stack after Full EOM
Belly River
Basal Belly River
Viking Medicine River Coal Mannville
12-27 12-27
Colorado
Results Synthetic model
(Lloyd and Margrave, 2010)
Estimating the velocities
0 500 1000 1500 2000 25001.5
2
2.5
3
3.5
Depth (m)
Gam
ma
Comparison of Gamma functions in depth
Gamma from Int. Vel.Gamma from RMS Vel.Gamma from Table 4.1 Layer Vp Vs γ
1 1500 500 3
2 2750 1100 2.5
3 3287 1700 1.93
4 3487 1800 1.94
5 4090 2000 2.03
6 4175 2050 2.04
P-P section
P-S section
Results NEBC
0 500 1000 1500 2000 2500 3000 3500 40001000
2000
3000
4000
5000
6000
Depth (m)
Vel
ocity
(m/s
)
Comparison of P interval velocities
Vp-int-Log(z)Vp-int-Picked(z)
Interval velocities
0 500 1000 1500 2000 2500 30000
500
1000
1500
2000
2500
3000
3500
Depth (m)
Vel
ocity
(m/s
)
Comparison of Shear interval velocities
Vs-int-Log(z)Vs-int-Estimated(z)
Estimating the velocities
0 500 1000 1500 2000 2500 30001
2
3
4
5
Depth (m)
Gam
ma γ
Gamma functions in depth from picked velocities
γ: Int. Vels. γ: RMS vels. γ: log
Summary
• This initial velocity Vc was estimated from RMS velocities Vp and an initial constant value for the Vp/Vs ratio .
• A reasonably accurate estimate of converted wave velocities Vc is required to form CCSP gathers as part of the Equivalent Offset Migration of converted waves.
• These converted wave velocities (second Vc) were then used to make an initial estimate of the shear wave velocities Vs. These velocities were then used with the Vp velocities for prestack migration.
Summary (continued)
• The common converted wave scatterpoint (CCSP) gathers were then used to create accurate estimates of Vc in order to improve Vs and .
• The quality of the method is demonstrated for the case of Hussar, synthetic model and NEBC dataset. The results show superior imaging when compared with alternative migration algorithms.
• The estimated interval velocities of Vp and Vs are compared with velocities obtained from well-logs and compare favorably.
γ
Acknowledgments
• All CREWES staff. Dr. Margrave Dr. Isaac
• All CREWES students • All CREWES sponsors Gedco Halliburton Husky Energy GeoKinetics Inova Nexen
Thanks