[email protected] .edu Multiple attenuation in the image space Paul Sava & Antoine Guitton Stanford University SEP
Dec 15, 2015
Multiple attenuation in the image space
Paul Sava & Antoine GuittonStanford University
SEP
Goal
• Method feasible in 3-D• Less expensive• Dense data requirement
• Exploit the data/imaging mismatch• Data: two-way propagation• Migration: one-way extrapolation
Key technology
• Migration by wavefield extrapolation (WEM)
• Angle-domain common-image gathers
• High resolution Radon Transforms
The big picture
WE prediction
S/N separation
S/N separation
Data
RT & Mute
Image
Data
NMO
RT & Mute
WE migration & ADCIG
Image
Multiple attenuation by RTs
– Moveout analysis• NMO
– Moveout analysis• WE migration
– S/N separation• RT + Mute
– S/N separation• RT + Mute
3-D depth imaging
• WE migration– Multi-arrival
• Angle-gathers– Single-valued
• Kirchhoff migration– Single-arrival
• Offset-gathers– Multi-valued
Biondi et al. (2003)Stolk & Symes (2002)
z
x
y
Which Radon transform?
2tan g
gqqz , Generic Radon Transform
2 gParabolic
Biondi & Symes (2003)
Tangent
q
g()
z
Synthetic example: S/N separation
primaries &multiples
primariesmultiplesART + muteART
Discussion
• PROs
– Cheap & robust– 3-D– Simple primaries
– Migration artifacts
• CONs
– Velocity model?
– Moveout function?– Interactive mute
– Inner angles– RT artifacts
Summary
WE prediction
S/N separation
S/N separation
Data
RT & Mute
Image
Data
NMO
RT & Mute
WE migration & ADCIG
Image
Summary
• Multiple attenuation after migration• WE migration• Angle gathers
• Cost/accuracy• Complex propagation• Cheap separation
• RT limitations • filtering approach