Embed Size (px)
Citation preview
Slide #1 [email protected]
Slide #2
(156990) Claerbout = 2003 KX18
Jon Claerbout (b. 1938) is a prolific contributor to the theory and art of exploration seismology. He was the first to demonstrate a practical method for imaging the Earth's interior using wavefields modeled in a computer. He also pioneered seismic interferometry, a method now used to probe the sun - Joe Dellinger (May 2011)
Slide #3
(156990) Claerbout = 2003 KX18
``I chose this asteroid`` because among those I have discovered solo it has a particularly inclined orbit (23 degrees), and "KX" in the designation, which makes me think of Fourier transforms.’’ Joe Dellinger
Slide #4
Wave-equation migration velocity analysis by residual
moveout fitting
Biondo Biondi
SEP 142 pp. 25-32 SEP 143 pp. 59-66
Slide #7
Ready for ``prime time’’?
§ Convergence – DSO gradient artifacts slows down convergence.
§ 3D WAZ – Poor source sampling causes artifacts in subsurface CIGs (offset or angle).
§ Resolution – Can we improve resolution by exploiting velocity information contained in back-scattered waves, and improve global convergence?
Slide #8
Ready for ``prime time’’?
§ Convergence – DSO gradient artifacts slows down convergence.
§ 3D WAZ – Poor source sampling causes artifacts in subsurface CIGs (offset or angle).
§ Resolution – Can we improve resolution by exploiting velocity information contained in back-scattered waves, and improve global convergence?
Slide #9
Ready for ``prime time’’?
§ Convergence – DSO gradient artifacts slows down convergence.
§ 3D WAZ – Poor source sampling causes artifacts in subsurface CIGs (offset or angle).
§ Resolution – Can we improve resolution by exploiting velocity information contained in back-scattered waves, and improve global convergence?
Slide #12 [email protected]
Velocity information in the data
Transmission kinematics & amplitudes
Yang Zhang’s & Ali Al-Momin’s
talks
Slide #13 [email protected]
Velocity information in the data
Transmission kinematics & amplitudes
Reflections kinematics & amplitudes
Ali Al-Momin’s talk
Yang Zhang’s & Ali Al-Momin’s
talks
Slide #15
Non-parametric objective function
Symes & Carazzone,
Geophysics 1991
Chavent & Jacewitz,
Geophysics 1995
Slide #16 [email protected]
Gradients with non-parametric objective functions
Stacking power maximization
Differential Semblance Optimization (DSO)
Gradients courtesy of Ali Al-Momin
Slide #25
z
Gradient evaluation – ΔρèΔI0
Δz(ρ)
Better-focused Image (I+)
Background Image ( )
Slide #26
z
Gradient evaluation – ΔρèΔI0
Δz(ρ)
Background Image ( )
Better-focused Image (I+)
Slide #29
Gradient evaluation – ΔI0èΔs
Biondi & Sava, SEG 1999
Luo & Schuster, Geophysics 1991
Slide #30 [email protected]
Challenges with one-parameter RMO of full-bandwidth data
Central frequency of 25 Hz
Slide #40
Summary § I presented a general methodology for
computing gradients of image-focusing objective functions defined by one (or more) kinematic parameters
§ Global convergence could be achieved in absence of low-frequency signal by properly smoothing objective functions along the axes of kinematic parameters.
Slide #41
Summary § I presented a general methodology for
computing gradients of image-focusing objective functions defined by one (or more) kinematic parameters
§ Global convergence could be achieved in absence of low-frequency signal by properly smoothing objective functions along the axes of kinematic parameters.
Slide #42 [email protected]
Slide #44 [email protected]
Challenges with single-parameter RMO of full-bandwidth data
Central frequency of 5 Hz
Slide #50 [email protected]
