SlideShare a Scribd company logo
1 of 32
Student Name: Benjamin Seive
Date presentation: 28/04/2015
Page 1 of 14
Agenda
1. Where does Reverse Time Migration (RTM)
algorithm fit within other algorithms
2. The three assumptions behind the RTM
migration
3. How RTM works: mathematics
4. Conclusion and modern RTM migration
applications
Page 2 of 14
• RTM = Reverse Time Migration
• FDM=Finite Difference Modelling
• ZO=zero offset time section
• f(x,z,t)=seismogram in x,z,t domain
• F(x,z,t)=Fourier Transform of seismogram
• kx=wavenumber Fourier Transform of
𝜕𝑃
𝜕𝑥
• kz=wavenumber Fourier Transform of
𝜕𝑃
𝜕𝑍
• P=pressure amplitude
• C=velocity (constant velocity field)
• V(x,z)=velocity (variable velocity field)
• 𝑡 = 𝑡𝑓 final time of the ZO at z=0
• 𝑡 = 𝑡0 exploding time within exploding reflector concept
• X-T domain = offset – time domain
• X-Z domain = offset – depth domain
Definition of Terms
Page 3 of 14
Where does RTM fit in Migration’s world ?
Page 5 of 14
Sources: Bancroft, 2007
Main types of migration methods:
• Kirchhoff migration
• FK migration
• Downward Continuation Migration (Phase Shift)
Downward Migration RTM Migration
• RTM
Sources: http://sep.stanford.edu/oldreports/sep40/40_01.pdf //
Loewenthal, 1983
RTM Assumptions: Exploding Reflectors
Page 6 of 14
Three Assumptions are:
• CDP stacked data time section ( Zero Offset Time section) in X-
Z plan that fits exploding reflection models/concept
Explosion at 𝑡 = 𝑡0Resulting CDP stacked
section at 𝑡 = 𝑡𝑓
Sources: Urosevic, Curtin University Lectures Notes, 2015
RTM Assumptions: backward time marching schemes
Page 7 of 14
Three Assumptions are:
• Exploding Reflector concept
• Solution by FDM of wave equation can be driven either forward or
backward in TIME and keep spatial frame X-Z fixed
RTM Assumptions: backward time marching schemes
Sources: Urosevic, Curtin University Lectures Notes, 2015
Page 7 of 14
Three Assumptions are:
• Exploding Reflector concept
• Solution by FDM of wave equation can be driven either forward or
backward in TIME and keep spatial frame X-Z fixed
RTM Assumptions: backward time marching schemes
Sources: Urosevic, Curtin University Lectures Notes, 2015
Page 7 of 14
Three Assumptions are:
• Exploding Reflector concept
• Solution by FDM of wave equation can be driven either forward or
backward in TIME and keep spatial frame X-Z fixed
RTM Assumptions: backward time marching schemes
Sources: Urosevic, Curtin University Lectures Notes, 2015
Page 7 of 14
Three Assumptions are:
• Exploding Reflector concept
• Solution by FDM of wave equation can be driven either forward or
backward in TIME and keep spatial frame X-Z fixed
RTM Assumptions: backward time marching schemes
Sources: Urosevic, Curtin University Lectures Notes, 2015
Page 7 of 14
Three Assumptions are:
• Exploding Reflector concept
• Solution by FDM of wave equation can be driven either forward or
backward in TIME and keep spatial frame X-Z fixed
Sources: Loewenthal, 1983
RTM Assumptions: Exploding Reflectors
Page 8 of 14
CDP section =U𝑛𝑚𝑖𝑔𝑟𝑎𝑡𝑒𝑑 𝑠𝑒𝑐𝑡𝑖𝑜𝑛 =
𝑓 𝑥, 𝑧, 𝑡 = 𝑡𝑓 = 𝑓(𝑥′
, 𝑧′
, 𝑡′
)
CDP section =𝑚𝑖𝑔𝑟𝑎𝑡𝑒𝑑 𝑠𝑒𝑐𝑡𝑖𝑜𝑛 =
𝑓 𝑥, 𝑧, 𝑡 = 𝑡0
𝑡′
= 2. 𝑡𝑖,𝑗 = 2. 𝑡𝑗,𝑖
Three Assumptions are:
• Exploding Reflector concept
• Solution of wave equation can be driven either forward or
backward in TIME and keep spatial frame fixedby FDM
RTM Assumptions: backward time marching schemes
Whole point of the RTM migration within the
exploding reflector concept is to move energy (wave
propagation) back in time to its reflection (i.e. its
explosion) time .
 Take successive steps BACKWARD in TIME
to go from 𝒕 = 𝒕 𝒇 𝒕𝒐 𝒕 = 𝒕 𝟎
Page 9 of 14
Sources: Loewenthal, 1983 // Levin,1985
RTM Assumptions: backward time marching schemes
Page 10 of 14
𝐹 𝑘 𝑥, 𝑘 𝑧, 𝑡 = 𝑓 𝑥, 𝑧, 𝑡 . exp 𝑖(𝑘 𝑥. 𝑥 + 𝑘 𝑧. 𝑧)
𝑍𝑋
Sources: Loewenthal, 1983
Three Assumptions are:
• Exploding Reflector concept
• Take Successive Steps Backward
• Easier and Faster to work in the space frequency domain
take Fourier transform of the whole zero offset time section
Zero offset time section
at time t
Fourier domain
representation of Zero
offset time section at time t
RTM Assumptions: backward time marching schemes
Page 10 of 14
Inverse Fourier Transform: we can compute the field at any
desired time level t from the corresponding value at t’
𝐹 𝑘 𝑥𝑝, 𝑘 𝑧𝑞, 𝑡′
= 𝑓 𝑚. ∆𝑥, 𝑛. ∆𝑧, 𝑡′
. exp 𝑖(𝑘 𝑥𝑝. 𝑚∆𝑥 + 𝑘 𝑧𝑞. 𝑛∆𝑧)
𝑁
𝑛=0
𝑀
𝑚=0
f 𝑚∆𝑥, 𝑛∆𝑧, 𝑡 = 𝐹 𝑘 𝑥𝑝, 𝑘 𝑧𝑞, 𝑡′
. exp −𝑖 𝑘 𝑥𝑝. 𝑚∆𝑥 + 𝑘 𝑧𝑞. 𝑛∆𝑧 + 𝜔 𝑝𝑞(𝑡 − 𝑡′
)𝑁
𝑞=0
𝑀
𝑝=0
adding some shift in time
Sources: Loewenthal, 1983
𝑀 = 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑡𝑟𝑎𝑐𝑒𝑠
𝑁 = 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑡𝑖𝑚𝑒 𝑠𝑎𝑚𝑝𝑙𝑒𝑠 𝑎𝑙𝑜𝑛𝑔 𝑡𝑟𝑎𝑐𝑒
∆𝑥 = 𝑡𝑟𝑎𝑐𝑒 𝑖𝑛𝑡𝑒𝑟𝑣𝑎𝑙
∆𝑡 = 𝑡𝑖𝑚𝑒 𝑠𝑎𝑚𝑝𝑙𝑖𝑛𝑔 𝑖𝑛𝑡𝑒𝑟𝑣𝑎𝑙
𝑣 𝑥, 𝑧 = 𝑣𝑒𝑙𝑜𝑐𝑖𝑡𝑦 𝑓𝑖𝑒𝑙𝑑
∆𝑧 = 𝑑𝑒𝑝𝑡ℎ 𝑖𝑛𝑡𝑒𝑟𝑣𝑎𝑙
𝜔 𝑝,𝑞 = 𝑡𝑒𝑚𝑝𝑜𝑟𝑎𝑙 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑑𝑒𝑝𝑒𝑛𝑑𝑒𝑛𝑐𝑒
𝑘 𝑥𝑝 =
2𝜋𝑝
(𝑀.∆𝑥)
=wavenumber x direction=
𝑑𝑖𝑠𝑐𝑟𝑒𝑡𝑒 𝑠𝑝𝑎𝑡𝑖𝑎𝑙 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑎𝑙𝑜𝑛𝑔 𝑥 𝑎𝑥𝑖𝑠
𝑝 = 0,1,2 … . 𝑀
2
𝑘 𝑧𝑞 =
2𝜋𝑞
(𝑁.∆𝑧)
=wavenumber z direction=
𝑑𝑖𝑠𝑐𝑟𝑒𝑡𝑒 𝑠𝑝𝑎𝑡𝑖𝑎𝑙 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑎𝑙𝑜𝑛𝑔 𝑥 𝑎𝑥𝑖𝑠
𝑞 = 0,1,2 … . 𝑁
2
∆𝑡 = 𝑡 − 𝑡′
> 0
RTM Assumptions: shift in time 𝝎 𝟐
𝒑,𝒒
Page 11 of 14
Wave equation Fourier Transform
𝜕2 𝑃
𝜕𝑡2
= 𝑉2.
𝜕2 𝑃
𝜕𝑥2
+
𝜕2 𝑃
𝜕𝑧2
𝜔2
𝑝,𝑞 = 𝑉2
. 𝑘2
𝑥𝑝 + 𝑘2
𝑧𝑞
Sources: Loewenthal, 1983
𝑘 𝑥𝑝 =
2𝜋𝑝
(𝑀.∆𝑥)
=wavenumber x direction=
𝑑𝑖𝑠𝑐𝑟𝑒𝑡𝑒 𝑠𝑝𝑎𝑡𝑖𝑎𝑙 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑎𝑙𝑜𝑛𝑔 𝑥 𝑎𝑥𝑖𝑠
𝑝 = 0,1,2 … . 𝑀
2
𝑘 𝑧𝑞 =
2𝜋𝑞
(𝑁.∆𝑧)
=wavenumber z direction=
𝑑𝑖𝑠𝑐𝑟𝑒𝑡𝑒 𝑠𝑝𝑎𝑡𝑖𝑎𝑙 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑎𝑙𝑜𝑛𝑔 𝑥 𝑎𝑥𝑖𝑠
𝑞 = 0,1,2 … . 𝑁
2
𝑉 = 𝑣𝑒𝑙𝑜𝑐𝑖𝑡𝑦
RTM problem solved: backward time marching schemes
Page 12 of 14
f 𝒎∆𝒙, 𝒏∆𝒛, 𝒕 𝒇 − ∆𝒕 = 𝑭 𝒌 𝒙𝒑, 𝒌 𝒛𝒒, 𝒕 𝒇 . 𝐞𝐱𝐩 −𝒊 𝒌 𝒙𝒑. 𝒎∆𝒙 + 𝒌 𝒛𝒒. 𝒏∆𝒛 − 𝝎 𝒑𝒒(∆𝒕)𝑵
𝒒=𝟎
𝑴
𝒑=𝟎
Take 𝑡′ = 𝑡𝑓 𝑎𝑛𝑑 𝑡 = 𝑡𝑓 − ∆𝑡
removing some shift ∆𝒕 in time
RTM Migration
Sources: Loewenthal, 1983
RTM problem solved: backward time marching schemes
Loewenthal synthetic applications:
• dipping layer 29°
• dipping layer 33°
• dipping layer 35°
• syncline
Ok errror < 1%
OK error <1%
Ok error 5%
OK
Sources: Loewenthal, 1983
Page 13 of 14
RTM problem solved: backward time marching schemes
Overall RTM do work well !
Sources: Loewenthal, 1983
Page 13 of 14
RTM : Conclusion
RTM advantages:
• Used today (in prestack mode) to improve imaging in complex
geology (e.g Salt Dome)
• Resulting images are usually easier to interpret
• Can handle severe combination of structural dip (steep and
even overturned reflectors) with high velocity contrast
RTM disadvantages:
• Computer intensive (was ?? )
Sources: Ion, 2015 // CGG, 2015 // Paradigm, 2015 // CSEG, recorder, 2009
Page 14 of 14
RTM : Conclusion
Sources: Ion, 2015
?
Page 14 of 14
RTM : Conclusion
Sources: Ion, 2015
Page 14 of 14
22
Thanks for your attention
RTM : Conclusion
Sources: Ion, 2015
Page 14 of 14
What is Migration: an example
Page 4 of 14
Sources: Yilmaz, 1987
Focusing effect of
a sharp syncline
Migration
algorithm
What is Migration ?
Page 4 of 11
Post Stack Migration = Dealing with circles
𝑇𝑎𝑛 𝛼 = 𝑆𝑖𝑛 (𝛽)
Sources: Bancroft, 2007 // CSEG recorder, 2012
• The mathematical process of re-arranging the reflection
events at their correct locations in the seismogram
• Migration will shortens dipping events and increase the dip
• Can be in X-T or X-Z domain and pre-stack or post stack
26
What is Downward Continuation in Depth
Sources: Bancroft, 2007
27
What is Downward Continuation in Time
Sources: Bancroft, 2007
28
What is Phase Shift
Sources: Bancroft, 2007
29
Sources: CSEG recorder, 2012
What is Migration ?
RTM Assumptions: backward time marching schemes
From 𝑡 = 𝑡𝑓 = 𝑡′ 𝑡𝑜 𝑡 = 𝑡0
 Take successive steps BACKWARD in TIME
until the desired migration is achieved
Page 9 of 11
Sources: Loewenthal, 1983 // Levin, 1985
RTM problem solved: backward time marching schemes
Page 10 of 11
∆𝑡
32

More Related Content

What's hot

A review of time­‐frequency methods
A review of time­‐frequency methodsA review of time­‐frequency methods
A review of time­‐frequency methods
UT Technology
 
Fundamentals of Seismic Refraction
Fundamentals of Seismic RefractionFundamentals of Seismic Refraction
Fundamentals of Seismic Refraction
Ali Osman Öncel
 
The role of kalman filter in improving the accuracy of gps kinematic technique
The role of kalman filter in improving the accuracy of gps kinematic techniqueThe role of kalman filter in improving the accuracy of gps kinematic technique
The role of kalman filter in improving the accuracy of gps kinematic technique
IAEME Publication
 

What's hot (20)

Automated seismic-to-well ties?
Automated seismic-to-well ties?Automated seismic-to-well ties?
Automated seismic-to-well ties?
 
Seismic presentation for Dip correction for convolution modelling
Seismic presentation for Dip correction for convolution modellingSeismic presentation for Dip correction for convolution modelling
Seismic presentation for Dip correction for convolution modelling
 
Seismic velocity analysis _ aghazade
Seismic velocity analysis _ aghazadeSeismic velocity analysis _ aghazade
Seismic velocity analysis _ aghazade
 
Siddhartha fft pptx
Siddhartha fft pptxSiddhartha fft pptx
Siddhartha fft pptx
 
Seismic refraction method lec22
Seismic refraction method lec22Seismic refraction method lec22
Seismic refraction method lec22
 
A review of time­‐frequency methods
A review of time­‐frequency methodsA review of time­‐frequency methods
A review of time­‐frequency methods
 
Fundamentals of Seismic Refraction
Fundamentals of Seismic RefractionFundamentals of Seismic Refraction
Fundamentals of Seismic Refraction
 
modeling of system electromechanical, Armature Controlled D.C Motor -Reduced ...
modeling of system electromechanical, Armature Controlled D.C Motor -Reduced ...modeling of system electromechanical, Armature Controlled D.C Motor -Reduced ...
modeling of system electromechanical, Armature Controlled D.C Motor -Reduced ...
 
Sufficient decrease is all you need
Sufficient decrease is all you needSufficient decrease is all you need
Sufficient decrease is all you need
 
Bilinear z transformaion
Bilinear z transformaionBilinear z transformaion
Bilinear z transformaion
 
Seismic geometric corrections
Seismic geometric correctionsSeismic geometric corrections
Seismic geometric corrections
 
3.Frequency Domain Representation of Signals and Systems
3.Frequency Domain Representation of Signals and Systems3.Frequency Domain Representation of Signals and Systems
3.Frequency Domain Representation of Signals and Systems
 
Filter- IIR - Digital signal processing(DSP)
Filter- IIR - Digital signal processing(DSP)Filter- IIR - Digital signal processing(DSP)
Filter- IIR - Digital signal processing(DSP)
 
Introduction to DFT Part 1
Introduction to DFT Part 1 Introduction to DFT Part 1
Introduction to DFT Part 1
 
Lti system(akept)
Lti system(akept)Lti system(akept)
Lti system(akept)
 
Seismic data processing
Seismic data processingSeismic data processing
Seismic data processing
 
F Giordano: spin-dependent effects in spin-averaged DIS
F Giordano: spin-dependent effects in spin-averaged DISF Giordano: spin-dependent effects in spin-averaged DIS
F Giordano: spin-dependent effects in spin-averaged DIS
 
The role of kalman filter in improving the accuracy of gps kinematic technique
The role of kalman filter in improving the accuracy of gps kinematic techniqueThe role of kalman filter in improving the accuracy of gps kinematic technique
The role of kalman filter in improving the accuracy of gps kinematic technique
 
DSP_FOEHU - Lec 11 - IIR Filter Design
DSP_FOEHU - Lec 11 - IIR Filter DesignDSP_FOEHU - Lec 11 - IIR Filter Design
DSP_FOEHU - Lec 11 - IIR Filter Design
 
Signals and Systems Ch 4 5_Fourier Domain
Signals and Systems Ch 4 5_Fourier DomainSignals and Systems Ch 4 5_Fourier Domain
Signals and Systems Ch 4 5_Fourier Domain
 

Similar to Comments_on_Paper_Mufti_1983

Semet Gecco06
Semet Gecco06Semet Gecco06
Semet Gecco06
ysemet
 
Introduction to Seismic Method
Introduction to Seismic MethodIntroduction to Seismic Method
Introduction to Seismic Method
Şarlatan Avcısı
 

Similar to Comments_on_Paper_Mufti_1983 (20)

Continuous and Discrete Crooklet Transform
Continuous and Discrete Crooklet TransformContinuous and Discrete Crooklet Transform
Continuous and Discrete Crooklet Transform
 
Fdtd ppt for mine
Fdtd ppt   for mineFdtd ppt   for mine
Fdtd ppt for mine
 
Design Method of Directional GenLOT with Trend Vanishing Moments
Design Method of Directional GenLOT with Trend Vanishing MomentsDesign Method of Directional GenLOT with Trend Vanishing Moments
Design Method of Directional GenLOT with Trend Vanishing Moments
 
Acoustic Model.pptx
Acoustic Model.pptxAcoustic Model.pptx
Acoustic Model.pptx
 
An introduction to Realistic Ocean Rendering through FFT - Fabio Suriano - Co...
An introduction to Realistic Ocean Rendering through FFT - Fabio Suriano - Co...An introduction to Realistic Ocean Rendering through FFT - Fabio Suriano - Co...
An introduction to Realistic Ocean Rendering through FFT - Fabio Suriano - Co...
 
What Is Fourier Transform
What Is Fourier TransformWhat Is Fourier Transform
What Is Fourier Transform
 
2014.10.dartmouth
2014.10.dartmouth2014.10.dartmouth
2014.10.dartmouth
 
Semet Gecco06
Semet Gecco06Semet Gecco06
Semet Gecco06
 
dynamics11lecture.ppt
dynamics11lecture.pptdynamics11lecture.ppt
dynamics11lecture.ppt
 
dynamics11lecture kinematics of particles.ppt
dynamics11lecture kinematics of particles.pptdynamics11lecture kinematics of particles.ppt
dynamics11lecture kinematics of particles.ppt
 
Hprec7.3
Hprec7.3Hprec7.3
Hprec7.3
 
Newton's Laws of motion Lec2
Newton's Laws of motion Lec2Newton's Laws of motion Lec2
Newton's Laws of motion Lec2
 
Від побудови сейсмічних зображень до інверсії
Від побудови сейсмічних зображень до інверсіїВід побудови сейсмічних зображень до інверсії
Від побудови сейсмічних зображень до інверсії
 
03 Motion in Two & Three Dimensions.ppt
03 Motion in Two & Three Dimensions.ppt03 Motion in Two & Three Dimensions.ppt
03 Motion in Two & Three Dimensions.ppt
 
The Removal of Large Space Debris Using Tethered Space Tug
The Removal of Large Space Debris Using Tethered Space TugThe Removal of Large Space Debris Using Tethered Space Tug
The Removal of Large Space Debris Using Tethered Space Tug
 
kuramoto
kuramotokuramoto
kuramoto
 
Basics of edge detection and forier transform
Basics of edge detection and forier transformBasics of edge detection and forier transform
Basics of edge detection and forier transform
 
Introduction to Seismic Method
Introduction to Seismic MethodIntroduction to Seismic Method
Introduction to Seismic Method
 
Fourier slide
Fourier slideFourier slide
Fourier slide
 
1.Day 1 Rectilinera Motion Part 01.pdf
1.Day 1 Rectilinera Motion Part 01.pdf1.Day 1 Rectilinera Motion Part 01.pdf
1.Day 1 Rectilinera Motion Part 01.pdf
 

Comments_on_Paper_Mufti_1983

  • 1. Student Name: Benjamin Seive Date presentation: 28/04/2015 Page 1 of 14
  • 2. Agenda 1. Where does Reverse Time Migration (RTM) algorithm fit within other algorithms 2. The three assumptions behind the RTM migration 3. How RTM works: mathematics 4. Conclusion and modern RTM migration applications Page 2 of 14
  • 3. • RTM = Reverse Time Migration • FDM=Finite Difference Modelling • ZO=zero offset time section • f(x,z,t)=seismogram in x,z,t domain • F(x,z,t)=Fourier Transform of seismogram • kx=wavenumber Fourier Transform of 𝜕𝑃 𝜕𝑥 • kz=wavenumber Fourier Transform of 𝜕𝑃 𝜕𝑍 • P=pressure amplitude • C=velocity (constant velocity field) • V(x,z)=velocity (variable velocity field) • 𝑡 = 𝑡𝑓 final time of the ZO at z=0 • 𝑡 = 𝑡0 exploding time within exploding reflector concept • X-T domain = offset – time domain • X-Z domain = offset – depth domain Definition of Terms Page 3 of 14
  • 4. Where does RTM fit in Migration’s world ? Page 5 of 14 Sources: Bancroft, 2007 Main types of migration methods: • Kirchhoff migration • FK migration • Downward Continuation Migration (Phase Shift) Downward Migration RTM Migration • RTM
  • 5. Sources: http://sep.stanford.edu/oldreports/sep40/40_01.pdf // Loewenthal, 1983 RTM Assumptions: Exploding Reflectors Page 6 of 14 Three Assumptions are: • CDP stacked data time section ( Zero Offset Time section) in X- Z plan that fits exploding reflection models/concept Explosion at 𝑡 = 𝑡0Resulting CDP stacked section at 𝑡 = 𝑡𝑓
  • 6. Sources: Urosevic, Curtin University Lectures Notes, 2015 RTM Assumptions: backward time marching schemes Page 7 of 14 Three Assumptions are: • Exploding Reflector concept • Solution by FDM of wave equation can be driven either forward or backward in TIME and keep spatial frame X-Z fixed
  • 7. RTM Assumptions: backward time marching schemes Sources: Urosevic, Curtin University Lectures Notes, 2015 Page 7 of 14 Three Assumptions are: • Exploding Reflector concept • Solution by FDM of wave equation can be driven either forward or backward in TIME and keep spatial frame X-Z fixed
  • 8. RTM Assumptions: backward time marching schemes Sources: Urosevic, Curtin University Lectures Notes, 2015 Page 7 of 14 Three Assumptions are: • Exploding Reflector concept • Solution by FDM of wave equation can be driven either forward or backward in TIME and keep spatial frame X-Z fixed
  • 9. RTM Assumptions: backward time marching schemes Sources: Urosevic, Curtin University Lectures Notes, 2015 Page 7 of 14 Three Assumptions are: • Exploding Reflector concept • Solution by FDM of wave equation can be driven either forward or backward in TIME and keep spatial frame X-Z fixed
  • 10. RTM Assumptions: backward time marching schemes Sources: Urosevic, Curtin University Lectures Notes, 2015 Page 7 of 14 Three Assumptions are: • Exploding Reflector concept • Solution by FDM of wave equation can be driven either forward or backward in TIME and keep spatial frame X-Z fixed
  • 11. Sources: Loewenthal, 1983 RTM Assumptions: Exploding Reflectors Page 8 of 14 CDP section =U𝑛𝑚𝑖𝑔𝑟𝑎𝑡𝑒𝑑 𝑠𝑒𝑐𝑡𝑖𝑜𝑛 = 𝑓 𝑥, 𝑧, 𝑡 = 𝑡𝑓 = 𝑓(𝑥′ , 𝑧′ , 𝑡′ ) CDP section =𝑚𝑖𝑔𝑟𝑎𝑡𝑒𝑑 𝑠𝑒𝑐𝑡𝑖𝑜𝑛 = 𝑓 𝑥, 𝑧, 𝑡 = 𝑡0 𝑡′ = 2. 𝑡𝑖,𝑗 = 2. 𝑡𝑗,𝑖 Three Assumptions are: • Exploding Reflector concept • Solution of wave equation can be driven either forward or backward in TIME and keep spatial frame fixedby FDM
  • 12. RTM Assumptions: backward time marching schemes Whole point of the RTM migration within the exploding reflector concept is to move energy (wave propagation) back in time to its reflection (i.e. its explosion) time .  Take successive steps BACKWARD in TIME to go from 𝒕 = 𝒕 𝒇 𝒕𝒐 𝒕 = 𝒕 𝟎 Page 9 of 14 Sources: Loewenthal, 1983 // Levin,1985
  • 13. RTM Assumptions: backward time marching schemes Page 10 of 14 𝐹 𝑘 𝑥, 𝑘 𝑧, 𝑡 = 𝑓 𝑥, 𝑧, 𝑡 . exp 𝑖(𝑘 𝑥. 𝑥 + 𝑘 𝑧. 𝑧) 𝑍𝑋 Sources: Loewenthal, 1983 Three Assumptions are: • Exploding Reflector concept • Take Successive Steps Backward • Easier and Faster to work in the space frequency domain take Fourier transform of the whole zero offset time section Zero offset time section at time t Fourier domain representation of Zero offset time section at time t
  • 14. RTM Assumptions: backward time marching schemes Page 10 of 14 Inverse Fourier Transform: we can compute the field at any desired time level t from the corresponding value at t’ 𝐹 𝑘 𝑥𝑝, 𝑘 𝑧𝑞, 𝑡′ = 𝑓 𝑚. ∆𝑥, 𝑛. ∆𝑧, 𝑡′ . exp 𝑖(𝑘 𝑥𝑝. 𝑚∆𝑥 + 𝑘 𝑧𝑞. 𝑛∆𝑧) 𝑁 𝑛=0 𝑀 𝑚=0 f 𝑚∆𝑥, 𝑛∆𝑧, 𝑡 = 𝐹 𝑘 𝑥𝑝, 𝑘 𝑧𝑞, 𝑡′ . exp −𝑖 𝑘 𝑥𝑝. 𝑚∆𝑥 + 𝑘 𝑧𝑞. 𝑛∆𝑧 + 𝜔 𝑝𝑞(𝑡 − 𝑡′ )𝑁 𝑞=0 𝑀 𝑝=0 adding some shift in time Sources: Loewenthal, 1983 𝑀 = 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑡𝑟𝑎𝑐𝑒𝑠 𝑁 = 𝑛𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑡𝑖𝑚𝑒 𝑠𝑎𝑚𝑝𝑙𝑒𝑠 𝑎𝑙𝑜𝑛𝑔 𝑡𝑟𝑎𝑐𝑒 ∆𝑥 = 𝑡𝑟𝑎𝑐𝑒 𝑖𝑛𝑡𝑒𝑟𝑣𝑎𝑙 ∆𝑡 = 𝑡𝑖𝑚𝑒 𝑠𝑎𝑚𝑝𝑙𝑖𝑛𝑔 𝑖𝑛𝑡𝑒𝑟𝑣𝑎𝑙 𝑣 𝑥, 𝑧 = 𝑣𝑒𝑙𝑜𝑐𝑖𝑡𝑦 𝑓𝑖𝑒𝑙𝑑 ∆𝑧 = 𝑑𝑒𝑝𝑡ℎ 𝑖𝑛𝑡𝑒𝑟𝑣𝑎𝑙 𝜔 𝑝,𝑞 = 𝑡𝑒𝑚𝑝𝑜𝑟𝑎𝑙 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑑𝑒𝑝𝑒𝑛𝑑𝑒𝑛𝑐𝑒 𝑘 𝑥𝑝 = 2𝜋𝑝 (𝑀.∆𝑥) =wavenumber x direction= 𝑑𝑖𝑠𝑐𝑟𝑒𝑡𝑒 𝑠𝑝𝑎𝑡𝑖𝑎𝑙 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑎𝑙𝑜𝑛𝑔 𝑥 𝑎𝑥𝑖𝑠 𝑝 = 0,1,2 … . 𝑀 2 𝑘 𝑧𝑞 = 2𝜋𝑞 (𝑁.∆𝑧) =wavenumber z direction= 𝑑𝑖𝑠𝑐𝑟𝑒𝑡𝑒 𝑠𝑝𝑎𝑡𝑖𝑎𝑙 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑎𝑙𝑜𝑛𝑔 𝑥 𝑎𝑥𝑖𝑠 𝑞 = 0,1,2 … . 𝑁 2 ∆𝑡 = 𝑡 − 𝑡′ > 0
  • 15. RTM Assumptions: shift in time 𝝎 𝟐 𝒑,𝒒 Page 11 of 14 Wave equation Fourier Transform 𝜕2 𝑃 𝜕𝑡2 = 𝑉2. 𝜕2 𝑃 𝜕𝑥2 + 𝜕2 𝑃 𝜕𝑧2 𝜔2 𝑝,𝑞 = 𝑉2 . 𝑘2 𝑥𝑝 + 𝑘2 𝑧𝑞 Sources: Loewenthal, 1983 𝑘 𝑥𝑝 = 2𝜋𝑝 (𝑀.∆𝑥) =wavenumber x direction= 𝑑𝑖𝑠𝑐𝑟𝑒𝑡𝑒 𝑠𝑝𝑎𝑡𝑖𝑎𝑙 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑎𝑙𝑜𝑛𝑔 𝑥 𝑎𝑥𝑖𝑠 𝑝 = 0,1,2 … . 𝑀 2 𝑘 𝑧𝑞 = 2𝜋𝑞 (𝑁.∆𝑧) =wavenumber z direction= 𝑑𝑖𝑠𝑐𝑟𝑒𝑡𝑒 𝑠𝑝𝑎𝑡𝑖𝑎𝑙 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑎𝑙𝑜𝑛𝑔 𝑥 𝑎𝑥𝑖𝑠 𝑞 = 0,1,2 … . 𝑁 2 𝑉 = 𝑣𝑒𝑙𝑜𝑐𝑖𝑡𝑦
  • 16. RTM problem solved: backward time marching schemes Page 12 of 14 f 𝒎∆𝒙, 𝒏∆𝒛, 𝒕 𝒇 − ∆𝒕 = 𝑭 𝒌 𝒙𝒑, 𝒌 𝒛𝒒, 𝒕 𝒇 . 𝐞𝐱𝐩 −𝒊 𝒌 𝒙𝒑. 𝒎∆𝒙 + 𝒌 𝒛𝒒. 𝒏∆𝒛 − 𝝎 𝒑𝒒(∆𝒕)𝑵 𝒒=𝟎 𝑴 𝒑=𝟎 Take 𝑡′ = 𝑡𝑓 𝑎𝑛𝑑 𝑡 = 𝑡𝑓 − ∆𝑡 removing some shift ∆𝒕 in time RTM Migration Sources: Loewenthal, 1983
  • 17. RTM problem solved: backward time marching schemes Loewenthal synthetic applications: • dipping layer 29° • dipping layer 33° • dipping layer 35° • syncline Ok errror < 1% OK error <1% Ok error 5% OK Sources: Loewenthal, 1983 Page 13 of 14
  • 18. RTM problem solved: backward time marching schemes Overall RTM do work well ! Sources: Loewenthal, 1983 Page 13 of 14
  • 19. RTM : Conclusion RTM advantages: • Used today (in prestack mode) to improve imaging in complex geology (e.g Salt Dome) • Resulting images are usually easier to interpret • Can handle severe combination of structural dip (steep and even overturned reflectors) with high velocity contrast RTM disadvantages: • Computer intensive (was ?? ) Sources: Ion, 2015 // CGG, 2015 // Paradigm, 2015 // CSEG, recorder, 2009 Page 14 of 14
  • 20. RTM : Conclusion Sources: Ion, 2015 ? Page 14 of 14
  • 21. RTM : Conclusion Sources: Ion, 2015 Page 14 of 14
  • 22. 22 Thanks for your attention
  • 23. RTM : Conclusion Sources: Ion, 2015 Page 14 of 14
  • 24. What is Migration: an example Page 4 of 14 Sources: Yilmaz, 1987 Focusing effect of a sharp syncline Migration algorithm
  • 25. What is Migration ? Page 4 of 11 Post Stack Migration = Dealing with circles 𝑇𝑎𝑛 𝛼 = 𝑆𝑖𝑛 (𝛽) Sources: Bancroft, 2007 // CSEG recorder, 2012 • The mathematical process of re-arranging the reflection events at their correct locations in the seismogram • Migration will shortens dipping events and increase the dip • Can be in X-T or X-Z domain and pre-stack or post stack
  • 26. 26 What is Downward Continuation in Depth Sources: Bancroft, 2007
  • 27. 27 What is Downward Continuation in Time Sources: Bancroft, 2007
  • 28. 28 What is Phase Shift Sources: Bancroft, 2007
  • 29. 29 Sources: CSEG recorder, 2012 What is Migration ?
  • 30. RTM Assumptions: backward time marching schemes From 𝑡 = 𝑡𝑓 = 𝑡′ 𝑡𝑜 𝑡 = 𝑡0  Take successive steps BACKWARD in TIME until the desired migration is achieved Page 9 of 11 Sources: Loewenthal, 1983 // Levin, 1985
  • 31. RTM problem solved: backward time marching schemes Page 10 of 11 ∆𝑡
  • 32. 32