From Imaging to Inversion             Ian F. Jones      SEG 2012 Honorary Lecture
Acknowledgements     Society of Exploration Geophysicists     Shell Sponsorship     ION GX Technology
SEG Membership SEG Digital Library - full text articles Technical Journals in Print and Online Networking Opportunities...
Student Opportunities Student Chapters available  Student Chapter Book Program  SEG/Chevron Student Leadership Symposiu...
From Imaging to Inversion             Ian F. Jones      SEG 2012 Honorary Lecture
But before I start…
But before I start…A big thanks to Judy Wall at the   SEG for her sterling work   organizing my schedule !
Talk OutlineTo a large extent, this presentation isspeculative, in that I’m looking at what ‘mightcome next’ moving beyond...
Talk Outline•   Hydrocarbon exploration•   Subsurface Imaging•   Waves versus rays•   Velocity model building•   Migration...
What is itthat hydrocarbon explorationgeoscientists set out to do …
Find oil and gas !
But how ?
Drill here ???
… or here ???
How do we decide where to drill?
How do we decide where to drill?… we use sound waves reflecting of the rocklayers to make pictures (similar to ultrasoundm...
The process currently involves several keystages:1) Removal of noise and undesired signal2) Velocity model building3) Migr...
The process currently involves several keystages:1) Removal of noise and undesired signal2) Velocity model building3) Migr...
Attribute estimationOnce we have estimated the speed of sound(velocity) in the different rock layers, and thenformed an im...
The geophysical problem                     We need to relocate recordedV1(x,y,z)            energy to its ‘true’ position...
What do these images ofthe subsurface look like?
Southern North Sea exampleImage dimensions are typically several hundred squarekilometres in area, extending to several ki...
Near-surface buried riverchannel, which distorts thedeeper image (unlesscorrectly dealt with)
How do we describe the way in which  sound travels through the earth?
Waves versus Rays
Waves versus RaysThe theoretical description of wave phenomenafalls into two categories:Ray-basedandWave- (diffraction or ...
A propagating wavefront…we can characterise its direction ofmotion, and speed, with a successionof normal vectors, constit...
A propagating wavefront…we can characterise its direction ofmotion, and speed, with a successionof normal vectors, constit...
A propagating wavefront…we can characterise its direction ofmotion, and speed, with a successionof normal vectors, constit...
A propagating wavefront…we can characterise its direction ofmotion, and speed, with a successionof normal vectors, constit...
A propagating wavefront…we can characterise its direction ofmotion, and speed, with a successionof normal vectors, constit...
A propagating wavefront…we can characterise its direction ofmotion, and speed, with a successionof normal vectors, constit...
Snell’s law at a flat interface                 θi                           Sinθi = Sinθr        vi                   vi ...
The high frequency approximation                            Seismic wavelength much                            smaller tha...
The high frequency approximation                              Seismic wavelength larger or                              si...
A propagating wavefront…Time = t             Time = t + 25ms      Time = t + 50msThe elements of some velocity features be...
Velocity Model Building
Common midpointsource                     receiver              CMP v1                                       Vrms1 v2     ...
CMP                                             Common midpoint gather                                                    ...
CMP                                             Common midpoint gather                                                    ...
 To estimate velocity for flat layers….
Conventional velocity analysis…..      0         Km         53.8S4.7          Input CMP data
Conventional velocity analysis…..      0         Km         53.8                 ΣS4.7          Input CMP data
Conventional velocity analysis…..      0         Km         5      Km          5 1.5     2.5     3.5                      ...
 To estimate velocity for dipping layers….
 To estimate velocity for dipping layers….  The notion of the CMP no longer has any meaning,  as the mid-points do not si...
Dipping layers         Common midpointsource                     receiver              CMP v1                             ...
 To estimate velocity for dipping layers….  The notion of the CMP no longer has any meaning,  as the mid-points do not si...
Tomographic velocity update…..Trace raypaths through the current version ofthe model and note arrival times
Tomographic velocity update…..Picks of reflection event   arrival times synthesizedarrival times from the      from ray tr...
Tomographic velocity update…..  Tomography iteratively modifies the  velocity model so as to minimize the  difference betw...
Iterative update                (1) PreSDM                                 (2) Autopickersmooth initial model output migra...
Iterative update                (1) PreSDM                                 (2) Autopickersmooth initial model output migra...
Iteration 1, 3D preSDM0                             Top Chalkkm2
Iteration 2, 3D preSDM0                             Top Chalkkm2
Iteration 3, 3D preSDM0                             Top Chalkkm2
Iteration 1 Velocities0km2
Iteration 2 Velocities0km2
Iteration 3 Velocities0km2
Migration:putting the recorded databack where it came from
Common midpointsource                     receiver              CMP v1                                       Vrms1 v2     ...
Common midpointsource                     receiver              CMP v1                                       Vrms1 v2     ...
Plot all the traces from various common midpoints toform a picture of the subsurface…
Common midpointSource                     Geophone             CMP                               tA                       ...
Common midpointSource                         Geophone             CMP                                     tA             ...
Main migration algorithms in use today       - Kirchhoff Ray   - Beam       - (GB, CRAM, CRS, CFP, ….)       - Wavefield e...
Migration algorithms relocate recorded energy to its ‘true’ position using an appropriate approximate solution to the two-...
Migration algorithms Primarily, the degree of approximation relates to how well the algorithm comprehends lateral velocity...
Migration algorithms Primarily, the degree of approximation relates to how well the algorithm comprehends lateral velocity...
Migration algorithms Primarily, the degree of approximation relates to how well the algorithm comprehends lateral velocity...
Migration algorithms Primarily, the degree of approximation relates to how well the algorithm comprehends lateral velocity...
1km           Velocity-depth model1490 m/s1600 m/s2000 m/s2200 m/s3500 m/s
Acoustic shot gather             3km   6km                            1sEnergy travelling in thewater (the ‘direct’ wave)R...
1km           Velocity-depth model1490 m/s1600 m/s2000 m/s2200 m/s3500 m/s
1km           preSDM1500 m/s1600 m/s2000 m/s2200 m/s3500 m/s
1km           preSTM1500 m/s (converted to depth)1600 m/s2000 m/s2200 m/s3500 m/s
Migration Issues:Lateral velocity variation:         Kirchhoff preSTM      vs Kirchhoff preSDM      vs RTM      Norwegian ...
Interval velocity modelAutopicking @50*50m   1km    Courtesy of ConocoPhillips NorwayTomo @250*250*50m
Kirchhoff preSTM (initial model)               1km      Courtesy of ConocoPhillips Norway
Kirchhoff preSDMAutopicking @50*50m   1km   Courtesy of ConocoPhillips NorwayTomo @250*250*50m
RTMAutopicking @50*50m   1km   Courtesy of ConocoPhillips NorwayTomo @250*250*50m
Migration Issues:In addition to the degree oflateral velocity change, we alsohave the issue of ray-pathcomplexity to consi...
Migration Issues:Multi-pathing:
What is multi-pathing? There is more than one path from a surface location to a subsurface point                          ...
Migration Issues:Multi-pathing:       Kirchhoff vs WEM      North Sea shallow water diapir example
1km          salt2km46          Vi(z)
1km2km46          Anisotropic Kirchhoff 3D preSDM
1km2km46          Anisotropic one-way SSFPI (WEM) 3D preSDM
Migration Issues:Two-way propagation:
What is two-way propagation?Conventional one-way propagation           Two-way propagation: requires a moreas assumed by s...
Migration Issues:Two way ray paths:     WEM vs RTM      North Sea shallow water diapir example
1km2km46          Anisotropic one-way SSFPI (WEM) 3D preSDM
1km2km46          Anisotropic two-way RTM 3D preSDM
1km2km46          Anisotropic two-way RTM 3D preSDM
Migration Issues:Two way ray paths:     WEM vs RTM      West African deep water diapir example
WEM       1km 1km
RTM       1km 1km
RTM       1km 1km
RTM       1km 1km
RTM       1km 1km
Once we have estimated velocity, andmigrated the data to obtain gathers intheir correct spatial location, we can     begin...
Extracting other rock attributes       (as well as velocity):    rock type, fluid type, density,saturation, pressure, atte...
Rock physics basics:(for isotropic materials)Stress (pressure) = force/area = F/AStrain = fractional change in volume = dV...
Common midpoint gather                                                   CMP                 t1   t2   t3   t4   t5For a C...
Common image gather                                                 CIG or CRP               t1   t2   t3   t4   t5       ...
Having obtained estimates of velocity:we can then estimate other parameters fromamplitude behaviour
Gathers output from preSDM - not exactly flat
After final residual event alignment and noise suppression                                        These data are          ...
After final residual event alignment and noise suppression                                        These data are          ...
After final residual event alignment and noise suppression                                        These data are          ...
Incident P wave                 Transmitted P wave                 Reflected P wave     textThe Knott-Zoeppritz equations ...
Rock physics basics:(for isotropic materials)                                     θ                                Vp     ...
3D preSDM Showing AVO Anomalies Over Producing Fields Near stack                   Far stack               AVO angle stack...
3D preSDM Showing AVO Anomalies Over Producing FieldsNear stack (0º-25º)               Far stack (25º-50º)      Average ab...
MacCulloch    15/24b-6 Far-angle stack EI InversionN                   S     W                          E                 ...
N                          15/25b-3                     S    Top Balder                               Brenda              ...
Unconventional (tight) reservoir - China     PS seismic line (PS time) through main producing wells                      P...
Unconventional (tight) reservoir - China     Characterizing Lithological Variations            shale               sandRec...
Unconventional (tight) reservoir - China     Full-wave explains well productivity – fracture characterization             ...
What we’ve reviewed so far, hasbeen the ‘state of the art’:    1) velocity model building    2) migration    3) attribute ...
What next?Can we do all this in one step?= full elastic waveform inversion
To accomplish this task, we must accuratelymodel the behaviour of the recorded data:
To accomplish this task, we must accuratelymodel the behaviour of the recorded data:- we start with initial estimates of t...
To accomplish this task, we must accuratelymodel the behaviour of the recorded data:- we start with initial estimates of t...
To accomplish this task, we must accuratelymodel the behaviour of the recorded data:- we start with initial estimates of t...
Real shot       Modelled shot            -                   = residual
Recall the conventional approach:(Tomographic velocity update)….. Tomography iteratively modifies the velocity model so as...
Waveform inversion update….. Waveform inversion iteratively modifies the parameter model so as to minimize the difference ...
What’s involved in getting the amplitude right?  -Visco elastic wave propagation (incorporates   attenuation and shear mod...
Ignoring density Reflection strength (amplitude) is related to impedance contrast: (ρ2v2 – ρ1v1)/(ρ2v2 + ρ1v1) By ignoring...
Where are the refractions? Perform some forward modelling to assess how deeply the diving waves penetrate The region of va...
Raytracing to show turning-ray paths- expected maximum depth of WFI update                             10km cable         ...
Wave modelling to show turning-ray pathsSnapshot (t=33ms)
Wave modelling to show turning-ray pathsSnapshot (t=1407ms)
Wave modelling to show turning-ray pathsSnapshot (t=1865ms)
Wave modelling to show turning-ray pathsSnapshot (t=2454ms)
Wave modelling to show turning-ray pathsSnapshot (t=3272ms)         Max Depth of Turning Rays ~3400m                 for c...
Do we obtain a better earth-model parameters? One way to confirm if FWI has produced better earth- model parameters is to ...
Gathers migrated with ray-tomography velocities            Courtesy of Chow Wang, GXT
Gathers migrated with waveform inversion velocities             Courtesy of Chow Wang, GXT
Shallow Section Before WFI            Courtesy of Chow Wang, GXT
Shallow Section After WFI            Courtesy of Chow Wang, GXT
BP in-house project: Valhall(courtesy of Jan Kommedal & Laurent Sirgue)                                   Courtesy of BP N...
BP ValhallRay tomography velocity modelWaveform inversion velocity model   Courtesy of BP Norway
BP Valhall: ray-based tomography                       Courtesy of BP Norway
BP Valhall: waveform tomography                      Courtesy of BP Norway
BP Valhall: waveform tomography                      Courtesy of BP Norway
175m depth slice of preSDM amplitudes                                    Courtesy of BP Norway
175m depth slice of FWI velocity                                   Courtesy of BP Norway
BP Valhall: 150m velocity slice                      Courtesy of BP Norway
BP Valhall: 150m velocity slice                      Courtesy of BP Norway
BP Valhall: 1050m velocity slice                      Courtesy of BP Norway
BP Valhall: 1050m velocity slice                      Courtesy of BP Norway
The ultimate goal of full waveform inversion…. At present, the limiting assumptions we make in waveform inversion limit wh...
The ultimate goal of full waveform inversion…. IFF we can move beyond the present limiting assumptions, then we may be abl...
The ultimate goal of full waveform inversion….    Vp         Vs          ρ        ε            δ                        mo...
The ultimate goal of full waveform inversion….    Courtesy of Joachim Mispel & Ina Wenske
The ultimate goal of full waveform inversion….                                                  Vp                        ...
In other words ….
Move from thislengthy disjointedprocess……
Extensive data pre-processing (remove multiples)                               Move from this                             ...
Extensive data pre-processing (remove multiples)                                Move from this                            ...
Extensive data pre-processing (remove multiples)                                Move from this                            ...
Extensive data pre-processing (remove multiples)                                Move from this                            ...
To this ……
CLEAN INPUT DATA(including multiples)                              To this ……             FWI                 rock        ...
But perhaps we shouldn’t    ‘hold our breath’        just yet !
Thank you !
Upcoming SlideShare
Loading in …5
×

Від побудови сейсмічних зображень до інверсії

1,118 views

Published on

У рамках програми «Підвищення кваліфікації фахівців нафтогазової галузі України для міжнародного співробітництва та роботи у західних компаніях», за підтримки компанії «Shell» 6 березня в аудиторії ВНЗ «Інститут Тутковського» відбулися курси підвищення кваліфікації на тему «Від побудови сейсмічних зображень до інверсії».

Published in: Education, Technology, Business
0 Comments
1 Like
Statistics
Notes
  • Be the first to comment

No Downloads
Views
Total views
1,118
On SlideShare
0
From Embeds
0
Number of Embeds
18
Actions
Shares
0
Downloads
138
Comments
0
Likes
1
Embeds 0
No embeds

No notes for slide

Від побудови сейсмічних зображень до інверсії

  1. 1. From Imaging to Inversion Ian F. Jones SEG 2012 Honorary Lecture
  2. 2. Acknowledgements  Society of Exploration Geophysicists  Shell Sponsorship  ION GX Technology
  3. 3. SEG Membership SEG Digital Library - full text articles Technical Journals in Print and Online Networking Opportunities Receive Membership Discounts on:  Continuing Education Courses  Publications (35% off list price)  Workshops and MeetingsSEG materials are available today!Join Online    http://seg.org/join
  4. 4. Student Opportunities Student Chapters available  Student Chapter Book Program  SEG/Chevron Student Leadership Symposium  Challenge Bowl Student Membership Resources  Scholarships  SEG/ExxonMobil Student Education Program  Annual Meeting Travel Grants  Student Expos  the Anomaly newsletter More information please visit:    http://seg.org/students
  5. 5. From Imaging to Inversion Ian F. Jones SEG 2012 Honorary Lecture
  6. 6. But before I start…
  7. 7. But before I start…A big thanks to Judy Wall at the SEG for her sterling work organizing my schedule !
  8. 8. Talk OutlineTo a large extent, this presentation isspeculative, in that I’m looking at what ‘mightcome next’ moving beyond the current industrialpractise of:- data pre-conditioning (multiple suppression),- velocity model building,- migrating data and then- analysing amplitude information….
  9. 9. Talk Outline• Hydrocarbon exploration• Subsurface Imaging• Waves versus rays• Velocity model building• Migration• Attribute estimation• Full waveform inversion
  10. 10. What is itthat hydrocarbon explorationgeoscientists set out to do …
  11. 11. Find oil and gas !
  12. 12. But how ?
  13. 13. Drill here ???
  14. 14. … or here ???
  15. 15. How do we decide where to drill?
  16. 16. How do we decide where to drill?… we use sound waves reflecting of the rocklayers to make pictures (similar to ultrasoundmedical imaging) and then analyse the amplitudebehaviour of the data to infer what types of rocksand fluids are present
  17. 17. The process currently involves several keystages:1) Removal of noise and undesired signal2) Velocity model building3) Migration4) Attribute estimation
  18. 18. The process currently involves several keystages:1) Removal of noise and undesired signal2) Velocity model building3) Migration4) Attribute estimation
  19. 19. Attribute estimationOnce we have estimated the speed of sound(velocity) in the different rock layers, and thenformed an image from the recorded data(‘migration’), we can analyse the amplitudes ofthe reflections to estimate rock properties(which helps us distinguish between oil, gas,water, etc)
  20. 20. The geophysical problem We need to relocate recordedV1(x,y,z) energy to its ‘true’ position using an appropriate approximate solution to the visco-elasticV2(x,y,z) two-way wave equationetc target location (and what is ‘appropriate’, depends on our objectives)
  21. 21. What do these images ofthe subsurface look like?
  22. 22. Southern North Sea exampleImage dimensions are typically several hundred squarekilometres in area, extending to several kilometres depthMigrated image Sea bed 30km chalk 30km 1600m/s 1800m/s 2000m/s 3000m/s 3500m/s 3.5 km anhydriteGas-bearing layers salt Sound speed in the rocks
  23. 23. Near-surface buried riverchannel, which distorts thedeeper image (unlesscorrectly dealt with)
  24. 24. How do we describe the way in which sound travels through the earth?
  25. 25. Waves versus Rays
  26. 26. Waves versus RaysThe theoretical description of wave phenomenafalls into two categories:Ray-basedandWave- (diffraction or scattering) based- Both migration and model update depend onone or other of these paradigms
  27. 27. A propagating wavefront…we can characterise its direction ofmotion, and speed, with a successionof normal vectors, constituting ‘rays’Time = t Time = t + 25ms
  28. 28. A propagating wavefront…we can characterise its direction ofmotion, and speed, with a successionof normal vectors, constituting ‘rays’Time = t Time = t + 25ms
  29. 29. A propagating wavefront…we can characterise its direction ofmotion, and speed, with a successionof normal vectors, constituting ‘rays’Time = t Time = t + 25ms
  30. 30. A propagating wavefront…we can characterise its direction ofmotion, and speed, with a successionof normal vectors, constituting ‘rays’Time = t Time = t + 25ms
  31. 31. A propagating wavefront…we can characterise its direction ofmotion, and speed, with a successionof normal vectors, constituting ‘rays’Time = t Time = t + 25ms
  32. 32. A propagating wavefront…we can characterise its direction ofmotion, and speed, with a successionof normal vectors, constituting ‘rays’Time = t Time = t + 25ms
  33. 33. Snell’s law at a flat interface θi Sinθi = Sinθr vi vi vr vr θr θr = Sin-1( vr Sinθi ) vi
  34. 34. The high frequency approximation Seismic wavelength much smaller than the anomaly we are trying to resolve Velocity anomaly The propagating wavefront can adequately be described by ray-pathsSnell’s law adequately describes the wave propagation… ray-based methods (Kirchhoff, beam, …) are OK
  35. 35. The high frequency approximation Seismic wavelength larger or similar to the anomaly we are trying to resolve Small scale-length velocity anomaly The velocity feature behaves more like a scatterer than a simple refracting surface element Trying to describe the propagation behaviour as ‘rays’ obeying Snell’s law, is no longer appropriateRay-based methods (Kirchhoff, beam, …) using the ‘highfrequency approximation’ begin to fail
  36. 36. A propagating wavefront…Time = t Time = t + 25ms Time = t + 50msThe elements of some velocity features behave more likepoint scatterers producing secondary wavefronts
  37. 37. Velocity Model Building
  38. 38. Common midpointsource receiver CMP v1 Vrms1 v2 Vrms2v3 Vrms3 v4 Vrms4
  39. 39. CMP Common midpoint gather CMP t1 t2 t3 t4 t5For a CMP gather, we have manyarrival time measurements for agiven subsurface reflectorelement
  40. 40. CMP Common midpoint gather CMP t1 t2 t3 t4 t5For a CMP gather, we have manyarrival time measurements for agiven subsurface reflectorelement This curvature is related to the velocity
  41. 41.  To estimate velocity for flat layers….
  42. 42. Conventional velocity analysis….. 0 Km 53.8S4.7 Input CMP data
  43. 43. Conventional velocity analysis….. 0 Km 53.8 ΣS4.7 Input CMP data
  44. 44. Conventional velocity analysis….. 0 Km 5 Km 5 1.5 2.5 3.5 Km/s3.8S4.7 Input CMP data scan along pick corresponding trajectories velocity
  45. 45.  To estimate velocity for dipping layers….
  46. 46.  To estimate velocity for dipping layers…. The notion of the CMP no longer has any meaning, as the mid-points do not sit above the same subsurface location for all offsets
  47. 47. Dipping layers Common midpointsource receiver CMP v1 Vrms1 Vrms2 v2 Vrms3v3 Vrms4
  48. 48.  To estimate velocity for dipping layers…. The notion of the CMP no longer has any meaning, as the mid-points do not sit above the same subsurface location for all offsets We have to assess the travel times for each offset separately
  49. 49. Tomographic velocity update…..Trace raypaths through the current version ofthe model and note arrival times
  50. 50. Tomographic velocity update…..Picks of reflection event arrival times synthesizedarrival times from the from ray tracing throughreal data the current velocity model
  51. 51. Tomographic velocity update….. Tomography iteratively modifies the velocity model so as to minimize the difference between observed arrival times on the real data, and ray-traced times through the current velocity model
  52. 52. Iterative update (1) PreSDM (2) Autopickersmooth initial model output migrated gathers Using continuous CRPs, calculate semblance, velocity & anisotropy (6) Interpretation (if required) error grids, & RMO stackPick constraint layer, insert ‘flood’ velocity, and migrate (3) TTI Tomography compute dip field demigrate picks & RMO stack update TTI velocity field remigrate picks & RMO stack (5) PreSDM with updated velocity No (4) Inversion QC (4) RMO & z Residual velocity error minimised? Yes (gathers flat) Depth error acceptable? acceptable?Final Volume
  53. 53. Iterative update (1) PreSDM (2) Autopickersmooth initial model output migrated gathers Using continuous CRPs, calculate semblance, velocity & anisotropy (6) Interpretation (if required) error grids, & RMO stackPick constraint layer, insert ‘flood’ velocity, and migrate This process (3) TTI Tomography usually involves demigrate picksdipRMO stack compute & field 6-8 iterations update TTI velocity field remigrate picks & RMO stack (5) PreSDM with updated velocity No (4) Inversion QC (4) RMO & z Residual velocity error minimised? Yes (gathers flat) Depth error acceptable? acceptable?Final Volume
  54. 54. Iteration 1, 3D preSDM0 Top Chalkkm2
  55. 55. Iteration 2, 3D preSDM0 Top Chalkkm2
  56. 56. Iteration 3, 3D preSDM0 Top Chalkkm2
  57. 57. Iteration 1 Velocities0km2
  58. 58. Iteration 2 Velocities0km2
  59. 59. Iteration 3 Velocities0km2
  60. 60. Migration:putting the recorded databack where it came from
  61. 61. Common midpointsource receiver CMP v1 Vrms1 v2 Vrms2v3 Vrms3 v4 Vrms4
  62. 62. Common midpointsource receiver CMP v1 Vrms1 v2 Vrms2v3 Vrms3 v4 Vrms4
  63. 63. Plot all the traces from various common midpoints toform a picture of the subsurface…
  64. 64. Common midpointSource Geophone CMP tA Reflector segment A B tA
  65. 65. Common midpointSource Geophone CMP tA Reflector segment A B tA ‘Migration’ moves the recorded data back to where it came from
  66. 66. Main migration algorithms in use today - Kirchhoff Ray - Beam - (GB, CRAM, CRS, CFP, ….) - Wavefield extrapolation (WEM)Wave - Reverse-Time (two-way)
  67. 67. Migration algorithms relocate recorded energy to its ‘true’ position using an appropriate approximate solution to the two-way visco-elastic wave equation (but what is ‘appropriate’, depends on our objectives)
  68. 68. Migration algorithms Primarily, the degree of approximation relates to how well the algorithm comprehends lateral velocity change
  69. 69. Migration algorithms Primarily, the degree of approximation relates to how well the algorithm comprehends lateral velocity change No Smooth Rapid lateral lateral lateral velocity velocity velocity change change change
  70. 70. Migration algorithms Primarily, the degree of approximation relates to how well the algorithm comprehends lateral velocity change No Smooth Rapid lateral lateral lateral velocity velocity velocity change change change Time Ray-based and RTM migration low-order FD (high-order FD) depth migration depth migration
  71. 71. Migration algorithms Primarily, the degree of approximation relates to how well the algorithm comprehends lateral velocity change No Smooth Rapid lateral lateral lateral velocity velocity velocity change change change Time Ray-based and RTM migration low-order FD (high-order FD) depth migration depth migration simple ray-paths complex ray-paths
  72. 72. 1km Velocity-depth model1490 m/s1600 m/s2000 m/s2200 m/s3500 m/s
  73. 73. Acoustic shot gather 3km 6km 1sEnergy travelling in thewater (the ‘direct’ wave)Reflection fromwater bottom 3sReflections from 4sdeeper rock layers 5s
  74. 74. 1km Velocity-depth model1490 m/s1600 m/s2000 m/s2200 m/s3500 m/s
  75. 75. 1km preSDM1500 m/s1600 m/s2000 m/s2200 m/s3500 m/s
  76. 76. 1km preSTM1500 m/s (converted to depth)1600 m/s2000 m/s2200 m/s3500 m/s
  77. 77. Migration Issues:Lateral velocity variation: Kirchhoff preSTM vs Kirchhoff preSDM vs RTM Norwegian Sea shallow water gas example
  78. 78. Interval velocity modelAutopicking @50*50m 1km Courtesy of ConocoPhillips NorwayTomo @250*250*50m
  79. 79. Kirchhoff preSTM (initial model) 1km Courtesy of ConocoPhillips Norway
  80. 80. Kirchhoff preSDMAutopicking @50*50m 1km Courtesy of ConocoPhillips NorwayTomo @250*250*50m
  81. 81. RTMAutopicking @50*50m 1km Courtesy of ConocoPhillips NorwayTomo @250*250*50m
  82. 82. Migration Issues:In addition to the degree oflateral velocity change, we alsohave the issue of ray-pathcomplexity to consider in themigration…
  83. 83. Migration Issues:Multi-pathing:
  84. 84. What is multi-pathing? There is more than one path from a surface location to a subsurface point saltA Kirchhoff scheme usually only computes travel times forone ray path… what happens to the energy from the restof the ray paths from input data?
  85. 85. Migration Issues:Multi-pathing: Kirchhoff vs WEM North Sea shallow water diapir example
  86. 86. 1km salt2km46 Vi(z)
  87. 87. 1km2km46 Anisotropic Kirchhoff 3D preSDM
  88. 88. 1km2km46 Anisotropic one-way SSFPI (WEM) 3D preSDM
  89. 89. Migration Issues:Two-way propagation:
  90. 90. What is two-way propagation?Conventional one-way propagation Two-way propagation: requires a moreas assumed by standard migration complete solution of the wave equationschemes to migrate such arrivals Nor from the reflection point back up to the surface The direction of propagation changes either on the way down from the surfaceNo change in propagation to the reflection point, or from thedirection on the way from reflection point back up to the surfacethe surface down to thereflection point
  91. 91. Migration Issues:Two way ray paths: WEM vs RTM North Sea shallow water diapir example
  92. 92. 1km2km46 Anisotropic one-way SSFPI (WEM) 3D preSDM
  93. 93. 1km2km46 Anisotropic two-way RTM 3D preSDM
  94. 94. 1km2km46 Anisotropic two-way RTM 3D preSDM
  95. 95. Migration Issues:Two way ray paths: WEM vs RTM West African deep water diapir example
  96. 96. WEM 1km 1km
  97. 97. RTM 1km 1km
  98. 98. RTM 1km 1km
  99. 99. RTM 1km 1km
  100. 100. RTM 1km 1km
  101. 101. Once we have estimated velocity, andmigrated the data to obtain gathers intheir correct spatial location, we can begin to analyse amplitude information
  102. 102. Extracting other rock attributes (as well as velocity): rock type, fluid type, density,saturation, pressure, attenuation, ….
  103. 103. Rock physics basics:(for isotropic materials)Stress (pressure) = force/area = F/AStrain = fractional change in volume = dV/VBulk modulus = pressure/strain = B = - (F/A)/(dV/V)Compressibility = 1/B B = λ + 2/3 μ
  104. 104. Common midpoint gather CMP t1 t2 t3 t4 t5For a CMP gather, we have manyarrival time measurements for agiven subsurface reflectorelement
  105. 105. Common image gather CIG or CRP t1 t2 t3 t4 t5 offsetAfter depth migration with anacceptable velocity model, allevents in the gather shouldline-up  ‘flat gathers’ Migrated depth
  106. 106. Having obtained estimates of velocity:we can then estimate other parameters fromamplitude behaviour
  107. 107. Gathers output from preSDM - not exactly flat
  108. 108. After final residual event alignment and noise suppression These data are now suitable for analyzing variations in amplitude:
  109. 109. After final residual event alignment and noise suppression These data are now suitable for analyzing variations in amplitude: vertically from reflector-to- reflector: (ρ2v2 – ρ1v1)/(ρ2v2 + ρ1v1)
  110. 110. After final residual event alignment and noise suppression These data are now suitable for analyzing variations in amplitude: vertically from reflector-to- reflector and laterally versus incidence angle at the reflectors
  111. 111. Incident P wave Transmitted P wave Reflected P wave textThe Knott-Zoeppritz equations relate theamplitude change as a function of incidentangle, to Vp, Vs, and density
  112. 112. Rock physics basics:(for isotropic materials) θ Vp Vp+δVp
  113. 113. 3D preSDM Showing AVO Anomalies Over Producing Fields Near stack Far stack AVO angle stack synthetics
  114. 114. 3D preSDM Showing AVO Anomalies Over Producing FieldsNear stack (0º-25º) Far stack (25º-50º) Average absolute amplitude Top Balder +50 - +200
  115. 115. MacCulloch 15/24b-6 Far-angle stack EI InversionN S W E 15/24b-6 15/24b-6 Top Balder Low EI Oil 650 Sand 600 550 500 450
  116. 116. N 15/25b-3 S Top Balder Brenda Field 650 600 Possible low EI Oil Sand on flank? 550 500 15/25b-3 Far-stack Inversion (inline) 450
  117. 117. Unconventional (tight) reservoir - China PS seismic line (PS time) through main producing wells Productive Interval Zone of interest11
  118. 118. Unconventional (tight) reservoir - China Characterizing Lithological Variations shale sandRecord P-wave onlyImpute shear-wave measurementusing simultaneous inversion (AVO)Attempt to infer sand-shale variations Record shear-waves directly More accurate depiction of sand-shale variations11
  119. 119. Unconventional (tight) reservoir - China Full-wave explains well productivity – fracture characterization Same lithology Note presence of No fractures fractures in No production producing zone New well location11
  120. 120. What we’ve reviewed so far, hasbeen the ‘state of the art’: 1) velocity model building 2) migration 3) attribute estimation
  121. 121. What next?Can we do all this in one step?= full elastic waveform inversion
  122. 122. To accomplish this task, we must accuratelymodel the behaviour of the recorded data:
  123. 123. To accomplish this task, we must accuratelymodel the behaviour of the recorded data:- we start with initial estimates of the rock physicsparameters (P-wave velocity, S-wave velocity,density, anisotropy, absorption, ..)
  124. 124. To accomplish this task, we must accuratelymodel the behaviour of the recorded data:- we start with initial estimates of the rock physicsparameters (P-wave velocity, S-wave velocity,density, anisotropy, absorption, ..)- make synthetic data and compare it to the realdata
  125. 125. To accomplish this task, we must accuratelymodel the behaviour of the recorded data:- we start with initial estimates of the rock physicsparameters (P-wave velocity, S-wave velocity,density, anisotropy, absorption, ...)- make synthetic data and compare it to the realdata- iteratively adjust the parameters until modelledand real data match
  126. 126. Real shot Modelled shot - = residual
  127. 127. Recall the conventional approach:(Tomographic velocity update)….. Tomography iteratively modifies the velocity model so as to minimize the difference between observed arrival times on the real data, and ray-traced times through the current velocity model
  128. 128. Waveform inversion update….. Waveform inversion iteratively modifies the parameter model so as to minimize the difference between observed amplitudes on the real data, and modelled amplitudes created using the current parameter model
  129. 129. What’s involved in getting the amplitude right? -Visco elastic wave propagation (incorporates attenuation and shear modes) -Elastic wave propagation (shear modes) -Acoustic wave propagation (P-wave only, thus ignoring density) -Anisotropy -Source wavelet (and are ghosts present?) -Source wavelet time delay -Cycle skipping (offset and frequency dependent)
  130. 130. Ignoring density Reflection strength (amplitude) is related to impedance contrast: (ρ2v2 – ρ1v1)/(ρ2v2 + ρ1v1) By ignoring density, we are saying that impedance is only a function of P velocity: Thus, if we invert using reflection events, we will have an amplitude error So, to avoid this error perhaps use only refractions (diving, turning waves)
  131. 131. Where are the refractions? Perform some forward modelling to assess how deeply the diving waves penetrate The region of validity of the model update will be related to this depth of penetration
  132. 132. Raytracing to show turning-ray paths- expected maximum depth of WFI update 10km cable H2OObserved depth ofupdate Insert Velocity Model Here with Rays for Cable we are using Maximum expected depth of WFI update Ray tracing performed in tomography derived sediment flood model
  133. 133. Wave modelling to show turning-ray pathsSnapshot (t=33ms)
  134. 134. Wave modelling to show turning-ray pathsSnapshot (t=1407ms)
  135. 135. Wave modelling to show turning-ray pathsSnapshot (t=1865ms)
  136. 136. Wave modelling to show turning-ray pathsSnapshot (t=2454ms)
  137. 137. Wave modelling to show turning-ray pathsSnapshot (t=3272ms) Max Depth of Turning Rays ~3400m for cable length
  138. 138. Do we obtain a better earth-model parameters? One way to confirm if FWI has produced better earth- model parameters is to use the FWI velocity to perform a new migration
  139. 139. Gathers migrated with ray-tomography velocities Courtesy of Chow Wang, GXT
  140. 140. Gathers migrated with waveform inversion velocities Courtesy of Chow Wang, GXT
  141. 141. Shallow Section Before WFI Courtesy of Chow Wang, GXT
  142. 142. Shallow Section After WFI Courtesy of Chow Wang, GXT
  143. 143. BP in-house project: Valhall(courtesy of Jan Kommedal & Laurent Sirgue) Courtesy of BP Norway
  144. 144. BP ValhallRay tomography velocity modelWaveform inversion velocity model Courtesy of BP Norway
  145. 145. BP Valhall: ray-based tomography Courtesy of BP Norway
  146. 146. BP Valhall: waveform tomography Courtesy of BP Norway
  147. 147. BP Valhall: waveform tomography Courtesy of BP Norway
  148. 148. 175m depth slice of preSDM amplitudes Courtesy of BP Norway
  149. 149. 175m depth slice of FWI velocity Courtesy of BP Norway
  150. 150. BP Valhall: 150m velocity slice Courtesy of BP Norway
  151. 151. BP Valhall: 150m velocity slice Courtesy of BP Norway
  152. 152. BP Valhall: 1050m velocity slice Courtesy of BP Norway
  153. 153. BP Valhall: 1050m velocity slice Courtesy of BP Norway
  154. 154. The ultimate goal of full waveform inversion…. At present, the limiting assumptions we make in waveform inversion limit what we can achieve: we can currently forward model with a priori parameters for: density, attenuation, anisotropy (and perhaps Vs) but invert only for P-wave velocity
  155. 155. The ultimate goal of full waveform inversion…. IFF we can move beyond the present limiting assumptions, then we may be able to invert so as to update all these parameters thereby recovering density, Vp, Vs, Q, and other parameters. Interpretation would then be performed on these parameter fields directly, rather than on inversions of migrated data obtained using the velocity parameter
  156. 156. The ultimate goal of full waveform inversion…. Vp Vs ρ ε δ model Inversion result Courtesy of Olga Podgornova
  157. 157. The ultimate goal of full waveform inversion…. Courtesy of Joachim Mispel & Ina Wenske
  158. 158. The ultimate goal of full waveform inversion…. Vp Vs/10 Vs Courtesy of Satish Singh
  159. 159. In other words ….
  160. 160. Move from thislengthy disjointedprocess……
  161. 161. Extensive data pre-processing (remove multiples) Move from this lengthy disjointed process……
  162. 162. Extensive data pre-processing (remove multiples) Move from this lengthy disjointed process……+ Iterative velocity model update and migration
  163. 163. Extensive data pre-processing (remove multiples) Move from this lengthy disjointed process……+ Iterative velocity model update and migration + elastic parameter inversion
  164. 164. Extensive data pre-processing (remove multiples) Move from this lengthy disjointed process……+ Iterative velocity model update and migration + elastic parameter inversion + rock property estimation
  165. 165. To this ……
  166. 166. CLEAN INPUT DATA(including multiples) To this …… FWI rock properties
  167. 167. But perhaps we shouldn’t ‘hold our breath’ just yet !
  168. 168. Thank you !

×