Dimensionality Reduction for Seismic Attribute Analysis Bradley C. - - PowerPoint PPT Presentation

▶
dimensionality reduction for seismic attribute analysis
SMART_READER_LITE
LIVE PREVIEW

Dimensionality Reduction for Seismic Attribute Analysis Bradley C. - - PowerPoint PPT Presentation

May 22, 2023 342 likes •970 views

Dimensionality Reduction for Seismic Attribute Analysis Bradley C. Wallet, Ph.D. University of Oklahoma ConocoPhillips School of Geology and Geophysics Where oil is first found is in the minds of men - Wallace Pratt Motivation Motivation

slide-1
SLIDE 1

Dimensionality Reduction for Seismic Attribute Analysis

Bradley C. Wallet, Ph.D. University of Oklahoma ConocoPhillips School of Geology and Geophysics

slide-2
SLIDE 2

Where oil is first found is in the minds’ of men

  • Wallace Pratt
slide-3
SLIDE 3

Motivation

slide-4
SLIDE 4

Motivation

slide-5
SLIDE 5

Outline

  • Seismic data
  • Seismic attributes
  • PCA
  • Image grand tour
  • Non-linear methods
  • Conclusions
  • Acknowledgements
slide-6
SLIDE 6

Outline

  • Seismic data
  • Seismic attributes
  • PCA
  • Image grand tour
  • Non-linear methods
  • Conclusions
  • Acknowledgements
slide-7
SLIDE 7

Why care about seismic data?

  • Single pre-stack data sets can be 10’s – 100’s of terabytes in size
  • Provide good spatial coverage exploration area
  • Used to make high dollar decisions
slide-8
SLIDE 8

Seismic shot

Courtesy of Bin Lyu

slide-9
SLIDE 9

Common midpoint gather

slide-10
SLIDE 10

Migration

slide-11
SLIDE 11

Convolutional model

Velocity Density Impedance

= x

Shale Sand Shale Sand Shale

Lithology Reflection Coefficients

⇒

*

⇒

Wavelet

slide-12
SLIDE 12

Seismic data

(Elebiju et al., 2009)

slide-13
SLIDE 13

Outline

  • Seismic data
  • Seismic attributes
  • PCA
  • Image grand tour
  • Non-linear methods
  • Conclusions
  • Acknowledgements
slide-14
SLIDE 14

These are features

slide-15
SLIDE 15

From one comes many

Seismic data Attribute 2 Attribute 3 Attribute 4 Attribute 5 Attribute 6 Attribute 7 Attribute 8 Attribute 1

slide-16
SLIDE 16

Coherence

inline inline

slide-17
SLIDE 17

Seismic

(Bahorich and Farmer, 1995) 5 km

slide-18
SLIDE 18

Coherence

(Bahorich and Farmer, 1995) salt 5 km

1.0 0.6 Coh

slide-19
SLIDE 19

Spectral decomposition

Synthetic Reflectivity CWT Magnitude Voices

CWT magnitude pos

(Matos and Marfurt, 2011)

Σ

Le Nozze di Figaro

slide-20
SLIDE 20

Spectral decomposition

(Laughlin et al., 2002)

A A′

15 Hz Map

A′ A

30 Hz Map

30 Hz 15 Hz A A′ Time (s)

slide-21
SLIDE 21

Spectral decomposition

18 Hz  Red 24 Hz  Green 36 Hz  Blue

(Bahorich et al., 2002)

slide-22
SLIDE 22

Dip attributes

y z x θy (crossline dip) θx (inline dip) a φ (dip azimuth) θ (dip magnitude) ψ (strike) n

(Marfurt, 2006)

slide-23
SLIDE 23

Dip attributes

Instantaneous dip = dip with highest coherence (Marfurt et al, 1998) Analysis Point

Minimum dip tested (-200) Maximum dip tested (+200) Dip with maximum coherence (+50)

slide-24
SLIDE 24

Dip attributes

Dip Azimuth Hue 180 360 Dip Magnitude Saturation High N E S W (c) 1.2 1.4

(Guo et al., 2008)

slide-25
SLIDE 25

How do we “assimilate” all these attributes?

slide-26
SLIDE 26

Outline

  • Seismic data
  • Seismic attributes
  • PCA
  • Image grand tour
  • Non-linear methods
  • Conclusions
  • Acknowledgements
slide-27
SLIDE 27

PCA

  • Rotates attribute space
  • New dimensions are called principal components
  • Var(pc1) > Var(pc2) > … > Var(pc d)
  • Defines variance as information
slide-28
SLIDE 28

PCA

(Wikapedia)

slide-29
SLIDE 29

Watonga survey

slide-30
SLIDE 30

Complex PCA

slide-31
SLIDE 31

Complex PCA

slide-32
SLIDE 32

PCA

slide-33
SLIDE 33

PCA

slide-34
SLIDE 34

PCA

slide-35
SLIDE 35

Outline

  • Seismic data
  • Seismic attributes
  • PCA
  • Image grand tour
  • Non-linear methods
  • Conclusions
  • Acknowledgements
slide-36
SLIDE 36

Linear projections

Poorly separated Somewhat separated

∑

=

=

d i i i

proj

1

) ( ξ α ξ

Well separated

slide-37
SLIDE 37

The Grand Tour (1750-1880’s)

slide-38
SLIDE 38

Defining the tour

slide-39
SLIDE 39

Image Grand Tour

7.005 10.95

  • 6.215
slide-40
SLIDE 40

View Locked Color IGT

slide-41
SLIDE 41

Outline

  • Seismic data
  • Seismic attributes
  • PCA
  • Image grand tour
  • Non-linear methods
  • Conclusions
  • Acknowledgements
slide-42
SLIDE 42

Latent spaces

slide-43
SLIDE 43

a) b)

N

Cartoon illustration of GTM

Generative topographical maps

slide-44
SLIDE 44

Canterbury Basin, offshore New Zealand

170° 30’ E 173° 00’ E 45° 30’ S 46° 30’ S

(Figure by Origin Energy) (Modified from Mitchell and Neil, 2012)

Waka 3D

slide-45
SLIDE 45

36

Seismic

slide-46
SLIDE 46

Peak Frequency

slide-47
SLIDE 47

38

Peak spectral magnitude

slide-48
SLIDE 48

39

Curvedness

slide-49
SLIDE 49

40

GLCM homogeneity

slide-50
SLIDE 50

41

Co-rendering

slide-51
SLIDE 51

GTM

slide-52
SLIDE 52

Waveforms as attributes

(Wallet et al, 2009)

slide-53
SLIDE 53

Watonga revisited

(Wallet et al, 2009)

slide-54
SLIDE 54

Diffusion maps

Form n-by-n similarity matrix Normalize rows to sum to 1 Perform PCA on diffusion matrix

slide-55
SLIDE 55

Diffusion maps

Advantages

  • Closed form solution
  • Direct calculation of inter-point

distances

  • Not tied to a Euclidean space
  • Eigenvalues

Disadvantages

  • Computationally intractable for

reasonable sized data sets

  • Out of training set data are not

defined in mapping

slide-56
SLIDE 56

Diffusion maps

slide-57
SLIDE 57

Outline

  • Seismic data
  • Seismic attributes
  • PCA
  • Image grand tour
  • Non-linear methods
  • Conclusions
  • Acknowledgements
slide-58
SLIDE 58

Conclusions

  • The human is still the best interpreter we have
  • Attribute overload can overwhelm interpeters
  • Dimensionality reduction produces highly interpretable images
slide-59
SLIDE 59

Acknowledgments

  • Prof. Kurt Marfurt (University of Oklahoma)
  • Mr. Victor Aarre (Schlumberger Norway Technology Center)
  • Mr. Tao Zhao (OU)
  • Dr. Marcilio de Matos (Petrobras)
  • CGG Veritas, Chesapeake Energy, Anadarko Petroleum, and the

Government of New Zealand

slide-60
SLIDE 60

Acknowledgments

slide-61
SLIDE 61

Questions?

bwallet@ou.edu http://geology.ou.edu/aaspi