Seismic Thickness Estimation: Three Approaches, Pros and Cons Gregory A. Partyka bp.

Seismic Thickness Estimation: Three Approaches, Pros and Cons

Gregory A. Partyka

bp

Outline

• IntroductionIntroduction

• Three Approaches

• Examples

• Pros and Cons

Outline

• IntroductionIntroduction


• Examples

• Pros and Cons

Blocky Wedge Model

366

400

Tra

vel T

ime

(m

s)

432

466

REFLECTIVITY

Temporal Thickness (ms)

0 5040302010

Blocky Wedge Model

366

400

Tra

vel T

ime

(m

s)

432

466


REFLECTIVITY

BANDLIMITED REFLECTIVITY (8-10-40-50hz)

+0.025

-0.025

Amplitude

366

400

Tra

vel T

ime

(m

s)

432

466


0 5040302010

0 5040302010

Blocky Wedge Model

366

400

Tra

vel T

ime

(m

s)

432

466


REFLECTIVITY


+0.025

-0.025

Amplitude

366

400

Tra

vel T

ime

(m

s)

432

466


0 5040302010

0 5040302010

Outline

• Introduction

• Three ApproachesThree Approaches

• Examples

• Pros and Cons

Three Approaches to Thickness Estimation

1. Conventional• peak-trough time separation• amplitude

2. Spectral Decomposition• 1st dominant frequency and amplitude

3. Spectral Decomposition• discrete frequency components


1.1. ConventionalConventional• peak-trough time separationpeak-trough time separation• amplitudeamplitude



Conventional Thickness Estimation

366

400

Tra

vel T

ime

(m

s)

432

466


REFLECTIVITY


+0.025

-0.025

Amplitude

366

400

Tra

vel T

ime

(m

s)

432

466


0 5040302010

0 5040302010


366

400

Tra

vel T

ime

(m

s)

432

466


REFLECTIVITY


+0.025

-0.025

Amplitude

366

400

Tra

vel T

ime

(m

s)

432

466


0 5040302010

0 5040302010

tuning thickness

Widess, M.B., 1973, How Thin is a Thin Bed?, Geophysics, vol. 38, pg 1176-1180.

Kallweitt, R.S. and Wood, L.C, 1982, The Limits of Resolution of Zero-Phase Wavelets, Geophysics, vol.47, pg 1035-1046.

-0.0250


366

400

Tra

vel T

ime

(m

s)

432

466


REFLECTIVITY


+0.025

-0.025

Amplitude

366

400

Tra

vel T

ime

(m

s)

432

466


0 5040302010

0 5040302010

0.00

5.00

10.00

15.00

20.00

25.00

30.00

35.00

40.00

45.00

50.00

0 5 10 15 20 25 30 35 40 45 50

-0.02500

-0.02000

-0.01500

-0.01000

-0.00500

0.00000


0 5040302010

50

40

Pe

ak-

Tro

ug

h T

ime

Se

pa

ratio

n (

ms)

30

20

10

0

-0.0000

La

rge

st N

eg

ativ

e A

mp

litu

de-0.0050

-0.0100

-0.0150

45

35

25

15

5

-0.0200

tuning thickness

tuning thickness =1

1.4 * frequencyupper

tuning thickness

Widess, M.B., 1973, How Thin is a Thin Bed?, Geophysics, vol. 38, pg 1176-1180.

Kallweitt, R.S. and Wood, L.C, 1982, The Limits of Resolution of Zero-Phase Wavelets, Geophysics, vol.47, pg 1035-1046.



2.2. Spectral DecompositionSpectral Decomposition• 11stst dominant frequency and amplitude dominant frequency and amplitude


Spectral Decomposition

• uses the discrete Fourier transform to:– quantify thin-bed interference, and– detect subtle discontinuities.

Spectral Interference

Source WaveletAmplitude Spectrum

Thin Bed ReflectionAmplitude Spectrum

Thin BedReflection

ReflectedWavelets

SourceWavelet

Thin Bed

ReflectivityAcousticImpedance

Temporal Thickness

FourierTransform

FourierTransform

Amplitude Amplitude

Fre

quen

cy

Fre

quen

cy

Temporal Thickness1

• The spectral interference pattern is imposed by the distribution of acoustic properties within the short analysis window.

Paryka, Gridley and Lopez, The Leading Edge, vol 18, no 3, 1999

Blocky Wedge Model

366

400

Tra

vel T

ime

(m

s)

432

466

REFLECTIVITY


0 5040302010

0 5040302010

0 5040302010


366

400

Tra

vel T

ime

(m

s)

432

466

50

150

Fre

qu

en

cy (

Hz)

100

200

0

250


REFLECTIVITY

SPECTRAL AMPLITUDE

0.0014

0

Amplitude


0 5040302010

0 5040302010

Spectral Interference and Frequency

366

400

Tra

vel T

ime

(m

s)

432

466

50

150

Fre

qu

en

cy (

Hz)

100

200

0

250


REFLECTIVITY

SPECTRAL AMPLITUDE

0.0014

0

Amplitude


The temporal thickness of the wedge (t), determines the period of notching in the amplitude spectrum (Pf) with respect to frequency.

Temporal Thickness1

Temporal Thickness

Pf = 1 / t

0 5040302010

0 5040302010

Spectral Interference and Thickness

366

400

Tra

vel T

ime

(m

s)

432

466

50

150

Fre

qu

en

cy (

Hz)

100

200

0

250


REFLECTIVITY

SPECTRAL AMPLITUDE

0.0014

0

Amplitude


The value of the frequency component (f), determines the period of notching in the amplitude spectrum (Pt) with respect to bed thickness.

Frequency1 Pt = 1 / f

0 5040302010

0 5040302010


366

400

Tra

vel T

ime

(m

s)

432

466

50

150

Fre

qu

en

cy (

Hz)

100

200

0

250


REFLECTIVITY

SPECTRAL AMPLITUDE

0.0014

0

Amplitude


0 5040302010

0 5040302010

Bandwidth8-10-40-50

366

400

Tra

vel T

ime

(m

s)

432

466


BANDLIMITED SPECTRAL AMPLITUDE

0.0014

0

Amplitude

50

150

Fre

qu

en

cy (

Hz)

100

200

0

250



+0.025

-0.025

Amplitude


Bandwidth8-10-40-50

0 5040302010

0 5040302010

366

400

Tra

vel T

ime

(m

s)

432

466



0.0014

0

Amplitude

25

75

Fre

qu

en

cy (

Hz)

50

100

0

125



+0.025

-0.025

Amplitude

Thickness via 1st Dominant Frequency and Amplitude

Bandwidth8-10-40-50

0 5040302010

0 5040302010

366

400

Tra

vel T

ime

(m

s)

432

466



0.0014

0

Amplitude

25

75

Fre

qu

en

cy (

Hz)

50

100

0

125



+0.025

-0.025

Amplitude

1st DominantFrequency

Frequencyupper and Frequency1st-dominant

= tuning thickness =1

1.4 * frequencyupper

1

2 * frequency1st-dominant



Bandwidth8-10-40-50

0 5040302010

0 5040302010

366

400

Tra

vel T

ime

(m

s)

432

466



0.0014

0

Amplitude

25

75

Fre

qu

en

cy (

Hz)

50

100

0

125



0.0000

0.0005

0.0010

0.0015

0.0020

0.0025

0

5

10

15

20

25

30

35

40


0 5040302010

0.0014

0.0012

1st D

om

ina

nt

Am

plit

ud

e

0.0008

0.0006

0.0002

0.0000

40

35

1st D

om

ina

nt

Fre

qu

en

cy30

25

20

10

5

0

15

+0.025

-0.025

Amplitude

tuning thickness tuning thickness

tuning thickness =1

2 * frequency

0.014sec =1

2 * 36hz

1st-dominant




3.3. Spectral DecompositionSpectral Decomposition• discrete frequency componentsdiscrete frequency components


Bandwidth8-10-40-50

0 5040302010

0 5040302010

366

400

Tra

vel T

ime

(m

s)

432

466



0.0014

0

Amplitude

25

75

Fre

qu

en

cy (

Hz)

50

100

0

125



+0.025

-0.025

Amplitude

Thickness via Discrete Frequency Components

0 5040302010

0 5040302010

366

400

Tra

vel T

ime

(m

s)

432

466



0.0014

0

Amplitude

25

75

Fre

qu

en

cy (

Hz)

50

100

0

125



10hz

0.00E+00

2.00E-04

4.00E-04

6.00E-04

8.00E-04

1.00E-03

1.20E-03

1.40E-03

0 5 10 15 20 25 30 35 40 45 50


0 5040302010

0.0014

0.0012

Am

plit

ud

e

0.0010

0.0008

0.0006

0.0004

0.0002

0.0000

+0.025

-0.025

Amplitude

10hz

tun

ing

th

ickn

ess

10hz amp

10hz

tun

ing

th

ickn

ess

tuning thickness =1

2 * frequency

0.050sec =1

2 * 10hz

1st-dominant


0 5040302010

0 5040302010

366

400

Tra

vel T

ime

(m

s)

432

466



0.0014

0

Amplitude

25

75

Fre

qu

en

cy (

Hz)

50

100

0

125



20hz

0.00E+00

2.00E-04

4.00E-04

6.00E-04

8.00E-04

1.00E-03

1.20E-03

1.40E-03

0 5 10 15 20 25 30 35 40 45 50


0 5040302010

0.0014

0.0012

Am

plit

ud

e

0.0010

0.0008

0.0006

0.0004

0.0002

0.0000

+0.025

-0.025

Amplitude

20hz

tun

ing

th

ickn

ess

20hz amp

20hz

tun

ing

th

ickn

ess

tuning thickness =1

2 * frequency

0.025sec =1

2 * 20hz

1st-dominant


0 5040302010

0 5040302010

366

400

Tra

vel T

ime

(m

s)

432

466



0.0014

0

Amplitude

25

75

Fre

qu

en

cy (

Hz)

50

100

0

125



30hz

0.00E+00

2.00E-04

4.00E-04

6.00E-04

8.00E-04

1.00E-03

1.20E-03

1.40E-03

0 5 10 15 20 25 30 35 40 45 50


0 5040302010

0.0014

0.0012

Am

plit

ud

e

0.0010

0.0008

0.0006

0.0004

0.0002

0.0000

+0.025

-0.025

Amplitude

30hz

tun

ing

th

ickn

ess

30hz amp

30hz

tun

ing

th

ickn

ess

tuning thickness =1

2 * frequency

0.017sec =1

2 * 30hz

1st-dominant


0 5040302010

0 5040302010

366

400

Tra

vel T

ime

(m

s)

432

466



0.0014

0

Amplitude

25

75

Fre

qu

en

cy (

Hz)

50

100

0

125



20hz30hz

10hz

0.00E+00

2.00E-04

4.00E-04

6.00E-04

8.00E-04

1.00E-03

1.20E-03

1.40E-03

0 5 10 15 20 25 30 35 40 45 50


0 5040302010

0.0014

0.0012

Am

plit

ud

e

0.0010

0.0008

0.0006

0.0004

0.0002

0.0000

+0.025

-0.025

Amplitude

20hz

tun

ing

th

ickn

ess

10hz

tun

ing

th

ickn

ess

20hz amp30hz amp

10hz amp

30hz

tun

ing

th

ickn

ess

20hz

tun

ing

th

ickn

ess

10hz

tun

ing

th

ickn

ess

tuning thickness =1


30hz

tun

ing

th

ickn

ess


0 5040302010

0 5040302010

366

400

Tra

vel T

ime

(m

s)

432

466



0.0014

0

Amplitude

25

75

Fre

qu

en

cy (

Hz)

50

100

0

125



20hz30hz

10hz

0.00E+00

2.00E-04

4.00E-04

6.00E-04

8.00E-04

1.00E-03

1.20E-03

1.40E-03

0 5 10 15 20 25 30 35 40 45 50


0 5040302010

0.0014

0.0012

Am

plit

ud

e

0.0010

0.0008

0.0006

0.0004

0.0002

0.0000

+0.025

-0.025

Amplitude

20hz

tun

ing

th

ickn

ess

10hz

tun

ing

th

ickn

ess

20hz amp30hz amp

10hz amp

30hz

tun

ing

th

ickn

ess

20hz

tun

ing

th

ickn

ess

10hz

tun

ing

th

ickn

ess

By choosing an appropriately-low frequency component, the entire range of possible thickness is forced below the tuning thickness, and therefore can be quantified using amplitude variability alone.

30hz

tun

ing

th

ickn

ess

0.0000

0.0005

0.0010

0.0015

0.0020

0.0025

0

5

10

15

20

25

30

35

40



Bandwidth8-10-40-50

0 5040302010

0 5040302010

366

400

Tra

vel T

ime

(m

s)

432

466



0.0014

0

Amplitude

25

75

Fre

qu

en

cy (

Hz)

50

100

0

125




0 5040302010

0.0014

0.0012

1st D

om

ina

nt

Am

plit

ud

e

0.0008

0.0006

0.0002

0.0000

40

35

1st D

om

ina

nt

Fre

qu

en

cy30

25

20

10

5

0

15

+0.025

-0.025

Amplitude

20hz

tun

ing

th

ickn

ess

10hz

tun

ing

th

ickn

ess

30hz

tun

ing

th

ickn

ess

20hz

tun

ing

th

ickn

ess

10hz

tun

ing

th

ickn

ess

tuning thickness =1


30hz

tun

ing

th

ickn

ess

Outline

• Introduction


• ExamplesExamples

• Pros and Cons

Example

• Using discrete frequency components to determine relative thickening/thinning.

Simple Channel Cross-Section

Animating from low to high frequency,causes amplitude contours to move from thick to thin.

REFLECTIVITY

SPECTRALAMPLITUDE

travel time (m

s)frequen

cy (Hz)

0

250

Example

• Using discrete frequency components to calibrate reservoir thickness.

Deep-Water Gulf of Mexico

8hz Spectral Amplitude Map

WELL #1

Zone 1Zone 1

1 mile

Partyka, Thomas, Turco and Hartmann, SEG 2000

Thickness Modeling

Thickness (ft)0 4020 8060 100

Fre

qu

enc

y (h

z)0

40

60

2010

30

50

70

amplitude1

0

Spectral SignaturesSpectral Signatures

Well-Log Interpretation(Zone 1)

shaleshale

Seismic Modeling(Zone 1)

Tw

o-W

ay T

rave

ltim

e (m

s)

0

200

100

150

50 amplitude1

-1

0

dep

th (

feet

)

01

sandsand oiloil

Temporal Wedge ModelTemporal Wedge Model

6hz6hz

8hz8hz


Thickness Calibration

6hz amplitude

8hz amplitude

Am

pli

tud

e

08hz Spectral Amplitude

Zone 1Zone 1

Thicknessfrom 6hz and 8hz energy

WELL #1

06hz Spectral Amplitude

Zone 1Zone 1

Modeled Spectral Signatures vs Thickness

Zone 1Zone 1

Fre

qu

ency

(h

z)

Modeled 6hz Spectral Amplitude vs ThicknessAriel - Zone A1

0.002

0.003

0.004

0.005

0.006

0.007

0.008

10 20 30 40 50 60 70 80 90 100

Gross Reservoir Thickness (ft)

En

erg

y 8hz

6hz

0.008

0.007

0.006

0.005

0.004

0.003

WELL #1

0

100

50

WELL #1

1

0

1 mile

1 mile

1 mile

0

40

2010

30

10 20 30 40 50 60 70 80 90 100Thickness (ft)

6hz6hz

8hz8hz


Outline

• Introduction


• Examples

• Pros and ConsPros and Cons


• Pros:– user and time intensiveness mandates careful QC.

• Cons:– two attributes are required to quantify thickness:

• peak-trough time-separation for thickness greater-than the tuning thickness, and

• amplitude for thickness less-than the tuning thickness.

– user and time intensiveness mandates careful QC.

1st Dominant Frequency and Amplitude

• Pros:– collapses the Tuning Cube into two maps. – does not require careful seismic event picking when the zone of

interest is relatively bright.• Cons:

– as in the conventional approach, two attributes are required to quantify thickness:

• 1st-dominant frequency for thickness greater-than the tuning thickness, and

• 1st-dominant amplitude for thickness less-than the tuning thickness.

Discrete Frequency Components

• Pros:– can be used qualitatively to determine relative thickening/thinning.– can be used quantitatively to calibrate reservoir thickness.– usually exhibit substantially more fidelity than full-bandwidth,

conventional amplitude/attributes. Can therefore selectively analyse frequencies exhibitting highest signal fidelity.

– usually provide superior rock mass (stratigraphic and structural) and fault definition.

– can be integrated with other appropriate information to yield a more comprehensive understanding of the reservoir.

– does not require careful seismic event picking when the zone of interest is relatively bright.

Discrete Frequency Components

• Cons:– complex layer distributions require seismic modeling analysis to

determine relationship between spectral response and thickness.

Seismic Thickness Estimation: Three Approaches, Pros and Cons Gregory A. Partyka bp.

Documents