Seismic inversion

Seismic inversion course, UPPA, 2008

Course Outline
General aspects of seismic inversion
ƒ Concept and purpose
ƒ Petrophysical basis
ƒ Data requirements
ƒ Overview of methods and basic inversion process

Poststack inversion – practical workflow

ƒ Data QC
ƒ Wavelet extraction
ƒ A priori model building
ƒ Inversion parameterization and QC of results

Prestack inversion – practical workflow

ƒ Data QC
ƒ Wavelet extraction
ƒ A priori model building
ƒ Inversion parameterization and QC of results

Advanced inversion techniques

ƒ Geostatistical inversion
ƒ Joint PP/PS inversion
ƒ 4D inversion

2 - Seismic inversion course, UPPA, 2008

 Geostatistical inversion

3 - Seismic inversion course, UPPA, 2008

Geostatistical inversion concept
Generate impedance models that
satisfy the statistics of the input data:
mean, variance, variogram

Impedance models are geologically

constrained:
• stratigraphic framework
• 3D a priori information
• variograms

Wavelet Vertical Horizontal variogram Statistical LF

covariance on seismic a priori model
Amplitude on Well log
γ (h)
0 1 AI
0 1

time
time 20ms
Mean
0 800 m h Mean +/- 2*s.d.
4 - Seismic inversion course, UPPA, 2008
 Variogram analysis
Variogram: Mean squared
Horizontal variogram
difference between pairs of points (seismic amplitude)
as a function of distance

Semi-variogram
Model
Data variance

Lag Distance (feet)

Vertical variogram Semi-variogram

Horizontal variograms (2D) are (AI well logs) Data variance

Combined Model
Model Structure 1
calculated parallel to Model Structure 2

stratigraphy

Vertical (1D) variograms along

the vertical direction

Lag Distance (feet)

(Courtesy O. Dubrule)

5 - Seismic inversion course, UPPA, 2008 2-22

 Geostatistical inversion algorithm

For each global realization

Define a random path through all
Trace location nodes (x,y) to be simulated
to be simulated For each node (x,y) perform a local
optimization
ƒ generation of a large number of local
realizations of acoustic impedance
(x,y) traces, such that input data statistics
are honoured
ƒ convolution with the wavelet
ƒ comparison with the actual seismic
ƒ retain the best trace which becomes
conditioning data

Go to next node

Actual seismic trace Local

realizations (Courtesy O. Dubrule)
6 - Seismic inversion course, UPPA, 2008
Dozens of realizations…

Real. 7 Real. 27

All honour the well and seismic data

Real. 13 Real. 31

7 - Seismic inversion course, UPPA, 2008

Main features of geostatistical inversions
One realisation
Impedance realizations honour
- seismic data
- well data

4 AI (km/s . g/cc) 9 Great number of AI realisations =>

Mean uncertainty on impedance results

High vertical frequency of impedance

realizations controlled by vertical
variogram model
SD
Impedance results are at high
resolution related to the stratigraphic
grid definition (~ 2 - 8m ; ~ 0.5 - 2ms)
0 σAI (km/s . g/cc) 0.7

8 - Seismic inversion course, UPPA, 2008

 Frequency bandwidth in a geostatistical inversion

Statistical layer constraints control low frequencies

Seismic amplitudes control medium frequencies

Variogram model controls high frequencies

Seismic bandwidth

(Courtesy E. Robein & L. Barens)

9 - Seismic inversion course, UPPA, 2008 5-9

 How to summarise geostatistical inversion results?

Threshold impedance probability

without seismic

Threshold impedance probability

with seismic
P (AI > 6.6)

10 - Seismic inversion course, UPPA, 2008 (Courtesy O. Dubrule)

 From impedance to petrophysical properties

50 Impedance realizations
50 Porosity realizations

Collocated
Impedance vs. Porosity cokriging

11 - Seismic inversion course, UPPA, 2008

Deterministic versus Stochastic inversion
The mean of all stochastic realizations looks like the deterministic, but it is different in details
w
Ra-10 Ra-1
w w
Ra-7 w
Ra-10 w
Ra-1 w
Ra-7

4ms sampling rate 0.25ms sampling rate

Reservoir Top
15ms Reservoir Top

Reservoir Base Reservoir Base

DETERMINISTIC
Red color indicates low AI STOCHASTIC REALIZATION

& porosity development w

Ra-10 w
Ra-1 w
Ra-7
Litho 3
∅ > 15%

Litho 2
W -10 10<∅ < 15%
Reservoir Top

Litho 1
W -1 ∅ < 10%

W -7 Reservoir Base
4 km
CO-SIMULATED LITHOTYPE
Depth Map

12 - Seismic inversion course, UPPA, 2008

 Joint PP/PS inversion

13 - Seismic inversion course, UPPA, 2008

 Joint PP/PS inversion

Simultaneous inversion of PP
and PS seismic volumes

PS data need to be
transformed into PP time and
reflections need to be
correlated

Calibration of PS data with

shear-wave logs is necessary PP prestack gather PS prestack gather

Joint PP/PS inversion often (From Margrave & Stewart, 2001)

constrains IS better than
single PP inversion

14 - Seismic inversion course, UPPA, 2008

 Joint inversion results
ΔIP

ΔIS

ΔIP-ΔIS
(From Margrave & Stewart, 2001)
15 - Seismic inversion course, UPPA, 2008
 Joint inversion case study: Impedance crossplots
PP single inversion PP-PS joint inversion

ΔIp/Ip
ΔIp/Ip

ΔIs/Is ΔIs/Is
LOG DATA Ip
ΔIs/Is from single PP inversion:
ƒ Wrong polarity at top reservoir
ƒ Appears to be simply a scaled version Oil sands
of ΔIp/Ip

ΔIs/Is from joint PP-PS inversion:

Is
ƒ Correct polarity at top reservoir
Shales
ƒ Independent parameter

Vclay
16 - Seismic inversion course, UPPA, 2008
 Joint inversion case study: 3D results
ΔF, joint inversion

ΔF is the fluid factor, the product of ΔIp/Ip and ΔIs/Is, which

highlights zones of gas

17 - Seismic inversion course, UPPA, 2008

 4D inversion

18 - Seismic inversion course, UPPA, 2008

3D Simultaneous Elastic Inversion in a Stratigraphic
Framework
Macro Stratigraphic Grid Framework Micro Stratigraphic Grid Framework
173 layers (5ms/5m)

19 interpreted horizons in an interval Vp,Vs and ρ: 3D kriging of low-pass

of 400 ms filtered log data

Model optimization: Simulated Annealing

Perturbations: Vp, Vs, ρ and time
Case study:
ƒ Δρ linked with ΔVp through Gardner’s relation

19 - Seismic inversion course, UPPA, 2008

 4D Coupled Inversion Workflow
Monitor a priori model building

Base a priori model building • Base inversion interpretation

• Base/Monitor time shift estimation
• Rock Physics Model

3D simultaneous 3D simultaneous
inversion inversion
Base acquisition Monitor acquisition
(P,S Impedance) (P,S Impedance)

Δ Ip, Δ Is
Base/Monitor
QC and interpretation
20 - Seismic inversion course, UPPA, 2008
 Layer 42 - P Impedance - Base Inversion
100 m 6500

gr/cm3 x m/s
10 ms

Well B

Well D
4000

42

Well C

Well E

Well A

21 - Seismic inversion course, UPPA, 2008

 Layer 42 - P Impedance - Monitor Inversion
100 m 6500

gr/cm3 x m/s
10 ms

Well B

Well D
4000

42

Well C

Well E

Monitor inversion
performed with the same Well A
a priori constraints as Base inversion

22 - Seismic inversion course, UPPA, 2008

10 ms Layer 42 - Delta P Impedance: Base-Monitor
Base +600

gr/cm3 x m/s
Well B

- 600 Well D
42

100 m
10 ms

Monitor Well C

Well E

42 Well A

Monitor inversion
performed with the same
a priori constraints as Base inversion
23 - Seismic inversion course, UPPA, 2008