Geology Reference
In-Depth Information
a)
b)
Vsand
Pay
Phi
Sonic
Den
Increasing pay
0
0
48 net pay ft
25
20
50
75
100
100
125
c)
d)
400
400
300
300
comp amp
trough
200
200
peak
100
100
0
0
0
10
20
30
40
0
20
40
60
Apparent thickness (ms)
Net pay thickness (ft)
Figure 10.19
(a) Log plot showing porous sands and shales, (b) modelled traces for incremental pay thicknesses, (c) apparent thickness vs
composite amplitude plot, (d) net pay thickness vs amplitude (re-drawn from Neff,
1990
).
(including net pay thickness) associated with the
match. Unlike traditional approaches to trace inver-
sion (
Chapter 9
) no further inversion steps (for
example from impedance to reservoir property) are
required. In addition, multiple matches can be used
to assess uncertainty.
Populate sands and shales with velocity and
density data from probability density functions,
and constraining cross correlations, derived from
rock physics analysis.
Calculate impedance traces at the appropriate
AVO angle.
Convolve impedance traces with the correctly
scaled bandpass wavelet relevant to the
bandlimited impedance dataset.
10.4 Time-lapse seismic
'
seismic refers to the use of
seismic data to detect changes in the reservoir over
time. Typically the technique is used in hydrocarbon
production but it can also be used for other purposes
such as monitoring gas or carbon dioxide storage.
The time-lapse premise is straightforward, with a
'
Time-lapse
'
or
'
4D
'
Pick upper and lower zero crossings and calculate
average bandlimited impedance and apparent
thickness.
Uncertainty is then estimated from the pseudo-well
data in the form of standard deviations of seismic N:G
(
Fig. 10.22
).
The logical application of pseudo-wells is to per-
form seismic inversion through trace matching (e.g.
Burge and Neff,
1998
;Ayeniet al.,
2008
; Spikes
et al.,
2008
). An attractive aspect of trace matching
is that when pseudo-wells provide simultaneous
matches to multiple angle stacks then the result
provides all petrophysical and geological parameters
survey typically acquired before produc-
tion and successive
baseline
'
surveys acquired at
periods (months or years apart) during the life of
the field. Individual surveys may be interpreted as
separate
'
monitor
'
but most modern time-lapse
projects aim to create differenced seismic volumes.
Provided that the seismic processing has optimised
the time-lapse signal, the differences between the
'
snapshots
'
236
