Geology Reference
In-Depth Information
There are clearly some challenges in extracting
fracture information from seismic even assuming that
seismic signal-to-noise, azimuth distribution and fold
are optimised (Johns et al., 2008 ).
0.5
0.45
Parallel to fractures
(fast shear wave)
0.4
If the fracture density is low the anisotropy might
be too weak to detect.
0.35
No fractures
Fractures in multiple azimuths can diminish the
magnitude of azimuthal anisotropy.
0.3
Steeply dipping beds can induce azimuthal
anisotropy as well as vertical fluid filled
fractures.
0.25
Perpendicular
to fractures
(slow shear wave)
0.2
Anisotropy at the reservoir level might be
influenced by anisotropy from above.
0.15
0
5 0 5 0 5 0 5
Angle of incidence (degrees)
5.4 The rock model and its applications
The previous discussion illustrates that in order to
effectively interpret seismic amplitude data the inter-
preter not only needs to understand the geology of
the basin (e.g. sedimentology, stratigraphy and
structural history) but also the rock physics that
explains the nature of the reflectivity. It is important
therefore to generate a database of rock properties
from an analysis of wireline log and other data. This
work involves detailed integration of petrophysical
and rock physics analyses, including log quality con-
trol and conditioning, fluid substitution and appli-
cation of various rock physics models ( Chapter 8 ).
As such, the job of putting together a rock physics
database is commonly done by specialists rather
than seismic interpreters. From the rock physics
database a rock model ( Fig. 5.49 ) can be established
which defines:
Figure 5.48
AVO model of VTI shale overlying fractured granite
(after Leaney
et al
., 1995 ).
Δγ ¼
change in the shear wave splitting parameter
γ
(note that
γ
is directly related to the crack
density).
Figure 5.48 illustrates a modelled solution for aniso-
tropic shale overlying a fractured granite using the
HTI assumption. The AVO responses perpendicular
and parallel to the fracture direction show significant
amplitude differences on far angles. In this particular
case the dimming of far offset amplitudes (i.e. steepest
negative AVO gradient) occurs perpendicular to frac-
tures and is consistent with Thomsen
s( 1995 ) sugges-
tion that the lesser AVAZ gradient is parallel to
fracture strike (Goodway et al., 2010 ). There are
numerous case studies in which fractures have been
interpreted from multi-azimuth seismic (e.g. Hall and
Kendall, 2003 ; Neves et al., 2003 ; Gray et al., 2002 ) but
there does not appear to be a rule of thumb with
regard to the interpretation of fracture orientation.
Far offset dimming may occur perpendicular or par-
allel to fractures. Clearly, whilst the AVO provides the
all-important spatial context, integration with other
data such as borehole image logs, mud loss measure-
ments and estimates of regional and local stresses (e.g.
borehole breakout studies) is required. Modelling of
course is also a key component in interpretation but it
is clear that care needs to be taken in how the models
are applied. Goodway et al.( 2010 ) have documented
potential ambiguities in the application of Rüger
'
(1) seismic lithofacies (based on crossplots and log
plots (upscaled data)) and statistical variation
(e.g. mean, standard deviation);
(2) thicknesses and stratigraphic ordering of
lithofacies;
(3) first order depth trends of
(a) elastic parameters of seismic lithofacies
(i.e. V p , V s , density and porosity)
(b) effective pressure and temperature
(c) fluid properties.
The rock model can be used in numerous applica-
tions including the generation of seismic rock
models for reflectivity interpretation ( Chapter 7 ),
s
two-term equation, suggesting that a three-term solu-
tion is required.
'
89
Search WWH ::




Custom Search