Geoscience Reference
In-Depth Information
Fig. 6.32
Facies interpretation of a platform margin
from outcrops near Rustrel, France (Leonide et al.
2012
)
(
left
) and a cellular representation of the same (
right
)
(Left
image redrawn from Leonide et al.
2012
,
#
SEPM Society
for Sedimentary Geology
[2012],
reproduced with permission)
In the example shown in Fig.
6.32
, outcrop
analogues for the Shuaiba reservoir are drawn
from examples in Provence (Leonide et al.
2012
) and detailed mapping of limestone facies
provides insight into the distribution of
reservoir types in age-equivalent Middle Eastern
reservoirs. Even here, though, the palaeo-
environments between reservoir and analogue
locations differ - the two areas lay on opposite
sides of the ancient Tethys Ocean and are
characterised by different faunal assemblages.
Other carbonate environments display
impressive lateral continuity and might appear
to make little call on the rock modelling toolbox.
This is reported from field studies by Palermo
et al. (
2012
) working on the Muschelkalk, where
laterally consistent reservoir properties have
been measured in platform carbonates and traced
over several 100's of metres, contrasting mark-
edly with very abrupt vertical variations. Similar
patterns are common in platform carbonates of
the Middle East (Fig.
6.33
).
This extreme anisotropy is also familiar in
carbonate-evaporite sequences where heteroge-
neities are controlled laterally by gentle basin-
wide chemical gradients but vertically by
fluctuations in basin inputs and outputs, such as
periodic connection and disconnection with
open seawater and periodic basin desiccation
(Fig.
6.34
).
Similar high frequency, laterally-correlatable
cyclicity is also common to chalk fields, and the
regularity can be picked out from vertical
variograms of porosity and permeability
measured on core (Almeida and Frykman
1994
)
(Fig.
6.35
).
Depositional environments can therefore dic-
tate the need for simple, very thinly layered
models, or heterogeneous object-based models,
although the latter potentially lack good ana-
logue data. In this respect, carbonate modelling
workflows may be comparable to workflows for
clastic reservoirs.
6.6.2 Pore Fabric
Where carbonates and clastics differ most mark-
edly is in their pore fabric. Clastic systems can
generally be broken down into elements with
reasonable porosity-permeability relationships,
reflecting a consistency of carbonate pore type
for a given model element. This is often not
the case in carbonates, where there may be
little or no relationship between porosity and
permeability.
Theunderlyingreasonisporesizedistribution,
which can vary over very short distances in
carbonates owing to the irregularity of pore shapes.
Not all carbonates behave this way: as pore shapes
become more uniform, such as in some chalks or
well sorted grainstones, regular k/
relationships
emerge. However, classifying core plug data using
the Dunham (
1962
) descriptive scheme, useful for
˕
