Reservoir Model Types - Reservoir Model Design - page 190

Fig. 6.23 Simulation of oil migration into shallow marine rock unit (Carruthers 1998 ) (Redrawn from Carruthers 1998
(Courtesy of D. Carruthers))

Other flow processes, such as gas injection,
would not benefit in the same way from the
geology so care is needed to ensure a rock
model design that fits the flow process at hand.
For a gas injection scheme, the most important
geological feature is likely to be the location and
continuity of the parasequence tops.

requires the seismic to be fortuitously resolved at
the REV scale pertinent to the model purpose.
This can be the case for gas reservoirs but,
recalling Flora's guiding rule (Chap. 2 ) , this is
unlikely to be the general case for oil reservoirs.
Sub-seismic architectural understanding is
usually required, and from reference to a com-
pendium of architectures (e.g. Nilson et al.
2008) it is immediately apparent that there are a
considerable range of possibilities. The question
to ask is: “What's inside the seismic loop?”, and
for model-related issues a consideration of con-
finement is a good place to start.

6.5 Deep Marine Sandstone

Reservoirs

Deep marine systems are dominated by processes
associated with density flows: gravity-driven
currents moving sediments in suspension or by
traction and depositing them along continental
margins. Deep-water systems include but are
not synonymous with 'turbidites' (Kneller 1995 ).
Emphasis in deep marine reservoirs has been
placed firmly on depositional geometries and
reservoir frameworks which are commonly
determined from seismic, in some cases of spec-
tacular quality. In reservoir modelling, the
strength has been the ability to integrate seismic
attributes into conditioned reservoir models; the
weakness has been in the underestimation of
small-scale heterogeneities and in not seeing
what lies below seismic resolution.
The tendency to miss significant reservoir
features has been encouraged by the observation
that seismic data and reservoir simulation work
at similar resolutions. It is therefore tempting to
avoid sub-seismic architecture and work directly
from 'seismic-to-simulation'. This can work, but

6.5.1 Confinement

Confinement describes the extent to which
a submarine gravity flow 'feels' physically
constrained by surrounding topography. Is the
flow being funnelled through a narrow canyon
('confined') or is it depleting on to the floor of a
large open basin ('unconfined')? Confinement is
important as a concept because it is the primary
underlying factor guiding the permeability archi-
tecture we are attempting to capture in reservoir
modelling and simulation.
In confined systems, new density flows tend to
erode into deposits of earlier flows and hence
sands from different flows tend to amalgamate
(Fig. 6.24 ). The erosional elements of the new
flow are typically sand-rich and the fill within the
erosional scour will also tend to be sand-rich,
whether deposited by the initial confined flow