Geoscience Reference
In-Depth Information
Well data
Logs re-scaled to
model cell size
k
k
4
Stochastic geological
reservoir model
3
2
1
Fig. 3.31 Handling net-sand within a facies model.
Block and upscaling can affect both the facies volume
fraction and the net-sand volume fraction (Redrawn from
Ringrose 2008 , # 2008, Society of Petroleum Engineers
Inc., reproduced with permission of SPE. Further repro-
duction prohibited without permission)
Sketch of geological
architecture within the
reservoir model grid cell
Sandstone facies
with shale layers and
cemented sand
patches
10m thick simulation grid cell
with 6m of 1000md sand
and 4m shale (N/G = 0.6)
k h = 600md
k v = undefined
Fig. 3.32 Simple example of a reservoir grid block
where the N/G assumption is correctly used to estimate
a horizontal block permeability of 600 mD in the case
where the net sand has an upscaled permeability of
1000 mD (Redrawn from Ringrose 2008 ,
2008, Society
of Petroleum Engineers Inc., reproduced with permission
of SPE. Further
#
reproduction
prohibited without
permission)
is a more comprehensive approach and is used in
many academic studies where economic cut-off
factors (e.g. oil reserves) are not a major concern.
This approach is especially appropriate where:
1. Reservoir properties are highly variable or
marginal;
2. Cementation is as important as the sand/shale
issue;
3. Carbonates comprise the main reservoirs.
The Total Property Modelling (TPM) method
is illustrated in Fig. 3.33 . Note that net-reservoir
is still defined but only after modelling and
upscaling. Since shaly or cemented rock units
are modelled explicitly alongside better quality
sandstones (or carbonates) it is easy to test the
effect of assuming different cut-offs - such as
“How will an 8 % versus a 10 % porosity cut-off
affect the reserves forecast?”
Search WWH ::




Custom Search