Chandra - Effective Integration of Reservoir Rock Typing and Res Simulation

Marine and Petroleum Geology 67 (2015) 307e326
Contents lists available at ScienceDirect
Marine and Petroleum Geology

journal homepage: www.elsevier.com/locate/marpetgeo
Research paper
Effective integration of reservoir rock-typing and simulation using

near-wellbore upscaling
Viswasanthi Chandra a, b, *, Andrew Barnett c, Patrick Corbett a, b, Sebastian Geiger a, b,
Paul Wright c, Richard Steele c, Paul Milroy c
a
Institute of Petroleum Engineering, Heriot-Watt University, Riccarton, Edinburgh, EH14 4AS, UK
b
International Centre for Carbonate Reservoirs, UK
c
BG Group, 100 Thames Valley Park, Reading, RG6 1PT, UK
a r t i c l e i n f o a b s t r a c t
Article history: Obtaining a fit-for-purpose rock-type classification that adequately incorporates the key depositional and
Received 1 August 2014 diagenetic heterogeneities is a prime challenge for carbonate reservoirs. Another prevailing issue is to
Received in revised form integrate the static and dynamic data consistently with the rock-typing scheme in order to correctly
4 May 2015
initialise the reservoir flow simulation model. This paper describes a novel near-wellbore rock-typing
Accepted 7 May 2015
Available online 20 May 2015
and upscaling approach adopted to address the crucial challenges of integrating reservoir rock-typing
and simulation in carbonate reservoirs. We demonstrate this workflow through a case study for a
highly heterogeneous Eocene-Oligocene limestone reservoir, Field X. Geological studies carried out in
Keywords:
Carbonate reservoirs
Field X suggested that the key permeability pathways are strongly related to the mechanism of reservoir
Rock-typing porosity and permeability evolution during late-burial corrosion. The rock-typing and upscaling meth-
Near-wellbore upscaling odology described in this paper involves the geological-petrophysical classification of the key reservoir
Initialisation heterogeneities through systematic evaluation of the main paragenetic events. Associations between the
Permeability depositional and late-burial corrosion features, and their impact on reservoir flow properties, were
Relative permeability accounted for in our workflow. Employing near-wellbore rock-typing and upscaling workflow yielded
Capillary pressure consistent initialisation of the Field X reservoir simulation model and therefore improved the accuracy of
J function
fluids-in-place calculation. Subsequently, the cumulative production curves computed by the reservoir
simulation model of Field X showed closer agreement to the historic production data. The revised Field X
simulation model is now much better constrained to the reservoir geology and provides an improved
geological-prior for history matching.
© 2015 Elsevier Ltd. All rights reserved.
1. Introduction 2013; Skalinski and Kenter, 2015). Another common challenge is

to integrate the dynamic data during rock-typing and upscale the
Rock-typing describes the process of characterising geological petrophysical properties of the rock types to the reservoir model
facies in terms of their dynamic behaviour. It is highly challenging scale using appropriate geostatistical tools to correctly initiate the
to obtain a fit-for-purpose rock-typing scheme that adequately reservoir simulation model. Furthermore, the difficulties of pre-
represents the influence of diagenetic processes on the reservoir dicting reservoir quality variations at inter-well scales have long
petrophysical properties, fluid in-place volumes, and hydrocarbon hindered the efficacy of carbonate reservoir rock-typing and
recovery efficiency. This is a classic issue for carbonate reservoirs, simulation (c.f. Agar and Geiger, 2015).
which typically contain multi-scale and multi-modal pore types Numerous authors have proposed carbonate rock-typing
that are difficult to be adequately incorporated into rock-typing workflows, trying to address the challenges discussed above.
(e.g. Gomes et al., 2008; Hollis et al., 2010; van der Land et al., Gomes et al. (2008) discussed the importance of obtaining good
understanding of the depositional and diagenetic processes in or-
der to establish better links between lithofacies, petrophysical
groups and rock types. Hollis et al. (2010) demonstrated a rock-
* Corresponding author. Institute of Petroleum Engineering, Heriot-Watt Uni-
versity, Riccarton, Edinburgh, EH14 4AS, UK. typing workflow based on pore system characterisation such that
E-mail address: [email protected] (V. Chandra). each rock type could be defined on the basis of both its
http://dx.doi.org/10.1016/j.marpetgeo.2015.05.005
0264-8172/© 2015 Elsevier Ltd. All rights reserved.
308 V. Chandra et al. / Marine and Petroleum Geology 67 (2015) 307e326
petrophysical properties and behaviour during hydrocarbon re-

covery. Van der Land et al. (2013) proposed a general approach
towards carbonate rock typing through pore-scale forward
modelling of the paragenetic sequences. Skalinski and Kenter
(2015) proposed a carbonate petrophysical rock-typing workflow
that could account for the role of diagenetic processes on reservoir
dynamics, and could also account for the role of fractures. In
contrast, the work presented in this paper is aimed to address all of
the above issues through an integrated near-wellbore rock-typing
and upscaling approach. The novel aspect of the Near-Wellbore
Upscaling (NWU) methodology applied in our study is that it
enabled explicit modelling of typical multi-scale carbonate het-
erogeneities such as leaching, stylolites and associated small-scale
fractures, which are difficult to account for in existing reservoir
rock-typing and simulation workflows. Our methodology allows us
to model these multi-scale geological-petrophysical features more
robustly at the reservoir grid-block scale in a reservoir simulation
Figure 1. Cumulative oil production curves from Field X. OPT corresponds to the
model. The rock-typing and upscaling methodology described in
history matched simulation model, which yields a cumulative oil production that is
this study involves the geological-petrophysical classification of identical to the historic production data OPTH. OPT1 is the simulated production after
multi-scale heterogeneities in the studied reservoir through sys- removing horizontal permeability multipliers (K-multipliers) from the zones. OPT2 is
tematic evaluation of the key paragenetic events with consider- the simulated production after removing zone and local well K-multipliers. OPT3 is the
simulated production after removing the well productivity multipliers in addition to
ation to the crucial parameters of near-wellbore modelling and
the zone and well K-multipliers. STB is the abbreviation for ‘Stock Tank Barrels’.
upscaling workflow. Figure modified from Chandra et al., 2015.
1.1. Case study: rock-typing and simulation challenges in Field X
The key porosity types present in Field X, such as chalky micro-

Field X is a giant offshore oil and gas field comprising
porosity, macroporosity including vuggy and moldic pores, leached
an Eocene-Oligocene limestone reservoir with long production
stylolites and associated tension gashes, were caused by late burial
history. It has a broad, low relief anticline trap structure and con-
(mesogenetic) dissolution (Wright and Barnett, 2011). Note that
sists of a gas column up to 50 m thick, an oil rim of about 20 m
microporosity in Field X is defined as pores with a pore throat
thickness, and an underlying aquifer of ground water. Permeability
diameter of 0.5 microns or less.
has been identified as one of the biggest uncertainties associated to
In order to revise the geomodel of Field X with considerations to
Field X's reservoir simulation model during previous field studies
the proposed late-burial corrosion model (Wright and Barnett,
(Chandra, 2014; Chandra et al., 2015). In addition, the performance
2011), it is vital to obtain a fit-for-purpose rock-typing scheme
predictions were also found to be sensitive to the volumes of fluids
that adequately incorporates the late-burial corrosion heteroge-
initially-in-place and the critical oil saturation. Major modifications
neities. However, conventional rock-typing workflows may not be
of the reservoir model were required in order to obtain a history
appropriate to Field X due to the difficulties associated with the
match. These modifications were not necessarily based in the
petrophysical characterisation and data sampling of the key
reservoir geology but comprised numerical adjustments in the
porosity types such as leached stylolites and associated tension
dynamic model. For example, horizontal permeability multipliers
gashes. Besides, upscaling these sub-grid-scale heterogeneities to
of 10 and 20 had to be applied in the main reservoir zones, along
incorporate them adequately into the reservoir simulation model is
with multipliers at the wells (Oates et al., 2012; Chandra, 2014;
another major challenge. The available core plug data suffers from
Chandra et al., 2015). Chandra et al., 2015 demonstrated how the
inherent sample bias and insufficiency due to rock-mechanical
computed cumulative oil production decreased when the perme-
constrictions and due to the shortcomings of using a regular sam-
ability modifications were removed in the dynamic model (Fig. 1).
pling interval of 1 m (Oates et al., 2012). Due to these challenges,
The irregularities observed between the distributions of fluids
Field X provides an ideal case study for the application of the near-
based on reservoir simulation predictions and the actual produc-
wellbore rock-typing and upscaling methodology. In this study, we
tion volumes also imply that the in-place volumes and production
thus aim to employ the near-wellbore rock-typing and upscaling
characteristics were not properly understood (Calvert and Ballay,
methodology to revise the full field geomodel of Field X and
2011). It was hence deemed necessary to re-evaluate the
generate a properly initialised reservoir simulation model with
geological-petrophysical model of Field X to increase the reliability
consistent static and dynamic reservoir properties.
of oil-in-place calculations and reservoir model initialisation for
simulation predictions.
Geological studies carried out by the operator suggest that the 1.2. Field X depositional and diagenetic history
key static and dynamic reservoir properties in Field X are strongly
related to the mechanism of reservoir porosity-permeability evo- Field X has a broad, low-relief anticlinal trap structure and
lution during late-burial corrosion (Wright and Barnett, 2011). In comprises an Eocene-Oligocene limestone reservoir (Fig. 2A) with
this study, late-burial corrosion is referred to as deep burial/mes- an oil rim and an underlying aquifer. The offshore basin
ogenetic corrosion associated with the corrosion of limestone by comprising Field X is a passive margin basin, split into longitudinal
burial-derived (hypogene) fluids. The uncertainty associated with horst and graben stripes by a series of basement controlled
permeability modelling in Field X was discussed by Oates et al. NorthWest-SouthEast to North-South trending faults. The major
(2012), Chandra (2014) and Chandra et al. (2013b,c, 2015). structural feature in the block containing Field X and its neigh-
Chandra et al. (2015) demonstrated that re-evaluating the perme- bouring hydrocarbon fields is the East-fault zone (Fig. 2B). Figure 2B
ability model of Field X with considerations to late diagenetic shows the porosity model of Field X and the four wells used in this
corrosion can significantly improve the reservoir simulation model. study. Porosity and permeability came from Routine Core Analysis
V. Chandra et al. / Marine and Petroleum Geology 67 (2015) 307e326 309
Figure 2. (A) Stratigraphic summary of Field X showing the main reservoir units A Zone and B Zone, the gas-oil contact (GOC) and the oil water contact (OWC) (Chandra, 2014). (B)
Field X geomodel showing the porosity distribution, main fault zones present in the model and the four wells used in this study. The East fault zone is the major structural feature in
the block containing Field X and its neighbouring hydrocarbon fields (Chandra, 2014). (C) Illustration of foraminifera distribution in the limestones of Field X with reference to a
carbonate ramp model zone. Facies key; CG e Coskinolina grainstones; CA e Coskinolina and Alveolinid facies; CH e Coskinolina hash facies; M e Miliolid facies; H e Hash facies with
fine skeletal debris; N e Nummulites matrix-rich limestones, ND e Nummulites-Discocyclinid facies. Modified from Wright and Barnett (2011).
(RCA) while capillary pressure and relative permeability data used for burial-derived fluids that caused a major phase of late-burial
in this work were part of the measured Special Core Analysis (SCAL) (mesogenetic) dissolution. During this process the stylolites and
data. RCA and SCAL were both provided by the operator. The associated tension gashes could have conducted reactive fluids
original geomodel of Field X and the history matched simulation containing sulphides, silica and aluminium, enabling them to
model were provided by the operator. The original model of Field X migrate into the surrounding host limestones. These reactive fluids
did not involve any rock-typing scheme. The geomodel comprises corroded the formerly tight cemented matrices by selectively
over five million grid blocks and a total of 59 layers. The grid block removing the micritic grains with high surface area (Wright and
dimensions are 50 m 50 m horizontally and an average of 2 m Barnett, 2011). The conduits feeding the reactive fluids to the
vertically. The two main zones in Field X hosting oil and gas are the reservoir are not known with certainty. Feed through faults and
Early Oligocene A Zone and the Eocene B Zone. The oil-water from the closely underlying basement are both possible. The
contact and the gas-oil contact are present in the B Zone. The presence of exotic minerals in the core such as pyrite, dickite and
stratigraphic framework of these zones was interpreted to be a saddle dolomite supports mixing corrosion mechanism as defined
stacked depositional sequence in a shallow ramp setting (Fig. 2C). A by Esteban and Taberner (2003). Late-burial dissolution (Esteban
Zone and B Zone are separated by a disconformity that is associated and Taberner 2002, 2003) played a crucial role in the evolution of
with sub-aerial exposure corresponding to an early Oligocene fall in reservoir static and dynamic properties in Field X (Wright and
relative sea level. A shale unit, which we refer to as ‘A/B supra- Barnett, 2011).
unconformity shale’, overlies this disconformity and acts as tran- The main porosity types present in Field X and probably the
sient local seal, capping the B Zone. Outer ramp Nummulitic facies majority of the reservoir porosity originated as a result of late-
are the dominant facies in the A Zone. In contrast, the B Zone burial corrosion of the A Zone and B Zone limestones. The key
mainly consists of inner ramp Coskinolina assemblage. A Zone and porosity types associated with late-burial corrosion are well con-
B Zone were interpreted to be of different depositional systems and nected micro- and macro-pore networks with leached stylolite and
were hence treated separately throughout the rock-typing work- tension gash porosity, all of which have significant impact on fluid
flow employed in this study. flow in the reservoir (Fig. 4) (Wright and Barnett, 2011; Chandra
The main paragenetic events that occurred in Field X were et al., 2015). The tension gashes observed in the core were at
discussed by Wright and Barnett (2011) and are summarised in times upto 6 mm wide and usually aligned with the direction of the
(Fig. 3). The A Zone underwent deeper phreatic stabilisation after stylolite peaks. They seemed to have initiated from the mechanical
deposition, whereas the B Zone sediments were stabilised and discontinuity along the stylolite surface and developed along the
cemented under shallow burial conditions. During intermediate direction of the lithostatic stress. Earlier geological studies suggest
burial, pressure dissolution features such as stylolites, micro- the presence of two generations of tension gashes, the most recent
stylolites and clay seams developed ubiquitously. The majority of ones tending to be open or partially filled with dickite and calcite.
the stylolites were associated with tension gashes, some of which Hence the two types of ‘fracture porosity’ observed in Field X are; a)
were cemented (Moshier, 1989; Alsharhan, 1990; Alsharhan and Stylolite associated tension gashes, which are few mm to several
Sadd, 2000). It was hypothesised that the stylolites later reop- cms long and visible only on or below core-scale resolution and b)
ened during a tectonic uplift event, allowing them to act as conduits 30e40 cm long fractures with no direct relation to stylolites that
Figure 3. Key paragenetic events that occurred in A Zone and B Zone (modified from Wright and Barnett, 2011). The depositional facies underwent extensive early cementation
followed by compaction and pressure solution developing stylolites and associated tension gashes. This was followed by a major phase of dissolution associated with saddle
dolomite and dickite precipitation.
truncate the stylolites and tension gashes and can be seen in FMI 2. Near-wellbore rock-typing and upscaling workflow
logs. This type of fractures was rarely observed in the studied wells.
In this paper, the porosity types associated with late-burial disso- In order to facilitate the effective description of this workflow,
lution are hereafter referred to as corrosion-enhanced porosity we have defined some key terminology at the outset. Lithofacies is
(CEP). Figure 5 illustrates the conceptual model of the multiscale defined as depositional facies classified based on sedimentary
connectivity of the CEP zones visualised across the Field X reservoir. texture, grain types and depositional structure (Dunham, 1962). A
In effect, the rock-typing methodology we have adopted for Field X ‘geological rock type’, GeoRT, is a division of a lithofacies type that
was based on the hypothesis that the enhanced permeability in went through similar diagenetic processes and whose pore
Field X was caused by late-burial corrosion, which needed to be network has the same genetic origin. A ‘near-wellbore rock type’,
accounted for in the geomodel and reservoir simulation model. NWRT, is a bin of GeoRTs that have similar petrophysical properties
Figure 4. Illustration of varying scales of corrosion-enhanced porosity caused by late Figure 5. Illustration of the conceptual model of multiscale connectivity of the CEP
burial corrosion within the corrosion-enhanced porosity (CEP) zones in A and B Zones zones across the Field X reservoir. The leached stylolites or tension gashes alone may
(modified from Chandra et al., 2015). (A) Vuggy/moldic porosity on core. (B) Leached not act as the high permeability conduits, but rather the combination of the corrosion-
stylolite and associated tension gashes. (C) Matrix micro- and macroporosity observed enhanced matrix porosity, stylolites and tension gashes is the potential contributor to
in thin-section. the enhanced permeability in Field X.
at core scale and similar near-wellbore modelling (NWM) aspects. the depositional and diagenetic models of the field. Subsequently
A ‘Geological-Porosity Derived System’, GeoPODS, is a geologically the GeoPODS are conditioned to the near-wellbore data and at the
realistic equivalent of the ‘hydraulic flow unit’ (c.f. Amaefule et al., same time spatially-commensurate with the reservoir dynamic
1993) at the reservoir grid-block scale. It consists of a bin of NWRTs simulator for distributing the dynamic properties away from the
that went through a near-wellbore upscaling workflow and display wells. The following sections describe how we applied the inte-
similar static and dynamic properties. grated near-wellbore rock-typing and upscaling workflow to Field
The rock-typing approach we adopted was iterative to ensure X. A series of ‘roadmaps’ has been included to highlight the key
seamless integration of the multi-scale and multi-disciplinary steps of the workflow in the corresponding sub-sections.
datasets. First, we classified the geological rock types (GeoRTs) by
characterising and grouping the gross depositional facies and the
3. Obtaining geological rock types (GeoRTs)
key porosity types associated with late-burial dissolution. This was
guided by the depositional model and interpretation of paragenetic
The specific types and distributions of the pores present within
sequences that occurred in the reservoir. The GeoRTs in our study
the reservoir rocks impart strong control on the production and
are hence comparable to the rock types developed by Hollis et al.
stimulation characteristics of carbonate reservoirs (e.g. Chilingarian
(2010) who grouped the lithofacies of the same genetic origin
et al., 1992; Honarpour et al., 1992; Jodry, 1992; Wardlaw, 1996;
and pore network to ensure that the entire geological evolution of
Mazullo, 2004; Chandra et al., 2015). It is often the case that the
the rock (including the diagenetic history) is incorporated into the
types of fabric-selective pores coincide with the depositional
rock type definition. The GeoRTs were then grouped into ‘near-
environment of the rock (Hollis et al., 2010; Van der Land et al.,
wellbore rock types’ (NWRTs) based on their petrophysical char-
2013). Understanding the crucial links between depositional and
acterisation using the RCA and SCAL data. This was done with
diagenetic models can provide valuable guidance towards unrav-
consideration to the NWM aspects of the GeoRTs to facilitate their
elling the relationships between lithofacies and petrophysical
input into the near-wellbore upscaling workflow. Near-wellbore
groups (Gomes et al., 2008), and towards the evaluation of the
upscaling yields averaged petrophysical properties at the reser-
reservoir flow properties (Mazullo, 2004). Hence the prime
voir simulation grid-block scale.
incentive for obtaining the GeoRTs is to capture the key relation-
Following near-wellbore upscaling, the centimetre-scale NWRTs
ships between the depositional and diagenetic features with con-
were further binned into reservoir-scale ‘Geological-Porosity
siderations to their impact on reservoir flow properties.
Derived Systems’ (GeoPODS) based on their upscaled petrophysical
Figure 6 illustrates the integrated near-wellbore rock-typing and
properties. In this work, GeoPODS emphasise the geological basis
upscaling workflow, highlighting the steps taken to obtain GeoRTs.
for the grouping and upscaling of rock types and are coherent with
As discussed in the earlier sections, late-burial corrosion played a
the definition of Porosity Derived Systems (PODS) (Kazemi et al.,
critical role in the evolution of reservoir petrophysical properties in
2012) at the reservoir grid block scale. Thus, GeoPODS represent
Field X (Wright and Barnett, 2011; Chandra et al., 2015). In order to
the impact of sub-grid scale heterogeneities on fluid flow at the
obtain the GeoRTs in Field X, the lithofacies types, stylolite types
reservoir grid-block scale and help define the Global Hydraulic
and the CEP types observed in the studied wells were characterised,
Elements (c.f. Corbett and Potter, 2004) concept at the appropriate
logged and digitised. For this we used core and thin-section data
modelling scale for full field simulation models. The evolution of
from over 300 m of well cores from 4 vertical wells and 1 highly
GeoPODS is strongly linked to GeoRTs, which in turn are based on
deviated pilot well. These core description logs were then used to
Figure 6. Summary of the integrated near-wellbore rock-typing and upscaling workflow. A geological rock type (GeoRT) is a division of a lithofacies type that went through similar
diagenetic processes. A near-wellbore rock type (NWRT) is a bin of GeoRTs that have similar petrophysical properties at core scale and near-wellbore modelling (NWM) parameters.
A Geological-Porosity Derived System (GeoPODS) is equivalent to a Global Hydraulic Element consisting of a bin of NWRTs grouped based on the upscaled porosity-permeability.
The steps taken to obtain GeoRTs are highlighted here in yellow. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this
article.)
Table 1 controls of burial corrosion that were considered for the classifi-
Summary of major lithotypes present in A and B zones. cation of GeoRTs.
# Description Interpretation Zone
1 Coskinolina-Grainstone Inner ramp B 3.1. Main controls of late-burial corrosion at reservoir scale
2 Coskinolina-Packstone Inner ramp B
3 Coskinolina Hash-Packstone/Wackestone Shallow mid ramp B At the regional scale, the East-fault zone and the A/B Supra-
4 Framestone with platy corals Shallow mid ramp B
unconformity shale seem to have played a critical role in the dis-
5 Floatstone with platy corals Shallow Mid ramp B, A
6 Rotalids-packstone Mid ramp B, A tributions of corrosion-enhanced porosity (CEP) in A Zone and B
7 Nummulitides-Packstone Outer ramp A Zone. Seismic data suggest that Field X comprises extensive breccia
8 Nummulitides-Grainstone/Wackestone Outer ramp A pipes and collapse features (Fig. 7B) that are hundreds of metres in
9 Nummulitides-Wackestone/Packstone Outer ramp A
diameter, and which appear to be associated with the NNW-SSE
10 Shale zones Unconformities A
trending strike-slip fault system. Based on this and earlier geolog-
ical interpretations, it is hypothesised that the East-fault zone was
probably one of the main conduits of the late-burial corrosion fluids
divide the near-wellbore region of the wells into centimetre to to enter Field X (Fig. 7A). Hence it is expected to see an increasing
decimetre scale intervals. Based on the gross depositional model, trend in the intensity of corrosion of the reservoir formation closer
the lithofacies observed in Field X were grouped in to 6 major to this fault zone. Statistical analysis of the digitised CEP logs from
lithofacies in A zone and 6 in B zone. In total, Field X contains 10 the studied wells suggests a good correlation between the distance
main lithofacies types, with some common to both zones (see of wells from the fault zone and the percent of CEP2 type porosity,
Table 1). which is associated with high intensity of burial corrosion (Fig. 7C).
The two main types of stylolite distributions observed in Field X It must be noted that although the breccia pipes can act as major
are; i) low amplitude stylolites (LAS) with low intensity of associ- vertical permeability enhancers, these were not included in the
ated tension gashes and ii) high amplitude stylolites (HAS) with reservoir model built by the operator since they were outside the
high intensity of associated tension gashes. The main types of CEP reservoir model boundaries. This is because the breccia pipes are
observed in Field X are; i) ‘CEP1’, representing the well intervals associated with the East Fault Zone, which was one of the bound-
with high proportions of well-connected chalky microporosity and aries of the reservoir model. On the other hand, the A/B Supra-
some mesoporosity, and ii) ‘CEP2’, which is mainly macroporosity unconformity shale (Fig. 7A) probably acted as an aquitard,
including vuggy, moldic and leached stylolites and tension gash causing the burial derived corrosive fluids to concentrate in the
porosity. The factors and processes that acted as key controls on the matrix of B Zone, resulting in more intense corrosion. This is evi-
intensity and spatial distribution of corrosion-enhanced porosity in denced by the significantly higher proportions of CEP2 type
Field X were reservoir stratigraphy, depositional setting and pres- porosity in B Zone (Fig. 8A), characterised by high proportions of
sure dissolution. In the following sub-sections we discuss the main leached macroporosity caused by advanced and intense corrosion.
Figure 7. (A) Idealised cross-section of Field X illustrating the East fault zone, the A/B Supra-unconformity shale and the flow of burial derived fluids into the reservoir formation
through the fault zone. (B) Collapse breccia pipe associated with the East fault zone observed on seismic data. (C) Histogram of cumulative percentages of CEP types in the studied
wells. Wells G5 and G6, which are closer to the East fault zone compared to wells G1 and G11, show higher percentage of CEP2. CEP2 is characterised by high proportions of leached
macroporosity caused by advanced and intense corrosion.
Figure 8. (A) Histogram of cumulative percentage of corrosion-enhanced porosity (CEP) types showing higher proportions of CEP2 type porosity in B Zone. (B) Cross-plot showing
higher porosity-permeability values in B Zone compared to A Zone. (C) Plot of pore size distributions demonstrating higher pore size distributions in B Zone compared to A Zone.
This hypothesis is also favoured by the field production data and Zone is dominated by inner ramp Coskinolina grainstones and
tracer data, which suggest that B Zone acts as a stratified high developed high amplitude stylolites and associated tension gashes.
permeability zone of much higher reservoir rock quality than A The HAS type stylolites in B Zone allowed the corrosive fluids to
Zone. This inference is supplemented by the comparison of selectively remove the fine grained walls of agglutinated miliolid
porosity-permeability values (Fig. 8B) and pore-size distributions foraminiferas in the early phase of corrosion. At a more advanced
(Fig. 8C) between A Zone and B Zone. Overall, A Zone has a lower stage, the sparite and more coarsely crystalline foraminiferas were
rock quality index and lower pore-size range compared to B Zone. extensively corroded. Hence, B Zone, which is predominated by
HAS and LAS type stylolites, contains higher percentages of CEP1
3.2. Main controls of late-burial corrosion at sub-grid scale and CEP2 compared to A Zone. This contrast in distribution of
porosity types in B Zone and A Zone could have also been amplified
Core and thin-section data suggest a strong correlation between by the effect the of A/B Supra-unconformity shale as discussed
the distribution and types of stylolites and the development of CEP before.
at centimetre to decimetre scale (Fig. 9A and B). A vast majority of
the stylolites and tension gashes are associated with halos of 3.3. Field X GeoRTs
moderately to highly corroded CEP zones that are highly perme-
able, as observed from the probe permeameter measurements. This Overall, 16 GeoRTs in A Zone and 16 in B Zone were identified
supports the hypothesis that stylolites acted as conduits to the flow based on the relationships observed between the main lithotypes,
of the corrosive fluids into the surrounding matrix during the late- presence of leached stylolites and associated tension gashes, and
burial corrosion phase in Field X (Wright and Barnett, 2011). the extent of matrix microporosity and macroporosity (Table 2).
Figure 10A illustrates the role of stylolites in the distribution of The GeoRT logs obtained with the help of core and thin-section
corrosion-enhanced porosity in the adjacent reservoir formation. descriptions were used to divide the near-wellbore region of the
Figure 10B illustrates the links between the depositional and wells into centimetre to decimetre scale intervals. The next section
diagenetic processes in Field X at core scale. The distribution of describes how these GeoRTs were further grouped into NWRTs
stylolite types is strongly related to the lithofacies distribution in based on petrophysical analysis.
the reservoir zones. Stylolitisation was relatively low in the outer
ramp Nummulitic packstones and wackstones of A Zone due to 4. Near-wellbore upscaling of GeoRTs for reservoir simulation
higher clay content. A Zone mainly consists LAS type stylolites and
comprises a much lower percentage of CEP2 type porosity The GeoRTs represent the geological-petrophysical heteroge-
compared to B Zone (Fig. 8A). The millimetre-sized clay seams and neity of the reservoir rock at the pore and core scale. This scale is
microstylolites in A Zone caused only low to moderately intense well below the practical size of a reservoir simulation grid-block. It
corrosion, which resulted in widespread microporosity develop- is hence necessary to efficiently upscale the GeoRTs to the field
ment in these formerly tight cemented limestones. In contrast, B scale so that they can be used in reservoir simulation studies.
Figure 9. (A) Histogram illustrating the percent of corrosion-enhanced porosity types associated with the various stylolite types in Field X. The high amplitude stylolites are
associated with higher amounts of CEP2, characterised by high amounts of leached macroporosity. (B) Correlation between percentage of corrosion-enhanced porosity per metre
and number of stylolites per metre for the studied wells. The extent of corrosion increased with the number of stylolites per metre.
Chandra et al. (2013a & c) demonstrated that NWM can effectively they were further grouped into NWRTs. This was based on petro-
capture and upscale centimetre to decimetre scale geological fea- physical characterisation of the GeoRTs using RCA and SCAL data
tures in large reservoir simulation models and improves the dy- and with consideration to the geometrical modelling aspects of the
namic calibration of geomodels. Near-wellbore models were associated sub-grid scale heterogeneities. For instance, the pres-
generated by recreating the sedimentary processes using a process- ence of high proportions of leached high-amplitude stylolites could
oriented modelling approach (Wen et al., 1998; Nordahl et al., strongly influence the effective flow properties of the simulation
2005). Process-oriented modelling approach formulates deter- grid-block. Hence, these stylolites were modelled explicitly using
ministic geological processes such as migration and deposition of the appropriate geometrical and petrophysical descriptors in the
sedimentological components in a stochastic framework (e.g. near-wellbore upscaling software. Figure 11 illustrates the inte-
Nordahl, 2004; Elfenbein et al. 2005). The depositional structures grated near-wellbore rock-typing and upscaling workflow, high-
were ‘overprinted’ with the appropriate diagenetic features in lighting the steps taken to obtain NWRTs.
these models using an object modelling function (Chandra et al.,
2013c; Dabek and Knepp, 2011). The resulting geometrical 4.1. Porosity and permeability characterisation of GeoRTs
models were populated with porosity and permeability values.
Flow-based upscaling was used to obtain effective porosity and The GeoRT logs from the near-wellbore regions were used in
permeability distributions at reservoir simulation grid-block scale. conjunction with the core plug and probe permeameter data to
To facilitate the input of the GeoRTs into near-wellbore upscaling, estimate porosity and permeability of the GeoRTs in the studied
K-B1 from B Zone (Table 3), containing GeoRTs with no or low

presence of any CEP types recorded much lower porosity-perme-
ability range (Fig. 12).
4.2. GeoRT capillary pressure characteristics
Capillary pressure (Pc) data provides information about pore

throat diameters, pore connectivity and their impact on fluid flow
in the reservoir (Bear, 1972). With respect to reservoir simulation,
the applications of Pc curves are mainly two-fold. First, Pc curves are
used to distribute the initial saturations in the reservoir simulation
gridblocks during initialisation of the simulation model. Second, Pc
curves contribute to the fluid-flow equations in the form of gradi-
ents of the phase pressures (e.g. Fanchi, 2006) that transfer oil
between low and high permeability regions during imbibition
(Schmid and Geiger, 2013). The former is one of the most crucial
contributors to the initial hydrocarbons-in-place calculations in the
reservoir simulation model. In order to ensure that the GeoRTs
reflect the dynamic effects of capillary pressure, they were recon-
ciled with the Mercury Injection-Capillary Pressure (MICP) data
provided by the operator. MICP data from 22 samples were used for
this study. The GeoRTs corresponding to high proportion of
corrosion-enhanced porosity (See Table 2) displayed low threshold
entry pressures of less than 2.25 psia. In contrast, the GeoRTs
associated with no or low corrosion (See Table 2) displayed
threshold entry pressures higher than 100 psia. The GeoRTs asso-
ciated with tight matrix (See Table 2) without any corrosion-
enhanced porosity displayed high initial water saturation values
and steep saturation-height curves. The GeoRTs with CEP1 type
porosity displayed higher initial saturation values compared to
those GeoRTs that contain CEP2. The former GeoRTs were abun-
dantly present in the A Zone and represent intermediate quality
rock with patchy microporosity grading into larger meso-pores. The
Figure 10. (A) Illustration of stylolites and tension gashes acting as conduits to the
GeoRTs with CEP2 type porosity generally represented the best
flow of the corrosive fluids into the surrounding matrix. (B) Illustration of the main
controls of reservoir properties at core-scale and the links established between the
quality rock in the reservoir and were dominated by macro-pores.
lithofacies types, stylolite types and extent of corrosion-enhanced porosity in Field X. These GeoRTs were abundantly present in the B Zone and charac-
terised by broadly shaped saturation curves and low initial water
saturation values. Consequently, the saturation-height functions
wells. As mentioned before, the available core plug data suffers were broadly classified into three main groups based on the
from sample insufficiency due to poor core recovery of the well corrosion-enhanced porosity of the GeoRTs. The saturation-height
intervals containing GeoRTs with high proportion of CEP types. This function curves corresponding to GeoRTs with tight matrix
is because the reservoir rock was mechanically weak as it was without any corrosion-enhanced porosity types and GeoRTs with
strongly altered by corrosive fluids. This resulted in a sample bias of CEP1 and CEP2 are shown in Figure 13.
RCA and SCAL data towards the GeoRTs that mainly contained the
uncorroded lithofacies. However, probe permeameter measure- 4.3. Relative permeability characterisation of GeoRTs
ments supplemented the core plug data and provided an estimate
of the upper and lower bounds of the permeability range. The The relative permeability characteristics of reservoir rock types,
uncertainties associated with the probe permeameter data (Corbett in particular the saturation and relative permeability end-points,
et al., 1999) and the issues related to the measurement of stylolite are fundamentally important for reliable estimation of ultimate
and tension gash porosity-permeability were discussed by Chandra hydrocarbon recovery (Dykstra and Parsons, 1950; Buckley and
et al. (2015). Based on probe permeameter and core plug data, the Leverett, 1952). Unsteady state water-oil and gas-oil relative
GeoRTs were binned into 7 porosity-permeability groups in A Zone permeability analysis results at 2000 psi and 70 F provided by the
and 7 in B Zone (Table 3). The range of probe permeameter mea- operator were used in this study. Figure 14 shows the groups of
surements and the porosity-permeability cross-plots of the relative permeability curves of the GeoRTs based on corrosion-
porosity-permeability groups are shown in (Fig. 12). enhanced porosity types CEP1 and CEP2. Initial water saturations
Due to the highly heterogeneous distribution of the CEP types, and the water relative permeability at critical oil saturation suggest
even at the core scale, a wide range of porosity-permeability values that the GeoRTs with high proportion of CEP1 could be strongly
were obtained for the GeoRTs. Nevertheless, there was a visible water-wet. In contrast, the GeoRTs with high proportion of CEP2
trend of increasing porosity-permeability values that correlate to tend to be intermediate-wet to weakly oil-wet. It was observed that
the increase in intensity of late-burial corrosion. Overall, groups K- the critical oil saturation values of the GeoRTs tend to decrease with
A6 and K-B6 (Table 3) show the highest porosity-permeability the increase in the proportion of CEP2 type porosity. The residual
range in A Zone and B Zone respectively (Fig. 12). These groups oil saturation varied between 0.19 and 0.36 for the samples from
constitute the GeoRTs with higher proportion of CEP2 type the B Zone, in which the oil rim is located. The initial water satu-
porosity, i.e. vuggy and moldic macroporosity associated with ration varied between 0.17 and 0.47 for the samples from the A
leached stylolites and tension gashes. Groups K-A1 from A Zone and Zone and 0.37e0.308 for those from the B Zone. The cross point of
Table 2
Summary of GeoRTs present in A Zone and B zone.
GeoRT A Zone Description GeoRT B Zone Description
A1 Shale B1 Inner ramp Coskinolia grainstone with CEP1

A2 Mid-ramp Rotalid packstone B2 Inner ramp Coskinolia packstone with CEP1
A3 Outer ramp Nummulitide packstone B3 Shallow mid-ramp Coskinolina Hash
packstone-wackstone with CEP1
A 4 Outer ramp Nummulite wackestone-packstone B 4 Mid-ramp-Rotalid-Packstone with CEP1
A 5 Shallow mid-ramp floatstone with platy corals B 5 Inner-ramp Coskinolina grainstone with CEP1 and LAS
A 6 Shallow mid-ramp floatstone with CEP1 B 6 Inner ramp Coskinolia packstone with CEP1 and LAS
A 7 Mid-ramp Rotalid packstone with CEP1 B 7 Shallow mid-ramp Coskinolina Hash
packstone-wackstone with CEP1 and LAS
A 8 Outer ramp Nummulitide packstone with CEP1 B 8 Shallow mid ramp framestone with LAS and CEP1
A 9 Outer ramp Nummulite wackestone-packstone with CEP1 B 9 Shallow mid ramp floatstone with LAS and CEP1
A 10 Shallow mid-ramp floatstone with LAS and CEP1 B 10 Mid-ramp Rotalid packstone with LAS and CEP1
A 11 Mid-ramp Rotalid packstone with LAS and CEP1 B 11 Inner ramp Coskinolia grainstone with HAS and CEP2
A 12 Outer ramp Nummulitide packstone with LAS and CEP1 B 12 Inner ramp Coskinolia packstone with HAS and CEP2
A 13 Outer ramp Nummulite grainstone-wackestone with LAS and CEP1 B 13 Shallow mid-ramp Coskinolina Hash
packstone-wackstone with HAS and CEP2
A 14 Outer ramp Nummulite wackestone-packstone with LAS and CEP1 B 14 Shallow mid ramp framestone with HAS and CEP2
A 15 Shallow mid-ramp floatstone with HAS and CEP2 B 15 Shallow mid ramp floatstone with HAS and CEP2
A 16 Outer ramp Nummulite grainstone-wackestone with HAS and CEP2 B 16 Mid-ramp Rotalid packstone with HAS and CEP2
LAS ¼ Low amplitude stylolites.

HAS ¼ High amplitude stylolites.
CEP1 ¼ Corrosion-enhanced micro- and mesoporosity.
CEP2 ¼ Corrosion-enhanced macroporosity including leached stylolites and tension gashes.
the water-oil relative permeability curves occurred at less than 50% 4.4. Near-wellbore upscaling of GeoRTs
water saturation for over 95% of the SCAL samples. The water
relative permeability at maximum water saturation ranged from Based on the petrophysical analysis discussed in the previous
0.30 up to 0.70 in A Zone and from 0.13 to 0.41 in B Zone. sections and the associated NWM parameters, the GeoRTs were
It is challenging to characterise each GeoRT with its own set grouped into 7 NWRTs in A Zone (Table 4) and 7 in B Zone (Table 5).
of relative permeability curves due to the limited availability of The NWM aspects considered for the grouping of GeoRTs are listed
SCAL data samples. Nevertheless, we were able to obtain some in Tables 4 and 5. The NWRTs were then modelled using the object
general inferences regarding the potential relationship between modelling feature in the NWM software, SBED™ (Chandra et al.,
the extent of late-burial corrosion and the relative permeability 2013c; Chandra, 2014). The geometries of near-wellbore models
characteristics of the GeoRTs. It appears that the samples from in this study were typically 20 cm 20 cm 20 cm in size. The
the wells that are located away from the East-fault zone (Wells main input required for NWM were the spatial and geometrical
G11 and G1) tend to be more water-wet compared to wells parameters and the statistics of the petrophysical properties asso-
which are located closer to the fault zone (Wells G6 and G5) ciated with the corresponding GeoRTs. As an example, Figure 15A
(Fig. 2B). This trend could be associated with the role of the East- illustrates the ‘network rod’ object model template in SBED™,
fault zone in the distribution of corrosion-enhanced porosity in which was used for modelling the NWRTs containing CEP types
Field X, causing higher proportions of CEP2 type porosity in the with micro- and macro-porosity. Table 6 summarises the geomet-
regions closer to the fault zone compared to the flank of the rical parameters applied to the network model template. In the
reservoir. NWRT scenarios with stylolite and tension gashes, various
Figure 11. Summary of the integrated near-wellbore rock-typing and upscaling workflow highlighting the steps taken to obtain Near-Wellbore Rock Types (NWRTs) from GeoRTs,
and modelling and upscaling them using NWU tools. To see this figure in colour, the reader is referred to the web version of this article.
Table 3 ‘U-shaped body’ object model template in SBED™ that was used for
Porosity-permeability groups in A Zone and B Zone. See Table 2 for GeoRT modelling tension gashes.
descriptions.
Core description suggested that the tension gashes were up to
A Zone B Zone 6 mm wide, few mm to several centimetres long and were oriented
GeoRT Porosity-permeability GeoRT Porosity-permeability vertically to sub-vertically, originating from the mechanical
group group discontinuity along the stylolite surfaces and developed nearly
A1 K-0 e e parallel to the direction of vertical stress (Chandra, 2014). Figure 16
A2, A3, A4, A5 K-A1 B4, B10 K-B1 illustrates exemplary SBED™ model realisations with varying
A7 K-A2 B8 K-B2 spatial distributions and orientations of the tension gashes. Two
A6, A8, K-A3 B16 K-B3
cases of tension gash lengths were tested using a constant height of
A11, A12
A10, A9 K-A4 B1, B2, B3, B7 K-B4 2 mm. In the first case, the lengths varied from 1 cm to 5 cm and in
A13, A14 K-A5 B5, B6, B9, B12 K-B5 the second case from 3 cm up to 10 cm. Since the tension gashes
A15, A16 K-A6 B11, B13, K-B6 observed on the core were oriented vertically to sub-vertically, the
B14, B15 dip angle was varied from 0 to 10 (Table 6). The azimuth, i.e.
orientation of the tension gashes with respect to the x-y plane, was
varied from 0 to 90 (Table 6). The spatial distribution of the
tension gashes was controlled in SBED™ by defining the number of
geometrical and spatial distribution aspects of leached stylolites
tension gashes that are randomly distributed in the NWM. In this
and associated tension gashes were superimposed on the back-
example, two cases of tension gash distributions were tested. The
ground matrix. The stylolites were explicitly modelled by applying
first case contained 100 tension gashes per cubic metre and the
a standard deviation of 1 to the roughness of the horizontal layers
second case contained 200 tension gashes per cubic metre. The
of the geometrical model (Chandra, 2014). Figure 15B illustrates the
former and the latter represented low density of tension gashes and
high density of tension gashes, respectively.
The resulting centimetre-decimetre scale near-wellbore
geological models were aimed to be realistic representations of
the geometries and distributions of the core-scale geological fea-
tures corresponding to the GeoRTs, for example the CEP types with
the leached stylolites and tension gashes. Multiple stochastic re-
alizations of high resolution porosity and permeability grids were
produced for each NWRT scenario using the mean, standard devi-
ation, minimum and maximum values of porosity-permeability
distributions within each of the NWRTs. Thus, the numerous
geometrical and petrophysical realisations of the NWRTs were
available to capture the extremes of the geological-petrophysical
heterogeneities in Field X that were observed from well data. The
high-resolution NWRTs at the core scale were then upscaled to the
reservoir scale using single-phase flow based upscaling to obtain
the effective porosity and permeabilities of each NWRT realisation
(Fig. 17). This yielded the upscaled porosity-permeability and Kv/Kh
distributions for the NWRTs, conditioned to the near-wellbore re-
gion. Further details of single-phase flow based upscaling in SBED™
are discussed by Chandra (2014) and references therein.
5. Geological-Porosity Derived Systems (GeoPODS)
The level of reservoir model complexity is generally case

dependent, based on the objectives of the reservoir simulation
study and the associated uncertainties (e.g. Fanchi, 2006). Although
it is computationally not possible to build reservoir simulation
models using grid blocks that resolve the small scale geological
features in Field X explicitly, the upscaled NWRTs from each zone
(Tables 4 and 5) can be used as the reservoir rock types that pre-
serve the small-scale and sub-grid heterogeneities in the reservoir
model of Field X. However, using all NWRTs would still result in a
relatively large number of rock types and Field X would comprise a
total of 12 reservoir rock types, resulting from 6 NWRTs in A Zone
and 6 in B Zone. As a consequence, the complexity of the reservoir
modelling and simulation workflows could increase significantly.
Hence the NWRTs from both A and B Zones were further grouped
into GeoPODS, i.e., the rock types from the two reservoir zones
were binned into corresponding GeoPODS. In effect, A and B Zones
were treated together from this point. The geomodel containing
Figure 12. Porosity-permeability cross-plots for the porosity-permeability groups
from (A) A Zone and (B) B Zone. Groups K-A1 and K-B1 (Table 3), containing no or low
GeoPODS was then used for reservoir simulation studies. Figure 18
presence of any CEP types, show much lower porosity-permeability compared to illustrates the integrated near-wellbore rock-typing and upscaling
groups K-A6 and K-B6 (Table 3) that represent vuggy and moldic macroporosity. workflow, highlighting the steps taken to obtain GeoPODS.
Figure 13. Three main groups of GeoRTs in Field X based on saturation-height functions, corresponding to tight matrix and corrosion-enhanced porosity types CEP1 and CEP2.
Figure 14. Groups of water-oil and gas-oil relative permeability curves of the GeoRTs based on the corrosion-enhanced porosity types CEP1 and CEP2.
The predominant porosity-permeability trends of the effective Rather than using trends from a regression analysis, image logs
porosity and permeability values resulting from near-wellbore could provide a more robust binning of the NWRTs at the wireline
upscaling were identified through regression analysis. The scale but such data were not available for this study. Figure 19A
trends provided the basis for grouping the NWRTs into GeoPODS. illustrates how GeoPODS were classified using the cross-plot
Table 4
Summary of NWRTs present in A Zone. See Table 2 for GeoRT descriptions.
# NWRT GeoRT A Zone Porosity-permeability Saturation-height Near-wellbore modelling aspects

group group (Fig. 13)
0 NWRT-A0 A1 K-A0 e Shale lamina

1 NWRT-A1 A2, A3, A4, A5 K-A1 Tight matrix Tight matrix, multiple porosity-permeability scenarios
2 NWRT-A2 A7 K-A2 CEP1 Multiple realisations of K-A2 overprinted on tight matrix
3 NWRT-A3 A6, A8, A11, A12 K-A3 CEP1 Multiple realisations of K-A3 overprinted on tight matrix
4 NWRT-A4 A10, A9 K-A4 CEP1 Multiple realisations of K-A4 overprinted on KA3
5 NWRT-A5 A13, A14 K-A5 CEP2 Multiple realisations of K-A5 overprinted on KA3
6 NWRT-A6 A15, A16 K-A6 CEP2 Multiple realisations of K-A5, explicitly modelled leached
stylolites and tension gashes overprinted on KA3
Table 5
Summary of NWRTs present in B Zone. See Table 2 for GeoRT descriptions.
# NWRT GeoRT B Zone Porosity-permeability Saturation-height Near-wellbore modelling aspects

group group (Fig. 13)
0 NWRT-B0 B1 K-B0 e Shale lamina

1 NWRT-B1 B4, B10 K-B1 Tight matrix Multiple realisations of KB1 overprinted on tight matrix
2 NWRT-B2 B8 K-B2 CEP1 Multiple realisations of KB2 overprinted on tight matrix
3 NWRT-B3 B16 K-B3 CEP1 Multiple realisations of KB4 overprinted on tight matrix
4 NWRT-B4 B1, B2, B3, B7 K-B4 CEP1 Multiple realisations of KB4 overprinted on KB3
5 NWRT-B5 B5, B6, B9, B12 K-B5 CEP2 Multiple realisations of K-B5, explicitly modelled leached
stylolites and tension gashes overprinted on KB4
6 NWRT-B6 B11, B13, B14, B15 K-B6 CEP2 Multiple realisations of K-B6, explicitly modelled leached
stylolites and tension gashes overprinted on KB4
obtained from near-wellbore upscaled porosity and horizontal Table 6

permeability values. Overall, the NWRTs were binned into four Geometrical parameters used for modelling the corrosion-enhanced matrix porosity
using the network rod and U-shaped body templates.
GeoPODS in Field X: shale, G0, G1 and G2 (Fig. 19A). The GeoPODS
G0 mainly constituted the NWRTs derived from GeoRTs with tight Shape Parameter Mean Standard deviation
matrix while G1 was dominated by NWRTs derived from GeoRTs Network rod Azimuth (degree) 0.00 30.00
with CEP1, i.e. corrosion-enhanced micro and meso-porosity Dip (degree) 0.00 30.00
types. The GeoPODS G2 comprises the NWRTs that correspond Rod length (cm) 3.00 1.00
Chain length (cm) 10.00 2.00
to GeoRTs with high CEP2, which includes leached stylolite and
Rod diameter (cm) 3.00 0.00
tension gash porosity types. The porosity-permeability values Chain diameter (cm) 2.00 0.00
obtained from upscaling the NWRTs were used to calculate the U-shaped body Azimuth (degree) 0.00 90
global permeability transforms representative of each GeoPODS. Dip (degree) 0.00 10
Width 1 (cm) 1 1
The cross-plot used to obtain the correlation between near-
Width 2 (cm) 1 1
wellbore upscaled Kv/Kh anisotropy and horizontal permeability Height (cm) 0.1 0.00
is shown in Figure 19B. This correlation was used to calculate the
vertical permeability of the GeoPODS. High and low case end-
members for the permeability transforms were also obtained for
each of the GeoPODS. These endmembers were used to analyse
how sample bias and incomplete porosity-permeability data at coherent with Lucia's Class 3 for non-vuggy rock-fabric type and
the core-scale could impact the field-scale reservoir simulation the K-transform from regression analysis corresponded to zone
results. GHE 3. The porosity-permeability distribution of G1 was spread
To provide further comparison of our rock types, we analysed between Lucia's class 2 and class 3 and showed equivalence to GHE
them with respect to two common classifications schemes. 4. G2 porosity-permeability distribution varied between Lucia's
Figure 19C compares the GeoPODS porosity-permeability trans- class 1 and class 2 rock fabric types and is spread across GHE 6 and
forms with Lucia's (1983) permeability transforms for non-vuggy GHE 5. The comparison of GeoPODS porosity-permeability relation
carbonate fabrics. Figure 19D superimposes the GeoPODS on the with GHE plot corroborated the presence of three main porosity
‘Global Hydraulic Elements’ (GHE) plot (Corbett and Potter, 2004). types in Field X at the reservoir grid-block scale. Table 7 summa-
The porosity-permeability distribution of GeoPODS G0 was rises petrophysical attributes of the four GeoPODS of Field X.
Figure 15. (A) The ‘network rod’ object template in SBED™ used for modelling NWRTs with CEP types containing micro- and macro-porosity. (B) The ‘U-shaped body’ object
template in SBED™ used for modelling NWRTs with leached tension gashes.
Figure 16. (A) Near-wellbore model representing distribution of corrosion-enhanced porosity surrounded by a tight matrix (B) Example of a near-wellbore model when stylolites
and associated tension-gashes are explicitly modelled. (C) Example model realisations of solution-enhanced porosity and tight matrix scenarios.
Figure 17. Cross-plot between effective porosity and effective horizontal permeability calculated using flow-based upscaling for the various near-wellbore rock types (NWRTs) from
A Zone (A) and B Zone (B). The core plug measurements have also been plotted to provide comparison between the porosity-permeability distributions in Field X before and after
upscaling in both A and B Zones.
Table 7 also lists the saturation-height function and relative distribution of initial-oil-in-place and the dependency of residual
permeability groups corresponding to the GeoPODS, inherited from oil saturation on initial water saturation require better characteri-
the GeoRTs. sation (Chandra, 2014). In order to accomplish this, we ensured that
the porosity-permeability distribution, capillary pressure, and
6. Employing GeoPODS in Field X reservoir simulation relative permeability characteristics are consistent within each
GeoPODS as discussed earlier (Table 7). We next discuss the steps
Not only the reservoir permeability of Field X, but also the initial taken to incorporate the static and dynamic properties of the
hydrocarbons-in-place and the critical oil saturation values are GeoPODS into the reservoir simulator.
associated with high degrees of uncertainty. Due to the capillary
transition zone present in the oil rim, Field X poses additional 6.1. Field X permeability and rock quality index
challenges for reliable characterisation and simulation of the
reservoir behaviour. The unrealistic water relative permeability Since well data from only four wells was available for this
endpoints and local initial saturation changes applied to the orig- study, an entirely new scenario of Field X geomodel based on
inal Field X geomodel to obtain history match suggest that the GeoPODS could not be obtained. Instead, the high resolution
Figure 18. Summary of the integrated near-wellbore rock-typing and upscaling workflow highlighting the steps taken to obtain Geological Porosity Derived Systems (GeoPoDS)
from NWRTs and incorporating them into reservoir simulation studies. To see this figure in colour, the reader is referred to the web version of this article.
porosity model provided by the operator was used to distribute possible uncertainty associated with the porosity model should be
the GeoPODS at the reservoir grid-block scale. Since image log given due consideration while performing a detailed history
data was unavailable for this study, the GeoPODS were distributed matching study.
in the reservoir geomodel using only porosity cut-offs. The
porosity cut-offs (Table 7) for each GeoPODS were obtained by
6.2. Relative permeability curves
reconciling the upscaled porosity-permeability cross-plots
(Fig. 17) with the wireline effective porosity logs in the near-
These relative permeability curve groups (Fig. 14) were averaged
wellbore regions. These porosity cut-offs were then applied to
and assigned to the GeoPODS. Figure 22 shows the averaged water-
the Field X porosity model (Fig. 2B), originally supplied by the
oil and gas-oil relative permeability curves of the GeoPODS
operator, to obtain the GeoPODS model. Following this, the
compared with those originally used for Field X history matching.
respective permeability and Kv/Kh transforms (Table 7) were
The relative permeability curves from each CEP group were aver-
applied to the porosity model to calculate the horizontal and
aged using the commercial software application called SCAL, which
vertical permeability distributions within each GeoPODS in both A
is a plug-in to the reservoir simulator ECLIPSE. The relative
and B Zones. Figure 20 shows the comparison between the
permeability curves were first normalised to remove the effect of
permeability model derived from conventional core porosity-
the different initial water saturation values and residual oil satu-
permeability transform and the permeability model resulting
ration values of the samples of the same group. Then the normal-
from the GeoPODS model. The reservoir rock quality index (RQI) in
ised curves were averaged to obtain a single relative permeability
Field X was calculated as (Amaefule et al., 1993)
curve, which was used as the representative of the group.
rffiffiffiffiffiffiffiffi
k 6.3. Capillary pressure-saturation correlations
RQI ¼ 0:0314 ; (1)
Ф
The three main groups of saturation-height functions (see
where k is permeability and Ф is porosity. Figure 21 illustrates the Fig. 13) that were obtained during the characterisation of GeoRTs in
comparison between the distributions of RQI in the GeoPODS case Section 4.2 were used to obtain the relationship between initial
and the original geomodel. The RQI calculated for the original water saturation and capillary pressure for the GeoPODS. As
geomodel (Fig. 21A) is significantly lower than that for the Geo- mentioned before, at this stage A and B Zones were treated
PODS case (Fig. 21B). together. We tested two methods to derive the relationship be-
The RQI calculated for the GeoPODS inherently captures the tween capillary pressure and saturation to be implemented in
enhanced flow properties caused by late-burial corrosion in Field X reservoir simulation. In the first method we obtained a Pc-Sw
that could not be modelled by conventional rock-typing. In effect, correlation for each GeoPODS based on the correlation between
the RQI distribution of the GeoPODS case implicitly represents the the logarithm of capillary pressure and water saturation (Fig. 23A).
heterogeneity associated with micro- and macro porosity, leached In the second method (Fig. 23B) the relationship between J-func-
stylolites and tension gashes. The characterisation of RQI in Field X tion, Pc and water saturation was used, which is defined as
hence improved due to the implementation of GeoPODS compared
rffiffiffiffiffiffiffiffi
to the original geomodel. Furthermore, this improvement in RQI 0:218 Pc k
characterisation has strong implications towards the calculation of J¼ ; (2)
sCosq Ф
initial water saturation, and subsequently the fluids in place cal-
culations in the simulation model when the J-function approach is
used for saturation modelling. It must be noted that the application Here s and q are interfacial tension and contact angle between
of porosity cut-offs to populate GeoPODS in Field X geomodel relied oil and water at reservoir condition, respectively. These parame-
on the basic assumption that the porosity model given by the ters have been provided by the operator. Thus, the J-function
operator is sufficiently accurate. It is hence recommended that the allowed us to scale capillary pressure to account for the variation
Figure 19. (A) Cross-plot between effective porosity and effective horizontal permeability obtained from NWRTs and the porosity-permeability transforms used to classify Geo-
PODS. (B) Cross-plot between calculated effective horizontal permeability and effective Kv/Kh anisotropy ratio of the GeoPODS G1 and G2. (C) Comparison between the GeoPODS
porosity-permeability crossplot and Lucia's (1983) permeability transforms for nonvuggy fabrics. (D) GeoPODS permeability transforms superimposed on the ‘Global Hydraulic
Element’ (Corbett and Potter, 2004) plot.
in porosity and permeability in the reservoir, i.e, the RQI. Some of curve of the GeoPODS G0 (Fig. 23A) could not represent this
the capillary pressure curves from G0 group, which represents the phenomenon adequately. In contrast, these curves were better
poor quality reservoir rock in Field X, were concave due to the characterised by the J-functions (Fig. 23B), which account for the
micritic nature of the rock. The normalised capillary pressure rock quality index of the GeoPODS.
Table 7
Summary of GeoPODS and their petrophysical attributes input into reservoir simulation. Note that ‘PHIE cut-offs’ represent the porosity cut-offs used to populate GeoPODS in
the porosity model provided by the operator.
GeoPODS Permeability transform Kv/Kh anisotropy Sw-H group Kr group PHIE cut-offs
Shale K ¼ 0:001 Kv ¼ Kh CEP0 CEP0 <0.01

Kv
G0 K ¼ 766:42Ф 3:2229 2
Kh ¼ 8E 07Kh þ 0:0016Kh þ 0:878 CEP0 CEP0 [0.01, 0.05)
Kv
G1 K ¼ 101278Ф 5:0483 2
Kh ¼ 8E 07Kh þ 0:0016Kh þ 0:878 CEP1 CEP1 [0.05, 0.15)
Kv
G2 K ¼ 663749Ф 5:5071 2
Kh ¼ 8E 07Kh þ 0:0016Kh þ 0:878 CEP2 CEP2 >0.15
Figure 20. (A) Permeability model derived from core permeability transform (B) Permeability model obtained from the permeability transforms of the GeoPODS. To see this figure
in colour, the reader is referred to the web version of this article.
Figure 21. (A) Reservoir rock quality index (RQI) distribution calculated using the permeability model derived from core permeability transform (B) RQI distribution calculated
using the permeability model obtained from the permeability transforms of the GeoPODS. The RQI distribution in the GeoPODS case is significantly higher compared to the original
geomodel, especially in the lower layers of the model that correspond to B Zone. To see this figure in colour, the reader is referred to the web version of this article.
6.4. Implications towards fluid-in-place calculations and reservoir The impact of the different capillary pressure- and J-saturation
simulation correlations on the distribution of initial water saturation and
therefore on the fluids-in-place calculation is demonstrated in
The capillary pressure-saturation relationships obtained for the Table 8. A difference of over 150 million STB of oil-in-place and up
GeoPODS using the two different methods discussed in the previ- to 700 million MSCF of gas-in-place was observed. The impact of
ous section were input as tables into the simulation model to test varying capillary pressure and J-function on the reservoir flow
how this impacts the fluids-in-place volumes. The capillary pres- simulation results is shown in Figure 24. The J-function approach
sure- and J-function-saturation correlations were applied to the was tested using two GeoPODS permeability cases, one using the
saturation column from the relative permeability-saturation tables near-wellbore upscaled permeability transform shown in
to calculate the Pc and J functions, respectively, for each of the Figure 19A and one using a high permeability transform case. Recall
GeoPODS. The input of J-function saturation tables was used by the that the impact of high- and low-endmember scenarios for the
simulator to compute the Pc values as GeoPODS was tested. The simulation scenarios that incorporate
GeoPODS and employ the respective J-functions showed better
rffiffiffiffiffiffi agreement of simulated and observed cumulative oil production
Ф compared to the simulation case that did not involve any rock-
Pc ¼ CF J TCosq ; (3)
k typing. The simulation scenario employing J-functions with high
permeability transform case showed up to 22% increment in the
where CF is the conversion factor and for capillary pressure in Psi
cumulative oil production compared to the simulation case without
(CF ¼ 4.61678).
Figure 22. (A) Average water-oil relative permeability curves obtained for the GeoPODS (CEP1 and CEP2) compared with those used for history matching (M17). (B) Average gas-oil
relative permeability curves obtained for the GeoPODS (CEP1 and CEP2) compared with those used for history matching (M17).
upscaling were in better agreement with the historic production

data compared to the scenario without rock-typing. Figure 25
compares the difference between the simulated and historic cu-
mulative fluid volumes in Field X after 22 years of production for
the original geomodel and the updated model that employs
GeoPODS.
After 22 years of production, the simulated cumulative pro-
duction value of the GeoPODS case using J-functions and with
NWM high permeability case showed only 9.46% error compared to
the historic data. In contrast, the error for the original geomodel
was almost 25%. It must be emphasised that the process of history
matching yields a non-unique solution. In a way, similar results
could have been obtained by simply applying permeability multi-
pliers in the simulation model and/or manipulating the relative
permeability end-points or the fault transmissibility values in the
simulator. However, such manipulations, when done without any
geological-petrophysical rationale, impose additional uncertainty
to the model and render it less reliable for reservoir management
and development studies, particularly if production mechanisms
change. The near-wellbore rock-typing and upscaling workflow
discussed in this study hence enables us to improve the charac-
terisation and initialisation of the reservoir model prior to history-
matching. This yielded a fit-for-purpose reservoir simulation model
that incorporates small-scale geological structures in a consistent
way and requires less modifications in order to obtain a good his-
tory match.
7. Conclusions
This study addressed key challenges associated with integrating

reservoir rock-typing and simulation by employing a novel rock-
typing workflow in conjunction with near-wellbore upscaling.
Our near-wellbore rock-typing and upscaling workflow involves
characterisation, modelling and upscaling of key small-scale
geological-petrophysical heterogeneities into reservoir grid-block
scale. This approach was demonstrated in a case study for a high-
ly heterogeneous carbonate reservoir, Field X. We showed that our
workflow improved the reservoir characterisation and simulation
Figure 23. (A) Average capillary pressure-saturation correlation curves obtained for of the field.
the GeoPODS (B) Average J functions obtained for the GeoPODS.
Geological rock types (GeoRTs) were generated with the specific
aim to account for the impact of late-burial corrosion. Late-burial
rock-typing. These differences can be explained as follows: When
corrosion is the key diagenetic phase in Field X and controls the
the J-function correlations are used, the effect of block dependent
evolution of reservoir properties. Petrophysical characterisation of
porosity and permeability is accounted for during initialisation in
the reservoir simulator (Eq. (3)). This approach will produce a
spatially varying transition zone which is not only a result of
spatially varying capillary pressures but also the RQI. The RQI is in
turn conditioned to the distribution of GeoPODS. Therefore, using
the J-function approach that employs GeoPODS in the reservoir
simulation provides internal consistency of the static reservoir
properties, i.e., porosity and permeability, as well as the dynamic
behaviour associated with capillary pressure and relative perme-
ability. Hence saturation modelling and initialisation is more reli-
able in the reservoir simulation model. As a result, the simulation
model predictions involving near-wellbore rock-typing and
Table 8
Sensitivity of volumes of oil- and gas-in-place to capillary pressure-saturation
correlation.
Correlation method STOIIP[*106 STB] GIIP [*106 MSCF]
Capillary pressure-saturation 1138.9 2721.3

Figure 24. Observed and simulated cumulative oil production curves in Field X. The
J-saturation using NWU permeability 965.7 2029
GeoPODS scenarios employing J-saturation method shows better agreement with
J-saturation using NWU permeability 1252.9 2685.2
historic data compared to the simulation model that does not involve a near-wellbore
high case
rock-typing and upscaling workflow.
Technologies, Geological Society special publication, vol. 406. Geological Society

of London, London, pp. 1e59.
Alsharhan, A.S., 1990. In: Brooks, J. (Ed.), Geology and Reservoir Characteristics of
Lower Cretaceous Kharaib Formation in Zakum Field, Abu Dhabi, United Arab
Emirates, Classic Petroleum Provinces. Geological Society Special Publications,
50, pp. 299e316.
Alsharhan, A.S., Sadd, J., 2000. Stylolites in Lower Cretaceous Carbonate Reservoirs.
In: U.A.E. Society of Economic Paleontology and Mineralogy Special Publication,
vol. 69, pp. 185e207.
Amaefule, J.O., Altunbay, M., Tiab, D., Kersey, D.G., Keelan, D.K., 1993. Enhanced
Reservoir Description: Using Core and Log Data to Identify Hydraulic (Flow)
Units and Predict Permeability in Uncored Intervals/Wells. Society of Petroleum
Engineers. http://dx.doi.org/10.2118/26436-MS.
Bear, J., 1972. Dynamics of Fluids in Porous Media. Elsevier, New York.
Calvert, S., Ballay, G., 2011. Basic Logs Unlock Complex Carbonate Pore Properties.
Society of Petrophysicists and Well-log Analysts. SPWLA-2011-MM.
Chandra, V., 2014. Improving Reservoir Characterisation and Simulation Using Near-
wellbore Upscaling (Ph.D. thesis). Heriot-Watt University, Edinburgh.
Figure 25. The percentage error between the historic cumulative production value of Chandra, V., Corbett, P.W.M., Hamdi, H., Geiger, S., 2013a. Improving reservoir
Field X after 22 years of production and the geomodel scenarios with and without characterisation and simulation with near-wellbore modelling. SPE Reserv. Eval.
Eng. 16 (2), 183e193. SPE Journal Paper 148104-PA.
GeoPODS.
Chandra, V., Steele, R., Milroy, P., Corbett, P.W.M., Geiger, S., 2013b. Using near-
wellbore modelling and dynamic calibration to improve permeability model-
ling in a giant carbonate field. In: Proceedings of 75th EAGE Conference &
Exhibition Incorporating SPE EUROPEC 2013. TU 14 15.
GeoRTs allowed us to group them into near-wellbore rock types Chandra, V., Geiger, S., Corbett, P.W.M., Steele, R., Milroy, P., Barnett, A., Wright, P.,
(NWRTs), which were then upscaled to the reservoir grid-block Jain, P., 2013c. Using near-wellbore upscaling to improve reservoir characteri-
scale using the near-wellbore upscaling workflow. The upscaled sation and simulation in highly heterogeneous carbonate reservoirs. In: Pro-
ceedings of SPE Reservoir Characterisation and Simulation Conference,
porosity-permeability cluster was used to group the NWRTs with September 2013. SPE 166033.
same porosity-permeability trends into Geological-Porosity Chandra, V., Wright, P.V., Barnett, A., Steele, R., Milroy, P., Corbett, P.W.M.,
Derived Systems (GeoPODS). GeoPODS are equivalent to upscaled Geiger, S., Mangione, A., 2015. Evaluating the Impact of a Late-burial
Corrosion Model on Reservoir Permeability and Performance in a Mature
plug-scale Hydraulic Flow Units and help to further define the Carbonate Field Using Near-wellbore Upscaling. Fundamental Controls on
Global Hydraulic Elements concept at the appropriate modelling Fluid Flow in Carbonates: Current Workflows to Emerging Technologies. In:
scale for full field simulation models. The resulting GeoPODS sce- Geological Society, London, Special Publications, pp. 427e445. http://
dx.doi.org/10.1144/SP406.11.
narios hence incorporate the small-scale depositional and diage- Chilingarian, G.V., Torabzadeh, J., Rieke, H.H., Metghalchi, M., Mazzullo, S.J., 1992.
netic heterogeneities in Field X. The GeoPODS were ultimately used Interrelationships among surface area, permeability, porosity, pore size, and
to populate the reservoir geomodel and ensure that the static and residual water saturation. In: Chilingarian, G.V., Mazzullo, S.J., Rieke, H.H. (Eds.),
Carbonate Reservoir Characterization: a Geologic-Engineering Analysis, Part I,
dynamic reservoir properties are internally consistent within each
Developments in Petroleum Science, vol. 30. Elsevier Publ. Co., Amsterdam,
GeoPODS. The porosity-permeability transforms, saturation-height pp. 379e397.
functions and relative permeability curves individually tailored for Corbett, P.W.M., Anggraeni, S., Bowen, D., 1999. The use of the probe permeameter
each GeoPODS were employed in the reservoir simulation model in carbonates e addressing the problems of permeability support and statio-
narity. Log Analyst 40, 316e326.
for Field X. The updated reservoir simulation model was hence Corbett, P.W.M., Potter, D.K., 2004. Petrotyping: a basemap and atlas for navigating
initialised consistently, providing an improved calculation of fluid through permeability and porosity data for reservoir comparison and perme-
in-place volumes. Simulation results indicate that reservoir simu- ability prediction. In: Paper Presented at International Symposium of the So-
ciety of Core Analysts, SCA2004-30.
lation models that use GeoPODS in conjunction with J-functions Dabek, L.B., Knepp, R., 2011. Bioturbation and its effects on permeability in wave-
yield cumulative production curves that agree well with the his- dominated shoreface rocks of the Spring Canyon Member, Blackhawk Forma-
toric production data. tion, Utah, USA. AAPG Search and Discovery Article # 50425. In: Proceedings of
AAPG Annual Convention and Exhibition, Houston, Texas, USA, April 10e13,
Overall, our study illustrates a new way of integrating multi- 2011.
scale static and dynamic data pertinent to heterogeneous carbon- Dunham, R.J., 1962. Classification of carbonate rocks according to depositional
ate reservoirs seamlessly into a reservoir simulation model for texture. In: Hamm, W.E. (Ed.), Classification of Carbonate Rocks, a Symposium.
American Association of Petroleum Geologists, pp. 108e121.
improved characterisation and dynamic calibration. The new Dykstra, H., Parsons, R.L., 1950. The prediction of oil recovery by waterflooding. In:
simulation model is hence much better constrained to the reservoir Secondary Recovery of Oil in the United States, second ed. API, Washington, DC,
geology and provides an improved geological-prior for subsequent pp. 160e174.
Elfenbein, C., Ringrose, P.S., Christie, M., 2005. Small-scale reservoir modeling tool
history matching.
optimizes recovery offshore Norway. World Oil 226 (10), 45e50.
Esteban, M., Taberner, C., 2002. Reconstructing fluid flow evolution from
compressional to extensional regimes. In: Late corrosive fluids: a key factor for
Acknowledgements porosity formation and enlargement. Proceedings of AAPG Hedberg Confer-
ence, Austin, p. 234.
We acknowledge Foundation CMG for supporting the PhD Esteban, M., Taberner, C., 2003. Secondary porosity development during late-burial
in carbonate reservoirs as a result of mixing and/or cooling of brines. J. Geo-
project of Viswasanthi. We would like to extend our thanks to Alex chem. Explor. 78e79, 355e359.
Assaf and Zoe Watt from BG Group, Mohamed Ahmed, Zeyun Jiang Fanchi, J.R., 2006. Fundamentals of reservoir simulation. In: Fanchi, John R. (Ed.),
and Eric Mackay from Heriot-Watt University for their valuable Principles of Applied Reservoir Simulation, third ed. Gulf Professional Pub-
lishing, Burlington, ISBN 9780750679336, pp. 162e186. http://dx.doi.org/
input into the project. Special thanks to the BG asset team for
10.1016/B978-075067933-6/50012-X.
providing the necessary data and support required for the project. Gomes, J.S., Ribeiro, M.T., Strohmenger, C.J., Negahban, S., Kala, M.Z., 2008. Car-
We also acknowledge CMG for IMEX, Schlumberger Information bonate Reservoir Rock Typing e the Link Betwen Geology and SCAL. SPE Paper,
p. 118284.
Solutions for access to Petrel and ECLIPSE, and to Geomodeling for
Honarpour, M.M., Chilingarian, G.V., Mazzullo, S.J., 1992. Permeability and relative
access to SBED™. permeability of carbonate reservoirs. In: Chilingarian, G.V., Mazzullo, S.J.,
Rieke, H.H. (Eds.), Carbonate Reservoir Characterization: a Geologic-
Engineering Analysis, Part I, Developments in Petroleum Science, vol. 30.
References Elsevier Publ. Co., Amsterdam, pp. 39x9e416.
Hollis, C., Vahrenkamp, V., Tull, S., Mookerjee, A., Taberner, C., Huang, Y., 2010. Pore
Agar, S.M., Geiger-Boschung, S., Jan 2015. In: Agar, S., Geiger, S. (Eds.), Fundamental system characterisation in heterogeneous carbonates: an alternative approach
Controls on Fluid Flow in Carbonates: Current Workflows to Emerging to widely-used rock-typing methodologies. Mar. Pet. Geol. 27, 772e793.
Jodry, R.L., 1992. Pore geometry of carbonate rocks and capillary pressure curves Schmid, K.S., Geiger, S., 2013. Universal scaling of spontaneous imbibition for
(basic geologic concepts). In: Chilingarian, G.V., Mazzullo, S.J., Rieke, H.H. (Eds.), arbitrary petrophysical properties: water-wet and mixed-wet states and Han-
Carbonate Reservoir Characterization: a Geologic-engineering Analysis, Part I, dy's conjecture. J. Pet. Sci. Eng. ISSN: 0920-4105 101, 44e61. http://dx.doi.org/
Developments in Petroleum Science, vol. 30. Elsevier Publ. Co., Amsterdam, 10.1016/j.petrol.2012.11.015.
pp. 331e377. Skalinski, M., Kenter, J.A.M., 2015. Carbonate Petrophysical Rock Typing: Integrating
Kazemi, A., Corbett, P.W.M., Wood, R., 2012. New approach for geomodeling and Geological Attributes and Petrophysical Properties while Linking with Dynamic
dynamic calibration of carbonate reservoirs using porosity determined system Behaviour. Fundamental Controls on Fluid Flow in Carbonates: Current Work-
(PODS). In: Presented at 74th EAGE Conference and Exhibition, Copenhagen, flows to Emerging Technologies. In: Geological Society, London, Special Publi-
Denmark, 4e7 June 2012. cations, vol. 406, pp. 229e259.
Lucia, F.J., 1983. Petrophysical parameters estimated from visual description of Van der Land, C., Wood, R.A., Wu, K., van Dijke, M.I.J., Jiang, Z., Corbett, P.W.M.,
carbonate rocks: a field classification of carbonate pore space. J. Pet. Technol. 35, Couples, G.D., 2013. Modelling the permeability evolution of carbonate rocks.
626e637. Mar. Pet. Geol. 48, 1e7. http://dx.doi.org/10.1016/j.marpetgeo.2013.07.006.
Mazullo, S.J., 2004. Overview of porosity evolution in carbonate reservoirs. Kans. Wardlaw, N.C., 1996. Factors affecting oil recovery from carbonate reservoirs and
Geol. Soc. Bull. 79, 1e2. prediction of recovery. In: Chilingarian, G.V., Mazzullo, S.J., Rieke, H.H. (Eds.),
Moshier, S.O., 1989. Development of microporosity in a micritic limestone reservoir, Carbonate Reservoir Characterization: a Geologic-Engineering Analysis, Part II,
Lower Cretaceous, Middle East. Sediment. Geol. 63, 217e240. Developments in Petroleum Science, vol. 44. Elsevier Publ. Co., Amsterdam,
Nordahl, K., Ringrose, P.S., Wen, R., 2005. Petrophysical characterisation of a het- pp. 867e903.
erolithic tidal reservoir interval using a process-based modeling tool. Pet. Wen, R., Martinius, A.W., Naess, A., Ringrose, P.S., 1998. Three-dimensional
Geosci. 11 (1), 17e28. simulation of small-scale heterogeneity in tidal deposits e a process-based
Nordahl, K., 2004. A Petrophysical Evaluation of Tidal Heterolithic Deposits: stochastic method. In: Proceedings of 4th Annual Conference of the Inter-
Application of a Near Wellbore Model for Reconciliation of Scale Dependent national Association of Mathematical Geology. Ischia De Frede, Naples,
Well Data (Ph.D. thesis). Norwegian University of Science and Technology, pp. 129e134.
Trondheim. Wright, V.P., Barnett, A.J., 2011. Burial Corrosion and Porosity Formation: from the
Oates, M., Mishra, G., Sultana, N., Nath, G., 2012. Resolving a permeability mismatch Seismic to Micropore Scale, but what Processes Do We Blame? Carbonate
in a mature carbonate field. In: Proceedings of SPE Oil and Gas India Conference Geochemistry: Reactions and Processes in Aquifers and Reservoirs, vol. 16. Karst
and Exhibition, Mumbai, March 2012. SPE Paper, p. 155458. Waters Institute Special Publication, pp. 81e83.

Chandra - Effective Integration of Reservoir Rock Typing and Res Simulation

Uploaded by

Copyright:

Available Formats

Chandra - Effective Integration of Reservoir Rock Typing and Res Simulation

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chandra - Effective Integration of Reservoir Rock Typing and Res Simulation

Uploaded by

Copyright:

Available Formats

Marine and Petroleum Geology 67 (2015) 307e326

Contents lists available at ScienceDirect

Marine and Petroleum Geology

Effective integration of reservoir rock-typing and simulation using

1. Introduction 2013; Skalinski and Kenter, 2015). Another common challenge is

petrophysical properties and behaviour during hydrocarbon re-

1.1. Case study: rock-typing and simulation challenges in Field X

The key porosity types present in Field X, such as chalky micro-

K-B1 from B Zone (Table 3), containing GeoRTs with no or low

4.2. GeoRT capillary pressure characteristics

Capillary pressure (Pc) data provides information about pore

GeoRT A Zone Description GeoRT B Zone Description

A1 Shale B1 Inner ramp Coskinolia grainstone with CEP1

LAS ¼ Low amplitude stylolites.

5. Geological-Porosity Derived Systems (GeoPODS)

The level of reservoir model complexity is generally case

# NWRT GeoRT A Zone Porosity-permeability Saturation-height Near-wellbore modelling aspects

0 NWRT-A0 A1 K-A0 e Shale lamina

# NWRT GeoRT B Zone Porosity-permeability Saturation-height Near-wellbore modelling aspects

0 NWRT-B0 B1 K-B0 e Shale lamina

obtained from near-wellbore upscaled porosity and horizontal Table 6

Shale K ¼ 0:001 Kv ¼ Kh CEP0 CEP0 <0.01

upscaling were in better agreement with the historic production

This study addressed key challenges associated with integrating

Correlation method STOIIP[*106 STB] GIIP [*106 MSCF]

Capillary pressure-saturation 1138.9 2721.3

Technologies, Geological Society special publication, vol. 406. Geological Society

You might also like

Correlation method STOIIP[106 STB] GIIP [106 MSCF]