Aguilera 2014

REE165360 DOI: 10.
2118/165360-PA Date: 30-April-14
Stage:
Page: 190
Total Pages: 19
Flow Units: From Conventional to

Tight-Gas to Shale-Gas to Tight-Oil
to Shale-Oil Reservoirs
Roberto Aguilera, SPE, Schulich School of Engineering, University of Calgary
Summary
Core data from various North American basins with the support of
limited amounts of data from other basins around the world have
shown in the past that process speed or delivery speed (the ratio of
permeability to porosity) provides a continuum between conventional, tight-, and shale-gas reservoirs (Aguilera 2010a). This work
shows that the previous observation can be extended to tight-oil
and shale-oil reservoirs. The link between the various hydrocarbon
fluids is provided by the word petroleum in the total petroleum
system (TPS), which encompasses liquid and gas hydrocarbons
found in conventional, tight, and shale reservoirs.
Results of the present study lead to distinctive flow units for
each type of reservoir that can be linked empirically to gas and oil
rates and, under favorable conditions, to production decline. To
make the work tractable, the bulk of the data used in this paper
has been extracted from published geologic and petroleum-engineering literature.
The paper introduces an unrestricted/transient/interlinear transition flow period in a triple-porosity model for evaluating the rate
performance of multistage-hydraulically-fractured (MSHF) tightoil reservoirs. Under ideal conditions, this flow period is recognized
by a straight line with a slope of 1.0 on log-log coordinates. However, the slope can change (e.g., to 0.75), depending on reservoir
characteristics, as shown with production data from the Cardium
and Shaunavon formations in Canada. This interlinear flow period
has not been reported previously in the literature because the standard assumption for MSHF reservoirs has been that of a pseudosteady-state transition between the linear flow periods.
It is concluded that there is a significant practical potential in
the use of process speed as part of the flow-unit characterization
of unconventional petroleum reservoirs. There is also potential for
the evaluation of production-decline rates by the use of the tripleporosity model presented in this study.
Introduction
Different hydrocarbons and reservoir types can be integrated
under the umbrella of a TPS. That is the premise for being able to
integrate, in this paper, flow units of conventional, tight-gas,
shale-gas, tight-oil, and shale-oil reservoirs and to estimate potential production rates. A previous paper (Aguilera 2010a) described
flow units in tight- and shale-gas reservoirs. That material is
extended in this work to the cases of tight oil and shale oil.
The names TPS (Magoon and Schmoker 2000) and petroleum
system (Magoon and Beaumont 1999) are used interchangeably
in the literature. The TPS is a unifying concept that encompasses
all the disparate elements and processes of petroleum geology
including a pod of active source rock and all genetically related
oil and gas accumulations.
The TPS includes all the geologic elements and processes
required for an oil-and-gas accumulation to exist. The word
petroleum includes high concentrations of any of the following
C 2014 Society of Petroleum Engineers
Copyright V
This paper (SPE 165360) was accepted for presentation at the SPE Western Regional &
AAPG Pacific Section Meeting, 2013 Joint Technical Conference, Monterey, California, USA,
1925 April 2013, and revised for publication. Original manuscript received for review 16 April
2013. Revised manuscript received for review 27 November 2013. Paper peer approved 16
December 2013.
190
substances: thermal and biological hydrocarbon gas found in conventional reservoirs, as well as in unconventional reservoirs (gas
hydrates, tight reservoirs, fractured shales, and coal); condensates;
crude oils; and natural bitumen. The word system describes the
interdependent elements and processes that form the functional
unit that creates hydrocarbon accumulations.
Magoon and Beaumont (1999) indicate that the essential elements of a TPS include the following: source rock, reservoir rock,
seal rock, and overburden rock. The TPS includes two processestrap formation and generation/migration/accumulation of
hydrocarbons. These essential elements and processes must be
correctly placed in time and space so that organic matter included
in a source rock can be converted into a petroleum accumulation.
A TPS exists wherever all these essential elements and processes
are known to occur or are thought to have a reasonable chance or
probability to occur.
The segments of the TPS described previously, dealing with
conventional and unconventional oil and gas reservoirs, are the
primary objectives of this paper. The significant paradigm shift is
that tight rocks that could not produce any petroleum in the past
or were nearly impermeable seals are now economical reservoir
rocks.
Several researchers (Archie 1950; Kwon and Pickett 1975;
MacKenzie 1975; Chopra et al. 1987; Ebanks 1987; Gunter et al.
1997a,b; Hartmann and Beaumont 1999; Nelson 2009; Clarkson
et al. 2012) have discussed the importance of pore and throat
structure (e.g., size, geometry, distribution, connectivity, and
composition) with respect to the flow unit and storage capacity of
porous media. Pore-throat apertures have been estimated on the
basis of knowledge of process speed [i.e., the ratio of permeability
to porosity (Kolodzie 1980; Aguilera 2002)]. In turn, these porethroat apertures have been used with reasonable success to anticipate flow rates that can be expected from given oil (Martin et al.
1997) or gas (Deng et al. 2011) wells. In general, this type of
work has been performed in the past in conventional carbonate
and siliciclastic reservoirs, and, more recently, in tight-gas and
shale-gas reservoirs. This paper shows, with the use of real data,
that the same concept can be extended quantitatively to the cases
of tight-oil and shale-oil reservoirs. To achieve that, data from
several conventional reservoirs around the world, from tight-gas
and from shale-gas reservoirs, and tight and oil reservoirs primarily in North America are examined in this study.
Aguilera (2010a) has indicated that tight-gas reservoirs are
best represented by at least dual-porosity models, whereas shalegas reservoirs are best represented by at least quadruple-porosity
models and more rigorously by quintuple-porosity models (Lopez
and Aguilera 2013). The approach permits estimating volumes of
petroleum in place that are larger than considered previously
(Aguilera 2010a), differentiating between viscous- and diffusiondominated flow in gas reservoirs and the contribution of each flow
mechanism with the use of a unified diffusion-/viscous-flow
model (Rahmanian et al. 2013). It must be noted, however, that
smaller volumes of original gas in place (OGIP) than those considered previously have also been discussed in the literature
(Ambrose et al. 2010).
An important component of the multiple porosities mentioned
previously is provided by natural fractures. All unconventional
reservoirs have to be hydraulically fractured. However, the tightgas, shale-gas, tight-oil, and shale-oil reservoirs with the best
May 2014 SPE Reservoir Evaluation & Engineering
ID: jaganm Time: 16:07 I Path: S:/3B2/REE#/Vol00000/140007/APPFile/SA-REE#140007
REE165360 DOI: 10.2118/165360-PA Date: 30-April-14
WORLD GAS RESOURCE PYRAMID

Reservoir Rock
1000 md
Endowment
Present
10 md
0.1 md
Increasing:
Decreasing:
Production
Costs and
Prices
Delivery
Speed
(k/)
Activation
Indexes
Pore
Throat
Apertures
Research
Unconventional Gas
Stage:
Page: 191
Total Pages: 19
Although process speed has been used for several decades in

conventional reservoirs in the oil-and-gas industry, the proposed
use of this concept for distinguishing between flow units in conventional and unconventional petroleum reservoirs is new. Comparisons of real data and theoretical simulations at the pore-throat
level are consistent and provide good support to the proposed
methodology. To make the results of the interpretations presented
in this paper tractable, the bulk of the core, cuttings, and production data used in the study have been extracted from referenced/
published geologic and petroleum-engineering literature.
Time
Future
Tight Gas
Shales Gas
CBM
Gas Hydrates
Bottom of Resource Pyramid Unknown
Aguilera et al., WPC, 2008, SPE 132845, SPE 162717
WORLD OIL RESOURCE PYRAMID
Present
Gravity
45 API
Endowment
30 API
Increasing:
Decreasing:
Production
Costs and
Prices
Oil Mobility
(k/)
Activation
Indexes
15 API
Unconventional Oil
Research
Time
2
0.1
Oil sands
Tight Oil
(bituminous sands)
2
0.1
Future
Heavy Oil
Oil Shale
Bottom of Resource Pyramid Unknown
Fig. 1Estimates of global natural-gas and oil endowment

(Aguilera et al. 2008, 2009, 2012, GFREE team).
production wells are associated with natural fractures. Examples

of natural-fracture characterization in tight-gas sandstones by
integrating mechanics and diagenesis have been presented by
Olson et al. (2009).
Fig. 1 shows worldwide gas-resource pyramids that include
the global endowment for conventional gas, tight-gas, shale-gas,
coalbed-methane, conventional-oil, tight-oil, shale-oil, and oilsands reservoirs. Endowment, as used in this paper, is the summation of cumulative gas production, reserves, and undiscovered
gas. The total natural-gas endowment, excluding natural-gas
hydrates, is gigantic, at approximately 68,000 Tcf, of which 70%
is estimated to be in tight and shale gas. Tight-gas endowment in
the United States and Canada is estimated to be 450 and 105 Tcf,
respectively. These volumes represent only 7% of the OGIP.
Shale-gas endowment in the United States has preliminary estimates of approximately 240 and 70 Tcf by the GFREE team
[GFREE refers to a multidisciplinary approach including geoscience (G); formation evaluation (F); reservoir drilling, completion,
and stimulation (R); reservoir engineering (RE); and economics
and externalities (EE)]. The endowment of natural gas, the cleanest-burning fossil fuel, will supply market needs for several decades. The pyramid shows that unconventional gas is associated
with very low permeabilities. As such, successful production of
unconventional gas will include increases in production, prices,
activation indices, research, and time. The activation index (Economides and Oligney 2000, p. 83) is a measure of the total investment required to establish access to new oil or gas expressed in
dollars per unit volume per day (e.g., USD/B/D or USD/Mcf/D)
of stabilized production. The pyramid also shows decreases in
process (or delivery) speed and pore-throat apertures, which are
some of the key topics discussed in this paper.
Process (or Delivery) Speed

Process speed or delivery speed (i.e., the ratio of permeability to
porosity) provides a relative indication of storage and how quickly
fluids can move through porous media. The concept has been
shown to be powerful for characterizing conventional oil-and-gas
reservoirs in various lithologies (Chopra et al. 1987; Gunter et al.
1997a,b), for predicting recoverable hydrocarbon volumes (Pickett and Artus 1970), and for determining flow units in tight-gas
and shale-gas reservoirs (Aguilera 2010a).
The flow unit (a function of permeability and porosity) is thus
a useful concept for linking geology, petrophysics, and reservoir
engineering because permeability and porosity are studied in
detail and used by all these disciplines (Aguilera 2004). The process speed (k/) is an important part of the diffusivity equation:
r2 p
lct @p
; . . . . . . . . . . . . . . . . . . . . . . . . . . 1
k=/ @t
where m is fluid viscosity and ct is total compressibility. Eq. 1 is at

the heart of fluid-flow calculations in porous media by reservoir
engineers. Thus, it is extremely important to have knowledge of
the process speed (k//) and hydraulic diffusivity [g mct/(k//)].
Although this paper concentrates on process speed, it is important
to note that the viscosity (m) and total compressibility (ct) play a
very important role on the fluid flow through porous media. For
example, production-decline data for the Bakken reservoir are
presented later in this paper in the section dealing with tight-oil
and shale-oil reservoirs.
Flow Units
Process speed is related directly to flow (or hydraulic) units, a
concept introduced by H.D. Winland (Kolodzie 1980) with data
from formations ranging in age from Ordovician to Tertiary,
including Simpson, Delaware, Tensleep, Nugget, Cotton Valley,
Muddy, Mesaverde, Terry, First Wall Creek, Frontier, Montrose,
Vicksburg, and Frio sandstone. Windlands concept has been used
successfully by MacKenzie (1975) in the Cardium sandstone of
Canada, Kolodzie (1980) in the Spindle field in Colorado, Pittman
(1992) in sandstone reservoirs of the United States, Ebanks
(1987) as an aid to reservoir description for engineering projects,
Aguilera (2010a) for flow-unit determination in tight-gas and
shale-gas reservoirs, and Clarkson et al. (2012) for the evaluation
of the Montney formation in Canada. A flow unit is defined as a
stratigraphically-continuous reservoir subdivision characterized
by a similar pore type (Hartmann and Beaumont 1999, pp. 97).
As such, the concept is powerful for helping define optimal layering in simulation work.
Pore-throat aperture (rp35) in micrometers can be calculated
from (Aguilera 2002, 2004):

0:45
k
: . . . . . . . . . . . . . . . . . . . . . 2
rp35 2:665
100/
The equation was developed with data from more than 2,500
sandstone and carbonate samples from the Aux Vases, Hoover,
Dakota, Nesson, Judith River, Lodgepole, Nisku dolomite, Morrow and Keyes, Hunton, Granite Wash, Venango, Cypress, Mission Canyon, Cherokee, Bartlesville, Stony Mountain, Swift,
Muddy, Tar Springs, Minnelusa, Red River, Desmoines, Devonian,
Benois, Trenton limestone, Silurian, and Edwards formations. The

191
1.E+00
1
0.55
micro
0.2
meso
1.E01
macro
Knudsen
1s
100s
0.1s
10s
0.01s
1s
1.0s
100s
0.1s
50s
0.05s
10s
0.01s
5s
0.005s
1s
1.E02
0.04
1.E04
0.01
1.E05
0.003
1.E06
0.0004
1.E07
1.E08
Diffusion Dominated Flow
1.E03
nano
Permeability (mD)
100s+
MSHF Horizontal
1.E+01
10s
MMscfd
Mbopd
1.E+02
Total Pages: 19
Vertical Wells
20
10
Page: 192
MMscfd
1.E+03
Viscous Flow
rp35
mega
1.E+04
Mbopd
CHART FOR ESTIMATING PORE THROAT APERTURE

(Extension to Conventional, Tight and Shale Petroleum)
Stage:
0.00008
1.E09
1.E10
0
10
15
20
25
30
Porosity (%)
Source: GFREE Research Team, U of Calgary, 2013.
Fig. 2(a) Flow units as a function of pore-throat radii (rp35) in micrometers, porosities (%), and permeabilities (md); and possible
ranges of oil rates (thousands of BOPD) for vertical wells published by Martin et al. (1997) and gas-flow rates (millions of scf/D) for
vertical wells published by Deng et al. (2011). Figure includes possible ranges of oil and gas rates for MSHF horizontal wells. All
the rates are only approximations and can be affected by individual rock characteristics. For example, the presence of well-developed natural fractures in the Bakken can increase the oil rates shown in the figure. The Knudsen number allows distinguishing
between viscous and diffusion-like flow in tight- and ultratight-gas reservoirs. However, the scale presented in Fig. 26 is only for
pore-throat apertures (not for Knudsen values).
data had been used originally by Kwon and Pickett (1975) for creating a pore-structure model and developing pore-structure
interrelationships.
Pore-size classes are grouped on the basis of pore-throat (port)
apertures as megaports (rp35 > 10 mm), macroports (2.5 to 10
mm), mesoports (0.5 to 2.5 mm), microports (0.1 to 0.5 mm), and
nanoports (0.01 to 0.1 mm) by following approximations suggested by Martin et al. (1997). Note that port prefixes (mega,
macro, meso, micro, nano), as used customarily in the geologic
literature, do not correspond to the mathematical meaning of such
prefixes.
Fig. 2a shows a crossplot of permeability vs. porosity for various pore-size classes (rp35) that include conventional, tight, and
shale petroleum reservoirs, and possible rates that can be
obtained from vertical oil wells (Martin et al. 1997) and vertical
gas wells (Deng et al. 2011). The figure also includes direct
observations of possible oil and gas rates that can be obtained
from MSHF horizontal wells. The figure suggests that if MSHF
horizontal wells with smaller rp35 are producing much better than
vertical wells with larger rp35, there is significant potential to
increase the rates and ultimate recoveries from some of those reservoirs with a larger rp35 with innovative drilling and stimulation
technologies as have been performed in tight and shale petroleum
systems.
However, all the rates in Fig. 2a are only approximations that
should be used with care because the rates can be affected by individual rock and fluid characteristics. Note in Eq. 1 that not only
does process speed play an important role in fluid flow but so also
do total compressibility (which can be affected significantly by
192
the presence of fractures), viscosity, and pressure. For example,

the presence of well-developed natural fractures in the Bakken
can increase the oil rates shown in the figure by factors between
approximately five and ten. The lines of constant rp35 in Fig. 2a
were developed with the use of Eq. 2. For comparison, note that
the radius of a methane molecule is 0.0002 mm, the radius of
water is 0.000137 mm, and the radius of oil ranges between
approximately 0.00025 and 0.005 mm.
Fig. 2a also shows an approximate separation between viscous- and diffusion-like flows. Viscous flow is present when the
architecture of the rock is dominated by megaports, macroports,
mesoports, and sometimes microports. Diffusion flow, on the
other hand, is observed sometimes during gas production at the
nanoport level. The approximate boundary between viscous- and
diffusion-dominated flow is estimated with the dimensionless
Knudsen number (Kn), which is defined as the ratio of the molecular mean gas mean free path length (k) to the pore diameter (d). A
methodology for using the Knudsen number on tight formations
has been presented by Ziarani and Aguilera (2012).
Corroboration of the validity of Eq. 2 used to generate Fig. 2a
has been demonstrated with the use of simulation at the porethroat level, as shown in Fig. 2b (Rahmanian et al. 2010).
Although the interpretation of pore-size classes described previously has been carried out mostly in conventional rocks in the
past, this paper shows that process (or delivery) speed provides a
continuum between conventional, tight-gas, shale-gas, tight-oil,
and shale-oil reservoirs. The surprising result, on the basis of core
data from various North American basins, leads to distinctive
flow units for each type of reservoir.
rp35
10
100
4
2
10
1
1
0.5
0.2
0.1
0.01
0.001
0.025
0.014
nanoports
0.04
microports
0.1
Total Pages: 19
0.0001
AA
A A
AA A A
A
A A A
A
A
A A
A A
AA
A A A BA
A
A
B
A
B
B
BA B
B
B
B
B
B
B B B
B BBB B
B BB
B
B
B
B
B
BBB B BB B BB
B
B
B
B
BC C
B
B
CC C
C
C C
CC
C
C
C C
C CC
C C CCCC
C
CCC
C CC
C
C
CC C
C
C CCD
DDDCD
D D DD
C D D
D
D
DD D
DD
DD D
D
D
D D
D D DD
DD
D DD
DD D DD
D
D
D
PENNSYLVANIA
SANDSTONES
AND
CONGLOMERATES,
OKLAHOMA
A = TYPE IA
B
B = TYPE IB
C = TYPE IC
D = TYPE ID
= TYPE II
+ = TYPE III
= CONGLOM
100
GAS PERMEABILITY MILLIDARCIES
20
1000
Page: 193
1000
megaports macroports mesoports
10000
Permeability, kmax (mD)
Stage:
10
1.0
0.1
0.01
0.00001
0.001
3
0.000001
0
10
15
20
Porosity (%)
rt : 1 4 micro meters
rt : 0.05 0.5 micro meters
25
30
rt : 0.1 1.5 micro meters

r t : 0.009 0.1 micro meters
Fig. 2(Continued) (b) There is consistency between porosities

and permeabilities from core analysis for conventional, tightgas, and shale-gas reservoirs (see Figs. 4 through 8) and porosities and permeabilities calculated from pore-scale modeling
shown previously. The empirical flow units represented by rp35
lines (Kolodzie 1980; Aguilera 2010a) are supported by porescale modeling calculations for pore throats (rt) ranging
between 1 and 4, 0.1 and 1.5, 0.005 and 0.5, and 0.009 and 0.1
mm (Rahmanian et al. 2010).
Conventional Petroleum Reservoirs

Fig. 2a has provided reasonable results compared against the average values of permeability and porosity for 10,481 worldwide carbonate reservoirs considered by Ehrenberg and Nadeau (2005).
The comparison (Aguilera 2006) indicates that 90% (P90) of the
reservoirs in all petroleum-producing countries (except Canada,
because the Ehrenberg and Nadeau data bank for Canada does not
include permeabilities) have pore-throat apertures (rp35) with a
maximal size of approximately 2.5 mm (mesoports); 50% of the
reservoirs have pore-throat apertures with a maximum of approximately 5 mm (macroports), and only 10% of the reservoirs have
pore-throat apertures with a maximum of approximately 10 mm
(megaports).
Fig. 2a has also provided reasonable results compared against
average values of permeability and porosity for 30,122 worldwide
siliciclastic reservoirs considered by Ehrenberg and Nadeau (2005).
At the P50 and P90 levels, the relative drop in siliciclastics porethroat apertures is more significant compared with the carbonate
case. This might be the result of a larger amount of fracturing and
microfracturing in carbonate rocks compared with siliciclastic reservoirs (Ehrenberg and Nadeau 2005; Aguilera 2006).
The results are significant because they indicate that more than
90% of the reservoirs have pore-throat sizes that are smaller than
2.5 mm (as in tight-gas and shale-gas reservoirs). This compares
well with the global-resource pyramids (Fig. 1) for oil and gas
that show that the size of the pore throats decreases continuously
as the size of the pyramid, toward the bottom, becomes larger.
The bottom of the resource pyramid is unknown, and this opens
7
10
15
POROSITY PERCENT
20
30
Fig. 3Porosity/permeability crossplot showing pore types of

Pennsylvanian sandstones and conglomerates, Elk City oil
field, Oklahoma (Sneider et al. 1983).
significant possibilities for the future of unconventional petroleum

resources throughout the world.
The validity of the rp35 approach is demonstrated with the use
of Fig. 3 that includes data published in Sneider et al. (1983), a
classic paper in which the authors developed an empirical qualitative methodology for evaluating drill cuttings with the use of a
binocular microscope at 20X magnification. The approach proved
to be of significant practical value for the economic development
of tight/clastic gas reservoirs in the Deep basin of the Western
Canada Sedimentary basin (WCSB).
To the authors knowledge, the data shown on Fig. 3 have
never been evaluated by the use of the rp35 method. Thus, it provides a good test for the flow-unit methodology presented in this
paper for the case of conventional reservoirs. The figure shows a
crossplot of permeability vs. porosity for Pennsylvanian sandstones and conglomerates of the Elk City field in Oklahoma. This
is a giant/conventional oil field in which Sneider et al. indicate
that porosity, permeability and pore geometry are related to grain
size and sorting, cementation and compaction, consolidation and
the amount of pore-filling clays.
Sneider et al. (1983) identified three types of rocks. (1) Type I
rocks are capable of producing petroleum without any type of
stimulation. Sneider et al. (1983) subdivided Type I rocks into A
(k >100 md), B (10 to 100 md), C (1 to 10 md), and D (0.5 to 1
md). (2) Type II rocks are capable of producing petroleum when
interbedded with Type I rocks or when affected by natural fractures and/or hydraulic fractures. Sneider et al. (1983) indicated
that Type II rocks have permeabilities larger than 0.07 md to 0.5
to 1 md. (3) Type III rocks are too tight to produce at commercial
rate even in the presence of some natural fractures and/or hydraulic fractures. Sneider et al. (1983) indicated that Type III rocks are
characterized by permeabilities smaller than 0.07 md. Note that
generally for all rock types (e.g., Type IC represented by the letter
C in Fig. 3), there is a general tendency for the data to display a
lower permeability as porosity decreases.
Fig. 4 shows the Sneider et al. (1983) data on the template
developed for the rp35 method. The figure clearly shows that the

193
10000
Total Pages: 19
rp35
microns
4.5
IA
E
TYP
2.3
B
I
PE
10
TY
TYPE ID
IC
P
TY
TYPE II
TYPE III
0.5
0.05
0.01
10
15
20
Porosity (%)
25
microports
0.1
mesoports
2
1
30
Fig. 4Porosity/permeability crossplot showing different rp35

flow units of Pennsylvanian sandstones and conglomerates,
Elk City oil field, Oklahoma. Data points from Sneider et al.
(1983) are the same as presented in Fig. 3.
Sneider et al. rock types can be properly placed into separate flow
units on the basis of the rp35 values. In this case, the rp35 template
captures properly the general tendency of permeability to
decrease with porosity for each rock type. The Type III rock could
not produce commercially when Sneider et al. published their paper, in 1983. These types of tight rocks, however, are now capable
of commercial production in several places in Canada and the
United States, thanks to technological innovations that include
horizontal drilling and multistage hydraulic fracturing.
Another conventional oil example is provided by the prolific
Cardium sandstone in Canada with the use of data published by
MacKenzie (1975) for the Pembina oil field. As in the previous
case, as far as I know, MacKenzies data have never been evaluated with the rp35 method. Fig. 5 shows MacKenzies data on the
rp35 template for his Type I, Type II, and Type III rocks. Note that
MacKenzies definitions of Types I, II, and III are different from
the definitions presented by Sneider et al. (1983). I retain their
original definitions in this paper to maintain consistency with their
work. What is important, from a practical point of view, is that in
both cases the rp35 values allow one to distinguish clearly unique
flow units in the Elk City (United States) and Pembina (Canada)
oil fields. On Fig. 4, the porosity and permeability data are presented as published by Sneider et al. (1983) without any modifications. The same holds true for Type I and Type II rocks published
by MacKenzie (1975).
The ranges of porosity and permeability for Cardium tight
rocks (equivalent to MacKenzies Type III) represented by open
squares in Fig. 5 were published by Hamm and Struyk (2011).
MacKenzie (1975) indicated that rocks with k// smaller than
0.05 corresponded to tight/nonproductive rock. From Eq. 2, a
value of k// equal to 0.05 corresponds to a pore radius (rp35)
equal to 0.69 mm. This value is represented by the green-dashed
line in Fig. 5. The comparison between the results of MacKenzies cutoff value, published in 1975, and the ranges of porosity
and permeability for tight Cardium rocks, published by Hamm
and Struyk in 2011, is truly remarkable. MacKenzie published a
figure with values of k// for Type III rocks but not separate values of porosity and permeabilities. Thus, the open circles for
Type III rocks in Fig. 5 are limited by the porosity and permeability ranges published by Hamm and Struyk (2011). The tight-oil
Cardium rocks, previously noncommercial, are now commercial
1.0E+02
Type I
Permeability (mD)
macroports
Permeability (mD)
15
megaports
E
ERAT
)
GLOM
CON CIRCLES
(RED
100
194
Page: 194
rp35
1000
0.001
Stage:
1.0E+01
Type II
1.0E+00
1.8
0.69
Type III
1.0E01
F
HR, B
1.0E02
0.20
0.04
1.0E03
0
12 15 18 21 24 27 30
Porosity
Fig. 5Porosity/permeability crossplot showing different rp35

flow units of Cardium sandstones, Pembina oil field, Canada.
Data points for rock of Type I (asterisks) and Type II (red circles)
were tabulated originally by MacKenzie (1975). Porosities and
permeabilities for Type III rocks (purple circles) were extracted
from k// data published by MacKenzie (1975). Ranges of porosities and permeabilities for tight-oil Cardium (open squares)
were published by Hamm and Struyk (2011).
in many instances. Tight-oil production and a new unrestricted

interlinear flow period for the tight Cardium are discussed later in
this paper.
Tight-Gas and Shale-Gas Reservoirs
Shales are defined in different ways by different organizations.
For example, the Energy Resources Conservation Board of
Alberta defines shale in Section 1.020(2) 27.1 of the Oil and Gas
Conservation Regulations as a lithostratigraphic unit having less
than 50% by weight organic matter, with: less than 10% of the
sedimentary clasts having a grain size greater than 62.5 micrometers; and more than 10% of the sedimentary clasts having a
grain size less than 4 micrometers (Province of Alberta 2013). In
some cases, the definition is simpler, and it considers only the size
of the fine-grained clastic sedimentary particles that make up the
rock (e.g., less than 0.0625 mm). The mineralogy of clays is quite
complex (Shaw and Weaver 1965) and includes, for example,
quartz (21.5%), feldspar (4.5%), clay minerals (66.9%), iron
oxides (less than 0.5%), carbonates (3.6%), other minerals (less
than 2%), and organic carbon (1%).
In the case of shale-gas formations, natural gas is generated in
the shale and remains within the shale. Consequently, the shale is
both source rock and reservoir rock. In commercial shale reservoirs, the shale is an excellent source rock. An important characteristic of both tight-gas and shale-gas reservoirs is that they
extend across very large areas, thus forming part of continuous
accumulations (Law 2002). Smocker (2005) has defined a continuous petroleum accumulation as those oil or gas accumulations
that have large spatial dimensions and indistinctly defined boundaries, and which exist more or less independently of the water
column. Grains and pores are smaller in shales compared with
tight and conventional gas formations.
Gas is trapped and stored in shale in different ways: (1) as gas
adsorbed on and dissolved into the kerogen material (Javadpour
et al. 2007), (2) as free gas trapped in nonorganic interparticle
(matrix) porosity, (3) as free gas trapped in microfracture porosity, (4) as free gas stored in hydraulic fractures created during the
stimulation of the shale reservoir, and (5) as free gas trapped in a
pore network developed within the organic matter or kerogen
(Ruppel and Loucks 2008; Wang and Reed 2009). Item 5 has significant/practical implications that can help explain the largerthan-anticipated gas rates and recoveries of natural gas from
some of these formations (e.g., from the Devonian shales of the
Appalachian basin). The different types of storage suggest that
shale-gas reservoirs should be represented, at least, by quadrupleMay 2014 SPE Reservoir Evaluation & Engineering
1.0E02
Permeability (mD)
rp35
microns
0.04
1.0E03
FAYETTVILLE
0.025
HORN RIVER
BARNETT
1.0E04
0.014
SOFT SHALES
1.0E05
0
10
12
14
16
Porosity
Fig. 6Permeability vs. porosity crossplot including data from

Horn River and soft shallow Cretaceous shales in Canada and
from the Fayetteville and Barnett shales in the United States.
Symbols of Fayetteville data, including blue triangles, correspond to individual wells. Green diamonds and black dots correspond to the Horn River (Muskwa formation) basin and soft
shales, respectively (Bustin et al. 2008); brown dots correspond
to the Barnett formation. Fayetteville data points fall below
0.025 mm at porosities smaller than 2% (Aguilera 2010a).
porosity models (Aguilera 2010a; Andrade et al. 2011) and more

rigorously by quintuple-porosity models (Lopez and Aguilera
2013).
Fig. 6 is a semilogarithmic crossplot of permeability vs. porosity that includes data from Horn River basin and soft Cretaceous
shallow shales in Canada, and Fayetteville and Barnett shales in
the United States. Lines for different values of rp35 (0.014, 0.025,
and 0.04 mm) were developed with the use of Eq. 2. The lines provide a reasonable fit to the data. However, it is acknowledged that
the lack of a well-defined protocol for estimating permeability
and porosity in shales is a limitation in the work process to make
these comparisons fully valid. In spite of this limitation, the rp35
results indicate that the values are, in all probability, at least directionally correct. The shale-gas flow units in Fig. 6 are used continuously for comparison against the tight-gas, shale-oil, and tightoil reservoirs discussed in this paper.
In the case of tight formations, natural gas is generated somewhere else (usually in a shale or coal) and migrates to the tight
reservoir in which it is trapped and stored in interparticle (matrix
porosity), slot, and microfracture porosity. This suggests that
Stage:
Page: 195
Total Pages: 19
tight-gas reservoirs should be represented at least by dual-porosity

models and preferentially by triple-porosity models. The use of a
triple-porosity model that permits estimating values on intergranular porosity, slot porosity, and/or fractures, and isolated or noneffective porosity has been demonstrated by Deng et al. (2011).
Fig. 7 reproduces the same data shown on Fig. 6 but also
includes cores and drill-cuttings data from the tight-gas Nikanassin formation (Upper Jurassic and Lower Cretaceous) in the
WCSB. Clearly, the tight-gas sandstone forms separate and distinctive flow units compared with shale-gas reservoirs. Permeabilities from drill cuttings are consistently lower than those in cores,
suggesting that some of the microfractures and slots are not preserved in cuttings because of the action of the drill bit on rocks.
Fig. 8 is a repeat of Fig. 7 but with added data from the Utica
shale gas in Quebec. Thus, also, imperfect data from shale-gas
reservoirs tend to form distinctive flow units for shale- and tightgas reservoirs.
The comparison is a good example of the continuum between
conventional, tight-gas, and shale-gas reservoirs presented in this
paper.
Tight-Oil and Shale-Oil Reservoirs

As in the case of tight gas and shale gas, tight oil and shale oil also
form part of the continuous accumulations. In the case of tight oil,
the oil is generated somewhere else and migrates into the tight formation. In the case of shale oil, the oil is generated in the shale and
remains within the shale. Thus, in this respect, the famous Bakken
(United States and Canada), although generally called a shale oil
reservoir (DOE 2012), is in reality a tight-oil reservoir. In fact, oil
was generated in the lower and/or upper Bakken shales and
migrates into the middle Bakken that has proved very prolific, particularly in North Dakota, although it is also found as a continuous
accumulation in South Dakota and Montana (United States) and in
Saskatchewan and Manitoba (Canada).
In the case of the Canadian Bakken, some of the oil might
have been generated in the general areas in which it is found, but
it is likely that a significant amount of the oil might have been
generated deep in the basin in the North Dakota area and might
have migrated north to the shallower Bakken in Canada. This is
suggested by the maturity levels of Bakken TPS presented in the
west/east schematic cross section shown in Fig. 9.
Although this paper concentrates on process speed (k//), it is
important to note from Eq. 1 that viscosity (m) and total
1.0E+01
rp35
microns
NIKANASSIN CORE
1.0E+00
0.55
1.0E01
Permeability (mD)
NIKANASSIN CUTTINGS
0.15
1.0E02
0.04
1.0E03
F
0.025
HR, B
1.0E04
0.014
SOFT SHALES
UTICA
1.0E05
0
10
12
14
16
Porosity
Fig. 7Permeability vs. porosity crossplot including shale data from Horn River (HR) and soft shales in Canada and from the
Fayetteville (F) and Barnett (B) in the United States, also shown on Fig. 6. The red dots in the upper part of the figure represent
core data from the Nikanassin tight-gas formation in Canada. The black squares with red crosses in the middle, and the red triangles, are data from Nikanassin drill cuttings (Solano 2010; Ortega and Aguilera 2013).
195
Stage:
Page: 196
Total Pages: 19
1.0E+01
rp35
microns
NIKANASSIN CORE
1.0E+00
Permeability (mD)
0.55
1.0E01
NIKANASSIN CUTTINGS
0.15
1.0E02
UTICA
1.0E03
0.04
F
0.025
HR, B
1.0E04
0.014
SOFT SHALES
UTICA
1.0E05
0
10
12
14
16
Porosity
Fig. 8Permeability vs. porosity crossplot including shale data from Horn River (HR) and soft shales in Canada and from the
Fayetteville (F) and Barnett (B) in the United States, also shown on Fig. 6. The red dots in the upper part of the figure represent
core data from the Nikanassin tight-gas formation in Canada. The black squares with red crosses in the middle, and the red triangles, are data from Nikanassin drill cuttings (Solano 2010; Ortega and Aguilera 2013). The yellow circles represent data from the
Utica shale in Quebec (Lavoie et al. 2011).
compressibility (ct) play a very important role in fluid flow

through porous media. For example, Bakken oil viscosities range
between 0.15 and 0.45 cp at reservoir conditions in some parts of
the Williston basin in the United States, and initial pressures range
between 5,000 and 8,000 psi. In Canada, the viscosities can be of
the same order of magnitude in some instances, but the pressures
can be smaller (e.g., 2,300 psi) compared with Bakken overpressures in thermal/mature areas. There is also much-lower gas content in migrated hydrocarbons. This will have a significant effect
on performance and oil recovery.
Furthermore, the Bakken is considered generally as a naturally
fractured tight-oil reservoir in North Dakota (Sonnenberg 2011).
The host permeability can be enhanced by natural fractures and
slot porosity stemming from dolomitization. This is the case of
the Elm Coulee, Parshall, and Sanish pools in North Dakota.
However, the evidence of natural fractures is not as clear in the
shallower Bakken present north of the border in Saskatchewan
and Manitoba, Canada. The presence or lack of natural fractures
can have a significant effect on total compressibility, and hence,
on oil recovery, because open/natural fractures tend to be more
compressible than the host rock.
Fig. 10 introduces a crossplot of permeability vs. porosity on
the rp35 template for the Bakken tight-oil reservoir in the United
States and Canada. The data for Bakken Foghorn, Brutus, and Jack-
son Rowdy were extracted from Almanza (2011). The big, red triangle highlights sweet spot properties (permeability 0.15 md,
porosity 6%) in the Bakken, as established by Sonnenberg
(2011). In this case, porosity and permeability can be much larger
as a result of dolomitization, natural fractures, and slot porosity.
Note also that porosities and permeabilities tend to be larger in the
shallower Saskatchewan Bakken in Canada. However, because of
the lack of natural fractures, the productivity of the Bakken Saskatchewan wells is generally smaller than the productivity of the
Bakken North Dakota wells. For comparison, Fig. 10 also includes
data from Horn River (HR) and soft shales in Canada and from
the Fayetteville and Barnett in the United States, included previously in Fig. 6. Different pore-throat apertures (rp35) are clearly
delineated.
Fig. 11 shows a crossplot of permeability vs. porosity for Cardium conventional oil reservoirs of Type I and Type II compared
with Cardium tight-oil reservoir (Type III). The data for Types I,
II, and III, published originally by MacKenzie (1975) and Hamm
and Struyk (2011), were also highlighted in Fig. 5. The Cardium
conventional reservoir has been one of the most prolific in Canada. Most recently, the tight-oil Cardium has been coming of
age. The variation in pore-throat aperture (rp35) can be associated
with different production profiles discussed later in this paper.
Fig. 11 highlights differences in pore-throat apertures between
Bakken Petroleum System
Ma
De
diso
vo
nia
nG
Charles Formation
rou
Mission Canyon Fm.
ole
gep
Source Rock
(Upper and Lower Shales)
Lod
Bakken
Three Forks
Reservoirs:
Middle Bakken & Three Forks
MATURITY
Nisk
Overpressure
Source Beds:
Upper & Lower Bakken Shales
Fig. 9Cross section of the Bakken petroleum system showing approximate maturity levels (Sonnenberg 2011). It is likely that
some of the oil generated in the Bakken in North Dakota migrated north (contrary to this page) to the Canadian Bakken.
196

Stage:
Page: 197
Total Pages: 19
rp35
microns
1.0E+02
4.5
1.0E+01
1.8
SASK
1.0E+00
0.55
Permeability (mD)
BRUTUS
1.0E01
FOGHORN
0.22
JACKSON
1.0E02
0.09
0.04
1.0E03
0.025
0.014
HR, B
1.0E04
0.004
1.0E05
1.0E06
0
12
15
18
21
24
27
30
Porosity
Fig. 10Permeability vs. porosity crossplot for the Bakken tight-oil reservoir in the United States and Canada. The big, red triangle
highlights sweet spot properties in the Bakken, as established by Sonnenberg (2011). Data for the Foghorn, Brutus, and Jackson
pools were extracted from Almanza (2011). In the Bakken, porosity and permeability can be larger because of dolomitization as in
the Elm Coulee, Parshall, and Sanish pools. The figure also includes, for comparison, shale data from Horn River (HR) in Canada
and from the Fayetteville (F) and Barnett (B) in the United States, also shown in Fig. 6.
the Cardium and shale-gas reservoirs. Each rock type corresponds clearly to different flow units. The data for the shale-gas
reservoirs, shown initially in Fig. 6, include those of the Horn
River in Canada and the Fayetteville and Barnett in the United
States.
Fig. 12 includes permeability and porosity data from the Monterey shale in California. Some data were published by Freeman
and Eller (2010) and were sourced from Venoco and Occidental.
The Monterey shale is composed of fractured chert and siliceous shales [with a very complex lithology that goes from Opal
rp35
microns
1.0E+02
4.5
Type I
1.8
1.0E+01
Permeability (mD)
Type II
1.0E+00
0.69
Type III
1.0E01
0.20
1.0E02
0.04
1.0E03
0.025
0.014
HR, B
1.0E04
0.004
1.0E05
1.0E06
0
12
15
18
21
24
27
30
Porosity
Fig. 11Permeability vs. porosity crossplot for Cardium conventional oil reservoirs of Type I and Type II in Canada compared with
Cardium tight-oil reservoir (Type III); plot highlights differences in pore-throat apertures. Each rock type corresponds clearly to different flow units. Cardium data extracted from MacKenzie (1975) and Hamm and Struyk (2011). The figure also includes, for comparison, shale data from Horn River (HR) in Canada and from the Fayetteville (F) and Barnett (B) in the United States, also shown
on Fig. 6.
197
Stage:
Page: 198
Total Pages: 19
rp35
microns
1.0E+02
4.5
1
LOWER OXY
1.0E+00
Permeability (mD)
1.8
VENOCO C
1.0E+01
1.0E01
VENOCO A
VENOCO B
UPPER OXY
0.55
MONTEREY
0.09
1.0E02
0.04
0.025
EAGLEFORD?
1.0E03
F
0.014
HR, B
1.0E04
0.004
1.0E05
1.0E06
0
12
15
18
21
24
27
30
Porosity
Fig. 12Permeability vs. porosity crossplot for the Monterey shale in California (source of data: Freeman and Eller 2010). Possible
Eagleford information is included. The figure also shows, for comparison, shale data from Horn River (HR) in Canada and from the
Fayetteville (F) and Barnett (B) in the United States, also shown on Fig. 6.
A (i.e., unaltered diatomite) to Opal CT (i.e., cristobalite tridymite) at larger depths] of the Upper Miocene (Regan 1953). There
are clearly different flow units for the shale-gas and shale-oil reservoirs in Fig. 12. The shale-gas samples show pore-throat radii
(rp35), varying between 0.014 and 0.004 mm. On the other hand,
62.5% of the Monterey samples show rp35 ranging between 0.55
and 2.2 mm, 31.3% between 0.15 and 0.25 mm, and 6.2% between
0.04 and 0.15 mm. Liquids require larger pore-throat apertures for
commercial production because of the larger viscosity of oil as
compared with natural gas. And, the more viscous the oil, the
larger must be the pore-throat apertures required to attain commercial production. In the case of the Monterey, the oil ranges
between less than 6 API and more than 30 API. For comparison
purposes, Fig. 12 also includes the same data shown on Fig. 6 for
the Horn River Basin in Canada; and Fayetteville and the Barnet
shale gas in the United States.
Fig. 13 shows permeability and porosity data developed by
Walls et al. (2011) for the Eagleford shale in Texas by use of an
integrated digital rock physics (DRP) process for analyzing the
rock properties of shales and other unconventional reservoirs. Data
rp35
microns
Permeability (mD)
1.0E+03
WELL A-DRP
10.0
WELL B-GRI-AR-AR
1.0E+02
4.5
1.0E+01
1.8
1.0E+00
0.55
5WELL A-GRI-DS
1.0E01
0.09
1.0E02
EAGLEFORD ?
1.0E03
1.0E04
0.04
0.025
HR, B
0.014
0.004
1.0E05
1.0E06
0
12
15
18
21
24
27
30
Porosity
Fig. 13Permeability vs. porosity crossplot for the Eagleford shale (Wells A-DRP, A-GRI-DS, and B-GRI-AR-AR). Source of data:
Walls et al. (2011). DRP 5 digital rock physics, GRI 5 Gas Research Institute, AR 5 as received, DS 5 Dean-Stark extracted and
dried. Additional possible Eagleford information is included (green triangle). The figure also shows, for comparison, shale data
from Horn River (HR) in Canada and from the Fayetteville (F) and Barnett (B) in the United States, also shown on Fig. 6.
198

Stage:
Page: 199
Total Pages: 19
rp35
microns
1.0E+02
CT I
4.5
MONTEREY
1.8
1.0E+01
CT II
SHENGLI
1
BAKKEN
VP
2 14
CT T
CP
4
SSK
SWEET SPOT
SM
BSASK2
BSASK3
VSK
5
1.0E01
FOGHORN
SHAUNAVON
BSASK1
UTICA
1.0E02
VRW
EAGLEFORD(?)
BRUTUS
JACKSON MARCELLUS
1.0E03
Permeability (mD)
1.0E+00
0.55
0.22
0.09
0.04
0.025
F
0.014
HR, B
1.0E04
1.0E05
0.004
UTICA
UPPER & LOWER B
1.0E06
0
12
15
18
21
24
27
30
Porosity
Fig. 14Composite of permeability vs. porosity crossplot for tight-oil and shale-oil reservoirs suggests pore-throat apertures
(rp35) ranging between approximately 0.09 and 4.5 mm. The figure includes data from the Shengli field (China), Cardium (CT I, CT II,
CT T), Monterey, Shaunavon and Bakken in Saskatchewan (BSASK), Bakken in the US (Foghorn, Brutus, Jackson), and Viking in
Canada (VSK, VRW). Also included are the Marcellus (red squares) and the Utica (two different patterns of yellow circles between
approximately 0.004 and 0.014 mm, and 009 and 0.55 mm). The lower part of the figure presents, for comparison, shale data from
Horn River (HR) in Canada and from the Fayetteville (F) and Barnett (B) in the United States, also shown on Fig. 6. At the bottom of
the figure are data for the Upper and Lower Bakken (B).
from Gas Research Institute (GRI) crushed sample analysis tend to

show lower permeability than the DRP results in the lower-porosity
range. Walls et al. (2011) indicate that the difference in porosity/
permeability trends between the two methods will be the subject of
further study. There is also a data point for the Eagleford shale (a
green triangle followed by a question mark) taken from Freeman
and Eller (2010). The question mark stems from the fact that the
author is not sure as to the validity of the data point. If the data
point is correct, a comparison with other tight-oil reservoirs discussed in this paper would suggest very-rapid production declines
for those wells with this porosity/permeability characteristic even if
the wells are drilled horizontally and hydraulically fractured in
multiple stages. Because there are wells that perform much better,
it is likely that there are areas with better rock characteristics, as
indicated by Walls et al. (2011). The Eagleford shale and the Austin chalk are considered as a TPS (Martin et al. 2011) in which the
Eagleford shale is the source for the naturally fractured Austin
chalk. Fig. 13 also includes the same data shown on Fig. 6 for the
Horn River Basin in Canada and for the Fayetteville and Barnett
shale gas in the United States.
Sleeping Giants
Compared with other flow units in shale and tight-oil reservoirs
discussed in this paper (Fig. 14), pore-throat apertures (rp35),
porosities, and permeabilities suggest that the Monterey is, in all
probability, a sleeping giant. This assertion is increased when the
Monterey area (approximately 1,650 sq miles) and thickness
(1,000 to 3,000 ft) are taken into account. Under these circumstances, it is reasonable to anticipate that, with proper GFREE
management, this combination of area, thickness, porosity, permeability, pore-throat apertures, large American Petroleum Institute (API) gravities in some areas, and innovative technology will
most likely lead to a gigantic recovery of unconventional Monterey oil.
As for natural gas, the flow units of the Utica shale in Quebec
presented in Figs. 8 and 14 suggest rock quality comparable to
other prolific shale-gas reservoirs. Ryder (USGS 2008) indicates
that, based on black shale reservoirs in the Utica shale of the St.
Lawrence Lowlands of Quebec (Aguilera 1978), a hypothetical
Utica shale reservoir is proposed in this report for the United
States parts of the Appalachian basin. The Utica Shale reservoirs
in Quebec are self-sourced, fractured, have porous zones that
range in thickness from 5090 ft, and have water saturations that
approach zero. Furthermore, fracture porosity for the Utica Shale
reservoir in Quebec averages 1.4%, and the reservoir pressure is
generally normal (Aguilera 1978). Natural fractures have been
observed in outcrop and in core for the Utica Shale in New York
State (Martin 2005). The 1.4% fracture porosity determined in
1978 compares well with more-recent values published in the literature, including 1.7% for the Marcellus, 1.5% for the Barnett,
and 1.2% for the Haynesville (Wang and Reed 2009). Because the
Utica shale in the US has an area that is larger than the Marcellus
and a thickness that is also larger than the Marcellus (Fig. 15), the
conclusion is reached that the Utica shale has a very large value
of petroleum in place.
Notice in Fig. 14 that the pore-throat apertures (rp35) of the
Marcellus (red squares) and the upper pattern of the Utica (yellow
circles) compare reasonably well. Furthermore, note that the Utica
in the US is underlain by basement rock. On the basis of the
authors experience, there is potential in the basement, if naturally
fractured, because the formation of natural fractures creates dilatancy and a vacuum that tends to suck the fluids (natural gas and
oil) present in the surroundings into the fractures. As in the case
of the Monterey, it is reasonable to anticipate that, with proper
GFREE management, the combination of area, thickness, porosity, permeability, pore-throat apertures, and innovative technology
will most likely lead to a gigantic recovery of unconventional
petroleum (oil and gas) in the Utica shale.
Cumulative-Production Distribution
Although shales are very heterogeneous and have multiple
porosities (Aguilera 2010a; Andrade et al. 2011; Lopez and
Aguilera 2013), the cumulative production of shale-gas reservoirs is significantly and surprisingly more homogeneous than

199
Stage:
Page: 200
Total Pages: 19
Generalized Cross Section

Utica and Marcellus Shale
New York to Pennsylvania
Utica Shale Gas Play

Geographic extent of potential source rock.
Marcellus Shale boundary in yellow.
NY
Pa
Oh
Utica Source Rock

Marcellus Boundary
Depth Below Sea Level (Thousand Feet)
B
NY PA
Sea Level
Mar
cell
us S
Utic
Naturally
10
hale
aS
hal
Fractured
Reservoir?
Basement
20
A
B
30
64
(C) SERVIPETROL LTD., Calgary,

Canada, 2011
Fig. 15Areal extension and cross section covering the Marcellus (yellow line) and Utica shales (green border) in the United
States. The Utica is thicker and larger than the Marcellus. Furthermore on the basis of the authors experience, there is also potential in the basement, if naturally fractured, because the formation of natural fractures creates a vacuum that tends to suck the fluids
(in this case, oil and natural gas) present in the surroundings. Thus, the Utica shale is a likely sleeping giant below the Marcellus
(Source of maps: King 2014).
the cumulative-gas-production distribution of tight-gas and

naturally fractured gas and oil reservoirs. Fig. 16 shows a production-variability plot of fractional cumulative gas vs. fractional cumulative number of production wells for the Barnett
shale-gas reservoir (Aguilera 2010b). The six curves represent
different operating companies. If the shales were completely homogeneous, all the curves would fall in the dashed 45 straight
line. The six curves separate from the dashed straight line, indicating a certain amount of heterogeneity.
Fig. 17 is a production-variability plot for the Nikanassin
tight-gas formation in six different areas of the Deep Basin
(WCSB) (Gonzalez and Aguilera 2011). The data show that, as
the fracture density decreases, the curvature (and, thus, the level
of heterogeneity) increases. The separation from the 45 dashed/
straight line is more significant in Fig. 17 than in Fig. 16, indicating a larger degree of heterogeneity in tight-gas reservoirs com-
pared with shale reservoirs. The result is surprising but corroborated by actual production data. At this time, there are not
enough production data from tight- and shale-oil reservoirs to reach
a conclusion with respect to the distribution of their cumulative oil
production.
Production Rates
The original attempt to correlate production rates and pore-throat
apertures was published by Martin et al. (1997). Pore-size classes
were grouped by these researchers as indicated previously in the
Flow Units section. Martin et al. (1997) indicated that, comparatively, megaports can reach medium-gravity oil-production rates
of tens of thousands of barrels per day if zonal thickness and
other factors are constant, and without mechanical constraints,
macroports can reach thousands of barrels per day, and mesoports
100
BASE
Devon
Burlington
Encana
XTO
Chief
% Cumulative gas
80
60
40
20
0
0
10
20
30
40
50
60
70
80
90
100
% Wells
Fig. 16Fractional production-variability plot for the Barnett fractured shale in Texas for various operating companies (Aguilera
2010b).
200

Stage:
Page: 201
Total Pages: 19
1.00
0.90
Fractional cumulative gas
0.80
0.70
Area A
Area B
Area C
Area D
Area E
Area F
0.60
to
ue n
d
s
tio
se duc
a
re re
inc ty
e ensi
r
u d
at
rv re
Cu actu
fr
0.50
0.40
0.30
0.20
0.10
0.00
0.00
0.10
0.20
0.30
0.40
0.50
0.60
0.70
0.80
0.90
1.00
Fractional number of wells

Fig. 17Fractional production-variability plot for six Nikanassin tight-gas areas in the Deep basin of Canada (Gonzalez and Aguilera 2011).
80
> 300 Mscfd
70
200300 Mscfd
100200 Mscfd
1 / q (1E6 / scf)
60
< 100 Mscfd
50
Linear (< 100 Mscfd)
40
30
20
10
0
0.0
0.5
1.0
1.5
2.0 2.5 3.0 3.5

Time0.5 (Years0.5)
4.0
4.5
5.0
Fig. 18Drawdown linear flow for wells producing from Devonian shales connected through natural fractures in an infiniteacting reservoir. The flow rates extend over a 25-year period
without reaching the boundary-dominated flow (Aguilera 1980,
p. 403).
hundreds of barrels per day. Microports can produce few to tens

of barrels per day on pump. However, they stated that microport
flow units are decidedly nonreservoir in this comparative completion of moderate thickness and medium gravity oil without mechanical constraints. These flow units are of far more interest as
potential seals for higher quality reservoir downdip.
Since Martin et al. (1997), microport and nanoport reservoirs
have become economical in many instances. Subsequently,
Deng et al. (2011) associated possible gas rates in conventional
reservoirs with pore-throat apertures. Oil and gas rates from
these efforts are presented in the upper-right-hand side of Fig.
2a. Martin et al. (1997) and Deng et al. (2011) estimates were
for vertical wells. Fig. 2a adds some estimates of oil and gas
rates for MSHF horizontal wells on the basis of the authors
observations. These rates are only approximations that are not
meant to replace any kind of testing. However, they can provide
an idea with respect to the potential production of wells in a

given reservoir.
This paper examines production rates and flow periods for
some unconventional petroleum reservoirs, and it makes a comparison with the corresponding pore-throat apertures. As early as
the 1970s, Aguilera (1980, p. 403) demonstrated the presence of
linear flow in vertical wells producing from Devonian shales in an
infinite-acting reservoir (Fig. 18). The start of the production data
went back to the 1950s, and the linear-flow period extended over
25 years without reaching boundary-dominated flow. The connectivity was interpreted to occur through natural fractures. This led
to the development of approximate solutions for the linear flow in
dual-porosity reservoirs (Aguilera 1987).
The linear-flow observation has also been made more recently
with data from wells drilled during the last few years, particularly
in wells that have been drilled horizontally and stimulated in multiple stages. Many rigorous solutions have been developed to handle the linear-flow observations in shale- and tight-gas reservoirs
(Wattenbarger et al. 1998; El Banbi 1998; El Banbi and Wattenbarger 1998; Arevalo-Villagran et al. 2006; Moghadam et al.
2010; Brohi et al. 2011), as well as simple, approximate solutions
(Aguilera 1987; Leguizamon and Aguilera 2011) that match very
well with the rigorous solutions. There also have been many useful and practical empirical models based on the actual recordings
of production rates in unconventional oil and gas reservoirs
(Reynolds and Munn 2010; Baihly et al. 2010, 2011; Martin et al.
2011; Hamm and Struyk 2011). As in the case of permeabilities
and porosities used for estimating pore-throat apertures (rp35),
the petroleum rates published in some of the aforementioned studies are used in this paper to make the proposed interpretations
tractable.
Gas Rates. Production data for the Barnett, Fayetteville, Woodford, and Haynesville shale published by Baihly et al. (2010, 2011)
are presented in the upper figure sections of Figs. 19, 20, 21, and
22, respectively. The authors considered 1,957 horizontal wells in
their study, and they shifted the production data to the day of first
production (DOFP). They grouped the wells according to the year
in which the wells started production. The middle sections in Figs.
19, 20, 21, and 22 (not published by Baihly et al.) present the data
as crossplots of rate vs. time on log-log coordinates (these sections
were not presented in the Baihly et al. study). For the Barnett,

201
BARNETT
q (Mcfd)
q (Mcfd)
1500
1000
500
500
0
0
12
24
36
48
Time (months)
60
72
84
12
24
36
48
Time (months)
60
72
84
FAYETTEVILLE
BARNETT
10000
10000
DOFP_YEAR_2003 25 WELLS
Linear Flow
LINEAR FLOW
q (Mcfd)
q (Mcfd)
2000
1000
Total Pages: 19
FAYETTEVILLE
2500
1500
Page: 202
3000
2500
2000
Stage:
1000
1000
100
100
1
1
10
Time (months)
10
Time (months)
100
100
FAYETTEVILLE
BARNETT
0.010
0.0040
Linear Flow
1/q (1/mcfd)
0.0030
0.0025
0.008
1/q (1/Mcfd)
0.0035
LINEAR
0.0020
0.006
0.004
0.0015
0.002
0.0010
0.0005
0.000
0
0.0000
0
4
6
Time0.5 (months0.5)
10
Fig. 19Production history of the Barnet shale normalized to

the day of first production (DOFP) is shown on the upper section of figure. The presence of linear flow is shown in the loglog middle section of figure (slope 5 0.5) and the lower section
of figure (1/q vs. square root of time). It is likely that some
groups of wells start showing boundary effects at t 0.5 equal to
6 to 8 months0.5 or 36 to 64 months (Baihly et al. 2010, 2011).
Fayetteville, and Woodford shale, there are linear trends with a

slope of 0.5, indicating an average linear flow. The possible beginning of boundary-dominated flow is also observed for some Barnett
wells, particularly in the group that started production in the year
2003. For the Haynesville shale, the production data fall below the
straight line with a slope of 0.5, indicating a possible lack of linear
flow. The actual production drop generates a straight line with a
202
3
4
5
Time0.5 (months0.5)
Fig. 20Production history of the Fayetteville shale normalized

to the DOFP is shown on the upper section of figure. The presence of linear flow is shown in the log-log middle section of figure (slope 5 0.5) and the lower section of figure (1/q vs. square
root of time). No boundary effects are noticeable (Baihly et al.
2010, 2011).
slope of 1.0 (green dashed line) that is interpreted to be the result

of fracture closure, something that happens quite often in overpressured/naturally fractured reservoirs.
The lower sections in Figs. 19, 20, 21, and 22 [not published
by Baihly et al. (2010, 2011)] present the data as crossplots of
1.0 divided by rate (1/q) vs. the square root of time on Cartesian
coordinates. There are indications of linear flow in the Barnett,
Fayetteville, and Woodford shales; in all these cases, the linear
trend extrapolates to the origin at zero time, a clear indication
Stage:
Page: 203
Total Pages: 19
WOODFORD
HAYNESVILLE
4000
10000
3500
3000
8000
7000
q (Mcfd)
2500
q (Mcfd)
9000
2000
1500
6000
5000
4000
3000
1000
2000
1000
500
0
0
0
0
12
24
36
48
Time (months)
60
72
12
24
36
48
Time (months)
84
60
72
84
HAYNESVILLE
WOODFORD
100000
10000
10000
q (Mcfd)
q (Mcfd)
LINEAR FLOW
1000
1000
LINEAR FLOW NOT PRESENT
POSSIBLE FRACTURE CLOSURE
100
100
1
10
Time (months)
100
10
100
Time (months)
HAYNESVILLE
WOODFORD
0.0010
0.0040
0.0035
0.0008
0.0030
LINEAR FLOW?
LINEAR FLOW
0.0025
1/q (1/Mcfd)
1/q (1/Mcfd)
0.0020
0.0015
0.0010
0.0006
0.0004
0.0002
0.0005
0.0000
0.0000
0
3
4
5
Time0.5 (months0.5)
Fig. 21Production history of the Woodford shale normalized

to the DOFP is shown on the upper section of figure. The presence of linear flow is shown in the log-log middle section of figure (slope 5 0.5) and the lower section of figure (1/q vs. square
root of time). It is likely that the upper two groups of wells start
showing boundary effects at t 0.5 equal to 5 to 6 months0.5 or 25
to 36 months (Baihly et al. 2010, 2011).
that there is no damage around the wellbore. For the Barnett shale
(Fig. 19), some wells have clearly reached boundary-dominated
flow. The Fayetteville shale does not reach boundary-dominated
flow, and the Woodford shale might have reached it for the wells
that started production in 2006. The Haynesville response is different. As indicated previously, the log-log crossplot of rate vs.
time yielded a production trend that fell below the 0.5 straight
line. If there were linear flow, the 1/q vs. square root of time
Time0.5 (month0.5)
Fig. 22Production history of the Haynesville shale normalized

to the DOFP is shown on the upper section of figure. Absence
of linear flow is shown in the log-log middle section of figure,
but the slope of 1.0 suggests closure of natural fractures. If
there is linear flow (not supported by the log-log plot), the lower
section of figure (1/q vs. square root of time) would indicate
improved conditions around the wellbore that would tend to
disappear as the reservoir is depleted (Baihly et al. 2010, 2011).
would extrapolate to a negative value of 1/q. This would be indicative of improved conditions around the wellbore that probably
stem from the overpressured status of the reservoir. The negative
effect, however, is that permeability and porosity decrease as the
net stress on the natural fractures becomes larger.
Thus, the type of flow observed in Barnett, Fayetteville, Woodford, and Haynesville horizontal gas wells is not too dissimilar

203
Stage:
Producing Day Rate (BOPD)
140
East Pembina 2009

East Pembina 2010
East Pembina 2011
West Pembina 2009
West Pembina 2010
West Pembina 2011
Type Well
150
100
50
0
0
12
18
Months
24
30
H2 2008
H1 2009
100
H2 2009
HI 2010
80
H2 2010
TYPE WELL
60
40
20
12
18
24
Months of Production
30
36
Production from the Shaunavon Tight Oil Reservoir, SW Saskatchewan

and Unrestricted Transition Between Linear Flow Periods (this study)
1000
East Pembina 2009

East Pembina 2010
East Pembina 2011
West Pembina 2009
West Pembina 2010
West Pembina 2011
Type Well
Transition slope 0.75
HI 2008
120
36
Production from the Cardium Tight Oil Reservoir

and Unrestricted Transition between Linear Flow Periods (this study)
100
100
Total Pages: 19

(Hamm and Struyk, 2011)
Production from the Cardium Tight Oil Reservoir

250
200
Page: 204
10
HI 2008
H2 2008
H1 2009
H2 2009
HI 2010
H2 2010
Transition slope 0.75
TYPE WELL
100
10
1
1
10
Months of Production
1
1
10
Months
100
100
1.E01
Unrestricted Transition between Linear Flow Periods (this study)
HI 2008
1.E01
East Pembina 2009
East Pembina 2010
East Pembina 2011
West Pembina 2009
West Pembina 2010
West Pembina 2011
Type Well
Unrestricted Transition
8.E02
7.E02
6.E02
5.E02
H2 2008
H1 2009
8.E02
H2 2009
7.E02
1/q (1/bopd)
9.E02
1/q (1/bopd)
9.E02
4.E02
HI 2010
H2 2010
6.E02
TYPE WELL
5.E02
Unrestricted Transition
4.E02
3.E02
3.E02
Flowback
2.E02
2.E02
1.E02
1.E02
0.E+00
0.E+00
0
0
10
15
20
10
t0.75
15
20
t0.75
Fig. 23The upper section of figure shows oil-production rates

vs. time for the tight-oil Cardium formation, as published by
Hamm and Struyk (2011). Their figure also includes a curve for
a Type Well developed from actual Cardium production. The figure in the middle shows a log-log crossplot of oil rate vs. time
with a slope of approximately 0.75. The bottom section of the
figure shows a Cartesian crossplot of 1/q vs. t 0.75, with straight
lines representing the unrestricted transition bracketing the
tight-oil Cardium data.
from the linear flow observed in hydraulically fractured vertical

Devonian shale wells, starting in the 1950s (Fig. 18). As such, the
data can be generally interpreted with methods available in the literature for handling linear (and bilinear) flow (Aguilera 1987; Wattenbarger et al. 1998; El Banbi 1998; El Banbi and Wattenbarger
1998; Arevalo-Villagran et al. 2006; Moghadam et al. 2010; Brohi
et al. 2011; Leguizamon and Aguilera 2011). The flow behavior of
tight-oil wells, however, can be different, as discussed next.
204
Fig. 24The upper section of the figure shows oil-production

rates vs. time for the tight southwest Saskatchewan Shaunavon
formation, as published by Hamm and Struyk (2011). Their figure also includes a curve for a Type Well developed from actual
Shaunavon production. The figure in the middle shows a loglog crossplot of oil rate vs. time with a slope of approximately
0.75. The bottom figure shows a Cartesian crossplot of 1/q vs.
t 0.75, with straight lines representing the unrestricted transition
bracketing the tight-oil Shaunavon data.
Oil Rates. Hamm and Struyk (2011) have presented very complete data sets for MSHF horizontal wells in the WCSB. Two
cases are presented in this paper in Figs. 23 and 24. On the basis
of the actual production data, they have developed empirical type
curves for various reservoirs with varying types of rock quality.
To generate their average type curves, they shift the production
data of all wells in a given formation to the same starting point.
The upper section in Fig. 23 shows oil-production rates vs. time
for the tight-oil Cardium formation, as published by Hamm and
Decline Rate with Restricted InterLinear Transition Flow

Triple Porosity Model Dominated by Linear Flow
1.E+04
Linear Flow (slope = 0.5)
1.E+03
1.E+02
RestrictedTransition (slope > 1.00)
1.E+01
qD
1.E+00
Linear
1.E01
1.E02
Transition
1.E03
1.E04
Linear
1.E05
1.E06
10
E+
1. 09
E+
1. 08
E+
1. 07
E+
1. 06
E+
1. 05
E+
1. 04
E+
1. 03
E+
1. 02
E+
1. 01
E+
1. 00
E+
1. 1
0
E
1. 2
0
E
1. 3
0
E
1. 4
0
E
1.
tD
Decline Rate with Unrestricted InterLinear Transition Flow
1.E+04
1.E+03
1.E+02
UnrestrictedTransition (slope = 1.00)
1.E+01
qD
1.E+00
Linear
1.E01
1.E02
Transition
1.E03
1.E04
Linear
1.E05
Stage:
Page: 205
Total Pages: 19
The sections in the middle of Figs. 23 and 24 (not published by

Hamm and Struyk) show log-log crossplots of oil rate vs. time, each
of which results in an approximate straight line with a slope of
0.75. The oil rates for the Cardium after 2 months of production
(approximate end of flowback) range between approximately 100
and 200 BOPD. The pore-throat radii (rp35) for the tight-oil Cardium range between approximately 0.2 and 0.69 mm (Fig. 11). The
oil rates for the Shaunavon, after 2 months of production, range
between approximately 95 and 125 BOPD. Note that flowbacks in
the Shaunavon take longer than the flowbacks in the Cardium. The
pore-throat radii for the Shaunavon range between approximately
0.09 and 0.55 mm (Fig. 14). Thus, the pore-throat and oil-production
ranges are smaller in the Shaunavon as compared with those in the
Cardium. Another difference is that the Shaunavon is charged with
20 to 22 API oil compared with 36 to 38 API oil in the Cardium.
Because the reservoir pressures are not the same, the performance,
recovery, and hydraulic-fracturing flowback can be quite different.
To the authors best knowledge, there are no references in the
literature that explain the 0.75 slope. Consequently, an approximation is presented next to try to explain this behavior. The bottom figures in Figs. 23 and 24 show a Cartesian crossplot of 1/q
vs. t0.75, with straight lines bracketing the tight-oil Cardium and
Shaunavon data. The explanation for the 0.75 slope has been
developed with the use of the following equation for a triple-porosity model dominated by linear flow in an infinite-acting reservoir adapted from Leguizamon and Aguilera (2011):
1 p
qD 2
s
ptD

:
x0d 1 x0d f tD ; sDd x0t 1 x0t f tD ; sDt
3
1.E06
10
E+
1. 09
E+
1. 08
E+
1. 07
E+
1. 06
E+
1. 05
E+
1. 04
E+
1. 03
E+
1. 02
E+
1. 01
E+
1. 00
E+
1. 01
E
1.
03
02
1.
04
1.
1.
tD
Decline Rate with Unrestricted InterLinear Transition Flow
1.E+03
1.E+02
1.E+01
UnrestrictedTransition (slope ~ 0.75)
qD
1.E+00
1.E01
Linear
1.E02
Transition
1.E03
1.E04
Linear
1.E05
1.E06
10
E+
1. 09
E+
1. 08
E+
1. 07
E+
1. 06
E+
1. 05
E+
1. 04
E+
1. 03
E+
1. 02
E+
1. 01
E+
1. 00
E+
1. 1
0
E
1. 2
0
E
1. 3
0
E
04
E
1.
1.
tD
Fig. 25The upper section of figure shows dimensionless rate

vs. dimensionless time for a triple-porosity model with restricted
(pseudosteady-state) interlinear flow. The transition between the
two linear-flow periods generates a trend line with a negative
slope much bigger than 1.00. The middle section of figure shows
unrestricted (transient) interlinear flow. The transition between
the two linear flow periods generates a straight line with a negative slope equal to 1.00. The bottom section of figure also shows
unrestricted interlinear flow. In this case, however, the transition
shows a negative slope of approximately 0.75, similar to what is
observed in the case of the real Cardium data shown on Fig. 23
and the Shaunavon data on Fig. 24.
Struyk (2011). Their figure section also includes a curve for a

Type Well developed from actual Cardium production. The upper
section in Fig. 24 shows the same information for the southwest
Saskatchewan Shaunavon formation.
For the case of restricted (pseudosteady-state) interlinear flow,

the function is given by

ccd
tD
;
f tD ; sDd 1 exp
sDd

cct
tD
;
f tD ; sDt 1 exp
4
sDt
where cc indicates empirical commingled completion exponents for the dual- and triple-porosity reservoirs (Leguizamon and
Aguilera 2011), and sDd and sDt are the approximate beginnings
of the straight lines for the dual- and triple-porosity responses,
respectively. For the case of unrestricted (transient) interlinear
flow, the functions are given by an extension of Najurieta (1980)
and Leguizamon and Aguilera (2011):
r
rccd
tD
sDd
tanh
;
f tD ; sDd
sDd
tD
r
r cct
tD
sDt
f tD ; sDt
tanh
:
5
sDt
tD
The x0d and x0t in Eq. 3 are a combination of storativity ratio
and hydraulic diffusivity ratios (for convenience, simply called
storativity/diffussivity), given by
x0d xd gDd ; x0t xd gDt ; . . . . . . . . . . . . . . . . . . 6
where the storativity ratios are given by
xd
Sf
Sf
; xt
: . . . . . . . . . . . . . . . . . 7
Sf Smd
Sf Smt
The fracture storage (Sf) is the product (/cth) for the fractures,
Smd is the product (/cth) for one matrix (or medium), and Smt is
the product (/cth) for the other matrix (or medium) in the tripleporosity reservoir. The dimensionless hydraulic diffusivity for the
dual- and triple-porosity media are given by gDd gf /gmd and
gDt gf /gmt. In the case of gas reservoirs, the storages are calculated at initial pressure.

205
The upper section in Fig. 25 shows examples of results with

the approximate model described previously for the case of restricted (pseudosteady-state) interlinear flow on the basis of the
following data: xd 0.1, xt 0.1, sDd 10,000, sDt 0.05,
ccd cct 1, gDd 10,000, gDt 10,000. The log-log crossplots
result in three parallel straight lines with slopes equal to 0.5,
indicating linear flow. The pseudosteady-state transition periods
result in very large slopes. This is what is generally assumed in
the literature for the analysis of MSHF horizontal wells.
The middle section in Fig. 25 uses the same input data but
assumes unrestricted (transient) interlinear flow. In this case,
the transition is recognized by a straight line with a slope equal
to 1.00. The lower section in Fig. 25 also assumes unrestricted
(transient) interlinear flow but reduces the hydraulic diffusivity
ratios to gDd gDt 550 (this reduces automatically the storativity/diffusivity to x0d x0t 55). In this case, the transition
results in an approximate linear trend with a slope of approximately 0.75. This is the flow period observed in the tight-oil
Cardium formation as shown by the real production-decline
data presented in the middle section of Fig. 23 and the Shaunavon in southwest Saskatchewan, as shown in the middle section
of Fig. 24. Thus, reducing the separation between the linearflow periods also decreases the negative value on the transition
slope. On the basis of this interpretation, it is likely that the
actual production rates shown in the middle of Figs. 23 and 24
will revert to linear-flow behavior in the future.
Conclusions
Process (or delivery) speed (i.e., the ratio of permeability to porosity) provides a continuum between conventional, tight-gas,
shale-gas, tight-oil, and shale-oil reservoirs.
There are distinctive flow units for each type of reservoir penetrated by vertical and horizontal MSHF wells that can be linked
empirically to possible gas and oil rates and under favorable
conditions to the type of production decline.
A new unrestricted transition flow period in tight-oil reservoirs
has been recognized by considering a triple-porosity model that
leads to a straight line with a negative slope equal to 1.00 on
log-log coordinates. This straight line occurs as a transition
between two linear-flow periods.
Reducing the values of storativity/diffusivity (x0d and x0t ) associated with the dual- and triple-porosity media decreases the
separation between the linear-flow periods and reduces the
value of the transition slope (for example, to 0.75).
To make the work tractable, the bulk of the data presented in
this paper has been extracted from published geologic and petroleum-engineering literature.
Nomenclature
c compressibility
cc empirical commingled completion exponent, dimensionless
h thickness
hf fracture width
k permeability
P pressure
pD dimensionless pressure
qD dimensionless rate
St storage of the second matrix in the triple-porosity system
(/cth)
Sm storage of the first matrix in the triple-porosity system
(/cth)
tD dimensionless time
gf hydraulic diffusivity of fractures
gmd hydraulic diffusivity of one matrix
gmt hydraulic diffusivity of the other matrix
m viscosity
x storativity ratio
xd storativity ratio in the dual-porosity reservoir
xt storativity ratio in triple porosity
/ porosity, fraction
206
Stage:
Page: 206
Total Pages: 19
sDd approximate beginning of the last straight line in the

dual-porosity reservoir, dimensionless
sDt approximate beginning of the last straight line in the triple-porosity reservoir, dimensionless
Acknowledgments
Parts of this work were funded by the Natural Sciences and Engineering Research Council of Canada (NSERC agreement 34782506), ConocoPhillips (agreement 4204638), Alberta Innovates
Energy and Environment Solutions (AERI agreement 1711), the
Schulich School of Engineering at the University of Calgary, and
Servipetrol Limited. The laboratory work with cuttings was carried out with Darcylog equipment provided to the GFREE
research team by Dr. Roland Lenormand of Cydarex in Paris,
France. Their contributions are gratefully acknowledged. I extend
my gratitude to my students in the Schulich School of Engineering, particularly Bukola Olusola, Peng Wu, and John Freddy
Ramirez, and to Maria Ester Aguilera for their help during preparation of this paper.
References
Aguilera, R. 1978. Log Analysis of Gas-Bearing Fracture Shales in the
Saint Lawrence Lowlands of Quebec. Paper SPE 7445 presented at the
53rd Annual Fall Technical Conference and Exhibition, Houston,
Texas, 13 October. http://dx.doi.org/10.2118/7445-MS.
Aguilera, R. 1980. Naturally Fractured Reservoirs, Tulsa, Oklahoma:
PennWell Books.
Aguilera, R. 1987. An Approximate Solution of Linear Flow in Naturally
Fractured Reservoirs. Paper SPE-16442-MS available from SPE,
Richardson, Texas.
Aguilera, R. 2002. Incorporating Capillary Pressure, Pore Aperture Radii,
Height Above Free Water Table, and Winland r35 Values on Pickett
Plots. AAPG Bull. 86 (4): 605624.
Aguilera, R. 2004. Integration of Geology, Petrophysics, and Reservoir
Engineering for Characterization of Carbonate Reservoirs Through
Pickett Plots. AAPG Bull. 88 (4): 433446.
Aguilera, R. 2006. Sandstone vs. Carbonate Petroleum Reservoirs: A
Global Perspective on Porosity-Depth and Porosity-Permeability Relationships: Discussion. AAPG Bull. 90 (5): 807810.
Aguilera, R. 2010a. Flow Units: From Conventional to Tight Gas to Shale
Gas Reservoirs. SPE Paper 132845 presented at the Trinidad and
Tobago Energy Resources Conference, Port of Spain, Trinidad, 2730
June. http://dx.doi.org/10.2118/132845-MS.
Aguilera, R. 2010b. A Method for Estimating Hydrocarbon Cumulative
Production Distribution of Individual Wells in Naturally Fractured
Carbonates, Sandstones, Shale Gas, Coalbed Methane and Tight Gas
Formations. J. Cdn. Pet. Tech. 49 (8): 5358. http://dx.doi.org/
10.2118/139846-MS.
Aguilera, R.F., Eggert, R.G., Lagos, G. et al. 2009. Depletion and the
Future Availability of Petroleum Resources. Energy J. 30 (1):
141174.
Aguilera, R.F., Harding, T., Krause, F. et al. 2008. Natural Gas Production
From Tight Gas Formations: A Global Perspective presented at the
19th World Petroleum Congress, Madrid, Spain, 29 June3 July.
Aguilera, R.F., Ripple, R.D., and Aguilera, R. 2012. Link Between Rocks,
Hydraulic Fracturing, Economics, Environment, and the Global Gas
Portfolio. Paper SPE 162717 presented at the SPE Canadian Unconventional Resources Conference, Calgary, Alberta, Canada, 30
October1 November. http://dx.doi.org/10.2118/162717-MS.
Almanza, A. 2011. Integrated 3D Geological Model of the Mississippian
Devonian Bakken Formation, Elm Coulee, Williston Basin: Richland
County, Montana, MSc thesis, Colorado School of Mines, 140 pages.
Ambrose, R.J., Hartman, R.C., Diaz-Campos, M. et al. 2010. New PoreScale Considerations for Shale Gas in Place Calculations. Paper SPE
131772 presented at the SPE Unconventional Gas Conference, Pittsburgh, Pennsylvania, 2325 February. http://dx.doi.org/10.2118/131772MS.
Andrade, J., Civan, F., Davegowda, D. et al. 2011. Design and Examination
of Requirements for a Rigorous Shale Gas Reservoir Simulator Compared to Current Shale Gas Simulators. Paper SPE 144401 prepared for
the SPE North American Unconventional Gas Conference, The Woodlands, Texas, 1416 June. http://dx.doi.org/10.2118/144401-MS.
Archie, G.E. 1950. Introduction to Petrophysics of Reservoir Rocks.
AAPG Bull. 34: 943961.
Arevalo-Villagran, J.A., Wattenbarger, R.A., and Samaniego-Verduzco, F.
2006. Some History Cases of Long-Term Linear Flow in Tight Gas
Wells. J. Cdn. Pet. Tech. 45 (3): 3137. http://dx.doi.org/10.2118/0603-01-MS.
Baihly, J., Altman, R., Malpani, R. et al. 2010. Shale Gas Production
Decline Trend Comparison Over Time and Basins. Paper SPE 135555
presented at the SPE Technical Conference and Exhibition, Florence,
Italy, 1922 September. http://dx.doi.org/10.2118/135555-MS.
Baihly, J., Altman, R., Malpani, R. et al. 2011. Study Assesses Shale
Decline Rates. In The American Oil and Gas Reporter.
Brohi, I., Pooladi-Darvish, M., and Aguilera, R. 2011. Modeling Fractured Horizontal Wells As Dual Porosity Composite Reservoirs
Application to Tight Gas, Shale Gas and Tight Oil Cases. Paper SPE
144057 presented at the SPE Western North American Regional
Meeting, Anchorage, Alaska, 711 May. http://dx.doi.org/10.2118/
144057-MS.
Bustin, R.M., Bustin, A.M.M., Cui, X. et al. 2008. Impact of Shale Properties on Pore Structure and Storage Characteristics. Paper SPE 119892
presented at the SPE Shale Gas production Conference, Fort Worth,
Texas, 1618 November. http://dx.doi.org/10.2118/119892-MS.
Chopra, A.K., Stein, M.H., and Ader, J.C. 1987. Development of Reservoir Descriptions To Aid in Design of EOR Projects. Paper SPE 16370
presented at the SPE California Regional Meeting, Ventura, California, 810 April. pp. 437449. http://dx.doi.org/10.2118/16370-MS.
Clarkson, C.R., Jensen, J.L., Pedersen, K. et al. 2012. Innovative Methods
for Flow-Unit and Pore-Structure Analyses in a Tight Siltstone and
Shale Gas Reservoir. AAPG Bull. 96 (2): 355374.
Deng, H., Leguizamon, R., and Aguilera, R. 2011. Petrophysics of Triple
Porosity Tight Gas Reservoirs With a Link to Gas Productivity. Paper
SPE 144590 presented at the SPE Western North American Regional
Meeting, Anchorage, Alaska, 711 May; SPE Res Eval & Eng 14 (5):
566577. http://dx.doi.org/10.2118/144590-MS.
DOE, US Department of Energy. 2012. New Models Help Optimize Development of Bakken Shale Resources, http://www.fossil.energy.gov/
news/techlines/2012/12003-New_Models_Aid_Bakken_Shale_Develo.
html(last entered March 3, 2013).
Ebanks, W.J. Jr. 1987. Flow Unit ConceptIntegrated Approach to Reservoir Description for Engineering Projects, Abstract. AAPG Bull. 71 (5):
551552.
Economides, M. and Oligney, R. 2000. The Color of Oil: The History, the
Money and the Politics of the Worlds Biggest Business, Spicewood,
Texas: Round Oak Publishing Company.
Ehrenberg, S.N. and Nadeau, P.H. 2005. Sandstone vs. Carbonate Petroleum Reservoirs: A Global Perspective on Porosity-Depth and Porosity-Permeability Relationships. AAPG Bull. 89 (4): 435445.
El Banbi, A.H. 1998. Analysis of Tight Gas Well Performance, PhD dissertation, Texas A&M University (May 1998).
El Banbi, A.H. and Wattenbarger, R.A. 1998. Analysis of Linear Flow in
Gas Well Production. Paper SPE 39972 presented at the SPE Gas
Technology Symposium, Calgary, Alberta, Canada, 1518 March.
http://dx.doi.org/10.2118/39972-MS.
Freeman, J. and Eller, D.L. 2010. A Golden Opportunity in the Golden
State: A Monterey Shale Primer, US Research, Raymond James and
Associates, December 20.
Gonzalez, L. and Aguilera, R. 2011. Effect of Natural Fracture Density on
Production Variability of Individual Wells in the Tight Gas Nikanassin
Formation. Paper CSUG/SPE 149222-PP presented at the Canadian
Unconventional Resources Conference, Calgary, Alberta, Canada,
1517 November; in press for 2013: J. Cdn. Pet. Tech. 52 (2):
131143. http://dx.doi.org/10.2118/149222-MS.
Gunter, G.W., Finneran, J.M., Hartmann, D.J. et al. 1997a. Early Determination of Reservoir Flow Units Using an Integrated Petrophysical
Method. Paper SPE 38679 presented at the SPE Annual Technical
Conference and Exhibition, San Antonio, Texas, 58 October. http://
dx.doi.org/10.2118/38679-MS.
Gunter, G.W., Pinch, J.J., Finneran, J.M. et al. 1997b. Overview of an
Integrated Process Model To Develop Petrophysical Based Reservoir
Descriptions. Paper SPE 38748 presented at the SPE Annual Technical
Stage:
Page: 207
Total Pages: 19
Conference and Exhibition, San Antonio, Texas, 58 October. http://

dx.doi.org/10.2118/38748-MS.
Hamm, B. and Struyk, E. 2011. Quantifying the Results of Horizontal Multistage Development in Tight Oil Reservoirs of the Western Canada
Sedimentary Basin: Technical and Economic Case Studies From a Reservoir Evaluators Perspective. Paper CSUG/SPE 149000 presented at
the Canadian Unconventional Resources Conference, Calgary, Canada,
1517 November. http://dx.doi.org/10.2118/149000-MS.
Hartmann, D.J. and Beaumont, E.A. 1999. Predicting Reservoir System
Quality and Performance. In Exploring for Oil and Gas Traps, AAPG
Treatise of Petroleum Geology, Handbook of Petroleum Geology, ed.
Edward A. Beaumont and Norman H. Foster, pp. 9-19-154.
Javadpour, F., Fisher, D., and Unsworth, M. 2007. Nanoscale Gas Flow in
Shale Gas Sediments. J. Cdn. Pet. Tech. 46: (10). http://dx.doi.org/
10.2118/07-10-06-PA.
King, H.M. 2014. Utica Shale - The Natural Gas Giant Below the Marcellus, http://geology.com/articles/utica-shale/ (accessed 28 January
2014).
Kolodzie, S.J. 1980. The Analysis of Pore Throat Size and Use of Waxman Smit to Determine OOIP in Spindle Field, Colorado. Paper presented at the SPE Annual Technical Conference and Exhibition,
Dallas, Texas, 2124 September. http://dx.doi.org/10.2118/ 9382-MS.
Kwon, B.S. and Pickett, G.R. 1975. A New Pore Structure Model and
Pore Structure Interrelationships. Paper presented at the SPWLA 16th
Annual Logging Symposium.
Lavoie, J.Y., Marcil, J.S., Dorrins, P.K. et al. 2011. Natural Gas Potential
in the Saint Lawrence Lowlands of Quebec: A Case Study. J. Cdn.
Pet. Tech. 50 (11): 7182. http://dx.doi.org/10.2118/137593-PA.
Law, B.E. 2002. Basin-Centered Gas Systems. AAPG Bull. 86:
18911919.
Leguizamon, J. and Aguilera, R. 2011. Optimizing Hydraulic Fracturing
of Naturally Fractured Tight Gas Formations. Paper SPE 142727 presented at the SPE Middle East Unconventional Gas Conference and
Exhibition, Muscat, Oman, 31 January2 February. http://dx.doi.org/
10.2118/142727-MS.
Lopez, B. and Aguilera, R. 2013. Evaluation of Quintuple Porosity in
Shale Petroleum Reservoirs. Paper SPE 165681 prepared for presentation at the SPE Eastern Regional Meeting, Pittsburgh, Pennsylvania,
2022 August. http://dx.doi.org/10.2118/165681-MS.
MacKenzie, W.T. 1975. Petrophysical Study of the Cardium Sand in the
Pembina Field. Paper SPE 5541 presented at the 50th Annual Technical Meeting of SPE, Dallas, Texas, 28 September1 October. http://
dx.doi.org/10.2118/5541-MS.
Magoon, L.B. and Beaumont, E.A. 1999. Petroleum Systems. Chapter 3 in
Exploring for Oil and Gas Traps, AAPG Treatise of Petroleum Geology, Handbook of Petroleum Geology, ed. Edward A. Beaumont and
Norman H. Foster, pp. 3.53.8.
Magoon, L.B. and Schmoker, J.W. 2000. The Total Petroleum System
The Natural Fluid Network That Controls the Assessment Unit. In
Chapter PS in US Geological Survey Digital Data Series 60, US Geological Survey World Petroleum Assessment.
Martin, A.J., Solomon, S.T., and Hartmann, D.J. 1997. Characterization of
Petrophysical Flow Units in Carbonate Reservoirs. AAPG Bull. 81:
(5): 734759.
Martin, J.P. 2005. The Utica and Hamilton ShalesThe Next Fractured
Shale Play: 4th Annual Gas Shales Summit: Production & Potential,
December 12, 2005, Denver, Colorado, New York: Strategic Research
Institute, 35 pp.
Martin, R., Baihly, J., Malpani, R. et al. 2011. Understanding Production
From Eagleford-Austin Chalk System. Paper SPE 145117 presented at
the SPE Annual Technical Conference and Exhibition, Denver, Colorado, 30 October2 November. http://dx.doi.org/10.2118/145117-MS.
Moghadam, S., Mattar, L., and Pooladi-Darvish, M. 2010. Dual Porosity
Type Curves for Shale Gas Reservoirs. Paper SPE/CSUG 137535 presented at the Canadian Unconventional Resources and International
Petroleum Conference, Calgary, Alberta, Canada, 1921 October.
http://dx.doi.org/10.2118/137535-MS.
Najurieta, H.L. 1980. A Theory for Transient Pressure Analysis in Naturally Fractured Reservoirs. J. Pet Tech 32 (7): 12411250. http://
dx.doi.org/10.2118/6017-PA.
Nelson, P.H. 2009. Pore-Throat Sizes in Sandstones, Tight Sandstones,
and Shales. AAPG Bull. 93 (3): 329340.

207
Olson, J.E, Laubach, S.E., and Lander, R.H. 2009. Natural Fracture Characterization in Tight Gas Sandstones: Integrating Mechanics and Diagenesis. AAPG Bull. 93 (11): 15351549.
Ortega C.E. and Aguilera, R. 2013. A Complete Petrophysical Evaluation
Method for Tight Formations From Only Drill Cuttings in the Absence
of Well Logs. Paper SPE 161875-PP presented at the SPE Canadian
Unconventional Resources Conference, Calgary, Alberta, Canada, 30
October1 November. Online: SPE J., May 24, 2013. http://
dx.doi.org/10.2118/161875-MS.
Pickett, G.R. and Artus, D.S. 1970. Prediction From Logs of Recoverable
Hydrocarbon Volume, Ordovician Carbonates, Williston Basin. Geophysics 35 (1): 113123.
Pittman, E.D. 1992. Relationship of Porosity and Permeability to Various
Parameters Derived From Mercury Injection-Capillary Pressure
Curves for Sandstone. AAPG Bull. 76 (2): 191198.
Province of Alberta. 2013. Oil and Gas Conservation Act, Oil and Gas
Conservation Rules. Alberta Regulation 151/1971, with amendments
up to and including Alberta Regulation 115/2013; www.qp.alberta.ca/
documents/Regs/1971_151.pdf.
Rahmanian, M., Aguilera, R., and Kantzas, A. 2013. A New Unified Diffusion-Viscous Flow Model Based on Pore Level Studies of Tight Gas
Formations. Paper CSUG/SPE 149223 presented at the Canadian
Unconventional Resources Conference, Calgary, Alberta, Canada,
1517 November. SPE J. 18 (1): 3849, February 2013. http://
dx.doi.org/10.2118/149223-PA.
Rahmanian, M., Solano, N., and Aguilera, R. 2010. Storage and Output
Flow From Shale and Tight Gas Reservoirs. Paper SPE 133611 presented at the SPE Western Regional Meeting, Anaheim, California,
2729 May. http://dx.doi.org/10.2118/133611-MS.
Regan, L.J. 1953. Fractured Shale Reservoirs of California. AAPG Bull.
37: 201216.
Reynolds, M.M. and Munn, D.L. 2010. Development Update for an
Emerging Shale Gas Giant FieldHorn River Basin, British Columbia, Canada. Paper SPE 130103 presented at the SPE Unconventional
Gas Conference. Pittsburgh, Pennsylvania, 2325 February. http://
dx.doi.org/10.2118/130103-MS.
Ruppel, S.C. and Loucks, R.G. 2008. Black Mudrocks: Lessons and Questions From the Mississippian Barnett Shale in the Southern Midcontinent. The Sedimentary Record, June.
Shaw, D.B. and Weaver, C.E. 1965. The Mineralogical Composition of
Shales. J. Sedimentary Petrology 35: 213222.
Smocker, J.W. 2005. U.S. Geological Survey Assessment Concepts for
Continuous Petroleum Accumulations 1995. In Chapter 13 of Petroleum Systems and Geologic Assessment of Oil and Gas in the Southwestern Wyoming Province, Wyoming, Colorado and Utah, Version 1.
Sneider, R.M., King, H.R., Hawkes, H.E. et al. 1983. Methods for Detection and Characterization of Reservoir Rock, Deep Basin Gas Area,
Western Canada. J. Pet Tech 35 (9): 17251734. http://dx.doi.org/
10.2118/10072-PA.
208
Stage:
Page: 208
Total Pages: 19
Solano, N.A. 2010. Reservoir Characterization of the Upper Jurassic

Lower Cretaceous Nikanassin Group, MSc thesis, Geoscience Department, GFREE Tight Gas Research Team, University of Calgary (September 2010).
Sonnenberg, S.A. 2011. The Bakken Petroleum System of the Williston Basin, An Example of a Tight-Oil Play. Paper presented at the
First Conference on Unconventional Oil and Gas Development,
China University of Petroleum (Eastern China), Qindao, China, 45
July.
Walls, J.D., Diaz, E., Derzhi, N. et al. 2011. Eagle Ford Shale Reservoir
Properties From Digital Rock Physics. Paper presented at the HGS
Applied Geoscience Mudrocks Conference (AGC), The Woodlands,
Texas, 78 February.
Wang, F.P. and Reed, R.M. 2009. Pore Networks and Fluid Flow in Gas
Shales. Paper SPE 124253 presented at the SPE Annual Technical
Conference and Exhibition, New Orleans, Louisiana, 47 October.
http://dx.doi.org/10.2118/124253-MS.
Wattenbarger, R.A., El Banbi, A.H., Villegas, E.M. et al. 1998. Production
Analysis of Linear Flow Into Fractured Tight Gas Wells. Paper SPE
39931 presented at the 1998 SPE Rocky Mountain Regional/Low Permeability Reservoirs Symposium, Denver, Colorado, 58 April. http://
dx.doi.org/10.2118/39931-MS.
Ziarani, A.S. and Aguilera, R. 2012. Knudsens Permeability Correction
for Tight Porous Media. Transport in Porous Media (Springer) 91 (1):
239260.
Roberto Aguilera is a professor and the ConocoPhillips-NSERCAIEE Chair in the Schulich School of Engineering, Chemical and
Petroleum Engineering Department, at the University of Calgary, a guest professor at the China University of Petroleum
(Eastern China); and a principal of Servipetrol Limited in Canada. He heads the GFREE tight-gas research program at the
University of Calgary. Aguilera is a petroleum-engineering graduate from the Universidad de America at Bogota, Colombia,
and holds MEng and PhD degrees in petroleum engineering
from the Colorado School of Mines. He was an American Association of Petroleum Geologists (AAPG) instructor on the subject of naturally fractured reservoirs from 1984 through 1996.
Aguilera has lectured, presented his course titled Naturally
Fractured Reservoirs, and/or has rendered consulting services
in more than 50 countries throughout the world. He was a Distinguished Author of the SPE Journal of Canadian Petroleum Technology (1993 and 1999), was a recipient of the Outstanding
Service Award (1994) and the Distinguished Service Medal
(2006) from the Petroleum Society of Canadian Institute of Mining Metallurgy and Petroleum (CIM), was an SPE Distinguished
Lecturer on the subject of naturally fractured reservoirs for the
20002001 season, was the recipient of the 2011 SPE Canada
Regional Distinguished Achievement Award for Petroleum Engineering Faculty, was a past Executive Editor of the SPE Journal
of Canadian Petroleum Technology, and is an SPE member.


Aguilera 2014

Enviado por

Dados do documento

Direitos autorais

Formatos disponíveis

Compartilhar este documento

Compartilhar ou incorporar documento

Opções de compartilhamento

Você considera este documento útil?

Este conteúdo é inapropriado?

Direitos autorais:

Formatos disponíveis

Aguilera 2014

Enviado por

Direitos autorais:

Formatos disponíveis

REE165360 DOI: 10.

2118/165360-PA Date: 30-April-14

Flow Units: From Conventional to

ID: jaganm Time: 16:07 I Path: S:/3B2/REE#/Vol00000/140007/APPFile/SA-REE#140007

REE165360 DOI: 10.2118/165360-PA Date: 30-April-14

WORLD GAS RESOURCE PYRAMID

Although process speed has been used for several decades in

Bottom of Resource Pyramid Unknown

Aguilera et al., WPC, 2008, SPE 132845, SPE 162717

WORLD OIL RESOURCE PYRAMID

Fig. 1Estimates of global natural-gas and oil endowment

production wells are associated with natural fractures. Examples

Process (or Delivery) Speed

where m is fluid viscosity and ct is total compressibility. Eq. 1 is at

May 2014 SPE Reservoir Evaluation & Engineering

REE165360 DOI: 10.2118/165360-PA Date: 30-April-14

Diffusion Dominated Flow

CHART FOR ESTIMATING PORE THROAT APERTURE

the presence of fractures), viscosity, and pressure. For example,

ID: jaganm Time: 16:07 I Path: S:/3B2/REE#/Vol00000/140007/APPFile/SA-REE#140007

REE165360 DOI: 10.2118/165360-PA Date: 30-April-14

Permeability, kmax (mD)

rt : 0.1 1.5 micro meters

Fig. 2(Continued) (b) There is consistency between porosities

Conventional Petroleum Reservoirs

Fig. 3Porosity/permeability crossplot showing pore types of

significant possibilities for the future of unconventional petroleum

May 2014 SPE Reservoir Evaluation & Engineering

REE165360 DOI: 10.2118/165360-PA Date: 30-April-14

Fig. 4Porosity/permeability crossplot showing different rp35

Fig. 5Porosity/permeability crossplot showing different rp35

in many instances. Tight-oil production and a new unrestricted

ID: jaganm Time: 16:08 I Path: S:/3B2/REE#/Vol00000/140007/APPFile/SA-REE#140007

REE165360 DOI: 10.2118/165360-PA Date: 30-April-14

Fig. 6Permeability vs. porosity crossplot including data from

porosity models (Aguilera 2010a; Andrade et al. 2011) and more

tight-gas reservoirs should be represented at least by dual-porosity

Tight-Oil and Shale-Oil Reservoirs

REE165360 DOI: 10.2118/165360-PA Date: 30-April-14

compressibility (ct) play a very important role in fluid flow

Bakken Petroleum System

Mission Canyon Fm.

May 2014 SPE Reservoir Evaluation & Engineering

REE165360 DOI: 10.2118/165360-PA Date: 30-April-14

REE165360 DOI: 10.2118/165360-PA Date: 30-April-14

May 2014 SPE Reservoir Evaluation & Engineering

REE165360 DOI: 10.2118/165360-PA Date: 30-April-14

from Gas Research Institute (GRI) crushed sample analysis tend to

May 2014 SPE Reservoir Evaluation & Engineering

REE165360 DOI: 10.2118/165360-PA Date: 30-April-14

Generalized Cross Section

Utica Shale Gas Play

Utica Source Rock

Depth Below Sea Level (Thousand Feet)

(C) SERVIPETROL LTD., Calgary,

the cumulative-gas-production distribution of tight-gas and

May 2014 SPE Reservoir Evaluation & Engineering

REE165360 DOI: 10.2118/165360-PA Date: 30-April-14

Fractional cumulative gas

Fractional number of wells

< 100 Mscfd

Linear (< 100 Mscfd)

2.0 2.5 3.0 3.5

hundreds of barrels per day. Microports can produce few to tens