See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/280092068

Impact of oil-water contacts, reservoir (dis)continuity, and reservoir

characteristics on spatial distribution of water, gas, and high-water – low-
bitumen saturated zones and varia...

Chapter · January 2013

DOI: 10.1306/13371580St643553

CITATIONS READS

11 4,113

6 authors, including:

Milovan Fustic Barry Bennett

Nazarbayev University Applied Petroleum Technology (UK) Ltd.
110 PUBLICATIONS 1,534 CITATIONS 191 PUBLICATIONS 3,810 CITATIONS

SEE PROFILE SEE PROFILE

Haiping Huang Thomas B P Oldenburg

The University of Calgary The University of Calgary
147 PUBLICATIONS 4,657 CITATIONS 132 PUBLICATIONS 2,844 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Milovan Fustic on 07 October 2015.

The user has requested enhancement of the downloaded file.

 6
Fustic, Milovan, Barry Bennett, Stephen M. Hubbard, Haiping Huang,
Thomas Oldenburg, and Steve Larter, 2013, Impact of reservoir heterogeneity
and geohistory on the variability of bitumen properties and on the distribution
of gas- and water-saturated zones in the Athabasca oil sands, Canada, in
F. J. Hein, D. Leckie, S. Larter, and J. R. Suter, eds., Heavy-oil and oil-sand petroleum
systems in Alberta and beyond: AAPG Studies in Geology 64, p. 163 – 205.

Impact of Reservoir Heterogeneity and

Geohistory on the Variability of Bitumen
Properties and on the Distribution of
Gas- and Water-saturated Zones in the
Athabasca Oil Sands, Canada
Milovan Fustic1
Statoil Canada Ltd., 3600, 308 4th Ave. SW, Calgary, Alberta, T2P 0H7, Canada
(e-mail: mfus@statoil.com)

Barry Bennett, Stephen M. Hubbard, Haiping Huang, Thomas Oldenburg, and Steve Larter
Petroleum Reservoir Group, University of Calgary, 2500 University Dr. NW, Calgary, Alberta, T2N 1N4,
Canada (e-mails: bennettb@gushor.com; shubbard@ucalgary.ca; huah@ucalgary.ca; toldenburg@ucalgary.ca;
slarter@ucalgary.ca)

ABSTRACT
The Athabasca oil sand deposit, the world’s largest petroleum accumulation, contains an es-
timated 1.7 trillion bbl of heavily to severely biodegraded oil, with API gravities ranging from
6 to 108. Although reservoir characterization has been the subject of many studies in the region,
very little attention has been given to petroleum (bitumen) characterization and particularly to
its reservoir-scale relationship with the host sediments.
In this study, variation in the bitumen physical and chemical properties were measured
on a suite of samples. These were obtained from numerous cores from various reservoir types
and geographic areas of the Athabasca oil sand deposits. The variation in bitumen viscosities
and changes in the hydrocarbon composition caused by varying levels of biodegradation were
interpreted using molecular markers. These data were integrated into the reservoir facies frame-
work and interpreted in the context of various reservoir configurations.
The bitumen composition and physical properties are highly variable on vertical and lateral
scales. In general, the viscosity of the bitumen residing at the base of an oil column may be an
order of magnitude greater than the bitumen located shallower in the reservoir. Although this
is a general rule, results also locally show anomalous inverse gradients and steps in compo-
sitional trends. These coincide with paleo- and/or present-day geologic features that include

1
Previous address: Petroleum Reservoir Group, University of Calgary, 2500 University Dr. NW, Calgary, Alberta, T2N 1N4, Canada.

Copyright n2013 by The American Association of Petroleum Geologists.

DOI:10.1306/13371580St643553

163
164 Fustic et al.

the presence and/or absence of oil-water contacts, vertical and lateral barriers that create res-
ervoir compartments, and variation in lithologies, porosity, and permeability. Qualitative and
quantitative analyses of bitumen molecular composition within the sedimentologic frame-
work demonstrate that each reservoir compartment behaves as an independent bioreactor;
bitumen is more degraded within porous and permeable strata than within the comparable
interbedded sand shale sequences, and bitumen biodegradation is intensified in highly water-
saturated zones within the petroleum column. In addition to biodegradation along the basal
(paleo-) oil-water contact, these complex variations are explained by the complex geohistory of
the petroleum charge and in-reservoir fluid (petroleum, water, and gas) mixing, including as-
sociated variations in oil biodegradation behavior.
In addition, our interpretation of the interplay through time and space of depositional setting
and biodegradation suggests that top water and other highly water-saturated zones are related
to the formation and the subsequent depletion of gas caps likely derived from microbial gas
generation that followed petroleum entrapment.
Implications to reservoir developments are immense. We suggest that integrated baseline
studies of sedimentologic and geochemical variation interpretation allow for mapping and
prediction away from well control of the extent of top gas, top water and low-bitumen, high-
water saturation zones, as well as bitumen properties variation throughout the reservoir.
Geochemical data from integrated baseline studies are a powerful tool for identifying barriers
or baffles to vertical fluid communication, whereas they may also be applied toward produc-
tion optimization and allocating production along horizontal wells, including the assessment
of steam chamber growth in oil sand operations.

INTRODUCTION et al., 1994, 1995; Larter and Aplin, 1995; McKinney and
Bland, 2003; Weissenburger and Borbas, 2004; Milkov
The Athabasca oil sand deposit, located in northeast- et al., 2007; Larter et al., 2008), and encouraging results
ern Alberta, Canada (Figure 1), contains an estimated are documented in oil sand reservoirs (Chan et al., 1992).
1.7 trillion bbl (Alberta Energy and Utilities Board, 2007) Because of the limited number of studies, the potential
of heavily degraded, mostly immobile petroleum, com- of geochemical analysis for defining the variation in
monly referred to as bitumen (Strausz and Lown, 2003). fluid properties in the Athabasca oil sands has yet to be
Current exploitation technologies involve surface min- realized.
ing (Kaminsky, 1974) and in-situ extraction methods This study includes and compares bitumen viscosity
such as steam-assisted gravity drainage (SAGD) (Butler, and molecular composition data from seven wells from
1994). The recovered bitumen is upgraded by thermal or two geographic areas located approximately 100 km
hydrocracking treatments to produce synthetic crude (62 mi) apart (Figure 1). The first study area, referred
oil. Both extraction and upgrading processes require a as the southern Athabasca deposit, is at a depth of about
large amount of energy and water to recover and up- 250 m (820 ft) below the present-day surface. Because
grade the bitumen into synthetic oil (Flint et al., 2005). it is too deep for mining exploitation, it is designated
The production exceeds 1.2 million bbl of oil/day, which for potential thermal in-situ development. The second
exceeds 40% of the total liquid petroleum production study area, referred to as the northern Athabasca de-
in Canada (Alberta Energy and Utilities Board, 2007). posit, is covered by only a few meters of glacial Pleis-
Bitumen molecular characterization studies in the tocene and/or Holocene deposits and extends to a depth
Athabasca oil sand region have been focused on pe- of about 100 m (328 ft). Because of the shallow depth of
troleum systems applications, such as oil-oil and oil- this deposit, this area is designated for potential mine
source rock correlations (Fowler and Brooks, 1987; development.
Brooks et al., 1988; Fowler and Riediger, 2000; Riediger The primary objective of this study is to indicate
et al., 2000, Higley et al., 2009; Adams et al., 2013). Uti- controls on present-day variations in bitumen viscos-
lization of reservoir geochemistry approaches and ap- ity and molecular composition and the origins of the
plications is still in the infant stages of development complex spatial distributions of gas, low-bitumen,
in the Athabasca oil sand region, although they have high-water saturation zones, and top waters. Biodeg-
been proven successful in many reservoir developments radation tends to be associated with close proximity to
worldwide (Kaufman et al., 1990, 1997; Smalley and oil-water contact areas; thus, the focus of this study is to
England, 1992, 1994; Hwang and Baskin, 1994; Nederlof investigate the oil compositions near oil-water contact
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 165

Figure 1. Location of study area,

including selected sites presented
in this chapter (indicated by stars).
Inset map shows the location of
the Athabasca oil sand deposit
in Alberta, Canada. Key: Township
(T) and ranges (R). The length
of townships and ranges is
6 mi (9.7 km). 10 km (6.2 mi).

zones and the effect on the scale and timing of degra- mentology, petrophysics, and geochemistry to develop
dation. In addition to the basal oil-water interface, this an improved understanding of the relative spatial
study also investigates the origin and impact of top distributions of rock and fluid types (i.e., bitumen, wa-
waters and low-bitumen, high-water saturation zones, ter, gas), including the subdivision of bitumen accord-
defined as bitumen-saturated zones containing mobile ing to its own characteristics (i.e., viscosity ranges and/
water, Sw > Sir (water saturation [Sw], irreducible water or the level of biodegradation). These studies may aid
saturation [Sir]), on the biodegradation levels of bitumen prediction and mapping bitumen properties through-
located within and close to the presence of these zones. out the reservoir. This information may also present a
An important part of our interpretation deals with in- powerful tool for identifying reservoir compartments,
reservoir fluid-mixing processes, including water inva- distinguishing between barrier and baffles to fluid
sion and fresh oil charging at various stages of the communication to aid production optimization (Fustic
reservoir history. The ultimate goal of this study is to et al., 2011), and for production allocation (i.e., steam
construct a baseline reservoir characterization program chamber growth assessment) (Bennett et al., 2010). To
that integrates numerous disciplines, including sedi- achieve the objectives described, this chapter examines
166 Fustic et al.

several case studies that show fluid property variations 1992). The possible impact of Barremian to early Aptian
within a range of reservoir settings, such as the pres- eustatic sea level drop on the formation of the incised
ence and absence of oil-water contact areas, vertical and valleys is a matter of an ongoing research (F. van Buchem,
lateral compartmentalization, bottom and top water, 2010, personal communication), whereas late Aptian –
low-bitumen, high-water saturation zones, and bitu- Albian eustatic sea level rise (Haq et al., 1988; Sahagian
mens encountered in a range of depositional facies. et al., 1996) resulted in the deposition of the marine de-
posits of the Clearwater Formation, which overlie bi-
tumen hosting McMurray Formation (Figure 2). Little
information exists on the function of tectonics on sedi-
GEOLOGIC SETTING AND BACKGROUND mentation during deposition of the McMurray For-
mation (late Aptian) except for studies of structural
Basin Setting features associated with the dissolution of the un-
derlying Prairie Evaporites Formation (McPhee and
The Athabasca oil sand deposit lies on the eastern Wightman, 1991; Wightman et al., 1991; Wightman
flanks of the Western Canada sedimentary basin (WCSB), et al., 1995). Westerly progressing salt dissolution
which is interpreted as a foreland basin (Leckie and formed paleotopographic lows of northwesterly trend-
Smith, 1992; Stockmal et al., 1992; Wright et al., 1994). ing paleovalley; these are developed on Devonian
Formation of the accommodation space in the Atha- carbonates (Wightman et al., 1989) on which bitu-
basca Basin and relative sea level rise are caused by men hosting McMurray sediments are deposited. The
basement flexure resulting from the accretion of alloch- McMurray Formation is Aptian in age (Leckie and
thonous terranes during the Colombian orogeny (Price, Smith, 1992). Radiometric dating suggests that the
1973; Stott, 1984; Leckie and Smith, 1992; Stockmal et al., age for the McMurray Formation sediments is about

Figure 2. Stratigraphic column for northeastern Alberta (modified from O’Donnell and Jodrey, 1984; Wightman et al., 1995).
LM1, LM2, MM1, and MM2 indicate the lower and upper subunits of the lower and middle McMurray Formation, respectively.
Depositional elements indicate dominant lithologies and architecture of each unit (i.e., lower McMurray 1 [LM1] is dominantly
sand; lower McMurray 2 [LM2] is laterally extensive flood-plain mud deposits; middle McMurray 1 [MM1] is subtidal sand bars;
middle McMurray 2 [MM2] consists of meandering river deposits comprising cross-bedded sand, inclined heterolithic strata
[IHS], and mud plug deposits; and the upper McMurray contains coarsening-upward shoreface deposits). Strong correlation
between large-scale depositional elements and their reservoir potential allows for using them to define reservoir flow units.
Generally, large-scale depositional elements dominated by (1) sand lithologies are classified as reservoir type 1 flow units;
(2) interbedded sand and mud as reservoir type 2; and (3) mud-dominated as nonreservoir flow units or reservoir flow barriers
(Fustic et al., 2013). LPB = lower point bar deposits; UPB = upper point bar deposits.
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 167

112 m.y. (Selby and Creaser, 2005) (Figure 2). Reser- entrapment of oil gas. In this work, consideration of the
voir sand intervals are unconsolidated. The lack of ce- spatial variation exhibited by the petroleum maturity
mentation is caused by a shallow burial history, there parameters, such as ratios of triaromatic and monoaro-
being less than 1 km (0.6 mi) of overburden removed matic steroids, shows that the trapping mechanism is
(Moshier and Waples, 1985; Nurkowski, 1985). This shal- caused by a combination of Clearwater Formation act-
low burial depth was insufficient to promote quartz ing as a regional vertical seal and a series of large-scale
dissolution and precipitation (Walderhaug, 1996). In (up to 40 m [131 ft] thick) mud plug deposits acting as
addition, oil-charged reservoirs reduce the water con- lateral barriers to the same extent.
nectivity needed for distributing silica or other cements
to the precipitation site, thus eliminating or reducing Biodegradation Processes and Products
the cementation rates (Marchand et al., 2000, 2001,
2002; Haszeldine et al., 2003). The lack of the cementa- Consensus exists that the high viscosity and low API
tion also suggests early petroleum (oil) charge, consis- gravity of the bitumen (6–108 API) in the Athabasca oil
tent with observation by Selby and Creaser (2005), but sand deposit is caused by the effects of very severe pe-
at the Athabasca, the low temperature means that this troleum alteration by biodegradation (Deroo et al.,
conclusion is not definitive. 1974, 1977; Tissot and Welte, 1984; Moshier and Waples,
1985; Brooks et al., 1988). The API gravities of the initial
charged oil were proposed to be between 25 and 308
PETROLEUM SYSTEM SUMMARY API (Deroo et al., 1977; Fowler et al., 2004), although
Adams et al. (2013) suggest that the charged oil was
The origin (source rocks) and timing of petroleum em- sulfur rich and likely about 258 API.
placement in the Athabasca oil sand region are a sub- Until recently, it was believed that severe petroleum
ject of considerable debate. Among many hypotheses, biodegradation was due to the incursion of oxygen-
the most widely accepted involves multiple source rocks bearing meteoric water associated with the shallow
(Creaney and Allan, 1990, 1992; Allan and Creaney, 1991; depth of the reservoir that supported aerobic mechan-
Creaney et al., 1994), with a predominant contribution isms for biodegradation (e.g., Tissot and Welte, 1984).
from the Exshaw Formation shale (Brooks et al., 1988; However, the presence of reduced naphthoic acids
Fowler and Riediger, 2000; Riediger et al., 2001). Recent that are formed only when hydrocarbon degradation is
work by Higley et al. (2009) suggests major contri- conducted under anaerobic conditions indicates that
bution from the Gordondale Member of the Nordegg biodegradation of the Athabasca oils occurred under
Formation (Jurassic Fernie Group), whereas Adams anaerobic conditions (Aitken et al., 2004). Recent work
et al. (2013) suggest that the petroleum in the Atha- on anaerobic biodegradation by Head et al. (2003) and
basca oil sand deposit is derived from at least two Larter et al. (2005) describes several aspects of hydro-
major source rocks, including the Exshaw and Gordon- carbon degradation via anaerobic pathways. These au-
dale, with Peace River having a larger Gordondale con- thors also demonstrated that anaerobic biodegrada-
tribution. Estimates for timing of petroleum charge tion occurs proximal to oil-water contact areas and that
range from 58 m.y. for all contributing source rocks downward diffusion of components in the oil column to
(Higley et al., 2009) and for the Mississippian Exshaw the oil-water contact results in vertical compositional
Formation to as early as 112 Ma (Selby and Creaser, gradients. Evidence for microbiological life occurs as
2005) and 110 to 80 Ma (Riediger et al., 2001). membrane lipids of microorganisms in basal waters of
the McMurray Formation (Oldenburg et al., 2009) and
Trapping Mechanism microbial communities including the methanomicro-
biales and epsilonproteobacteria that are indigenous to
The Clearwater Formation (Figure 2) provides a regional biodegraded petroleum reservoirs (Hubert et al., 2012).
seal for entrapped petroleum (Ranger, 1994; Ranger and Adams et al. (2006) explains that the geologic con-
Gingras, 2006), but the trapping mechanisms are de- ditions in the Athabasca oil sand deposit was the most
batable. The trapping mechanism theories include a favorable of all the deposits in the western Canadian
range from the stratigraphic pinch-outs (Vigrass, 1968), sedimentary basin for destructive anaerobic biodegra-
a subtle regional anticline (Vigrass, 1968; Ranger, 1994; dation. Increasing reservoir temperatures westward
Tozer et al., 2012), and updip seals formed by biode- were less favorable for microbial communities, result-
graded viscous petroleum (Mossop, 1980). Fustic et al. ing in progressively less biodegraded oils occuring
(2012b) suggests sequential filling of reservoir compart- toward the west of the basin. The westernmost reser-
ments with progressively more mature oil, a process voirs that have been buried to temperatures in excess
that resembles Gussow’s (1954) concept of differential of 808C are the least biodegraded; this is considered a
168 Fustic et al.

temperature threshold for pasteurization of the reser- most widely accepted model is that the lower, middle,
voir microbiota (Head et al., 2003; Adams et al., 2006; and upper units are composed of fluvial (braided river
Larter et al., 2006). Thus, Adams et al. (2006) concluded [lower McMurray 1 {LM1}] and associated flood-plain
that reservoir temperature is a key control on the gen- deposits [lower McMurray 2 {LM2}]); fluvioestuarine
eral level of the biodegradation on the basinwide scale. (subtidal open-estuarine bars [middle McMurray 1
Vertical viscosity gradients in the petroleum col- {MM1}]) and tidally influenced meandering river chan-
umns of the McMurray Formation were documented nel deposits (middle McMurray 2 [MM2]); and shallow-
by Seyer and Gyte (1989) but did not attract much at- marine shoreface facies (upper McMurray [UM]) de-
tention from the oil sand reservoir developers until posits, respectively (Figure 2). The marine Clearwater
recent reservoir simulation studies (Gates et al., 2008) Formation overlies the McMurray Formation. This over-
demonstrated that the variability and ranges in viscos- all upward transition from the continental to fully ma-
ities in the Athabasca oil sand deposit are important rine deposits is interpreted as an overall transgressive
components for reservoir development planning and setting (Mellon and Wall, 1956; Carrigy, 1959, 1971;
optimization. Flach, 1984; Ranger, 1994).
Petroleum compositional and viscosity gradients in The lower McMurray Formation occupies the lows
a single vertical column in heavy-oil reservoirs world- on the Devonian unconformity (Figure 2). Both braided
wide have been interpreted as the effects of anaerobic river sands (LM1) and associated flood-plain deposits
biodegradation (Head et al., 2003; Huang et al., 2004; (LM2) of that unit are characterized by extensive lateral
Larter et al., 2006), including those in the Athabasca oil continuity that is laterally bounded by Devonian sub-
sand deposit (e.g., Larter et al., 2006, 2008). The biodeg- crops (Figure 2). Subsurface correlation and mapping
radation level is commonly assessed using the Peters of these deposits are commonly straightforward using
and Moldowan (1993) scale (PM) that ranks biodegra- features recognized in gamma-ray logs.
dation level on a scale of 1 to 10, with level PM1 rep- However, the straightforward correlation is not the
resenting slightly biodegraded oil and PM10 repre- case with the middle McMurray Formation unit open-
senting the most severely biodegraded oil. The bitumen estuarine subtidal bars and tidally influenced mean-
from the Athabasca oil sand deposit ranges between dering river deposits. The preserved morphology and
PM5 and PM9 (Bennett et al., 2006). the extent of the subtidal bars (MM1) highly depend
on the amount of the erosion by later incised mean-
Reservoir Charging and In-reservoir Fluid Mixing dering rivers (Wightman and Pemberton, 1997). In
many areas, this unit (MM1) is completely eroded. The
Fluids in petroleum reservoirs are constantly mixing, architecture of tidally influenced meandering river de-
with a tendency to attain equilibrium (England et al., posits comprises several basic large-scale depositional
1987; England, 1994; Stainforth, 2004). Biodegradation elements, such as abandoned channel-fill (mud plug)
or physical leaking results in some fluid components deposits, sandy lower point-bar deposits, and inter-
being removed from the reservoir, whereas other com- bedded sand and mud of the upper point bar (Wightman
ponents such as methane and CO2 are generated in the and Pemberton, 1997). Interbedded sand and mud of
reservoir as by-products of biodegradation (Zengler the upper point bar are referred to as inclined hetero-
et al., 1999; Larter et al., 2005). These processes are lithic strata (IHS) (Thomas et al., 1987; Smith, 1988;
poorly documented in studies dealing with the Atha- Smith et al., 2009). Recently documented counter point-
basca oil sand deposits, but it is clear that mixing times bar deposits (Smith et al., 2009) and translation point-
with these viscous fluids may exceed reservoir age bar deposits (Hubbard et al., 2011; Fustic et al., 2012a)
(Adams et al., 2013). are morphologically distinct from point-bar deposits
and are also characterized by lower trough cross-bedded
Stratigraphy, Sedimentology, and the sand and upper IHS. The meandering river depositional
Reservoir Architecture elements create a complex mosaic of reservoir and
nonreservoir facies (Fustic et al., 2012a; Hubbard et al.,
Stratigraphically, the Lower Cretaceous McMurray For- 2011).
mation is informally divided into the lower, middle,
and upper McMurray units (Flach and Mossop, 1985) Reservoir Characterization
(Figure 2). These units are further subdivided into the
lower and the upper subunits (members) (e.g., O’Donnell Reservoir characterization studies commonly include
and Jodrey, 1984; Paul and Fustic, 2002) (Figure 2). lithofacies description, petrophysical analysis, and lab-
Although the McMurray Formation depositional envi- oratory measurements of the bitumen saturation. Re-
ronments are still a matter of considerable debate, the gardless of the interpreted depositional environment,
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 169

the clean sands in all units commonly have poros- into conceptual 3-D objects (geobodies). Unlike a series
ities exceeding 30% and permeabilities reaching 7 d of cross sections and plan view maps, conceptual 3-D
(Strobl et al., 1997; Fustic et al., 2008, 2013). The strong geobodies allow for easy visualization and analysis of
relationships between porosity and permeability and various data sets (Fustic et al., 2013), including geo-
sedimentary structures have been documented by chemical analysis. Geobody types are (1) reservoir flow
Strobl et al. (1997). Relationships between porosity and units type 1 (clean bitumen-saturated sand of the lower
permeability and large-scale depositional elements point-bar deposits); (2) reservoir flow unit type 2 (sandy
are documented by Fustic et al. (2013). Strobl et al. IHS of the upper point and breccia of the channel-lag
(1997) and Fustic et al. (2008, 2013) suggest that (1) deposits); (3) nonreservoir or reservoir flow barrier units
sands of the lower point-bar deposits are character- (mud plug and muddy flood-plain and/or overbank
ized by excellent and relatively consistent reservoir deposits); and (4) reservoir flow impairments in top gas,
properties ranging between 30 and 37% porosity and bottom water, top water, and low-bitumen, high-water
from 5 to 7 d permeability; (2) the IHS deposits show saturation zones.
significant changes from low to 35% porosity and from Fluid characterization studies consisted of labora-
low to 7 d permeability over the intervals of a few cen- tory measurements of mass fractions of bitumen, water,
timeters; and (3) abandoned channel deposits are pre- and solids in each reservoir sample. These were ob-
dominantly mudstone characterized with very low tained by routine Dean Stark analysis (Dean and Stark,
effective porosities and permeabilities. The strong cor- 1920). Sampling was continuous through the bitumen-
relation between depositional elements and reservoir saturated sand intervals, and results were posted as
properties allowed to incorporate depositional elements histograms at 10-cm (4-in.) intervals along the wellbore.
as reservoir flow units for mapping oil sand reservoir Samples that were selected for viscosity measure-
quality (Strobl et al., 1997; Fustic et al., 2008, 2013). The ments were obtained from recently drilled quickly fro-
strong relationship between depositional elements and zen cores that minimize postdrilling storage effects on
reservoir flow units allows the use of these terms inter- the bitumen composition and properties principally
changeably between disciplines (Figure 2) (Fustic et al., caused by evaporation of volatile compounds (Adams
2013). et al., 2008). Representative samples were selected from
various reservoir facies to interpret the relationship be-
tween the reservoir facies and petroleum in the reser-
METHODOLOGY voir. In each well, three to six samples were selected
through the petroleum column for viscosity measure-
Data used in this study include reservoir core descrip- ments. The dead oil viscosity was measured on bitu-
tions, petrophysical logs that include dipmeter logs, men recovered from core samples by mechanical ex-
laboratory measurements of bitumen weight and vis- traction using a compaction-based device referred to
cosity, and bitumen molecular characterization by as a plunger (Gushor Inc.). The dead oil viscosity was
gas chromatography-mass spectrometry (GC-MS). measured using Brookfield viscometers at several tem-
The same type of data was used in both sites (Figure 1), peratures, generally between 20 and 808C.
except that three-dimensional (3-D) seismic information Core samples selected for the molecular compo-
was available in the southern Athabasca study area. sition analysis of bitumen were obtained from four
For each case study, (1) a geologic (depositional ele- recently drilled and three previously drilled wells. Ap-
ments) framework was created; (2) the distribution of proximately 10 to 15 samples were taken per well at a
water, gas, and bitumen was mapped within each geo- 2- to 3-m (6.6- to 10-ft) spacing. Bitumen from oil sand
logic framework; (3) high-resolution sampling and mea- samples selected for geochemical molecular composi-
surement of bitumen saturation, dead oil viscosity, and tion analysis was extracted using dichloromethane sol-
oil molecular composition on core from seven wells were vent. Saturated and aromatic hydrocarbon compounds
made; and finally, (4) bitumen characterization results were separated from bitumen extracts using the pro-
were analyzed within the context of each geologic frame- cedure described by Bennett and Larter (2000). The
work to determine the origin of bitumen heterogeneity saturated and aromatic hydrocarbon fractions were
and the occurrences of water, gas, and low-bitumen, analyzed using GC-MS on a Hewlett-Packard (HP)
high-water saturation zones. 5890 GC (using splitless injection) interfaced to an HP
To build the geologic framework, the first step is to 5970B quadruple mass-selective detector. The effects
define depositional environments and large-scale de- of biodegradation were assessed on both saturated
positional elements, as well as presence of water, gas, and aromatic hydrocarbon compounds.
and low-bitumen, high-water saturation zones in each Geochemical and oil viscosity results were posted
well location and then to correlate them among wells on interpreted geologic cross sections and analyzed in
170 Fustic et al.

the context of present-day reservoir conditions. Final- and the upper point-bar deposits decreases in the di-
ly, a schematic summary diagram was created for both rection of the channel accretion (Figure 4). Channel C
case studies to illustrate the relative influence of reser- deposit (well 3 above 285 m [935 ft]) is sandy, and de-
voir charging, in-reservoir biodegradation, and fluid- positional dips show the easterly oriented lateral accre-
mixing histories in space and time on oil composition tion. The contact of channel C with channel B has been
and distribution. interpreted as an erosional avulsion surface.
The interpreted stacking patterns of the large-scale
depositional elements (Figure 4D) show that stacked
RESULTS lower point-bar deposits of A and B (well 1 in Figure 4A,
D) and lower point bars of B and C (well 3 in Figure 4C,
Southern Athabasca Study Region D) form thick type 1 reservoir flow units. Other res-
ervoir flow units include the type 2 reservoir flow unit
The oil sands in the study area in the southern Atha- (IHS), nonreservoir flow units and a low-bitumen, high-
basca region, within township 85 and ranges 6 and 7 water saturation zones (Table 1). The low-bitumen, high-
(Figure 1), are interpreted to consist of the series of water saturation zones are encountered between reser-
lateral reservoir compartments separated by mud plug voir flow unit types 1 and 2 in wells 1 and 2, whereas it
deposits, which are 400- to 500-m (1310- to 1640-ft)- has not been detected as a distinct unit in well 3. Water
wide curvilinear features (Figure 3). The mud plug saturation and porosity crossplots for each reservoir flow
deposits (Figures 3B, C; 4, 5) create lateral compart- unit (Figure 7) show clear clustering of data points. The
mentalization of the reservoir sands. low-bitumen, high-water saturation zone has the same
range of porosities as type 1 reservoir flow unit ranging
from 30 to 37%, whereas sandy and muddy IHS reser-
Southern Athabasca: Compartment B voir flow units show a wide range of porosities from
10 to 37% (Figure 7; Table 1). Although characterized
Geology and Present-day Fluid Distribution by the same range of porosities, the type 1 and the low-
bitumen, high-water saturation zones have different
The study area was characterized using data obtained water saturations (Figure 7; Table 1). Water satura-
from three closely spaced wells. The distance between tion in sandy and muddy IHS is influenced by clay-
wells 1 and 3 is about 650 m (2133 ft) (Figure 4), which bound water. Core photographs (Figure 8) and geo-
is a common length for SAGD well pairs in oil sand physical logs (Figure 4) do not show any evidence of
reservoir developments. sedimentologic break between the type 1 reservoir
The McMurray Formation is approximately 60 m flow unit and the low-bitumen, high-water satura-
(197 ft) thick. Based on seismic, dipmeter, and core tion zone.
information, we interpreted it to consist of parts of three Data from three wells show that three stacked chan-
stacked and/or interfingered meandering channel nels (A, B, and C) form the continuous petroleum col-
deposits. The oldest to the youngest channel deposits umn composed of reservoir flow unit types 1 and 2,
are identified as A, B, and C (Figures 3, 4). Channel A with the base of petroleum column in well 1 being
occupies the bottom 13 to 17 m (43–56 ft) of each well. significantly lower than in wells 2 and 3 (Figure 4).
Well 1 penetrated through the sandy lower point-bar Nonreservoir (impermeable) flow units are laterally to
part of channel A, but wells 2 and 3 penetrated the the left and at the base of wells 2 and 3 (Figure 4).
nonreservoir muddy IHS part of channel A. Lateral
changes in the lithology of channel A from being sandy Fluid Property Variations within the Geologic
in well 1 to muddy in wells 2 and 3 are typical features Framework: Results
ascribed to the laterally accreting deposition in the me-
andering river environment. In the absence of reliable Oil viscosity measurements were obtained from 3-m
dipmeter data in the case of channel A, lateral accre- (10-ft)-thick composite samples (Figure 4; Table 2). Based
tion processes can be interpreted from left toward the on their relative position in the well, samples are
right. The thickness of channel B exceeds 40 m (131 ft) named bottom (B), middle (M), and top (T) (Figure 4;
(Figure 4). Clean sand characterized by trough cross- Table 2). Results are generally downward increasing
beds at the lower part and the IHS at the upper part of dead oil viscosity in wells 2 and 3. This is not the case
the deposit are interpreted as lower and upper point- for well 1, in which the most viscous oil is found in
bar deposits, respectively. Seismic and dipmeter data the middle sample, adjacent to the overlying low-
reveal that this channel was accreting toward the west– bitumen, high-water saturation zone (266–272 m [873–
northwest (Figures 3, 4). The proportion of the lower 892 ft]) (Figure 9; Table 2).
Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 171

Figure 3. The southern Athabasca

study area with interpreted lat-
eral compartments. (A) Three-
dimensional seismic depth slice
flattened on the McMurray (slice
is 10 m [33 ft] below the top of
the McMurray); (B) interpreted
extent of the mud plug deposits
in plan view suggests the lateral
compartmentalization of the res-
ervoir sands; (C) geologic cross
section shown in Figure 3B dis-
plays the present-day distribution
of the bitumen-saturated reser-
voir sands, water, and gas, as
well as nonreservoir depositional
elements. Comp. A, B, and C
are interpreted lateral compart-
ments; numbers 1 to 5 show
locations of wells used in this
study; AC is abandoned channel-
fill deposit, filled with mud and
classified as nonreservoir flow
unit; MP (mud plug) is the well
location drilled through the
abandoned channel fill. (i – iii)
Location of three tie wells (not
characterized by geochemistry)
used to complete the eastern part
of the cross section showing
significant changes in bitumen/
water saturation between adja-
cent compartments separated by
mud plug 2. Arrow points out
interpreted reservoir charge
pathway below spill point (Fustic
et al., 2012b). ch. A, B, C, X, and Y
= channel deposits A, B, C, X, and
Y. 1 km (0.6 mi). IHS = inclined
heterolithic strata.
172 Fustic et al.

Figure 4. Geology and reservoir flow units within compartment B. Upper part (A – D): Composite logs of wells MP, 1, 2, and 3.
Gamma-ray (GR) log has colored yellow cutoff at 608 API. The resistivity log is logarithmic scale with green color cutoff on
20 ohms. Viscosity and gas chromatography-mass spectrometry identification (GC-MS ID) columns show sample locations
analyzed in this study. Viscosity sample intervals are labeled in red letters B (bottom), M (middle), and T (top). The bitumen %
column shows laboratory-measured bitumen saturation using Dean Stark method. The scale is 0 to 20 mass %. The last
two columns show dipmeter log used for determining the vertical continuity of channel deposits, direction of lateral point-bar
growth, and well-to-well correlations. A, B, and C show the thickness and the lateral extent of each of the stacked channel
deposits. LPB and UPB indicate the lower and the upper point-bar deposits, respectively. Core image shown in Figure 8 is
located by a box labeled core. (E) Schematic cross section between three study wells. MIHS = muddy inclined heterolithic
strata; MP = mud plug; SIHS = sandy inclined heterolithic strata. 250 m (820 ft).

Molecular analysis of the bitumen includes data and downhole preservation of the more resistant com-
from 16 samples obtained from wells 2 and 3 with pounds are caused by the different levels of biodeg-
14 samples from well 1 (Figure 4). Samples were se- radation through the oil column. For example, the
lected from reservoir flow units type 1 and 2, as well biodegradation-susceptible acyclic isoprenoid and n-
as from low-bitumen, high-water saturation zone alkanes, C0-C3-naphthalenes, C0-C2-dibenzothiophenes,
(Figure 4). alkylbenzenes, and alkyltoluenes were not detected in
The concentrations of the selected molecular marker any of the analyzed samples, whereas the biodegrada-
from saturated (regular sterane, diasterane, pregnane, tion is limited for the very resistant triaromatic and
and tetracyclic and pentacyclic terpanes) and aromatic monoaromatic steroid hydrocarbons (Figure 12). Ster-
fractions (alkylnaphthalenes, methylphenanthrenes, ane and diasterane concentration profiles show two
and dimethyl- and ethyl-phenanthrenes) show dif- distinct clusters of data (Figure 10A), whereas the more
ferent profiles (Figures 10 – 12). Sequential removal biodegradation-resistant pregnane (C21 sterane) has
of biodegradation-susceptible molecular compounds narrow ranges of concentration variation along all
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 173

Figure 5. Geology and reservoir

flow units defined in compart-
ment A. Composite logs of (A)
well 5 and (B) well 4. The key to
the description of log columns is
in Figure 4. The X, Y, and Z show
the thickness and the lateral ex-
tent of each of the stacked chan-
nel deposits; (C) geologic cross
section with interpreted large-
scale depositional elements and
top water and gas interval. 60 m
(197 ft).

three columns (Figure 10A). Significant changes in panes (PT) show a sharp decrease in concentration at
the concentration of steranes (Figure 10D) and dia- the base of all three wells (Figure 10C).
steranes coincide with the boundary between the low- The aromatic molecular markers show similar pro-
bitumen, high-water saturation zone, and the reservoir files to those observed from the saturated molecular
type flow unit (Figure 10A) in wells 1 and 2. The well 3 markers. The C4-napthalenes (C4N) represent the lightest
boundary coincides with that between reservoir type 1 aromatic hydrocarbon compound detected in some sam-
and the reservoir type 2 flow units (Figure 10A). Sim- ples (Figure 11A), whereas other more biodegradation-
ilarly, bicyclic sesquiterpane (SST) concentration pro- resistant compounds (i.e., methylphenanthrene, dimeth-
files exhibit an overall downward decrease in con- ylphenanthrenes, and trimethylphenanthrenes) are
centration (Figure 10B). Deviation from the smooth present in a greater number of samples and in higher
downward-decreasing profile occurs within and in the concentrations (Figure 11B, C, D). All of these com-
proximity of the low-bitumen, high-water saturation pounds show a sharply defined contact along the bound-
zone, and at the top of each well (Figure 10B). The tri- ary of the low-bitumen, high-water saturation zone, and
cyclic terpanes (TT) show very little variation in all three the overlying reservoir type 2 (cf. trends in concentra-
well profiles (Figure 10C), whereas the pentacyclic ter- tions of steranes, diasteranes, and SST; see Figure 10A)
174 Fustic et al.

Figure 6. A schematic diagram

representing the key geologic
events associated with reservoir
charging, in-reservoir mixing, and
biodegradation processes in (A)
early stages of reservoir charging;
(B) early stages of reservoir mix-
ing with biogenic gas trapped
under low-permeability layers;
(C) later escape of entrapped gas
with subsequent reoccupation
by water; (D) the present-day
distribution of bitumen, water,
and gas. IHS = inclined hetero-
lithic strata.
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 175

Table 1. Summary table showing thicknesses, porosity, and bitumen and the associated physical properties of the
water saturation values of interpreted reservoir flow units in bitumen. This is indicated by preferential depletion of
wells 1 to 3. 1 m (3.3 ft). more biodegradable compounds and preservation of
the biodegradation-resistant compounds. The primary
control of biodegradation within the study area was
paleo-oil-water contact that migrated downward as the
reservoir was charging. Vertical gradients in hydrocar-
bon compositions and viscosity are typical of the Atha-
basca that correspond to the downward increase in
biodegradation intensity (Bennett et al., 2006; Gates
et al., 2008; Larter et al., 2008).
However, in our experience, we commonly see trends
in physical property and chemical composition that are
interrupted or offset from what are viewed as the typ-
ical vertical gradients (Larter et al., 2008). Bitumen in
low-permeability lithologies have lower levels of bio-
degradation, as shown by greatly increased concentra-
tions of steranes, diasteranes, and SST (Figure 10A);
C4N, methylphenanthrenes (MP), ethylphenanthrenes
(C2P), and trimethylphenanthrenes (C3P) (Figure 11);
and the short-chained TAS and MAS (Figure 12) within
the type 2 reservoir (IHS) flow unit in wells 1 and 2.
Specifically, biodegradation level appears to be more
intense in the clean porous sand units of reservoir flow
unit type 1 and is limited or restricted in the more
varying interbedded sand and shale units (IHS) of res-
ervoir flow unit type 2. Significant offsets coincide with
lithologic boundaries in the trends of molecular con-
centrations of most of the compounds including C3P
(Figure 11D) and SST (Figure 10B) below 250 m (820 ft)
elevation in well 2. This break coincides with the bound-
ary between reservoir flow unit type 1 and the under-
lying muddy nonreservoir flow unit (Figure 4). Samples
from nonreservoir flow units were collected from rare
and apparently isolated bitumen-saturated sand lam-
inae. Oil in mud-dominated lithologies with associated
reduced porosity and permeability and, thus, signifi-
cantly decreased pore network connectivity may have
and a relatively consistent downward-decreasing con- reduced nutrient supply, or as they are thin, the reser-
centration gradient in well 3 (cf. SST) (Figure 10B). voir filled more rapidly and lost their internal water
The depth profiles of long-chained triaromatic (TAS) legs, thus retarding the level of biodegradation (cf.
and monoaromatic (MAS) steroid hydrocarbons show Larter et al., 2008). We conclude that low-permeability
very little variation (Figure 12A). However, the ratio of thin lithologies preserve oil quality.
short-chained TAS and MAS relative to the side (long- The third subset of bitumen properties is com-
chain) chain containing TAS and MAS shows signifi- posed of samples in low-bitumen, high-water saturation
cant changes along the boundary of the low-bitumen, zones in the middle of wells 1 and 2. Stratigraphically,
high-water zone and the overlying reservoir type 2, these zones are located above the top of the reservoir
whereas in well 3, this ratio displays a consistent flow unit type 1 and beneath the overlying reservoir flow
downward-decreasing trend (Figure 12B). unit type 2 (IHS unit). Within and adjacent to low-
bitumen, high-water saturation zones, the vertical com-
Synthesis: Interpretation positional gradients of some (e.g., bicyclic SST in well 3,
Figure 10B), but not all, molecular compounds are sig-
Biodegradation was the dominant factor leading to nificantly deviated from the consistently and smooth-
the present-day molecular distributions within the ly increasing depth normal molecular compositional
176 Fustic et al.

Figure 7. Crossplots of porosity – water saturation and corresponding reservoir flow units. Yellow dots = reservoir type 1;
cyan dots = high-water, low-bitumen saturation zones or reservoir impairment; orange dots = sandy IHS (reservoir type 2);
gray dots = muddy IHS (nonreservoir). IHS = inclined heterolithic strata.

gradients. For instance, the concentration of steranes even by percolation from the surface (Figure 6C)
and diasteranes is the same in the low-bitumen, high- (Barson et al., 2001). These new incursions of water may
water saturation zone and the underlying type 1 reser- have restimulated local microbial activity by enhanc-
voir flow unit (Figure 10A), whereas the concentration ing mineral dissolution (Head et al., 2003) and con-
of SST shows a decrease within the low-bitumen, high- tributing to the increased depletion of SST (Figure 10B).
water saturation zones. The impact of the low-bitumen, high-water satura-
The sample taken in the proximity of the lean zone tion zone extends to the top of the underlying res-
shows increased viscosity (Figure 9), indicating that ervoir type 1. This is evident from the increased oil
its increase is related to the intensified biodegradation viscosity in the middle sample of well 1 (Table 2) and
within and in the proximity of the high-water saturation anomalously decreased concentrations of SST imme-
zones. The low-bitumen, high-water saturation zones diately below that unit (Figure 10B) as well as changes
are characterized by bitumen concentrations of about 4 to a blocky, compartmental profile compared with
wt. % (Figure 4) and oil stain on all sand grains through- smoothly downward-decreasing concentration gra-
out the interval (Figure 8). The oil stain on sand grains dients of steranes, diasteranes, C4N, MP, C2P, and
(Figure 8) suggests that this unit was initially saturated C3P in wells 1 and 2 (Figures 10, 11). The latter com-
with oil. ponents show quite consistent downward-decreasing
Methane generated along the paleo-oil-water con- gradients through reservoir flow unit type 1 in well 3
tact during biodegradation migrated up the oil col- (Figures 10, 11), suggesting that the blocky profiles are
umn by diffusion to form a gas phase below local seals caused by enhanced biodegradation from the low-
(Larter et al., 2003; Jones et al., 2008; Adams et al., 2013). bitumen, high-water saturation zones, coupled by rel-
This process would occur until the water leg was com- atively unrestricted molecular mixing within the very
pletely expelled by the charging oil. Because of low porous type 1 reservoir flow unit.
formation pressures (shallow depth), gas-phase meth- Deviations from the smooth downward-decreasing
ane was trapped against the first muddy layer that molecular concentration gradients are also evident from
marks the base of reservoir flow unit type 2 (IHS unit) subtle relative decreased concentrations of the SST,
(Figure 6B). Preserved bitumen (about 4 mass %) sug- C4N, MP, C2P, and C3P (Figures 10, 11) at the very top
gests that biodegradation was already advanced when samples of all three wells. These samples are taken
biogenic gas arrived into that zone while displacing from the zones of decreased bitumen saturation (de-
most oil to residual oil saturation. Over geologic time, creasing from 14 to about 10 mass %) and increased
methane likely escaped, whereas high bitumen viscos- water saturation (Figure 4). We speculate that increased
ity prevented oil remigration, leaving the space for water saturation is either caused by the increased orig-
reincursion of mobile water, perhaps, sourced from the inal connate water that has not been replaced by pe-
surrounding shale layers, nearby aquifer systems, or troleum during reservoir charge because of the lower
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 177

Figure 8. Core images of depth interval from 197 to 220 m (646 – 722 ft) from well 1 with posted locations of GC-MS sample 6
to 11. Each tube is about 75 cm (30 in.) long. In each tube, the bottom is on the right side, and the top is on the left side.
White lines show interpreted contacts between the type 1, the low-bitumen, high-water saturation zone, and the type 2
reservoir flow units. Note the apparent lack of any sedimentologic barrier between type 1 and low-bitumen, high-water
saturation zone. The border of these two units is indicated by changes in color of sand from black (type 1) to brown
(low-bitumen, high-water saturation zone). This point is indicated by a sharp break on the resistivity log (Figure 4B). Residual
oil stain on all sand grains in low-bitumen, high-water saturation suggests that this zone was probably previously occupied
by oil.

permeabilities or more likely caused by the water that most part of the reservoir related to gas leakage and
percolated into the reservoir from the overlying units mobile water reinjection.
over geologic time (Barson et al., 2001). Although bitumen-
saturated zones are generally considered as aquitards, Southern Athabasca Deposit: Compartment A
when water is injected in the formation under high
pressure, displacement followed by imbibition and Geology and Present-day Fluid Distribution
subsequent migration through the once saturated bi-
tumen zone may occur even on a human time scale This study area includes fluid characterization data from
(Aherne and Maini, 2008). We conclude that another two stratigraphic wells (Figures 3, 5) that are approxi-
factor controlling biodegradation locally in compart- mately 640 m (2100 ft) apart. This is comparable to typ-
ment B is increased water saturations in the upper- ical SAGD well pair lengths in oil sand developments.
178 Fustic et al.

Table 2. Wells 1, 2, and 3 bitumen viscosity values measured at 35, 55, and 758C, respectively.

Midpoint
Well ID Position Sample Name Sampled Interval Elevation (m) Viscosity (cp)
358C 558C 758C
Well 1 Top T 273.2 – 276.2 274.7 84,975.3 8371.8 1524.2
Well 1 Middle M 258.2 – 261.2 259.7 275,007.5 17,707.0 3260.3
Well 1 Bottom B 255.2 – 258.2 256.7 200,389.6 18,785.6 3423.0
Well 2 Top T 282.1 – 285.1 283.6 48,966.9 5462.8 1144.1
Well 2 Bottom B 264.3 – 267.3 265.8 262,985.8 20,404.3 2932.3
Well 3 Top T 284.5 – 287.5 286.0 46,830.4 5925.6 1144.0
Well 3 Middle M 272.9 – 275.9 274.4 58,830.5 6413.6 1271.8
Well 3 Bottom B 255.8 – 258.8 257.3 187,433.6 15,414.0 2774.4

The thickness of the McMurray Formation ranges Z were accreting to the east, whereas channel Y was
between 60 and 70 m (197 and 230 ft) (Figure 5). Based accreting toward the west. The lower part of each chan-
on seismic, dipmeter, and channel scours observed in nel deposit is characterized by clean sand, and data
cores, it has been interpreted to consist of an approx- from wells 4 and 5 show a somewhat different order of
imately 7-m (23-ft)-thick continental flood-plain de- large-scale depositional units. The base of channel X in
posit at the base of well 5; major stacked meandering well 5 contains a several-meters-thick channel-lag brec-
river channel deposits X and Y and a smaller mean- cia deposit, and the sandy IHS (type 2 reservoir flow
dering river channel Z (Figure 5). The preserved thick- unit) of channel X extends to the base of the channel
ness of channel X ranges between 25 m (82 ft) in well 5 deposit in well 4 (Figure 5). Channel Y at well 4 is char-
and 27 m (89 ft) in well 4, whereas the thickness of the acterized by clean trough cross-bedded sand through
channel Y is about 25 m (82 ft) at both locations. the entire preserved thickness, whereas in well 5, it
Channel Z (well 4) is about 7 m (23 ft) thick (Figure 5). fines upward into sandy IHS (Figure 5). Petrophysical
Dipmeter data (Figure 5) suggest that channels X and and core analyses show that the upper parts of both

Figure 9. Bitumen viscosity profiles measured at 358C superimposed on the interpreted reservoir flow units. IHS = inclined
heterolithic strata. 10 m (33 ft).
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 179

Figure 10. Concentration (Mg/g oil) profiles of the selected saturated molecular compounds posted on interpreted reservoir
flow units (see legend). (A) Concentration profile of regular steranes, diasteranes, and pregnanes; (B) concentration
profile of sesquiterpanes (SST); (C) concentration profile of tetracyclic (TT) and pentacyclic (PT) terpanes; (D) selected
mass chromatogram (m/z) 217 traces from well 1 shows downward-decreasing concentration of steranes and diasteranes
(i.e., C – F) and intact pregnanes (i.e., traces A and B) indicative of the sequential removal of the more biodegradation-
susceptible molecular compounds. Key for identification of C21-C30 steranes: aaa abb, A-C21 aaa + abb pregnane, B-C 22
aaa + abb pregnane, C-C27 ba 20R diasterane, D-C 29 ba 20R diasterane, E-C 29 aaa 20S sterane, F-C29 aaa 20R sterane,
1-14 are relative locations of geochemical samples sorted by elevation (for sample location see Figure 4B). The deepest
sample, labeled 1, is from 239 m (784 ft) above sea level (ASL) and the shallowest sample (14) is from 294 m (965 ft) ASL.
10 m (33 ft).

wells are characterized by high water saturation and IHS), whereas nonreservoir flow units include top water
the presence of gas (Figure 5). and gas (Figure 5C). Although some overlap exists, water
The interpreted stacking patterns of the large-scale saturation versus porosity crossplots (Figure 13) show
depositional elements and water- and gas-occupied in- clustering of data points within each of the selected
tervals (Figure 5C) show that the stacked clean sands of reservoir flow units. Sandy IHS at the base of well 4 is
the lower point-bar deposits of channel X and channel Y characterized by porosities exceeding 25%, inferring
(well 5 in Figure 5A) form a thick type 1 reservoir flow excellent reservoir properties, whereas gas-occupied
unit. Other reservoir flow units include type 2 (sandy IHS units at the top of well 5 are characterized by
180 Fustic et al.

Figure 11. Concentration (Mg/g oil) profiles of the selected aromatic hydrocarbon compounds posted on the top of the
interpreted reservoir flow units: (A) C4 alkylnaphthalenes (C4N); (B) methylphenanthrenes (MP); (C) dimethylphenanthrenes
and ethylphenanthrenes (C2P); (D) trimethylphenanthrenes (C3P); (E) selected mass chromatograms (m/z 184) from
well 3 showing systematic changes of C4 alkylnaphthalenes. Key for identification of C4N: 1-1,3,5,7 TeMN; 2-1,3,6,7 TeMN;
3-1,2,4,6+1,2,4,7+1,4,6,7 TeMN; 4-1,2,5,7 TeMN; 5-2,3,6,7 TeMN; 6-1,2,6,7 TeMN; 7-1,2,3,7 TeMN; 8-1,2,3,6 TeMN; 9-
1,2,5,6+1,2,3,5 TeMN. 1-16 are relative locations of geochemical samples sorted by elevation. The deepest sample labeled 1
is from 238 m (781 ft) above sea level (ASL) and the shallowest sample (14) is from 295 m (968 ft) ASL (for sample location
see Figure 4B). 10 m (33 ft).

porosities as low as 15% (Figure 13; Table 3). Geologic Fluid Property Variations within the Geologic
cross sections between wells 4 and 5 show a continuity Framework: Results
of flow units through the study area and the lateral
trapping of the top water and gas to the east by a mud Oil viscosity measurements were obtained from six
plug deposit (Figure 3). samples from bitumen-saturated zones of reservoir
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 181

Figure 12. (A) Concentration (Mg/g oil) profiles of the triaromatic (TAS) and monoaromatic (MAS) steroid hydrocarbons posted
on top of the interpreted reservoir flow units; (B) ratios of short side chain versus long side chain containing TAS and MAS.
10 m (33 ft).

type 1 and type 2 (Figure 5; Table 4) in both studied and 4 (Figure 5). The concentration of the saturated
wells. In well 5, the dead oil viscosity (208C) results hydrocarbon molecular markers shows different pro-
display a linear downward-increasing gradient from files in the studied wells (Figures 15– 18). Sequential
about 4  105 cp (at 208C) at the top to almost 1  107 cp removal of molecular compounds more susceptible to
(at 208C) at the base (Figure 14). In well 4, the data show biodegradation and preservation of the more resistant
two clusters. The bottom 4 samples increase down- compounds is indicative that variation in molecular
ward from about 3.6  106 cp (at 208C) in the middle of marker concentrations is caused by different levels of
the column to almost 9  106 cp (at 208C) at the base biodegradation. The biodegradation-susceptible acy-
(Figure 14). Results from the top three samples are quite clic isoprenoid and n-alkanes, as well as C0-C3-
uniform, ranging from 3.3  106 to 3.7  106 cp (Table 3; naphthalenes, C0-C2-dibenzothiophenes, alkylben-
Figure 14). zenes, and alkyltoluenes, were not detected in any of
Molecular analysis of the bitumen was performed the analyzed samples, whereas biodegradation-resis-
on 15 and 16 bitumen samples, respectively, and col- tant TAS and MAS appear to be intact. Sterane and
lected from all identified reservoir flow units, includ- diasterane concentration profiles (Figure 15A), SST
ing the top water and the top gas zones from wells 5 (Figure 15B), and pentacyclic terpanes (Figure 15C)
182 Fustic et al.

Figure 13. Crossplots of porosity-water saturation shown for each interpreted reservoir flow unit.

show different patterns in the two wells. Deviations MAS shows very little variation along both well profiles,
from the consistently and smoothly increasing depth whereas the ratio of short-chained TAS to long-chained
molecular compositional gradients coincide with TAS shows subtle zonation in well 4 and a strong
reservoir flow unit contacts and/or sedimentologic relatively consistent linear downward-decreasing ver-
breaks (i.e., for well 5 at 280 and 304 m [919 and 997 ft] tical gradient in well 5 (Figure 17B). The ratio between
[Figure 15A–C] and for well 4 at 275 and 290 m [902 and C29-ab-hopane and C30-ab-hopane in the very bottom
951 ft] [Figure 15A–C]). samples of both wells (Figure 18) greatly increases, indi-
The aromatic hydrocarbon molecular markers are cating that C30-ab-hopane is more susceptible to bio-
intensely depleted. The C0-C2-alkylnaphthalenes, C0-C2 degradation than C29-ab-hopane (Bennett et al., 2006).
alkyl-dibenzothiophenes, alkylbenzenes, and alkylto-
luenes were not detected or were present as trace com-
ponents. The C3-alkyl naphthalenes (C3N) represent Synthesis: Interpretation
the lightest aromatic hydrocarbon compound detected
at the top of well 5 only. Other more biodegradation- Preferential depletion of more biodegradable com-
resistant compounds (i.e., tetramethylnaphthalenes, pounds and preservation of biodegradation-resistant
methylphenanthrenes, and dimethylphenanthrenes) compounds suggest that biodegradation was the dom-
are present in a greater number of samples and in higher inant factor leading to the present-day molecular dis-
concentrations, but only at the upper part of the petro- tributions within the bitumen. Vertical compositional
leum columns of wells 4 and 5 (Figure 16A – C). The and viscosity gradients showing an overall downward-
molecular concentration profiles of TAS and MAS as increasing biodegradation intensity suggest that the
the most resistant to the biodegradation component, primary control on biodegradation within the study area
show very little variations in both wells (Figure 17A). was in proximity to the paleo-oil-water contact. Thus,
The ratio of short-chained MAS over long-chained overall, the most degraded and viscous petroleum is
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 183

Table 3. Summary table showing thicknesses, porosity, and high. Petrophysical logs (Figure 5) and core image inves-
water saturation values of interpreted reservoir flow units in tigation (Figure 19) do not show any evidence of the li-
wells 4 and 5. 1 m (3.3 ft). thologic boundary that may have controlled this contact.
Well 5 shows a change in slope at 280 m (919 ft) (the
slope of the concentrations of steranes and diasteranes;
Figure 15) and subtle offset in the vertical gradient
for the changes in concentrations of pregnanes, TT, PT,
as well as TAS and MAS. This break does not coin-
cide with any interpreted reservoir flow unit or de-
positional boundary. We speculate that even slightly
increased water saturation may be responsible for the
increased biodegradation. The bitumen concentration
in that area (at 280 m [919 ft]) slightly drops from 14 to
12 mass %.
Unlike the integrated results shown in the compart-
ment B case study, the results from this study area
show little correlation between bitumen composition
and mapped reservoir flow units. Bitumen encoun-
tered in well 4 has an overall higher degradation level
than in well 5. This is evident from the five to eight
times higher viscosity at the top of well 4 (Table 4) than
for the top 2 viscosity samples of well 5. It is also seen
from the low concentrations of steranes and diaster-
anes (Figure 15A), as well as the complete absence of
C3N and a smaller number of samples with detectable
C4N, MP, and C2P (Figure 16).
In addition to an overall downward-decreasing con-
centration bitumen molecular compositional gradient,
the similarity between the two wells includes strikingly
typically found at the bottom of the reservoirs, whereas similar viscosity (5.36  106 cp and 5.7  106 cp at
the least degraded is located at the top of the reservoir. 208C) at an elevation of about 268 m (879 ft) (Figure 14;
Major departures from a smoothly downward- Table 4) and similar concentrations of steranes, diaster-
increasing biodegradation level as commonly seen in anes, SST, TT, PT, TAS, and MAS from 280 m (919 ft) to
oil sand reservoirs are locally indicated by the reduced the base of the reservoir (Figures 15–17).
concentrations of saturated and aromatic hydrocarbon
molecular markers. Examples include reduced con-
centrations of steranes, diasteranes, SST, TT, and PT in Table 4. Wells 4 and 5 bitumen viscosity values measured at
20, 54, and 808C. 1 m (3.3 ft).
sample 15 (Figure 15), and C4N, MP, C2P (Figure 16),
which are taken from the top of the top gas zone in well
Elevation
5 (Figure 5). Sample 14 was taken from the same top gas
Well ID (m) Viscosity (cp)
zone, but it does not show decreased concentration of
the analyzed molecular markers. The same zone in well 208C 548C 808C
4 has evidence of the virtually complete depletion of all Well 5 296.5 401,344 6429 819.1
aromatic molecular markers (Figure 16) and noticeable Well 5 290.7 590,359 8566 970.2
depletion of SST (Figure 15B). Well 5 281.1 2,467,270 24,458 2072.0
Sample 13 in well 5 was taken from the bitumen-gas Well 5 275.7 4,068,020 28,575 2405.0
transition zone (Figure 5). The gas and bitumen-gas Well 5 266.2 5,367,802 32,891 2778.0
transition zones are identified based on decreased Well 5 253.4 9,693,794 49,279 3720.0
neutron-log readings in the porous interval, which is a Well 4 288.6 3,307,333 25,946 2324.0
diagnostic pattern for pores occupied with gas instead Well 4 278.4 3,740,594 26,144 2318.0
of water and/or oil (low response comes from the re- Well 4 272.4 3,591,764 26,392 2225.0
Well 4 264.2 5,700,400 32,657 2769.0
duced density of the hydrogen atoms in low-density
Well 4 257.1 7,716,800 41,387 3336.0
gas relative to water or oil). Dean Stark analysis results
Well 4 249.7 8,912,600 42,638 3358.0
(Figure 5) indicate that the bitumen saturation is still
184 Fustic et al.

Figure 14. Bitumen viscosity (cp)

profiles measured at 208C super-
imposed on the interpreted res-
ervoir flow units. 10 m (33 ft).

The above molecular information and physical prop- concentration of many compounds up to 10 m (33 ft)
erty data may be used to interpret the reservoir charg- below the present-day bitumen –top water contact in
ing and mixing history. We suggest that the reservoir well 4 (Figures 15–17).
was first filled by low-maturity oil (Adams et al., 2013) Gas located at the top of the reservoir suggests the
from the same or very similar source rocks. In the ab- presence of a competent cap rock that prevents gas es-
sence of major barriers, the initial oil-water contact cape from the present-day accumulation. Thus, we sug-
migrated downward as the reservoir filled. Methane gest that fissures that allowed water incursion closed
generated along the oil-water contact was mobilized over time and isolated fluid and bacteria within the
upward by buoyancy and trapped against muddy given compartment. The origin of the present-day gas
overlying layers (Figure 5) at the top (Clearwater For- may derive from either biodegradation at the bottom of
mation) and mud plug deposits laterally (Figure 5C). the reservoir or locally generated along the top water–
Considering the laterally compartmentalized nature bitumen contact, or both processes may have contrib-
of the reservoir (Figures 3B, C; 4) in the study area, it uted. Isolation of a gas zone in well 4 between two thick
is difficult to accept the lateral migration of large water zones and a significant amount of solution gas
amounts of water encountered at the top of this com- in water support our suggestions for the present-day
partment. Leakage of all or part of the generated bio- formation of the biogenic gas, which is in line with
genic gas through the seal would have allowed the discoveries of polar lipid membranes (Oldenburg
gas to escape upward and be replaced by a downward et al., 2009) and the presence of the methanomicrobiales
intrusion of meteoric water. Decreased concentrations (Hubert et al., 2012) in oil sand reservoirs. Because of
of SST in the top water zone suggest that top water relatively shallow depths, the recently generated bio-
may have encouraged local petroleum degradation genic gas is under pressure, and even a minor perme-
because water supports microbial life and nutrient ability barrier results in accumulation.
diffusion for their metabolism. The increased level of
biodegradation is observed within the invaded zones Comparison of Fluids and Processes between A
and also in bitumen samples from the underlying and B Compartments
layers likely because of diffusion of nutrients along the
mineral-clay grain contacts. The increased level of the Molecular maturity parameters that are sensitive to
biodegradation is evident from reduced viscosity and changes in thermal maturity (TAS/[TAS + MAS]) show
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 185

Figure 15. Concentration (Mg/g oil) profiles of saturated hydrocarbons posted on interpreted reservoir flow units (see legend):
(A) steranes, diasteranes, pregnanes; (B) SST; (C) TT and PT terpanes; (D) partial reconstructed mass chromatogram (m/z 217)
displaying the systematic removal of biodegradation-sensitive sterane and diasterane compounds in well 5. See Figure 11
for the key for identification of compounds. Peaks A – F same as in Figure 11. Numbers 1 to 14 are the relative locations of
geochemical samples sorted by elevation. The deepest sample labeled 1 is from 252 m (826 ft) above sea level (ASL) and
the shallowest sample (14) is from 305 m (1001 ft) ASL (Figure 5A).

that the oil in compartment A was charged with oil has been severely altered by biodegradation, and thus,
of slightly higher thermal maturity than compart- most of the initial compositional differences are over-
ment B (Fustic et al., 2012b). The molecular compo- printed by the effect of in-reservoir biodegradation
sition shows that the petroleum in both compartments processes.
186 Fustic et al.

Figure 16. Concentration (Mg/g oil) profiles of selected aromatic hydrocarbons posted on interpreted reservoir flow units
(see legend): (A) tetramethylnaphthalenes (C4N); (B) methylphenanthrenes (MP); (C) dimethylphenanthrenes (C2P);
(D) m/z 192 from well 5 shows systematic removal of methylphenanthrene isomers. Peaks 3, 2, 9, and 1 point different isomers.
1–15 are relative locations of geochemical samples sorted by elevation. The deepest sample labeled 1 is from 253.4 m (831 ft)
above sea level (ASL) and the shallowest sample (15) is from 296.5 m (973 ft) ASL (for sample location see Figure 5A).

Similarities between the two compartments include the concentrations of many compounds in and near the low-
overall downward trend of decreasing concentrations of bitumen, high-water saturation zones, and top waters.
molecular compounds that are biodegradable under se- Differences between two compartments include the
vere levels of biodegradation (PM6–PM9) and decreased stratigraphic position of the low-bitumen, high-water
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 187

Figure 17. (A) Concentration

(Mg/g oil) profiles of TAS and
MAS posted on interpreted res-
ervoir flow units (see legend);
(B) ratio of short side chain TAS
over side chain containing TAS
and ratio of short side chain
MAS over chain containing MAS
posted on interpreted reservoir
flow units (see legend). TG = top
gas; TWG = Top water and top
gas; NR = nonreservoir facies.

saturation zones encountered in the middle of the pe- biodegradation processes in time and space, includ-
troleum column in compartment B and at the top of the ing their effect on present-day spatial distributions
column in compartment A. Although occurring at dif- and fluid properties of bitumen, water, gas, and low-
ferent stratigraphic intervals, the origins of the low- bitumen, high-water saturated zones. The reservoir was
bitumen, high-water saturation zones, and top waters sequentially filled from the west by low-maturity oil in
are interpreted similarly as former biogenic gas caps, in compartment A, then reservoir compartment B, and
which, after gas escapes, the pores were reoccupied by finally reservoir compartment C (Figure 6A) (Fustic
water. Additional differences include the strong corre- et al., 2012b). The buoyancy-driven oil first occupied
lation between reservoir flow units and bitumen mo- the most porous and uppermost parts of each reser-
lecular composition in compartment B and the asso- voir compartment and then filled the rest of the pores
ciated lack of correlation in compartment A. This until it spilled into the next compartment (Fustic et al.,
suggests that the biodegradation in compartment A 2012b). When reservoir charging ceased, petroleum did
crossed some lithologic boundaries, a situation pre- not fill the easternmost compartment C (Fustic et al.,
viously described in McCaffrey et al. (1996). 2012b) (Figure 6A). Initial oil in the reservoir compart-
The above information, summarized in Figure 2, ments was attacked by microbial communities in the
illustrates reservoir charging, reservoir mixing, and formation waters (Figure 6A). During reservoir charging,
188 Fustic et al.

Figure 18. Plot of the (A) ratio of C29 AB hopane/C30 AB hopane posted on interpreted reservoir flow units (see legend); (B) C29
AB hopane/C30 AB hopane with selected m/z 191 mass chromatograms showing the changes in relation to peak heights
between C29 and C30 hopanes, in the very bottom sample of both wells. 10 m (33 ft).

the oil-water contact in each separate compartment impermeable layers, such as IHS in compartment B
migrated downward. In each compartment, the reser- and cap-rock mudstones in compartment A (Figure 6B).
voir oil-water contact gets broader and broader as it Over geologic time, some, if not all, trapped gas es-
fills, allowing more extensive biodegradation. Gen- caped. Locally, gas escape might have been via frac-
erated gas migrated up the column until trapped by tures in the overlying rocks caused by structural events,
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 189

occur when the accumulated petroleum column ex-

ceeds the capillary threshold pressure of the cap rock
especially at shallow burial depths. At the time gas
escaped, the bitumen was already immobilized and,
thus, not able to refill these zones. Thus, water oc-
cupied previous gas caps (Figure 6C). These waters,
along with the nutrients, or possibly even new bac-
terial consortia, probably reactivated biodegradation.
The above processes likely occurred after the cessation
of the reservoir charge associated with cessation of the
orogeny (Stott, 1984) and subsequent basin uplifting.
Although cessation of the orogeny stopped petroleum
generation and reservoir charging, this could not stop
fluid migration within the compartments, which is
continuous even today.
In addition to the overall downward-decreasing ver-
tical gradients caused by downward-moving oil-water
contacts and vertical component redistributions by dif-
fusion, the present-day bitumen compositions show
departures from that gradient. Departures from gradi-
ent include reduced degradation in zones of reduced
porosity and permeability and enhanced degradation in
proximity to low-bitumen, high-water saturation zones
regardless of their position in the reservoir (middle or
top). The petrophysical data from wells drilled in com-
partment C (not included in geochemical studies) show
the presence of a thick gas cap. This is likely biogenic gas
generated over geologic time by biodegradation, which
continues today (Figure 6D).

Northern Athabasca Study Region

This study area is located in the northern Athabasca

region, within the mine development designated part
of the Athabasca oil sand deposit, township 95 and
range 9 (Figure 1). Data presented in this study were
collected from two vertical delineation wells (Figure 20A,
B) positioned about 1 mi (1.6 km) apart. Schematic geo-
logic cross sections (Figure 20C) between two wells
were supported with data from a dozen wells (not
shown) between them.

Figure 19. Location of the top three samples in well 5. Geology and Present-day Fluid Distribution
Samples 14 and 15 are from a low-bitumen saturation
zone filled up by gas. Sample 13 is from a high-bitumen Within the study area, the McMurray Formation is
saturation zone in which the neutron-density crossover approximately 100 m (328 ft) thick (Figure 20). Based
indicates the presence of gas (Figure 5A).
on facies descriptions from core studies and dipmeter
analysis, the reservoir has been interpreted to consist of
such as the dissolution of underlying Prairie Evapo- stacked braided river channel deposits (LM1, Figure 2),
rite Formation, or glacial unloading, or simply by the flood-plain deposits (LM2, Figure 2), open-estuarine
vertical migration or leakage through the matrix pore subtidal sand bars (MM1, Figure 2), and tidally
network or permeability of incompetent top seals by influenced meandering river channel deposits (MM2,
either diffusion or bulk vertical flow of trapped pe- Figure 2). Geologic interpretation between the studied
troleum (Jokanola 2007; Jokanola et al., 2009). This may wells suggests that the braided river and associated
190 Fustic et al.

Figure 20. Geology and reservoir

flow units in the northern Atha-
basca wells. Composite logs repre-
senting (A) well 6 and (B) well 7
(key for description of log columns
same as in Figure 4); (C) schematic
(not to scale) geologic cross section
outlining the extent of the major
interpreted reservoir flow units be-
tween two studied wells. The M
and N are interpreted as two ver-
tically stacked reservoir compart-
ments separated by impermeable
continental flood-plain mud de-
posit (CFP). GC-MS ID = gas
chromatography-mass spectrom-
etry identification. 80 m (262.5 ft).
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 191

Figure 21. Crossplots of porosity and water saturation for wells (A) 6 and (B) 7. Yellow dots are sample points from the reservoir
type 1, orange are from the reservoir type 2, blue represents bottom water, white is from the low-bitumen, high-water
saturation zone, dark gray is from the flood-plain deposit and light gray is from the nonreservoir muddy inclined heterolithic
strata (IHS) unit.

overlying flood-plain deposits are laterally continuous, saturated by bitumen (Figure 20). Intervals with in-
and that river channel deposits have complex and creased and high water saturation in the lower part of
significant lateral facies changes (cf. lateral compart- the study area are classified as, low-bitumen, high-water
mentalization, Figures 3, 6). The upper McMurray saturation zone, and bottom-water flow units, re-
Formation sedimentary deposits in the southern Atha- spectively (Figure 20). The laterally extensive flood-
basca case study are absent here because of Pleistocene plain deposit separates the reservoir into two vertical
glacial erosion. The McMurray Formation is covered compartments (compartments M and N in Figure 20C).
only by a couple of meters of Pleistocene sediments Compartment N in well 6 is interpreted to consist of
and/or Holocene muskeg deposits. stacked open-estuarine subtidal sand bars and two
Porosity–water saturation crossplots (Figure 21) show stacked meandering river channel deposits (note two
that despite different genetic origins, different deposi- fining-upward patterns on the gamma-ray [GR] log;
tional units and elements have the same or very sim- Figure 20A), whereas in well 7 (Figure 20B), that com-
ilar reservoir rock properties. This allows grouping of partment is interpreted as a single, 45-m (148-ft)-thick
them into a narrower range of reservoir flow units. meandering channel deposit, as shown by continu-
For instance, lower McMurray braided river channel ous fining-upward pattern on GR log (Figure 20B). As
sand, middle McMurray subtidal sand bars, and lower a result, reservoir flow units in well 6 show an inter-
point-bar deposits all have porosities exceeding 30% layered pattern of type 1 and 2 reservoir flow units,
(Figures 20, 21). Thus, when saturated by bitumen, whereas in well 7, the reservoir quality generally dete-
these units are classified as reservoir flow units of type 1 riorates upward. With most porosity values exceed-
(cf. Fustic et al., 2013). Nonreservoir flow units (flow ing 30%, the reservoir quality of compartment M is
barriers) are flood-plain deposits and muddy IHS consistently excellent throughout the unit (Figures 20,
deposits. Reservoir type 2 flow units are IHS facies 21). However, a significant part of that reservoir
192 Fustic et al.

Table 5. Summary table showing thicknesses (m) and Fluid Property Variations within the Geologic Framework
porosity and water saturation values of interpreted reservoir
flow units in wells 6 and 7. 1 m (3.3 ft).
The salinity and water chemistry of formation and
bottom-water zones were determined from resistivity
logs and Pickett plots. The results show uniform sa-
linity of about 1700 ppm in the entire upper compart-
ment. In the lower compartment, the salinity increases
from 2700 ppm in well 6 to 5500 ppm in well 7.
Oil viscosity measurements were obtained from five
samples in both studied wells (Figure 20A, B; Table 6).
In the upper compartment (compartment N) in both
wells, the results show typical vertical viscosity gra-
dients where oil viscosity increases with depth (Table 6;
Figure 22). However, samples from the lower compart-
ment (compartment M) show opposing trends in the
studied wells. Although the bottom two samples of
well 6 show somewhat lower viscosity than the low-
est sample in the upper compartment, the lowest sam-
ple in well 7 shows a significant increase in viscosity
compared with the upper compartment in the same
well (Table 6; Figure 22).
Molecular analysis of the bitumen from wells 6 and
7 was performed on 11 and 12 samples, respectively
(Figure 20). Samples were collected from both com-
partments and from all mapped reservoir flow units
saturated by bitumen (Figure 20).
The concentration of the molecular markers shows
different profiles in the studied wells (Figures 23 – 26).
Sequential removal of the biodegradation-susceptible
molecular compounds and preservation of the bio-
degradation-resistant compounds indicate that the var-
iation in molecular marker concentrations is caused
by different levels of biodegradation. For example,
acyclic isoprenoid and n-alkanes as well as C0-C2-
naphthalenes were not detected in any of the analyzed

Table 6. Wells 6 and 7 bitumen viscosity (cp) values

measured at 38, 54, and 808C.

Well ID Depth (m) Viscosity (cp)

compartment is occupied by bottom water and/or low-
bitumen, high-water saturation zone (Figure 20C). In 388C 548C 808C
well 6, an 18-m (59-ft)-thick high-water, low-bitumen, Well 6 7.4 12,419 2687 403.3
high-water saturation zone is identified between thin Well 6 27.1 18,256 3622 508.1
bottom water and the overlying bitumen (Figure 20A), Well 6 27.1 21,977 4410 580.4
whereas in well 7 (Figure 20B), a 24-m (79-ft)-thick bot- Well 6 41.9 51,594 8702 965.4
tom water is overlain with a 7-m (23-ft)-thick low- Well 6 49.4 25,896 4723 620.8
bitumen, high-water saturation zone (Figure 20B, C). Well 6 57.4 39,341 6921 853.3
Reservoir properties for each flow unit (Table 5) show Well 7 17.2 11,179 2401 372.6
a narrow range of porosities among bottom-water and Well 7 23.5 18,331 3643 503.5
Well 7 37.2 23,912 4629 608.8
high-water, low-bitumen and type 1 reservoir flow units,
Well 7 52.3 32,693 5801 727.6
although water saturation significantly varies among
Well 7 74.6 78,384 12,039 1245.0
them.
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 193

Figure 22. Bitumen viscosity (cp)

profiles measured at 208C super-
imposed on the interpreted res-
ervoir flow units. Arrows indicate
downward-increasing viscosity
gradients in each compartment.

samples, indicating that they are vulnerable to bio- Synthesis: Interpretation

degradation, whereas the triaromatic and monoaro-
matic steroids appear intact because of their high The study of the source rock thermal maturity pa-
resistance to biodegradation (cf. Peters et al., 2005) rameters from the above samples shown suggests that
(Figure 26A). both reservoir compartments are charged with the oil
In addition, the concentrations of steranes, diaster- from the same or very similar source rocks and of the
anes, and even pregnanes (Figure 23A), pentacyclic same thermal maturity (Fustic et al., 2012b). Thus, the
terpanes (Figure 23B), as well as C3N, C4N, and C5N present-day bitumen heterogeneities are primarily
(Figure 24A), and MP and C2P (Figure 24B) show a caused by different levels of biodegradation. Molec-
systematic downward decrease in concentration within ular analysis suggests that biodegradation was the
compartment N. However, the concentration of the mo- dominant factor causing present-day variations in the
lecular markers within compartment M is opposite. Al- molecular composition of the bitumen. Bitumen vertical
though the concentration of the molecular markers molecular compositional (Figures 23–26) and viscosity
from compartment M in samples from well 6 shows a (Figure 22) gradients in the upper compartment (N)
relative increase in component concentration compared suggest that the primary control of biodegradation was
with the bottom samples of compartment N, bitumen access to the paleo-oil-water contact that migrated
samples from well 7 show a significant decrease in con- downward as the reservoir was charged (Figure 27B –
centration relative to the samples from compartment N D). Comparison of panels B and C of Figure 27 shows
(Figures 23, 24). that the oil-water contact within each compartment
The ratio between 9MP and 1MP, used as biodeg- increases as reservoir charges, and that eventually, oil-
radation indicator (Peters and Moldowan, 1993), shows water contacts of different compartments may emerge
significant changes at the bottom part of the upper (Figure 27C). The same effect and explanation applies to
compartment in well 6 (Figure 25A), whereas these the bitumen compositional gradients seen in the lower
ratios in well 7 show two distinct clusters, coinciding compartment (Figures 23–26), where a present-day oil-
with reservoir type 1 and 2 flow units, respectively water contact still exists. However, the comparison of
(Figure 25A, B). the molecular markers between the upper and the low-
The ratio of short-chained MAS over longer chain er compartment (Figures 23–26) indicates that the pro-
MAS shows very little variation along both well pro- cesses occurring within them were independent from
files, whereas the ratio of short-chained TAS over the each other in the two isolated compartments. In both
long-chained TAS shows some systematic downward- compartments, the primary control on the level of the
increasing trends in the upper compartment from both biodegradation was access to the paleo-oil-water con-
wells (Figure 26B). tact that migrated downward as the reservoir was
194 Fustic et al.

Figure 23. Concentration (Mg/g oil) profiles

superimposed on interpreted reservoir
flow units (see legend) for (A) steranes,
diasteranes, and pregnanes and (B) TT and
PT in wells 6 and 7. LB-HW = low-bitumen,
high-water saturation zone (transition
zone). 10 m (33 ft).

charged. Thus, overall, the most degraded petroleum degradation level in compartment N in well 7 is more
is found at the bottom of each compartment, whereas intense than in well 6 despite the higher water salinity
the least degraded is at the top of each compartment. (2700 ppm in well 6 to 5500 ppm in well 7), which sug-
The slight differences in initially charged petroleum in gests that water salinity is not a major factor affecting
different compartments are most likely overprinted by biodegradation. Alternatively, water salinity may have
biodegradation, except differences in the concentration increased in the later stages of the reservoir history be-
of the biodegradation-resistant triaromatic and mono- cause of the westerly progressing salt dissolution of the
aromatic steroids (Fustic et al., 2012b). underlying Prairie Evaporates Formation (McPhee
The second level of control on the overall level of and Wightman, 1991) after the oil was already heavily
the biodegradation appears to be the persistence of an degraded.
oil-water contact in the lower compartment. The bio- High-resolution geochemical sampling reveals ad-
degradation along this contact is intensified in well 7 ditional compositional anomalies, particularly with-
but is not as evident in well 6. We interpret it as a in the upper compartment. Concentration profiles of
function of the thickness and persistence of the water steranes, diasteranes, pregnanes (Figure 23A), and
leg and increased water saturation in the lower com- PT (Figure 23B) show slightly anomalous increases
partment (Figure 20B). Thicker water leg in well 7 within the reservoir type 2 flow unit (IHS unit) and
(Figure 20B, C) allowed for more intensified biodegra- within sand laminae of the nonreservoir muddy IHS
dation, as documented by Huang et al. (2004). The bio- deposit and flood-plain deposits, suggesting that locally,
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 195

Figure 24. Concentration (Mg /g oil)

profiles superimposed on interpreted res-
ervoir flow units (see legend) for (A) C3N,
C4N, and C5 and (B) MP and DMP (C2P)
in wells 6 and 7. LB-HW = low-bitumen,
high-water saturation zone (transition zone).
10 m (33 ft).

low-permeability lithologies retard the level of the Petroleum at the top of the lower compartment in
biodegradation. well 6 is less degraded than that at the base of the
The lower compartment was not fully charged by upper compartment, suggesting a more recent charge
petroleum. Well 7 reservoir has been only partially or shorter exposure to the paleo-oil-water contact.
filled at the top, whereas well 6 reservoir contains bitu- The same phenomenon is absent in well 7 because of
men almost to the base of the reservoir column. Bitumen the enhanced biodegradation associated with the ex-
saturated well 6 that is situated west of well 7. In both tremely thick bottom-water zone or persistence of the
wells, the bottom water is overlain by a low-bitumen, oil-water contact. Within the study area, the charg-
high-water saturation zone. The base of the low-bitumen, ing of the upper compartment may have been com-
high-water saturation zone is quite flat (Figure 20C). pleted when the lower compartment was still receiv-
However, the origin of that zone is not clear. We spec- ing oil.
ulate that the zone formed as a transition zone dur- A summary diagram depicting geologic framework
ing the charging of the lower compartment, but also and events, including reservoir charging in reservoir
that water saturations might have increased over geo- mixing and biodegradation processes in the northern
logic time by vertical readjustment of the oil-water, Athabasca study area, is shown in Figure 27. The
perhaps before oil was completely immobilized by lower McMurray (LM1 and LM2 units in Figure 2)
biodegradation. was preserved only in structural and/or erosional lows
196 Fustic et al.

Figure 25. (A) Profile of 9MP/1MP ratios

posted on interpreted reservoir flow units;
(B) a schematic presentation of selected
chromatograms (m/z 192) representing
methylphenanthrene distribution in com-
partments M and N separated by flood-
plain deposits. Numbers 1 – 3 and 9 repre-
sent MP isomers. The M and N are two
vertically stacked reservoir compartments.
10 m (33 ft).

on unconformity and contains laterally extensive braided petroleum occupied the sediments of compartment N
river sand deposits capped by the laterally extensive first, perhaps progressing eastward and spilling under
flood-plain mudstone deposits. The floodplain mud- lateral compartment seals, such as mud plug deposits
stones are occasionally breached because of erosion. (Fustic et al., 2012b). Once biodegradation progressed
These gaps are providing pathways for communication to the point where the incoming petroleum could not
of fluids between the lower and the middle McMurray force the existing petroleum farther to the east, it en-
Formation (Figure 27C). The Clearwater Formation tered the lower McMurray through occasional gaps in
sediments provided the overlying seal at the time of the otherwise laterally continuous flood-plain mud-
reservoir charging (Figure 27A). Driven by buoyancy, stone deposits. This scenario implies charging the lower
petroleum entered the McMurray Formation. Because compartment with fresh oil that, at the same time, mixed
of the low reservoir temperature history, the McMurray with the petroleum of the middle McMurray For-
Formation waters contained abundant microbial life mation and occasionally refreshed upper parts of the
that immediately attacked incoming petroleum. The biodegraded columns within the middle McMurray
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 197

Figure 26. (A) Concentration (Mg/g oil)

profile of TAS and MAS posted on inter-
preted reservoir flow units (see legend);
(B) ratios of long- and short-chained TAS
and MAS. 10 m (33 ft).

Formation. Petroleum in the lower compartment moved Because of regional seals being eroded by Pleisto-
laterally under the floodplain mudstone that provided cene glaciations, gas caps have not been found in this
a relatively continuous barrier. Intensive biodegrada- study area. However, note that Oldenburg et al. (2009)
tion occurred along the water contact, resulting in the and Hubert et al. (2012) reported evidence of micro-
vertical compositional gradients in that unit too or the bial life from this field area, indicating that biogenic
contact was more persistent. Biodegradation was par- gas is still generated in the area, although at slow rates.
ticularly intensive in areas where the bottom water The slow rates of current gas formation are perhaps miti-
was thicker. Samples farther from the bottom water- gated by low reservoir temperatures and locally in-
bitumen contact are less degraded because of limited creased water salinity in the lower McMurray Formation,
exposure to microbes and perhaps slow mixing rates. and any gas that is presently produced quickly leaks
Once the petroleum charging had ceased, biodegra- because of the absence of extensive seals. Increased sa-
dation continued along water-bitumen contacts. Oil- linity is probably caused by salty water influx from the
water contacts may have also moved upward, pushing underlying Prairie Evaporite Formation (Figure 27C).
part of undegraded oil upward and leaving residual Additional evidence of gas generation in the area
heavy oil (bitumen) in situ, enhancing the thickness includes bubbling water that releases gases that ignite
of the low-bitumen, high-water saturation zone or tran- on contact with flames, a phenomenon that is com-
sition zone. monly observed in the open pit mine benches in oil
198 Fustic et al.
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 199

sand mine operators (personal experience, 2006), mi- optimization (Gates et al., 2008), and future in-situ pro-
crobial mats on outcrops along the Athabasca River, duction allocation studies (Bennett et al., 2010). This
and bubbling gas in summer along the river (F. Hein, study and Fustic et al. (2011) demonstrate that molecular
2010, personal communication). Considering the lack of composition can also be used for identifying reservoir
cap rocks in the northern study area and the deep in- compartments and mapping their geographic extent.
cision of the Athabasca River, one may infer that ini- Anomalies in well vertical compositional gradients
tially charged gas and early generated biogenic gas may be used for distinguishing between barriers and
would have to leak out, and that present-day gas leaks baffles. For example, subtle deviations in the compo-
are indicative of neoformation of gas. sitional gradients evident in samples 5 and 6 in well 6
Inferred former gas caps may have provided an addi- (Figures 23, 24) indicate a baffle that slows but does
tional driving mechanism for fluid movement within the not stop the mixing of fluids, inferring that steam ad-
reservoir. Gas cap expansion caused by changes in pres- vances will behave the same way. Greatly changed
sure in the reservoir because of a decrease in pressure compositions between samples 8 and 10, for example,
resulting from regional (basinwide) uplifting, erosion, in well 6 suggest that no communication existed between
and/or melting of overlying glaciers may have allowed the two reservoir units, inferring that the mudstone
pushing oil-gas contacts downward, inferring that some between them is laterally extensive. Thus, geochemical
oil was still mobile when these events occurred. data from the integrated baseline studies can be used
Finally, geochemical data from neither studied well as a powerful tool for distinguishing between barriers
show evidence of enhanced biodegradation at the top and baffles (Bennett et al., 2010; Fustic et al., 2011).
of the reservoir, despite being only a few meters below Finally, comparison of the molecular composition of
the glacial and Holocene sediments, inferring intensive produced fluids with baseline studies of compositional
percolation of surface waters in these areas and, thus, variations in well profiles may be used as a tool for pro-
diffusion of nutrients. This suggests that the last 10 to duction allocation studies and steam chamber growth as-
20 k.y. since glaciers removed the cap rock have not sessments (Bennett et al., 2010). For example, in cases
been sufficient for bitumen at the top to further de- when certain compounds are not identified in a lower
grade. The cold climate in the region during the last 10 part of the reservoir column (i.e., C4N in well 2, Figure 11)
to 20 k.y. likely contributed to the low biodegradation but start occurring in upper parts of the reservoir, the first
rates evident from the lack of the intensified biodeg- occurrence of that compound in produced fluids may
radation at the top of wells 6 and 7 (Figures 23– 26). be used, along with four-dimensional seismic data as an
Bottom-hole temperatures at the base of the reservoir indicator that the steam chamber has reached the upper
average 10 to 158C, whereas the optimal reservoir tem- reservoir (280 m [919 ft] in case of well 2; Figure 11). For
perature for biodegradation is about 408C. more sophisticated and precise assessment, a multivar-
iate statistic and chemometric study might be used.
This approach is the subject of an ongoing research.
IMPLICATIONS FOR RESERVOIR DEVELOPMENTS

The spatial distribution of the water, gas, and low- SUMMARY AND CONCLUSIONS
bitumen, high-water saturation zones is predictable
when interpreted in a geologic context that includes We have incorporated detailed case studies of fluid
understanding biodegradation processes. Results indi- property variations in a range of reservoir settings,
cate that integrating the data from a detailed baseline including the presence and absence of oil-water con-
geochemical study into the context of sedimentologic tacts; vertical and lateral compartmentalization; bot-
framework will ultimately provide a baseline reference tom- and top-water, top gas, low-bitumen, high-water
point that allows for predicting and mapping the varia- saturation zones, transition zones; and bitumens en-
tions in physical properties and chemical composition. countered in a range of depositional facies. Our in-
The use of this kind of information may be very beneficial tegrated geoscience approach used sedimentologic
for mine operations and for SAGD planning, production interpretation and petrophysical analysis to define

Figure 27. Summary diagram depicting major geologic features and events contributing to the present-day bitumen
accumulation, including reservoir charging, in-reservoir mixing, and biodegradation processes in northern Athabasca study area:
(A) early stages of reservoir charging; (B) later stages of reservoir charge (note suggested gas caps and merging oil-water
contacts between compartments); (C) present-day distribution of bitumen and water (including existing microbial life along the
present-day oil-bitumen contact identified in the study area by Oldenburg et al. (2009); Hubert et al. (2012). LB-HW = Low-
bitumen–high-water saturated zone; IHS = inclined heterolithic strata. 100 m (328 ft).
200 Fustic et al.

the depositional units and reservoir flow units and tumen in less porous and less permeable layers is less
qualitative and quantitative geochemistry to assess and biodegraded, probably a function of the tortuosity as-
compare the levels of biodegradation across the reser- sociated with restricted diffusion or mixing of fluids
voir. Finally, geochemical results were integrated and during reservoir mixing processes and, hence, limited
analyzed in the geologic context. Results show that the nutrient availability.
bitumen in the Athabasca oil sand deposit is very het- The third most important control are postaccumu-
erogeneous on vertical and lateral scales. Heterogene- lation changes that have affected the bitumen compo-
ity is a defining characteristic of oil sand reservoirs, sition. The main reason is occasional incursion of fresh
which is derived from or a function of the geologic water, perhaps with renewed nutrient supply that
framework, varying levels of fluid properties in terms of changed fluid distributions and possibly even brought
biodegradation, and multiple phases of charging and new microbial communities locally into the shallow
mixing, with changes in water chemistry (i.e., salinity) reservoirs (Figures 6, 27). The recent water incursions
introducing additional complexities. In the study area, that are probably related to glacial erosion (10 ka) do
biodegradation was controlled by an interplay of reser- not noticeably alert the level of the biodegradation in
voir facies framework, petroleum charging histories, our study area (Figure 27), whereas those documented
and postdepositional changes. These factors controlled in the southern Athabasca (Figure 6) case had more
the availability and distribution of nutrients in space significant impacts on the bitumen composition. This
and time and also perhaps biodegradation pathways. suggests that the fresh water encountered in them is
Commonly, in heavy-oil petroleum system studies, likely not related to (post) glacial events but to meteoric
including the Western Canada sedimentary basin, the water percolation over longer periods of geologic time
most important control on the degree of biodegrada- (Barson et al., 2001).
tion is reservoir temperature history (Adams et al., Evidence from both saturated and aromatic com-
2006). In the case of the Athabasca area, the tempera- pounds shows that biodegradation was the main pro-
ture range was low enough to be favorable for growth cess shaping the present-day composition of the bitumen.
of microbial communities throughout its history. Thus, However, as variations in the level of biodegradation of
biodegradation occurred and persisted for the entire aromatic compounds are more profound and variable,
Athabasca area and continues today (Oldenburg et al., these compounds should be the primary targets for
2009; Hubert et al., 2012). production allocation studies. Saturated and aromatic
Results from this study suggest that controls on the compound results suggest that biodegradation level
level of the biodegradation in studied areas may be ranges from 6 to 9 PM at the top and at the base of the
grouped into several process settings. The most im- reservoir, respectively. Biodegradation is not complete.
portant control factor on biodegradation level was the Present-day reservoir temperatures coupled with a dy-
presence or absence of an oil-water contact through namic hydrogeologic system are still favorable for
space and time. Areas with extensive present-day or microbial community growth. The origin of the low-
paleo-oil-water contacts contain more biodegraded bi- bitumen, high-water saturation zones and top water
tumen. When saline, instead of fresh or brackish, water intervals appears to result from the formation and the
was in contact with bitumen, biodegradation inten- subsequent depletion of gas caps. Because the oil is
sity may have been mitigated, but our examples do highly viscous, it cannot naturally move buoyantly into
not show it as a major factor (Figure 27C). This implies the pore spaces, whereas water is a much more mobile
that salinity here is not sufficient to offset biodeg- fluid and, thus, more readily imbibes into once oil (and
radation, or that salinity increased later in the reservoir gas)-saturated pore spaces. This is based on the as-
history when biodegradation level was already intense. sumption that most gas generation followed petro-
Vertical and lateral compartmentalization is the sec- leum entrapment. A conceptual diagram for the occur-
ond most important control on bitumen heterogeneity. rence of the gas and low-bitumen, high-water saturation
Results show that after charging is complete, each reser- zones in the McMurray Formation reservoirs is given in
voir compartment behaves as an independent bioreactor, Fustic et al. (2013) (Figure 3).
resulting in correlatable and mappable spatial distribu- Implications for reservoir developments are im-
tions of water; gas; low-bitumen, high-water saturation portant. We recommend that baseline studies integrat-
zones; and bitumen composition variation within each ing sedimentologic and geochemical characterization
compartment. However, adjacent compartments may will allow for predicting and mapping bitumen proper-
exhibit very different behaviors (Figures 6, 27). ties in the reservoir. Geochemical data from the inte-
Lithology and associated reservoir porosity and per- grated baseline studies can be used as a powerful tool for
meability properties also had an important function on production optimization and production allocation anal-
bitumen biodegradation (Figures 9–12). In general, bi- ysis, including assisting with distinguishing between
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 201

barriers and baffles (Bennett et al., 2010; Fustic et al., 2011). REFERENCES CITED
Finally, present-day biodegradation may be enhanced
by groundwater providing nutrients to microbes that Adams, J., C. L. Riediger, M. G. Fowler, and S. R. Larter,
then produce biogenic methane (Larter et al., 2008). 2006, Thermal controls on biodegradation around the Peace
River tar sands: Paleo-pasteurization to the west: Journal
of Geochemical Exploration, v. 89, p. 1 – 4, doi:10.1016
ACKNOWLEDGMENTS /j.gexplo.2005.11.004.
Adams, J., C. Jiang, B. Bennett, H. Huang, T. Oldenburg,
K. Noke, L. Snowdon, I. Gates, and S. Larter, 2008, Vis-
We thank Nexen Inc. and Shell-Albian for providing
cosity determination of heavy oil and bitumen: Cau-
sample material for this study. We thank the Alberta
tions and solutions: Proceedings for the World Heavy Oil
Ingenuity Centre for In Situ Energy and AAPG Grants- Congress, Edmonton, Alberta, Canada, Paper 2008-443,
in-Aid for support during the Ph.D. research project, 11 p.
titled Geological Controls on Reservoir and Bitumen Het- Adams, J., S. Larter, B. Bennett, H. Huang, J. Westrich, and
erogeneity in Athabasca Oil Sands Deposit, that this pa- C. van Kruisdijk, 2013, The dynamic interplay of oil mix-
per is a part of. We also thank Martin Fowler (Geological ing, charge timing, and biodegradation in forming the Al-
Survey of Canada, now Talisman Energy Inc.) for berta oil sands: Insights from geologic modeling and bio-
useful discussions. Fran Hein (Energy Resource and geochemistry, in F. J. Hein, D. Leckie, S. Larter, and J. R.
Conservation Board), Debra Higley-Feldman (U.S. Geo- Suter, eds., Heavy-oil and oil-sand petroleum systems in
logical Survey), and two anonymous reviewers are ac- Alberta and beyond: AAPG Studies in Geology 64, p. 23–
102.
knowledged for providing constructive comments that
Aherne, L. A., and B. Maini, 2008, Fluid movement in the
improved the final version of this manuscript. Mathew SAGD process: A review of the Dover project: Journal of
Smith (Nexen Inc.) is thanked for drafting most of the Canadian Petroleum Technology, v. 47, p. 31 – 37, doi:10
included figures. .2118/08-01-31.
Aitken, C. M., D. M. Jones, and S. R. Larter, 2004, Anaerobic
APPENDIX 1: CHEMICAL NAMES hydrocarbon biodegradation in deep subsurface reservoirs:
APPENDIX 1. Chemical names. Nature, v. 431, p. 291 – 295, doi:10.1038/nature02922.
Alberta Energy and Utilities Board, 2007, Alberta’s energy
reserves 2006 and supply/demand outlook 2007 – 2016,
TAS Triaromatic steroids
ST98-2007, 218 p.: http://www.eub.ca/docs/products/sts
MAS Monoaromatic steroids
/st98_current.pdf (accessed August 5, 2008).
TAS/(TAS + MAS) Thermal maturity parameters
Allan, J., and S. Creaney, 1991, Oil families of the Western
describing the molecular
Canada basin: Bulletin of Canadian Petroleum Geology,
concentration ratios between
v. 39, p. 107 – 122.
triaromatic steroids and
Barson, D., S. Bachu, and P. Esslinger, 2001, Flow systems in
monoaromatic steroids
the Mannville Group in the east-central Athabasca area
C3N Trimethylnaphthalene
and implications for steam-assisted gravity drainage
C4N Tetramethylnaphthalene
(SAGD) operations for in-situ bitumen production: Bulle-
C5N Pentamethylnaphthalene
tin of Canadian Petroleum Geology, v. 49, p. 376 – 392,
MP Methylphenanthrene
doi:10.2113/49.3.376.
C2P Dimethylphenanthrene
Bennett, B., and S. R. Larter, 2000, Quantitative separation of
C3P Trimethylphenanthrene
aliphatic and aromatic hydrocarbons using silver ion-
C4P Tetramethylphenanthrene
silica solid-phase extraction: Analytical Chemistry, v. 72,
1, 2, 3, and 9MP Methylphenanthrene isomers
p. 1039 – 1044, doi:10.1021/ac9910482.
(the same composition but
Bennett, B., M. Fustic, P. Farrimond, H. Huang, and S. R. Larter,
the methyl group attached
2006, 25-Norhopanes: Formation during biodegradation
at the different position on
of petroleum in the subsurface: Organic Geochemistry,
the molecule)
v. 37, p. 787– 797, doi:10.1016/j.orggeochem.2006.03.003.
9MP/1MP The molecular ratio based on
Bennett, B., M. Fustic, and S. Larter, 2010, Geochemical so-
the 9-methylphenanthrene and
lutions to production problems (abs.): AAPG Hedberg
1-methylphenanthrene has been
Research Conference, Vail, Colorado, June 7 – 10, 5 p.
applied to indicate relative
Brooks, P. W., M. G. Fowler, and R. W. Macqueen, 1988,
levels of the biodegradation
Biological marker and conventional organic geochemis-
TT Tricyclic terpanes
try of oil sands heavy oils, Western Canada Basin: Or-
PT Pentacyclic terpanes
ganic Geochemistry, v. 12, p. 519 – 538, doi:10.1016/0146
C29 AB hopane/
-6380(88)90144-1.
C30 AB hopane ratio
Butler, R., 1994, Steam-assisted gravity drainage: Concept,
202 Fustic et al.

development, performance and future: Journal of Cana- Handbook, 2d ed.: Calgary, Canada, Canadian Heavy Oil
dian Petroleum Technology, v. 33, p. 44 – 50. Association, p. 575 – 576.
Carrigy, M. A., 1959, Geology of the McMurray Formation: Fowler, M. G., and P. W. Brooks, 1987, Organic geochem-
Part III. General geology of the McMurray area: Alberta istry of Western Canada basin tar sands and heavy oils:
Research Council Memoir 1, 130 p. 2. Correlation of tar sands using hydrous pyrolysis of as-
Carrigy, M. A., 1971, Deltaic sedimentation in Athabasca tar phaltenes: Journal of Energy and Fuels, v. 1, p. 459 – 467,
sands: AAPG Bulletin, v. 55, no. 8, p. 1155 – 1169. doi:10.1021/ef00006a002.
Chan, M. Y., F. J. Altebaumer, W. P. Brooks, and G. G. Fowler, M. G., and C. Riediger, 2000, Origin of the Athabasca
Sarioglu, 1992, Application of biomarkers to tracing the tar sands, in D. Barson, R. Bartlett, F. Hein, M. Fowler,
source of oil-producing stratum in oil sands reservoirs: S. Grasby, and C. Riediger, eds., Hydrogeology of heavy
Society of Petroleum Engineers Annual Technical Confer- oil and tar sand deposits: Water flow and supply, migra-
ence and Exhibition, Washington, D.C., SPE Paper 24815: tion and degradation: GeoCanada 2000 Field Trip Guide-
http://www.onepetro.org/mslib/servlet/onepetropreview book 14, Stop 14, p. 117 – 127.
?id=00024815&soc=SPE (accessed August 5, 2009). Fowler, M. G., C. L. Riediger, L. D. Stasiuk, and M. Obermayer,
Creaney, S., and J. Allan, 1990, Hydrocarbon generation and 2004, Reassessing the origin of the heavy oils and bitumens
migration in the Western Canada sedimentary basin: Clas- of the Western Canada Basin, the world’s largest biode-
sic petroleum provinces: Geological Society (London) Spe- graded hydrocarbon deposit, CD-ROM.
cial Publication 50, p. 189 – 202. Fustic, M., L. Skulski, D. Vanhooren, P. Bessette, D. Hinks,
Creaney, S., and J. Allan, 1992, Petroleum systems in the L. Bellman, and D. Leckie, 2008, Data integration im-
foreland basin of western Canada, in R. W. Macqueen and proves geological mapping and reservoir characterization
D. A. Leckie, eds., Foreland basins and fold belts: AAPG of oil sands reservoir: World Oil Magazine, v. 229, no. 10,
Memoir 55, p. 279 – 308. p. 117 – 122.
Creaney, S., J. Allan, K. S. Cole, M. G. Fowler, P. W. Brooks, Fustic, M., B. Bennett, J. J. Adams, H. Huang, B. MacFarlane,
K. G. Osadetz, R. W. Macqueen, L. R. Snowdon, and C. L. D. Leckie, and S. R. Larter, 2011, Bitumen geochemistry:
Riediger, 1994, Petroleum generation and migration in the A tool for distinguishing barriers from baffles in oil sands
Western Canada sedimentary basin, in G. Mossop and I. reservoirs: Bulletin of Canadian Petroleum Geology, v. 59,
Shetsen, compilers: Geological atlas of the Western Can- p. 295–316.
ada sedimentary basin, p. 455 – 470. Fustic, M., S. M. Hubbard, R. S. Spencer, G. D. Smith, D. A.
Dean, E. W., and D. D. Stark, 1920, A convenient method for Leckie, B. Bennett, and S. Larter, 2012a, Recognition of down-
the determination of water in petroleum and other or- valley translation in tidally influenced meandering fluvial
ganic emulsions: Industrial and Engineering Chemistry, deposits, Athabasca oil sands (Cretaceous), Alberta, Canada:
v. 12, p. 486 – 490, doi:10.1021/ie50125a025. Journal of Marine and Petroleum Geology, v. 29, p. 219–232.
Deroo, G., T. G. Powell, R. G. McGrossan, and E. Der, 1974, Fustic, M., B. Bennett, H. Huang, and S. Larter, 2012b,
Geochemistry of the heavy oils in Alberta, in L. V. Hills, Differential entrapment of charged oil: New insights on
ed., Oil sands fuel of the future: Canadian Society of Pe- McMurray Formation oil trapping mechanisms: Journal
troleum Geologists Memoir 3, p. 148 – 167. of Marine and Petroleum Geology, v. 36, p. 50 – 69.
Deroo, G., T. G. Powell, B. Tissot, and R. G. McGrossan, Fustic, M., D. Cadiou, D. Thurston, A. Al-Dliwe, and D. A.
1977, The origin and migration of petroleum in the west- Leckie, 2013, Reservoir modeling by constraining stochas-
ern Canadian sedimentary basin, Alberta: A geochemical tic simulation to deterministically interpreted three-
and thermal maturation study: Geological Survey of Can- dimensional geobodies: Case study from Lower Cretaceous
ada Bulletin, v. 262, 136 p. McMurray Formation, Long Lake steam-assisted gravity
England, W. A., 1994, Secondary migration and accumula- drainage project, northeast Alberta, Canada, in F. J. Hein,
tion of hydrocarbons, in L. B. Magoon and W. G. Dow, D. Leckie, S. Larter, and J. R. Suter, eds., Heavy-oil and oil-
eds., The petroleum system: From source to trap: AAPG sand petroleum systems in Alberta and beyond: AAPG
Memoir 60, p. 211 – 217. Studies in Geology 64, p. 565 – 604.
England, W. A., A. S. MacKenzie, D. M. Mann, and T. M. Gates, I. D., J. J. Adams, S. R. Larter, and I. D. Gates, 2008,
Quigley, 1987, The movement and entrapment of petro- The impact of oil viscosity heterogeneity on the produc-
leum fluids in the subsurface: Journal of Geological So- tion characteristics of tar sand and heavy oil reservoirs:
ciety (London), v. 144, p. 327 – 347. Part II. Intelligent, geotailored recovery processes in com-
Flach, P. D., 1984, Oil sands geology: Athabasca deposit positionally graded reservoirs: Journal Canadian Petro-
north: Alberta Research Council Bulletin, v. 46, 31 p. leum Technology, v. 47, no. 9, p. 40– 49.
Flach, P. D., and G. D. Mossop, 1985. Depositional environ- Gussow, W. C., 1954, Differential entrapment of oil and gas:
ments of Lower Cretaceous McMurray Formation, Atha- A fundamental principle: AAPG Bulletin, v. 38, p. 816 –
basca oil sands, Alberta: AAPG Bulletin, v. 69, p. 1195 – 853.
1207. Haq, B. U., J. Hardenbol, and P. R. Vail, 1988, Mesozoic and
Flint, L., R. Kerr, H. Tran, M. Singleton, and T. de Bruijn, Cenozoic chronostratigraphy and cycles of sea level
2005, Bitumen/heavy oil upgrading: The synopsis, in C. change, in C. K. Wilgus, H. Posamentier, B. S. Hastings, J.
Palmgren, ed., Canadian Heavy Oil Association (CHOA) Van Wagoner, C. A. Ross, and C. G. St. C. Kendall, eds.,
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 203

Sea-level changes: An integrated approach: SEPM Special Kaufman, R. L., H. Dashti, C. S. Kabir, J. M. Pederson, M. S.
Publication 42, p. 71– 108. Moon, R. Quttainah, and H. Al-Wael, 1997, Characteriz-
Haszeldine, S. R., A. J. Cavanagh, and G. L. England, 2003, ing the greater Burgan field: Use of geochemistry and oil
Effects of oil charge on illite dates and stopping quartz fingerprinting: Society of Petroleum Engineers Proceed-
cement: Calibration of basin models: Journal of Geo- ings of Middle East Oil Show, SPE Paper 37803, p. 385 –
chemical Exploration, v. 78 – 79, p. 373 – 376, doi:10.1016 394.
/S0375-6742(03)00151-1. Larter, S. R., and A. Aplin, 1995, Reservoir geochemistry:
Head, I., D. M. Jones, and S. R. Larter, 2003, Biological activity Methods, applications and opportunities, in J. M. Cubitt
in the deep subsurface and the origin of heavy oil: Nature, and W. A. England, eds., The geochemistry of reser-
v. 426, p. 344 – 352, doi:10.1038/nature02134. voirs: Geological Society (London) Special Publication 86,
Higley, K. D., D. M. Lewan, N. R. L. Roberts, and M. Henry, p. 5 – 32.
2009, Timing and petroleum sources for the Lower Cre- Larter, S. R., A. Wilhelms, I. Head, M. Koopmans, A. Aplin,
taceous Mannville Group oil sands of northern Alberta R. di Primio, C. Zwach, M. Erdmann, and N. Telnaes, 2003,
based on 4-D modeling: AAPG Bulletin, v. 93, no. 2, The controls on the composition of biodegraded oils in the
p. 203 – 230, doi:10.1306/09150808060. deep subsurface: Part 1. Biodegradation rates in petro-
Huang, H., S. Larter, B. F. J. Bowler, and T. Oldenburg, 2004, leum reservoirs: Organic Geochemistry, v. 34, p. 601–613.
A dynamic biodegradation model suggested by petroleum Larter, S. R., I. M. Head, H. Huang, B. Bennett, M. Jones,
compositional gradients within reservoir columns from A. C. Aplin, A. Murray, M. Erdmann, A. Wilhelms, and
the Liaohe Basin, NE China: Organic Geochemistry, v. 35, R. di Primio, 2005, Biodegradation, gas destruction and
p. 299 – 316, doi:10.1016/j.orggeochem.2003.11.003. methane generation in deep subsurface petroleum reser-
Hubbard, S. M., D. G. Smith, H. A. Nielsen, D. A. Leckie, M. voirs: An overview, in A. G. Dore and B. Vining, eds., Pe-
Fustic, R. J. Spencer, and L. Bloom, 2011, Seismic geomor- troleum geology: Northwest Europe and global perspec-
phology and sedimentology of a tidally influenced river tives: 6th Petroleum Geology Conference, Geological Society
deposit, Lower Cretaceous Athabasca oil sands, Alberta, (London), p. 633–640.
Canada: AAPG Bulletin, v. 95, p. 1123 – 1145, doi:10.1306 Larter, S. R., H. Huang, J. Adams, B. Bennett, O. Jokanola, T.
/12131010111. Oldenburg, M. Jones, I. Head, C. Riediger, and M. Fowler,
Hubert, C. R. J., et al., 2012, Massive dominance of Epsilon- 2006, The controls on the composition of biodegraded oils
proteobacteria in formation waters from a Canadian oil in the deep subsurface: Part II. Geological controls on sub-
sands reservoir containing severely biodegraded oil: En- surface biodegradation fluxes and constraints on reservoir-
vironmental Microbiology, v. 14, p. 387– 404. fluid property prediction: AAPG Bulletin, v. 90, no. 6,
Hwang, R. J., and D. K. Baskin, 1994, Reservoir connectivity p. 921 – 938, doi:10.1306/01270605130.
and oil homogeneity in a large-scale reservoir, in M. I. Larter, S. R., J. Adams, I. D. Gates, B. Bennett, and H. Huang,
Al-Husseini, ed., Middle East Petroleum Geoscience 2008, The origin, prediction and impact of oil viscosity
Geo ’94, v. 2, p. 529 – 541. heterogeneity on the production characteristics of tar sand
Jokanola, A. O., 2007, Pore pressure, seal risking and petro- and heavy oil reservoirs: Journal of Canadian Petroleum
leum column height predictions in leaky petroleum sys- Technology, v. 47, p. 52 – 61.
tems: Some case studies: Ph.D. thesis, University of Cal- Leckie, A. D., and G. D. Smith, 1992, Regional setting, evolu-
gary, Calgary, Alberta, Canada, 463 p. tion, and depositional cycles of the Western Canada fore-
Jokanola, A. O., A. C. Aplin, S. R. Larter, and K. Kurtev, land basin, in A. D. Leckie and G. D. Smith, eds., Foreland
2009, Petroleum leakage in the Snorre oil field, North Sea: basins and fold belts: AAPG Memoir 55, p. 5 – 46.
A case for change in cap-rock wettability and dynamic Marchand, A. M. E., S. R. Haszeldine, I. C. Macaulay, R.
seal risking (abs.): 2nd International Conference on Fault Swennen, and E. A. Fallick, 2000, Quartz cementation
and Top Seals: From Pore to Basin Scale Montpellier, inhibited by crestal oil charge: Miller deepwater sand-
France, September 21 – 24, 2009, European Association of stone, UK North Sea: Clay Miner, v. 35, p. 205 – 214.
Geoscientists and Engineer, p. 3. Marchand, A. M. E., S. R. Haszeldine, C. P. Smalley, I. C.
Jones, D. M., et al., 2008, Crude-oil biodegradation via meth- Macaulay, and A. E. Fallick, 2001, Evidence for reduced
anogenesis in subsurface petroleum reservoirs: Nature, quartz cementation rates in oil-filled sandstones: Geol-
v. 451, p. 176 – 180. ogy, v. 29, p. 915–918, doi:10.1130/0091-7613(2001)029<0915
Kaminsky, V. P., 1974, Selection of a mining scheme for a tar :EFRQCR>2.0.CO;2.
sands extraction plant, in L. V. Hills, ed., Oil sands fuel Marchand, A. M. E., C. P. Smalley, S. R. Haszeldine, and
of the future: Canadian Society of Petroleum Geologists E. A. Fallick, 2002, The importance of hydrocarbon-fill for
Memoir 3, p. 212 – 222. reservoir quality prediction: AAPG Bulletin, v. 86, p. 1561–
Kaufman, R. L., A. S. Ahmed, and R. J. Elsinger, 1990, Gas 1571.
chromatography as a development and production tool McCaffrey, M. A., A. H. Legarre, and J. S. Johnson, 1996,
for fingerprinting oils from individual reservoirs: Appli- Using biomarkers to improve heavy oil reservoir man-
cations in the Gulf of Mexico, in D. Schumaker and B. F. agement: An example from the Cymric field, Kern Coun-
Perkins, eds., 9th Annual Research Conference of the ty, California: AAPG Bulletin, v. 80, p. 898 – 913.
SEPM, p. 263 – 282. McKinney, D., and K. Bland, 2003, Time lapse geochemistry
204 Fustic et al.

at the Ursa field: A surveillance tool for understanding Peters, K. E., C. C. Walters, and J. M. Moldowan, 2005, The
complex, heterogeneous reservoir fluids (abs.): 21st In- biomarker guide: New York, Cambridge University Press,
ternational Meeting on Organic Geochemistry, Krakow, v. 2, 1155 p.
Poland, p. I/96. Price, R. A., 1973, Large-scale gravitational flow of supra-
McPhee, D. A., and D. M. Wightman, 1991, Timing of the crustal rocks, southern Canadian Rockies, in K. A. De Jong
dissolution of Middle Devonian Elk Point Group evapo- and R. A. Scholten, eds., Gravity and tectonics: New York,
rates: Twps. 47 to 103 and Rges. 15 W3M to 20 W4M Willey-Interscience, p. 491–502.
(abstract): Bulletin of Canadian Petroleum Geology, v. 39, Ranger, M., 1994, Basin study of southern Athabasca oil sands:
p. 218. Ph.D. thesis, University of Alberta, Edmonton, Alberta,
Mellon, G. B., and J. H. Wall, 1956, Foraminifera of the Canada, 290 p.
upper McMurray and basal Clearwater Formation: Part Ranger, M. J., and M. K. Gingras, 2006, Geology of the Atha-
1. Research Council of Alberta Report 72, p. 5 – 29. basca oil sands: Field guide and overview: Edmonton,
Milkov, V. A., E. Goebel, L. Dzou, A. Fisher, A. Kutch, N. Alberta, Canada, Department of Earth and Atmospheric
McCaslin, and F. D. Bergman, 2007, Compartmentaliza- Science, University of Alberta, 119 p.
tion and time-lapse geochemical reservoir surveillance of Riediger, C., S. Ness, M. G. Fowler, and T. Akpulat, 2000,
the Horn Mountain oil field, deep-water Gulf of Mexico: Timing of oil migration, Paleozoic and Cretaceous bitu-
AAPG Bulletin, v. 91, p. 847–876, doi:10.1306/01090706091. men and heavy oil deposits, eastern Alberta (abs.): Geo-
Moshier, S. O., and D. W. Waples, 1985, Quantitative evalua- Canada 2000 Conference, Calgary, Alberta, Canada,
tion of Lower Cretaceous Mannville Group as source rock p. 419.
of Alberta’s oil sands: AAPG Bulletin, v. 69, p. 161 – 172. Riediger, C. L., S. Ness, M. Fowler, and N. T. Akpulat, 2001,
Mossop, G. D., 1980, Geology of the Athabasca oil sands: Timing of oil generation and migration, northeastern British
Science, v. 207, p. 145 – 152, doi:10.1126/science.207.4427 Columbia and southern Alberta: Significance for under-
.145. standing the development of the eastern Alberta tar sands
Nederlof, P. J., M. A. Gijsen, and M. A. Doyle, 1994, Ap- deposits (abs.): AAPG Annual Convention and Exhibition,
plication of reservoir geochemistry to field appraisal, in v. 10, p. A168.
M. I. Al-Husseini, ed., The Middle East Petroleum Geo- Sahagian, D., O. Pinous, A. Olferiev, and V. Zakharov, 1996,
sciences Geo ’94: Bahrain, Gulf-Petrolink, v. 2, p. 709 – 722. Eustatic curve for the Middle Jurassic – Cretaceous based
Nederlof, P. J. R., F. M. van der Veen, and G. A. van den Bos, on Russian platform and Siberian stratigraphy: Zonal res-
1995, Application of reservoir geochemistry in Oman, in olution: AAPG Bulletin, v. 80, p. 1433 – 1458.
J. O. Grimalt and C. Dorronsoro, eds., Organic geochem- Selby, D., and R. A. Creaser, 2005, Direct radiometric dating
istry: Developments and applications to energy, climate, of hydrocarbon deposits using rhenium-osmium isotopes:
environment and human history: p. 329 – 331. Science, v. 308, p. 1293 –1295, doi:10.1126/science.1111081.
Nurkowski, J. R., 1985, Coal quality and rank variation within Seyer, F. A., and G. W. Gyte, 1989, Viscosity, in L. E. Hepler
the Upper Cretaceous and Tertiary sediments, Alberta and C. His, eds., AOSTRA technical handbook on oil sands,
plains region: Alberta Research Council Earth Science bitumens and heavy oils: Alberta Oil Sands Technology and
Report 85-1, 39 p. Research Authority Technical Publication Series 6, 376 p.
O’Donnell, D. N., and J. M. Jodrey, 1985, Geology of the Smalley, P. C., and W. A. England, 1992, Assessing reservoir
Syncrude mine site and its application to sampling and compartmentalization during field appraisal: How geo-
grade control, in R. A. Metz, ed., Applied mining geol- chemistry can help: European Petroleum Conference, SPE
ogy: Problems of sampling and grade control: New York, Paper 25005, p. 423 – 431.
Society of Mining Engineers of the American Institute of Smalley, P. C., and W. A. England, 1994, Reservoir com-
Mining, Metallurgical, and Petroleum Engineers, p. 93 – partmentalization assessed with fluid compositional data:
102. Society of Petroleum Engineers Reservoir Engineering,
Oldenburg, T. B. P., S. R. Larter, J. J. Adams, A. K. Rowan, A. v. 9, p. 175 – 180.
Brown, I. M. Head, A. A. Grigoriyan, G. Voordouw, M. Smith, D. G., 1988, Modern point bar deposits analogous to
Fustic, and C. Hubert, 2009, Methods for recovery of the Athabasca oil sands, Alberta, Canada, in P. L. de Boer,
microorganisms and intact microbial polar lipids (IPLs) A. van Gelder, and S. D. Nio, eds., Tide-influenced sedi-
from oil-water mixtures: Lab experiments and natural mentary environments and facies: Dordrecht, Holland,
well-head fluids: Analytical Chemistry, v. 81, p. 4130 – Reidel Publishing, p. 417– 432.
4136, doi:10.1021/ac8025515. Smith, D. G., S. D. Hubbard, D. Leckie, and M. Fustic, 2009,
Paul, A. H., and M. Fustic, 2002, Geologic investigation of Counter point bars in modern meandering rivers: Recog-
the Muskeg River mine, Ft. McMurray, Alberta: Society nition of morphology, lithofacies and reservoir signifi-
of Manufacturing Engineers Annual Meeting Preprint cance, examples from Peace River, AB, Canada: Sedimen-
02-135. tology, v. 56, p. 1655 – 1669, doi:10.1111/j.1365-3091.2009
Peters, K. E., and J. M. Moldowan, 1993, The biomarker guide: .01050.x.
Interpreting molecular fossils in petroleum and ancient Stainforth, J. G., 2004, New insights into reservoir filing and
sediments: Englewood Cliffs, New Jersey, Prentice Hall, mixing processes, in J. M. Cubitt, W. A. England, and S. R.
363 p. Larter, eds., Understanding petroleum reservoirs: Toward
 Variability of Bitumen Properties in the Athabasca Oil Sands, Canada 205

an integrated reservoir engineering: Geological Society phase and compartmentalization: Magnolia field case
(London) Special Publication 168, p. 115 – 132. study, deepwater Gulf of Mexico, U.S.A., in J. M. Cubitt,
Stockmal, G. S., D. J. Cant, and J. S. Bell, 1992, Relation- W. A. England, and S. R. Larter, eds., Understanding
ship of the stratigraphy of the Western Canada foreland petroleum reservoirs: Toward an integrated reservoir en-
basin and cordilleran tectonics: Insights from geodynamic gineering: Geological Society (London) Special Publica-
models, in R. A. MacQueen and D. A. Leckie, eds., Foreland tion 237, p. 231 – 256.
basin and fold belts: AAPG Memoir 55, p. 107 – 124. Wightman, D. M., and S. G. Pemberton, 1997, The Lower Cre-
Stott, D. F., 1984, Cretaceous sequences of the Foothills of taceous (Aptian) McMurray Formation: An overview of the
the Canadian Rocky Mountains, in D. F. Stott and D. J. McMurray area, northeastern Alberta, in S. G. Pemberton
Glass, eds., The Mesozoic of middle North America: Can- and D. P. James, eds., Petroleum geology of the Cretaceous
adian Society of Petroleum Geologists Memoir 9, p. 85 – lower Mannville Group: western Canada: Canadian So-
107. ciety of Petroleum Geologists Memoir 18, p. 312 – 344.
Strausz, O. P., and E. M. Lown, 2003, The chemistry of Alberta Wightman, D. M., B. Rottenfusser, J. Kramers, and R. Harrison,
oil sands, bitumens and heavy oils: Calgary, Alberta, 1989, Geology of the Alberta oil sands deposits, in L. G.
Canada, Alberta Energy Research Institute, 695 p. Hepler and C. His, eds., AOSTRA technical handbook on
Strobl, R. S., W. K. Muwais, D. M. Wightman, D. K. Cotterill, oil sands, bitumens and heavy oils: Alberta Oil Sands
and L.-P. Yuan, 1997, Geological modeling of McMurray Technology and Research Authority Technical Publication
Formation reservoirs based on outcrop and subsurface Series 6, p. 1 – 9.
analogues, in S. G. Pemberton and D. P. James, eds., Pe- Wightman, D. M., J. R. MacGillivary, D. McPhee, H. Berhane,
troleum of the Cretaceous Mannville Group, western Can- and T. Berezniuk, 1991, McMurray Formation and Wabis-
ada: Canadian Society of Petroleum Geologists Memoir 18, kaw Member (Clearwater Formation): Regional perspec-
p. 292–311. tives derived from the North Primrose area, Alberta,
Thomas, R. G., D. G. Smith, J. M. Wood, J. Visser, E. A. Canada, in R. F. Meyer, ed., 5th UNITAR international
Calverley-Range, and E. H. Koster, 1987, Inclined hetero- conference on heavy crude and tar sands: UNITAR Centre
lithic stratification-terminology, description, interpretation for Heavy Crude and Tar Sands, v. 1, p. 285 – 320.
and significance: Sedimentary Geology, v. 53, no. 1 – 2, Wightman, D. M., M. N. Attala, D. A. Wynne, R. S. Strobl, H.
p. 123 – 179. Berhane, D. K. Cotterill, and T. Berezniuk, 1995, Resource
Tissot, B., and D. H. Welte, 1984, Petroleum formation and characterization of the McMurray/Wabiskaw deposit in
occurrence, 2d ed.: New York, Spring-Verlag, 699 p. the Athabasca oil sands area: A synthesis: Alberta Oil
Tozer, R., A. Choi, J. Pietras, and D. Tanasichuk, 2012, Atha- Sands Technology and Research Authority Technical Pub-
basca oil sands: Mega-trap geometry and timing, (abs.): lications Series 10, 220 p.
Annual Joint Convention of Canadian Society of Petroleum Wright, G. N., M. E. McMechan, D. E. G. Potter, and M. E.
Geologists, Canadian Society of Exploration Geophysists Holter, 1994, Structure and architecture of the Western
and Canadian Well Logging Society, Calgary, Canada, Canada sedimentary basin, in G. D. Mossop and I. Shetsen
May 14 – 18, http://www.geoconvention.com/uploads (compilers), Geological atlas of the Western Canada sed-
/2012abstracts/168_GC2012_Athabasca_Oil_Sands.pdf imentary basin: Canadian Society of Petroleum Geolo-
(accessed May 1, 2012). gists and Alberta Research Council Special Report 4: http
Vigrass, L. W., 1968, Geology of Canadian heavy oil sands: ://www.ags.gov.ab.ca/publications/wcsb_atlas/atlas.html
AAPG Bulletin, v. 52, p. 1984 – 1999. (accessed August 5, 2008).
Walderhaug, O., 1996, Kinetic modeling of quartz cementa- Zengler, K., H. H. Richnow, R. Rossello-Mora, W. Michaelis,
tion and porosity loss in deeply buried sandstone reser- and F. Widdel, 1999, Methane formation from long-chain
voirs: AAPG Bulletin, v. 80, p. 731 – 745. alkanes by anaerobic microorganisms: Nature, v. 401,
Weissenburger, K. S., and T. Borbas, 2004, Fluid properties, p. 266 – 269.
View publication stats