Gigantic, gaseous mushwads in Cambrian shale: Conasauga Formation, southern Appalachians, USA

International Journal of Coal Geology 103 (2012) 70–91

Contents lists available at SciVerse ScienceDirect

International Journal of Coal Geology journal homepage: www.elsevier.com/locate/ijcoalgeo

Gigantic, gaseous mushwads in Cambrian shale: Conasauga Formation, southern Appalachians, USA Jack C. Pashin ⁎, David C. Kopaska-Merkel, Ann C. Arnold, Marcella R. McIntyre, William A. Thomas Geological Survey of Alabama, P.O. Box 869999, Tuscaloosa, AL 35486‐6999, USA

a r t i c l e

i n f o

Article history: Received 15 December 2011 Received in revised form 26 April 2012 Accepted 12 May 2012 Available online 9 July 2012 Keywords: Conasauga Shale gas Alabama Appalachian Thrust belt

a b s t r a c t Giant duplexes called mushwads in the southern Appalachian thrust belt are estimated to contain a natural gas resource base of about 17.7 Tcm (625 Tcf). Early development efforts in mushwads of the Cambrian-age Conasauga Formation encountered significant challenges and highlight the need for an integrated, multidisciplinary approach to shale gas exploration. To facilitate development of this gigantic resource, an integrated geological analysis of stratigraphy, sedimentation, structural geology, basin hydrodynamics, petrology, geochemistry, gas storage and mobility, and reservoir volumetrics was conducted. The results of this analysis demonstrate the myriad geological factors that need to be considered when developing shale gas resources and suggest that the greatest reservoir potential lies deep in the mushwads, where free gas is concentrated. Meeting the challenges posed by deformed shale masses in thrust belts is a major frontier for hydrocarbon exploration that could result in a major expansion of natural gas reserves. © 2012 Elsevier B.V. All rights reserved.

1. Introduction The Cambrian-age Conasauga Formation contains significant natural gas resources in the Appalachian thrust belt of Alabama, where exploration is focused on shale formations of Cambrian, Devonian, and Mississippian age (Fig. 1). Unconventional gas plays require an integrated, multidisciplinary approach to exploration and development (e.g., Hill and Jarvie, 2007; Ross and Bustin, 2008). Few broadly applicable geologic models of resource distribution and producibility have been developed for shale gas reservoirs, however, including the structurally complex reservoirs that constitute major exploration targets in foreland thrust belts (Pashin, 2008, 2009; Pashin et al., 2010a, 2010b). Development of Conasauga gas resources poses a broad range of technical challenges, and many of these challenges relate to incomplete characterization of the regional geologic framework, as well as an inadequate understanding of the fundamental geologic controls governing the producibility of shale gas. Integrated geologic models are needed for emerging shale plays like the Conasauga that take into account a range of geologic variables and reservoir conditions (Fig. 2). Indeed, key aspects of reservoir quality, such as composition, thickness, and continuity, are determined in the original depositional environment. Reservoir quality and geometry evolve in response to structural deformation, the hydrodynamic and geothermal environment of the host sedimentary basin, and the petrologic and

⁎ Corresponding author. E-mail address: [email protected] (J.C. Pashin). 0166-5162/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.coal.2012.05.010

geochemical changes that occur during burial and unroofing of the host basin. The composition and pore structure of shale, moreover, determine how hydrocarbons are stored and how these compounds may be mobilized. After a geologic framework is defined and the fundamental controls on gas storage and mobility are understood, then effective strategies for development can be formulated, and resources and reserves assessed. In this paper, we use the basic approach outlined above to provide a systematic evaluation of the shale gas potential of the Conasauga Formation (Fig. 2). The paper begins with an overview of the regional geologic setting of the shale gas plays and continues with a review of the analytical methods employed in this study. A systematic evaluation of stratigraphy, sedimentology, geologic structure, hydrodynamics, geothermics, petrology, geochemistry, gas storage, and permeability of the shale forms the heart of this paper. The paper concludes with an assessment of gas resources and production potential, as well as some insights on shale gas development in thrust belts. 2. Regional setting 2.1. Shale gas development Natural gas has been produced commercially from organic-rich shale of the Appalachian region of the eastern United States for well over a century; the first recorded production was in New York in 1821 (Martin, 2009) and in Kentucky between 1863 and 1865 (Nuttall et al., 2009). Recent developments in hydraulic fracturing technology have facilitated high gas production rates from shale and have had a strong impact on U.S. gas supply and markets (Hill and

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

71

EXPLANATION

Black Warrior Basin

Neal (Floyd) Shale (Mississippian)

st

ian

lt Be

u hr

T

Devonian Shale Conasauga Shale (Cambrian) Shale gas development

ch

la pa

Ap

Index Map

0 0

50 mi 50 km

Fig. 1. Generalized map showing location of shale-gas development areas in Alabama.

Jarvie, 2007). Significant shows of natural gas were first documented in the Conasauga Formation of Alabama in the J. J. Young 34–2 #1 well, which was spudded by Amoco Production Company in St. Clair County in November 1984. The well was spudded in Cambrian–Ordovician carbonate rocks; it penetrated more than 2700 m (9000 ft) of shaly Conasauga strata and reached a total depth of 3022 m (9915 ft) (Raymond, 1991). At that depth, drill pipe twisted and was stuck in the well, which was subsequently abandoned in February 1985 after an unsuccessful fishing attempt. In 2005, Dominion Exploration and Production, Incorporated, began developing the natural gas in the area of the J. J. Young well. The Big Canoe Creek Field (Figs. 3, 4) was established in February 2007, and the Dawson 34-03-01 well produced gas from Conasauga shale and was therefore designated the discovery well. This event marked the first commercial gas production from shale in Alabama. The Conasauga further has the distinction of being geologically the oldest and most structurally complex shale from which production of natural gas has been established. The field is characterized by large volumes of gas-in-place, as well as locally high reservoir pressure (Pashin et al., 2011; Williams, 2007). To date, the field has produced only 5297×103 scm (187,047×103 scf) of gas, underscoring the technical challenges that will be discussed in this paper. Shale gas potential extends far beyond the confines of the Big Canoe Creek Field (e.g., Cook and Thomas, 2010; Pashin, 2008), and so regional exploration is in its infancy. Energen Resources and Chesapeake Energy

Stratigraphy, Sedimentology Resources, Reserves

Structural Geology Shale gas Producibility

Gas Storage, Mobility Petrology, Geochemistry

Hydrodynamics, Geothermics

Fig. 2. Conceptual model of shale gas producibility based on key geologic variables.

drilled an exploratory well (Marchant 22–16 #1) to a total depth of 3781 m (12,406 ft) near the southwest end of the Conasauga prospect area (Fig. 3). The well penetrated more than 3230 m (10,600 ft) of Conasauga strata and encountered several prominent gas shows between depths of 2103 and 3572 m (6900-11,720 ft). Exploration activity is also taking place northeast of Big Canoe Creek Field in Alabama and northwest Georgia. 2.2. Geologic setting In the thrust belt of Alabama, many exposures of the Conasauga Formation contain fossils of Late Cambrian age. In some thrust sheets in the southeastern part of the thrust belt, where pre-Conasauga strata are exposed, Middle Cambrian fossils have been identified in the Conasauga (Butts, 1910, 1926, 1927; Kidd and Neathery, 1976). The Conasauga is composed of interbedded shale, limestone, and dolostone and was deposited in a spectrum of platform, shoal, ramp, and intrashelf basin environments (Fig. 5) (Astini et al., 2000; Pashin, 2008; Thomas et al., 2000). The Conasauga contains a spectrum of carbonate rock types, including peloidal, oolitic, intraclastic, and stromatolitic limestone and dolostone. The productive shale facies contains abundant nodules and thin beds of limestone and was deposited primarily in ramp and intrashelf basin environments. Thick Conasauga shale was deposited in the Birmingham graben, which was initiated during late Proterozoic–Early Cambrian Iapetan rifting. Original shale thickness within the graben probably was more than 2000 m (6500 ft) (Thomas, 2007a). Facies and thickness changes in Cambrian–Ordovician carbonate rocks indicate that a structural inversion of the Birmingham graben occurred during the Ordovician-age Taconian orogeny (e.g., Butts, 1926; Thomas, 2001, 2007a). The graben was reactivated during the Mississippian and Pennsylvanian. Taconian uplift and subsequent reactivation in the northwestern part of the graben may have helped make Conasauga shale prone to deformation associated with detachment tectonics during later orogenic events. Indeed, the shale was deformed into giant antiformal stacks called mushwads (i.e., malleable,

72

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

.

10

0

tn

20 km

d an

M

W

rg

Co

u

M

o Lo

e

m Ro

Shale Gas Field

Warrior

Ca

ha

ba

Be

ss

Va

em

nd

iv

er

er B irm

ing

ha

m

Basin

Pa

lm er

Black

Ca ha ba

da le

q Se

t ou

y lle a V . en d n s t s e ad me re t M h G n p Ro ur lou M B Big Canoe Creek

al bu

e

hi

c at

.

tn

k

S

10 mi

ills

10

0

Va lle y

10

sa

EXPLANATION

C

oo

Stratigraphy Mesozoic- green, yellow Cenozoic Mississippian- purple, blue Pennsylvanian Cambrian- - pink, orange Devonian

Conasauga Formation (Cambrian)

Index Map

Structures

Coosa - Major structure Vandsden - Mushwad

Marchant 22-16 #1 well Fig. 3. Geologic map of the frontal Appalachian structures showing areas of shale gas development (after Pashin, 2008, 2009). Geology from Osborne et al. (1988).

unctuous shale, weak-layer accretion in a ductile duplex) during late Paleozoic Appalachian thrusting (Figs. 3, 4) (Thomas, 2001). The Appalachian thrust belt of Alabama contains carbonate and siliclastic rocks of Cambrian through Pennsylvanian age (Figs. 3, 4) (e.g., Thomas, 1985; Thomas and Bayona, 2005). The thrust belt is dominated by thin-skinned deformation in which Paleozoic strata have been translated northwestward above a basal detachment in Cambrian shale (Rodgers, 1950; Thomas, 1985). The Cambrian shale, including the Conasauga Formation, forms a weak lithotectonic unit that not only hosts the basal detachment, but is in places duplicated into giant mushwads that are locally thicker than 4000 m (13,000 ft) (Thomas, 2001, 2007b). The mushwads formed above a large-scale frontal ramp in the basal detachment along the northwest margin of the Birmingham graben. The Conasauga contains three named mushwads in this area (Gadsden, Palmerdale, and Bessemer; Thomas, 2001), and shale gas exploration has been conducted in the Gadsden and Bessemer mushwads (Fig. 3). The Cambrian shale section is overlain by a thick succession of Cambrian–Ordovician carbonate rocks called the Knox Group. The Knox Group constitutes a stiff layer that controls the geometry of major frontal and lateral thrust ramps and forms the roof of the Conasauga mushwads (Thomas, 2001, 2007b; Thomas and Bayona, 2005). Devonian through Pennsylvanian strata are preserved in broad, deep synclines. This section is dominated by siliciclastic rocks and forms a weak geomechanical unit that was translated cratonward

atop the stiff carbonates. These siliciclastic strata locally host upper-level and secondary detachments (Pashin and Groshong, 1998; Thomas, 1985; Thomas and Bayona, 2005). This mechanical stratigraphy has resulted in a distinctive suite of faults, folds, and fracture systems that should be taken into consideration when formulating shale gas development strategies. 3. Analytical methods The multidisciplinary approach of this study (Fig. 2) necessitates that a great range of materials and methods be applied to characterize shale gas reservoirs. Basic materials used include outcrops, cores, well logs, and seismic profiles. Methods include standard stratigraphic, sedimentologic, and structural procedures for rock description; correlation; mapping; construction, balancing, and restoration of cross sections; and interpretation. Hydrodynamic and geothermic analysis relied mainly on well records, as well as application of fundamental hydrogeologic principles to the regional geologic framework. A battery of petrologic procedures, including tight-rock analysis and isotherm analysis, was employed to characterize petrology, geochemistry, gas storage, and permeability. A deterministic approach to reservoir volumetrics was used to estimate gas resources and reserves. To determine the stratigraphic, sedimentologic, and structural architecture of Conasauga shale, geophysical well logs were correlated and analyzed, and incorporated into a database using Petra software. Cores and outcrops were described using standard field procedures.

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

73

Northwest

Southeast

A

A’

Murphrees Valley Big Canoe Creek Field anticline Sequatchie Blount Mtn. syncline Flatwoods S-D Coosa synclinorium anticline Coalburg syncline M-P

S-D

M-P

S-D

C-O C

C

Basement

Birmingham Graben

GADSDEN MUSHWAD

B Birmingham Sequatchie PALMERDALE MUSHWAD anticlinorium anticline S-D M-P Coalburg syncline S-D C-O

B’ Cahaba synclinorium S-D

Birmingham Graben

No vertical exaggeration

10 km

M-P C-O

C

Basement

5

Coosa synclinorium

M-P

C-O

0

M-P

M-P

C-O

C-O

Index map M-P S-D C-O C Basement

A

Mississippian-Pennsylvanian Silurian-Devonian Cambrian-Ordovician carbonate Cambrian Shale Crystalline basement

B

A’ B’

Alabama

Fig. 4. Balanced structural cross sections of the southern Appalachian thrust belt in Alabama (after Thomas and Bayona, 2005).

Cores are stored at the Geological Survey of Alabama, and access is open to the public. Restored and balanced structural cross sections (Thomas and Bayona, 2005) were used to constrain the thickness and geometry of the Conasauga mushwads. Hydrologic and geothermic information, including pressure data and bottom-hole temperature, were obtained from geophysical well logs and file reports made available by the State Oil and Gas Board of Alabama. Gas samples were collected from 5 wells producing from Conasauga shale using an isotube system and were analyzed for bulk and isotopic composition by Weatherford Laboratories. Billets and thin sections for petrologic analysis were made from core samples. Typical and diagnostic features in these samples were identified and documented using digital photomicrography. Stable isotopic analysis was performed to determine δ13C and δ18O values for vein-filling calcite cement, which provide insight into thermal and geochemical conditions during burial and hydrocarbon

Intrashelf basin

Ramp

Shale, thinly bedded micrite, Micritic limestone, and nodular micrite ribbon rocks, and shale

generation (e.g., Friedman and O'Neil, 1977). Isotopic composition was measured with a GasBench-IRMS system using methods similar to those described by Debajyoti and Skrzypek (2007). Isotope values are expressed in per mil (‰) relative to the Vienna Pee Dee Belemnite (VPDB) scale by use of the NBS-19 standard. Rough sample surfaces were examined with a JEOL 7000 field emission SEM to investigate and document the size, shape, fabric, and chemical composition of particles in shale. Core and cutting samples of shale were sent to Terra Tek (Schlumberger) and Weatherford Laboratories for TOC analysis, rock-eval pyrolysis, X-ray diffraction, tight-rock analysis (TRA), and methane adsorption isotherm analysis. TOC and rock-eval pyrolysis provide information on organic content, kerogen type and quality, residual hydrocarbon content, and thermal maturity. TRA provides information on the grain density, porosity, fluid saturation,

Shoal complex Oolitic and oncolitic limestone

Platform lagoon Stromatolitic limestone

Sea level

Conasauga limestone and dolostone

Fair-weather wave base Storm wave base

Euxinic basin

Conasauga shale Rome Formation Basement

IAPETAN RIFT COMPLEX

Fig. 5. Generalized depositional model of the Conasauga Formation in Alabama (after Astini et al., 2000; Markello and Read, 1982).

74

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

and permeability of shale, and is important for estimating free gas storage. Permeability was determined parallel to bedding by the pressure-decay method. Adsorption isotherm measurements were used to determine how adsorbed gas storage capacity varies with pressure at reservoir temperature. Gas resources were estimated using a deterministic approach. An isopach map showing the tectonic thickness of the Conasauga shale was made in Petra and provided primary control on reservoir volume. Porosity logs and cores helped constrain the proportions of microporous shale and nonporous carbonate. Isotherm and porosity data were used to estimate the volume of sorbed and free gas in the Conasauga. Reservoir pressure in the Conasauga is sufficiently high that the amount of sorbed gas could be estimated as Langmuir volume. Accordingly, the sorbed gas estimate can be considered as an upper limit. Free gas volume, by contrast, was estimated on the basis of porosity, reservoir pressure, and reservoir temperature. Reservoir pressure was assumed to be normal hydrostatic for the calculation. Sorbed and free gas volumes were gridded and contoured in Petra. The grids were then added together and contoured to develop maps of original gas-in-place (OGIP). In addition, estimates of reservoir area, sorbed, free, and total OGIP were derived using the basic volumetric functions in Petra. 4. Stratigraphy and sedimentation Conasauga shale is in facies relationship with a spectrum of carbonate rocks and was deposited as part of a regionally extensive carbonate ramp system (Fig. 5) (e.g., Astini et al., 2000; Markello and Read, 1981, 1982; Srinivasan and Walker, 1993; Thomas et al., 2000). The Conasauga Formation gradationally overlies the Lower Cambrian Rome Formation, which contains interbedded shale, siltstone, sandstone, and limestone and includes redbeds. The Conasauga Formation is overlain conformably by interbedded dolostone, chert, and sandstone of the Copper Ridge Dolomite, which forms the base of the Upper Cambrian–Lower Ordovician Knox Group. The Conasauga is characterized by complex internal stratigraphy and facies relationships, which are closely related to the tectonic framework of deposition within and outside the Birmingham graben (Butts, 1910, 1926, 1927; Osborne et al., 2000). In general, shale dominates the lower part of the formation, and limestone and dolostone predominate in the upper part. Palinspastically within the graben, shale generally dominates all but the uppermost part of the Conasauga. In contrast, palinspastically outside the graben, shale is limited to only the lowermost part, and most of the Conasauga interval is carbonate (Osborne et al., 2000). The carbonates include oolitic and intraclastic grainstones, although in many places the original textures are obscured by dolomitization. The carbonate succession of the upper Conasauga is also subdivided into the Brierfield, Ketona, and Bibb formations, which are distinguished on the basis of presence or absence of chert. In a grand sense, the Conasauga represents a giant shoaling-upward succession in which shale deposited in intrashelf basin and ramp environments passes into platform carbonate deposits. The age and character of the transition, however, vary depending on the relationship of the original depositional setting to Iapetan basement structures (Fig. 5) (Astini et al., 2000). 4.1. Dawson 33–09 #2A core Core from the Dawson 33–09 #2A well provides an important glimpse of Conasauga shale in Big Canoe Creek Field at depths between 2298 and 2309 m (7540 and 7577 ft) (Fig. 6). The shale is dark gray to very dark gray, is commonly calcareous and dolomitic, and forms laminae to thick beds. It is brittle in dry samples and retains water when wet. Internally, the shale is laminated (Fig. 6A), and the laminae are defined by variations in the proportions of clay, organic matter, carbonate, and quartz silt. Locally, low-relief ripple cross-laminae are preserved (Fig. 7). Sand- to granule-size pyrite nodules are present locally. No bioturbated shale has been observed in this core.

Fig. 6. Photographs of Conasauga shale in core from the Dawson 33–09 #2A well, Big Canoe Creek Field. (A) Laminated shale, 2302.2 m (7,553.2 ft). (B) Laminated shale with micrite nodules, 2302.2 m (7553.2 ft). (C) Micrite bed with flame structure, internal grading, and burrows, 2303 m (7555.8 ft). (D) Imbricate micrite clasts in shale, 2299.4 m (7544 ft). Core diameter = 10 cm.

Limestone in the Dawson 33–09 #2A core is typically medium gray to light gray and brittle (Fig. 6). Peloidal micrite predominates, although much of the limestone has been recrystallized. The thickest limestone beds are argillaceous, and some have sharp bases with flame structures and gradational tops (Fig. 6C). Some beds contain faint burrows and ripple cross-laminae. Although most limestone beds and nodules appear to be in facies relationship with adjacent strata, some nodules have been rotated, and layers of imbricate micrite pebbles have been preserved locally (Fig. 6D).

4.2. Dawson 34-03-01 exploratory core The Dawson 34-03-01 exploratory core penetrates 386 m (1265 ft) of section and provides an exceptional record of the upper part of the Conasauga shale mass in Big Canoe Creek Field (Fig. 8). The core is dominated by interbedded limestone and shale; the proportion of limestone increases upward. Calcilutite is most common in the lower 150 m (500 ft) of core. Higher in section, the carbonates are mainly calcarenite with abundant peloids and coated grains. Layers of conglomeratic limestone (i.e., calcirudite) are scattered through the section. Shale in the exploratory core is dark gray to very dark gray and mostly fissile. Laminae are not as clearly defined as in the deeper core. Another contrast is that the shale in this core does not appear to retain water when sprayed. Physical sedimentary structures in shale include mud ripples; pinstripe bedding formed by interlamination of shale and limestone

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

75

Uninterpreted photograph

Photograph with tracing Subparallel laminae

Ripple cross-laminae Subparallel laminae

Fig. 7. Subparallel and ripple cross-laminae in Conasauga shale from the Dawson 33–09 #2A well, Big Canoe Creek Field, 2302.6 m (7554.6 ft). Core diameter = 10 cm.

is common in parts of the core (Fig. 9). The carbonate pinstripes are commonly paired. In some sections of the core carbonate laminae form bundles of 10 or more pairs. Biogenic structures are common in medium to thick shale beds and include horizontal burrows (Planolites) and trilobite trails (Cruziana) (Fig. 10). Macrofossils include linguloid brachiopods, Agnostus sp., and ptychoparioid trilobites. Ribbons and flakes of organic material may be algal or other phytic debris. The ptychoparioid trilobites include complete individuals ranging in size from a few mm to about 2 cm in length (Fig. 10D); they are preserved on planar and rippled bedding surfaces in shale. All of the largest specimens are conspecific but are too poorly preserved for generic identification. Some are associated with Cruziana. By contrast, some laminae contain abundant impressions of trilobite debris, including more or less whole agnostids mingled with small (b 5 mm long) and disarticulated ptychoparioids. The best preserved impression of Agnostus sp. is an apparent exuvium because one thoracic segment is missing and the other is displaced (Fig. 11). Thinly interbedded shale and limestone can be considered the signature lithology of the ramp and basinal facies; and thin to medium beds of wavy, lenticular, and nodular limestone are interbedded with shale in most of the core (Fig. 12). The limestone has a variety of sedimentary structures and textures, including ripple cross-laminae (Fig. 12A), fenestral fabric (Fig. 12C), chert nodules, vertical burrows or borings, and synaeresis cracks (Fig. 12D). The overall coarsening-upward trend observed in the core is composed of four higher order coarsening-upward successions, or parasequences, ranging from about 45 to 200 m (150 to 650 ft) in thickness (Fig. 8). The basal part of the core (deeper than 399 m; 1,310 ft) consists mainly of interlaminated shale and micrite. This succession is punctuated by beds of peloidal calcilutite and calcarenite containing thin-walled fossils; the limestone beds are as much as 0.3 m (1 ft) thick and increase upward in abundance. The second parasequence extends upward to about 330 m (1085 ft) and contains a similar assemblage of rock types. These strata are overlain sharply by a third parasequence containing beds of calcirudite as much as 2 m (6.6 ft) thick in the upper part. The calcirudite includes clast- and matrix-supported fabrics (Fig. 13A, B) and soft-sediment deformation (Fig. 13B, C). The youngest parasequence is shallower than 268 m (880 ft) and is the thickest and most calcareous such unit in the core (Fig. 8). In fact, strata in the upper 122 m (400 ft) of the core display different rock types and sedimentary structures than those deeper in the

formation. Interbedded limestone and shale include ribbon rock, and stylonodular and fitted fabrics are common. The uppermost part of the core includes cross-bedded, oolitic calcarenite with echinoderm ossicles and oncoidal limestone with irregular dolostone laminae (Fig. 8). Microbial boundstone includes thrombolite and spar-cemented shelter voids. Mixed-particle calcarenite in this part of the core includes sponges, fragments of boundstone, intraclasts, ooids, brachiopods, trilobite sclerites, and abundant peloids. 4.3. Depositional processes and environments Following the lead of Markello and Read (1981, 1982), Astini et al. (2000) applied carbonate ramp and platform models to the Conasauga Formation of Alabama (Fig. 5). The shale facies of the Conasauga was recognized by Astini et al. (2000) as an outer ramp deposit of an oxygen-deficient intrashelf basin. The predominance of laminated shale and a lack of bioturbation in the deep Dawson 33–09 #2A core (Fig. 6A) supports the inference of euxinic sedimentation. Ripples in the shale and calcilutite provide evidence for episodic currents that swept the sea floor (Fig. 7). Graded micrite beds (Fig. 6C) suggest distal storm deposits (e.g., Aigner, 1985), although it seems that most shale sedimentation occurred below storm wave base. Bioturbation within the micrite beds indicates that the water in which the carbonate mud was transported was sufficiently oxygenated to support an ephemeral infauna. Divergence of laminae around micrite nodules (Fig. 6B) and angular micrite clasts (Fig. 6D) indicates early lithification of the carbonate and considerable compaction of the enveloping mud during burial. The presence of imbricate limestone clasts, moreover, confirms early cementation and demonstrates that currents were episodically strong enough to rip up, transport, and redeposit fragments of the calcilutite layers. The Dawson 34-03-01 exploratory core contains evidence for sedimentation in shallow water and is important for characterizing depositional processes on the Conasauga carbonate ramp. Bioturbation and body fossils suggest more consistent oxygenation of the water column (Fig. 10). Fragile skeletal elements preserved in peloid calcarenite, even in the lower part of the core, suggest a thriving in situ fauna, at least episodically. Agnostid trilobites are considered by many to have been pelagic organisms (Fortey and Theron, 1994; Müller and Walossek, 1987). They are highly abundant in a thin lamina containing small and fragmentary ptychoparioid trilobites, which is probably a storm deposit; and taphonomic evidence cited above and shown in Fig. 11 indicates deposition as exuviae. Linguloid brachiopods and ptychoparioid trilobites were

76

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

200 ft

Dawson 34-03-01 Exploratory Core Thrombolitic structures Sec. 34, T. 13 S., R. 4 E., 865 FNL, 2346 FWL Shelter voids 33.864° N. lat., 86.21191 W. long. Big Canoe Creek Field Ribbon rocks St. Clair County Sponge spicules

300

Dolostone laminae EXPLANATION 400

Rock types

500

600

Parasequence 4

Limestone Interbedded limestone and shale (nodular, lenticular, wavy beds) Shale Oil stain Sedimentary structures Cross-beds

700 Clast-supported Tidal bundles 800

Tidal bundles

Ripple cross-laminae Horizontal laminae Convolute bedding Intraclasts

1000

Parasequence 3

900

Matrix-supported Tidal bundles

Breccia Ooids Oncoids

Matrix-supported Tidal bundles

Synaeresis cracks

Biological structures Parasequence 2

1200

Burrows

Parasequence 1

1100

Trilobite fragments

Lingulid brachiopod Ptychoparioid trilobite Agnostid trilobite

1300

Tectonic structures Extensive tectonic deformation

Faults and microfaults Recumbent fold

SH Inter-

bedded

lutite arenite rudite

1400

Open fold Isoclinal fold

LS Fig. 8. Graphic log of the Dawson 34-03-01 exploratory core, Big Canoe Creek Field, Gadsden mushwad.

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

Fig. 9. Pinstripe-laminated shale and limestone in the Conasauga Formation (Dawson 34-03-01 core, depth = 316.1 m; 1,037 ft; diameter = 10 cm). Note common pairing of pale limestone laminae.

A

77

Fig. 11. Agnostid trilobite with broken and missing thoracic segments (arrow), suggesting origin by exuviation. Specimen is 1.2 mm long.

1029.2 ft

B

876.2 ft

10 mm 3 mm

C

772.5 ft

D 1036.8 ft p

p

3 mm

2 mm

a

Fig. 10. Trace and body fossils in the Dawson 34-03-01 exploratory core, Big Canoe Creek Field. (A) Planolites burrows, (B) Cruziana trail, (C) lingulid brachiopods, (D) ptychoparioid (p) and agnostid (a) trilobites.

78

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

B

The regional distribution of the black shale facies of the Conasauga Formation is imprecisely known. This is because few wells penetrate the shale, and nearly all outcrops and cores of the shale are in the Gadsden mushwad. One possibility is that the black shale facies is extremely widespread, which would indicate shale gas potential over a large part of northern Alabama. Alternatively, prospective black shale may have been deposited only in the Birmingham graben, a large, late Iapetan basement graben that houses a significant part of the Appalachian thrust belt in Alabama (e.g., Thomas, 1985; Thomas and Bayona, 2005).

A

5. Structural geology 5.1. Conasauga mushwads 450.5 ft 214 ft

C

D c

f s

487 ft

330 ft

Core diameter = 2.3”

Fig. 12. Nodular- and wavy-bedded limestone and shale in the Dawson 34-03-01 core. (A) Wavy-bedded limestone with ripple cross-laminae and fenestrae. (B) Wavy-bedded limestone with nodules and possible relict ripples. (C) Nodular limestone with spar-filled fenestrae (f). (D) Limestone with abundant synaeresis cracks (s) and a chert nodule (c).

benthic. Preservation on shale laminae of Cruziana, as well as complete brachiopod valves and intact juvenile and adult ptychoparioids, indicates that trilobites did indeed inhabit Conasauga mud substrates. Pairing of pinstripe laminae (Fig. 9) indicates paired sedimentation events resembling flood and ebb tide or semidiurnal tides. Indeed, bundling of paired laminae into groups of 10 or more suggests spring-neap tidal bundles resembling those described by Kvale et al. (1989). Some of the wavy, lenticular, and nodular limestone beds (Fig. 12) also may have formed in response to tidal action, although most may be storm deposits similar to those in the Dawson 33–09 #2A core. Clast-supported calcirudite (Fig. 13A) may have been deposited as ravinement and channel lags. Matrix-supported calcirudite (Fig. 13B, C), by contrast, indicates transport as debris flows, which may have been generated by slope and channel-bank failures. Strata in the upper part of the core were apparently deposited in shoal and platform lagoon environments (Figs. 5, 8). Indeed, crossbedded calcarenite verifies shoal development. Oncoidal limestone is likely lagoonal, and dolostone laminae indicate supratidal conditions. Thrombolitic strata with shelter voids may represent microbial buildups that grew in sheltered areas of the platform lagoon. A lack of stromatolitic rock types indicates that the youngest Conasauga strata are not represented in the core, although abundant stromatolites have been identified in the roof of the Gadsden mushwad near Big Canoe Creek Field (Thomas and Pashin, 2011). The two Conasauga cores together tell a tale of overall progradation in a range of depositional settings from outer ramp or intrashelf-basin floor to near sea level.

Surface mapping, drilling, and seismic exploration reveal that at least three Conasauga mushwads are preserved in the Alabama Appalachians (Figs. 3, 4) (Thomas, 2001). Exploration has focused primarily on the southeastern part of the Gadsden mushwad, in which a large area of Conasauga shale is exposed at the surface and forms a low-relief topography known locally as the Flatwoods. The Flatwoods area constitutes a crestal pop-up structure that is bounded on the northeast and southwest by a high-angle forward thrust and a mirror-image backthrust, respectively (Fig. 4). In cross section, the Gadsden mushwad is a nearly symmetrical structure in which a large part of the deformed shale mass is tucked below the Coosa synclinorium to the southeast of the Flatwoods and the Blount Mountain syncline to the northwest. The northeastern and southwestern margins of the Gadsden mushwad are defined by major lateral ramps that terminate major northeast-striking synclines, namely the Cahaba synclinorium and the Lookout Mountain syncline (Fig. 3). The Palmerdale and Bessemer mushwads are closely related and constitute the core of the Birmingham anticlinorium (Figs. 3, 4). These mushwads are strongly asymmetrical in cross section, reflecting the northwest-verging geometry of the Birmingham anticlinorium (Fig. 4). The crestal regions of the Palmerdale and Bessemer mushwads are overlain by a thin roof dominated by brittle Conasauga and Knox carbonate rocks, and little Conasauga shale is exposed at the surface. The roof of these mushwads contains a complex array of forward thrusts and backthrusts but appears fairly smooth in cross section. The northeastern margin of the Palmerdale mushwad is marked by major cross-strike structural changes associated with structural relay between the Blount Mountain syncline and the Cahaba synclinorium (Fig. 3). By contrast, the boundary between the Bessemer and Palmerdale mushwads is defined by relatively subtle along-strike variations within the Birmingham anticlinorium. The southwestern margin of the Bessemer mushwad is concealed below Mesozoic-Cenozoic cover of the Gulf of Mexico coastal plain and thus is not well defined. In addition to the giant mushwads described herein, additional thick shale bodies may be concealed below the shallow Rome thrust sheet in northeast Alabama (Maher, 2002) and in structures northwest of the Rome thrust sheet in Georgia (Cook and Thomas, 2010; Mittenthal and Harry, 2004). An isopach map of the deformed Conasauga shale was made using the balanced cross sections of Thomas and Bayona (2005) to control thickness (Fig. 14). The resulting map suggests that, together, the three mushwads form a giant antiformal ridge that strikes about N. 45° E. The Gadsden mushwad is locally thicker than 4000 m (13,000 ft), and the contour pattern reflects the general symmetry of the mushwad. A large part of the shale mass is exposed at the surface (Fig. 3), and so the shale mass must have been much thicker at one time (Thomas, 2001). The asymmetrical contour pattern of the Palmerdale and Bessemer mushwads reflects the northwest structural vergence of the Birmingham anticlinorium (Fig. 14). The Palmerdale mushwad corresponds to a structural saddle in the roof of the mushwad, and most of the shale mass is thinner than 2600 m (8500 ft). The Bessemer mushwad is much thicker, and maximum shale thickness

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

A

760 ft B

888 ft

79

C 984 ft

Fig. 13. Conglomeratic limestone (calcirudite) in the Dawson 34-03-01 core. (A) Clast-supported conglomerate underlying laminated shale. (B) Poorly sorted clasts floating in argillaceous matrix. (C) Deformed shale laminae and edgewise limestone clasts.

exceeds 4200 m (14,000 ft) at the southwestern end of the mapped shale mass. 5.2. Internal deformation Thomas (2001) recognized that the Conasauga mushwads constitute zones of bulk ductile deformation; they are composed of small, regionally dipping thrust horses that are internally deformed and are separated by zones of intense disharmonic deformation. Exposures within the Gadsden mushwad provide the best views of this deformation (Pashin et al., 2010a; Thomas and Pashin, 2011). The regionally dipping horses contain minor folds, abundant joints (Fig. 15A), and carbonate-filled veins (Fig. 15B, C). Regionally, the horses generally strike N. 55° E. and dip about 10–35° SE. Intensely deformed strata between the structurally coherent horses range from mylonitic fault gouge (Fig. 15D) to chaotic deformation zones with abundant folds and faults (Fig. 15E, F). Fold types within the chaotic zones are dominated by chevron, isoclinal, and similar folds. Limb attitude is extremely variable, and most folds plunge steeply. One possibility is that these intensely deformed strata are accommodation zones that help fill space along the margins of the regionally dipping thrust horses. Similar deformation structures are in the Dawson 34-03-01 exploratory core (Fig. 16). Bedding typically dips between 10 and 30°, and the core largely preserves original stratigraphic order. Numerous

faults and shear fractures are distributed through the core. Shear fractures include conjugate fractures and tension gashes, which are filled with calcite spar. Locally, oil is present in voids within the calcite fill. Stylolites are also important deformation structures in the core. Most are bed-parallel, whereas others are perpendicular to bedding. Some stylolites truncate bedding, indicating that they may have formed along faults or that a large amount of carbonate has been dissolved. Deformation is especially intense in the basal 30 m (100 ft) of the core and may record shearing associated with a thrust zone in the mushwad. 5.3. Natural fractures and veins Natural fracture networks are important recorders of the stress history of sedimentary basins and generally have a strong influence on the reservoir properties of shale (Engelder et al., 2009; Gale et al., 2007). Fractures in Conasauga shale include orthogonal joint networks and shear fracture networks. Joints are best developed within the coherent structural panels in outcrop (Fig. 15A); they are perpendicular to bedding and form orthogonal networks consisting of systematic joints and cross joints. Systematic joints in the Conasauga tend to be planar fractures that extend laterally for several meters. These fractures typically follow regional strike (N. 45° E.) and are spaced as close as 2 cm in some laminated shale. Cross joints effectively strike perpendicular to systematic joints and commonly terminate at intersections with systematic joints. In the Conasauga, cross-joints are poorly developed. Close fracture

80

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

EXPLANATION > 3750 m 3000-3750

Etowah

2250-3000 1501-2250

Blount

750-1500 1-750 0

EN

Contour interval = 75 m

D DS

GA

A

LE

Structural transect

SE

M

ER

PA

Jefferson

LM ER

D

St. Clair

B

ES

Shelby

Tuscaloosa

Bibb 10 10

0 0

20

10 10

30 km 20 mi

Fig. 14. Isopach map of deformed Conasauga shale constituting the Gadsden, Palmerdale, and Bessemer mushwads. Thickness control from balanced structural cross sections of Thomas and Bayona (2005).

spacing in organic-rich strata, such as coal and shale, is thought to be a product of stresses generated by tectonic forces, as well as stresses generated by devolatilization and hydrocarbon generation during burial and thermal maturation (e.g., Engelder et al., 2009; Laubach et al., 1998). Joints are abundant in outcrop, but they have not been observed in core and may thus be of limited significance for hydrocarbon exploration. Shear fractures include a variety of bed-parallel and dipping fractures. Mineralized bed-parallel fractures, or veins, are abundant in shale of the Conasauga Formation (Fig. 15B). These veins locally have kinematic aperture exceeding 3 cm and contain fibrous calcite that grew perpendicular to the vein walls. The fibers typically terminate at a keel line in the interior of the vein. Some of the veins, by contrast, cut across bedding, and many fill en echelon tension gashes (Fig. 15C). These types of structures have been observed in core from Big Canoe Creek Field and are thus significant to shale gas exploration. This fibrous cement is typical of synkinematic crack-seal textures (Laubach et al., 2004); thus vein filling is considered to be contemporaneous with mushwad formation. The wide kinematic aperture of these veins, moreover, indicates a high degree of dilatancy associated with fracturing and mineralization.

6. Hydrodynamics and geothermics The stratigraphic and structural framework of a sedimentary basin provides the basic container in which hydrodynamic and geothermic processes operate. These processes have a strong influence on the generation, composition, storage, and mobility of reservoir fluid (Fig. 17). The principal elements used to characterize basin hydrodynamics and geothermics

are porosity, permeability, fluid chemistry, reservoir pressure, and reservoir temperature. The Conasauga Formation and Knox Group in the Appalachian thrust belt contain shale and carbonate units that have a broad range of reservoir properties. Shale characteristically has porosity of 1–6% and matrix permeability on the order of 0.1 μD (e.g., Soeder, 1988). Limestone and dolostone in the Conasauga Formation and Knox Group have varied hydrologic properties (Ortiz et al., 1993). Porosity varies from negligible to more than 20%, and permeability ranges from the μD scale to more than 100 mD. Accordingly, some carbonate units have sealing capacity, whereas others, like those in the Copper Ridge Dolomite of the basal Knox Group, serve as aquifers that can transmit fluid long distances. For example, fresh-water recharge of the Copper Ridge in the Murphrees Valley anticline (Figs. 3, 4), has resulted in fresh-water plumes that extend at least 50 km to the west, thus limiting possibilities for disposal of produced water and making air drilling through the roof thrust into Conasauga shale difficult. Several wells penetrating Cambrian Conasauga shale in the Gadsden and Bessemer mushwads have reported significant gas shows, demonstrating that gas pressure exceeded fluid pressure during drilling. Blowout of the Andrews 27–14 well in Big Canoe Creek Field (Williams, 2007), moreover, indicates overpressure (Fig. 17). Reservoir temperature varies depending on reservoir depth and geothermal gradient. Sparse data from the Conasauga Formation indicate a maximum recorded reservoir temperature of 74 °C (165 °F) at a depth of 3781 m (12,406 ft). On the basis of bottom-hole temperature data from multiple wells, geothermal gradient in the Conasauga is only 14–16 °C/km (8–9 °F/1000 ft).

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

81

Fig. 15. Outcrop photos of deformation in Conasauga shale in the Gadsden mushwad near Pinedale Lake, St. Clair County, Alabama. (A) Closely spaced joints in interbedded shale and limestone. (B) Bed-parallel vein filled with fibrous calcite. (C) En echelon tension gashes filled with fibrous calcite. (D) Mylonitic fault gouge. (E) Chevron folds. (F) Isoclinal fold.

Although the Conasauga is not warm enough to generate thermogenic hydrocarbons today, vitrinite reflectance data indicate that it had been heated enough in the past to have generated large volumes of natural gas. The woody plant material most commonly associated with vitrinite did not exist during Cambrian time. Instead, herbaceous vitreous kerogen can be used as a surrogate for vitrinite for reflectance studies of Cambrian strata (Burchard and Lewan, 1990). Reflectance of the Conasauga ranges from 1.1 to 1.9% and increases with depth, indicating that the shale is in the thermogenic gas window (Fig. 18). Below a depth of 1070 m (3500 ft), the shale lies in the main gas generation window. The geochemistry of produced gas from the Conasauga further supports a thermogenic origin (Table 1). A plot of δ 13C1 and δDCH4 isotopic ratios supports thermogenesis (Fig. 19), and a dryness index (100*C1/C1–5) of 95 to 97 suggests derivation from a sapropelic source. The ratio of C2 to C3 hydrocarbons is between 7 and 10, and the difference between δ 13C2 and δ 13C3 is between − 5 and 0, which are consistent with an origin by secondary cracking

of oil and gas (see Lorant et al., 1998). The increase of reflectance with depth is log-linear and fairly uniform (Fig. 18). The absence of major anomalies of thermal maturity corresponding with folding and thrusting within the Gadsden mushwad indicates that thermal maturation and gas generation were effectively post-kinematic. Therefore, it appears that the mushwad was emplaced relatively early in the evolution of the southern Appalachian thrust belt. Indeed, burial and thermal history models in the region indicate that major thermal maturation occurred near maximum burial, probably during the Permian (e.g., Carroll et al., 1995; Telle et al., 1987). 7. Petrology and geochemistry 7.1. Shale composition Conasauga shale is a variable mixture of clay minerals, silica, carbonate, pyrite, and organic matter (Fig. 20; Tables 2, 3). Few samples can be classified as clay shale, and many can be classified

82

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

1440 ft

0.0

1443.5 ft

If

Early Gas Generation

Main Gas Generation

0.5

Depth (km)

Of

Sf

1.0

1.5

M Mf 2.0

Core diameter = 2.3 “

1.5

1.0

Of Sf If Mf M-

Open fold Sheared fold Isoclinal fold Microfault Mylonite

2.0

Vitrinite Reflectance (Romax) Fig. 18. Plot of vitrinite-like kerogen reflectance versus depth in the Conasauga shale of Big Canoe Creek Field, St. Clair County, Alabama.

M

Fig. 16. Tectonic deformation structures in the Dawson 34-03-01 core.

Conasauga Production

Chattanooga Production Fresh-water recharge, shallow, cool reservoirs

Join ts Listric ramp

Thermogenic gas, deep, warm reservoirs

Hinge-zone sweet spots? M-P S-D C-O

Karst C-O

S-D

Hinge

Cretaceous

Mushwad

Fold hinge

Cambrian shale

Sh Ramp-flat

ea

r

Bedding Fault

C

?

C

S-D

C-O Cambrian-Ordovician carbonate

PM C-O

M-P Mississippian-Pennsylvanian Silurian-Devonian

K

?

C

Thermogenic gas, hydrocarbon pressure in fractures K

Devonian Gas Shows

Structural form line (hypothetical) Subsurface flow potential

Basement Fig. 17. Hydrodynamic model of shale gas reservoirs in the Appalachian thrust belt of Alabama (after Pashin, 2009).

S-D C-O

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

83

Table 1 Results of geochemical analysis of natural gas produced from the Conasauga Formation in Big Canoe Creek Field (analyses courtesy of Weatherford Laboratories). Well

Gas composition (%)

Beason E33-06-14 Dawson 33–09 #2A Dawson 34-03-01 Bearden E26-11-29 Oakes E23-11-26

Isotopic data (‰)

CH4

C2H6

C3H8

iC4

nC4

iC5

nC5

C6+

CO2

δ13C1

δ13C2

δ13C3

δ13iC4

δ13nC4

δ13CCO2

δDCH4

94.94 96.56 96.06 96.27 94.24

3.36 3.05 3.39 3.23 4.13

0.40 0.30 0.43 0.38 0.54

0.03 0.02 0.03 0.03 0.02

0.03 0.02 0.03 0.02 0.04

0.01 0.00 0.01 0.00 0.00

0.00 0.00 0.00 0.00 0.00

0.01 0.01 0.00 0.00 0.00

1.22 0.04 0.05 0.06 1.03

−39 −39 −40 −39 −38

−36 −37 −36 −35 −39

−31 −33 −32 −31 −39

−23

−30

−11

−130 −129 −130 −130 −136

as argillaceous dolomitic limestone (e.g., calcilutite and calcisiltite). Whereas much of the material in the shale can be classified as clay and silt, some detrital and biogenic particles are as large as coarse sand. Examination of shale in thin section risks sampling bias because the most friable rock types cannot be prepared. This was a particular problem in the Conasauga, where friability and fluid sensitivity made preparation difficult. This section of the report describes Conasauga shale based on X‐ray diffraction results and petrographic examination of the rock types that could be examined in thin section. X-ray diffraction indicates that Conasauga shale contains a diverse mineral suite (Tables 2, 3). Non-clay minerals are dominated by calcite, dolomite, and quartz. Calcite content is highly variable, ranging from 8 to 55%. Dolomite content increases with depth and is as great as 25%. Quartz content is remarkably consistent, typically between 12 and 20%. Most quartz appears biogenic or authigenic. Feldspar constitutes as much as 19% of some shale and appears to be associated with the clay fraction. Clay minerals account for 12 to 50% by weight of the samples analyzed and are dominated by illite, smectite, and mica. Chlorite composes as much as 14% of some samples, whereas kaolinite is a minor constituent. The provenance of the siliciclastic fraction of the Conasauga is unknown, although the high proportion of quartz and feldspar suggests denudation of granitic basement uplifts and recycling of quartzofeldspathic sediment and sedimentary rock. Indeed, determining the original detrital composition of Conasauga shale is difficult because of extensive diagenetic recrystallization and replacement of carbonate minerals. Aside from clay, quartzofeldspathic silt, carbonate, and organic matter (Fig. 20), the principal detrital components are fossil fragments and siliceous ovoids (Fig. 21). The ovoids are of enigmatic origin; some are filled with polycrystalline quartz, and some are hollow, suggesting a biogenic origin. Some of these ovoids are apparently remains of radiolarians, although many may be chertified peloids and coated grains.

7.2. Vein fills Mineral cements in fractures are sensitive recorders of geochemical processes (Laubach et al., 2004; Olson et al., 2009). Carbon and oxygen isotopes are valuable for determining the relationships of vein-filling calcite cement to burial and hydrocarbon generation. In coalbed methane reservoirs of the Black Warrior basin, for example, δ 13C values have been used to distinguish cementation in brine from that associated with late-stage bacterial methanogenesis, and δ 18O values have been used to constrain the temperature of mineralization (Pitman et al., 2003) (Fig. 22). Stable isotopic data from Alabama shale formations plot in a narrow range compared to the coalbed methane reservoirs. In the Conasauga Formation, δ 13C values are positive near the surface and are depleted at depth (Fig. 23). Positive δ13C values suggest near-surface bacterial activity, which may include the digestion of light hydrocarbons, and the values trend downward toward those of normal marine carbonate (~0 ‰). Values of δ18O as high as −2 ‰ indicate precipitation at or near surface temperature, and decreasing values with depth are consistent with modern burial temperature. Values from the deep Dawson 33–09 #2A core are not appreciably more depleted in 13C and 18O than those from the shallow core. Why are carbon and oxygen isotopic data from the calcite veins more consistent with shallow marine carbonate than deep burial cement? The answer is probably related to the abundance of marine limestone beds within and adjacent to prospective shale gas formations. Much of the carbon and oxygen in the vein fills was probably derived from dissolution of limestone by basinal fluids during burial, and stylolites provide the most obvious evidence for this dissolution. The scavenged carbonate apparently retained the isotopic signature of the parent carbonate, thereby enriching the basinal fluids in 13C and 18O. The strongly dilatant habit of the vein fills (Fig. 15B, C), moreover, suggests that cementation was contemporaneous with the high fluid pressures associated with overthrusting and gas generation.

The Conasauga Formation predates the earliest known land plants, which are of Silurian age, and so the types of kerogen preserved differ

Bacterial CO2 reduction

-100

δ13C methane (‰)

−12

7.3. Organic petrology

-120

-80

−35

Bacterial methyl-type fermentation

-60

Conasauga shale samples

Mixed The

rmo gen esi ing t s her ma l m atu rity

Increas

-40

-20 -400

-300

-200

-100

δD methane (‰) Fig. 19. Cross-plot of stable isotope data showing thermogenic origin of natural gas produced from the Conasauga Formation in Big Canoe Creek Field.

Fig. 20. Photomicrograph showing detrital organic matter in dolomitic Conasauga shale, Dawson 33–09 #2A well, Big Canoe Creek Field, 2308.5 m (7573.8 ft).

84

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

Table 2 Non-clay mineralogy of Conasauga shale. Numbers in weight% unless otherwise indicated. Well

Depth (m)

Quartz

Plagioclase

Calcite

Ankerite, Ferroan Dolomite

Dolomite

Pyrite

Fluorapatite

Barite

Total Non-Clay

Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Statistics n Mean Minimum Maximum Std. deviation

176 257 260 266 276 277 289 289 359 367 372 398

7 19 20 21 22 17 14 13 16 17 17 15

K-Feldspar 5 8 8 6 13 2 1 6 2 1 1 0

3 7 7 8 6 6 1 5 6 6 5 3

40 13 11 12 8 55 41 49 20 30 17 43

5 7 3 3 2 1 6 1 6 7 10 6

4 0 0 0 1 0 25 1 23 11 17 20

1 1 1 1 1 0 0 1 1 1 1 1

0 0 0 1 0 0 1 0 0 0 0 1

0 0 0 0 2 0 0 0 0 0 0 0

66 54 50 52 54 81 88 75 74 72 67 88

12 16 7 22 4

12 4 0 13 4

12 5 1 8 2

12 28 8 55 16

12 5 1 10 3

12 8 0 25 10

12 1 0 1 0

12 0 0 1 0

12 0 0 2 0

12 68 50 88 13

from those in younger black shale units. Identifiable organic particles include kerogen types II through IV. Type II kerogen is oil-prone and includes palynomorphs, or sporinite, which consisted primarily of acritarchs during the Cambrian. As mentioned in the section on geothermics, herbaceous kerogen resembling vitrinite was used for reflectance analysis (Fig. 18). In addition, small amounts of inertinite are present, which is type IV kerogen that principally represents oxidized and fungal organic matter; the latter did not evolve until after the Cambrian. Type IV kerogen lacks significant potential for hydrocarbon generation but may store hydrocarbons. The dominant type of organic matter in Conasauga shale is matrix bituminite. Matrix bituminite is amorphous kerogen that is dispersed throughout the argillaceous rock matrix and gives the shale dark color. Matrix bituminite is

oil-prone, gas-prone, or inert, and is generally considered type II kerogen. Geochemical analysis of Conasauga shale indicates that TOC content averages 0.4% but is higher than 1.7% in some samples (Fig. 24A). Rock-eval pyrolysis reveals that the shale plots geochemically as a type IV kerogen source rock on the pseudo-Van Krevelen diagram, indicating that much of the hydrocarbon generation and expulsion potential has been exhausted (Fig. 24B). Some of the most organic-rich shale plots as a mixed type II-III kerogen source rock, which stands in stark contrast to the bulk of the shale. One possibility is that layers rich in acritarchs or herbaceous kerogen (Fig. 20) help elevate the organic and hydrogen content.

7.4. Microfabric Table 3 Clay mineralogy of Conasauga shale. Numbers in weight% unless otherwise indicated. Well

Depth Smectite Illite-Smectite Illite, (m) Mica

Kaolinite Chlorite Total Clay

Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Statistics n Mean Minimum Maximum Standard deviation

176

12

5

15

0

1

34

257

1

17

15

1

13

46

260

2

26

9

0

14

50

266

0

20

14

0

14

48

276

1

15

15

2

14

47

277

0

0

15

0

4

19

289

0

0

11

0

2

12

289

0

7

12

0

7

25

359

1

10

12

1

3

26

367

1

7

15

1

4

28

372

1

11

13

1

7

32

398

0

0

9

1

2

12

12 1 0 12 3

12 10 0 26 8

12 13 9 15 2

12 1 0 2 1

12 7 1 14 5

12 31 12 50 13

Shale fabric represents a network of detrital minerals, authigenic minerals, organic matter, and porosity. As such, the rock fabric is ultimately the product of sedimentation, compaction, diagenesis, and hydrocarbon generation. SEM microscopy is an excellent way to explore the fabric of fine-grained rocks and is thus the topic of this discussion. Platy clay minerals are a fundamental part of the petrologic framework of shale and are the most conspicuous elements of rock fabric. In Conasauga shale, illite plates commonly are weakly aligned (Fig. 25A). This weak alignment of clay plates may reflect high fluid pressure that arrested compaction during formation of the Gadsden mushwad and may include relict fabric related to clay flocculation (e.g., Slatt and O'Brien, 2011). Platy clay is locally folded around quartz, feldspar, and carbonate particles and commonly forms coats on such particles (Fig. 25B, C). This combination of alignment and folding is interpreted to be a product of compaction and displacive mineral growth.

Fig. 21. Photomicrograph showing siliceous ovoids in Conasauga carbonate, Dawson 33–09 #2A core, Big Canoe Creek Field, 2303 m (7555.7) ft.

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

Basinal brine

-2

A. KEROGEN QUALITY

Bacterial methanogenesis

16

Remaining Hydrocarbon Potential (S2, mg/g)

0

-6 -8

Heat

δ18O calcite (‰ VPDB)

-4

-10 -12

Marine carbonate

-14 -16

-5

0

5

10

15

20

12

Type I

10

Type II

8 6 4

Mixed Type II-III

2

Type III Type IV 0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

Total Organic Carbon (wt. %)

-20 -10

14

0 0.0

Pennsylvanian shale, coal Floyd (Neal) Shale Devonian shale Conasauga shale

-18

85

B. KEROGEN TYPE

25

δ13C calcite (‰ VPDB)

1000

Fig. 22. Cross-plot comparing the stable isotopic characteristics of vein fills in shale gas reservoirs in Alabama to those in coalbed methane reservoirs of the Black Warrior basin.

800

Hydrogen Index

Euhedral laths and fibers of illite and apatite (Fig. 25A, D) are rare and demonstrate that a fraction of the clay is diagenetic. These laths and fibers require open space in which to grow, and thus provide the best evidence for open porosity. The laths are not folded or broken, indicating formation after most compaction had occurred. Quartz, feldspar, and carbonate minerals are dispersed throughout the shale matrix. Quartz and feldspar grains typically appear as rounded bodies but are difficult to record because they commonly are overgrown with clay and have backscatter properties similar to clay and carbonate (fig. 25B). Carbonate minerals are in distinct rhombs and thus are readily identified (Fig. 25C, D). Pyrite forms framboids, loose clusters, and isolated crystals of varying size (Fig. 26A). Common crystal habits include cubes, bipyramids, and pyritohedra. Framboidal pyrite is locally abundant, and most framboids

l

II 600

400

200

III IV

0

0

20

40

60

80

100

Oxygen Index Fig. 24. Results of kerogen analysis in the Conasauga shale of Big Canoe Creek Field.

Calcite isotopic value (‰ VPDB)

0.5

-4.0

-2.0

0.0

2.0

4.0

6.0

δ13C Remobilized marine carbonate

1.5

bonate

1.0

2.0

ce rfa nt u s e ar- hm Ne nric e

δ18O Remobilized marine car

Depth (km)

-6.0

Lo w ce -te m mp en e ta ra tio tu n re

-12.0 -10.0 -8.0 0.0

are b10 μm in diameter. Most pyrite was deposited directly from the water column or formed within pre-existing sediment or rock (e.g., Wignall and Newton, 1998). Framboids are a source of microporosity in shale, and the interstices among crystals hosted crystalline illite growth (Fig. 26B). Organic matter in gas shale takes many forms and has a varied expression in SEM secondary electron images. Because kerogen is less dense than minerals, discrete kerogen particles appear dark. Matrix bituminite coats virtually all surfaces within organic-rich black shale. These coatings are very thin and thus barely perceptible in SEM photomicrographs. The bituminite, however, is readily vaporized by the microprobe scanning beam. Porosity is difficult to image using an SEM. The principal difficulty is the inability to distinguish plucked voids from dissolution voids (Fig. 25D). Accordingly, some of the most convincing microscopic evidence of porosity in shale comes from pores containing euhedra that had to grow into open space. Pore throats in shale typically range in size from 0.1 to 0.0005 μm and thus approach or overlap the size of gas and hydrocarbon molecules (e.g., Nelson, 2009). Microporosity in Conasauga shale appears widespread, but SEM imaging allows only qualitative analysis. Quantifying porosity, permeability, and gas storage is the focus of the next section of this paper. 8. Gas storage and permeability

Fig. 23. Relationship of isotopic composition of calcite veins to depth in the Conasauga Formation.

Shale is a dual-porosity reservoir rock (e.g., Kuuskraa et al., 1992; Montgomery et al., 2005; Ross and Bustin, 2008). Gas is stored primarily

86

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

A

B f

d

i f

C

D Labels

a

f

r c r

f - Folded clay i - Illite lath d - Detrital grain c - Carbonate rhomb a - Apatite r - Rhombic void

Fig. 25. SEM secondary electron images of Conasauga shale in the Dawson 33–09 #2A core, Big Canoe Creek Field. (A) Poorly aligned and folded clay platelets, 2308.9 m (7,575 ft). (B) Detrital grain (quartz or feldspar) coated with illite, 2308.9 m (7575 ft). (C) Carbonate rhomb in matrix of platy illite, 2308.9 m (7575 ft). (D) Apatite laths in platy clay matrix, 2303.9 m (7558.8 ft).

in a free state in the interstitial porosity of the shale matrix and in an adsorbed state in the microporosity of the organic matter (Fig. 27). Gas stored in a free state is expected to follow the pressure-volume-temperature relationships dictated by ideal gas law. Assuming constant temperature, therefore, free gas mobility is predicted to be fairly constant across a broad range of reservoir pressure. By contrast, adsorbed gas is attracted to free surfaces by Van der Waals forces and thus can be characterized using Langmuir parameters. As such, adsorbed gas can be mobilized by small pressure changes at low reservoir pressure, where the slope of the Langmuir isotherm is steep. Larger pressure changes are required to mobilize gas at elevated pressure, where the slope is low. Regardless of storage mechanism, however, the permeability of shale is very low, typically on the order of 0.1 μD (e.g., Soeder, 1988). Free gas is thought to flow primarily by Darcian processes and thus responds directly to pressure changes. Although adsorption is a pressure-sensitive phenomenon, gas flows in organic matrix primarily by Fickian processes, or diffusion, which is driven by concentration gradients rather than pressure. The contrasting storage and flow mechanisms in shale give rise to reservoirs with characteristics resembling a mix of tight sandstone and coal. 8.1. Porosity, permeability, and fluid saturation Even though Conasauga shale is thermally mature and has effectively exhausted the potential for hydrocarbon generation, core analysis demonstrates that the shale is capable of holding large volumes of natural gas (Table 4). On the basis of the Dawson 34-03-01 core, shale in the Conasauga Formation has effective porosity between 1.4 and 5.4%. On average, gas occupies 66.5% of this pore volume. Gas saturation increases with depth to a maximum of 93.5% of pore volume and is inversely

related to water saturation. Water saturation decreases from more than 52.6% of pore volume shallower than 275 m (900 ft) to less than 24.3% at greater depth. Oil saturation averages 2.5% of pore volume and is fairly consistent throughout the core. Matrix permeability was measured parallel to bedding using the pressure-decay method. Bed-parallel permeability ranges from 0.105 to 0.180 μD and averages 0.133 μD, which is comparable to other shale-gas formations (e.g., Ross and Bustin, 2008; Soeder, 1988). These values are more typical of confining units than reservoir rocks. Hence, as in other tight reservoirs, the mobility of free gas is too low without significant gas pressure and natural or induced fractures to support commercial flow rates. 8.2. Adsorption Adsorption isotherms were run for methane on shale samples at reservoir temperature as estimated from bottom-hole temperature observations. Results from Conasauga shale of the Dawson 34-03-01 core indicate that CH4 adsorption capacity is generally low (Table 5). The performance of the samples analyzed can be characterized in terms of Langmuir volume and Langmuir pressure. Langmuir volume is the adsorption capacity at infinite pressure. Langmuir pressure, by comparison, is the pressure at which adsorption capacity is 50% of Langmuir volume. Langmuir pressure is a measure of isotherm shape. Isotherms with low Langmuir pressure have a steep slope at low pressure that flattens greatly at high pressure; those with high Langmuir pressure tend to maintain slope at elevated pressure. Langmuir volume is very low in Conasauga shale, ranging from 0.29 to 0.89 scc/g (Table 5). Langmuir volume correlates strongly with TOC content (Fig. 28). The y-intercept of the regression line is 0.28 scc/g, indicating that 31 to 96% of the gas is adsorbed on clay and other inorganic

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

87

Gas capacity (scc/g)

10

sm

il ob

Limited gas mobility

it y

hg

a

Adsorption

ity

Hi g

bil

o sm

rm

Free storage

ga

ifo

Un 0

0

Pressure (MPa)

20

Fig. 27. Generalized diagram showing storage mechanisms for natural gas in shale and their effect on the mobility of gas.

Fig. 26. SEM secondary electron images of pyrite in the in the Dawson 33–09 #2A core, Big Canoe Creek Field. (A) Cluster of spherical to oblate framboids, 2308.9 m (7575 ft). (B) Detail of pyrite framboid showing illite overgrowths on pyrite crystals, 2299.5 m (7544.2 ft).

surfaces. This is unusual for prospective shale-gas formations and results partly from the low TOC content of most of the shale. In addition, inorganic adsorption values in Conasauga shale are 3.95 times higher than those determined in Devonian shale in Alabama (Pashin et al., 2010b). This suggests that the surface area of the mineral framework in Conasauga shale is substantially higher than that in most black shale and may be related to the weak alignment of clay matrix in the Conasauga (Fig. 25). Langmuir pressure is moderate in Conasauga shale, ranging from 3.63 to 4.62 MPa (Table 5). This narrow range contrasts sharply with Langmuir pressure in coalbed methane reservoirs of the Black Warrior basin, which ranges from 1.8 to 6.1 MPa, indicating that the mobility of adsorbed gas is more variable than in the shale reservoirs (Pashin, 2010). The high consistency of Langmuir pressure values in Conasauga shale facilitates prediction of the mobility of the adsorbed gas fraction. The relatively high Langmuir pressure of the shale, moreover, helps the adsorbed gas maintain mobility at elevated reservoir pressure.

9. Gas resources Volumetric analysis of the Conasauga Formation in the Gadsden, Palmerdale, and Bessemer mushwads indicates that an enormous natural gas resource base exists in Cambrian shale of the Appalachian thrust belt (Table 6; Fig. 29). Free gas resources are estimated to exceed 14,800 Bcm (523,800 Bcf), whereas sorbed gas resources are estimated at nearly 2900 Bcm (101,200 Bcf). Accordingly, the total gas resource in the area evaluated may exceed 17.7 Tcm (625 Tcf) and is stored mainly in a free state. This value is remarkable, considering

that these estimates were discounted to account for 50% limestone in the mushwads. Mapping OGIP in the mushwads indicates that resource concentration follows isopach contours and locally exceeds 10.9 Bcm/km 2 (1000 Bcf/mi 2) (Fig. 29). Additional potential exists within the Appalachian thrust belt in northeast Alabama and northwest Georgia (e.g., Cook and Thomas, 2010), and possibly within undeformed strata northwest of the Birmingham graben and below the basal thrust detachment within the graben. Thus, the Conasauga is unquestionably a vast exploration target. Recoverability and reserves in the Conasauga Formation are difficult to assess because of the technical challenges encountered during initial development. Indeed, the Energy Information Administration website currently carries no proven shale gas reserves in Alabama and carried a maximum proven reserve estimate of only 2 Bcf in 2008 as production was established in Big Canoe Creek Field (http://www.eia.gov/dnav/ng/ ng_enr_shalegas_dcu_sal_a.htm). Estimates of the long-term recoverability of shale gas vary greatly, and uncertainty exists regarding the types of production decline curves that should be used for reserve estimation (Seidle and O'Connor, 2011; SPEE, 2001). Not only are the long-term decline characteristics of Conasauga wells unknown, but the response of wells to restimulation is undetermined. Because no data are available to bracket the long-term performance of Conasauga shale wells, simple estimates of 10 to 20% of OGIP were made to provide baseline information on the volumes of gas that may be technically recoverable. Accordingly, the amount of technically recoverable gas in the Gadsden, Palmerdale, and Bessemer mushwads is estimated to be 1.8 to 3.5 Tcm (62 to 125 Tcf) (Table 6). Volumetric analysis indicates that more than 80% of OGIP is in a free state, which indicates that the quantity of gas per unit volume of rock effectively increases linearly with depth. The sorbed gas resource is limited by low TOC content, although clay sorption appears to play a stronger role in the Conasauga than in most other shale formations. Hence, development efforts should perhaps focus on the free gas fraction, which is concentrated in deep reservoirs and has greater mobility than sorbed gas at elevated reservoir pressure (Fig. 27). The main problems encountered during initial development include difficulty drilling deformed shale masses with significant gas pressure and identifying ways to stimulate an intensely faulted and fractured formation. Thus, realizing this giant resource potential will require the identification and application of procedures and technology for drilling and completion that are tailored to the development of intensely deformed shale masses in thrust belts. For example, underbalanced drilling techniques, such as air drilling, minimize the invasion of fines and the reaction of fluid with carbonate and clay in the reservoir, yet zones of high gas pressure can make underbalanced wells difficult to control. In addition, experience in other formations

88

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

Table 4 Basic reservoir properties of Conasauga shale as determined by core analysis. Well

Depth (m)

Effective porosity (%)

Gas-filled porosity (%)

Gas Saturation (%)

Water Saturation (%)

Oil saturation (%)

Permeability (μD)

Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Statistics n Mean Minimum Maximum Standard deviation

176 257 260 266 276 277 289 289 359 367 372 398

5.4 3.5 3.5 3.1 4.4 1.4 1.8 2.6 4.1 4.3 4.6 4.0

2.5 1.5 1.2 1.1 1.5 1.1 1.7 1.8 3.8 3.7 4.3 3.7

45.8 43.7 33.5 34.1 32.7 78.9 92.9 72.3 92.0 86.4 91.8 93.5

52.6 53.8 64.0 63.1 65.3 17.7 2.2 24.3 5.9 11.6 7.1 4.2

1.6 2.5 2.5 2.8 2.0 3.4 4.9 3.4 2.1 2.0 1.0 2.2

0.180 0.147 0.107 0.125 0.133 0.137 0.134 0.152 0.158 0.112 0.105 0.110

12 3.6 1.4 5.4 1.1

12 2.3 1.1 4.3 1.2

12 66.5 32.7 93.5 25.0

12 31.0 2.2 65.3 25.2

12 2.5 1.0 4.9 1.0

12 0.133 0.105 0.180 0.022

indicates that faults and major fracture zones may be conduits for major leak-off of stimulation fluid, thus limiting the effectiveness of hydrofracturing operations in shale (Bowker, 2007). The challenges posed by the Conasauga mushwads are almost certainly not unique, and meeting the development challenges posed by deformed shale masses in thrust belts has the potential to substantially increase domestic and global natural gas reserves. 10. Summary and conclusions Development of natural gas resources in Cambrian Conasauga shale has been affected by uncertainty about best practices for reservoir evaluation, exploration, and completion. To address this uncertainty, an integrated, multidisciplinary approach was undertaken. Key geologic variables considered include stratigraphy, sedimentation, structure, hydrodynamics, geothermics, petrology, geochemistry, gas storage, permeability, and resource base. Many characteristics of Conasauga reservoirs were established in the original depositional environment. The shale was deposited in an extensional tectonic setting during the late stages of Iapetan rifting. Sedimentation occurred in an intrashelf basin and was intimately associated with progradation of a large carbonate ramp. As such, the shale represents a transition from a euxinic basin to a

shoal rim where deposition was influenced by gravity, storm, and tidal processes, as well as organic productivity. The result of these processes was a complex stratigraphic architecture that gave rise to heterogeneous facies and reservoir quality. Geologic structure affects the geometry, continuity, and permeability of shale gas reservoirs. Thrust belt structures in Alabama include giant antiformal ductile duplexes, i.e. mushwads, in weak Cambrian Conasauga shale and ramp-flat structure in younger strata, which were translated cratonward along with a thick succession of stiff Cambrian–Ordovician Knox carbonate rocks. Structural deformation and tectonic thickening in the Conasauga mushwads resulted in shale masses thicker than 4000 m (13,000 ft). Fracture networks, including joints and shear zones, are common in the shale and appear to form important hydraulic conduits. The fracture networks are typically cemented with calcite. Petrologic analysis indicates that most fractures are synkinematic and hosted fluids containing carbon derived by dissolution of marine carbonate. Significant reservoir pressure exists in the Conasauga mushwads and is apparently relict gas pressure related to hydrocarbon generation. By contrast, fluid chemistry and pressure in Knox aquifers preserved in the roof and flanks of the mushwads are determined by meteoric recharge along Appalachian frontal structures. Reservoir temperature and geothermal gradient within the mushwads are low, and petrologic evidence indicates that the shale reservoirs were substantially warmer in the geologic

Table 5 Summary results of adsorption isotherm analysis in the Conasauga Formation.

1.0

Depth (m)

TOC (%)

VL (g/scc)

VP (MPa)

Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Dawson 34-3-1 Statistics n Mean Minimum Maximum Standard deviation

176 257 260 266 276 277 289 289 359 367 372 398

0.3 0.3 0.2 0.2 0.2 1.8 0.6 0.6 0.4 0.2 0.4 0.3

0.37 0.35 0.29 0.32 0.37 0.89 0.58 0.58 0.49 0.33 0.48 0.35

3.83 3.69 3.95 4.24 4.07 4.58 4.62 4.42 4.11 3.63 3.79 4.31

12 0.5 0.2 1.8 0.4

12 0.45 0.29 0.89 0.16

12 4.10 3.63 4.62 0.32

VL = Langmuir volume; VP = Langmuir pressure.

Langmuir volume (scc/g)

0.9 Well name

0.8

n = 12 y = 0.36x + 0.28 r = 0.96

0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

TOC content (%) Fig. 28. Cross-plot of Langmuir volume vs. TOC content in Conasauga shale, Dawson 34-03-01 core, Big Canoe Creek Field.

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

source material toward thermally mature kerogen that has exhausted most if not all of its generative potential. Although the generative potential is largely exhausted, the shale is capable of storing large volumes of natural gas. Shale is a dual-porosity reservoir in which some gas is stored in a free state, and some is adsorbed on organic matter and minerals with high surface area. Effective porosity averages 3.6%, and about 67% of this pore volume is capable of storing free gas. The shale is weakly sorptive because of low TOC content; Langmuir volume ranges from 0.29 to 0.89 scc/g. Adsorption of gas on mineral surfaces is anomalously high in the Conasauga and appears to reflect increased surface area associated with the open clay fabric. Permeability of the shale averages 0.133 μD, which is typical of shale reservoirs and indicates that natural or induced fractures are required to support commercial flow rates. Volumetric analysis indicates that the Conasauga mushwads contain OGIP of about 17.7 Tcm (625 Tcf) and that more than 80% of this gas is stored in a free state. Free gas is mobile across a spectrum of pressure-temperature conditions, whereas sorbed gas has greatest mobility at low reservoir pressure. Hence, Conasauga exploration should perhaps concentrate on deep reservoirs, where the free gas fraction is highly concentrated. The Conasauga clearly represents a major target for exploration and development, although determining best practices for drilling and completion remains a significant challenge that must be overcome to realize the economic potential of deformed shale masses. Meeting these challenges, however, can result in a major expansion of domestic and global natural gas reserves.

Table 6 Estimated natural gas resources in the Conasauga mushwads of the southern Appalachian thrust belt in Alabama. Unit

Metric

Area Free gas Sorbed gas Total Gas Total Gas 10% recovery 20% recovery

4411 14,831 2864 17,695 17.7 1.8 3.5

English km2 Bcm Bcm Bcm Tcm Tcm Tcm

1,703 523,827 101,165 624,992 625.0 62.5 125.0

89

mi2 Bcf Bcf Bcf Tcf Tcf Tcf

past than they are today. Indeed, the mushwads apparently lie entirely within the thermogenic gas window, and plotting thermal maturity versus depth indicates that maturation was largely post-kinematic. Analysis of gas composition, moreover, indicates a thermogenic origin involving the thermal cracking of hydrocarbons to methane. Detrital minerals are dominated by clay, carbonate, quartz, and feldspar. Authigenic minerals include pyrite, calcite, dolomite, silica, illite, and apatite. Weak alignment of platy clay minerals in Conasauga shale is perhaps a product of overpressuring related to tectonic deformation and hydrocarbon generation. Organic matter is dominated by matrix bituminite and includes minor amounts of liptinite, inertinite, and herbaceous kerogen resembling vitrinite. Rock-eval pyrolysis indicates that Conasauga reservoirs are dominated by type IV kerogen. Geochemical evidence points toward an evolution from sapropelic

EXPLANATION > 1000 Bcf/mi2 (10.9 Bcm/km2) 750-1000 (8.2-10.9) 500-750 (5.5-8.2) 250-500 (2.7-5.5) 0-250 (0-2.7) 0

Etowah Blount

EN

Contour interval = 250 Bcf/mi2

D DS

A

LE

GA

LM

B ES

SE

M

ER

PA

Jefferson

ER

D

St. Clair

Shelby Tuscaloosa Bibb 10 10

0 0

20

10 10

30 km 20 mi

Fig. 29. Map of estimated OGIP in the Conasauga mushwads of the southern Appalachian thrust belt.

90

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91

Acknowledgments Funding for this project is provided by the Research Partnership to Secure Energy for America (RPSEA) under contract 07122–17 through the “Ultra-Deepwater and Unconventional Natural Gas and Other Petroleum Resources” program authorized by the U.S. Energy Policy Act of 2005. RPSEA (http://www.rpsea.org) is a nonprofit corporation whose mission is to provide a stewardship role in ensuring the focused research, development and deployment of safe and environmentally responsible technology that can effectively deliver hydrocarbons from domestic resources to the citizens of the United States. RPSEA, operating as a consortium of premier U.S. energy research universities, industry, and independent research organizations, manages the program under a contract with the U.S. Department of Energy's National Energy Technology Laboratory. Brian Cardott and an anonymous reviewer made suggestions that substantially improved the quality of this contribution. The project team at the Geological Survey of Alabama would like to thank Keith Greaves of Terra Tek, who ran adsorption isotherms and determined basic reservoir properties of core samples. We would also like to thank Albert Maende of Weatherford Laboratories for donating source rock data and gas composition data to this project. Stable isotopic analysis of calcite was performed by Paul Aharon and Joe Lambert of the University of Alabama. References Aigner, T., 1985. Storm Depositional Systems. Lecture Notes in Earth Sci., 3. Springer-Verlag, Berlin (174 pp.). Astini, R.A., Thomas, W.A., Osborne, W.E., 2000. Sedimentology of the Conasauga Formation and equivalent units, Appalachian thrust belt in Alabama. In: Osborne, W.E., Thomas, W.A., Astini, R.A. (Eds.), The Conasauga Formation and equivalent units in the Appalachian thrust belt in Alabama. Alabama Geol. Soc. 31st Ann. Field Trip Guidebook, pp. 41–71. Bowker, K.A., 2007. Barnett Shale gas production, Fort Worth basin: issues and discussion. AAPG Bulletin 91, 525–533. Burchard, B., Lewan, M.D., 1990. Reflectance of vitrinite-like macerals as a thermal maturity index for Cambrian–Ordovician Alum Shale, southern Scandinavia. AAPG Bulletin 74, 394–406. Butts, C., 1910. Description of the Birmingham Quadrangle, Alabama: U.S. Geol. Surv. Geol. Atlas, Folio, 175. 24 pp. Butts, C., 1926. The Paleozoic rocks. In: Adams, G.I., Butts, C., Stephenson, L.W., Cooke, C.W. (Eds.), Geology of Alabama: Alabama Geol. Surv. Spec. Rept., 14, pp. 40–230. Butts, C., 1927. Bessemer–Vandiver folio: U.S. Geol. Surv. Geol. Atlas, Folio, 221 (22 pp.). Carroll, R.E., Pashin, J.C., Kugler, R.L., 1995. Burial history and source-rock characteristics of Upper Devonian through Pennsylvanian strata, Black Warrior Basin, Alabama: Alabama Geol. Surv. Circ., 187 (29 pp.). Cook, B.S., Thomas, W.A., 2010. Ductile duplexes as potential natural gas plays: an example from the Appalachian thrust belt in Georgia, USA. Geological Society, London, Special Publications 348, 57–70. Debajyoti, P., Skrzypek, G., 2007. Assessment of carbonate-phosphoric acid analytical technique performed using GasBench II in continuous flow isotope ratio mass spectrometry. International Journal of Mass Spectrometry 262, 180–186. Engelder, T., Lash, G.G., Uzcátegui, R.S., 2009. Joint sets that enhanced production from Middle and Upper Devonian gas shales of the Appalachian basin. AAPG Bulletin 95, 857–889. Fortey, R.A., Theron, J., 1994. A new Ordovician arthropod Soomaspis and the agnostid problem. Palaeontology 37, 841–861. Friedman, I., O'Neil, J.R., 1977. Compilation of stable isotope fractionation factors of geochemical interest, In: Fleischer, M. (Ed.), Data of Geochemistry, 6th ed. U.S. Geol. Surv. Prof. Paper 440-KK (12 pp.). Gale, J.F.W., Reed, R.M., Holder, J., 2007. Natural fractures in the Barnett Shale and their importance for hydraulic fracture treatments. AAPG Bulletin 91, 603–622. Hill, R.J., Jarvie, D.M. (Eds.), 2007. Barnett Shale: AAPG Bull., 91, pp. 399–622. Kidd, J.T., Neathery, T.L., 1976. Correlation between Cambrian rocks of the southern Appalachian geosyncline and the interior low plateaus. Geology 4, 767–769. Kuuskraa, V.A., Wicks, D.E., Thurber, J.L., 1992. Geologic and reservoir mechanisms controlling gas recovery from the antrim shale: SPE Paper, 24883, pp. 209–224. Kvale, E.P., Archer, A.W., Johnson, H.R., 1989. Daily, monthly, and yearly tidal cycles within laminated siltstones of the Mansfield Formation (Pennsylvanian) of Indiana. Geology 17, 365–368. Laubach, S.E., Marrett, R.A., Olson, J.E., Scott, A.R., 1998. Characteristics and origins of coal cleat: a review. International Journal of Coal Geology 35, 175–207. Laubach, S.E., Reed, R.M., Olson, J.E., Lander, R.H., Bonnell, L.M., 2004. Coevolution of crack-seal texture and fracture porosity in sedimentary rocks: cathodoluminescence observations of regional fractures. Journal of Structural Geology 26, 967–982. Lorant, F., Prinzhover, A., Behar, F., Huc, A.Y., 1998. Carbon isotopic and molecular constraints on the formation and the expulsion of thermogenic hydrocarbon gases. Chemical Geology 147, 249–264.

Maher, C.A., 2002. Structural deformation of the southern Appalachians in the vicinity of the Wills Valley anticline, northeast Alabama. Univ. Alabama, unpub. M.S. thesis, 45 pp. Markello, J.R., Read, J.F., 1981. Carbonate ramp to deeper shelf transitions of an Upper Cambrian intrabasin shelf, Nolichucky Formation, southwest Virginia Appalachians. Sedimentology 28, 573–597. Markello, J.R., Read, J.F., 1982. Upper Cambrian intrashelf basin, Nolichucky Formation, southwest Virginia Appalachians. AAPG Bulletin 28, 860–878. Martin, J.P., 2009. Marcellus Shale natural gas: the resources: New York State Energy Research and Development Authority, unpub. PDF, 19 pp. Mittenthal, M.D., Harry, D.L., 2004. Seismic interpretation and structural validation of the southern Appalachian fold and thrust belt, northwest Georgia: Georgia Geol. Soc. Guidebook, 42, pp. 1–12. Montgomery, S.L., Jarvie, D.M., Bowker, K.A., Pollastro, R.M., 2005. Mississippian Barnett Shale, Fort Worth Basin, north-central Texas: gas shale play with multi-trillion cubic foot potential. AAPG Bulletin 89, 155–175. Müller, K.J., Walossek, D., 1987. Morphology, ontogeny and life habit of Agnostus pisiformis from the Upper Cambrian of Sweden. Fossils and Strata 19, 1–124. Nelson, P.H., 2009. Pore-throat sizes in sandsones, tight sandstones, and shales. AAPG Bulletin 93, 329–340. Nuttall, B.C., Drahovzal, J.A., Eble, C.F., Bustin, R.M., 2009. Regional assessment of organic-rich gas shales for carbon sequestration: an example from the Devonian shales of the Illinois and Appalachian basins, Kentucky: AAPG Studies in Geol., 59, pp. 173–190. Olson, J.E., Laubach, S.E., Lander, R.H., 2009. Natural fracture characterization in tight gas sandstones: integrating mechanics and diagenesis. AAPG Bulletin 93, 1535–1549. Ortiz, I., Weller, T.F., Anthony, R.V., Frank, J., Linz, D., Nakles, D., 1993. Disposal of produced waters: underground injection option in the Black Warrior Basin. 1993 Int. Coalbed Methane Symposium Proc, pp. 339–364. Osborne, W.E., Szabo, M.W., Neathery, T.L., Copeland, C.W., Jr., 1988. Geologic map of Alabama: Alabama Geol. Surv. Special Map 220. Osborne, W.E., Thomas, W.A., Astini, R.A., Irvin, G.D., 2000. Stratigraphy of the Conasauga Formation and equivalent units, Appalachian thrust belt in Alabama. In: Osborne, W.E., Thomas, W.A., Astini, R.A. (Eds.), The Conasauga Formation and equivalent units in the Appalachian thrust belt in Alabama. Alabama Geol. Soc. 31st Annual Field Trip Guidebook, pp. 1–17. Pashin, J.C., 2008. Gas shale potential of Alabama. Univ Alabama, 2008 Int. Coalbed & Shale Gas Symposium Proc., Paper 0808 (13 pp.). Pashin, J.C., 2009. Shale gas plays of the southern Appalachian thrust belt. Univ Alabama, 2009 Int. Coalbed & Shale Gas Symposium Proc., Paper 0907 (14 pp.). Pashin, J.C., 2010. Variable gas saturation in coalbed methane reservoirs of the Black Warrior basin: implications for exploration and production. International Journal of Coal Geology 82, 135–146, http://dx.doi.org/10.1016/j.coal.2009.10.017. Pashin, J.C., Groshong Jr., R.H., 1998. Structural control of coalbed methane production in Alabama. International Journal of Coal Geology 38, 89–113. Pashin, J.C., Carroll, R.E., McIntyre, M.R., Grace, R.L.B., 2010a. Geology of unconventional gas plays in the southern Appalachian thrust belt. SEPM Guidebook, Field Trip 7, AAPG Ann. Conference and Exposition, New Orleans, Louisiana (86 pp.). Pashin, J.C., Grace, R.L.B., Kopaska-Merkel, D.C., 2010b. Devonian shale plays in the Black Warrior basin and Appalachian thrust belt of Alabama. Univ Alabama, 2010 Int. Coalbed & Shale Gas Symposium Proc., Paper 1016 (20 pp.). Pashin, J.C., Kopaska-Merkel, D.C., Arnold, A.C., McIntyre, M.R., 2011. Geological foundation for production of natural gas from diverse shale formations. Research Partnership to Secure Energy for America (RPSEA) Final Report 07122.17.01 (156 pp.). Pitman, J.K., Pashin, J.C., Hatch, J.R., Goldhaber, M.B., 2003. Origin of minerals in joint and cleat systems of the Pottsville Formation, Black Warrior basin, Alabama: implications for coalbed methane generation and production. AAPG Bulletin 87, 713–731. Raymond, D.E., 1991. New subsurface information on Paleozoic stratigraphy of the Alabama fold and thrust belt and the Black Warrior basin: Alabama Geol. Surv. Bull., 143 (185 pp.). Rodgers, J., 1950. Mechanics of Appalachian folding as illustrated by the Sequatchie anticline, Tennessee and Alabama. AAPG Bulletin 34, 672–681. Ross, D.J.K., Bustin, R.M., 2008. Characterizing the shale gas resource potential of Devonian– Mississippian strata in the Western Canada sedimentary basin: application of an integrated formation evaluation. AAPG Bulletin 92, 87–125. Seidle, J., O'Connor, L., 2011. Well performance & economics of selected U.S. shales. Soc. Petrol. Eval. Eng. Ann. Meeting Proc (31 pp.). Slatt, R.M., O'Brien, N.R., 2011. Pore types in the Barnett and Woodford gas shales: contribution to understanding gas storage and migration pathways in fine-grained rocks. AAPG Bulletin 95, 2017–2030. Soeder, D.J., 1988. Porosity and permeability of eastern devonian gas shale. SPE Formation Evaluation, pp. 116–124 (March, 1988). SPEE, 2010. Guidelines for the practical evaluation of undeveloped reserves in resource plays: Soc. Petrol. Eval. Eng. Monograph 3. (83pp.). Srinivasan, K., Walker, K.R., 1993. Sequence stratigraphy of an intrashelf basin carbonate ramp to rimmed platform transition: Maryville Limestone (Middle Cambrian), southern Appalachians. GSA Bulletin 105, 883–896. Telle, W.R., Thompson, D.A., Lottman, L.K., Malone, P.G., 1987. Preliminary burial-thermal history investigations of the Black Warrior basin: implications for coalbed methane and conventional hydrocarbon development. Univ. Alabama, 1987 Coalbed Methane Symposium Proc, pp. 37–50. Thomas, W.A., 1985. Northern Alabama sections. In: Woodward, N.B. (Ed.), Valley and Ridge Thrust Belt: Balanced Structural Sections, Pennsylvania to Alabama: Univ. Tennessee Dept. Geol. Sci. Studies in Geol., 12, pp. 54–60. Thomas, W.A., 2001. Mushwad: Ductile duplex in the Appalachian thrust belt in Alabama. AAPG Bulletin 85, 1847–1869.

J.C. Pashin et al. / International Journal of Coal Geology 103 (2012) 70–91 Thomas, W.A., 2007a. Balancing tectonic shortening in contrasting deformation styles through a mechanically heterogeneous stratigraphic succession: GSA Special Paper, 433, pp. 277–290. Thomas, 2007b. Role of the Birmingham basement fault in thin-skinned thrusting of the Birmingham anticlinorium, Appalachian thrust belt in Alabama. American Journal of Science 307, 46–62. Thomas, W.A., Bayona, G., 2005. The Appalachian thrust belt in Alabama and Georgia: thrust-belt structure, basement structure, and palinspastic reconstruction: Alabama Geol. Surv. Monograph, 16 (48 pp.). Thomas, W.A., Pashin, J.C., 2011. Of Mushwads and Mayhem: Disharmonically Deformed Shale in the Southern Appalachian Thrust Belt. Alabama Geol. Soc. 48th Ann. Field Trip Guidebook (226 pp.).

91

Thomas, W.A., Astini, R.A., Osborne, W.E., Bayona, G., 2000. Tectonic framework of deposition of the Conasauga Formation. In: Osborne, W.E., Thomas, W.A., Astini, R.A. (Eds.), The Conasauga Formation and equivalent units in the Appalachian thrust belt in Alabama. Alabama Geol. Soc. 31st Ann. Field Trip Guidebook, pp. 19–40. Wignall, P.B., Newton, R., 1998. Pyrite framboid diameter as a measure of oxygen deficiency in ancient mudrocks. American Journal of Science 298, 537–552. Williams, P., 2007. Conasauga Saga. Oil and Gas Investor. (Sept. 6, 2007, 4 pp.).