Sequence Stratigraphy and Provenance of the Bakken Formation in Southeast Alberta and Southwest Saskatchewan
University of Calgary PRISM: University of Calgary's Digital Repository
Graduate Studies The Vault: Electronic Theses and Dissertations
2015-05-04 Sequence Stratigraphy and Provenance of the Bakken Formation in Southeast Alberta and Southwest Saskatchewan
Mohamed, Tarig Ibrahim
Mohamed, T. I. (2015). Sequence Stratigraphy and Provenance of the Bakken Formation in Southeast Alberta and Southwest Saskatchewan (Unpublished master's thesis). University of Calgary, Calgary, AB. doi:10.11575/PRISM/27758 http://hdl.handle.net/11023/2241 master thesis
University of Calgary graduate students retain copyright ownership and moral rights for their thesis. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission. Downloaded from PRISM: https://prism.ucalgary.ca UNIVERSITY OF CALGARY
Sequence Stratigraphy and Provenance of the Bakken Formation in
Southeast Alberta and Southwest Saskatchewan
by
TarigMohamed Ibrahim
A THESIS
SUBMITTED TO THE FACULTY OF GRADUATE STUDIES IN PARTIAL
FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER SOFCIENCE
GRADUATE PROGRAM IN GEOLOGY AND GEOPHYSICS
CALGARY, ALBERTA
APRIL, 2015
© Tarig Ibrahim Mohamed 2015 Abstract: Three transgressive-regressive systems tracts are interpreted in the Bakken Formation in southeastern Alberta and southwestern Saskatchewan based on analysis of cores and well logs. These systems tracts include a basal transgressive systems tract TST1, a regressive systems tract RST and a transgressive systems tract TST2. Three distinct systems tracts are interpreted within Bakken RST; these are highstand systems tract (HST), Falling stage systems tract (FSST) interpreted within the Bakken Formation and a lowstand systems tract within the Exshaw Formation to the west. Fluctuations in base level that resulted from changes in relative sea level and tectonics controlled the deposition of the Bakken Formation. Detrital zircon geochronology and reworked conodonts from the Bakken middle member indicate sediment transport from multiple sources. Thrust-belt sources associated with the Franklinian Orogeny to the north and northeast contributed the majority (no less than 25%) of the Bakken clastic budget. The two mechanisms for Bakken sediments delivery are continental- scale fluvial systems and longshore drift of sediments.
ii ACKNOWLEDGEMENT
This project was financially supported by a Natural Science and Engineering Research Council of Canada (NSERC) Discovery grant held by Dr. Charles M. Henderson. Conocophillips Canada provided financial support to the candidate (Tarig Ibrahim) through “Conocophillips Scholarship for applied Basin Studies). I would like to acknowledge and thank my supervisor Dr. Charles Henderson for the countless hours of discussions, feedback, guidance and support. The time and effort Dr. Charles
Henderson put into reviewing several of my “wordy” drafts, identifying conodonts and re- drafting figures are much appreciated. I thank my committee Dr. Benoit Beaucha mp, Dr. Per Pederson and Dr. Yvonne Martin for their in-depth reviews and valuable comments. Dr. Pederson provided valuable feedback and comments during several stages of this study. My office mates and colleagues Chad Morgan, Elinda Dehari and David Terrill provided ideas and suggestions during our weekly lab meetings. I would like to thank Paul Starr for reviewing and editing the draft of this thesis, and Makram Hedhli for the valuable discussions. My “landlord and roommate” Dr. Alan Hildebrand provided support and acted as a pseudo- family -when it was much needed- throughout the time I spent in graduate school. I thank my sister El- Mahadia Ibrahim for her endless support; without her guidance I am sure my life course would’ve been significantly different, I highly doubt I would’ve ended up in Calgary if not for her help and guidance. This work is dedicated to my mother Ihsan Al-Zain whom her support is beyond measure.
iii Table of Contents Approval Page...... I Abstract...... II Acknowledgements ...... III Table of Contents ...... IV List of Figures...... VI List of Tables...... VII Appendices...... VII
CHAPTER 1: Introduction 1. Introduction...... 1 2. Study Area…...... 2 3. Geologic Background ...... 5 3.1. Stratigraphy ...... 5 3.2. Tectonic and Paleogeographic setting...... 8 3.3. Biostratigraphy of the Bakken Formation ...... 12 3.4. Provenance of the Bakken Formation Clastics ...... 12 4. Methods...... 13 4.1. Core Description...... 13 4.2. Detrital Zircon Geochronology...... 13 4.3. Conodont Samples Processing...... 14 4.4. Scanning Electron Microscopy...... 15 4.5. Subsurface Cross Sections...... 15 5. Role of Student...... 16 6. Depositional Environments and Sequence Stratigraphic Terminologies...... 17 6.1. Facies and depositional systems terminologies...... 17 6.2. Sequence stratigraphic terminologies...... 18 CHAPTER 2: Sequence Stratigraphy and Provenance of the Bakken Formation in Southeast Alberta and Southwest Saskatchewan Extended abstract ...... 23 1. Introduction...... 25 2. Geologic Setting...... 29 3. Facies and Sequence Stratigraphy...... 33 3.1. Introduction...... 33 3.2. Methods...... 36 3.3. Lithofacies...... 37 3.4. Interpretation and Discussion...... 40 Lower and upper Bakken shale...... 40 Bioturbated siltstone Mb1...... 42 Fossiliferous wackestone Mb2 ...... 42 Rippled Laminated Sandstones of Mb3 and Mb4...... 46 Tidal deposits of Mb5...... 48 Fossiliferous wackestone Mb6...... 50
iv 4. Sequence Stratigraphy...... 51 4.1. Sequence 1...... 53 4.1.1. Sequence Boundary (SB 1) and Maximum Regressive Surface (MRS1)...... 53 4.1.2. Lower Transgressive Systems Tract TST 1 ...... 55 4.1.3. Maximum Flooding Surface MFS 1...... 55 4.1.4. High Stand Systems Tract HST...... 56 4.1.5. Regressive Surface of Marine Erosion (RSME) and Basal Surface of Forced Regression (BSFR)...... 56 4.1.6. Falling Stage Systems Tract...... 57 4.1.6. Lowstand Systems Tract...... 57 4.2. Sequence 2...... 63 4.2.1. Sequence Boundary (SB 2) and Maximum Regressive Surface (MRS 2) ...... 63 4.2.2. Upper Transgressive Systems Tract TST 2…...... 64 4.3. Conclusions, Discussion and Implications...... 65 5. Provenance of the Bakken Clastic Deposits ...... 70 5.1. Introduction...... 70 5.2. Samples and Methods 5.2.1. Samples...... 71 5.2.2. U-Pb Geochronology...... 71 Method...... 71 Results...... 73 Bk1...... 74 Bk2...... 74 Bk3...... 76 Ex1...... 76 Interpretation...... 78 Sources of Detrital Zircons...... 78 5.2.3. Conodont Colour Alteration Index CAI Analysis...... 82 Analysis...... 82 Results...... 83 Interpretation...... 84 5.3. Discussion...... 87 5.3.1. Significance of Young Grain Ages (370 to 700 Ma) and High CAI Conodont Elements ...... 87 5.3.2. Implications for Sediment Transportation and Palegeography during the late Devonian-Earliest Mississippian...... 87 5.3.3. Implications for Depositional Style of the WCSB during the Devonian- Mississippian ...... 90 6. Conclusions...... 90
CHAPTER 3: Summary and Future work 3.1 1.Conclusions ...... 94 3.2. Future Work...... 96
References...... 99
v List of Figures
CHAPTER 1 Figure 1a: Regional extent of upper Devonian- Mississippian strata in WCSB ...... 3 Figure 1b: Location of study area and cores and well logs used in the study ...... 4 Figure 2: Informal stratigraphy of the Bakken Formation ...... 7 Figure 3: Middle Devonian to early Mississippian Paleogeography ...... 9 Figure 4: Williston Basin location and Devonian tectonic elements ...... 11
CHAPTER 2 Figure 1: Location of study area and cores and well logs used in the study ...... 28 Figure 2: Williston Basin location and Devonian tectonic elements ...... 31 Figure 3: Generalized Bakken core log ...... 35 Figure 4: Core photographsr of lowe part of the Bakken Formation well 10-03-032-28W3 ...... 41 Figure 5: Core photographs of lithofacies Mb3 and Mb4 ...... 44 Figure 6: Core log showing the sharp based shoreface deposits of lithofacies Mb3 and Mb4 ...... 45 Figure 7: Core photographs of upper part of the Bakken Formation well 10-03-032-28W3 ...... 47 Figure 8: Depositional model of the Bakken Formation ...... 49 Figure 9: Sequence stratigraphic cross section of the Bakken Formation ...... 52 Figure 10: Core photograph of lower and middle members of the Bakken Formation in well 111/4-06-032-27W3 ...... 54 Figure 11: Core data of well 111/06-33-028-2W4 two parasequences within shoreface and foreshore deposits of Mb4 ...... 59 Figure 12: High resoultion subsurface correlation of the FSST deposits of the Bakken Formation ...... 60 Figure 13: Stratigraphic section of uppermost Palliser Formation, Exshaw Formation and lower Banff Formation at Goat Creek, Rundle range ...... 62 Figure 14: Core photograph of middle and upper members of the Bakken Formation in well 111/10-03-032-28W3 showing position of sequence stratigraphic surfaces and systems tracts ...... 64 Figure 15: Schematic cross section depicting the sequence stratigraphic surfaces and systems tracts present in the Bakken Formation and their basinal continuation westward ...... 66
vi Figure 16: Map of study area with sample location ...... 72 Figure 17: Detrital zircon probability plot for Bakken Formation samples ...... 75 Figure 18: Detrital zircon probability plot for Exshaw Formation sample ...... 77 Figure 19: Provenance Interpretation of the Bakken Formation ...... 79 Figure 20a: Conodonts from the Bakken Formation ...... 84 Figure 20b: Conodonts CAI ...... 85 List of Tables Table 1: Sedimentary Facies Description Summary and Depositional Environments ...... 38 Table 2: Detrital zircon sample information ...... 71
Appendices:
Appendix 1: Lithologs with sample locations ...... 106 Appendix 2: U-Pb ages of Detrital Zircon grains from the Bakken Formation (Bk1, Bk2 and Bk3) and Exshaw Formation (Ex1) ...... 118 Appendix 3: Cross sections ...... 121 Appendix 4: Isopach maps of the Bakken lithofacies ...... 124
vii 1
CHAPTER 1
INTRODUCTION
1- Introduction
This thesis investigates the depositional history, sequence stratigraphy and the provenance of the Devonian-Mississippian Bakken Formation within southeastern Alberta and southwestern
Saskatchewan. Deposition of Bakken deposits took place in a shallow marine setting (the
Devonian- Mississippian seaway) where several depositional environments coexisted resulting in deposition of various lithofacies.
Sedimentary core logging and well log analysis allowed identification of the Bakken sedimentary facies. Identification and interpretation of depositional environments of Bakken lithofacies provided a method to interpret the influence of sea level fluctuations and sediment supply and to establish a sequence-stratigraphic framework for the Bakken Formation. Furthermore, conodont biostratigraphy, colour alteration index and preservational observations of detrital conodonts, as well as detrital zircon geochronology data constrained the interpretation of provenance and transportation mechanisms of the Bakken clastic deposits. This thesis is organized in a paper format; this paper will be submitted to the American Association of
Petroleum Geologists AAPG Bulletin.
The first part of this paper outlines the depositional environments and sequence stratigraphy of the Bakken Formation in southeastern Alberta and southwestern Saskatchewan. Bakken facies generally suggest a shallow marine setting where several depositional environments occupied different parts of the basin simultaneously; fluctuations in sea levels resulted in lateral migration of these depositional environments. Eustasy and fluctuations in sediment supply resulted in formation of distinct system tracts that are bounded by stratigraphic surfaces. 2
The goal of the first part of the paper is to define and interpret the depositional environments
of the Bakken Formation, as well as the stratigraphic surfaces and system tracts and to
construct a sequence stratigraphic framework that illustrates the Bakken depositional history.
Constraining the provenance and outlining dispersal systems of the Bakken clastics are the key
scope of this study, and are discussed in the second part of this paper. Significant amount of
clastic deposits are present in the Bakken middle member sandy and silty lithofacies. These
clastic deposits are located in a basin predominantly occupied by carbonate units. This
abnormal and abrupt occurrence of clastics can be explained by constraining the provenance and dispersal systems of the clastic deposits. The increased clastic supply observed in the
Bakken Formation suggests a major tectonic episode that created, uplifted and exhumed
highlands supplying clastic sediments to the Bakken. This study investigates the provenance of
Bakken clastics using a novel approach that incorporates detrital zircon geochronology data
and colour alteration index (CAI) as well as preservational observations of detrital conodont elements. This paper presents the first “published” detrital zircon ages and provenance
interpretation for the Bakken Formation in the Williston Basin.
2- Study area
This study used 11 cores and 150 well logs from the Bakken Formation in the subsurface of the
Williston Basin in southwestern Saskatchewan and southeastern Alberta. The part of the
Bakken Formation studied here represents the northwestern limit of the Bakken deposits. To the west of the study area, shallow marine deposits of the upper part of middle Bakken
Formation grades into siltstone deposits of the upper Exshaw member (Johnston et al., 2010)
(Fig. 1a). The study area covers approximately 2200 km2 (Fig. 1b) (townships 28 to 34 and 3
Fig. 1a. Showing regional extentr of uppe Devonian- lower Mississippian strata in Western Canada Sedimentary Basins. Grey box outlines the location of study area (slightly modified from Halabura et al., (2007) and references therein. 4
Fig. 1b. Map of study area showing wells with cores and with well logs used in this study. Dashed line marks the Alberta-Saskatchewan provincial boundary. 5
range 27 west of third meridian to the third range west of fourth meridian). The cores are
described and illustrated in detail in the next chapter and the supplementary appendices.
3- Geologic Background:
3.1- Stratigraphy:
In Western Canada Sedimentary Basin (WCSB), the Bakken Formation is part of the Upper
Devonian- Lower Mississippian Three Forks Group. This group consists of the Torquay, Big
Valley and Bakken formations (ascending order) and represents cyclical deposition of at least five 3rd order shallow marine transgressive-regressive T-R sequences (Halbertsma, 1994).
The Torquay Formation and its basinal correlative Stettler Formation consist of carbonate and evaporite rocks deposited in low-energy, lagoonal to subtidal depositional environments and represent an overall regressive sequence with "intermittent" transgressive episodes
(Halbertsma, 1994). The Big Valley Formation and its basinal equivalent, the Costigan Member of the Palliser Formation, were deposited following the end of Torquay deposition. The Big
Valley Formation and its equivalent consist of wackestones and packstones deposited in low energy marine settings (Richards, 1991). The end of Big Valley/ Palliser deposition was marked by a final regression that formed a basin-wide unconformity; this unconformity is clearly observed in Bakken cores as a sharp contact, rich in reworked phosphatic nodules, which is also
observed in Exshaw Formation outcrops in the eastern Rocky Mountain Front Ranges as a thin
layer of conglomeratic coarse sandstone with phosphatic nodules, pyrite and sphalerite
(Richards, 2002). Overlying this unconformity is the Bakken Formation and its basinal
equivalent, the Exshaw Formation. 6
The Bakken Formation is a thin (30 m av.), mixed clastic- carbonate unit that comprises three informal members, lower, middle and upper. The lower and upper members consist of black,
laminated, organic-rich shale and average a thickness of 5 and 3.5 m respectivly. The middle
member comprises fine sandstone, siltstone and wackestone and averages 21 m in thickness.
The deposition of the Bakken Formation was the result of the advancing transgression of the
late Devonian-Mississippian seaway (Meijer-Drees et al., 1996; Richards et al, 2002;
Maldonado, 2012). The initial transgression of the seaway allowed deposition of the fissile,
finely laminated and organic-rich lower Bakken member in outer shelf depositional
setting characterized by a stratified anoxic water column and very low intensity currents
(Christopher 1961; Richards et al., 2002). Conodont biostratigraphy suggests a late Devonian
age for the lower Bakken shale, correlative to the lower part of the lower shale member of
the Exshaw Formation in Alberta Foreland Basin (Johnston et al., 2010). The middle Bakken member consists of several lithologies that vary from clastic to carbonate rocks (Fig. 2). These lithologies include siltstone, horizontally bedded fine grained sandstone, cross stratified and trough bedded fine to medium grained sandstone, interbedded very fine grained sandstone and mudstone, wackestone and packstone; several subdivisions for the
Middle Bakken member have been proposed by previous workers (Fig. 2). The middle
Bakken Member represents an overall regressive systems tract that was deposited in shallow marine, inner ramp environments (Richards, 2002). The upper Bakken member consists of black, fissile and organic-rich shale that was deposited during marine transgression in an outter ramp setting (Smith and Bustin, 2000). The upper Bakken shale is correlative to the lower shale member of Banff Formation of the Alberta Basin (Johnston et al., 2010). 7 member member publishe d Bakke n from of middle n descriptio subunits compile d a nd lithological units of including thicknesses
, Formation, n (1991)) Bakke et al ., the of Le Fevre from stratigraphy
Informal 2. Fig. subunits (modifie d literature. 8
Thick carbonates of Lodgepole Formation and its basinal equivalent the upper member
of Banff Formation accumulated during a high sea level in an outer shelf to basinal depositional environments following the end of upper Bakken shale deposition
(Richards, 2002).
3.2- Tectonics and Paleogeographic setting:
During the Middle Devonian-Mississippian, three distinct orogenies were shaping the
Laurentian margins. These orogenies are the Middle Devonian to Mississippian Antler Orogeny
(Root, 2001), the Middle Devonian to Late Mississippian Acadian phase of the Appalachians
Orogeny (Faill, 1997) and the Ellesmerian phase of the Caledonian Orogeny (Patchett et al,
1999) (Fig. 3). These orogenies resulted in several episodes of uplift and subsidence within the margins as well as in the interior of the Laurentian Craton (Ross et al, 1991), subsequently
resulting in changes in base levels and rates of sediment supply (Fig. 3) (Patchett et al, 1999;
Anfinson et al, 2012, Gehrels and Picha, 2014).
The WCSB was an extensive, tectonically active and submerged continental margin, located 5° to 10° north of the Devonian-Mississippian equator during the Late Devonian-Mississippian
(Richards, 1989). Several basins existed within the Devonian-Mississippian continental margin concurrently (Kent, 1994). The distal, westernmost part of the margin underwent convergent deformation in response to the Middle Devonian-Mississippian Antler and Cariboo orogenies
(Fig. 3) (Roots, 2001). The Williston Basin occupied the relatively shallower eastern parts of the margin (Fig. 3); while the Elbow Sub-basin and Central Montana Trough occupied the central part of the margin in southwestern Saskatchewan, southeastern Alberta and Montana (Fig. 4)
(Kent and Christopher, 1994). 9
Fig.3. Depicts middle Devonian to early Mississippian paleogeography, orogenies, as well as the distribution of clastic sediments throughout North America during early to Middle Paleozoic (Cook and Bally, 1975), slightly modified from Patchett et al., (1999). 10
Reactivation of deep seated basement faults is believed to have controlled the subsidence rate within the Devonian passive margin (Greggs, 2006).
Deposition of the Bakken Formation took place throughout the Williston Basin of central North
America (Fig. 3). The Williston Basin is an ellipsoidal shaped basin (Fig. 3, 4) that covers most of
North Dakota, the northern parts of South Dakota and Montana and extends to cover the southern part of Saskatchewan, southeastern Alberta and southwestern Manitoba (Gerhard and Anderson, 1988). Tectonically, the Williston Basin is a stable intracratonic basin created along weak zones produced in the Precambrian basement following the Trans-Hudson Orogeny
(Gibson, 1995). The Sioux Arch borders the basin in North Dakota, South Dakota and southeastern Montana, the Punnichy Arch in Saskatchewan borders the northeastern edge of the basin and the Sweetgrass Arch of southeastern Alberta and northern Montana forms the western border of the Williston Basin and separates the Williston Basin from the Mesozoic
Alberta Foreland Basin to the east. The Swift Current Platform represents the stratigraphic manifestation of the eastern limits of the Sweetgrass Arch (Kent and Christopher, 1994) (Fig 4).
Initial subsidence and deposition within the basin started during the Cambrian with significant subsidence during the Ordovician, Silurian and Devonian that resulted in thick accumulations of carbonate deposits (Gerhard and Anderson, 1988). Thermal cooling and reactivation of fault blocks in the Precambrian Basement were likely responsible for subsidence of the Williston
Basin (Ahern and Mrkvicka, 1984). Movement along these fault blocks resulted in the formation of positive, dome-like basement structures; several of these structures were identified proximal to the Swift Current Platform in southern Saskatchewan (Kent and
Christopher, 1994). 11
Fig. 4. Devonian tectonic elements of the Williston Basin, the red rectangle indicates location of study area (slightly modified from Kent and Christopher, 1994). 12
3.3- Biostratigraphy of the Bakken Formation
Johnston et al. (2010) assigned an age of Lower to Upper expansa conodont zone (Late
Famennian) for the lower shale member within southeastern Alberta. Conodonts from the top of the Middle sandstone member and Upper Bakken shale suggest a Lower crenulata conodont zone (Early Mississippian). These findings resulted in placement of the Devonian-
Mississippian boundary somewhere within the Bakken middle member.
3.4- Provenance of the clastics of the Bakken Formation:
The Bakken Formation contains significant clastic deposits. Deposition of sand and silt- sized
clastic deposits within a carbonate dominated basin clearly indicates a fundamental change in
the depositional style in the basin.
Previous workers have proposed several sources for the Bakken clastics. Christopher (1961) suggested that Bakken clastics were likely sourced from the Canadian Shield east of the
Williston Basin based on petrography and proximity of the Canadian Shield. Toews (2005) suggested that Bakken clastics were sourced from the Antler highlands to the southwest. Nd isotope studies by Patchett et al. (1999) in the Exshaw and Banff formations (Bakken basinal equivalents) indicated that these deposits were likely sourced from the Franklinian Belt in the
Canadian Arctic. 13
4- Methods:
4.1- Core Description:
This study utilized eleven drill cores sampled from the subsurface of southeastern Alberta and
southwestern Saskatchewan. Cores obtained from Southeastern Alberta (four cores) were
examined at the Alberta Energy and Utilities Board Centre in Calgary Alberta; the other seven
cores are from the subsurface of Saskatchewan and were examined at the Saskatchewan
Subsurface Geological Laboratories in Regina, Saskatchewan. Cores were described for
lithology, sedimentary structures, fossils and stratigraphic surfaces. Core descriptions were
then integrated with gamma ray logs to interpret and correlate other well logs used in the
study. Approximately 150 well logs from different wells have been examined in this study; most
of these well logs were used to construct cross sections and subsurface isopach maps. Detrital zircon samples were collected for U-Pb geochronology from selected sand-rich intervals.
Conodont biostratigraphy samples were obtained from fossil-rich layers. Detailed core logs and
are presented in appendix section.
4.2- Detrital Zircon Geochronology:
Sample preparation and analysis were conducted at the Isotope Geology Laboratory at Boise
State University. Samples were crushed with a jaw crusher and zircon grains were separated
with water table and heavy liquid separation techniques. Separated zircon grains were annealed at 900˚C for 60 hours in a furnace. The zircon grains were then mounted in epoxy and polished to expose the interior of the grains. Detrital zircon ages were obtained using a
ThermoElectron X-Series II quadrupole Inductively Coupled Plasma Mass Spectrometer 14
(ICPMS) and New Wave Research UP-213 Nd: YAG UV (213 nm) laser ablation system. Zircon
grains were ablated with a 25 µm wide laser spot, using fluence and pulse rate of 5 J/cm2 and
10 Hz respectively, excavating a pit, 25 µm in depth, during a 45 second analysis (30 second
abalation, 15 second gas blank). A 1.2 L/min He gas stream carried the ablated material to the
nebulizer’s plasma flow. Background count rates were obtained prior to each spot analysis and
later subtracted from the raw count rate of each analysis. Instrumental fractionation of the
background-subtracted ratios was corrected and dates were calibrated with respect to
measurements of the Plešovice zircon standard (Sláma et al., 2008). A zircon secondary
reference was analyzed as an unknown in groups of two analyses every 20 unknown analyses to
assess instruments accuracy. Between 85 and 129 ages were recorded for each sample.
The U-Pb ages obtained were plotted in Isoplot 3.0 (Ludwig, 2003) to calculate weighted mean
dates. Standard calibration uncertainties for 207Pb/206Pb dates ranged from 0.46 to 0.64% (2σ),
and 3.1% to 3.5% (2σ) for 206Pb/238U dates. The data reported are 206Pb/238U ages for grains younger than 1000 Ma and 207Pb/206Pb ages for grains older than 1000 Ma. Errors on the dates
are given at 2σ.
4.3- Conodont Sample Processing:
Conodont samples have been processed using in-house analytical protocols and standard
materials at the micropaleontology lab, University of Calgary.
Rock samples were crushed into approximately 1 cm sized fragments using a jaw crusher. The crushed samples were then placed in plastic buckets and dissolved in 10% acetic acid solution for three days. Acidized samples were washed after three days and re-acidized for another 15 three days in 10% acetic acid. Dissolved samples were washed and sieved using no. 16 and no.
200 seive to separate the fine sand fraction (75 µm to 1.18 mm) that typically contains conodont grains. Coarse fractions were labeled and stored for future use, if needed. Fine sand residue was bleached in detergent, washed and then dried. Heavy liquid separation is the technique used to reduce the amount of fine-sand fraction that might contain condonts.
Tetrabromoethane (specific gravity of 2.85) was used to separate the heavy, conodont- bearing fraction from the fine-sand residue. Conodont elements, as well as any other significant material, were picked using a binocular microscope and mounted on a micro-slide.
Conodont fauna recovered were identified and CAI values were determined by directly comparing the conodont colour to a standard conodont CAI collection provided by Anita Harris of the US Geological Survey.
4.4- Scanning Electron Microscopy:
High magnification images were obtained using a scanning electron microscope (FEI XL30) at the Microscopy and Imaging Facility, University of Calgary. Specimens were mounted on a stub, and for each stub a corresponding map has been made to show the location of the specimen and the sample id number. Stubs were coated with gold-palladium and high resolution images were obtained. CorelDraw and CorelPhoto-Paint were used to produce conodont plates.
4.5- Subsurface Cross Sections:
Four high resolution cross sections were generated in this study. Each of these cross sections was developed in three stages. The first stage consisted of the integration of core descriptions and interpretations of core gamma ray logs; these logs were used later to 16
facilitate well log correlations throughout the study area. The second stage involved the
selection of good quality gamma ray logs to generate cross section; well logs were obtained
using Geoscout software. Well logs were correlated manually and cross sections were revised
and refined several times before the final process of digital representation. The graphic
illustration of cross sections included digitization of well logs and illustration of interpreted
stratigraphic surfaces and systems tracts. The final process was performed using CorelDraw.
Although not incorporated directly in this paper, isopach maps were generated for Bakken
lithofacies to observe any depositional patterns that could be attributed to the Bakken
depositional environments. Isopach maps did not provide significant information to the
scope of this paper, however, are presented in Appendices section.
5- Role of Student:
The paper presented in the following chapter is the product of research carried by the first author (candidate) and supervised by the second author. The first author conducted field work, sample preparation, analysis and interpretation. Cross sections, core plates and maps presented in this paper were produced by the first author. The second author, who is the candidate supervisor, provided guidance, discussions, valuable feedback and financial support throughout different stages of this research. The second author identified conodonts and edited several drafts of the paper as well as figures before submission of this thesis. 17
6‐ Depositional Environments and Sequence Stratigraphic Terminologies:
6.1- Facies and depositional systems terminologies:
The middle Bakken deposits are dominated by inner shelf, shallow marine strata . Stacking
patterns and sedimentary structures of these strata suggest deposition in a prograding
shoreface and marginal marine environments, as will be discussed in the following chapter.
This section will briefly define some of the re‐occurring terminologies of shallow marine
environments and sequence stratigraphic terminologies used in this study.
In this study, the shoreface is defined as the part of a shelf where sediments are deposited above fair‐weather wave base and below low tide level. Sediment transportation and deposition in shoreface settings are controlled and driven by breaking waves (Plint, 2010).
Shoaling waves generate landward, oscillatory and longshore currents that break along the shoreline in the surf zone (swash zone) (Clifton, 2006). The landward movement of waves is stronger than the seaward movement of waves, which results in deposition of coarser‐grained sediments in the upper parts of the shoreface (Plint, 2010). The seaward translation of water is demonstrated by rip currents that flow seaward and disperse sediments to few hundred meters seaward (Hunter et al., 1979). In this study, only lower and upper subdivisions of the shoreface profile are used. The distinction between lower and upper shoreface deposits in the Bakken
Formation is based on grain size and the sedimentary structures present in the rocks. Lower shoreface deposits are hetrolithic (silt to fine sand sized grains and often contain mud interbeds) and are typically rippled and wavy cross stratified; whereas upper shoreface deposits are relatively mud‐free, fine sandstone that demonstrate trough‐cross stratification and minor 18 low‐angle lamination (Plint, 2010). The foreshore setting occupies the intertidal zone (above the low tide level and below the high tide level). Foreshore deposits exhibit parallel laminations that slightly dipping seaward (Walker and Plint, 1992).
6.2- Sequence stratigraphic terminologies:
Sequence stratigraphy aims at subdividing the rock record into systems tracts within a chronological framework (Catuneanu et al., 2009). This is achieved by identification of key surfaces that bound the systems tracts, and develop as a product of interplay between changes in accommodation and sediment supply. Several sequence stratigraphic models are currently in use by workers (e.g. depositional sequences II, III and IV, genetic sequences and T‐R sequences).
The applicability of any of these models is highly dependent on several factors including basin configuration, sediment types (clastic, carbonate and/or evaporites) and the data available for the study (seismic, outcrop, core and well data) (Catuneanu et al., 2009).
Although the sequence stratigraphic concepts and terminologies used in this study are heavily adopted from the transgressive‐regressive (T‐R) sequence model (Embry and Johannessen,
1992), this study follows a model‐independent approach (Catuneanu et al., 2009) to develop a sequence stratigraphic framework for the Bakken Formation in the study area. T‐R sequence
concepts are used here due to the thin character of Bakken deposits (30 m average) that
renders any stratigraphic geometric observations (downlap, offlap, onlap …etc) almost
impossible to make; as well as, the availability of cores and well log data that offer sparse
information of the subsurface. Identification of key stratigraphic surfaces in T‐R 19
sequence model is based on observations of the physical characteristics of surface rather than an interpretive approach (Embry and Johannessen, 1992).
The sequence boundaries used in this study follow that of Embry (2002). A sequence boundary
(SB) is defined as a discrete surface that develops throughout base level fall; in shallow marine strata SB is marked by subaerial unconformity and brackish/non marine strata that onlap onto
the SB. In distal marine strata, SB marks a shift from coarsening (shallowing) upward to
fining (deepening) upward succession, this surface is known as the maximum regressive surface
(MRS) (Embry, 2002). This surface (MRS) was assigned the descriptive term “Transgressive surface” in other sequence stratigraphic schools (e.g. Posamentier et al., 1988; Van Wagoner et
al., 1988), since it is onlapped by deposits accumulated during marine trangression.
The regressive surface of marine erosion (RSME) is an erosional surface that develops below progradational, shallowing‐upward shallow marine strata during forced marine regression
(Plint, 1988). In reality, RSME comprises several erosional surfaces that develop throughout the
entire period of forced marine regression (Plint and Nummedal, 2000; Embry, 2002). The high
diachroneity of RSME lead some workers to develop the descriptive term (basal surface of
forced regression BSFR) which marks the onset of base level fall (Hunt and Tucker, 1992).
Several sequence stratigraphic schools (e.g. depositional sequences II, III and IV) identify the
RSME surface as their interpreted sequence boundary since it marks base level fall. The
correlative conformity (CC, also known as BSFR) is the distal correlative to the aforementioned
interpreted sequence boundary, this surface (CC) is ambiguous and hard to identify. RSME is used in 20
this study to correlate shoreface higher order sequences (parasequences), as well as to identify the base of strata formed during forced marine regression.
The maximum flooding surface (MFS) is a significant stratigraphic surface that is generated at the end of marine transgression and the onset of normal marine regression and marks the maximum landward extent of the shoreline. The MFS marks a change from a deepening upward trend to a shallowing upward trend (Embry, 2002). In this study this surface is recognized (interpreted) and used to separate transgressive deposits from overlying normal regressive deposits.
This study uses the broad terms of transgressive system tract (TST) and regressive systems tract
(RST) (sensu Embry, 2002) to group the linked depositional environments demonstrated by the
strata of the Bakken Formation. The term RST used in this study is a generalized term that encompasses all the strata deposited during normal and forced marine regression. These regressive strata comprise several systems tracts; these include a highstand systems tract (HST), a falling stage systems tract (FSST) and a lowstand systems tract (LST).
The HST encompasses all the strata deposited during normal marine regression, and is bounded by a maximum flooding surface (MFS) from below and a regressive surface of marine erosion from above (RSME) (Embry, 2002). The FSST develops above the HST during a phase of forced marine regression, and consists of offlapping, downstepping shallow marine strata (Plint and
Nummedal, 2000). Regressive surface of marine erosion (RSME) are readily recognized at the base of FSST. The upper surface of FSST represents a sequence boundary that typically demonstrates subaerial exposure and deposition of brackish and/or non marine strata above
(Embry, 2002; Plint and Nummedal, 2000). The lowstand systems tract (LST) interpreted in this 21
study resembles the LST wedge defined by Posamentier and Vail (1988). This LST consists mostly of fine‐grained deposits that accumulated in an offshore slope setting contemporaneous to deposition of incised valley fills in more shoreward areas (Posamentier and Vail, 1988).
Deposition of LST wedge occurs during late stages of marine regression and early marine transgression as incised valleys and fluvial channels provide fine sediments that bypass the shoreline (Posamentier and Vail, 1988, Plint and Nummedal, 2000). 22
CHAPTER 2
To be submitted to the: American Association of Petroleum Geologists AAPG Bulletin Sequence Stratigraphy and Provenance of the Bakken Formation in Southeastern
Alberta and Southwestern Saskatchewan
Tarig Ib. Mohamed
Charles M. Henderson
University of Calgary 23
Extended abstract:
Sedimentologic analysis and geophysical well logs interpretation of the Bakken Formation indicate the presence of three systems tracts in southeast Alberta and southwest Saskatchewan and a distal systems tract in the Exshaw Formation to the west. These systems tracts are grouped into a lower transgressive-regressive T-R sequence and an overlying transgressive systems tract. Biostratigraphic data suggest that Bakken T-R sequence(s) are third order sequences. The basal transgressive systems tract comprises most of lower Bakken member, and rests on a sequence boundary which grades into a maximum regressive surface basinwards. The regressive systems tract is made of shallowing upward succession of upper part of lower Bakken member, offshore siltstone, offshore transition wackestone and shoreface and foreshore deposits of the middle Bakken member. Our data allow subdivision of Bakken regressive systems tract into a highstand systems tract (HST) and a falling stage systems tract (FSST) and a distal lowstand systems tract (LST). The HST encompasses the upper portion of the lower Bakken shale and lower offshore siltstone and offshore transition wackestone of the middle Bakken member. The FSST consists of shoreface and foreshore deposits of the middle Bakken member. The distal lowstand systems tract consists of most of lower Exshaw shale member (except lowest part) and upper Exshaw siltsone member. Higher order parasequences are identified within FSST in the Bakken Formation. These parasequence are separated by flooding surfaces demarcated by sharp facies transition from shoreface sandstone to black homogenous mudstone. This demonstrates that pulses of sea level rises interrupted a period of overall relative sea level fall responsible for deposition of the middle Bakken member. The upper transgressive systems tract consists of a succession of tidal channel fill, upper offshore wackestone and outer ramp shale deposits. The transition from tidal deposits to offshore transition wackestone deposits demonstrates a rise in sea level that bypassed and/or eroded other shallow marine environments. This TST is separated from underlying sequence by a sequence boundary. A maximum flooding surface is interpreted to rest within the upper part of the upper Bakken shale. 2
4 This T-R model represents a more refined sequence stratigraphic representation of the Bakken Formation that highlights significant stratigraphic surfaces, as well as lateral relationships between the middle Bakken lithofacies.
Detrital zircon U-Pb analysis and reworked conodonts from three samples of middle Bakken member indicate sediment transport from local basement as well as thrust-belt sources to the north and northeast. Four detrital zircon age populations include: (1) early Paleozoic to late Neoproterozoic 370- 700 Ma grains that were shed from the Franklinian, Ellesmerian and Caledonian orogens, (2) Mesoproterozoic 850- 1500 Ma grains derived from the Grenville orogeny, (3) Paleoproterozoic 1600- 2100 Ma grains similar to Yavapai-Mazatzal provinces, and (4) early Paleoproterozoic to late Archean grains 2400-2900 Ma that were shed from the Canadian Shield to the east. Population 1 is accompanied by reworked conodonts of Chirognathidae and Protopanderodontidae suggesting M-U Ordovician sources (445- 470 Ma) and other reworked elements may be as young as L-M Devonian (385- 420 Ma). These conodonts have much higher CAI values compared to in-situ specimens indicating that they may have been buried to depths in excess of 3 km prior to uplift and exhumation. Our findings suggest derivation of at least 25% of Bakken clastics from northern sources exhumed during the middle Devonian Ellesmerian orogeny. This interpretation suggests long distance sediment transportation most likely by continental scale fluvial systems and possibly by long-shore drift. Outlining sediment provenance and dispersal systems has implications for rock property distribution and prediction and reservoir facies modeling in the Bakken Formation. 25
1. Introduction:
The Bakken Formation is a mixed clastic-carbonate unit that accumulated across the Williston
Basin of central North America during the Late Devonian-Early Mississippian. Generally, the
Bakken Formation comprises three members: a lower black shale member, middle silty
sandstone member and anupper black shale member; all of which are oil producing units in
Canada and the United States (Kries et al, 2006; Sonnenberg and Pramudito, 2009). Currently,
the Bakken Formation is regarded as one of the largest tight oil plays in the world with
recoverable volumes of 7.4 billion barrels of oil (USGS, 2013).
Bakken deposition has been linked to changes in base level in reponse to relative sea level
cyclicity (Richards, 1989; Caplan and Bustin, 1998; Smith and Bustin, 2000; Angulo and Buatois,
2011). Changes in base level in the sedimentary record are recognized by employing sequence
stratigraphic principles (Posamentier, 1988; Van Wagoner et al., 1988; Galloway, 1989; Embry
and Johannessen, 1992). Previous workers (Richards, 1989; Caplan and Bustin, 1998; Smith
and Bustin, 2000; Angulo and Buatois, 2012) have demonstrated that Bakken black shale
members were deposited during a sea transgression. However, controversy still exists over the
base level changes that shaped the Bakken middle sandstone member. Previous
interpretations include transgressive shallow marine (Christopher, 1961), forced regression
systems tract, regressive incised estuary (Angulo et al., 2008), low stand systems tract (Smith
and Bustin, 2000; Kohlruss and Nickel, 2009) and transgressive and high stand systems tracts
(Angulo and Buatois, 2012). Sequence stratigraphic resolution and constraining the provenance and sediment dispersal systems of the middle member clastic deposits remain as two primary challenges. 26
This study employs a sequence stratigraphic model based on a combination of Transgressive-
Regressive (T-R) model (Embry, 2002) and Exxon sequence model (Posamentier, 1988) to
present a new depositional and sequence stratigraphic model for the Bakken Formation. This
model outlines sequence stratigraphic surfaces and systems tracts that have been previously overlooked in the Bakken Formation, specifically within the Bakken middle sandstone deposits.
This model is based on sedimentologic analysis of cores from southeastern Alberta and
southwestern Saskatchewan, allowing new insights into the Bakken depositional and sediment
dispersal systems. Provenance and dispersal systems of Bakken clastics are the key scope of this
study, and are investigated by means of detrital zircon geochronology and detrital conodont
colour alteration index (CAI) and preservational data.
Detrital zircon U-Pb geochronology is a proven technique in identifying crystallization ages of
the grains, constraining potential source areas for sediment and reconstructing sediment
dispersal systems (Gehrels and Dickinson, 1995; Lawton et al., 2009; Gehrels, 2011). However,
multi-cycle sedimentation and recycling of sedimentary rocks have always been a challenge in
defining sedimentary provenance accurately (Thomas, 2011). Conventionally, this challenge
has been addressed by combining petrography and detrital geochronology (Dickinson, 1985), a technique proved to be successful in some cases (e.g. Lawton et al., 2009).
In this study we employ a novel approach to address this challenge for the Bakken deposits; this
approach utilizes two types of detrital materials including detrital zircons and reworked
conodonts that are rounded and have elevated CAI values. Detrital zircon geochronology is used to identify and constrain source rock ages while detrital conodonts and their colour 27
alteration index (CAI) is used to define ages as well as the thermal history of the sedimentary
source area for the Bakken clastic sediments. This combined approach may constrain possible
source areas for sediment with a great degree of precision compared to zircon only studies.
Here, we present over 300 U-Pb ages obtained from three samples collected from the Bakken middle sandstone member in southeastern Alberta and southwestern Saskatchewan (Fig. 1); these ages represent the first documented U-Pb ages of detrital zircon obtained from the
Bakken Formation. These data offer new insight into the late Devonian-early Mississippian paleogeography of central North America, and constrain the provenance and paleogeographic drainage for Bakken clastic deposits.
To summarize, this study uses detrital zircon geochronology and conodont colour alteration index (CAI) and preservational data to (1) constrain the provenance and investigate sediment dispersal systems of the Bakken clastic deposits. Also, this study (2) presents a revised
depositional and sequence stratigraphic model for the Bakken Formation based on
sedimentologic analysis of 11 cores and our current understanding of Bakken sediment
dispersal systems. Finally, this paper will discuss the implications of Bakken provenance on
sediment dispersal systems, as well as depositional environments of the Bakken Formation
present in the study area. 28
Fig. (1) Map of study area depicting wells with cores and with well logs used in this study. Dashed line marks the Alberta-Saskatchewan provincial boundary. 29
2. Geological Setting
The Bakken Formation was deposited in the Williston Basin during the late Devonian- early
Mississippian. Conodont biostratigraphic data suggest that deposition of the Bakken
Formation started during the late Famennian (lower expansa conodont zone) and continued to the middle Tournaisian (lower crenulata conodont zone) (Johnston et al., 2010).
Deposits of the Bakken Formation comprise three main members: (1) Lower black shale member; (2) Middle sandstone-siltstone member; and (3) Upper shale member. The middle member can be further subdivided into six lithofacies; grey siltstone, lower wackestone, laminated and rippled very fine sandstone, cross- and parallel-stratified fine sandstone, interlaminated very fine muddy sandstone and upper fossiliferous wackestone.
Bakken deposits were initially deposited along an intracratonic margin that covered western
North America (Richards, 1989; Halbertsma, 1994). The passive margin constituted a broad, continuous depositional system in which the Williston Basin, the cratonic platform, the Prophet
Trough and the Peace River Embayment were the primary tectonic elements (Richards, 1989). A brief description of the tectonic elements of the Williston Basin is provided here. However, for a detailed background and tectonic evolution of the basin, see Kent and Christopher (1994). The
Williston Basin is an intracratonic, ellipsoidal shaped basin that covers most of North Dakota, the northern parts of South Dakota and Montana and extends to cover the southern part of
Saskatchewan, southeastern Alberta and southwestern Manitoba. Structurally, the Williston
Basin is a stable basin created along weak zones formed in the Precambrian basement following the Trans-Hudson Orogeny (Gibson, 1995). 30
The Sioux Arch in North Dakota, South Dakota and southeastern Montana borders the southern edge of the basin, whereas the Punnichy Arch in Saskatchewan and the Sweetgrass
Arch of southeastern Alberta and northern Montana border the northern and western flanks of the basin (Kent and Christopher, 1994) (Fig 2). Initial deposition within the basin started
during the Cambrian. Recurrent activation of basement faults during the Ordovician, Silurian and Devonian resulted in significant subsidence of the basin and produced fault controlled
sub-basins (Gerhard et al., 1990; Richards et al., 1994). Dissolution of Devonian salt units positively contributed to basin accommodation (Fuzesy, 1960). Williston Basin development from Late Devonian to Early Mississippian is marked by a basin-wide shift from carbonate to siliciclastic deposition (Christopher, 1961), and deposition of accumulations of shale
throughout the basin. Shale accumulations during this time span is thought to be a global event (e.g. central North America Bakken, Exshaw and Banff shales, the Appalachian
Chattanooga shale, Woodford shale of the Permian Basin, Hangenberg shale of Germany and
Montagne Noire, France and Cumana Formation in Bolivia (Isaacson et al., 2008; Kaiser,
2012)). Isaacson et al. (2008) concluded that a glacial episode coupled with mass extinction
(Hangenberg event) might have influenced global depositional patterns during the late
Devonian to Early Mississippian time, leading to worldwide clastic deposition .
A regional unconformity separates the late Devonian-Early Mississippian Bakken Formation
from underlying late Devonian Big Valley carbonate in the subsurface of south central and
southwest Saskstchewan, and from underlying Torquay Formation in southeastern
Saskatchewan (Christopher, 1961). 31
Fig (2) Williston Basin location and Devonian tectonic elements, the red rectangle indicates location of study area (slightly modified from Kent and Christopher, 1994). 32
Late Devonian transgression is recorded by deposition of the offshore black shale of lower
Bakken member and its basinal equivalent the lower part of lower member of Exshaw
Formation (Alberta Basin) (Johnston et al, 2010). Nd-isotope data suggest that Exshaw shale deposits were derived from northern sources (Franklinian Basin in Arctic Canada) during the late Devonian (Patchett et al., 1999).
Clastic (sandstone and siltstone) deposition in a shallow marine environment dominated the
Williston Basin during deposition of the Bakken middle member (Le Fever et al., 1991; Smith and Bustin, 2000). The transition to a shallow marine environment suggests deposition during low relative sea level (Richards, 1989; Smith and Bustin, 2000; Angulo and Buatois, 2012) in which the primary sediment source is thought to be the North American Craton located to the east (Christopher, 1961). Upper Bakken shale and Banff shale deposition indicate a second episode of marine transgression and deposition in open marine environments (Caplan and
Bustin, 1998; Smith and Bustin, 2000). 33
3. Facies and Sequence Stratigraphy
3.1. Introduction:
Depositional sequences are generated by changes in the rate of accommodation change relative to the rate of sediment supply (Vail et al, 1977; Posamentier, 1988; Van Wagoner et al., 1988; Embry, 2002). Generally, Bakken deposition demonstrates a significant basin-wide change in base levels as indicated by the shift from carbonate (Big Valley Formation) to clastic deposition (Bakken and Exshaw formations). The Bakken Formation was deposited on an unconformity surface that records a period of non-deposition, erosion, and reworking of sediments (Kent, 1959; Christopher, 1961; Richards, 2002). This surface has long been described in the literature as the Acadian unconformity (Wheeler, 1963), a regional unconformity that separates the upper and lower Kaskaskia sequence in North America
(Sloss,1962; Smith and Bustin, 2000).
Age determinations are limited in the Bakken deposits. Conodont biostratigraphy is the most reliable, if not the only, criterion for age determination and biostratigraphic correlations. Ages obtained from Bakken conodonts indicate that the lower shale member was deposited during the lower expansa conodont zone while the upper shale member bears conodonts indicating a lower crenulata zone (Johnston et al., 2010). These conodont biozones allow an estimated duration of deposition of approximately 7 to 8 Myrs (Fig. 3). This age duration suggests that one or two 3rd order sequences might be represented by the Bakken deposits, and also points to the condensed character of Bakken deposits accounting for the high abundance of conodonts in these deposits (Johnston and Henderson, 2005). 34
Facies stacking patterns of Bakken deposits indicate two regional marine transgressions
intervened by a regression. Lower and upper shale members are widely accepted as
transgressive and highstand deposits (Fig 3) (e. g. Smith and Bustin 1998; 2000, Richards,
2002; Angulo and Buatois, 2012). However, the depositional conditions leading to the Bakken
middle member have long been debated. Previous studies proposed several sequence-
stratigraphic interpretations for this member, including forced regressive systems tract, regressive incised estuary (Angulo et al., 2008), low stand systems tract (Smith and Bustin,
2000; Kohlruss and Nickel, 2009) and high stand systems tract (Angulo and Buatois, 2012).
The disagreement in these interpretations arises from lack of solid understanding of middle
Bakken depositional systems, particularly, the mechanisms of sediment delivery and dispersal. 35
Fig. (3) Generalized Bakken core log depicting sedimentologic and biostratigraphic characteristics of the Bakken Formation and the interpreted depositional environments and sequence stratigraphic surfaces and systems tracts. The average thickness of the Bakken Formation in southeastern Alberta and southwestern Saskatchewan is approximately 30 m. Conodont zonation adopted from Johnston et al., (2010); age correlations are from Gradstein et al., (2012); litholog modified after Angulo and Buatois (2012). 36
3.2. Methods
This study utilized 11 cores that cover the various units of the Bakken Formation in
southeastern Alberta and southwestern Saskatchewan. The study area covers approximately
2200 Km2 (townships 28 to 34 and range 27 west of third meridian to the third range west of
fourth meridian) (Fig. 1). Detailed descriptions of cores were integrated with petrophysical
well logs, specifically gamma ray logs, in order to interpret and correlate other well logs used
in this study. This study examined more than 150 well logs from different wells, most of which
were used to construct cross sections and isopach maps (See appendecis).
The sequence stratigraphic concepts and terminology in this paper are heavily based on the
transgressive-regressive T-R sequence model (Embry, 2002), as well as the Exxon model
(Posamentier and Vail, 1988). The T-R model is based on recognizable, low diachroneity surfaces as a “quasi-chronostratigraphic” markers for facies correlation (Embry et al., 2007); the T-R model is very applicable in ramp settings.
Three sandstone samples from the Bakken Formation were analyzed for detrital zircon U-Pb
ages to identify the provenance of the Bakken clastic deposits in the study area. An additional
detrital zircon sample was obtained from Exshaw Formation in Rundle Range, Alberta for
provenance correlation. Detrital zircon samples were analyzed at the Isotope Geology
Laboratory, Boise State University. Detailed description of the methodology used in detrital zircon U-Pb analysis is presented in next chapter. Interpretation of the sediment provenance
and source area of the Bakken Formation was further constrained by conodont colour
alteration index (CAI) and preservational observations of Bakken detrital conodonts. Conodont 37
samples were obtained from Bakken cores and were processed following the standard
procedure. Conodont colour alteration indices (CAI) were obtained using a binocular
microscope and reference CAI material.
3.3. Lithofacies
Eight distinct lithofacies were identified in the Bakken Formation within the study area in
southeastern Saskatchewan and southwestern Alberta. The lower and upper lithofacies (Lower
and Upper shale members in the literature) are made of organic-rich, fissile, black shale. This
shale is characterized by laterally consistent lithology and sparse fossils (Table 1, Fig. 4).
Historically, several subdivisions have been established for the Bakken middle member (Smith
and Bustin 1996, 2000; Kries et al., 2006; Angulo and Buatois, 2011). In this study, six lithofacies have been recognized in this member. These lithofacies exhibit distinct lithologies in cores and have distinct responses in gamma ray logs. Interpretation of sedimentary facies and
depositional environments indicate an overall shallowing upward depositional trend. This trend is demonstrated by a shift from lower offshore deposition of lithofacies Mb1 bioturbated
siltstone, to upper offshore deposits of Mb2 fossiliferous wackestone to shoreface and
foreshore deposits of Mb3 and Mb4 and finally tidal deposits of lithofacies Mb5 (Fig. 3). This
shallowing upward succession is overlain by a deepening upward trend demonstrated by
lithofacies Mb 6 upper offshore wackestone and finally upper Bakken shale (Table 1) (Fig. 3). 38 39 40
3.4. Interpretation and discussion:
- Lower and upper Bakken Shale:
Deposition of Bakken lower and upper black shale is thought to represent deposition below wave and storm base level (Kent, 1958; Christopher, 1961; Richards, 1989; Smith and
Bustin, 1996; Kreis et al., 2006) in stratified, anoxic bottom water (Meissner,
1978; Caplan and Bustin, 1998). The consistent monolithic character (Table 1, Fig. 4a) and lack of storm deposits in the lower and upper shale suggest deposition below storm wave base in quiet outer ramp setting, mainly by settling of fine grained clay particles from suspension. The fine laminae observed in upper Bakken shale (Figs, 7c) may be interpreted as a result of suspension settling of fine grains and possibly sporadic higher energy currents that supplied and reworked the silty materials contained in these laminae; these laminae consist of concentrations of conodont elements and silt sized grains. Several patterns of conodont concentrations have been identified and described from the upper Bakken shale (Johnston and Henderson, 2005). One of these concentrations was suggestive of reworking by intermittent, relatively high energy conditions likely storm events during deposition; other types of concentration were consistent with low, quiet energies during slow rates of deposition (Johnston and Henderson, 2005). The sparse fossil distribution in lower and upper
Bakken shale suggests inhospitable, stratified anoxic water during deposition (e.g. Meissner,
1978; Caplan and Bustin, 1998). The regional extent of Bakken shale and its basinal equivalents (Exshaw and Banff formations) throughout WCSB clearly indicate abundance of fine-grained clastics supply during the late Devonian-Mississippian time. The low settling velocities of suspended, fine-grained particles likely resulted in longer transportation 41
Fig. 4. Core photographs of lower part of the Bakken Formation well 10-03-032-28W3, a) outer ramp deposits of lower Bakken shale. b) Lower offshore siltstone deposits of Mb1 showing Nereites (Ne) and Planolites (Pl) ichnofossils. c) Upper offshore to offshore transition deposits of Mb2 fossiliferous wackestone, fossils include brachiopods (Br) and crinoids (Cr). 42
distances and extensive deposition of shale deposits by gravity settling from suspension and possibly flocculation.
Middle Bakken
Generally, sedimentary structures within sandstone and siltstone of middle Bakken
member indicate shallow marine deposition. Lithofacies of middle Bakken member include:
Bioturbated siltstone Mb1
Lithofacies Mb1 consists of grey siltstone with abundant grey and black mud and isolated green
shale lenses. This lithofacies is intensely bioturbated (Chondrites isp, Teichichnus, Nereites missouriensis) with poorly developed lenticular laminations (Table 1). Within the basal part,
lithofacies Mb1 is dark grey and gradually overlies lower Bakken shale. Carbonate mud and fossil fragments increase upwards, reflected as a slightly lighter grey colour. The black muddy
matrix and green mud laminations decreases gradually upward towards the transition to the
overlying unit Mb2.
The fine grained lithology, pervasive bioturbation and sparse sedimentary structures within the
siltstone beds of lithofacies Mb1 (Table, 1; Fig. 4b) imply deposition well below fair weather
base and just above storm wave base in an offshore setting. Isolated mud lenses and drapes suggest periods of storm deposition. The sparse fossils and dark colour in the lower part of
lithofacies Mb1 suggest deposition in a stratified dysoxic conditions. Environmental conditions gradually changed upward as reflected by the more abundant bioturbation and fossil
fragments. The presence of Chondrites isp in Mb1 is suggestive in a lower offshore 43
depositional setting (cf. Angulo and Buatois, 2012), Chondrites isp is typically associated with low oxygen levels and abundant organic matter (Pemberton et al., 1992).The transition from the lower shale to Mb1 siltstone is gradual in all six of the cores studied herein.
Fossiliferous wackestone Mb2
Fossiliferous wackestone of Mb2 consists largely of light grey carbonate mudstone interbedded with very fine to fine grained sandstone (Table 1). Lithofacies Mb2 is rich in brachiopod, crinoid and bryozoan fragments (Fig. 4c). Sedimentary structures are scarce possibly due to bioturbation with rare interpreted hummocky cross stratification, in which sand-rich wackestone is interbedded with dark coloured mudstone deposits. Soft sediment deformation and dewatering structures are common in Mb2.
The predominant muddy lithology and sparse sedimentary structures reflect deposition in an offshore setting below fair-weather wave base. Abundant fossils fragments indicate deposition in storm-influenced setting with well oxygenated, open marine environmental conditions above storm wave base. Hummocky cross stratification outlined by san d-rich laminae may also reflect periods of storm deposition, in which storm waves delivered sandy material from shallower, more shoreward settings. The heterozoan fossil assemblage within Mb2 suggests deposition in mid to outer ramp depositional setting. This interpretation implies that fossils within Mb2 may have accumulated initially below storm wave base and were later reworked by storm waves as a result in changes in relative sea level. Reworking of Mb2 fossils is demonstrated by the fragmentary character of fossils. 44
Fig. 5. Core photographs of coarsening upward succession (See Fig. 6 core log) of lithofacies Mb3 and Mb4. a) lithofacies Mb3 lower shoreface deposits, fine to very fine grained sandstone with cross laminations outlined by darker material, photograph is from well 111/4-06-032-27W3, depth 839.5 m. b) Lithofacies Mb4 upper offshore and foreshore deposits, fine grained sandstone with parallel and cross laminations, photograph is from well 111/16-34-028-3W4, depth 946 m. 45
Fig. 6. Core log showing the sharp based shoreface deposits of lithofacies Mb3 and Mb4 of the Bakken Formation. Core data is showing sedimentology, sedimentary structures, fossils and sharp based contact between upper offshore of Mb2 and lower shoreface of Mb3. 46
Rippled Laminated Sandstone of Mb3 and Mb4
Lithofacies Mb 3 and Mb 4 are composed of fine grained sandstone interbedded with siltstone and mudstone (Mb 3) that coarsens upward into blocky medium grained sandstone
(upper portions of Mb4) (Table 1, Fig. 5, 6). Basal portion of lithofacies Mb3 sharply overlies
Mb2 deposits, and composed of wavy, low angle cross stratified (interpreted as hummocky cross stratification), fine to very fine grained sandstone interbedded with discrete mud drapes and laminations.
The sedimentary structures in Mb3 suggest a sudden increase in energy conditions from upper offshore of lithofacies Mb2 to a lower shoreface setting above fair weather base. The transition from lithofacies Mb3 to the overlying rippled, horizontally laminated, cross bedded and massive sandstone of Mb 4 is gradual in most cores studied here; however, in few instances the contact surface is sharp and demarcated by a Glossifungites surface. The coarsening-upward succession of Mb4 suggests deposition in an upper shoreface setting based on the planner cross bedding and medium grain size and mud-free lithology. Parallel laminated and massive sandstone are common in upper portions of Mb4; these sde a posit re interpreted as foreshore deposits based on the parallel laminations (Fig 5b), which are characteristic of foreshore deposition. Similar shoreface successions are reported to be the main conventional reservoir within the Bakken Formation in southeastern Saskatchewan
(Kreis et al., 2006). North of the study area, Mageau et al. (2001) suggested that lithofacies
Mb3 and Mb4 (described there as upper middle Bakken) were partially eroded and reworked into NE-SW shelf tidal ridges. 47
Fig. 7. Core photographs of upper part of the Bakken Formation well 10-03-032-28W3, a) Tidal channel deposits of lithofacies Mb5 demonstrating rhythmic laminated sandstone and mudstone laminae and mud drapes. Note the upward increase in mud laminae. Ichnofossils include Teichichnus (Te) and Planolites (Pl) ichnofossils. b) Upper offshore to offshore transition deposits of Mb6 fossiliferous wackestone, fossils include brachiopods (Br) and crinoids (Cr). c) Upper Bakken shale. Note the fine horizontal laminations. 48
Tidal deposits of Mb5
Lithofacies Mb5 consists of rhythmically laminated very fine grained sandstone and mudstone couplets. The sedimentary structures are mostly lenticular bedding and flaser bedding (Table
1, Fig. 7a). Mud drapes increase upward within Mb5. The lithology and sedimentary structures in Mb5 are indicative of tidal cyclicity in which sand-rich laminae were deposited in high energy regime while mudstone interlaminations settled during periods of low energy. This lithofacies is interpreted to have been deposited in a tidal dominated setting. Lithofacies
Mb5 is locally present in the study area; in several cores Mb5 is absent and lithofacies Mb6 sharply overlies shoreface deposits of Mb4 (e.g. see appendices well 131/10-3-30-29W3 and
101/10-03-32-28W3; also see subsurface cross sections Fig. 9 and Fig. 15). The erosional contact between tidal deposits of Mb5 and underlying upper shoreface/ foreshore deposits of Mb4 (Fig. 7, 14) and upward increase in mud laminations and mud drape content suggest that deposition of tidal deposits of Mb5 likely occurred in incised valleys/channels that eroded the underlying deposits of Mb4. Preservation of lithofacies Mb5 was likely dependent on distribution of the incised valleys/channels that accommodated tidal deposits. Subsurface cross sections (Fig. 9, appendix 3) suggests that these incised valleys were oriented approximatly NE-SW (Fig. 8). North of the study area (cactus lake area), Mageau et al.,
(2001) postulated that lithofacies Mb5 was preserved in inter-tidal ridge areas, bounded by
NE-SW elongate tidal ridges. These tidal ridges have not been encountered in this study.
However, such tidal ridges could possibly represent subtidal and/or intertidal bar ridges, common in tide-dominated deltas in mesotidal and microtidal settings (Dalrymple, 2010). 49
Tidal flat
Incised valley
Lower shoreface
Upper shoreface Sequence boundary SB2
RSME
Upper offshore
FWWB
Tidal ridges SWB
Sequence boundary SB1 Interlaminated Subaerially exposed rocks, Fine Sandstone not preserved in study area
SB Rippled/ Laminated Fossiliferous Fine Sandstone Sandy Wackestone MFS V.Fine to Laminated Mudstone, Siltstone/ Fine Sandstone V. Fine Sandstone RSME Grey Siltstone Shale Fig. 8. Interpreted distribution of the depositional environments in the Bakken Formation prior to deposition of transgressive deposits of upper wackestone and upper Bakken shale. Incised valleys/channels were likely formed during late stages of forced marine regression. Tidal flats, channel fills and tidal ridges are interpreted to have been deposited during early stages of marine transgression. NE-SW tidal ridges were interpreted nearby to study area in Cactus lake area (Mageau et al., 2001), and possibly represent tidal ridges deposited inmacro-tidal settings. SWB= Storm wave base. FWWB= Fair-weather wave base. 50
In either cases, this is significant because it indicates that tidal processes played a sigificant role during deposition of lithofacies Mb5, as well as it constrains a NW-SE orientation for the shoreline during deposition of the Bakken (Fig. 8).
Fossiliferous wackestone Mb6
Wackestone deposits of Mb6 resemble upper offshore deposits of Mb2 but with more abundant fine-sand and silty material (Table 1, Fig. 7b). The muddy lithology and sparse sedimentary structures of Mb6 reflect deposition in offshore setting below fair-weather wave base. Sand-rich laminae and diverse fossil fragments that consist of brachiopod, crinoid and bryozoan indicate deposition in well oxygenated, storm- influenced open marine setting.
Lithofacies Mb6 contains more abundant very fine grained sandstone laminae, probably eroded and deposited during periods of storm activity. Further deepening is recorded by transition from tidal deposition of lithofacies Mb5 to upper offshore deposits of Mb6 to black shale deposits of outer ramp/ shelf settings. 51
4. Sequence Stratigraphy:
This study defines three systems tracts in the Bakken Formation, a basal transgressive systems tract (TST 1), regressive systems tract (RST) and an upper transgressive systems tract (TST 2)
(Fig. 3, 9). These systems tracts were grouped into a basal sequence (sequence 1) that include
TST1 and RST and a lower part of a sequence (sequence 2) that consists of TST 2 and an overlying RST (within Lodgepole and Banff formations, that is not studied herein). The basal transgressive systems tract (TST1) comprises the lower part of the lower Bakken shale. The regressive systems tract RST encompasses the upper portion of the lower Bakken shale and the middle Bakken member, with tidal deposits of lithofacies Mb5 and upper wackestone of lithofacies Mb6 being an exception; the RST can be further subdivided into a highstand systems tract (HST), a falling stage systems tract (FSST) and a western, distal lowstand systems tract (LST) interpreted within the Exshaw Formation. The upper transgressive systems tract
TST 2 consists of lithofacies Mb5, Mb6 and lower part of the upper shale.
Five sequence stratigraphic surfaces were identified within this succession (Fig. 3, 9), including (1) basal sequence boundary (SB) (sensu Embry, 2002) - and it is basinal correlative maximum regressive surface (MRS1)- underlying the lower Bakken shale, (2) Maximum flooding surface (MFS1) within the lower Bakken shale, (3) regressive surface of marine erosion RSME that separates Mb2 below from Mb3 above, (4) maximum regressive surface
(MRS2) separating Mb4 from Mb5 and finally (5) maximum flooding surface (MFS2) within the upper shale . 52
Fig. 9. Sequence stratigraphic cross section showing systems tracts defined in the Bakken Formation in the study area. Study area and cross section location are shown in the upper left corner. Lower transgressive- regressive T-R sequence is bounded by SB1 and SB2 both of which define the maximum base level fall preceding marine transgression, upper TST2 overlies MRS2. Regressive system tract RST comprises a lower highstand systems tract HST and an upper falling stage systems tract FSST, FSST deposits consist of shallow marine shoreface to tidal flat deposits (Mb3, Mb4 and Mb5). RSME separates HST from FSST. Several marine flooding events are demonstrated by offshore deposits overlying FSST strata (See Fig. 10. for core photograph of flooding surfaces). Cross section is attempting to depict the prograding character of FSST as stacked shoreface successions that prograde westward with ongoing marine regression; note that schematic progradation does not reflect obvious picks in the logs, but this is shown in more detail in Fig. 12. Shazam lines indicate that shallow marine Mb1, Mb2 and Mb3 might be time-synchronous within a shoreface parasequence. The size and extent of shoreface parasequences is based on models of modern shoreface parasequences (Stive and Vriend, 1995; Charvin et al., 2011). 53
4.1. Sequence 1:
4.1.1. Sequence Boundary (SB 1) and Maximum Regressive Surface (MRS 1)
Sharp lithologic transition from shallow mari ne carbonate deposits to outer ramp/shelf black shale demonstrates a sequence boundary (SB 1) that separates Big Valley Formation below from lower Bakken shale above (Fig. 10). This sharp transition indicates non-deposition and/ or erosion prior to deposition of black shale. This surface has been identified as a regional unconformity throughout North America that separates the upper and Lower Kaskasia mega sequences (Acadian unconformity) (Sloss, 1962; Smith and Bustin, 2000). Basinal continuation of this surface is represented by sharp lithologic contact between lower black shale member of Exshaw Formation and upper Costigan member of Palliser Formation at Goat Creek section in Rundle Range. Well preserved Glossifungites ichnofacies along the contact surface suggest a period of non-deposition and development of a firm ground. This contact has been documented elsewhere in the Rocky Mountains (Richards, 1989; Richards et al., 2002) and from the subsurface of WCSB (Johnston et al., 2010).
In Bakken cores, this surface is interpreted as a sequence boundary (SB 1) (sensu Embry, 2002) since it indicates a depositional hiatus marked by development of firm grounds and
Glossifungites ichnofacies, as well as a major disruption in facies patterns, demonstrated by sudden deposition of black shale above the surface. In the subsurface of present-day Alberta
Foreland Basin, the distal continuation of this SB 1 is interpreted as maximum regressive surface (MRS 1) (Fig. 13, Fig. 15) that rests below the Exshaw Formation and records the shallowest sea level that preceded the relatively deep offshore deposition of lower shale member of the Exshaw Formation. The MRS1 marks a change from a regressive depositional 54
Fig. 10. Core photograph of lower and middle members of the Bakken Formation in well 111/4- 06-032-27W3. See Fig. 6 for detailed sedimentological log and interpretation of depositional environments. Note the abrupt facies changes across SB 1; location of MFS1 is interpreted within the upper part of lower Bakken shale. Note the lag surface demarcating RSME. RSME separates normal regression deposits below from forced regression deposits above. 55
style below, observed in Big Valley/ Torquay formations (Halbertsma, 1994) to a transgressive depositional setting manifested by lower part of lower Exshaw member and lower Bakken shale.
4.1.2. Lower Transgressive Systems Tract TST 1:
Rapid vertical transition from carbonate-dominated Big Valley Formation to shale deposits
(Fig. 10 of lower Bakken indicates change from shallow marine setting to a relatively deeper, open marine deposition below fair-weather and storm wave-base. This transition reflects a rise in relative sea level. Lower part of lower Bakken shale was deposited during marine transgression following the development of SB 1and MRS (Fig. 8, Fig. 9). Landward movement of shoreline and base-level rise during marine transgression might have resulted in reducing sediment input into the basin and progressive deepening of depositional setting allowing condensed deposition of mudstones. In addition, marine transgression might have positively contributed to long-shore drift and transportation of fine-grained sediments from source areas.
4.1.3. Maximum Flooding Surface MFS 1
Identification of the maximum flooding surface (MFS) surface within the study area is somewhat challenging since all of the lower Bakken shale represent condesed deposition of fine-grained sediments (Catuneanu et al., 2009). As a result, maximum flooding surface MFS 1 is interpreted within the upper part of lower Bakken shale (Figs. 8, 9, 15). The interpreted
MFS1 indicates the maximum landward transgression of the shoreline and marks the point of deepest deposition in the lower Bakken shale. Within shallower part of the basin (shoreward facies), MFS1 likely coincides with a distinct shift in depositional trend from fining upward deposition below to coarsening upward deposition above, and records increased coarse- grained sediment supply. 56
Regressive Systems Tract RST
4.1.4. High Stand Systems Tract HST
Deposition of Bakken lower shale came to an end following enhanced supply of silt-sized
sediments. The increased sediment supply is recorded by deposition of siltstone deposits of
Mb1 (Fig. 4, 10). Lithofacies Mb1 and Mb2 indicate deposition during highstand sea level with
frequent, sporadic pulses of silt-sized sediment supply during normal marine regression. The
lithology mud-rich siltstone deposits of of Mb1 (Table 1)suggests that marine transgression
ceased, but sea level was still high. Fine-grained (clay-sized) sediments continued to settle
from suspension while silt-sized sediments were delivered from more shoreward settings
during periods of storms. Very fine to fine grained sand component significantly increases in
Mb2 (20- 25% of total grains component) as opposed to the silty component of Mb1, which further support a supply of coarser sediment and development of normal marine regression.
This implies that this gradational succession (upper part of lower Bakken shale and lithofacies
Mb1 and Mb2) represents an overall progradational highstand system tract.
4.1.5. Regressive Surface of Marine Erosion (RSME) and Basal Surface of Forced Regression
(BSFR)
The erosional contact between the fossiliferous wackestone of Mb2 and shoreface deposits of
Mb3 and Mb4 (Fig. 10) indicates a significant base level fall. This surface is interpreted as a regressive surface of marine erosion (RSME), in which shallow marine shoreface deposits (Mb3
and Mb 4) sharply overly deeper offshore deposits of Mb2 (Fig. 9, 10). Catuneanu (2002) and
Embry (2002) demonstrated that a RSME is a highly diachronous surface, and is typically
associated with forced marine regression (Plint and Nummedal, 2000). 57
The disconformable nature of this surface is due to erosion and reworking of shallower parts of inner shelf regions by waves as base level fell (Plint, 1988; Embry et al., 2007). This surface
has been previously interpreted as a sequence boundary (Smith and Bustin, 2000), but, here it is interpreted as a regressive surface of marine erosion (RSME) since no subaerial
unconformity along the surface has been reported and the change from upper offshore
deposition to lower shoreface deposition across the surface suggests no major disruption of
facies pattern (Embry, 2002). Although the RSME shows evidence of erosion in cores in which lower shoreface deposits of lithofacies Mb3 sharply overlies upper offshore deposits of lithofacies Mb 2 (Fig, 10), the vertical facies transition from upper offshore deposits to lower
shoreface deposits indicates no major disruption of facies succesion, and suggests that RSME erosion was not very deep and the time represented by this disconformity may be short.
RSME is a highly diachronous surface that consists of multiple surfaces (Fig. 10, 11, 15) that develop throughout the entire period of base level fall, making this surface an invalid time
marker (Embry, 2007, Plint and Nummedal, 2000)). This is demonstrated by the regional extent of shoreface deposits of Mb3 and Mb4 within the study area and most of the Canadian
part of the Williston Basin (Kreis, 2006), that implies relatively long period of deposition in
which base levels continued to fall and the shoreline prograded further seaward. Therefore,
the RSME identified here represents a good lithostratigraphic marker (material surface) that
generally outlines basinward progradation of base level fall and accumulation of overlying
FSST deposits.
4.1.6. Falling Stage Systems Tract
The facies succession of Mb3 and Mb4 represents deposition in shoreface to foreshore setting. 58
This is based on the sandy and silty character, cross lamination, hummocky and parallel lamination (Table 1, Fig. 5) that indicate a high-energy environment. The sharp base and coarsening upward succession (Mb3 and Mb4) suggest that this facies succession was formed by progressive fall in base level and progressive offlap of shallow marine environments. As a result, this succession is part of a falling stage systems tract FSST (cf. Angulo and Buatois, 2012) that makes the upper part of a regressive systems tract (RST).
Within the falling-stage succession, at least two higher-order parasequences are identified in well (111/06-33-28-2W4) (Fig. 11). The lower parasequence consists of sharp-based massive sandstone similar to lithofacies Mb4 and bounded by a flooding surface. On well logs, the parasequences are laterally correlatable, especially when flooding surfaces are marked and delineated into the shoreface succession (Fig. 9); in cross section, flooding surfaces are low- angle and westward dipping (Fig 12). Parasequences within the upper part of FSST grade laterally westward into time equivalent offshore, highstand deposits of Mb2 and Mb1. Stacking patterns of these parasequences suggest a downstepping, westward progradation (Fig. 12) direction in which sediments were likely sourced from the east or northeast. This demonstrates that this period of overall base level fall and formation of FSST was interrupted by several minor and high-frequency flooding events; this is the first attempt to correlate parasequences within the Bakken Formation in the study area.
4.1.7. Lowstand Systems Tract
Upper siltstone member of Exshaw Formation has been previously interpreted as a regressive systems tract (Richard, 2002). In this study, upper part of lower member and upper siltstone member of Exshaw Formation are interpreted as a distal lowstand systems tract bounded by a basal surface of forced regression (BSFR) at the base and capped by a MRS 2 (Fig. 13, Fig. 15). 59
Fig. 11. Core data of well 111/06-33-028-2W4 two parasequences within shoreface and foreshore deposits of Mb4. Parasequences are bounded by flooding surfaces shown in core photograph; depth 922 m. RSME 2 suggests return of forced marine regression and deposition of another parasequence above. 60
101/12-02-032-28W3/00 <=965.7m=> 191/10-02-032-28W3/00 <=252.9m=> 140/09-02-032-28W3/00 <=399.2m=> 121/12-01-032-28W3/00 <=413.5m=> 101/11-01-032-28W3/00 <=501.9m=> 111/10-01-032-28W3/00 <=579.1m=> 141/09-01-032-28W3/00 W E
GR (gapi) GR (gapi) GR (gapi) GR (gapi) GR (GAPI) 0.00 150.00 0.00 150.00 0.00 150.00 0.00 150.00 0.00 150.00 GR (gapi) GR (gapi) 0.00 150.00 0.00 150.00 800 800
800
800
Mbakken_U Mbakken_U Mbakken_U Mbakken_M Mbakken_M Mbakken_U Mbakken_M 800
Mbakken_L 800
850 Mbakken_L Mbakken_L TD 836.1m 850 Mbakken_L GR (gapi) Mbakken_L Mbakken_L 850 Mbakken_L 0.00 150.00
TD 842.2m GR (gapi) 0.00 150.00 TD 827.5m GR (gapi) TD 870.5m 0.00 150.00 GR (GAPI) TD 847.5m 0.00 150.00 TD 883.0m GR (gapi) 0.00 150.00 GR (gapi) 0.00 150.00 Sequence boundary Flooding surface TD 891.0m GR (gapi) 0.00 150.00 Fig. 12. High resolution subsurface cross-section of the falling stage systems tract (FSST) of the middle Bakken parasequences (higher order sequences). The map shows the location and orientation of the cross-section. FSST strata were deposited during forced marine regression and comprises several parasequences, illustrated in various colours seen in the cross section, i.e. yellow, orange, blue, green...etc. Parasequences of FSST are sharp based and bounded by Flooding surfaces and/or regressive surfaces of marine erosion (RSME). RSME are expressed as erosional contacts (see Fig. 10) rich in pebbles and reworked material, while flooding surfaces are marked by finner mud-rich deposits overlying coarsening upwards, fine to medium sandstone deposits (see Fig. 11). Note west-downstepping charecter of parasequences, and offlaping of younger parasequences on older ones. 61
Conodont biostratigraphy indicates that lower part of lower Exshaw shale member is correlative with lower Bakken shale member (Johnston et al., 2010). This suggests that lowest part of
Exshaw shale member was deposited coeval to transgressive deposits of lower Bakken shale
(Richard, 1989; 2002; Johnston et al., 2010). Deposits higher in the lower Exshaw shale member were deposited during lowstand sea level following deposition of FSST deposits of the Bakken
Formation (Fig. 15). The last point is further supported by U-Pb ages of 365 Ma (this study, see provenance section) obtained from detrital zircons extracted from reworked lag surfaces (1.28 from base of Exshaw Formation, sample EX1) (Fig. 13) as well as ages of 359-360 Ma obtained from ash beds within upper part of lower Exshaw member (Bundy et al., 2013). These ages suggest that lag surfaces within lower Exshaw and ash beds of upper part of lower Exshaw shale member may have been deposited during praesulcata to kockeli conodont zones, and thus correlative to parts removed from middle Bakken FSST deposits during forced marine regression. A coarse grained sand and silt-rich lag surface (0-10 cm thick) within lower Exshaw member at Rundle Range (Goat Creek, 1.28 m from base of Exshaw Formation) suggest erosion and reworking of nearby or underlying beds. This erosion indicate marine regression associated with fall in base level. Although delineation of the basal surface of forced regression BSFR and start of forced marine regression is speculative in distal or basinal deposits (Embry, 2002), the reworked layers observed in lower member of Exshaw Formation at Goat Creek might constrain the position of BSFR. Thus, a BSFR is interpreted to lie below the reworked layers observed in lower member of Exshaw Formation (Fig. 13 & Fig 15). Silt-size deposits of upper siltstone member of Exshaw Formation suggest progradation of shallower depositional settings and deposition during a lowstand sea level (Richards, 2002). 62
Fig. 13. Stratigraphic section of uppermost Palliser Formation, Exshaw Formation and lower Banff Formation at Goat Creek, Rundle range, showing lithologies of the units and sequence stratigraphic interpretation. Red line indicate reworked lag layer, red star shows location of 365 Ma detrital zircon sample (Ex1). Exshaw Fm. rests on the Costigan Member of the Palliser Formation. Section from Johnston et al., (2010). For sequence stratigraphic interpretation of Banff formation see Richards et al., (2002). U-Pb ages are from Bundy et al., (2013). 63
4.2. Sequence 2
4.2.1. Sequence Boundary (SB 2) and Maximum Regressive Surface (MRS 2)
Truncation of upper part of shoreface and foreshore deposits of Mb4 is clearly visible in cores
(Fig. 14, photo) and well logs (Fig, 9). Fall in base level resulted in the erosion of Mb4 likely by tidal channels and subaerial exposure. Although no direct evidence of subaerial exposure has been reported in Bakken deposits in this part of the basin, it is interpreted that subaerial exposure of upper part of Mb4 strata might have occurred, but was eroded during successive marine transgression.
Correlation of well logs show that the lithologic change across the surface that seperates Mb4 from Mb5 deposits corresponds to a sharp increase in gamma ray values. The fining upward trend in overlying strata of Mb5 (Fig. 14) is interpreted to reflect rise in base level and increase in accommodation space. As a result, this surface is interpreted as a sequence boundary (SB 2) that grades basin-ward into a maximum regressive surface (MRS 2 (Fig. 9, 15).
In Exshaw Formation outcrops (e.g. Goat Creek section), MRS 2 separates the coarsening-upward strata of upper siltstone member of Exshaw Formation from the overlying fining-upward strata of lower Banff shale member (Fig, 13 & Fig. 15) (See Richards,
1989; 2002 for comprehensive descriptions of Exshaw sections). Angulo and Buatois (2012) interpreted the surface seperating Mb4 from Mb5 deposits as an amalgamated sequence boundary and transgressive surface that separates a highstand systems tract below from a transgressive systems tract above. Others (Smith and Bustin, 2000) assigned a transgressive surface interpretation for this surface. Here, this surface is interpreted as a sequence boundary (SB 2) since there is (1) clear evidence of erosion of underlying shoreface deposits 64
Fig. 14. Core photograph of middle and upper members of the Bakken Formation in well 111/10-03-032-28W3 showing the position of sequence stratigraphic surfaces and systems tracts. a) Sharp facies transition across SB 2 from FSST deposits below to TST 2 deposits above. Flooding surface (FS) marks change from tidal channel deposits below to upper offshore/ offshore transition deposits above following a rise in relative sea level. MFS 2 is interpreted within upper part of upper Bakken shale.
(Mb4) below, (2) an overall fining upward succession above the surface and (3) the likely
possibility of subaerial exposure of upper parts of Mb4 strata. SB 2 separates a falling
stage systems tract below FSST from a transgressive systems tract (TST 2) above (Fig. 9).
4.2.2. Upper Transgressive Systems Tract TST 2:
The rhythmic laminations of very fine grained sandstone and mudstone present in lithofacies
Mb5 (Fig. 7, 14) suggest deposition in a tidal setting. Sand-rich laminae indicate deposition in
high energy regime while mudstone interlaminations indicate periods of low energy. The sharp
contact between Mb5 and underlying upper shoreface and offshore deposits of Mb4 suggest
that tidal channels likely eroded older marginal marine deposits providing the mud clasts and
very fine sandstone and siltstone seen in the laminae. 65
The well developed rhythmic laminations suggest that this setting was likely tidally influenced.
These tidally influenced strata were either deposited and preserved in a bay-head delta during falling-stage or deposited as tidal channel or estuary fills as a result of increase in accommodation during marine transgression. The latter interpretation is favoured because of the erosional contact between Mb4 and Mb5, which suggests that upper part of Mb4 have been eroded during base level fall and possibly during earliest stages of marine transgression, after which, Mb5 was deposited during early stages of marine transgression. As a result, Mb5 is interpreted as part of a transgressive systems tract rather than a FSST. Fossiliferous wackestone of Mb6 and upper Bakken shale abruptly overlie tidal flat deposits of Mb5. Lithofacies Mb6 is interpreted to have been deposited below, but close to, fair-weather base in upper offshore to offshore transition. Fine grained deposits of upper Bakken shale reflect deposition in deeper outer ramp/offshore setting. This demonstrates marine transgression and increase in accommodation space due to rise in base level. This marine transgression resulted in re- establishment of open marine setting similar to that of lower Bakken shale and lithofacies Mb2.
Deposition of Bakken upper shale came to an end following end of marine transgression and development of a maximum flooding surface (MFS2). Deposition of Lodgepole Formation and
Banff upper member carbonates dominated the basin during a highstand sea level. Maximum flooding surface MFS2 is placed within the upper Bakken shale.
4.3. Conclusions, Discussion and Implications:
In summary, this study describes a T-R sequence (sequence 1) and a half sequence (TST2 of sequence 2) (Fig. 9, 15) within the Bakken Formation based on the sedimentologic and 66
W Goat Creek, Rundle Range, AB E Williston Basin Approximate U.Sh Location of Study Area Mb 6 TST2 Mb 5
Mb 4
b b Removed FSST
alw
. B nff
M r MFS 2 SB 2 Mb 3 TST2 RST MRS E Mb 2 MRMRSS 2 HST Ex. Mb1 Mbr SFSF
up. up.
LST L.Sh 359-360 Ma TST1 365 Ma SB 1 MFS 1 Peripheral Bulge
Low aw r xsheE m br m MRS 1 TST1
LST = Lowstand Systems Tract SB TST = Transgressive Systems Tract Up per of f sho r e/ o f f sh or e tr an siti on Ex sha w o ffshore Marg ina l m ari ne / tidal depo si t s Si ltstone d eposi t s MFS HST = Highstand Syst ms Tracte Low er of f sh or e deposit s Upp er shoreface/ foreshore Ex shaw sandy l ag RSME RS T = Re gre s s i v e S y s te ms T ra c t depo s its s ur f ace s and sandy MRS Outer ramp/ shelf deposits Lower shoref ace de pos its FSST = Falling Stage Systems Tract bentonit e l aye r s
Fig. 15. Schematic cross section depicting the sequence stratigraphic surfaces and systems tracts present in the Bakken Formation and their basinal continuation westward. Note the prograding character of FSST deposits. Coarse sand and silt lag surfaces within the lower member of Exshaw Formation might have possibly formed during early stages of forced regression observed in the Bakken deposits. Offshore siltstone deposits of upper Exshaw member represent the distal continuation of forced regression deposits of the Bakken Formation. Some elements in this figure (removed strata and peripheral bulge) are adopted from Johnston et al., (2010). U-Pb ages of Exshaw Formation (359- 360 Ma) obtained from Bundy et al., (2013). 67 stratigraphic characteristics of Bakken deposits in cores and well logs. Conodont biostratigraphy ages (Fig 3) suggest a duration of 6.5 Ma for deposition of Bakken Formation
(Johnston et al., 2010). This duration implies that Bakken depositional sequences are 3rd order sequences. Two sequence boundaries (SB1 and SB2) and their basinal continuation
(MRS 1 and MRS 2) have been identified; SB 1 separates the lower Bakken sequence from underlying Big Valley carbonates while the upper SB2 separates sequence 1 from overlying TST
2 of the overlying sequence. These surfaces are easily recognized in Bakken cores (Fig. 10) and
Exshaw cores (Johnston and Henderson, 2005) and outcrops sections (Richards, 1989;
Richards et al., 2002) by truncation of underlying deposits and accumulation of fine-grained, deepening upwards deposits above. The regional extent of these surfaces and the major changes of facies across these surfaces suggest that these sequence boundaries were caused by allocyclic processes (relative sea level fluctuations and regional tectonics).
Two main challenges are present when attempting sequence stratigraphy in the Bakken
Formation: (1) the formation is exclusively subsurface, which results in typical uncertainties associated with core observations and interpretations and (2) thin nature of the Bakken deposits significantly limits well log correlations and hinders identifying interfingering relationship between different facies that are time-equivalent. As a result of these limitations, previous sequence stratigraphic interpretations (e.g. Smith and Bustin, 2000; Angulo and
Buatois, 2012) followed a “flat-layer cake analogy” approach, in which time relationships between the various facies observed in the Bakken Formation are either ignored or vaguely addressed. The sequence stratigraphic model we provide here attempts to establish a framework that accounts for the diachroneity of middle Bakken deposits. Well log correlations 68 of marine flooding surfaces demonstrates the genetic relationship between the normal regression HST siltstone and wackestone deposits of Mb1, and with the overlying shoreface
FSST deposits of Mb3 and Mb4 (Fig. 9, 15). It is possible to view the interfingering facies in terms of variation in accommodation space. Occurrence of deeper, offshore deposits separating shoreface parasequences of Mb3 and Mb 4 (Fig. 11) suggests that rise in base level and increased accommodation to sediment supply ratio during an overall regression. This demonstrates that several minor and high-frequency relative sea level rise and flooding events interrupted deposition of FSST during a period of an overall base level fall. Furthermore, this facies association outlines the highly diachronous character of forced regression and FSST deposits; during which, several flooding events can be recognized (Fig. 11).
The sharp upward change to open marine condition between tidal deposits of lithofacies Mb5 and upper offshore to offshore transition deposits of Mb 6 indicate a rapid rise in base level and marine transgression. This marine transgression is likely responsible for erosion of underlying upper shoreface to foreshore deposits of Mb4 and partial preservation of tidally influenced deposits of Mb5. Furthermore, marine transgression could possibly account for the high sand content of lithofacies Mb 6 which likely resulted from erosion of older strata. Rapid marine transgression might also explain the absence of shallow marine sedimentary environments that bypassed the area or eroded away during the change to open marine settings.
From a reservoir perspective, spatial distribution of reservoir facies is controlled by several parameters (physical, chemical and biological) (Angulo and Buatois, 2012). Base level changes are the main controlling factor of these parameters. Therefore, examination of base level changes within a sedimentary unit is crucial in identifying potential reservoir facies. 69
Previous interpretations assumed a “laterally persistent” facies distribution of Bakken
deposits that grades basinward to Exshaw deposits. This assumption could be misleading if one considers the complex lateral variation of middle Bakken lithofacies locally and transition to distal lowstand deposits of Exshaw Formation. We suggest integrating reservoir
characterization studies with high resolution sequence stratigraphic data for effective
reservoir exploration, evaluation and development. High resolution sequence stratigraphic
data will allow interpretation of sudden lithologic changes resulting from higher frequency
base level shifts. 70
5. Provenance of the Bakken Clastic Deposits
5.1. Introduction
The middle Bakken member consists predominantly of clastic deposits that accumulated during
a stage of marine regression within a shallow marine shoreface and foreshore environments
(Smith and Bustin, 1996; Kries et al., 2006; Angulo and Buatois, 2012; this study). Whereas the
sedimentology and petrophysical properties of these clastic deposits are reasonably well
established (Le Fever et al., 1991; Pitman et al., 2001), fundamental questions regarding
sediment provenance and sediment transportation patterns remain unanswered. The
Franklinian, Acadian and Antler orogenies have been recognized along the margins of North
America during the Devonian- early Mississippian (Fail, 1997; Root, 2001; Mckerrow et al.,
2000), however, the relationship between these orogenies and the sediment supply to the
Williston Basin remains poorly understood. Reconstruction of Devonian-Mississippian (D-M) paleogeography is crucial in refining the depositional models and stratigraphic frameworks of
D-M strata. This can be constrained by sediment provenance and understanding the processes
involved in sediment transportation.
Increasing interest in U-Pb crystallization ages of zircon grains of igneous, metamorphic and sedimentary rocks has provided robust detrital zircon reference ages that have been used to constrain provenance and paleogeography of several sedimentary units across North America
(Gehrels et al., 1995, Gehrels and Pecha, 2014).
Our objective is to constrain the provenance of Bakken deposits in southeastern Alberta and southwestern Saskatchewan and to investigate sediment dispersal systems for these deposits.
This investigation combines detrital zircon U-Pb ages and detrital conodont analysis. 71
U-Pb ages of detrital zircons are used to determine the original igneous and/or metamorphic
and/ or sedimentary provenance for the deposits, while detrital conodonts CAI are used to
determine sedimentary source for the deposits. Data presented here represent the first U-Pb
detrital age data from the Bakken Formation, and provide important insights into the
paleogeography of central North America during latest Devonian- early Mississippian time.
5.2. Samples and Methods:
5.2.1. Samples:
Three samples were collected from the Bakken Formation in southeastern Alberta and southwestern Saskatchewan and analyzed for detrital geochronology and conodont colour alteration index (CAI) and preservation (See Fig. 16 for sample location, Fig. 2 for stratigraphic position and Table 2 for samples information). An additional sample (Ex1) was taken from a thin sandstone interval located 1.28 m from the base of the Exshaw Formation
(Fig. 16, Table 2) within the black shale member in the Rundle Range (Goat Creek). A description of thesamples and sample locations are summarized in Table (2).
5.2.2. U-Pb Geochronology
Method
U-Pb samples were prepared and analyzed at the Isotope Geology Laboratory, Boise State
University. Samples were crushed with a jaw crusher and zircon grains were separated with water table followed by heavy liquid separation. Separated zircon grains were annealed at
900˚C for 60 hours in a furnace. The zircon grains were then mounted in epoxy and polished to expose the interior of the grains. 72
Fig. 16. Map of study area, red circles indicate location of samples used in this study. Sample Ex1 was obtained from Rundle Range (Goat Creek), Alberta. 73
Detrital zircon ages were obtained using a ThermoElectron X-Series II quadrupole Inductively
Coupled Plasma Mass Spectrometer (ICPMS) and New Wave Research UP-213 Nd:YAG UV (213 nm) laser ablation system at the Boise State University Isotope Geology Laboratory. Zircon grains were ablated with a 25 µm wide laser spot, using fluence and pulse rate of 5 J/cm2 and
10 Hz respectively, excavating a pit, 25 µm in depth, during a 45 second analysis (30 second abalation, 15 second gas blank). A 1.2 L/min He gas stream carried the ablated material to the nebulizer’s plasma flow. Background count rates were obtained prior to each spot analysis and later subtracted from the raw count rate of each analysis. Instrumental fractionation of the background-subtracted ratios was corrected and dates were calibrated with respect to measurements of the Plešovice zircon standard (Sláma et al., 2008). A zircon secondary reference was analyzed as an unknown in groups of two analyses every 20 unknown analyses to assess instruments accuracy. Between 85 and 129 ages were recorded for each sample. The
U-Pb ages obtained were plotted in Isoplot 3.0 (Ludwig, 2003) to calculate weighted mean dates. Standard calibration uncertainties for Pb207/Pb206 dates ranged from 0.46 to 0.64%
(2σ), and 3.1% to 3.5% (2σ) for Pb206/U238 dates. The data reported are Pb206/U238 ages for grains younger than 1000 Ma and Pb207/Pb206 ages for grains older than 1000 Ma. Errors on the dates are given at 2σ.
Results
Detrital zircon U-Pb ages as young as 370 Ma and as old as 2900 Ma are present in the Bakken deposits. Zircon ages can be subdivided into four distinct age groups that are present in similar concentrations in all three samples (Bk1, Bk2 and Bk3). These age groups are (1) middle
Paleozoic to late Neoproterozoic (370- 700 Ma), (2) late Neoproterozoic to Mesoproterozoic 74
(800- 1500 Ma), (3) Paleoproterozoic (1600- 2200 Ma), and (4) early Paleoproterozoic to
Meso-archean (2400-2900) Ma. The consistency observed in the U-Pb ages of Bk1, Bk2 and
Bk3 is evident in the graphical similarity of their cumulative probability curves (Fig. 17) in which major peaks nearly overlap, suggesting that these units were likely sourced from the same provenance.
Bk1
Eighty-four zircon grains were analyzed from sand-rich wackestone of Mb2. Projected ages of these grains, as illustrated in Fig (17), yield one main peak at 415 Ma and subordinate peaks at
1020, 1140, Ma and 2680 Ma. Paleoproterozoic to late Archean detrital zircon grains count for
8% of the total zircon population of this sample. A conspicuous spike is observed at 2680 Ma and no zircons were found in the range from 2340 Ma to 2600 Ma. Paleoproterozoic grains are relatively sparse in this sample and represent 19% of the total zircon grain population; however, no significant spikes are observed in this age range. Forty grains of the total population of this sample (48%) falls within the Mesoproterozoic age group (800 to 1500 Ma), and two spikes at 1020 to 1140 are observed in this age range. Early Paleozoic to late
Neoproterozoic grains falls in the range of 378 Ma to 625 Ma and count for 23% of total grains population. The main spike in this sample falls within this age range at 415 Ma, and no zircons were found in the range of 625 to 850 Ma in this sample. Thermal ionization mass spectrometry (TIMS) analysis indicate that the youngest grain in this population is dated at 378
± 0.2 Ma.
Bk2
Sample Bk2 was obtained from shoreface sandstone (Mb3 and Mb4). The 129 grains analyzed 75
Fig. 17. Detrital zircon probability curves for Bakken Formation samples. Zircons ages can be subdivided into four distinct age groups that are present in similar concentrations in all three samples (Bk1, Bk2 and Bk3). Age groups are outlined with grey rectangles shown in the figure. The youngest detrital zircon population is accompanied by reworked conodonts of Chirognathidae and Protopanderodontidae (Fig 20 a & b), possibly, of Ordovician age (445- 470 Ma) and other elements may be as young as L-M Devonian (385- 420 Ma). from this sample yielded a main peak that nearly overlaps that of Bk1 at 423 Ma. Subordinate peaks were observed at 1040, 1620, 1740, 1990 and 2610 Ma. Paleoproterozoic to late Archean detrital zircon grains are relatively scarce in this sample and count for only 7% of total grain population. A subordinate spike was observed at 2610 Ma and no zircons were found in the range from 2200 Ma to 2500 Ma. Grains within the range of 1500 Ma to 2200 Ma
(Paleoproterozoic) counts for 22% of the total population, and contains three spikes at 76
1620, 1740 and 1990 Ma. Mesoproterozoic grains occupying 800 to 1500 age range represent
50% of the grains analyzed in this sample with a spike observed at 1040 Ma. The youngest age population (Early Paleozoic to late Neoproterozoic) counts for 19% of total grain population and contains the main spike in this sample at 423 Ma.
Bk3
Sample Bk3 is the youngest of the three samples and was collected from sand-rich wackestone of Mb6 (Table 2). The relative age probability curve is nearly identical to that of Bk2 with
slightly older age peaks. The 112 grains analyzed yielded a main peak at 426 Ma and
subordinate peaks at 1055, 1640, 1770, 1890 and 2600 Ma (Fig. 17). Paleoproterozoic to late
Archean detrital zircon grains count for 11% of total grain population of this sample and
demonstrate a sparse distribution in the age range from 2400 Ma to 2900 Ma while no grains
were observed in the age range from 2100 to 2400. Paleoproterozoic grains (27% of total
grains) are distributed within the age range of 1600 to 2060 Ma and exhibit three subordinate spikes at 1640, 1770, 1890 Ma. Mesoproterozoic grains represent 40% of total grains analyzed in this sample with a spike observed at 1055 Ma. Early Paleozoic to late Neoproterozoic grains fall in the range from 380 Ma to 710 Ma and represent 20% of the total grains analyzed in the sample. The zircon age spectrum shows a main spike at 426 Ma in this age population.
Ex1
The sandstone collected from the lower member of Exshaw Formation contains a large number of zircon grains. Ninety nine grains were analyzed and yielded ages as young as 340
Ma and as old as 2900 Ma (these are LA-ICPMS ages, reported TIMS ages yield 360 Ma from yonger ash layers (Bundy et al., 2013), with an age distribution that significantly varies from that of the Bakken samples (Fig. 18). 77
Fig. 18. Upper graph illustrate detrital zircon probability curve for Exshaw Formation sample, age groups are outlined with grey rectangles shown in the figure. Middle Paleozoic population (340 to 400 Ma) constitutes 65% of the sample population, grains within this population (340-400 Ma) were likely supplied from a nearby volcanically active source related to the collision of Kootenay Terrane with the western margin of North America (Richards et al., 2002). Note how age probability curve of Ex 1 differs from that of Bakken samples Bks, suggesting distinct sources as well as paleo-sediment dispersal systems for detrital zircons within each Formation.
Paleoproterozoic to late Archean detrital zircon grains count for 9% of the total zircon population of the sample. Paleoproterozoic grains (1600- 2400 Ma) represent 18% of the total sample grains, and show a subordinate spike at 1820 Ma. Mesoproterozoic grains are scarce in this sample (8% of total sample grains) compared to the Bakken samples, and occupy a range from 1100 Ma to 1500 Ma. 78
Middle Paleozoic grains (340 Ma to 400 Ma) are abundant in this sample (65%). The predominant
spike is located at 365 Ma and no grains were observed in the age range from 400 Ma to 1000
Ma.
Interpretation:
Sources of Detrital Zircons
Archean and Paleoproterozoic zircon grains > 1800 Ma in North American sedimentary strata are attributed to the Canadian Shield Archean Basement (Ross, 1991; Stewart et al., 2001; Gehrels and Pecha, 2014). Archean grains (2400 to 2900 Ma) in Bakken deposits are interpreted to be derived from the Canadian Shield east of the Williston Basin (Fig. 19). Similarly, grains older than
1800 Ma within age population 1600 to 2200 Ma are attributed to Paleoproterozoic juvenile crust and were likely sourced from the Canadian Shield and/ or Wyoming Province (Fig. 19) (Ross,
1991; Gehrels et al., 2011). Grains within the range of 1600- 1800 Ma are consistent with derivation from juvenile crust of Yavapai- Mazatzal provinces (Fig. 19) reported by several workers from strata of various ages in southern California (Wooden et al., 2013; Gehrels and
Pecha, 2014).
Mesoproterozoic (800- 1500 Ma) grains are attributed to several provinces. Grains within age range from 1300 to 1500 Ma were sourced from granitoids of the U.S. midcontinent (Goodge and
Vervoort, 2006; Gehrels and Pecha, 2014). The dominant age range in Bakken Mesoproterozoic zircon ages is within 1000 to 1200 Ma; this age range is consistent with Grenville provinces and is documented in several paleozoic strata in North America (Gehrels et al., 2000). Zircon grains of
800 to 1000 Ma correspond to zircons derived from southern Appalachian province reported by
Muller et al., (2008). 79
Fig. 19. Provenance interpretation of the Bakken Formation. The map depicts the paleogeography of North America during Early Mississippian. Detrital zircon data suggest three sediment sources (1) The Franklinian belt along the northern margin of North America, (2) The Canadian Shield east of the Williston Basin and (3) southern sources related to the Grenville and Yavapai- Mazatzal orogens. Eastern and southern sourced sediments were likely delivered via fluvial systems and later dispersed within the basin by longshore drift. Northern sediments were likely delivered through fluvial systems and reworked by longshore drift of sediments (Interpretation 2). For interpretation 1 and 2 refer to text in page 77-78. Green- outlined areas indicate regions occupied by Exshaw Formation in "present-day" Alberta Foreland Basin. Paleogeography map is based on reconstructions from Ron Blakey, http:// jan .ucc .nau .edu /rcb7/nam .html. 80
Similar detrital zircon ages (900 to 2100 Ma) were reported from Upper Devonian strata from the Canadian Arctic (Anfinson et al. 2012) and southern British Columbia (Gehrels and Pecha,
2014). These ages were attributed to the Greenland Caledonides (Embry and Klovan, 1976) and possibly to an exotic landmass (Crockerland) that collided with the northern margin of
Laurentia during early to middle Devonian time (Anfinson et al., 2012). Detrital Zircon U-Pb ages, solely, are not enough to identify the sources of this broad population (800 Ma to 2200
Ma) in the Bakken deposits because of possible derivation from multiple provinces; these provinces include south Laurentian provinces (Yavapai- Mazatzal and Grenville provinces) and northern sources associated with the Greenland Caledonides and Timanide Orogeny (Anfinson et al., 2012). Recycling of lower Paleozoic clastic units in the Williston Basin (e.g. Deadwood
Formation) may have also contributed to the sedimentary budget of the Bakken Formation.
The first appearance of 370- 700 Ma zircons in Paleozoic strata in North America is recorded along the Arctic and northern Cordilleran margins within middle Devonian (Beranek et al.,
2010; Gehrels and Pecha, 2014). This age population was also in upper Devonian strata from eastern Alaska and southern British Columbia (Gehrels and Pecha, 2014) and was interpreted to be derived from magmatic and orogenic belts of Crockerland terrain, which was accreted to the northern Laurentian margin during the Franklinian orogeny (Anfinson et al., 2012). The 370-
700 Ma detrital zircon grains recorded in the Bakken Formation are similar to grains attributed to magmatic and orogenic belts of Franklinian mobile belt (Fig, 19) (Anfinson et al., 2012;
Gehrels and Pecha, 2014).
The Franklinian belt comprises upper Neoproterozoic to upper Devonian strata that underwent orogenic deformation during early to middle Silurian and later during late Devonian to 81
Mississippian (Anfinson et al., 2012). The age of 370- 700 Ma Bakken zircon grains overlap ages of strata as well as deformation of the Franklinian belt; this suggests that 370 to 700 Ma zircons of the Bakken Formation were likely recycled from older Franklinian strata or originated from the magmatic activity that was associated with Franklinian deformation and older Silurian deformation phase.
A shift in zircon-age spike from 415 Ma in Bk1 to 423 Ma in Bk2 to 426 Ma in Bk 3 (Fig, 17) suggest that increasingly older rocks supplied the grains to sample Bk2 and Bk 3. This could be explained by progressive unroofing of the Franklinian belt strata.
The Exshaw sample is characterized by a predominant spike at 365 Ma and, as mentioned, this middle Paleozoic population (340 to 400 Ma) constitutes 65% of the sample population. The predominance of a narrow age group suggests that the provenance of these grains was a single source that was proximal to the final deposition area of the grains (Stewart et al.,
2001). The likely source for the abundant zircons of this population is tuff deposits from a nearby volcanic arc (Richards, 2002). Tuff layers are present higher in the Exshaw Formation and have been extensively studied and dated (Richards et al., 2002; Kaufmann, 2006; Bundy et al., 2013) and are possibly the result of volcanic activity arising from the collision of Kootenay
Terrane to the western margin of North America (Richards et al., 2002) during the mid
Devonian to Mississippian Antler orogeny (Root, 2001). Absence of 340 to 365 Ma detrital zircon population in the Bakken Formation suggest that it is improbable that component of the Bakken clastics had a western provenance, and also constrain the extent of the latest
Devonian volcanic ash falls documented in the Exshaw Formation (Fig. 19). 82
5.2.3 Conodont Colour Alteration Index CAI Analysis:
Conodont colour alteration index (CAI) analysis is a technique used to evaluate the organic metamorphism of conodont elements (Epstein, 1977). The organic metamorphism is manifested by a progressive, gradual and irreversible change in element colour from pale yellow (CAI 1) to light to dark brown (CAI 1.5 to 4), to black (CAI 5), and subsequently to grey
(CAI 6), white (CAI 7) and finally translucent (CAI 8) (Voldman et al., 2010). Change in conodont colour is a result of carbon-fixing processes within the conodont element, in which temperature and time are the controlling factors of these changes (Epstein, 1977,
Königshof, 2003). Experimental and field studies postulated that conodont colours can be linked to specific temperatures (Epstein et al, 1977) and time durations (Rejebian et al.,
1987; Burnett, 1987). Therefore, evaluating the conodont colour enables the assessment of the geothermal gradient experienced by the element after burial. Varying structural forms of different conodont types (Barnes et al., 1973; Cook, 1986), host rock lithology (Epstein et al., 1977; Mayr et al., 1978) and, to some extent, conodont size and shape might influence conodont colour alteration (Epstein et al., 1977).
Here, we utilize conodont CAI values and preservational observations to constrain the age of sedimentary source of these elements as well as to investigate any recycling aspect present in the Bakken clastics prior to their final deposition.
Analysis:
Three samples Bk1, Bk2 and Bk3 (see Table 2) were processed for conodonts using the standard procedures following Collinson (1963) and Stone (1987). Conodont fauna recovered were identified and CAI values were determined using a reflected-light binocular 83
microscope by directly comparing the conodont colour to a standard conodont CAI collection.
High magnification images were obtained using a scanning electron microscope (FEI XL30) at the Microscopy and Imaging Facility, University of Calgary.
Results:
Two conodont populations with distinct conodont CAIs were recognized in Bk1 and Bk3 (lower and upper sandy wackestone respectively), whilst sample Bk2 yielded only fragments. The lower CAI (~1.25) (Fig. 20b) conodont population is typified by the conodont species
Polygnathus communis, Siphonodella isosticha and Bispathodus aff. stabilis (Branson &Mehl).
Other species in this fauna include polygnathid biofacies represented by Polygnathus cf.
inornatus (Branson &Mehl), Polygnathus cf. longiposticus and possibly Polygnathus
longiusculus (Fig. 20a). These elements are well preserved and show pale yellow to amber colours with smooth, silky surfaces.
The other population had a higher CAI (~4) (Fig. 20b) and contains reworked, abraded and dark brown elements of Chirognathidae and Protopanderodontidae that overlap zircon ages.
Interpretation:
Occurrence of Siphonodella isosticha and Siphonodella crenulata in the lower CAI (~1.25) conodont population indicates an age of Lower crenulata conodont zones for the upper
wackestone lithofacies (Mb6) (Johnston et al., 2010). Presence of Polygnathus communis in this fauna suggests middle to outer ramp environments (Johnston et al., 2010), which reconciles with the interpreted depositional environment from the lithologies. The CAI value
of ~1.25 suggests that this fauna had only been buried to depths of ~1 km; this combined with the generic environmental preferences indicate that this population is indigenous to Bakken 84 85
Fig. 20 a. Conodonts from the Bakken Formation. All elements are from well 100/06-33-028-2W4/00, (depth 919 m) upper wackestone unit Mb6, unless indicated otherwise. Scale bar= 200µ 1-4. Polygnathus communis, 1-3B are oblique lateral views. 5. Polygnathus longiusculus ? 6. Elictognathid element, P2 element of Siphonodella. 7-10. Polygnathus cf. inornatus (Branson&Mehl). 11. Polygnathus cf. longiposticus (Branson &Mehl). 12. Polygnathus communis carinus (Hass) 13. Polygnathus cf. germanus. 14. Siphonodella isosticha (Haas). 15- 17. Siphonodella crenulata (Branson &Mehl). 18- 19. Bispathodus aff. stabilis (Branson &Mehl). 20-25. Icriodus sp. (Branson &Mehl), note the rounding of the elements, CAI is (=4) for this assemblage. 26-28. Reworked elements of Chirognathidae and Protopanderodontidae, X150, from 101/10-03-30- 29W3 well (depth 888 m) lower wackestone subunit Mb2. Note the rounding of the elements. Scale bar= 200µ
Fig. 20 b. Conodont plate showing two conodont populations with distinct CAIs recognized in the Bakken Formation. Population 1 (elements 1 to 6, scale bar= 200µ) has a CAI of 1.25 which suggests that this fauna had only been buried to depths of ~ 1 km; this combined to the generic age and environmental preferences indicate that this population is indigenous to Bakken deposits. Population 2 (elements 7- 12, scale bar= 200µ consists of reworked elements of Chirognathidae and Protopanderodontidae) has a CAI value of ~4, elements, exhibits a greater degree of roundness and suggests an age range from latest Silurian to middle Devonian. The CAI (~ 4) observed in this population indicates that these elements were exposed to a temperature range of 190°C to 300°C (Epstein et al, 1977). Colours are slightly distorted due to graphic reproduction. 86
deposits. The species recovered from Bk1 and Bk2 have been previously reported from the
Bakken Formation (e.g. Johnston and Henderson, 2005; Johnston et al., 2010).
Presence of Chirognathidae and Protopanderodontidae in the second population (CAI ~4) suggests an age range from latest Silurian to middle Devonian, which partially overlaps ages of
375 to 700 Ma zircon grains. These elements were exposed to temperatures in the range of
190°C to 300°C as indicated by their high CAI (~4) (Epstein, 1977). Such temperatures are typically associated with a depth range of 4 to 6.8 Km based on an average geothermal gradient of 25°C/Km. It is improbable that these elements have originated in the WCSB or in the cratonic platform to the east, because the sedimentary cover was thin during the Devonian and is not likely to generate such high temperatures, and no known igneous intrusions in the vicinity of the basin to provide elevated temperatures. The broad age range and high CAI of these elements suggest that they were likely sourced from Paleozoic sedimentary units of various ages that had undergone orogenic deformation and exhumation. Conodonts are fined sand- sized particles and are transported as such. The wear patterns demonstrated by high CAI elements in the Bakken deposits suggest reworking and transportation prior to final deposition.
Ages and alteration (high CAI) of population 2 conodonts highly resemble the age as well as alteration resulting from deformation of the Franklinian strata. This interpretation strongly supports previous detrital zircon interpretation of possible derivation of Bakken detrital grains from the Franklinian mobile belt in the Canadian Arctic. 87
5.3. Discussion:
5.3.1. Significance of Young Grain Ages (370 to 700 Ma) and High CAI Conodonts Elements
One of the critical findings of this study is the identification of 370- 700 Ma zircon grains in the
Bakken deposits. These grains (370 to 700 Ma) are attributed to North Laurentian volcanic island arcs and continental arcs (Crockerland) associated with the Franklinian, Ellesmerian and
Caledonian Orogens (Fig. 19) (Embry and Klovan, 1976; Anfinson et al., 2012). This interpretation is supported by lack of 370 to 700 Ma volcanic rocks in the North American basement (Gehrels et al., 2011; Hadlari et al., 2012).
Presence of 370- 700 Ma grains and Franklinian-sourced detrital conodonts in the Bakken
Formation not only suggest volcanic and orogenic activity of that age along the North
Laurentian margins, but also indicate regional, continental scale sediment dispersal systems
that transported detrital grain to central parts of North America (e.g. Bakken Formation), as
well as other strata of middle Devonian age in Alaska and southeast British Columbia (Gehrels and Pecha, 2014). Furthermore, the presence of the 370 to 700 Ma detrital zircon grains accompanied with the high CAI conodont elements in the Bakken deposits suggest that a
significant part of the clastics present in the Bakken Formation underwent recycling prior to transportation and deposition in their final location. This interpretation implies that the
Franklinian mobile belt and basin were significant sources of sediment during the middle
Devonian to Mississippian.
5.3.2. Implications for Sediment Transportation and Paleogeography during the late
Devonian- Earliest Mississippian
Clastic-sediments in intracratonic sag basins provide a record of the tectonic evolution of the 88
basins and surrounding cratons, since the sediments originate from the cratons surrounding
the margins of these basins, with no to little clastics being sourced from the seaward side
(Kingston et al., 1983; Mitchell and Reading, 1986). Drainage patterns responsible for terrestrial sediment delivery to interior sag basins typically consist of fluvial systems that transport sediments from the eroded nearby cratons (Einsele, 2000, p575). Data presented in this study suggest that Bakken clastic deposits may have been sourced from eastern, southern and northern provinces (Fig. 19). The presence of detrital zircon grains that originated in the craton
to the east within the Bakken suggests that fluvial systems likely delivered the zircon grains to the Williston Basin. Although no major fluvial systems haven reported bee in the Bakken
Formation, evidence of what could be interpreted as incised valley (Rocanville trend) (Fig. 19)
has been reported by Christopher (1961). The feature known as the Rocanville trend was first reported in southeastern Saskatchewan (Fig. 19) (Christopher, 1961), and represents a
northeast-southwest linear trend in which the lower Bakken shale is often absent and the the
upper part of middle Bakken member is abnormally thick and sharply overlies the Torquay
Formation (Christopher, 1961; Le Fever et al., 1991; Kreis et al., 2006). The absence of lower
Bakken shale and sharp contact between over-thickened middle Bakken and underlying
Torquay Formation within the Rocanville trend suggest deep incision likely during forced marine regression of middle Bakken member. Furthermore, the northeast- southwest strike of the Rocanville trend is oriented more or less perpendicular to the paleo-shoreline strike, which is consistent with orientation of incised valleys developed during marine regression
(Posamentier and Vail, 1988) (Fig 19). Further research is needed to fully understand the nature of the Rocanville trend, as well as to investigate other incised valleys elsewhere in the
Bakken Formation. 89
Tidal channels and/or incised valleys identified in this study likely acted as a sediment transport medium that enhanced sediment reworking by erosion of underlying shoreface and foreshore deposits and delivering sediments to more seaward locations (e.g. Exshaw lowstand systems tract) during late stages of forced marine regression. Significant portion (> 25%) of the detrital zircons of the Bakken Formation was sourced from Franklinian province along the northern margin of Laurentia. Two interpretations could explain the southward transportation of the Franklinian-sourced clastic grains. The first interpretation requires continental-scale fluvial systems that drained in the Williston Basin and transported sediments from source areas in the North and possibly from Canadian Shield as well (Fig. 19). Evidence of detrital zircons sourced from the Canadian Shield (2400-2900 Ma) east of the Williston Basin, as well as zircon grains originated in Yavapai-Mazatzal province south of Williston Basin in Bakken deposits (Fig. 19) requires a fluvial network in which several fluvial systems provided recycled clastics to the Williston Basin, this supports the first interpretation.
The second interpretation consists of long-shore drift of sediments along the shoreline of the
Devonian- Mississippian seaway from source areas to the Williston Basin southward (Fig. 19); this scenario requires shallow marine transportation of sediments across distances in excess of approximately 3500 to 4000 km. The second interpretation is somewhat weak since Northern parts of "present-day" Canada Foreland Basin are occupied by thick accumulations of Exshaw shale (Fig. 19) (Richards et al., 1994; Pedersen, personal communication) with no evidence of late Devonian to early Mississippian shallow marine sandstone deposits ever been reported within these regions. It is worthy to mention that present day limit of the Devonian-
Mississippian interval is an erosional limit (Fig. 19); this implies that a shallow marine system have existed in Northern parts of the passive margin during Devonian-Mississippian time. 90
However, for the time being the first interpretation is favoured, but long shore drift played
critical role in reworking and distributing the sediments along the paleo-shoreline.
5.3.3. Implications for Depositional Style of the WCSB during the Latest Devonian- Earliest
Mississippian
Stratigraphically, upper Devonian-Mississippian strata of WCSB record a major tectonic change manifested by a shift in depositional style and sediment provenances; two pieces of evidence
support this interpretation. First, a basin-wide unconformity separates the basal surface of the
Bakken Formation and its westerly correlative unit the Exshaw Formation from the underlying
Big Valley and Palliser formations respectively. Second, a shift in depositional style is manifested by an abrupt change from carbonate deposition in the underlying Palliser and Big
Valley formations into clastic deposition of the Bakken and Exshaw formations. These changes in depositional style and sediment composition resulted from sea level changes and tectonic activity along the Laurentian margins. Progradation of the Devonian- Mississippian shoreline
to the west and influx of clastic sediments are clearly demonstrated in sharp-based shoreface
deposits of middle Bakken member. These interpretations are consistent with base level falls
associated with forced regressions, and enhanced supply of clastic sediments to the Basin.
6. Conclusions:
First, this study employed a novel approach that utilized U-Pb geochronology of detrital zircons and CAI of detrital conodonts to constrain the provenance and paleogeographic drainage for
Bakken clastic deposits. We acquired over 300 U- Pb ages that represent the first documented
U-Pb ages of detrital zircon obtained from the Bakken Formation. The approach utilized in this 91 study provides four main contributions to understanding the sediment provenance and sediment dispersal systems of Bakken clastic deposits.
1. The Bakken clastics were sourced from multiple sources. These sources include:
early Paleozoic to late Neoproterozoic (370- 700 Ma) Franklinian and Caledonian
sources, Mesoproterozoic (850- 1500 Ma) Grenville province, Paleoproterozoic (1600-
2100Ma) Yavapai-Mazatzal provinces, and early Paleoproterozoic to late Archean
(2400-2900 Ma) Canadian Shield.
2. The Bakken Formation contains reworked conodonts of Chirognathidae and
Protopanderodontidae, suggesting middle to late Ordovician sources (445- 470 Ma) and
other reworked elements may be as young as early to middle Devonian (385- 420 Ma).
These conodonts have much higher CAI values compared to in-situ specimens
indicating that they may have been previously buried to depths in excess of 3 km and/
or exposed to elevated temperatures associated with nearby igneous intrusions prior
to uplift and exhumation. The presence of detrital conodonts of wide time- span in
Bakken deposits indicate profound reworking of older clastics prior to final
transportation and deposition.
3. Our data suggest derivation of at least 25% of Bakken clastics from northern sources
exhumed during the middle Devonian Franklinian orogeny. This interpretation
underlines the significance of the Ellesmerian orogeny in providing sediments to the
interior of North America during middle Paleozoic time.
4. The significant amount of sediment transported southward from the Franklinian belt suggests long distance sediment transportation most likely by fluvial and long-shore 92
drift. Sediments from eastern and southern sources (Canadian Shield, Grenville,
Yavapai-Mazatzal provinces) were likely delivered via fluvial systems and reworked by
shallow marine processes or reworked from deposits to the North that was originally
sourced from the south. Second, using sedimentologic analysis of cores and well log analysis, the Bakken Formation in southeastern Alberta and southwestern Saskatchewan can be subdivided into four systems tracts. 1- The basal systems tract is a transgressive systems tract (TST1) bounded by a sequence boundary (SB 1) and its basinal continuation a maximum regressive surface (MRS1)
below and capped by a maximum flooding surface (MFS1). This transgressive systems
tract encompasses most of the Bakken lower shale member.
2- The basal transgressive systems tract (TST1) transitions into a regressive systems tract
that consists of the lower two-thirds of the middle Bakken Member and the upper
portion of the Bakken lower shale; this regressive systems tract can be further
subdivided into lower high stand systems tract (HST) bounded by a maximum flooding
surface and a basal surface of forced regression regressive(BSFR)/ regressive surface of
marine erosion (RSME) and a falling stage systems tract (FSST) overlying the RSME/
BSFR and capped by a maximum regressive surface (MRS2).
3- At this time progressive portions of the Bakken Formation were exposed and most
of the lower Exshaw member as well as upper Exshaw member were deposited as
lowstand systems tract.
4- The final systems tracts is a transgressive systems tract (TST2) that consists of tidal flat
and channel deposits, the upper fossiliferous wackestone and most of the upper
Bakken shale and is bounded by sequence boundary SB2 and maximum regressive 93
surface MRS2 at the base and maximum flooding surface (MFS2) above.
Cores observations indicate that several higher-order parasequences are present within the
FSST deposits of the Bakken Formation. These parasequences prograde basin-ward, and are bounded by flooding surfaces that are recognizable in cores. Furthermore, core and well log data indicate localized presence of tidal channels and or incised deposits in the study area. While these channels were likely formed during relative sea level fall, accumulation and preservation of tidal deposits may have happened during early transgression following increase in accommodation space during relative sea level rise. 94
CHAPTER 3
3.1. Conclusions:
The work described in the first segment of this paper is concerned with the sequence stratigraphy of the Bakken Formation in southeast Alberta and southwest Saskatchewan. Three systems tracts are interpreted in the Bakken Formation, as well as a lowstand systems tract that is interpreted to include most of the lower shale member and upper siltstone member of the
Exshaw Formation. The Bakken systems tract are (1) a basal transgressive systems tract TST1,
(2) regressive systems tract (RST) and (3) a second trangressive systems tract (TST2). Two sequence boundaries are interpreted here; the first rests at the base of lower Bakken shale, and the second lies on top of foreshore deposits of the middle Bakken. Sequence boundaries translate into maximum regressive surfaces MRS in a basinward direction.
Lower part of middle Bakken member and the upper portion of the lower Bakken shale are interpreted to represent a highstand systems tract (HST). The HST comprises the lower part of
Bakken RST. Upper portion of RST resembles a falling stage systems tract (FSST) deposited during stages of forced marine regression. Lag surfaces within lower Exshaw shale constrain timing of base level fall to be consistent with that of base level fall of the middle Bakken member; this is further supported by detrital zircon ages of ash layers from lower Exshaw member and conodont biostratigraphic ages.
Higher- order parasequences are present within the FSST deposits of the Bakken Formation as observed from cores. Flooding surfaces characterized by sharp transitions from shallow marine sandstone to mudstone beds outline these parasequences in cores, these parasequences prograde in a basinward direction. Cores and well log data indicate localized presence of tidal deposits in the study area. It is interpreted that incised valleys in the Bakken Formation 95
were likely formed during relative sea level fall due to fall in base level and subsequent
erosion. Accumulation and preservation of tidal channel fills likely occurred during early
transgression following increase in accommodation space during relativea level se rise.
Provenance and dispersal systems of the Bakken clastic deposits were the scope of the second
segment of this paper. Detrital zircon U-Pb analysis and reworked conodonts indicate sediment
transport from local basement as well as thrust-belt sources to the north, northeast and
south. These sources include (1) Franklinian and Caledonian sources (370- 700 Ma), (2)
Grenville province (850-1500 Ma grains), (3) Yavapai-Mazatzal provinces (1600- 2100 Ma
grains), and (4) early Paleoproterozoic to late Archean Canadian Shield source (grains
2400-2900 Ma). The reworked Chirognathidae and Protopanderodontidae conodonts suggest Middle to Upper
Ordovician sources (445- 470 Ma) that could be as young Lower to Middle Devonian (385- 420
Ma). The higher CAI values (CAI ~4) suggest that they may have been buried to depths in excess
of 3 km prior to uplift and exhumation. Ages and high CAI of these elements suggest that they were likely sourced from the Franklinian mobile belt that consists of deformed Upper
Neoproterozoic to Upper Devonian clastic rocks (Embry, 1991; Anfinson et al., 2012).
Our data suggest derivation of at least 25% of Bakken clastics from northern sources exhumed and recycled during the middle Devonian Franklinian orogeny. This interpretation underlines the significance of the Ellesmerian orogeny in providing sedimentary budget to the interior of
North America during middle Paleozoic time. This interpretation supports long distance sediment transportation most likely by fluvial and long-shore drift. Sediments from eastern and southern sources (Canadian Shield, Grenville, Yavapai-Mazatzal provinces) were likely delivered via fluvial systems and reworked by shallow marine processes. 96
3.2. Future work:
Although the results presented in this paper highlight several aspects of the depositional systems, sequence stratigraphy and provenance of the Bakken Formation, numerous questions still remain to be researched. The following are some of these questions and some suggestions on how to resolve these questions.
- How can conodont biostratigraphy contribute to refining the correlations of Bakken
parasequences?
Ideally, a parasequence is a relatively conformable succession of related beds and bedsets that were formed in distinct depositional and ecological environments. These parasequences dip basinward and consist of predictable facies succession. A group of parasequences will constitute a parasequence set that can prograde or retrograde depending on changes in base level. As a result of this lateral migration, parasequences of different ages, environments and different fossil assemblage could be juxtaposed laterally. Therefore, by incorporating high- resolution conodont biostratigraphy to sequence stratigraphy identifying and mapping parasequences could be achieved. Although the discrimination of immediately-adjacent
Bakken parasequences is not likely to be achieved by conodont biostratigraphy (since adjacent parasequences are within subzonal level), regional time correlations and chronological order of the Bakken Formation deposition throughout the basin could be achieved by conducting detailed sequence-biostratigraphy. 97
- What are the mechanisms and processes responsible for shale deposition in the
Bakken and Exshaw formations?
Gravity settling of fine grains below wave and storm base level has been proposed as a mechanism of Bakken shale deposition (Kent, 1958; Christopher, 1961; Richards, 1989; Smith and Bustin, 1996; Kreis et al., 2006). This interpretation was based on the homogeneous character and by the lack current structures in the shale deposits, yet, it fails to fully explain the long distance transportation of the fine particles and the abundance of organic matter within the shale deposits. By incorporating x-ray diffraction techniques and high resolution
SEM microscopy, shale composition and depositional bedforms could be demonstrated. This will provide results that could possibly explain the transportation mechanisms of fine particles. Such interpretations will have direct implication on refining the depositional model of the Bakken and Exshaw shale as well as explaining the nature of organic matter aggregates within these deposits.
- Is there any non-marine facies (fluvial deposits) present in the eastern regions of the
Bakken Formation?
In principle, if the shoreline of the Bakken Formation prograded several hundred kilometers seaward to the southwest, the accompanying fluvial system must have migrated similar distances. Yet, no fluvial deposits have been reported in the Bakken Formation to date. The lack of fluvial deposits in the study area suggests that marine transgression, following deposition of the regressive systems tracts of the Bakken, might have been slow and erosive resulting in reworking of any fluvial deposits in the region. The problem could be 98
resolved by investigating the sedimentology of the Bakken deposits in regions where the
Bakken deposits are anomalously thick, for example the Rocanville trend in southeast
Saskatchewan. Fluvial channels typically play a significant role as conduits for hydrocarbon migration, and could be targeted for hydrocarbon production as well. 99
References Ahern, J. L., and S. R. Mrkvicka (1984), A mechanical and thermal model for the evolution of the Williston Basin, Tectonics, 3(1), 79–102, doi:10.1029/TC003i001p00079. Anfinson, O. A., Leier, A. L., Embry, A. F., & Dewing, K. (2012). Detrital zircon geochronology and provenance of the Neoproterozoic to Late Devonian Franklinian basin, Canadian Arctic Islands. Geological Society of America Bulletin, 124(3-4), 415-430. Anfinson, O. A., Leier, A. L., Gaschnig, R., Embry, A. F., & Dewing, K. (2012). U–Pb and Hf isotopic data from Franklinian Basin strata: insights into the nature of Crockerland and the timing of accretion, Canadian Arctic Islands. Canadian Journal of Earth Sciences, 49(11), 1316-1328. Angulo, S., Buatois, L.A., Halabura, S., 2008. Paleoenvironmental and sequence stratigraphic Reinterpretation of the Upper Devonian–Lower Mississippian Bakken Formation of subsurface Saskatchewan: integrating Sedimentological and ichnological data. Summary of Investigations 2008, v. 1, Saskatchewan Geological Survey, Saskatchewan Ministry of Energy and Resources, Miscellaneous Report 2008-4.1.
Angulo, S., & Buatois, L. A. (2011). Petrophysical characterization of sedimentary facies from the Upper Devonian–Lower Mississippian Bakken Formation in the Williston Basin, southeastern Saskatchewan. Canada: Summary of Investigations, 2011-4.
Angulo, S., & Buatois, L. A. (2012). Ichnology of a Late Devonian–Early Carboniferous low- energy seaway: The Bakken Formation of subsurface Saskatchewan, Canada: Assessing paleoenvironmental controls and biotic responses. Palaeogeography, Palaeoclimatology, Barnes, C. R., Rexford, C. B., and Miller, J. F. 1973. Lower Paleozoic conodonts provincialism. In Conodont paleoecology. Edited by F. H. T. Rhodes. Geological Society of America, Special Paper 141, pp. 157- 190. Beranek, L. P., Mortensen, J. K., Lane, L. S., Allen, T. L., Fraser, T. A., Hadlari, T., & Zantvoort, W. G. (2010). Detrital zircon geochronology of the western Ellesmerian clastic wedge, northwestern Canada: Insights on Arctic tectonics and the evolution of the northern Cordilleran miogeocline. Geological Society of America Bulletin, 122(11-12), 1899-1911.
Burnett, R. D. (1987). Regional maturation patterns for late Visean (Carboniferous, Dinantian) rocks of northern England based on mapping of conodont colour. Irish journal of earth sciences, 165-185.
Bundy, K., Ekhoff, J., & Schmitz, M. (2013). Geochronology in the Exshaw Formation: correlation with the Hangenberg extinction event. Poster, GSA annual meeting in Denver 2013.
Caplan, M.L. and Bustin, R.M., 1998. Sedimentology and sequence stratigraphy of Devonian- Carboniferous strata, Southern Alberta, Bulletin of Canadian Petroleum Geology, v 46, p. 487– 514 100
Catuneanu, O. (2002). Sequence stratigraphy of clastic systems: concepts, merits, and pitfalls. Journal of African Earth Sciences, 35(1), 1-43.
Catuneanu, O., Abreu, V., Bhattacharya, J. P., Blum, M. D., Dalrymple, R. W., Eriksson, P. G., ... & Winker, C. (2009). Towards the standardization of sequence stratigraphy. Earth-Science Reviews, 92(1), 1-33. Christopher, J.E., 1961. Transitional Devonian–Mississippian Formations of southern Saskatchewan, Saskatchewan Department of Mineral Resources, Report 66, 103 p.
Charvin, K., Hampson, G. J., Gallagher, K. L., Storms, J. E., & Labourdette, R. (2011). Characterization of controls on high-resolution stratigraphic architecture in wave-dominated shoreface–shelf parasequences using inverse numerical modeling. Journal of Sedimentary Research, 81(8), 562-578.
Clifton, H. E. (2006). A reexamination of facies models for clastic shorelines. SPECIAL PUBLICATION- SEPM, 84, 293.
Cook T.D., and Bally A.W. (1975) Stratigraphic atlas of North and Central America. Univ Press, Princeton, 272 pp Collinson, C. (1963). Techniques for the collecting and processing of conodonts. Ill. Geol. Surv. Circ, 343, 1-16.
Einsele, G. (2000). Sedimentary basins: evolution, facies, and sediment budget. Springer. P575
Embry, A., & Klovan, J. E. (1976). The Middle-Upper Devonian clastic wedge of the Franklinian geosyncline. Bulletin of Canadian Petroleum Geology, 24(4), 485-639.
Embry, A. F., & Johannessen, E. P. (1992). TR sequence stratigraphy, facies analysis and reservoir distribution in the uppermost Triassic-Lower Jurassic succession, western Sverdrup Basin, Arctic Canada. Arctic geology and petroleum potential, 2, 121-146.
Embry, A.F., 2002. Transgressive-regressive (T-R) sequence stratigraphy. In: J.M. Armentrout and N.C. Rosen (eds.). Sequence Stratigraphic Interpretations for Exploration and Production. Gulf Coast Section SEPM, Conference Proceedings, Houston, Texas, p. 151–172.
Embry, A., Johannessen, E., Owen, D., Beauchamp, B., & Gianolla, P. (2007). Sequence Stratigraphy as a “Concrete” Stratigraphic. Epstein, A. G., Epstein, J. B. and Harris, L. D. 1977. Conodont color alteration — an index to organic metamorphism. U.S. Geological Survey Professional Paper, 995: 1–27.
Faill, R. T. (1997). A geologic history of the north-central Appalachians; Part 1, Orogenesis from the Mesoproterozoic through the Taconic Orogeny. American Journal of Science, 297(6), 551-619. 101
Fuzesy, L. M. (1960). Correlation and subcrops of the Mississippian strata in southeastern and south-central Saskatchewan. Saskatchewan Department of Mineral Resources, Geological Sciences Branch, Geology Division.
Galloway, W. E. (1989). Genetic stratigraphic sequences in basin analysis II: application to northwest Gulf of Mexico Cenozoic basin. AAPG Bulletin, 73(2), 143-154.
Gehrels, G.E., and Dickinson, W.R., (1995) Detrital zircon provenance of Cambrian to Triassic miogeoclinal and eugeoclinal strata in Nevada: American Journal of Science, v. 295, p. 18–48, doi: 10.2475 /ajs .295 .1.18.
Gehrels, G.E., Dickinson, W.R., Riley, B.C.D., Finney, S.C., and Smith, M.T., (2000). Detrital zircon geochronology of the Roberts Mountains allochthon, Nevada, in Gehrels, G.E., and Soreghan, M.J., eds., Paleozoic and Triassic Paleogeography and Tectonics of Western Nevada and Northern California: Boulder, Colorado, Geological Society of America, Special Paper 347, p. 19–42, doi: 10.1130 /0-8137 - 2347 -7.19.
Gehrels, G. E., Blakey, R., Karlstrom, K. E., Timmons, J. M., Dickinson, B., & Pecha, M. (2011). Detrital zircon U-Pb geochronology of Paleozoic strata in the Grand Canyon, Arizona. Lithosphere, 3(3), 183-200
Gehrels, G. (2011). Detrital zircon U-Pb geochronology: Current methods and new opportunities. Tectonics of sedimentary basins: Recent advances, 45-62.
Gehrels, G., & Pecha, M. (2014). Detrital zircon U-Pb geochronology and Hf isotope geochemistry of Paleozoic and Triassic passive margin strata of western North America. Geosphere, 10(1), 49-65. Gerhard, L. C., & Anderson, S. B. (1988). Geology of the Williston basin (United States portion). Sedimentary Cover; North American Craton; US Geological Society of America, The Geology of North America, 2, 221-241. Gerhard, L. C., Anderson, S. B., & Fischer, D. W. (1990). Petroleum geology of the Williston Basin. Interior cratonic basins: AAPG Memoir, 51, 507-559. Gibson, R. I. (1995). Basement tectonics and hydrocarbon production in the Williston Basin: An interpretive overview. Williston Basin Symposium. Goodge, J. W., & Vervoort, J. D. (2006). Origin of Mesoproterozoic A-type granites in Laurentia: Hf isotope evidence. Earth and Planetary Science Letters, 243(3), 711-731. Gradstein, F. M., Ogg, G., & Schmitz, M. (Eds.). (2012). The Geologic Time Scale 2012 2-Volume Set. Elsevier. Greggs, D., (2006). The effects of reactivating Basement structure on the overlying Devonian sedimentary cover in the Western Canada Sedimentary Basin. Phd Thesis, University of Calgary. 102
Hadlari, T., Davis, W. J., Dewing, K., Heaman, L. M., Lemieux, Y., Ootes, L. B.R. Pratt & Pyle, L. J. (2012). Two detrital zircon signatures for the Cambrian passive margin of northern Laurentia highlighted by new U-Pb results from northern Canada. Geological Society of America Bulletin, 124(7-8), 1155-1168. Halabura, S., Buatois, L.A., Angulo, S. and Piche, L., (2007). From source to trap: A review of the Bakken petroleum system, Upper Devonian-Mississippian, southeast Saskatchewan, Summary of Investigations 2007, Saskatchewan Industry and Resources, CD-ROM paper A-4, 8 p. Halbertsma, H. L. (1994). Devonian Wabamun Group of the western Canada sedimentary basin. Geological Atlas of the Western Canada Sedimentary Basin, 203-220. Hunt, D., & Tucker, M. E. (1995). Stranded parasequences and the forced regressive wedge systems tract: deposition during base-level fall—reply. Sedimentary Geology, 95(1), 147-160. Hunter, R. E., Clifton, H. E., & Phillips, R. L. (1979). Depositional processes, sedimentary structures, and predicted vertical sequences in barred nearshore systems, southern Oregon coast. Journal of Sedimentary Research, 49(3).
Isaacson, P. E., Diaz-Martinez, E., Grader, G. W., Kalvoda, J., Babek, O., & Devuyst, F. X. (2008). Late Devonian–earliest Mississippian glaciation in Gondwanaland and its biogeographic consequences. Palaeogeography, Palaeoclimatology, Palaeoecology, 268(3), 126-142. Johnston, D. I., & Henderson, C. M. (2005). Disrupted conodont bedding plane assemblages, upper Bakken Formation (Lower Mississippian) from the subsurface of Western Canada. Journal Information, 79(4). Johnston, D.I., Henderson, C.M. and Schmidt, M.J.,( 2010). Upper Devonian to Lower Mississippian conodont biostratigraphy of uppermost Wabamun Group and Palliser Formation to lowermost Banff and Lodgepole formations, southern Alberta and southeastern British Columbia, Canada: implications for correlations and sequence stratigraphy; Bulletin of Canadian Petroleum Geology, v. 58, no. 4, p. 295–341.
Kaiser, S.I., Becker, R.T., Steuber, T, Aboussalam, Z.S., (2011). Climate-controlled mass extinctions, facies and sea-level changes around the Devonian-Carboniferous boundary in the eastern Anti-Atlas (SE Morocco), Palaeogeography, Palaeoclimatology, Palaeoecology 310 (2011) 340–364
Kaufmann, B. (2006). Calibrating the Devonian time scale: a synthesis of U–Pb ID–TIMS ages and conodont stratigraphy. Earth-Science Reviews, 76(3), 175-190. Kents, P. (1959). Three Forks and Bakken stratigraphy in west central Saskatchewan. Department of Mineral Resources, Geological Sciences Branch, Sedimentary Geology Division. Kent, D. M. (1994). Paleogeographic evolution of the cratonic platform-Cambrian to Triassic. Geological Atlas of the Western Canada Sedimentary Basin, 69-86. 103
Kent, D. M., & Christopher, J. E. (1994). Geological history of the Williston Basin and Sweetgrass arch. Geological atlas of the Western Canada Sedimentary basin, published jointly by the Canadian Society of Petroleum Geologists and the Alberta Research Council, 421-429.
Kingston, D. R., Dishroon, C. P., & Williams, P. A. (1983). Global basin classification system. AAPG bulletin, 67(12), 2175-2193.
Kohlruss, D., and E. Nickel, 2009, Facies analysis of the Upper Devonian–Lower Mississippian Bakken Formation, southeastern Saskatchewan: Summary of Investigations 2009, Saskatchewan Geological Survey, Saskatchewan Ministry of Energy and Resources, Miscellaneous Report 2009-4.1, Paper A-6, v. 1, 11 p.
Königshof, P. (2003). Conodont deformation patterns and textural alteration in Paleozoic conodonts: examples from Germany and France. Senckenbergiana lethaea, 83(1-2), 149-156.
Kreis, K., Costa, A., & Osadetz, K. (2006). Hydrocarbon potential of Bakken and Torquay formations, southeastern Saskatchewan; In: Gilboy, C.F., Whittaker, S.G. (Eds.), Saskatchewan and Northern Plains Oil & Gas Symposium 2006: Saskatchewan Geological Society Special Publication, 19, pp, 118-137
Lawton, T. F., Bradford, I. A., Vega, F. J., Gehrels, G. E., & Amato, J. M. (2009). Provenance of Upper Cretaceous–Paleogene sandstones in the foreland basin system of the Sierra Madre Oriental, northeastern Mexico, and its bearing on fluvial dispersal systems of the Mexican Laramide Province. Geological Society of America Bulletin, 121(5-6), 820-836.
LeFever, J.A., Martiniuk, C.D., Dancsok, E.F.R., Mahnic, P.A., 1991. Petroleum potential of the middle member Bakken Formation, Williston Basin, In: Christopher, J.E., Haidl, F. (Eds.), Sixth International Williston Basin Symposium: Saskatchewan Geological Society Special Publication, 11, pp. 74–94.
Ludwig, K. R. (2003). User's manual for Isoplot 3.00: a geochronological toolkit for Microsoft Excel (No. 4). Kenneth R. Ludwig. Mageau, K. R., Leckie, D., & Maguire, R. (2001). The Bakken Formation of West-Central Saskatchewan and East-Central Albert: A Depositional History, Stratigraphy and Facies Distribution of Bakken Shelf Sand Ridges Using the Cactus Lake Field as a Working Model.
Maldonado, D. N. (2014). Chemostratigraphy And Geochemical Constraints On The Deposition Of The Bakken Formation, Williston Basin, Eastern Montana And Western North Dakota. Unpublished MSc thesis, University of Texas Arlington, 90 pp. Mayr, U., Uyeno, T. T. & Barnes, C. R. 1978. Subsurface stratigraphy, conodont zonation, and organic metamorphism of the Lower Paleozoic succession, Bjorne Peninsula, Ellesmere Island, District of Franklin 393–398. Geological Survey of Canada Paper 78–1A 104
McKerrow, W. S., Mac Niocaill, C., & Dewey, J. F. (2000). The Caledonian orogeny redefined. Journal of the Geological Society, 157(6), 1149-1154. Meijer-Drees, N., & Johnston, D. I. (1996). Famennian and Tournaisian biostratigraphy of the Big Valley, Exshaw and Bakken formations, southeastern Alberta and southwestern Saskatchewan. Bulletin of canadian petroleum geology, 44(4), 683-694. Meissner, F. F. (1978). Petroleum geology of the Bakken Formation Williston basin, North Dakota and Montana. 19-42. Mitchell, A. H. G., & Reading, H. G. (1986). Sedimentation and tectonics. Sedimentary environments and facies, 471-519.
Patchett, P. J., Roth, M. A., Canale, B. S., De Freitas, T. A., Harrison, J. C., Embry, A. F., & Ross, G. M. (1999). Nd isotopes, geochemistry, and constraints on sources of sediments in the Franklinian mobile belt, Arctic Canada. Geological Society of America Bulletin, 111(4), 578-589. Pemberton, S. G., MacEachern, J. A., & Frey, R. W. (1992). Trace fossil facies models: environmental and allostratigraphic significance. Facies models: response to sea level change, 47-72. Pitman, J.K., Price, L.C., and LeFever, J.A. 2001. Diagenesis and Fracture Development in the Bakken Formation, Williston Basin: Implications for Reservoir Quality in the Middle Member. U.S. Geological Survey Professional Paper 1653. Plint, A. G. (1988). Sharp-based shoreface sequences and “offshore bars” in the Cardium Formation of Alberta: their relationship to relative changes in sea level. Sea-Level Changes: An Integrated Approach: SEPM, Special Publication, 42, 357-370. Plint, A. G., & Nummedal, D. (2000). The falling stage systems tract: recognition and importance in sequence stratigraphic analysis. Geological Society, London, Special Publications, 172(1), 1-17. Plint, A.G. 2010. Wave- and storm-dominated shoreline and shallow marine systems. In Facies models. 4th ed. Edited by R.W. Dalrymple and N.P. James. Geological Association of Canada. Chap. 8, pp. 167-199.
Pollastro, R. M., Roberts, L. N. R., Cook, T. A., & Lewan, M. D. (2008). Assessment of Undiscovered Technically Recoverable Oil and Gas Resources of the Bakken Formation, Williston Basin, Montana and North Dakota, 2008. AAPG Search and Discovery Article #90092©2009 Posamentier, H. W. (1988). Eustatic controls on clastic deposition II—sequence and systems tract models. An Integrated Approach: SEPM Special Publication No.42. Rejebian, V. A., Harris, A. G., & Huebner, J. S. (1987). Conodont color and textural alteration: An index to regional metamorphism, contact metamorphism, and hydrothermal alteration. Geological Society of America Bulletin, 99(4), 471-479. 105
Richards, B. C., Barclay, J. E., Bryan, D., Hartling, A., Henderson, C. M., & Hinds, R. C. (1994). Carboniferous strata of the Western Canada sedimentary basin. Geological Atlas of the Western Canada Sedimentary Basin, GD Mossop and I. Shetsen (comp.), Canadian Society of Petroleum Geologists and Alberta Research Council, Special Report, 4, 221-250. Richards, B. C., 1989. Upper Kaskaskia sequence—uppermost Devonian and Lower Carboniferous, in B. D. Ricketts, ed., Western Canadian sedimentary basin, a case history: Calgary, Alberta, Canadian Society of Petroleum Geologists, p. 165–201. Richards, B.C., Ross, G.M., Utting, J., 2002. U–Pb geochronology, lithology and biostratigraphy of tuff in the upper Famennian to Tournaisian Exshaw Formation: evidence for a mid-Paleozoic magmatic arc on the northwestern margin of North America. Memoir – Canadian Society of Petroleum Geologists 19, 158–207. Ross, G. M., Parrish, R. R., Villeneuve, M. E., & Bowring, S. A. (1991). Geophysics and geochronology of the crystalline basement of the Alberta Basin, western Canada. Canadian Journal of Earth Sciences, 28(4), 512-522. Root, K. G. (2001). Devonian Antler fold and thrust belt and foreland basin development in the southern Canadian Cordillera: implications for the Western Canada Sedimentary Basin. Bulletin of Canadian Petroleum Geology, 49(1), 7-36. Sláma, J., Košler, J., Condon, D. J., Crowley, J. L., Gerdes, A., Hanchar, J. M., ... & Whitehouse, M. J. (2008). Plešovice zircon—a new natural reference material for U–Pb and Hf isotopic microanalysis. Chemical Geology, 249(1), 1-35. Sloss, L. L. (1962). Stratigraphic models in exploration. Journal of Sedimentary Research, 32(3). Smith, M.G., Bustin, R.M., 1996. Lithofacies and paleoenvironments of the Upper Devonian and Lower Mississippian Bakken Formation, Williston Basin, Bulletin of Canadian Petroleum Geology 44 (3), 495–507. Smith, M.G., Bustin, R.M., 1998. Production and preservation of organic matter during deposition of the Bakken Formation (Late Devonian and Early Mississippian), Williston Basin, Palaeogeography, Palaeoclimatology, Palaeoecology 142, 185–200. Smith, M.G., Bustin, R.M., 2000. Late Devonian and Early Mississippian Bakken andE xshaw black shale source rocks, Western Canada Sedimentary Basin: a sequence stratigraphic interpretation. American Association of Petroleum Geologists Bulletin 84, 940–960. Sonnenberg, S. A., Pramudito, A. 2009. Petroleum geology of the giant Elm Coulee field, Williston Basin: American Association of Petroleum Geologists Bulletin, v. 93, p. 1127 1153. Stewart, J. H., Gehrels, G. E., Barth, A. P., Link, P. K., Christie-Blick, N., & Wrucke, C. T. (2001). Detrital zircon provenance of Mesoproterozoic to Cambrian arenites in the western United States and northwestern Mexico. Geological Society of America Bulletin, 113(10), 1343-1356. Stive, M. J., & De Vriend, H. J. (1995). Modelling shoreface profile evolution. Marine Geology, 126(1), 235-248.
Stone, J. (1987). Review of investigative techniques used in the study of conodonts. Conodonts: Investigative Techniques and Applications, Ellis Horwood Limited, Chichester, 17-34. Thomas, W. A. (2011). Detrital-zircon geochronology and sedimentary provenance. Lithosphere, 3(4), 304-308.
Toews, C.N. (2005) Sedimentology and stratigraphic architecture of the Bakken Formation (Devonian-Mississippian) west- central Saskatchewan, Unpublished M.Sc thesis, University of Regina, 314 pp.
Vail, P. R., Mitchum Jr, R. M., & Thompson III, S. (1977). Seismic Stratigraphy and Global Changes of Sea Level: Part 4. Global Cycles of Relative Changes of Sea Level.: Section 2. Application of Seismic Reflection Configuration to Stratigraphic Interpretation. Van Wagoner, J. C. (1988). An overview of the fundamentals of sequence stratigraphy and key definitions. Voldman, G. G., Bustos-Marún, R. A., & Albanesi, G. L. (2010). Calculation of the conodont Color Alteration Index (CAI) for complex thermal histories. International Journal of Coal Geology, 82(1), 45- 50. Walker, R.G. and Plint, A.G. (1992) Wave-and Storm-Dominated Shallow Marine Systems. In: Walker, R.G. and James, N.P., Eds., Facies Models: Response to Sea Level Change, Geological Association of Canada, Newfoundland, 219-238. Wheeler, H. E. (1964). Baselevel, lithosphere surface, and time-stratigraphy. Geological Society of America Bulletin, 75(7), 599-610. Wooden, J. L., Barth, A. P., & Mueller, P. A. (2013). Crustal growth and tectonic evolution of the Mojave crustal province: Insights from hafnium isotope systematics in zircons. Lithosphere, 5(1), 17-28.