sign in sign up
<<
Home , Antrim Shale, Bedford Shale, Bell Shale, Berea Sandstone, Coldwater Shale, Ellsworth Shale

Western University ScholarWorks at WMU

Master's Theses Graduate College

8-2016

Stratigraphy of the Upper -Lower : Review and Revision with an Emphasis on the Ellsworth System

Bryan J. Currie

Follow this and additional works at: https://scholarworks.wmich.edu/masters_theses

Part of the Geology Commons

Recommended Citation Currie, Bryan J., "Stratigraphy of the Upper Devonian-Lower Mississippian Michigan Basin: Review and Revision with an Emphasis on the Ellsworth Petroleum System" (2016). Master's Theses. 721. https://scholarworks.wmich.edu/masters_theses/721

This Masters Thesis-Open Access is brought to you for free and open access by the Graduate College at ScholarWorks at WMU. It has been accepted for inclusion in Master's Theses by an authorized administrator of ScholarWorks at WMU. For more information, please contact [email protected]. STRATIGRAPHY OF THE UPPER DEVONIAN-LOWER MISSISSIPPIAN MICHIGAN BASIN: REVIEW AND REVISION WITH AN EMPHASIS ON THE ELLSWORTH PETROLEUM SYSTEM

by Bryan J. Currie

A thesis submitted to the Graduate College in partial fulfillment of the requirements for the degree of Master of Science Geology Western Michigan University August 2016

Thesis Committee: Dave Barnes, Ph.D., Chair William Harrison III, Ph.D. Peter Voice, Ph. STRATIGRAPHY OF THE UPPER DEVONIAN-LOWER MISSISSIPPIAN MICHIGAN BASIN: REVIEW AND REVISION WITH AN EMPHASIS ON THE ELLSWORTH PETROLEUM SYSTEM

Bryan J. Currie, M.S. Western Michigan University, 2016

Compression associated with the formation of the Transcontinental Arch and the

Acadian Mountains initiated subsidence in the Michigan Basin and lead to a depositional switch in the Michigan Basin from an oxic shallow-water carbonate platform (Traverse

Group and Squaw Bay ) to a deep anaerobic sea floor and the beginning of

Upper Devonian Antrim . The vertical distribution of the different Upper

Devonian-Lower Mississippian formations and members demonstrates large scale cycles of anaerobic-dysaerobic transitions attributed fluctuations in sea-level, different rates of subsidence and sediment influx triggered by different orogenic events to.

The regional understanding and the distribution of the different lithologies is still poorly understood. This work expands on the mapping of the regional deposits of Upper

Devonian-Lower Mississippian by incorporating modern logs and high resolution chemostratigraphic data. New stratigraphic relationships suggest the expanded much further east than previously suggested and the Bedford-Berea sequence postdates the Ellsworth Shale eliminating any possibilities of inter-tonguing. In addition to the expansion of the regional framework, special consideration is given to the

Ellsworth Shale petroleum system. © 2016 Bryan J. Currie

iv ACKNOWLEDGMENTS

I would first and foremost like to sincerely thank my committee chair, Dr. Dave

Barnes for not only encouraging me throughout this project but for also introducing me to new ideas in and out of the classroom. I was fortunate enough to take a majority of the courses he offered prior to his retirement. I would also like to thank the additional members of my committee, Dr. William Harrison III and Dr. Peter Voice for their constant guidance and presence. Dr. Voice is an encyclopedia of resources and is always quick to give feedback. He has supplied me with numerous documents which helped evolve my research. Dr. Harrison’s knowledge of the Michigan Basin is astonishing. He is always thinking of different ways to keep the students involved with the industry and community. The work Dr. Harrison and his wife, Linda Harrison, have put into the

MGRRE facility really shows and is a large asset to Western’s Geoscience Department.

I would also like to thank Jennifer Trout, Jon Garrett, Matthew Rine, Agam Arief

Suhaimi, Cameron Manche, Zaid Naseer Nadhim Nadhim, Steve Kaczmarek, Nick

Panyard, Kirk Wagenvelt, Kyle Cox, Frank Sattler, and all other students and faculty who helped make my experience at Western Michigan University a pleasant one.

Lastly, I would like to thank my parents, Peter and Linda Currie, and my siblings,

Sean, Lauren, and Austin for their constant love and support throughout my academic career. You each have contributed in different ways and I will forever be grateful.

Bryan J. Currie

ii TABLE OF CONTENTS

ACKNOWLEDGMENTS ...... ii LIST OF FIGURES ...... vi LIST OF TABLES ...... ix ...... 1 INTRODUCTION ...... 1 Purpose of Study ...... 1

Regional Stratigraphy ...... 1 Ellsworth Petroleum System ...... 5 Objective ...... 5 Geologic Setting ...... 7 Stratigraphy ...... 10

Squaw Bay Limestone and ...... 12

Antrim Shale ...... 14

Ellsworth Shale ...... 18

Bedford Shale ...... 18

Berea ...... 19

Sunbury Shale ...... 20 Previous Research...... 20 Early Nomenclature ...... 21

United States Geological Survey (USGS) Mapping Project ...... 21

Eastern Project (ESGP) ...... 23 Production History ...... 26

Antrim Shale ...... 26

Ellsworth Shale ...... 27

Berea Sandstone ...... 28 ...... 30 METHODOLOGY AND DATA ...... 30 Wireline Log Analysis Method ...... 30 Gamma Ray...... 31 iii Table of Contents-Continued

Bulk Density ...... 32

Neutron Porosity ...... 33

Photoelectric ...... 33 Cross Sections and Maps Data ...... 34 X-Ray Fluorescence Method ...... 36 X-Ray Fluorescence Data ...... 39

Limitations...... 40 Ellsworth Petroleum System Method ...... 41 Source Data ...... 43

Kerogen Type ...... 43

Maturation ...... 44

TOC ...... 48

Gas Analysis ...... 49 ...... 51 RESULTS AND DISCUSSION ...... 51 Regional Stratigraphy ...... 51 ...... 55

Antrim Shale ...... 57

Norwood Member Results ...... 59

Paxton Member Results ...... 63

Lachine Member Results ...... 67

Ellsworth Shale Results ...... 70

Upper Antrim Member Results ...... 77

Bedford Shale Results ...... 81

Berea Sandstone Results ...... 84

Sunbury Shale Results ...... 88 Ellsworth Petroleum System Results and Discussion ...... 92

Source ...... 92

Reservoir ...... 106 iv Table of Contents-Continued

Seal ...... 107 ...... 108 CONCLUSION ...... 108 Regional Stratigraphy ...... 108 Ellsworth Petroleum System ...... 110 FUTURE WORK ...... 111 REFERENCES ...... 113 APPENDIX A ...... 121 CROSS SECTIONS ...... 121 APPENDIX B ...... 127 CROSS SECTION WELL INFORMATION ...... 127 APPENDIX C ...... 136 FORMATION TOP PICKS ...... 136

v LIST OF FIGURES

1.1 Cross-section of Ells (1979) Units...... 3 1.2 Cross-section of Ells (1979) Units with Antrim Members...... 4 1.3 Map of the Upper Devonian paleogeography...... 8 1.4 Map of the sturctural features associated with Devonian Michigan Basin...... 9 1.5 Stratigraphic nomenclature defined by Gutschick and Sandberg (1991)...... 11 1.6 Isopach map of the Upper Devonian-Lower Mississippian...... 12 1.7 Structure map of the Squaw Bay Limestone ...... 13 1.8 Type log for the Upper Devonian - Lower Mississippian...... 17 1.9 Regional correlation from Gutschick and Sandberg (1991)...... 25 1.10 Map of oil and natural gas fields...... 27 2.1 Location of the 193 wells used in this study...... 31 2.2 Cross section network ...... 35 2.3 Geographic locations of XRD cores...... 37 2.4 A simplified diagram of the handheld XRF device...... 39 2.5 Location of wells with maturation data...... 46 2.6 Well locations of wells with TOC data ...... 49 3.1 Type log facies ...... 53 3.2 Structure map of the Squaw Bay Limestone ...... 56 3.3 Isopach map of the Lower Antrim...... 58 3.4 XRF Si/Al cross-plot...... 59 3.5 Norwood isopach ...... 61 3.6 Norwood isopach comparisson...... 63 3.7 Paxton isopach ...... 64 3.8 XRF for the Krocker 1-17...... 66 3.9 Lachine isopach ...... 68 3.10 XRF cross-section for S and Fe ...... 69 3.11 Depositional model for the Lachine-Ellsworth transition period ...... 70 3.12 Ellsworth isopach...... 72 3.13 XRF cross-section ...... 75

vi List of Figures-Continued

3.14 Upper Antrim isopach...... 78 3.15 Depositional model for the Upper Antrim Member ...... 79 3.16 Erosional model from Matthews (1989) ...... 80 3.17 Bedford isopach ...... 82 3.18 Bedford prodelta lobes...... 84 3.19 Berea isopach ...... 85 3.20 Cross-section of Berea Channel...... 87 3.21 Inferred orientation of the Berea Channels...... 87 3.22 Sunbury isopach ...... 89 3.23 Map of formations below the Sunbury Shale ...... 91 3.24 The TOC spatial distribution in the Antrim members ...... 93 3.25 TOC spatial distribution in the Ellsworth Shale ...... 95 3.26 Rock-Eval data plotted on a Van Krevelen Diagram...... 97 3.27 Map of the maturation spatial distribution...... 99 3.28 The maturation spatial distribution in relation to burial depth...... 101 3.29 The Ellsworth fields in relation to the MMGA...... 103 3.30 Gas and oil fields in relation to occurances of hydrothermal dolomite ...... 104 3.31 Map of the spatial distribution of gas chemistry...... 105 3.32 Cross-section of the Ellsworth reservoir ...... 106 A.1 Cross-section network...... 121 A.2 Cross-section A-A’ ...... 122 A.3 Cross section B-B’...... 122 A.4 Cross section C-C’ ...... 123 A.5 Cross section D-D’...... 123 A.6 Cross section E-E’...... 124 A.7 Cross section F-F’...... 124 A.8 Cross section G-G’...... 125 A.9 Cross section H-H’...... 125 A.10 Cross section I-I’...... 126

vii List of Figures-Continued

A.11 Cross section J-J’...... 126

viii LIST OF TABLES

2.1 List of cores in which XRD was run...... 40 2.2 List of well with source rock data ...... 44 2.3 List of wells with maturation data ...... 47 2.4 Table summarizing the gas analysis data ...... 50 3.1 Summary of the average log responses ...... 51 B.1 Cross section will information ...... 127 C.1 Formation top picks ...... 136

ix INTRODUCTION

Purpose of Study

Regional Stratigraphy

Upper Devonian-Lower Mississippian strata have produced hydrocarbons in the

Michigan Basin since the 1920’s with the first well drilled in the Saginaw field located in

Saginaw County, Michigan. Until the 1970’s, the Berea Sandstone was the primary

producer in the Upper Devonian. Focus was shifted to the Antrim Shale in response to

the energy shortages of the 1970’s when the Federal Government funded research into

the feasibility of producing hydrocarbons from oil and other unconventional

energy sources (Matthews, 1989). The Eastern Gas Shales Project (ESGP) was initiated

by the US Department of Energy (DOE) in 1976 to map and document the regional

stratigraphic framework, characterize the shales, and develop new stimulation techniques

and improve on existing ones in the Michigan, Appalachian, and Basins

(Matthews, 1989). In the Michigan Basin, the DOE signed a 4-year contract with Dow

Chemical Company who had undertaken experimental investigation of the Antrim Shale

as an oil shale since the mid-1950’s (Matthews, 1989). As a result, more than 40 reports

were issued, which focused on a variety of characteristics of the Upper Devonian

sequence with an emphasis on the physical and mechanical properties of the Antrim

Shale.

1 Garland D. Ells was one of the many who was involved in the ESGP. His report

divided the Upper Devonian-Lower Mississippian sequence into 9 units (Figure 1.1),

correlated these units throughout the Michigan Basin using 99 gamma-ray logs, and

created six cross sections. Ells (1979) divided the units based on gamma-ray patterns and

previous studies (Cohee and Underwood, 1944, 1951; Lemone, 1964; Asseez, 1967;

Michigan Basin Geological Society, 1969; Lilienthal, 1974, p. 8, p. 12). Ells’ work was a

major contribution to the ESGP and set the framework for future studies. However, some

of the unit boundaries were inconsistent and could be improved upon.

The Antrim Shale is comprised of units 1-6 with unit 1 having three subunits A,

B, and C. The Bedford Shale includes units 7 and 8, and the Berea Sandstone is unit 9.

Figure 1.1 demonstrates one of the cases where the Antrim-Bedford contact (unit 6-unit

7) could be argued. The transition from the Antrim into the above Bedford is typically noted by a significant drop in gamma-ray like the well in Sanilac County (Figure 1.1).

The same drop in gamma-ray occurs in the well in Livingston County but it is picked in unit 5 rather than the unit 6 and 7 boundary. An argument could also be made about the

Bedford-Berea contact (unit 8-unit 9) in Livingston County being in the middle of unit 7.

2 Figure 1.1 Well logs from Sanilac and Livingston Counties showing Ells’ (1979) unit picks based on the different gamma-ray signatures. Transitioning west from Livingston to Sanilac County, there is a notable inconsistency in the Antrim-Bedford boundary (unit 6-unit 7 contact). The contact between the Bedford and Berea could also be argued to be lower. Modified from Ells (1979).

Gutschick and Sandberg (1991) established biofacies in the Michigan

Basin, and helped develop new insights into the bio- and chronostratigraphy for the

Upper Devonian-Lower Mississippian interval. In addition to the establishing the

Devonian-Mississippian boundary, Gutschick and Sandberg (1991) divided the Antrim into four separate members (Norwood, Paxton, Lachine, and Upper Antrim). However, the regional distribution of each member was determined by correlating the gamma-ray

3 units defined by Ells (1979) with the different members (Figure 1.2). The units defined by Ells (1979) can be improved upon by incorporating modern wireline logs.

Figure 1.2 Cross-section from Livingston to Sanilac County, Michigan. The lateral extent of the different Antrim Members was determined using the unit boundaries defined by Ells (1979). The cross-section demonstrates how the regional distribution of the different Antrim Members defined by Gutschick and Sandberg (1991) can be further improved upon. Modified from Ells (1979).

Since the ESGP, new technological advancements in wireline logging have been

more regularly applied to the Michigan Basin. Using only gamma-ray for log facies

differentiation has limitations. Some lithologies can appear equivocal when gamma-ray

is the only data available. Modern logs (bulk density, RHOB; neutron porosity, NPHI;

photoelectric, PEF) allow for a high resolution analysis of the different rock properties.

4

There is a historical lack of high resolution wireline logs to define the different Upper

Devonian-Lower Mississippian units.

Ellsworth Petroleum System

The Ellsworth is sandwiched between two organic rich shales (Norwood/Lachine and Upper Antrim) and produces both oil and gas at shallow depths (900-2000 ft.). The total play production exceeds three Bcf but is not well-documented because most of the activity took place between the late 1930’s through the 1950’s and was used for lease and domestic use. There are 16 fields with Ellsworth production which were commonly referred to as “Berea” in state of Michigan production records. The largest Ellsworth field is the Ravena (31 Ellsworth wells) with an average initial production (IP) of 800

MCF/day. The Cedar Creek Field had the highest cumulative production results averaging 91,000 MCF/well. The typical field production life is around 10 years or less

(Harrison, 2007). Despite having economic potential and historic production, the

Ellsworth Shale has yet to be analyzed in regards to the different components of a petroleum system.

Objective

What is the regional lithostratigraphy of the Upper Devonian-Lower

Mississippian in the Michigan Basin and how does it compare to previous interpretations? This study expands on previous work and incorporates modern wireline logs and X-ray fluorescence (XRF) data for regional correlation of the Upper Devonian-

Lower Mississippian units in the Michigan Basin to help explain the depositional history.

Correlating high resolution XRF data with wireline logs allows for a better understanding 5 of the lithology. XRF data highlights subtle changes in the geochemistry which will

further aid in the distinction between the different shale bodies.

What are the attributes and mechanisms responsible for the shallow gas produced out of the Ellsworth Shale? The general trend/distribution of Ellsworth producing fields in western Michigan suggests controls on petroleum occurrences. Ellsworth fields follow a northwest-southeast trend much different than the irregular distribution of Antrim production in and around Otsego County (Figure 1.10). This suggests a structural and/or stratigraphic element to the distribution of Ellsworth production. A fundamental question regarding the Ellsworth petroleum system, assuming the source of the hydrocarbons is the relatively shallow Antrim Shale, is the geological mechanisms for hydrocarbon generation. Is the probable, Antrim Shale, source rock thermally mature enough to produce oil and gas? An important focus of this study is to incorporate the new understanding of stratigraphy with source rock data to better explain past production and to aid in future exploration projects in the Ellsworth Shale and other Upper Devonian units.

In summary, this work has two objectives. The first objective is to expand on the

mapping of Upper Devonian-Lower Mississippian strata accomplished by Ells (1979) for

the Michigan Basin. Through the use of previous knowledge, modern logs and new, high

resolution chemostratigraphic (XRF) data, Upper Devonian-Lower Mississippian,

prospective hydrocarbon producing shaly formations can be more confidently identified

in the subsurface. The second objective is to give special consideration to the Ellsworth

Shale as a petroleum system, due to the lack of literature on the unit despite its apparent

economic potential. 6 Geologic Setting

Upper Devonian-Lower Mississippian strata in the Michigan Basin and adjacent basins demonstrate the tectonic and sedimentary response of foreland compression and isostatic compensation produced by the Acadian and Antler orogenies (Gutschick and

Sandberg, 1991). The Late Devonian Michigan Basin was a small, roughly circular, persistently subsiding basin within the Eastern Interior seaway. This seaway separated the North American craton from the Old Red continent (Gutschick and Sandberg, 1991)

(Figure 1.3). The Michigan Basin is bounded to the west by the Arch and the

Wisconsin Highlands on the north and northeast by the , on the east by the Algonquin arch, on the southeast by the Findlay arch in northwestern , and the southwest by the in and Illinois. The Michigan Basin communicated with the Appalachian, Illinois, and Moose River Basins through a series of inlets (Figure 1.4).

7 Figure 1.3 Map showing the location of the Late Devonian Michigan Basin (blue highlight) in relation to adjacent basins in the Eastern Interior Seaway as depicted by Gutschick and Sandberg (1991). The position of the paleoequator was taken from Scotese (1986). The black dots show the locations in which Upper Devonian calcareous have been documented. The Catskill deltaic complex (CAT) of the Appalachian Basin demonstrates similar sequences seen in the adjacent Michigan Basin. Modified from Gutschick and Sandberg (1991).

The Middle Devonian-Late Mississippian strata in the Michigan Basin are part of the transgressive Kaskaskia Sequence (Sloss, 1963). Sea-level was generally high and marine transgression was widespread during episodes of anaerobic shale deposition followed by isostatic rebound and eustatic fall resulting in the regressive Bedford-Berea

8 sequence (Gutschick and Sandberg, 1991). During this time the Michigan Basin was located within equatorial belt (Figure 1.3; Scotese, 1984).

Figure 1.4 Map showing the Michigan Basin and surrounding structural features. The red highlighted area shows the Upper Devonian subcrop belt. The dotted line is the Late Devonian shoreline. Modified from Gutschick and Sandberg (1991).

9 Stratigraphy

The Michigan Basin contains more than 17,000 ft. of sedimentary rocks

(Matthews, 1989). The Upper Devonian-Lower Mississippian is greater than 900 ft. thick. The Upper Devonian-Lower Mississippian contains the Antrim, Ellsworth,

Bedford, Berea, and Sunbury formations with the Antrim subdivided into four members

(Norwood, Paxton, Lachine, and Upper Antrim) (Figure 1.5). An isopach map of the

Upper Devonian-Lower Mississippian (base Antrim to top Sunbury) shows a thinning towards the basin center. This has been interpreted to be the result of the Ellsworth

(westerly source) and Berea (north-easterly source) deltas (Asseez, 1969; Matthews,

1989) (Figure 1.6). The Upper Devonian sequence in the Michigan Basin was previously interpreted to be a westward extension of the cyclical facies of the Catskill Deltaic

10 Complex in the Appalachian Basin (Orton, 1879, 1888, 1893; Rich, 1951 a,b; Pepper et al., 1954).

Figure 1.5 Stratigraphic nomenclature for the western and eastern Michigan Basin. Gutschick and Sandberg (1991) divided the Antrim into four separate members (Norwood, Paxton, Lachine, and Upper Member). The black dots represent intervals in which were identified. Modified from Gutschick and Sandberg (1991).

11 Figure 1.6 Isopach map of the Upper Devonian-Lower Mississippian (base of Antrim-top of Sunbury) with 50 ft. contours. Original map from this report that will be discussed in further detail.

Squaw Bay Limestone and Traverse Group

The top of the Squaw Bay Limestone (or Traverse Fm. in older literature) marks the base of the Antrim and the beginning of the Upper Devonian-Lower Mississippian

12 strata discussed in this report. The Traverse Group, below the Squaw Bay, in subsurface includes from base to top: and Traverse Limestone. The subsea depth to the top of the Squaw Bay Limestone is displayed in Figure 1.7.

Figure 1.7 Structure map of the Squaw Bay Limestone and commonly referred to as the “” in older literature (Riggs, 1938). Original map from this report that will be discussed in further detail.

13 Antrim Shale

The Antrim is part of a large Devonian black shale system which extends across most of the eastern cratonic region of North America (Dellapenna, 1991). The lithostratigraphic equivalent units around the region include the Kettle Point Shale of southern , the of the , the in Ohio,

The Chattanooga Shale of the Appalachian Basin, the Woodford Shale of Texas and

Oklahoma, and the Bakken Shale of the Williston Basin (Dellapenna, 1991). The Antrim extends throughout the Michigan Basin with the exception of areas towards the northern and southern margin where the Antrim shale is truncated by erosion. The Antrim has been described as representing a change from normal carbonate marine conditions to a clastically dominated system (Dellapenna, 1991).

The Norwood, Lachine, and Upper Antrim are primarily black shales and the

Paxton Member is a light black-gray shale. The color variation can be attributed, in part, to the amount of oxygen present when the sediment was being deposited. Oxygenated sediment tends to be lighter in color while anoxic sediment is black due to the amount of organics preserved in the sediment. The Antrim has been interpreted as being deposited in an oxygen-stratified basin with a fluctuating pycnocline producing thin laminations in the shale (Dellapenna, 1991).

The Norwood and Lachine Members are primarily composed of quartzose, organic-rich, thinly-laminated black shale beds containing abundant nodules

(Dellapenna, 1991). The quartz can make up to 50% of the matrix and is in the form of silt sized grains (both detrital and polycrystalline) and authigenic cement (Dellapenna,

14 1991). Other minerals include illite, chlorite, kaolinite, muscovite, pyrite, and sparse dolomitic concretions (Dellapenna, 1991). These units have noteworthy high gamma-ray signature ranging between 126 and 441 API .

The Paxton Member divides the Norwood from the Lachine Members (Figure

1.8) and is primarily composed of carbonate-rich, organic-lean, moderately to intensely bioturbated gray shale beds with abundant carbonate concretions. The mineralogy of the

Paxton is primarily calcite, dolomite, quartz, illite, chlorite, kaolinite, muscovite, pyrite, and sparse concretions of pyrite (Dellapenna, 1991). The Paxton Member marks a transition in the basin’s history when anaerobic shale deposition was interrupted by dysaerobic, lighter colored muds and lime muds (Gutschick and Sandberg, 1991).

Literature suggests the transition from anaerobic to dysaerobic sedimentation was initially the result of isostatic rebound producing sediment associated with a prodelta environment (Gutschick and Sandberg, 1991). This was suggested because the conodont

(Lower gigas Zone) in the lower Paxton Member is related to a transgression. This implies the rate of isostatic rebound was greater than the rate of global sea-level rise

(Gutschick and Sandberg, 1991). The conodont linguiformis was identified towards the

top of the Paxton and is related to a drop in global sea-level and mass extinction event

(Gutschick and Sandberg, 1991). The evidence for the Paxton Member being deposited

in a prodelta environment is not well documented. The gamma-ray signature decreases

when compared to the above and below Lachine and Norwood Members and ranges

between 72 and 300 API.

The Upper Antrim (or Upper Member) interfingers with the Ellsworth Shale and

is directly below the Bedford Shale in the central and eastern basin (Gutschick and 15

Sandberg, 1991). The Upper Antrim is another organic-rich black shale similar to the

Norwood and Lachine Members, however, no mineralogical data has been collected in the Upper Antrim. The gamma-ray ranges between 100 and 328 API.

16 Figure 1.8 A type log for the Upper Devonian-Lower Mississippian from Midland County. Prior to 1991, the Antrim Shale was not subdivided into members. This type log shows that on the basis of patterns in the gamma-ray log, that the Antrim Shale can be readily subdivided into the Norwood, Paxton, Lachine and Upper Antrim Members.

17 Ellsworth Shale

The Ellsworth Shale is comprised of greenish-gray shale with interbedded dolomite and limestone. The mineralogical composition of the Ellsworth is variable but generally contains as much as 25% dolomite and less quartz than the Antrim (Ruotsala,

1980). The Ellsworth also contains illite, chlorite, and small concentrations of pyrite

(Dellapenna, 1991). The average gamma-ray for the Ellsworth is around 168 API.

The origin of the Ellsworth is still poorly understood, but published literature suggests it was deposited as a prodelta shale sourced from a river system to the west

(Asseez, 1969; Gutschick and Sandberg, 1991). The Ellsworth is believed to locally interfinger with the Bedford and Antrim Shales in and around Clare and Calhoun

Counties (Tarbell, 1941; Asseez, 1969; Ells, 1979). The presence of limestone and dolomite beds towards the top of the Ellsworth indicates a decrease in detrital sedimentation (Asseez, 1969).

Bedford Shale

The Bedford Shale consists predominantly of silt- and deposited prior to the Berea Sandstone on the eastern portion of the Michigan Basin and directly overlies the Upper Antrim (Gutschick and Sandberg, 1991). The Bedford Shale is described as being composed of highly fissile interlaminated shale, silty shale, micaceouse shale, and minor thin-bedded in variable proportions (Balthazor, 1991). The gamma-ray signature varies between 50 and 200 API.

On regional isopach maps, the Bedford Shale exhibits a westward thinning of the sequence suggesting an eastern source for the unit (Gutschick and Sandberg, 1991). 18 Newcombe (1932) proposed that the Bedford-Berea sequence was deposited in a deltaic

setting. Uplift from the Late Devonian and/or a regional regression

apparently provided the clastic sediment supply for the delta system (Pepper et al., 1954;

Sloss, 1963; Pashin and Ettensohn, 1995).

Berea Sandstone

The Berea Sandstone overlies the Bedford Shale and is composed of fine- to medium-grained sandstone that is generally well-cemented with silica and dolomite, micaceous, pyritic, and very silty and shaly in places (Cohee and Underwood, 1944;

Balthazor, 1991). In areas like the Williams field in Bay and Midland Counties, the

Berea Sandstone is divided into five lithofacies with various mineralogies. The clean sandstone facies is composed of primarily quartz with small amount of feldspar

(Balthazor, 1991). The high percentage of quartz results in an overall low average gamma-ray response (87 API) compared to the other Upper Devonian-Lower

Mississippian intervals.

The Bedford-Berea sequence is a terrestrial clastic wedge sourced from the eastern Algonquin Arch and thins toward the center of the basin demonstrating a deltaic geometry (Asseez, 1969). The Berea Sandstone was deposited in a fluvial-dominated, shallow water delta (Balthazor, 1991). The different Berea lithofacies described by

Balthazor (1991) in the Williams Field are associated with different sub-environments in a delta plain setting.

19 Sunbury Shale

The Sunbury Shale marks the end of the Devonian and the beginning of the

Mississippian (Gutschick and Sandberg, 1991). The Sunbury Shale overlies the

Ellsworth and Bedford-Berea throughout most of the Michigan Basin. This bituminous black shale is in contact with the underlying Antrim Shale where the Bedford-Berea sequence and the Ellsworth Shale are not present in the north and south central parts of the Michigan Basin. No mineralogical data from the Sunbury Shale has been collected in the Michigan Basin. The gamma-ray signature is similar to the Antrim black shales averaging 203 API.

The Sunbury Shale represents regional transgression above coarser-grained clastics and the re-establishment of an open marine, clastics starved environment

(Asseez, 1969). The black color suggests the basin was anoxic when the sediment was being deposited.

Previous Research

The understanding of the Upper Devonian-Lower Mississippian evolved through time, primarily initiated by different economic events which led to an increased demand for natural resources. This section highlights some of the events and authors that contributed to a better understanding of the Upper Devonian-Lower Mississippian stratigraphy.

20 Early Nomenclature

The Upper Devonian-Lower Mississippian was first recognized with the formal description of what is now called the Bedford-Berea sequence by the First Geological

Survey of Ohio (Briggs, 1838; Whittlesey, 1838). Newberry (1870) separated the sequence and named the Bedford Shale and Berea Sandstone from exposures in northern Ohio. The subsurface studies during the growth of the petroleum industry in the late 19th century lead to an increased understanding of the Bedford-Berea strata in Ohio (Orton, 1879, 1888,

1893). Orton showed that the Berea Sandstone extends into , West ,

Kentucky, and Michigan.

The Antrim Shale was first name by A.C. Lane in 1901 based on exposures in

Antrim County, Michigan which was previously known as the St. Cleric Shale. Newcombe

(1932) first described the Ellsworth Shale from cuttings from a well drilled in the

Muskegon field in the Michigan Basin. Newcombe described the Ellsworth Shale as interbedded blue shale and limy sandstone.

United States Geological Survey (USGS) Mapping Project

The U.S. Geological Survey started an 11-year mapping project to help increase hydrocarbon production during World War II (Hale, 1941; Cohee and Underwood, 1944;

Pepper et al., 1946; Demarest, 1946; de Witt, 1946, 1951; Rittenhouse, 1946; Baltrusaitis et al., 1948).

Hale (1941) documented the sedimentation patterns of Lower Mississippian units in the western portion of the Michigan Basin, which at the time of publication were believed to include the Coldwater, Ellsworth, and upper Antrim formations. The 21 observations were based on well logs and cuttings. Baltrusaitis et al. (1948) was the first

to suggest a facies relationship between the Antrim and Ellsworth stating, “The greenish

gray Ellsworth Shale of the western half of Michigan overlying the black shale of Antrim

is contemporaneous with and interfingering with the upper Antrim of central and eastern

Michigan.”

The project primarily focused on the Bedford-Berea sequence The most noteworthy early was done by Rich (1951 a, b) and Pepper et al. (1954), which has since been used as a paradigm for deltaic sedimentation models (van Siclen, 1958; Krumbein and Sloss, 1963; Asquith, 1970; Wanless et al., 1970; Heckel, 1972; Woodrow and Isley,

1983; Frazier and Schwimmer, 1987; Stanley, 1989). Rich’s original model of the

Bedford-Berea recognized that the depositional geometry of the Bedford-Berea sequence could be described as clinoforms. Pepper et al. (1954) compiled extensive research and developed a detailed regional model of the Bedford-Berea sequence. Multiple scenarios for the timing of deposition between the Michigan and Appalachian Basins were suggested by Pepper et al. (1954). The most favored theory suggests that the Berea in Michigan was slightly younger than the Berea of Ohio but derived from the same source (Ontario River).

The “Red Bedford Delta” found in Ohio is absent in the Michigan Basin implying that the source supplying red sediment had been eroded and sediment transported into the

Appalachian Basin prior to the deposition of the Bedford in Michigan. A modern analogue was drawn with the Hwang-Ho River in China where in 1938 the rivers main channel switched diverting the mouth 300 miles south of the original outlet.

Asseez (1969) developed a better understanding of the paleogeography and distribution of the thick shale sequences of the Upper Devonian-Lower Mississippian units 22 (Antrim, Ellsworth, Bedford, Berea, and Sunbury) in the Michigan Basin. Asseez (1969)

used drill cuttings from 443 wells and differentiated them based on texture and color. The

color variation in the shales suggests a changing environment of deposition. For example,

the Ellsworth Shale is predominantly a green shale that thins from the west towards the

center of the basin. The green color is from an abundance of green clay minerals

(glauconite) and poor preservation of the organic matter (OM) (Asseez, 1969). The

Ellsworth Shale also exhibits changes in lithology, with the upper Ellsworth composed of

interbedded sand and grainy carbonate strata. Asseez (1969) used the changes in lithology

and thickness to suggest that the Ellsworth Shale represented a second deltaic system

coming into the basin from the west. He interpreted the oolitic and grainy carbonate strata

to be more proximal delta deposits, while the green shales represent prodelta deposits. The

delta front deposits have likely been removed through erosion.

Eastern Shale Gas Project (ESGP)

In the late 1970’s stratigraphic characterization of the Devonian-Mississippian shale sequence in the Michigan Basin accelerated due to increased interest in sediment- hosted energy resources. The funding was sourced from the US Department of Energy.

As a result, more than 40 reports were written outlining the applicability of producing hydrocarbons out of unconventional resources in the Appalachian, Michigan, and Illinois

Basins (Matthew, 1989). The data gathered in the Michigan Basin was taken from Antrim cores in Sanilac and Otsego Counties, cuttings, and 7 outcrop locations. A majority of the data focused on evaluating geochemical characteristics (Matthews and Humphrey, 1977;

Leddy and others, 1980; Ruotsala, 1980; Young, 1980), lithological properties (Hathon and others, 1980; Ruotsala, 1980; Young, 1980), and source rock potential (Matthews and 23 Humphrey, 1977; Matthews, 1980; Matthews and others, 1980; Cercone, 1984) of the

Antrim Shale.

Ells (1979) examined the Upper Devonian through Lower Mississippian regional stratigraphy of the Michigan Basin by constructing six cross sections extending from the

Antrim to Mississippian shale units. Formation tops were picked by Ells using index gamma-ray signatures to discern the 6 units in the Antrim and Ellsworth Shales. In Ells type log, the Bedford Shale is represented by units 7 and 8 and the Berea Sandstone by unit

9 (Figure 1.1). In his cross-sections, Ells (1979) noted an east to west facies relationship between the Ellsworth and both the Bedford and part of the Antrim Shales. Moving from east to west in the basin the Berea pinches out and the Sunbury thins and eventually merges into the uppermost Ellsworth Shale. Ells interpreted the Bedford and Berea as

Mississippian in age and the Antrim as Devonian in age, implying that the Ellsworth straddles the Upper-Devonian-Lower Mississippian.

The work done by Matthews (1989) summarizes the 40 reports from the ESGP.

Matthews further emphasizes the importance of this study by stating “The true significance of thinning and the appropriate choice of boundaries for units that are to be considered for thickness study remain debatable points for the Devonian-Mississippian shale sequence in

Michigan.” Mathews further suggested that in order to resolve correlation issues more precise mapping must be done to better define the transition zone between the Ellsworth and the Antrim Shales.

Gutschick and Sandberg (1991) generated biostratigraphic data for the Upper

Devonian-Lower Mississippian units in the Michigan Basin and established conodont

24 biofacies, which proved useful for placing the Devonian-Mississippian boundary below the

Sunbury Shale (Figure 1.9). The boundary was established through the recognition of the spore Retispora lepidophyta in both the Berea and Bedford formations. The spore is a global index in latest Devonian strata (Gutschick and Sandberg, 1991). Coincident with this work Gutschick and Sandberg, (1991b) combined their biostratigraphic data and the stratigraphic work done by Ells (1979) to describe the evolution and history of the Late

Devonian Michigan Basin. The paper establishes a chronostratigraphic relationship between the Appalachian, Michigan, and Illinois Basins.

Figure 1.9 Regional correlation chart for Upper Devonian and Lower Mississippian strata in the Michigan (highlighted in blue) and Appalachian Basins. Age assignments of these units are tied to conodont zones. Gutschick and Sandberg (1991) correlated the Antrim members based on the units developed by Ells (1979). Modified from Gutschick and Sandberg (1991).

25 Production History

Antrim Shale

The Antrim Shale has produced gas in wells in Antrim, Crawford, Kalkaska,

Otsego, Manistee, Grand Traverse, and Montmorency counties (Figure 1.10). There are over 3,600 gas wells in Otsego County alone. The gas produced is in access of 2.5 trillion cubic ft. (Tcf). The wells produced primarily methane-rich biogenic gas (Martini,

2002; Wen, 2015) at shallow depths no deeper than about 2,500 ft. The Antrim is considered to be an unconventional reservoir because it is self-sourced and in order to economically produce, typically requires additional fracture and stimulation techniques.

Antrim gas in the northern trend (Figure 1.10) is well documented to be primarily biogenic in origin associated with fresh water influx during the Wisconsin glaciation (10 to 79 ka) (Martini, 1998; Wen, 2015). Wen (2015), through the use of noble gasses, did find traces of thermogenic methane. Whether gas production was thermogenic versus biogenic, was difficult to establish due to brine migration and in-situ production from the

Norwood and Lachine. Wen concluded that the thermogenic methane most likely migrated up-dip from deeper Antrim Shale or a different source rock below the Antrim, possibly age rocks of the A-1 Carbonate. Recent published research has given additional insight into source rock potential within the Michigan Basin (Wagenvelt, 2015;

Wen, 2015).

26 Figure 1.10 Map showing the distribution of oil and natural gas fields in Upper Devonian producing formations. Red polygons are gas producing fields, green polygons are oil, and yellow are both gas and oil.

Ellsworth Shale

More than 120 wells have produced hydrocarbons in the Ellsworth. Producing fields trend northwest-southeast (Figure 1.10). The Ellsworth has produced more than 3 billion cubic feet (Bcf) of gas and 0.11 million barrels of oil (MMbo). The source rock for these accumulations is uncertain but previous studies suggest the hydrocarbons are self-sourced or are from the Antrim Shale (Charpentier, 1994). Thin silt stringers within the Ellsworth serve as the reservoir. The silt was originally termed “Berea” by wellsite geologists, on State of Michigan drillers’ reports, implying the sediment source was the

27 same as the eastern Bedford-Berea delta. This would imply that the Bedford-Berea

sequence is contemporaneous with the Ellsworth Shale. The more proximal coarser

clastic sediment of the Ellsworth wedge that would represent the delta plain is believed to

have been removed by subsequent erosion (Asseez, 1969).

Berea Sandstone

The Berea Sandstone has produced both gas and oil in just over 300 wells in

Michigan. Gas production typically occurs further east at shallower depths compared to oil production (Figure 1.10). The source of hydrocarbons in the Berea has not been well documented but is believed to be organic rich shale of the Antrim and Sunbury Shales. A majority of the 29 different Berea fields are related to anticlinal structures and often also production from the . The best documented producing field in the

Berea is the Larkin-Williams Field located in Bay and Midland Counties (Balthazor,

1991). This pool has produced over 3 MMbo to date.

Field studies were presented in the Michigan Basin by Addison (1940), Balthazor

(1991), and Duszynski (1991) to better understanding the reservoir architecture and

trapping mechanisms of Berea Sandstone reservoirs. Addison (1940) analyzed the South

Buckeye field in Gladwin County. This field is a structural trap mainly producing out of

the Middle Devonian Dundee formation. Addison (1940) characterizes the Dundee

reservoir but does mention that the Berea has little to no water within the formation. There

has been additional production in the Berea Sandstone in the South Buckeye field since

this publication. Balthazor (1991) did a detailed assessment of the Williams and Larkin

fields in Bay and Midland Counties and Duszynski (1991) focused on the New Lothrop

28 field in Shiawassee and Genesee counties, MI. Both authors developed similar conclusions suggesting the trapping mechanism was both structural and stratigraphic. Plunging anticline structures trap hydrocarbons at a seal formed by a truncation overlain by mud- filled channel deposits. Balthazor’s facies mapping work in the Larkin-Williams field has since contributed to more effective enhanced recovery projects in the Williams and Larkin fields (Tim Maness, pers. com.?). Dolton (1996) suggests the hydrocarbons are most likely sourced from the underlying Antrim Shale and some contribution is possible from the overlying Sunbury Shale.

29 METHODOLOGY AND DATA

Wireline log analysis and XRF is used for the purpose of defining the regional lithostratigraphy. Understanding the distribution and relationships of the different lithologic units will also set up the framework for understanding the Ellsworth petroleum system. In addition, any source rock and gas chemistry data collected in the Upper

Devonian-Lower Mississippian will be summarized in this report to help explain the different attributes and mechanisms responsible for hydrocarbon generation.

Wireline Log Analysis Method

The purpose of well log analysis is to better understand the regional stratigraphy and basin history during the Upper Devonian-Lower Mississippian. Using modern logs gives more complete information on the different units by making measurements on the rock which reflect the mineralogy and pore space properties. The cross sections and maps in this document have been created using a GIS based software Petra v3.4.3. A majority of the figures were edited in Adobe Illustrator CS6. The isopach and structure maps were constructed from 193 digital logs. The logs were selected based on spatial distribution and availability of modern wireline log tracks (Figure 2.1).

30 Figure 2.1 Location of the 193 wells used in this study to develop the different maps and cross sections. Colors indicate the different logs that were taken in each well.

Gamma Ray

Gamma-ray (GR) logs measure the natural radioactivity of a rock emitted primarily from radioactive isotopes of potassium, uranium, and thorium (Shlumberger,

1972, p. 57). Organic particles in black shales or other organic-rich rocks absorb 31 different radioactive elements which is why shale tends to demonstrate a higher gamma-

ray signature.

Clean composed of quartz and clean limestone composed mostly of calcite are generally poor emitters of gamma-rays (Kelley, 1969, p. 10). However, clean sandstones with low shale content might produce a high gamma-ray response if the sandstone contains potassium feldspars, micas, glauconite, or uranium-rich waters

(Asquith and Krygowski, 2004). Gamma-ray is usually plotted on the left track in

American Petroleum Institute (API) units which vary in scale based on the lithology of interest.

Bulk Density

Bulk density (RHOB) is the density of the entire formation (solid and fluid parts) as measured by the logging tool and recorded in grams per cubic centimeter (g/cm).

The logging tool emits gamma-rays into the formation which interact with electrons in the rock media. Some gamma-rays are reflected back into the logging device and counted by a scintillometer. The count is proportional to the rock density. The density logging tool has a relatively shallow depth of investigation, and as a result, the tool is held against the side of the borehole during logging to maximize its response to the formation (Asquith and Krygowski, 2004).

The scale in the cross-sections for RHOB ranges from 2 to 3 g/cm and is displayed on the second track. Sandstones have the lowest bulk density around 2.644 g/cm. Carbonates have a higher density with limestone around 2.710 g/cm and

32 dolomite close to 2.877 g/cm (Halliburton, 1991). Shale varies in bulk density due to the heterogeneity in composition.

Neutron Porosity

Neutron porosity logs (NPHI) measure the hydrogen concentration in a formation.

In clay-free formations where the porosity is filled with water or oil, the neutron log measures liquid-filled porosity. The neutron logging tool emits neutrons which collide with the nuclei of the formation. The maximum energy loss occurs when the neutron collides with a hydrogen atom because it is almost equal to the mass of the neutron.

When enough collisions occur a gamma-ray is emitted (Asquith and Krygowski, 2004).

The first neutron logs detected the gamma-rays that were products of this collision but more modern detectors measure the neutrons that were not absorbed.

Neutron porosity measures the pore space within a rock and is not always representative of the primary rock fabric. Shale has higher neutron porosity due to the bound water content (Doveton, 1994).

Photoelectric

Photoelectric-effect (PEF) is commonly referred to as Litho or Spectral tool and was introduced around 1978 (Asquith and Krygowski, 2004). PEF is a density logging tool and records the absorption of low-energy gamma-rays in units of barns per electron

(Doveton, 1994).

33 The PEF log is commonly scaled on a range between 1 and 10 and is displayed on

the second track with NPHI and RHOB. PEF is an indicator of mineralogy. The basic

mineral reference values are: quartz 1.81; dolomite 3.14; calcite 5.08 barns/electron.

Cross Sections and Maps Data

The cross sections and maps were developed through analysis of 193 digital logs

(Figure 2.1). These logs were selected based on spatial distribution, well log quality and availability. The isopach and structure maps incorporate all 193 wells and the cross sections were developed using 102 wells. There are ten total cross sections with five trending EW (AA’-EE’) and five trending NS (FF’-JJ’) (Figure 2.2).

34 Figure 2.2 Cross section network developed through the use of 102 wells selected based on spatial distribution and well log availability. Cross-sections are included in the appendix.

35 X-Ray Fluorescence Method

Energy-dispersive x-ray fluorescence, commonly known as XRF, can quickly and nondestructively determine the elemental composition of rocks and any other item needing elemental analysis in both weight percent and parts per million (ppm).

Geochemical data was collected in all eight Upper Devonian-Lower Mississippian units to better understand the variations in chemostratigraphy. XRF has many uses but the main purpose for using XRF in this project is to correlate the variation in chemical composition with wireline log response and mineralogy to project the vertical and lateral distribution of the different lithologies with greater certainty.

The Thermo Fisher Scientific Niton XL3t 950 and the Bruker TITAN S1 800-8 hand held XRF analyzers were used to gather chemical data on 6 different cores (Figure

2.3).

36 Figure 2.3 Geographic locations of the six cores with XRD data in the Michigan Basin.

The XRF device is placed directly on a sample for a selected amount of time. For the TITAN, each sample was tested for 50 seconds at one foot intervals. The first 10 seconds measure the heavier elements which are easier to detect and the last 40 seconds measure the lighter elements. Samples tested with the Niton were analyzed for 90 seconds at every half foot. The time was selected based on observation of when the values stabilized. Both instruments test approximately 45 elements ranging from magnesium to uranium. Both instruments have an eight millimeter beam spot.

37 The physics behind XRF is similar to the process in which humans see color. The eye emits photons which interact with molecules in a sample and are reflected back, scattered, or absorbed. The photons reflected back to the eye exhibit different energies which the brain interprets as different colors. The fundamental difference is the eye measures molecular information and XRF measures atomic information. This is because the photons emitted from XRF are much higher energy (1,000-40,000 eV) than that of the human eye (1-3 eV) allowing the phonons to reach the internal electron orbitals (L and

K) of an atom (Figure 2.4). In some cases, the photon “bumps” the electron out of the inner shell (K) of an atom making the atom unstable. To stabilize the atom an electron from a higher shell (L) fills the vacancy. This process requires energy which is released in the form of another photon. The energy in the released photon is what is recorded in the detector of the XRF device. The energy required to move from the L orbital to the K orbital depends on the number of protons in the nucleus and the variation in protons is what differentiates elements. The detector in the XRF device analyzes the returning photons for a selected period of time resulting in the sum of the different elements for that particular sample (Drake, 2016).

38 Figure 2.4 A simplified diagram showing how the XRF device gathers the elemental data. The x-ray source emits photons (orange arrow) which “knock out” electrons (red, purple and green spheres) from an orbital (K, L, or M). An electron from the outer orbital fills the void to stabilize the atom. The energy required to do this is released in the form of another photon (red arrow) and recorded by the detector. This image was modified from Thermo Fisher Scientific, 2007.

X-Ray Fluorescence Data

The Bruker S1 TITAN 800-8 hand held XRF analyzer was used to gather data on the DOW #100, #103, and #104 wells in Sanilac County and the Amos C1-4 well in

Grand Traverse County (Figure 2.3). The CONN D2-8 well in Shiawassee County was tested with both the TITAN and the Thermo Scientific Niton XL3t 950 XRF analyzer for comparison. The Krocker #1-17 in Clare County was analyzed by Dr. Jay Zambito from the Wisconsin Geological Survey using the Thermo Scientific analyzer. All of the cores

39 are curated at the Michigan Geological Repository for Research and Education

(MGRRE) in Kalamazoo, Michigan.

Well Permit Sample Device Sample Name # County Type Used Operator Interval Amos Grand whole Bryan C1-4 59019 Traverse core TITAN Currie 1 ft. CONN whole TITAN, Bryan 1 ft. (TITAN) D2-8 48772 Shiawassee core Niton Currie 1/2 ft. (Niton) DOW 624- whole Bryan 100 771-474 Sanilac core TITAN Currie 1 ft. DOW 687- whole Bryan 103 771-474 Sanilac core TITAN Currie 1 ft. DOW 686- whole Bryan 104 771-474 Sanilac core TITAN Currie 1 ft. Krocker powdered Jay 1-17 60134 Clare core Niton Zambito random Table 2.1 List of cores in which XRD was run. Samples were collected in all formations in the Upper Devonian-Lower Mississippian.

Limitations

The instrument must first be calibrated in order to obtain valid quantitative elemental results. The proper calibration is done using a standard for the formation being analyzed (Rowe, 2012). The purpose of this study is not to determine the exact quantity of a particular element in a sample, but rather to establish noticeable trends when transitioning from one formation to the next. Although XRF values are displayed in the results, it is important to note that the numbers should be interpreted with caution. As previously mentioned, the main purpose of running XRF is to observe noticeable elemental trends and how they relate to the mineralogy and wireline logs.

Not all of the cores where analyzed using the same instrument. The samples in the CONN D2-8 were measured with both devices to ensure a consistent correlation.

40 When comparing the results both showed similar trends. However, the TITAN gave

higher values for calcium (Ca), magnesium (Mg), aluminum (Al), potassium (K),

phosphorus (P), and silicon (Si) because it reports the results as oxide compound

percentage rather than elemental percentage for those particular elements. Cross plotting

the samples for the above elements showed a good correlation which established a

relationship for correction to get elemental percentage.

The Krocker #1-17 was sampled using powdered core rather than placing the instrument directly on the whole core. The data from the Krocker displayed more variation and less of a dominant element than the other five wells. When measuring whole core, a very small surface area is being analyzed but when powdering the sample, a larger area is being measured resulting in a greater variation in elemental data.

Ellsworth Petroleum System Method

The primary components of a petroleum system include the source, reservoir, and

seal. The regional mapping of the lithostratigraphy is important to understand a

petroleum system because it demonstrates the spatial distribution and relationship each

component has in relation to the other.

The Michigan Basin contains substantial, thermally immature (with respect to potential

liquid and gaseous hydrocarbon generation) Paleozoic rock strata when compared to

other basins around the region. Evaluation of the thermal alteration of potential source

rocks is critical in the evaluation of shallow petroleum systems. Knowing the kerogen

type, total organic carbon (TOC), and maturity of the sediment is important because all

three components determine if hydrocarbon production occurred and at what temperature. 41 Different kerogen types have different temperatures at which the onset of oil and gas

takes place (McCarthy, 2011). The type of kerogen can also suggest the environment in

which the sediment was deposited. TOC is analyzed to determine if there is enough OM

to produce hydrocarbons assuming the maturation is sufficient. A rock that has TOC

greater than 2 (wt. %) is considered to be a good-excellent source rock (McCarthy, 2011).

The maturity of source rock depends upon the kerogen type and the amount of time, heat, and pressure the rock has been exposed to. Maturation can be measured using multiple techniques. Reflectance values are measured and calculated to a normalized reflectance value (Ro) for rocks with little or no vitrinite, bitumen, or other OM. Vitrinite reflectance is the most common technique but it is time-consuming and subject to the interpretation of the analysts (Wust et al., 2013). Rock Eval or pyrolysis records the temperature (⁰C) when the maximum amount of HC from cracking of kerogen (peak of

S2) occurs during testing (McCarthy et al., 2011). Spore coloration index (SCI) is a

technique that uses the color of palynomorphs (pollen and spores) to determine the

maturation of the sample (Moyer, 1985). There is no direct standardization between the

different measurements because Ro and SCI are both subject to the interpretation of the analysts.

The gas chemistry can suggest where the gas formed and what mechanism was responsible for gas generation (bacterial methanogenesis or thermal cracking of kerogen).

Where the gas was formed and how it was generated is indicated by the gas chemistry.

The northern Antrim fields are suggested to be primarily biogenic gas (Martini, 2003) which could also explain the Ellsworth production. Biogenic gas is typically associated

42 with higher concentrations of methane compared to other uniquely thermogenic natural

gas components.

Source Rock Data

Kerogen Type

Core laboratories and Weatherford conducted a source rock analysis on 20 different wells (Table 3.3) in Michigan at multiple intervals in the Upper Devonian-

Lower Mississippian. The samples were tested for Tmax (temperature (°C) when maximum release of hydrocarbons (S2) from cracking of kerogen during pyrolysis), TOC,

S1 (free HC which are volatilized below 300⁰C), S2 (HC compounds produced by cracking of kerogen as temperature increases to 600⁰C), S3 (quantity of CO2 released during pyrolysis), hydrogen index (hydrogen carbon ratio ((100*S2)/TOC)), and oxygen index (oxygen carbon ratio ((100*S3)/TOC)).

43 API # Source 21035128680000 Weatherford 21035346110000 Weatherford 21035440040000 Weatherford 21039429480000 Weatherford 21069406340000 Weatherford 21073136730000 Weatherford 21107397270000 Weatherford 21113418300000 Weatherford 21117479880000 Weatherford 21129409540000 Weatherford 21129423960000 Weatherford 21133177750000 Weatherford 21133355510000 Weatherford 21137415590000 Core Labs 21165350990000 Weatherford 21009476730000 Weatherford 21029348240000 Weatherford 21113346060000 Weatherford 21143345370000 Weatherford 21165346120000 Weatherford

Table 2.2 List of well API numbers and the source of data for wells in which source rock analysis was conducted.

Maturation

Moyer (1982) determined the maturation of different shale formations in Michigan using the SCI technique. The data includes 16 wells with samples taken from the Antrim

Shale and one core measurement in Newaygo County sampled in the Ellsworth Shale.

Rullkötter et al. (1992) collected Antrim samples in Alpena and Missaukee counties for source rock evaluation but did not list well or outcrop locations. Wagenvelt (2015) combined new (Wagenvelt Pyrolysis Data, 2014) and previously published maturation data

(Cercone and Pollack, 1991; Dellapenna, 1991; Everham, 2004; Amoco, Weatherford) in

44 multiple formations throughout the basin. Within the Wagenvelt Pyrolysis dataset, 33

wells have samples from the Upper Devonian-Lower Mississippian shales.

There are a total of 52 wells with thermal maturation values in this combined dataset (Figure 2.5). The thermal maturation data was recorded using vitrinite reflectance

(Ro), Rock Eval or pyrolysis (Tmax), and SCI. Table 3.4 shows the 52 total wells with different wells and the maturation values. The values were averaged if multiple samples were taken in the Upper Devonian-Lower Mississippian.

45

Figure 2.5 Locations and permit numbers for all wells in Michigan with thermal maturation data in the Upper Devonian – Lower Mississippian shale formations.

46

Well Name Permit # Maturation Source SATTELBERG 1 23890 0.60 Geochem data for Thermal Benson 1-14 34612 438.57 Weatherford TOC data Butcher Unit 2 30615 0.38 Cercone & Pollack Ro data CHISTIANS EN 4-32 35248 428.33 Wagenvelt_Pyrolysis Conn D2-8 48772 437.64 Data_2014TOC_RE_Amoco Granger 1-22 44004 439.20 Weatherford TOC data Hart 1-21 39727 440.83 Weatherford TOC data Jackson D2-1 58397 424.53 TOC Antrim and Traverse Jelinek & Ferris Unit 1 27907 426.56 Cercone & Pollack Ro data Kaverman 1-23 35551 0.79 Weatherford TOC data Keller 1 17775 442.25 Weatherford TOC data Kizer 1 25868 438.07 Everham Ro data Kovacs L 1 13673 0.56 Weatherford TOC data Krabill Estate 1 12868 440.36 Weatherford TOC data LAKE HORICON CORP 1 25873 434.31 Middleton, Ruth 1 47988 0.56 Weatherford TOC data North Michigan Land & Oil 1-27 34824 442.30 Weatherford TOC data Reno 1-28 40634 437.00 Weatherford TOC data Riverside 1-15 41830 441.25 Weatherford TOC data State Chester 18 33875 0.52 Cercone & Pollack Ro data STATE FOSTER 1 25099 0.59 Everham Ro data State Foster 1-12 42396 436.00 Weatherford TOC data State Liberty 1-18 35099 439.50 Weatherford TOC data State South Branch 1-29 42948 437.29 Weatherford TOC data State Union Unit "M" 1-21 31364 0.52 Cercone & Pollack Ro data TAGGART UNIT 32 17789 0.58 Everham Ro data THALMANN 1 26112 0.52 Everham Ro data Visser et al 3- 35 34606 436.00 Weatherford TOC data Warren 1-20 40954 436.07 Weatherford TOC data Weingartz 1-7 34611 438.60 Weatherford TOC data Latuszek B 1-32 41559 434.80 Rock-Eval data_Dellapenna State Custer 1-21 47673 431.03 Rock-Eval data_Dellapenna 4-40 Club 1-35 33405 433.00 Rock-Eval data_Dellapenna NIX, L S 1 9582 6.0 Moyer TRAUTNER, EDMAN 1 11597 5.1 Moyer METHNER, OTTO 1-B 10639 5.6 Moyer MIO UNIT AREA 1 11995 5.8 Moyer STATE SECORD A-1 13632 5.6 Moyer GLADSTONE, LLOYD 1 8482 6.4 Moyer MCCLEAR, TRESSA 1 16690 4.8 Moyer HARDEN, M. D. ESTATE 1 8270 4.3 Moyer MCINTYRE, GEORGE 1 10936 5.4 Moyer SATTERLEE, CRIS 1 3298 6.8 Moyer LARSON, LOUISE 1 10661 4.5 Moyer CITIES SERVICE OIL 1 16734 6.3 Moyer RICE, PEARL O 1 9897 7.0 Moyer RODDENBERRY 1 10792 4.1 Moyer HEWETT, RICHARD 1-20 30974 4.9 Moyer SPARKS R&J 1-8 29739 5.3 Moyer VANDERLEY 1-5 34234 4.5 Moyer Alpena 427 Rullkotter Missaukee 432 Rullkotter

Table 2.3 List of well names, permit numbers, maturation (Tmax, Ro, and SCI), and the source of data. 47 TOC

TOC data was collected in the Upper Devonian-Lower Mississippian in 44 different wells (Figure 2.6) from multiple sources (Moyer, 1982; Rullkötter et al., 1992;

Unpublished MGRRE TOC database). TOC values were averaged by formation for the wells that included sample depths. Average TOC values are in the following order from least to greatest: Ellsworth Shale (1.23 wt. %), Sunbury Shale (1.51 wt. %), Berea

Sandstone (1.55 wt. %), Upper Antrim Member (2.77 wt. %), Paxton Member (3.57 wt.

%), Lachine (4.45 wt. %), and Norwood (6.08 wt. %). The Bedford Shale has no TOC data. The TOC data supplied from Moyer (1982) was sampled in unknown intervals in the

Antrim and was notably lower than the other Antrim data. Excluding the TOC data from

Moyer (1984), the average TOC for the Antrim undifferentiated is 4.58 wt. %. The lower values may be the result of different sampling preparation and procedures. The TOC data in relation to each formation will be further discussed in chapter 4.

48 Figure 2.6 Well locations with permit number (PN) for wells with TOC data in the Upper Devonian – Lower Mississippian.

Gas Analysis

Upper Devonian-Lower Mississippian gas analysis data was collected from the

Michigan Public Service Commission (MPSC) and a report done by Martini et al. (2003).

A majority of the gas samples were taken from Antrim producing wells in northern

49 Michigan. The MPSC sampled gas out of five wells near the eastern margin producing gas

out of the Berea Sandstone which has been suggested to be sourced from the Antrim Shale

(Dolton, 1996). The gas was tested for the amount of methane (C1), ethane (C2), propane

(C3), iso-butane (IC4), normal butane (IC4), iso-pentane (IC5), normal pentane (NC5), hexane (C6), heptane (C7), nitrogen (N2), and CO2. Martini et al. (2003) sampled Antrim gas in 73 wells in northern, northwestern, eastern, central and southern Michigan. The purpose of Martini’s (2003) study was to discriminate microbial from thermogenic origin natural gas samples in the Antrim in five separate geographic locations listed above.

Within the dataset, 35 wells were sampled for C1 - C3, N2, and CO2 which will allow for comparison with the MPSC dataset. After combining the data, 82 total wells were used to analyze the variation in Antrim gas composition in relation to spatial distribution (Table

3.5). The relative amounts of C1, C2, and C3 have commonly been used to distinguish bacterial from thermogenic gas because microbes do not produce significant quantities of hydrocarbons other than C1 (Bernard et al., 1978). Although the variation in C1 – C3 can suggest the origin of natural gas, it is not a direct indicator. Secondary processes, such as migration, bacterial oxidation, and mixing, further complicate identification of gas- generation mechanisms (Martini, 2003).

# of Wells Location C1 C2 C3 N2 CO2 Log(C1/C2+C3)) 5 Central 68.34 7.93 3.26 15.39 0.11 0.79 6 East 75.36 8.57 2.85 11.93 0.40 0.82 55 North 89.16 1.40 0.55 1.56 15.44 1.66 5 South 58.68 3.67 0.86 34.13 3.88 1.11 11 NW 84.58 0.58 0.29 8.03 5.74 1.99

Table 2.4 Table summarizing the gas analysis data used in this study to help suggest the processes responsible for natural gas production. In the eastern locations, five of the six wells were sampled from Berea produced gas.

50 RESULTS AND DISCUSSION

Regional Stratigraphy

The stratigraphy of the Upper Devonian-Lower Mississippian was mapped based

on a combination of the wireline log and XRF responses. The results are displayed using

isopach maps for each stratigraphic unit and 10 cross sections (five EW and five NS)

shown in the appendix. Formation tops were initially picked based upon variations in

wireline log response using wells with full suites (GR, NPHI, RHOB, and PEF). Table

3.1 summarizes the average log values for the different Upper Devonian-Lower

Mississippian intervals. In addition to wireline log analysis, XRF was used to further

confirm variations in lithology especially in regards to the different shale intervals.

Avg GR Avg NPHI Avg RHOB Avg Interval (API) (%) (g/cm3) PEF Sunbury 203 29.7 2.490 3.144 Berea 87 23.3 2.520 2.769 Bedford 113 23.8 2.607 3.244 U Antrim 204 30.2 2.468 3.058 Ellsworth 168 28.4 2.531 3.186 Lachine 274 32.9 2.416 3.156 Paxton 164 26.7 2.531 3.304 Norwood 297 33.7 2.365 3.117 Table 3.1 Summary of the average log responses in relation to the different Upper Devonian- Lower Mississippian intervals.

The different formations and members in the Upper Devonian-Lower

Mississippian displayed three distinct log facies when transitioning up section (Figure

3.1). After understanding the mineralogy of the different units (Leddy et al., 1980; 51 Ruotsala, 1980; Young, 1980; Balthazor, 1991; Dellapenna, 1991), different inferences are made to help explain the well log responses. The black shales (Norwood, Lachine,

Upper Antrim, and Sunbury) have a lower PEF and RHOB and a higher GR and NPHI when compared to the units above and below. Organic-rich shale typically has a higher

GR and a lower RHOB because of the greater concentration of organic material. Organic material has a low RHOB and tends to absorb radioactive elements especially U, and Th.

The increase in NPHI is attributed to the increase in hydrogen, present in the bound water of clay minerals and/or the high organic content, in these shales. The light black-gray shales (Paxton, Ellsworth, and Bedford) demonstrate the opposite with a higher PEF and

RHOB and a lower GR and NPHI when compared to the associated units above and below. The log response is mainly attributed to both a decrease in organic material and an increase in carbonate material (especially in the Paxton and Ellsworth). The Berea

Sandstone demonstrates and overall drop in GR, NPHI, RHOB, and PEF mainly because of the abundance of quartz and decrease in clay mineral and organic material content.

52 Figure 3.1 Type log which demonstrates the different Upper Devonian-Lower Mississippian log facies in this project. There is a consistent transition from organic-rich black shale to light black-gray shale. 53 Cross-sections were developed after distinguishing the different log facies and

how they correlate to the different Upper Devonian-Lower Mississippian lithologic units.

The cross-sections are displayed in the appendix and the color fill between wells was

accomplished using the interpretive fill tool in Petra to help demonstrate the variation and

patterns in GR distribution. The API color range is from 0 (bright yellow) to above 450

(black) API units. Prior to using the interpretive fill, GR curves were normalized in Petra

by adjusting the digital GR log curves to a common scale range by stretching or

compressing curves using the computed average high and low range to diminish variation

caused by different GR log calibrations from well to well.

The distinctions amongst the Ellsworth, Lachine and Upper Antrim Shales in the eastern basin is better defined using XRF. The log facies are less apparent as the

Ellsworth thins to the east. The XRF data offers higher resolution discrimination of these units. The Ellsworth log facies, in the eastern basin is identified on the basis of an increase in elements associated with carbonates (Ca and Mg) and a decrease in elements associated with quartz (Si). This is discussed in greater detail in the Ellsworth section.

Combining lithologic interpretations from log (petrophysical) interpretations with the isopach map geometry is a crucial step in understanding the changes and controls of the depositional environment of subsurface formations. Isopach maps were developed for each Upper Devonian-Lower Mississippian formation and member to infer geological controls on changes in the depositional setting through time and discuss some of the possible paleogeographic implications .

54 Squaw Bay Limestone

The structure map on the top Squaw Bay Limestone (Figure 3.2) demonstrates an extremely gently dipping, nearly circular basin with the basin center located more than

2,400 feet below sea-level in Clare County. This surface forms the base of the Antrim

Shale and marks a transition from shallow marine carbonate deposition to a period primarily controlled by deep marine sedimentation. The major structure of the northwest-southeast trending Howell anticline can be traced in and around Livingston

County where shallower subsea contours extend further towards the center of the basin.

55 Figure 3.2 Structure map of the Squaw Bay Limestone and the base of the Antrim Shale. The center of the basin is located in Clare County where the base of the Antrim is 2,400 feet below sea level. The structure lines bulging towards the central basin in SE Michigan are from the Howell Anticline.

56 Antrim Shale

Figure 3.3 is an isopach map of the Lower Antrim which includes the Norwood,

Paxton, and Lachine Members. The Lower Antrim is equivalent to Ells’ (1979) units 1A,

1B, 1C, and 2 (Figure 1.1). The overall thickness ranges from around 35 ft. to over 185 ft. in the northwestern part of Clare County. The north and southern margins both demonstrate an erosional truncation. The isopach map suggests the sediment distribution of the Lower Antrim was primarily controlled by basin centered subsidence. The thickness of the individual members will be discussed in greater detail.

57 Figure 3.3 Isopach map of the Lower Antrim, which is the gross thickness of the Norwood, Paxton, and Lachine Members. The Lower Antrim is thickest in Clare County.

A majority of silica in the Antrim Members was precipitated as authigenic quartz from formation waters or as silt sized grains both detrital and biogenic (Hathon, 1979;

Hathon et al., 1980; Dellapenna, 1991). The Si/Al ratio of many of the samples from the

58

Upper Devonian-Lower Mississippian shales of the Michigan Basin tend to plot above

the detrital trendline (Figure 3.4) developed by Blood et al. (2013). This further

confirms that the Antrim shales are enriched in non-detrital silica. The Paxton Member,

however, has a greater percentage of samples that plot closer to the detrital proxy.

Figure 3.4 Cross-plot of Si/Al demonstrating a positive correlation. The black line is a detrital proxy developed by Blood et al. (2013). A majority of the Antrim members have higher Si concentrations than the detrital proxy suggesting possible biogenic Si enrichment (Hathon et al., 1980). A majority of the samples from the Paxton Member plot closer to the detrital proxy.

Norwood Member Results

The Norwood has the highest GR signature of the Antrim members, averaging

297 API (Table 3.1). The black, organic-rich Norwood Member extends throughout most of the basin and ranges from 0 to just under 40 feet in thickness. The thickness increases towards the north central part of the basin with the depocenter centered in southern 59 Gladwin County (Figure 3.5). There is some uncertainty of the Norwood thickness in the west-southwest basin near Kent County because the Paxton Member is very thin to absent (Figure 3.7) making it difficult to distinguish the Norwood from the Lachine. The zero edge near the northern and southern margins is due to an erosional truncation of the formation. The Norwood Member is equivalent to Ells’ (1979) unit 1C (Figure 1.1).

60 Figure 3.5 The organic-rich Norwood Shale is thickest in the north central basin and thins to the south. The shale is thickest in Gladwin County where there is over 35 feet of the Norwood Member. The Norwood top becomes difficult to pick in the southwest due to the thinning of the Paxton, making it difficult to distinguish the Norwood from the Lachine Member.

61

Norwood Member Discussion

Rapid subsidence in the Michigan Basin induced the deposition of stagnant, organic-rich, black mud which was devoid of oxygen, preserving OM. The preservation of OM could have also been the result of syndepositional pyrite. Initial deposition of the

Norwood is equivalent to the in the Appalachian Basin (Gutschick and

Sandberg, 1991) and the Blocher Member of the Illinois Basin (Collins, 1967) (Figure

1.9). Upper Devonian subsidence in the Michigan and Illinois Basins was initiated at the same times based on conodont faunal ages (Gutschick and Sandberg, 1991). Based on the age of the sediment (Gutschick and Sandberg, 1991) and the thickness of the member, the maximum rate of deposition at this time was around 18 ft./Ma. The Norwood isopach map suggests the detrital sediment may have been restricted in the southeast and western margins by the Findlay and Wisconsin arches (Figure 1.4). The thickness distribution was primarily controlled by different rates in subsidence.

The Norwood isopach map is very similar to the map developed by Gutschick and

Sandberg (1991) (Figure 3.6). Gutschick and Sandberg (1991) used Ells’ (1979) unit 1C picks for the Norwood Member (Figure 1.1). There is a variation between the interpretations in Saginaw County and the southern basin margin. Gutschick and

Sandberg (1991) have the Norwood thickening in these two areas (over 45 ft. in Saginaw

County) whereas in this study the Norwood thins in Saginaw County and is less than 10 ft. towards the southern margin. Cross-section I-I’ has two wells in Saginaw County and demonstrates the thickening and thinning of the Norwood log facies towards the southern margin.

62 Figure 3.6 Gutschick and Sandberg’s (1991) isopach map of the Norwood Member based on Ells’ (1979) unit 1C. The map is similar to the interpretation from this study with the exception of the thickening highlighted in the box. Cross-section II’ demonstrates the thickening and thinning of the Norwood log facies.

Paxton Member Results

The top and base for the Paxton is, in most cases, more obvious due to the lower

GR (164 API) compared to the Norwood below and Lachine above. The PEF value is the highest of the three members because of the higher concentration of calcite and dolomite

(Dellapenna, 1991).

The Paxton Member has two depocenters in Michigan (Figure 3.7). The northern depocenter is centered on Clare, Missaukee and Osceola Counties, where the Paxton

Member exceeds 70 ft. in thickness. The southern depocenter is centered on Jackson and

Ingham counties, and the Paxton Member is over 40 feet thick. The member thins in the central part of Michigan following an east-west trend. The Paxton Member is equivalent

63 to Ells’ (1979) unit 1B (Figure 1.1). There is an erosional truncation towards the northern and southern basin margin.

Figure 3.7 Isopach map of the Paxton Member of the Antrim Shale. The unit has two depocenters. The Paxton is thickest in the north exceeding 7 ft. in thickness.

64 Paxton Member Discussion

An abrupt transition in deposition from anaerobic, organic-rich mud of the

Norwood Member below, to dysaerobic, lighter colored muds and lime muds of the

Paxton Member occurred in the Michigan Basin during the Late Devonian. The beginning of Paxton deposition is correlated to the (base-middle of the

Ohio Shale Huron Member) which is the start of the cycle 4 in the Catskill Delta in the

Appalachian Basin (Figure 1.9; Ettensohn, 1985; Gutschick and Sandberg, 1991).

The lithology and fauna in the Paxton Member mark a significant period in the

Upper Devonian. Lithologic changes in the organic-rich Norwood to lower organic content in the Paxton apparently resulted from a drop in sea-level and the development of a more oxygenated water column coincident with the increase in fine-grained clastic and carbonate sediments into the basin (Gutschick and Sandberg, 1991). The Paxton was deposited faster in the northern depocenter (47 ft./Ma) than the southern (27 ft./Ma).

This is most likely due to a difference in accommodation space and sediment supply.

The middle-lower Paxton contains the conodont fauna of the Lower gigas Zone and

marks the time of a major eustatic rise throughout Euramerica (Sandberg et al., 1988).

However, the lithology of the Paxton suggests a drop in sea-level. Gutschick and

Sandberg (1991) suggest that an explanation for the difference in relative sea level trends

in the Michigan Basin is the result of isostatic rebound due to diminished Acadian

orogenic compression. The upper Paxton contains the linguiformis Zone which coincides with the time of mass extinction and a severe sea-level fall in Euramerica and North

Africa during the end of the Fransnian (Sandberg et al., 1988).

65 The geochemistry from the XRF data marks some notable changes in the Paxton with a decrease in U, Mo, and Si and an increase in and Ca (Figure 3.8) compared to stratigraphically adjacent members. Mo and U are indicative of reducing environments

(Algeo and Tribovillard, 2009), which further suggests the Paxton was deposited in a more oxygenated environment. The cross-plot in Figure 3.4 also suggests the Si in the

Paxton is primarily from a detrital source.

Figure 3.8 XRF data from the Krocker 1-17 well demonstrating the differences in geochemistry when compared to the Norwood and Lachine Members. The changes are primarily due to a relative followed by a eustatic drop in sea-level resulting in the deposition of dysaerobic muds in a more oxygenated water column (Gutschick and Sandberg, 1991).

66

The suggestion that the Paxton was deposited during a major eustatic rise in sea- level is within the realm of possibilities but the rate of isostatic rebound would have to be relatively fast to overcompensate the rate of sea-level rise. It is possible that the fauna of the Lower gigas Zone was reworked from the Norwood and later deposited in the Paxton during the severe sea-level drop at the end of the Fransnian.

There is not enough evidence to characterize the Paxton as a prodelta sediment.

A delta sequence typically demonstrates a coarsening upward with coarser detrital dominated sediment at the top of the sequence which is not noted in the Paxton. The lithology and thickness distribution of the Paxton suggests the sediment was deposited in a shallower more oxygenated basin primarily controlled by different rates of subsidence and accommodation space. The concentration of Si and Ca fluctuate in opposition of one another (Figure 3.8) suggesting a possible seasonal variation which would result in a cyclic transition from clastics to carbonate-dominated sedimentation.

Lachine Member Results

The Lachine Member has a higher GR signature, similar to the Norwood Member, averaging 274 API units. The Lachine Member is thickest towards the west (over 100 ft.) and thins to the southeast (Figure 3.9). The Lachine-Ellsworth contact in the northwest is gradational over a greater thickness interval than the contact in the east-southeast. In this study the gradational interval is incorporated into the thickness of the Ellsworth Shale.

Ells’ (1979) units 1A and 2 were grouped into the Lachine Member (Figure 1.1).

67

Figure 3.9 Isopach map of the Lachine Member. Note that the member thins to the southeast from a maximum thickness of 100 ft.

Lachine Member Discussion

The Lachine is correlated to the Dunkirk Shale (middle of the Huron Member of the Ohio Shale), which coincides with the start of cycle 5 in the Catskill Delta in the

Appalachian Basin (Figure 1.9; Ettensohn, 1985; Gutschick and Sandberg, 1991). The

68 black, organic-rich mud lithology and slow rate of deposition (25 ft./Ma) also suggests the basin was starved and oxygen deprived at this time.

The contact between the Paxton and Lachine is believed to be the -

Famennian extinction boundary (Gutschick and Sandberg, 1991; Formolo et al., 2014).

Elevated trace metal enrichments at the base of the Lachine coincide with the extinction event and suggest euxinia may be a leading contributor to the mass extinction events

(Formolo et al., 2014). The XRF data demonstrates an increase in S and Fe at the

Paxton-Lachine contact (Figure 3.10) most likely associated with pyrite commonly noted in the Lachine (Leddy et al., 1980; Dellapenna, 1991).

Figure 3.10 A cross-section of three wells with XRF data displaying and increase in S and Fe (most likely attributed to pyrite) at the transition from the Paxton to the Lachine Member. Formolo et al. (2014) describes the contact as the Frasnian- extinction boundary triggered by euxinic conditions and an elevated chemocline.

69

The position of the depocenter for the Lachine (Figure 3.9) suggests that sediment

was primarily restricted in the southeast by the Findlay Arch. The presence of siliceous

organic-rich black shale suggests a resumption of a deep oxygen deprived basin

transitioning from the Paxton. Subsidence towards western Michigan is greatest based on

the thickness distribution. This was possibly the result of both greater compression to the

west associated with the Antler orogeny and lessened compressional forces from the

Acadian orogenic event. The gradational transition into the Ellsworth also suggests uplift

further west supplying the sediment for the Ellsworth Shale (Figure 3.11).

Figure 3.11 Depositional model for the Lachine-Ellsworth transition period. Greater tectonic forces to the west (possibly associated with the Antler orogeny) caused uplift and greater subsidence towards the western part of Michigan. Modified from Cloetingh (1992).

Ellsworth Shale Results

The Ellsworth Shale is more easily differentiated from the Lachine (below) and the Upper Antrim (above) in the western portion of the Michigan Basin using only GR.

In the eastern Michigan Basin where the Ellsworth is much thinner (Figure 3.12) a

70 combination of available petrophysical and geochemical data including GR, XRF, PEF,

RHOB, and NPHI was used to map the extent of this unit.

The Ellsworth Shale is the thickest unit in the Upper Devonian exceeding 700 feet in the west-northwest (Figure 3.12). The Ellsworth uniformly thins towards the east- southeast where it is less than 20 ft. thick. There is some interfingering with the underlying Lachine Shale and the overlying Upper Antrim but the Ellsworth extends further east than previous studies had mapped (Asseez, 1969; Ells, 1979; Fisher, 1980;

Matthews, 1989; Gutschick and Sandberg, 1991). The GR signature increases in the

Ellsworth towards the east away from the westerly source of the sediment (Asseez, 1969;

Matthews, 1989).

71 Figure 3.12 Isopach map of the Ellsworth Shale extending throughout the basin. The formation demonstrates a uniform thinning towards the east-southeast in Michigan.

Ellsworth Shale Discussion

Previous studies, pertaining to the thickness of the different Upper Devonian formations, generally combine the Ellsworth Shale and the Upper Antrim in central and

72 eastern Michigan (Matthews, 1989). Gutschick and Sandberg (1991) describe the

Ellsworth as the prodelta facies of a delta complex sourced from the west, based on the

geometry of the isopach map pattern. The Ellsworth is correlated approximately with the

greenish gray in the Appalachian Basin during the Lower marginifera

Zone (Gutschick and Sandberg, 1991).

The contact between the Lachine and Ellsworth is a gradational coarsening upward sequence, common in delta complexes. The isopach map of the Ellsworth thickens to over 700 ft. to the west, suggesting sediment sources from the

Transcontinental Arch and Wisconsin Uplands (Figure 3.11), which were deposited at a rate of up to 100 ft./Ma. The Ellsworth isopach indicates a uniform, regional thinning to the east.

The Upper Antrim, Bedford, and Berea thin and pinch out to the west placing the

Sunbury, when present, directly over the Ellsworth in the western Michigan Basin. The

Bedford may extend into Muskegon and Ottawa County such that the Bedford directly overlies the Ellsworth in these counties. However, the high gamma-ray signature of the

Sunbury, which separates the Coldwater from the Bedford-Berea sequence, is less apparent or absent in the western basin. In this location it is difficult to distinguish the

Bedford from the . The Coldwater directly overlies the Ellsworth in areas to the northwest and southeast where the Upper Antrim, Bedford, Berea, and

Sunbury are all absent.

The Lachine-Ellsworth and Ellsworth-Upper Antrim contacts in the central and eastern Michigan Basin are difficult to identify using only GR log. Gutschick and

73 Sandberg (1991) used the appearance of Protosalvinia (Foerstia) during the Upper rhomboidea Zone as an indicator for the beginning of the Ellsworth. Although a chronostratigraphic approach for the identification of this contact is ideal, core and outcrop in the Ellsworth is scarce and cuttings cannot give a precise depth.

To address this uncertainty and determine the eastern extent of the Ellsworth

Shale, XRF elemental data was compared in wells in Sanilac County (DOW 103 and

104) with the Amos C1-4 and Krocker 1-17 wells located in western and central

Michigan, respectively (Figure 2.3). The geochemistry of the Ellsworth in the Krocker and Amos wells, compared to the Lachine, has less Si, Mo, U, Th, and K and a higher concentration of Ca and Mg. The DOW 103 and 104 cores demonstrate a similar geochemistry at a small interval (10 ft.) around 1250-1260 ft. MD (Figure 3.13).

74 Figure 3.13 Four of the six wells cored in the Upper Devonian Michigan Basin that were analyzed using XRF. The Ellsworth has a higher concentration of Mg and Ca and a lower concentration of Si, Mo, and U+Th+K compared to the Lachine and Upper Antrim. Track 1: GR, track 2: MD, track 3: U+Th+K (black), track 4: Mo (black), track 5: S (orange), track 6: Ca (blue), Mg (purple), Si (yellow), track 7: Al (black), Fe (orange).

75 The interfingering relationship between the two Antrim members is much more

complicated than displayed in the cross-sections. However, the XRF and wireline log

data do suggest that facies similar to the Ellsworth are present on the eastern part of the

basin. A more detailed analysis could be done on this interval to better understand the

relationship and to definitively determine the geological origin of the Ca and Mg in the

Ellsworth Shale.

The origin of the Ca and Mg in the Ellsworth is unclear and does not fit well with the prodelta interpretation because carbonates do not typically coexist with clastic dominated systems. The Ca and Mg was either precipitated during the progradation of the Ellsworth or was diagenetically incorporated through later cementation. There are modern analogues (e.g. Northern Great Barrier reef, Red Sea Reefs, Caribbean reefs and platforms off Nicaragua, Abrolhos and Recife reefs off Brazil, and reefs in the Java archipelago) which have specific environmental conditions that allow for carbonate deposition in close vicinity, if not directly within, siliciclastic settings (Wilson, 1967;

Chave, 1967; Mount, 1984; Roberts, 1987; Friedman, 1988; Riegl and Piller, 1996;

Leinfelder, 1997; Eliuk and Wach, 2014). Leinfelder (1997) summarizes the different analogs, some of which fit the setting of the Ellsworth. The Red Sea reef stretches over

6.5 million ft. along the coast of Egypt, Sudan, and Eritrea and is surrounded by coeval siliciclastic sedimentation. The reefs are able to grow because the aridity of the climate reduces clastic sedimentation events, giving the carbonate factory time to flourish (Riegl and Piller, 1996). During the Late Devonian, the Michigan Basin was located near the paleoequator in the Eastern Interior Seaway (Figure 1.3) and the climate was believed to be arid due to a rain shadow from the Acadian Mountains to the south and east

76 (Ettensohn, 1985; Algeo and Tribovillard, 2009). Similar settings in the Red Sea may

have been present during Ellsworth deposition towards the western margin and

contributed Ca and Mg as a marl component in the shale lithology. Longshore currents

and depositional trapping in estuaries have also been cited to reduce the amount of clastic

input in areas of carbonate production (Roberts, 1987; D’Aluisio-Guerrieri and Davis,

1988; Leinfelder, 1997).

The presence of silt stringers in the middle and upper part of the Ellsworth towards the western basin margin suggest there is a western source of coarser grained clastic sediment. The thickness distribution of the Ellsworth is similar to sediment that was deposited in a prodelta environment suggesting a detrital source primarily controlled the isopach patter. Any further evidence of a delta plain has been removed during glacial erosion and the formation of . The depositional interpretation fits with the model suggested in Figure 3.11.

Upper Antrim Member Results

The Upper Antrim Shale is thickest to the east-northeast and thins towards the west (Figure 3.14). The pinchout of the Upper Antrim follows a north-south trend in the western part of Michigan. The Upper Antrim is over 145 feet thick in Arenac and northern Bay County. The contact with the overlying Bedford Shale is typically sharp when the Bedford Shale is present.

77 Figure 3.14 Isopach map of the Upper Antrim Shale in Michigan. The shale is thickest to the east and thins to the west until it eventually pinches out.

Upper Antrim Shale Discussion

The Upper Antrim is equivalent to Ells’ (1979) units 3, 4, 5, and 6 and was correlated by Matthews (1983) with the Cleveland Member of the Ohio Shale in the

Appalachian Basin. Upper Antrim deposition corresponds to a black-shale depositional 78 phase at the start of cycle 6 of the Catskill deltaic facies (Figure 1.9; Ettensohn, 1985).

Deposition of the Cleveland started when deep water spread to the northern Appalachian

Basin and joined to the Michigan Basin through the Chatham Strait (Figure 1.4;

Gutschick and Sandberg, 1991). The Upper Antrim isopach map (Figure 3.14) also

suggests that the sediment was sourced from the northeast and southeast at a maximum

rate of 32 ft./Ma.

The pinchout of the Upper Antrim towards the western basin margin is the result of either a facies pinchout or an unconformity. The facies pinchout model would imply that the basin center was further to the east and it was an oxygen-stratified basin. This scenario suggests the deposition of the Upper Antrim took place during the deposition of the Ellsworth and the lateral distribution of facies is primarily controlled by the position of the pycnocline. Figure 3.15 is a depositional model which was proposed by

Dellapenna (1991) for the Antrim Shale. An oxygen-stratified basin would explain the different transitions from dysaerobic sediment to anaerobic sediment.

Figure 3.15 Depositional model for the Upper Antrim Member in an oxygen-stratified basin. This model explains the facies pinchout towards the western basin margin. Modified from Dellapenna (1991).

79 The second scenario suggests the Upper Antrim facies was deposited further west in the basin but was later exposed and eroded during a drop in sea-level. The time of exposure has been suggested to take place during the post-Sunbury—pre-Coldwater

(Matthews, 1989). Uplift in the west would expose the western region while not affecting the central and eastern basin (Figure 3.16). In southwestern Michigan, Bishop

(1940) noted a very apparent unconformity at the top of the Ellsworth. Matthews (1989) used this observation to suggest the erosional event taking place pre-Coldwater which explains the truncation of the Upper Antrim, Bedford, and Sunbury Shales to the west.

There is not enough evidence in this report to suggest the extent of erosion or when erosion took place but an erosional event is feasible to explain the pinchout of the Upper

Antrim and the truncation of the Bedford and Sunbury Shales.

Figure 3.16 Erosional model from Matthews (1989) used to explain the truncation of the Upper Antrim, Bedford, and Sunbury Shales.

80

The first scenario is most likely the main control on the thickness distribution of the Upper Antrim Member. As the Ellsworth prograded into the basin from west, the eastern part of the basin continued to subside (possibly from a combination of sediment loading and increased compression from the Acadian orogeny) creating more accommodation space. The rise in sea-level forced the retreat of the Ellsworth and the expansion of the Upper Antrim.

Bedford Shale Results

The Bedford Shale is over 200 ft. thick towards the eastern basin margin. The

Bedford thins and eventually pinches out to the west (Figure 3.17). The Bedford has a localized thickening in and around Kent County towards the western basin margin. The

Bedford Shale decreases in thickness in Bay, Saginaw, and Genesee Counties where the

Berea Sandstone is thicker (Figure 3.19). Ells’ (1979) units 7 and 8 are equivalent to the

Bedford Shale (Figure 1.1).

81 Figure 3.17 The Bedford Shale isopach map demonstrating a thinning to the west until the shale eventually pinches out in the west.

Bedford Shale Discussion

The Bedford Shale is the prodelta of the Bedford-Berea sequence and was deposited in both the Michigan and Appalachian Basins (Asseez, 1969; Matthews, 1989;

Balthazor, 1991). A drop in sea-level caused by isostatic rebound was the driving 82 mechanism for the westward progradation of the delta sequence into the Michigan Basin.

The river most likely branched northeast of the basin near the Old Red Continent flowing

south-southwest along both sides of the Algonquin Arch (Figure 1.4). Initiation of

Bedford Shale deposition in the Appalachian Basin was initiated earlier than deposition

in the Michigan Basin (Pepper et al., 1954). This relationship was first suggested on the

basis of red beds in the Bedford Shale in Ohio not observed in Michigan. The source of

the red sediment in the Bedford of Ohio may have been eroded before the river diverged

west of the Algonquin Arch into the Michigan Basin. In addition, cores in Midland and

Sanilac Counties in the Upper Antrim show no sign of soft sediment deformation

(Matthews, 1989), which is noteworthy at the top of the (Lewis, 1982) in

Ohio. The contact between the Upper Antrim and Bedford is sharp in most areas of the

Michigan Basin (similar to the one shown in Figure 1.8) suggesting a paraconformity

rather than a continuous gradational transition into the Bedford. These relationships

suggest that the Bedford Shale was deposited more slowly and possibly following a

depositional hiatus above the Upper Antrim Shale in Michigan precluding soft sediment

deformation in the Upper Antrim.

The geometry of the Bedford suggests there was a northern and southern lobe associated with Bedford-Berea sequence (Figure 3.18). The southern lobe extended further southwest but was later removed by post-Sunbury—pre-Coldwater erosional truncation, possibly concurrent with the pre-Burlington truncation of the Hannibal-

Saverton Shale interval in the northwestern Illinois Basin (Bishop, 1940; Cluff et al.,

1981; Matthews, 1989).

83 Figure 3.18 Map showing the approximate size and orientation of the Bedford prodelta lobes based on the geometry of the isopach map (left). The map suggests the sediment was deposited directly west-southwest from the Algonquin Arch. The dashed line towards the southwestern basin margin is the approximate location of the pre-Coldwater nonconformity (Bishop, 1940; Matthews, 1989).

Berea Sandstone Results

The Berea is thickest in the east and thins to the west until it pinches out (Figure

3.19). The predominantly sandstone unit is over 160 feet thick in Bay, Saginaw, and

Genesee Counties where there is a thinning of the underlying Bedford Shale (Figure

3.17). The Berea Sandstone is equivalent to Ells’ (1979) unit 9 (Figure 1.1) and marks the end of Devonian deposition in the Michigan Basin (Gutschick and Sandberg, 1991).

84 Figure 3.19 Isopach map of the Berea Sandstone. The sandstone becomes thickest when there is a thinning within the underlying Bedford Shale most likely associated with the major channels incising down into the Bedford (Figure 3.17).

Berea Sandstone Discussion

The Berea Sandstone is a deltaic deposit comprising delta front, destructive marine shale, abandoned distributary channel-fill, interdistributary bay-fill, and

85 transgressive marine sandstone facies (Balthazor, 1991). Balthazor (1991) describes in detail different facies in the Berea Sandstone. These facies are the result of autocyclic events associated with delta environments that facilitate deposition of muddy sediment

(abandoned channels, interdistributary bays, floodplains, etc.) in some locations, whereas in adjacent regions coarser sediment is deposited.

During a regression and the progradation of the sequence, channels incised down into areas of the Bedford Shale (Figures 3.20). The modern well log control diminishes to the east (Figure 2.1) making it difficult to properly map the channels and other sand bodies in detail. However, the approximate orientation of feeder channels can be estimated based on the Berea Sandstone isopach (Figure 3.21).

A majority of previous literature places the Berea Sandstone in the Mississippian but the Bedford Shale and Berea Sandstone both contain the spore Retispora lepidophyta which is a global indicator of the latest Devonian (Gutschick and Sandberg, 1991). It is possible the sediment was reworked from an older unit and redeposited in the Berea, however, the Berea was deposited during a major regression that coincides with a eustatic fall during a glacially induced global mass extinction event (Sandberg et al., 1988).

86

Figure 3.20 Eastern portion of cross-section D-D’ demonstrating how the Berea Sandstone incised down into the Bedford Shale in areas the channel was most likely present.

Figure 3.21 Inferred and approximate orientation and location of the major channels responsible for the deposition of the Berea Sandstone in the Michigan Basin. The sediment was most likely sourced from the north and adjacent Algonquin Arch. The orientation and location of the Algonquin Arch was modified from Gutschick and Sandberg (1991).

87 Sunbury Shale Results

The shale is thickest in the east around Sanilac County and thins to the west

(Figure 3.22). The Sunbury Shale has a higher gamma-ray signature (203 API average) than either the underlying Berea Sandstone or the overlying Coldwater Shale, which makes the unit a good marker for correlation (Figure 1.8). A notable truncation of the

Sunbury Shale occurs in southwestern Michigan. Towards the eastern basin margin the

Sunbury becomes much thicker exceeding over 120 ft. There is another thickening of the

Sunbury towards the western basin margin in Newaygo County. Ells (1969) does not assign a unit number for the Sunbury Shale.

88 Figure 3.22 Isopach map of the Sunbury Shale. The Sunbury Shale is thickest in Sanilac and Huron counties and thins to the west.

The entire Upper Devonian – Lower Mississippian ranges in thickness from under

250 feet to over 850 feet (Figure 1.6). The increase in thickness to the east and northwest

89

is attributed to the Bedford-Berea and Ellsworth both thickening towards eastern and

western basin margin (respectively) and thinning towards the center of the basin.

Sunbury Shale Discussion

The Sunbury marks the beginning of the Mississippian and is associated with a major eustatic rise in the Lower crenulata Zone (Gutschick and Sandberg, 1991). The formation directly below the Sunbury is the Ellsworth, Upper Antrim, Bedford, or Berea and was determined by overlying the isopach maps from the Berea-Ellsworth tracing the pinchout of each formation (Figure 3.23). The Sunbury Shale is much thicker towards the eastern margin which could be a result of accommodation space due to the eastern tilt of the basin (Howell and van der Pluijm, 1996) and/or rapid subsidence due to flexure and deltaic sediment load (Gutschick and Sandberg, 1991). The previously mentioned pre-Coldwater erosional event could also explain the truncation to the west (Matthews,

1989).

90 Figure 3.23 Map showing the spatial distribution of the formations directly below the Devonian-Mississippian boundary.

91 Ellsworth Petroleum System Results and Discussion

The regional stratigraphy discussed in the previous section will help in understanding the Ellsworth petroleum system. This section will break down the different components of a petroleum system (source, reservoir, and seal) to better understand the different mechanisms and attributes responsible for the production.

Source

The source for the hydrocarbons has been primarily theorized to be either the

Ellsworth or the Antrim Shale. Oil generation in the Upper Devonian is believed to have taken place during the (Cercone and Pollack, 1991). Table 3.2 summarizes some of the different source rock attributes and the requirements to generate hydrocarbons.

The first source rock component analyzed is the TOC. Without enough OM hydrocarbon production cannot take place. The variation in TOC within each formation is most likely attributed to the position of the sediment in the water column when it was deposited. TOC is typically highest just below wave base and decrease both above and below (Moyer, 1982). For this study TOC values were averaged by formation to get a broad sense of organic content for each formation rather than specific target depths. TOC values for each member and formation are highlighted green if the values are 2 (wt. %) or greater which is an estimated value for good-excellent source rock potential (Table 3.1;

McCarthy et al., 2011).

92 Figure 3.24 The TOC spatial distribution in the Norwood, Paxton, Lachine, and Upper Antrim. Green values represent TOC values that have good-excellent source rock potential (McCarthy et al., 2011).

The Norwood Member is one of the primary unconventional reservoirs in the

Antrim because of the higher TOC ranging from 1-14 (wt. %) (Figure 3.24). TOC in the

Paxton varies from 0.75-6.5 (wt. %) and is not typically targeted in the Antrim.

However, some of the TOC values are high enough in areas to produce hydrocarbons

(Figure 3.24). There are thin black shale beds within the Paxton Member. Samples may 93 have been picked in the black beds which do not properly represent the bulk lithology.

The Lachine is another exploration target in the Antrim with TOC values ranging from

1.75-8 (wt. %). The Upper Antrim Member is another unconventional target in the

Antrim and has also been suggested to be the source rock for the Berea petroleum system

(Dolton, 1996). Overall, TOC values in each of the Antrim members appear to be sufficient enough to produce hydrocarbons.

94

Figure 3.25 TOC spatial distribution in the Ellsworth Shale. TOC values are lower in the Ellsworth but there are recorded TOC values considered to have good source rock potential (McCarthy et al., 2011).

The TOC in the Ellsworth Shale is relatively low (average 1.23 wt. %) when compared to the Antrim Members above and below, which range from 0.33-7.75 (wt. %)

(Figure 3.25). The Ellsworth is most likely not the primary source for the hydrocarbons produced out of the Ellsworth petroleum system. 95 The next piece of data to analyze is the type of kerogen in the source rock.

Previous literature pertaining to kerogen type in the Antrim divided the Antrim into black and gray Antrim. Most source rock data in the black Antrim suggests the kerogen is amorphous Type I derived from Tasmanites which is a marine algal phytoplankton common in epieric seas (Moyer, 1982; Dellapenna, 1991; Martini, 2003). The gray

Antrim is believed to be Type II and III kerogens with a higher concentration of terrestrial OM which was more highly oxidized at the time of deposition (Martini, 2003;

Dellapenna, 1991). The combined Rock Eval data, when plotted on the Van Krevelen

Diagram, further suggests primarily Type I and II kerogens (Figure 3.26). Type I and II kerogens are both associated with marine environments (Type I can be lacustrine as well) consistent with the depositional setting of the Upper Devonian. The kerogen type is important because the temperature at which oil and gas generation takes place varies with the different kerogens (Table 3.2; McCarthy et al., 2011). Type I kerogen starts generating oil at around 445°C and Type II kerogen at around 435°C (Table 3.2).

96 Figure 3.26 Rock-Eval data plotted on a Van Krevelen Diagram to determine the kerogen type in the different source rocks. The Antrim Shale appears to be primarily Type I and II kerogen.

Table 3.2 Summarization of source rock criteria from McCarthy et al. (2011). Most of the source rocks in the Upper Devonian-Lower Mississippian are characterized as being Type I and II kerogens which begins the onset of oil generation at around 445⁰C and 435⁰C respectively. Modified from McCarthy et al. (2011).

The thermal maturation data is plotted throughout the Michigan Basin with the field map to demonstrate the maturation distribution with respect to the Ellsworth 97 producing fields (highlighted in gray, Figure 3.27). The maturation control diminishes around the Ellsworth producing fields but the few samples around the area appear to be in the transition window or are still relatively immature.

98 Figure 3.27 Map showing the maturation spatial distribution. The area where there is Ellsworth production is highlighted in gray. The color of the number represents the type of measurement used to determine the maturation. The color of the circle represents the level of maturation. The samples around the Ellsworth production are either immature or in the transition window.

Several scenarios are proposed for hydrocarbon migration from Antrim source rock to Ellsworth reservoirs located on the west side of the Michigan Basin (Figure 3.27).

99 The hydrocarbons may have migrated up dip from the east, where the Antrim Shale is

more deeply buried and mature, or sourced from the Antrim directly below and

transported through vertical fault networks, or formed through biogenic processes similar

to the large Antrim gas fields in (Figure 1.10).

The first scenario is suggested because the Antrim is too shallowly buried, on the western margin of the Michigan Basin, to have reached thermal maturity and produced gas or oil, assuming a regional Michigan Basin geothermal gradient of approximately

20.1 ⁰C/Km (Cermak and Raybach, 1982). Deeper in the basin, to the east, oil and much lesser amounts of gas may have been produced in these deeper basinal settings, but natural gas is the primary hydrocarbon produced from Ellsworth reservoirs in the western basin possibly because gas has higher relative permeability compared to oil and would more likely migrate substantial distances, updip (Figure 3.28). This scenario would also explain the Berea Sandstone fields producing gas further updip in Ogemaw, Iosco, and

Arenac Counties (Figure 1.10).

100 Figure 3.28 The maturation spatial distribution in relation to burial depth. With the exception of a few outliers, the maturation generally increases deeper in the basin.

The other proposed scenario suggests the hydrocarbons migrated vertically from the Antrim directly below, or in close proximity to the western Michigan fields, through localized fractures or fault trends possibly related to the Mid-Michigan Gravity Anomaly

(MMGA). The Ellsworth fields follow the western edge of a NW-SE trending gravity high in southwestern, Lower Michigan (Figure 3.29) suggesting there may be 101 unidentified structures defining the Ellsworth field trends in this area, analogous to faults

trends associated with Berea fields in east-central Lower Michigan. Similar well-

documented structures like the Albion-Scipio and Lucas Faults follow the same trend and

penetrate the crystalline (Fisher et al., 1988). These proposed,

basement rooted faults could serve as conduits for hydrothermal fluid migration which

would cause “forced maturation” or increased maturation in the source rocks in close

proximity to the fault/fracture conduits. Wagenvelt (2014) proposed a “forced

maturation” mechanism to explain higher maturation values throughout the Michigan

Basin, compared to maturation due to burial and regional geothermal gradients.

The process of hydrothermal dolomitization requires hydrothermal fluid. There are areas where the Trenton-Black River and the Middle Devonian Dundee have been hydrothermally altered from fluids transported from different fault networks.

There is a notable relationship when mapping the distribution of hydrothermal dolomite

(Barnes et al., 2008) in relation to gas versus oil Ellsworth and Berea fields (Figure 3.30).

This further supports the “forced maturation” on the Antrim source rock. Seismic data and more source rock data around the Ellsworth producing area would help confirm or disprove this theory.

102 Figure 3.29 The Ellsworth fields have a NW-SE trend similar to the MMGA. The Albion-Scipio and Lucas Faults are two well documented structures in the Michigan Basin that follow the same NW-SE trend (Fisher et al., 1988). This suggests there may be similar structures around the Ellsworth producing area that not only enhance vertical hydrocarbon migration but also caused force maturation on the source rocks the hydrothermal fluid came into contact with.

103 Figure 3.30 Map showing the distribution of Ellsworth and Berea oil and gas fields (green, red, and yellow) in relation to underlying Trenton-Black River and Dundee limestone and hydrothermal dolomite (light blue and purple). The Ellsworth and Berea gas fields are in close proximity to areas with documented hydrothermal dolomite which further supports the “forced maturation theory”. The distribution of dolomite and limestone is modified from Barnes et al. (2008).

The biogenic origin for Ellsworth gas occurrences is unlikely because of the thickness of the Ellsworth Shale and the distance of the producing fields from the subcrop

(Figure 3.31). The influx of glacially derived, surface waters, responsible for the biogenic origin of natural gas in the Antrim to the north, almost certainly could not have 104 migrated downwards through a much thicker bedrock overburden, including the reservoirs in the Ellsworth Shale, to transport surface-derived microbes to the Antrim and induce biogenic gas production. The concentration of methane decreases towards the central and southern basin, further suggesting the gas produced from the Ellsworth was thermogenic.

Figure 3.31 Map of the spatial distribution of gas chemistry of samples taken from the Antrim or Berea. The Antrim gas fields in the north are much closer to the Devonian subcrop and have a higher concentration of methane which is typically associated with biogenic gas (Martini, 2003). The concentration of methane decreases in the central and southern basin.

105 Reservoir

The Ellsworth contains silt stringers in the middle and upper section, which serve as the primary fluid reservoirs in the formation (Figure 3.32). The stringers are composed of gray, very-fine, well-sorted, subrounded, siltstone. A majority of well files label the silt or producing interval as “Berea” but the Ellsworth was deposited before the

Berea eliminating any possibility of an interfingering relationship. The silt stringers are difficult to map because of the well log control and thickness of the interval.

Figure 3.32 A cross-section showing the Ellsworth reservoir (siltstone). The reservoir is typically less than 10 ft. thick.

106 Seal

The Ellsworth green-gray shale above the silt stringers is the primary seal. The trapping mechanism is most likely a combination of structure and stratigraphy. The regional structural trends and the orientation of the Ellsworth fields suggests the trapping of hydrocarbons primarily relies upon structure. It is also likely the trapping is caused by a facies pinchout of the Ellsworth siltstone. A more detailed attempt to map the different silt stringers would further define the primary trapping mechanism.

107 CONCLUSION

Regional Stratigraphy

The Antler orogeny in the west (Poole and Sandberg, 1977) and the Acadian orogeny to the east (Ettensohn, 1985) of the Eastern Interior seaway (Figure 1.3) are two

important events relevant to interpreting the geological controls on sedimentation and

stratigraphy of Upper Devonian-Lower Mississippian strata in the Michigan Basin.

Catskill deltaic facies (Ettensohn, 1985), generated by cyclic, tectonic pulses correspond

to cycles in the Middle and Upper Devonian sequences in the Michigan Basin (Gutschick

and Sandberg, 1991). Compression associated with the formation of the Transcontinental

Arch and the Acadian Mountains initiated subsidence in the region between (Eastern

Interior Seaway, Michigan Basin, Appalachian Basin, and Findlay Arch; Gutschick and

Sandberg, 1991). This event leads to a depositional switch in the Michigan Basin from

an oxic shallow-water carbonate platform (Traverse Group and Squaw Bay Limestone) to

a deep anaerobic sea floor.

Initial deposition of the organic-rich Norwood Member took place in a deep

oxygen deprived basin restricted in the southeast and western margins by the Findlay and

Wisconsin arches. The distribution of thickness was primarily controlled by a greater

rate in subsidence in the northeastern part of the basin.

A brief major regression in the Late Frasnian leads to the development of a more

oxygenated basin and the deposition of the Paxton Member. The thickness distribution

108 suggests the basin had two depocenters. The geochemical fluctuations within the Paxton suggest possible seasonal variations which led to cyclic transitions from clastic to pelagic carbonate deposition.

Increased compressional forces from the west forced greater subsidence in western Michigan and led to the resumption of anaerobic organic-rich black shale of the

Lachine Member during the beginning of the Famennian. The sediment was primarily restricted in the southeast by the Findlay Arch. Sediment influx from the west increased towards end of the Lachine as the Ellsworth prodelta began to prograde further east into the Michigan Basin forming a gradational transition.

The Ellsworth continued to prograde east as organic-rich shale deposition continued downdip beneath the pycnocline in an oxygen-stratified basin. This developed an interfingering relationship between the Ellsworth and Lachine. The Ellsworth prodelta prograded further east in the Michigan Basin than previously suggested (Asseez, 1969;

Ells, 1979; Fisher, 1980; Matthews, 1989; Gutschick and Sandberg, 1991).

The retreat of the Ellsworth and the deposition of the Upper Antrim was triggered by a rise in sea-level. The distribution of the Upper Antrim was controlled by an influx of sediment from the Chatham Strait, increased subsidence in eastern Michigan, and restriction from the Ellsworth Shale. A depositional hiatus occurred at the end of the

Upper Antrim.

Uplift of the Acadian Mountains from the Acadian orogenic event and a regression initiated the deposition of the clastic dominated Bedford-Berea sequence sourced from the east. The Bedford prodelta has been noted to inter-tongue with the

109 Ellsworth suggesting that the two formations are time equivalent (Asseez, 1969). The

interpretations presented in this study, and in accordance with the conclusions of

Matthews (1989), suggest the Upper Antrim was deposited after or towards the end of

Ellsworth time and before the Bedford making the Ellsworth older than the Bedford-

Berea sequence. The Bedford has a sharp contact with the underlying Upper Antrim and

gradually coarsens up into the Berea Sandstone. The contact between the Bedford and

Berea is difficult to pick without modern logs. The major channels associated with the

Berea delta plain incise down into the Bedford Shale. The Berea sandstone contains the

spore Retispora lepidophyta which is a global indicator of the latest Devonian (Gutschick and Sandberg, 1991).

The Sunbury marks the beginning of the Mississippian and is associated with a major eustatic rise in the Lower crenulata Zone (Gutschick and Sandberg, 1991). The distribution of the Sunbury Shale could be a result of accommodation space due to the eastern tilt of the basin (Howell and van der Pluijm, 1996) and/or rapid subsidence due to flexure and deltaic sediment load (Gutschick and Sandberg, 1991). The pre-Coldwater erosional event could explain the truncation to the west (Matthews, 1989).

Ellsworth Petroleum System

The gas produced from the Ellsworth Shale is a thermogenic gas primarily sourced from the Antrim Shale. The process responsible for the presence of gas in the

Ellsworth shallow reservoirs can be attributed to a combination of Antrim hydrocarbon migration from deeper in the basin, and a “forced maturation” mechanism which

110 increased the source rock maturity through contact with hydrothermal fluids transported through deep fault and fracture networks.

The reservoir is composed of gray, very-fine, well-sorted, subrounded, siltstone stringers sourced from the west. The Berea Sandstone was deposited after the Ellsworth

Shale eliminating any association with the Ellsworth reservoir.

The Ellsworth green-gray shale above the silt stringers is the primary seal. The trapping mechanism is most likely a combination of structure and stratigraphy.

FUTURE WORK

The Upper Devonian-Lower Mississippian should be analyzed in greater detail to better understand the complex relationship between the Lachine, Ellsworth, and Upper

Antrim. X-ray diffraction studies have not been done on the Antrim since the Antrim was divided into the four members by Gutschick and Sandberg (1991). A more detailed look into the mineralogy would allow for a better distinction between the different members. It is still unclear when Lachine deposition starts and Upper Antrim begins.

The Devonian-Mississippian boundary in the western part of the basin is still unclear. The pre-Coldwater unconformity and absence of the Sunbury Shale makes it difficult to distinguish the Coldwater from the Ellsworth. Asseez (1969) notes carbonate beds at the top of the Ellsworth but it is also possible the carbonates are associated with the Coldwater “Red Rock”.

The material presented in this report pertaining to the Ellsworth Petroleum

System primarily relied upon public data. This subject could be analyzed in far greater

111 detail. More Rock Eval and pyrolysis data in the Antrim Shale in the western basin in and around the Ellsworth fields would help prove or disprove the “forced maturation” theory. Seismic data around the fields would also benefit the understanding of the mechanisms responsible for hydrocarbon generation. Mapping the silt reservoirs would not only increase the understanding of the trapping mechanism but would also benefit future exploration attempts.

112 REFERENCES

Addison, C. C., 1940, Buckeye Oil Field, Gladwin County, Michigan: Bulletin of the American Association of Petroleum Geologists, v. 24, no. 11, p. 1950–1982. Algeo, T. J., and J. B. Maynard, 2004, Trace-element behavior and redox facies in core shales of Upper -type cyclothems: Chemical Geology, v. 206, no. 3-4, p. 289–318, doi:10.1016/j.chemgeo.2003.12.009. Algeo, T. J., and N. Tribovillard, 2009, Environmental analysis of paleoceanographic systems based on molybdenum-uranium covariation: Chemical Geology, v. 268, no. 3-4, p. 211–225. Asseez, O. L., 1969, Paleogeography of Lower Mississippian Rocks of Michigan Basin: Bulletin of the American Association of Petroleum Geologists, v. 53, no. 1, p. 127– 135. Asseez, O. L., 1967, Stratigraphy and Paleogeography of the Lower Mississippian Sediments of the Michigan Basin: unpublished PhD dissertation, Michigan State University. Balthazor, D. A., 1991, Sedimentology of the Bedford-Berea Sequence (Early Mississippian), Williams Field, Michigan: unpublished MS thesis, Western Michigan University. Barnes, D.A., Parris, T.M., and Grammer, M.G., 2008, Hydrothermal Dolomitization of Fluid Reservoirs in the Michigan Basin, USA: American Association of Petroleum Geologist Annual Convention, Search and Doscovery Article 50087. Bernard, B. B., J. M. Brooks, and W. M. Sackett, 1978, Light Hydrocarbons in Recent Texas Continental Shelf and Slope Sediments: Journal of Geophysical Research, v. 83, no. C8, p. 4053–4061. Bishop, M. S., 1940, Isopachous Studies of Ellsworth to Traverse Limestone Section of Southwestern Michigan: Bulletin of the American Association of Petroleum Geologists, v. 24, no. 12, p. 2150–2162. Blood, R., G. Lash, and L. Bridges, 2013, Biogenic Silica in the Devonian Shale Succession of the Appalachian Basin, USA. American Association of Petroleum Geologists Search and Discovery Article 50864. Brown, A., 2011, Identification of source carbon for microbial methane in unconventional gas reservoirs: Bulletin of the American Association of Petroleum Geologists, v. 95, no. 8, p. 1321–1338. Brown, T. C., and F. Kenig, 2004, Water column structure during deposition of Middle Devonian-Lower Mississippian black and green/gray shales of the Illinois and Michigan Basins: A biomarker approach: Palaeogeography, Palaeoclimatology, 113 Palaeoecology, v. 215, p. 59–85. Budai, J. M., A. M. Martini, L. M. Walter, and T. C. W. Ku, 2002, Fracture-fill calcite as a record of microbial methanogenesis and fluid migration: a case study from the Devonian Antrim Shale, Michigan Basin: Geofluids, v. 2, no. 3, p. 163–183. Carlson, C. G., 1927, Geology of the Saginaw Oil Field, Michigan, and Discussion of Michigan’s Oil Prospects: Bulletin of the American Association of Petroleum Geologists, v. 11, p. 959-965. Cercone, K. R., 1984, Thermal History of Michigan Basin: Bulletin of the American Association of Petroleum Geologists, v. 68, no. 2, p. 130–136. Cercone, K. R., and H. N. Pollack, 1991, Thermal maturity of the Michigan Basin, Catacosinos, P.A., Daniels, P.A., Early Sedimentary Evolution of the Michigan Basin: Geological Society of America, no. Special Paper 256, p. 1-11. Charpentier, R. R., 1988, A summary of petroleum plays and characteristics of the Michigan Basin: United States Geological Survey Open file Report 87-450-R. Cliffs Minerals, I., 1982, Analysis of the Devonian Shales in the Michigan Basin. Cliffs Minerals, I., 1981a, Corrected Phase II Report Preliminary Labratory Results: Michigan #1 Well, Otsego County. Cliffs Minerals, I., 1981b, Phase II Report Preliminary Labratory Results: Dow Chemical Cores #100, #103, and #205 Sanilac County Michigan. Cloetingh, G., 1992, Intraplate Stress and Sedimentary Basin Evolution, Brown, G., Hawkesworth, C., Wilson, C., Understanding the Earth: Department of Earth Sciences, Cambridge University, p. 426. Coburn, T. C., P. A. Freeman, and E. D. Attanasi, 2012, Empirical Methods for Detecting Regional Trends and Other Spatial Expressions in Antrim Shale Gas Productivity, with Implications for Improving Resource Projections Using Local Nonparametric Estimation Techniques: Natural Resources Research, v. 21, no. 1, p. 1–21. Cohee, G. V, 1979, Paleotectonic Investigations of the Mississippian Systems in the United States, Part I: Introduction and Regional Analysis of the Mississippian System: Geological Survey, no. 1010-D. Cole, G. A., R. J. Drozd, R. A. Sedivy, and H. I. Halpern, 1987, Organic Geochemistry and Oil-Source Correlations, Paleozoic of Ohio: Bulletin of the American Association of Petroleum Geologists, v. 71, no. 7, p. 788–809. Cooper, J. R., 1943, Flow Structures in the Berea Sandstone and Bedford Shale of Central Ohio: The Journal of Geology, v. 51, no. 3, p. 190–203. Cox, K., 2015, Anomalous Thermal Indicators from Authigenic Minerals in Upper Paleozoic Strata of the Michigan Basin: unpublished MS thesis, Western Michigan 114 University. Curtis, J. B., 2002, Fractured shale-gas systems: Bulletin of the American Association of Petroleum Geologists, v. 86, no. 11, p. 1921–1938. Decker, D. A., 1992, Log-based gas content and resource estimates for the Antrim Shale, Michigan Basin: Gas Shales Technology Review, v. 8, no. 1, p. 1–10. Decker, D. A., J.-M. P. Coates, and D. Wicks, 1992, Stratigraphy, Gas Occurrence, Formation Evaluation and Fracture Characterization of the Antrim Shale, Michigan Basin: Lakewood, CO. Decker, A. D., D. G. Hill, and D. E. Wicks, 1993, Log-Based Gas Content and Resource Estimates for the Antrim Shale, Michigan Basin: Society of Petroleum Engineers, no. 25910, p. 659–669. Dellapenna, T. M., 1991, Sedimentological, Structural, and Organic Geochemical controls on Natural Gas occurrence in the Antrim Formation in Otsego County, Michigan: unbublished PhD dissertation, Western Michigan University. Dellapenna, T. M., and W. B. Harrison III, 1993, Mineralogic, Petrophysical and Organic Geochemical Properties Antrim Shale Michigan Basin: unpublished MGRRE core workshop. Dolton, B. G. L., 1996, Michigan Basin Province ( 063 ): Kansas Geological Society, no. 063, p. 1–32. Drake, L., 2014, XRF User Guide, XRF Guru. Dyman, T. S., 1981, Eastern Gas Shales Project (EGSP) data files: A final report. Elam, T. D., 1981, Stratigraphy and Paleoenvironmental aspects of the Bedford-Berea sequence and the Sunbury Shale in Eastern and South-Central : The University of Kentucky. Ells, G. D., 1958, Notes on the Devonian-Silurian in the Subsurface of Southwest Michigan: Michigan Geological Survey Progress Report, v. 18, 55p. Ells, G. D., 1978, An Appraisal of known Antrim Shale and Berea Oil and Gas Pools: DOE Report EGS-88, p. 281-290. Ells, G. D., 1979, Stratigraphic Cross Sections extending from Devonian Antrim Shale to Mississippian Sunbury Shale in the Michigan Basin: Michigan Geological Survey Progress Report, v. 22, 27p. Ettensohn, F. R., 1985a, Controls on development of Catskill Delta complex basin-facies: Geological Society of America Special Paper, v. 201, p. 65–77. Ettensohn, F. R., 2004, Modeling the nature and development of major paleozoic clastic wedges in the Appalachian Basin, USA: Journal of Geodynamics, v. 37, no. 3-5, p.

115 657–681. Ettensohn, F. R., and T. D. Elam, 1985, Defining the nature and location of a Late Devonian-Early Mississippian pycnocline in eastern Kentucky.: Geological Society of America Bulletin, v. 96, no. 10, p. 1313–1321. Ettensohn, F. R., and T. R. Lierman, 2012, Large-scale Tectonic Controls on the Origin of Paleozoic Dark-shale Source-rock Basins: Examples from the Appalachian Foreland Basin, Eastern United States: Bulletin of the American Association of Petroleum Geologists, v. 100, p. 95–124. Everham, W. D., 2004, Thermal history of Michigan Basin: unpublished PhD dissertation, Michigan Technological University. Fisher, J.H., 1980, Stratigraphy of the Upper Devonian-Lower Mississippian of Michigan, Report FE-2346-30, p. 21. Fisher, J., M. Barratt, J. Droste, and R. Shaver, 1988, Michigan Basin: The Geological Society of America, v. D-2, no. 13, p. 361–382. Floyd, J., 2015, Subsurface Geological analyses of the Berea Petroleum System in Eastern Kentucky: unpublished MS thesis, University of Kentucky. Formolo, M. J., N. Riedinger, and B. C. Gill, 2014, Geochemical evidence for euxinia during the Late Devonian extinction events in the Michigan Basin (U.S.A.): Palaeogeography, Palaeoclimatology, Palaeoecology, v. 414, p. 146–154. Girard, J.-P., and D. A. Barnes, 1995, Illitization and paleothermal regimes in the Middle Ordovician St Peter sandstone, Basin: K-Ar, oxygen isotope, and fluid inclusion data: Bulletin of the American Association of Petroleum Geologists, v. 79, no. 1, p. 49–69. Gunn, G. R., 1988, Stratigraphy, reservoir characteristics of Berea sandstone in Michigan fields: Oil and Gas, v. 86, no. 10, p. 57–60. Gutschick, R. C., and C. A. Sandberg, 1991a, Late Devonian history of Michigan Basin, Catacosino, P.A., Daniels, P.A., Early sedimentary evolution of the Michigan Basin: Geological Society of America, Special Paper no. 256, p. 181–202. Gutschick, R. C., and C. A. Sandberg, 1991b, Upper Devonian biostratigraphy of Michigan Basin, Catacosino, P.A., Daniels, P.A., Early sedimentary evolution of the Michigan Basin: Geological Society of America, Special Paper no. 256, p. 155–179. Hake, B. F., 1938, Geologic Occurrence of Oil and Gas in Michigan: Bulletin of the American Association of Petroleum Geologists, v. 22, no. 4, p. 393–415. Hale, L., 1941, Study of Sedimentation and Stratigraphy of Lower Mississippian in Western Michigan: Bulletin of the American Association of Petroleum Geologists, v. 25, no. 4, p. 713–723.

116 Harrell, J., C. Hatfield, and G. R. Gunn, 1991, Mississippian System of the Michigan Basin; Stratigraphy, sedimentology, and economic geology: Geological Society of America, no. 256, p. 203–219. Hathon, C. P., 1979, The Origin of the Quartz in the Antrim Shale: unpublished MS thesis, Michigan State University. Howell, P. D., and B. A. van der Pluijm, 1999, Structural sequences and styles of subsidence in the Michigan basin: Bulletin of the Geological Society of America, v. 111, no. 7, p. 974–991. Jackson, D. S., 1985, The Petrology, Porosity and Permeability of the Berea Sandstone (Mississippian) Perry Township, Ashland County, Ohio: University of . Jin, H., S. A. Sonnenberg, and J. F. Sarg, 2015, Source Rock Potential and Sequence Stratigraphy of Bakken Shales in the Williston Basin: The Society of Petroleum Engineers, no. 178657. Kelley, D. R., 1981, A Summary of Major Geophysical Logging Methods: Harrisburg. Leddy, D. G., V. R. Sandel, G. L. Swartz, D. H. Kenny, W. M. Gulick, and H. S. El Khadem, 1980, Chemical Composition of Antrim Shale in the Michigan Basin: Report FE-2346-89, 231 p. Leinfelder, R., 1997, Coral Reefs and Carbonate Platforms within a Siliciclastic Setting. General Aspects and Examples from the Late of Portugal: Institute of Geology and Paleontology, v. 2, p. 1737–1742. LeMone, D. V, 1964, The Upper Devonian and Lower Mississippian Sediments of the Michigan Basin and Bay County, Michigan: unpublished PhD dissertation, Michigan State University. Lilienthal, R. T., 1974, Subsurface Geology of Barry County, Michigan: Report of Investigation 15, Geological Survey Division, Michigan Department of Natural Resources, p. 1-36. Luczaj, J. A., W. B. Harrison, and N. S. Williams, 2006, Fractured hydrothermal dolomite reservoirs in the Devonian Dundee Formation of the central Michigan Basin: Bulletin of the American Association of Petroleum Geologists, v. 90, no. 11, p. 1787–1801. Ma, L., M. C. Castro, and C. M. Hall, 2009, Crustal noble gases in deep brines as natural tracers of vertical transport processes in the Michigan Basin: Geochemistry, Geophysics, Geosystems, v. 10, no. 6, p. 1–24, doi:10.1029/2009GC002475. Martini, A. M., L. M. Walter, J. M. Budat, T. C. W. Ku, C. J. Kaiser, and M. Schoell, 1998, Genetic and temporal relations between formation waters and biogenic methane: Upper Devonian Antrim shale, Michigan Basin, USA: Geochimica et Cosmochimica Acta, v. 62, no. 10, p. 1699–1720, doi:10.1016/S0016- 7037(98)00090-8. 117

Martini, A. M., L. M. Walter, T. C. W. Ku, J. M. Budai, J. C. McIntosh, and M. Schoell, 2003, Microbial production and modification of gases in sedimentary basins: A geochemical case study from a Devonian shale gas play, Michigan basin: Bulletin of the American Association of Petroleum Geologists, v. 87, no. 8, p. 1355–1375, doi:10.1306/031903200184. Matthews, D., 1989, Review and Revision of the Devonian-Mississippian Stratigraphy in the Michigan Basin: U.S. Geological Survey Bulletin, p. 101–189. Maynard, J. B., 1981, Carbon Isotopes as indicators of dispersal patterns in Devonian- Mississipian shales of the Appalachian Basin: Geology, v. 9, no. June, p. 262–265. McCarthy, K., K. Rojas, M. Niemann, D. Palmowski, K. Peters, and A. Stankiewicz, 2011, Basic Petroleum Geochemistry for Source Rock Evaluation: Oilfield Review, v. 23, no. 2, p. 32–43. Moyer, R. B., 1982, Thermal Maturity and Organic Content of selected Paleozoic Formations-Michigan Basin: unpublished MS thesis, Michigan State University. Newcombe, R. B., 1931, Oil and Gas Fields of Michigan: A discussion of depositional and structural features of the Michigan Basin: unpublished PhD dissertation, University of Michigan. Over, D. J., 2007, Conodont Biostratigraphy of the Chattanooga Shale, Middle and Upper Devonian, Southern Appalachian Basin, Eastern United States: Journal of Paleontology, v. 81, no. 6, p. 1194–1217. Over, D. J., R. Lazar, G. C. Baird, J. Schieber, and F. R. Ettensohn, 2009, Protosalvinia Dawson and Associated Conodonts of the Upper Trachytera Zone, Famennian, Upper Devonian, in the Eastern United States: Journal of Paleontology, v. 83, no. 1, p. 70–79. Pashin, J. C., and F. R. Ettensohn, 1992, Paleoecology and sedimentology of the dysaerobic Bedford fauna (Late Devonian), Ohio and Kentucky (USA): Palaeogeography, Palaeoclimatology, Palaeoecology, v. 91, p. 21–34. Pashin, J. C., and F. R. Ettensohn, 1995, Reevaluation of the Bedford-Berea Sequence in Ohio and Adjacent States: Forced Regression in a Foreland Basin, : Geological Society of America, Special Paper no. 298, p. 1–69. Pepper, J. F., W. De Witt, and D. F. Demarest, 1954, Geology of the Bedford Shale and Berea Sandstone in the Appalachian Basin: United States Geological Survey, Professional Paper no. 259, 111 p. Poole, F. G., and C. A. Sandberg, 1977, Mississippian Paleogeography and Tectonics of the Western United States, Stewart, J.H., Stevens, C.H., Gritsche, A.E., Paleozoic Paleogeography of the Western United States: Pacific Section, Society of Economic Paleontologists and Mineralogists Pacific Coast Paleogeography Symposium 1, p. 67–85.

118

Potter, P. E., J. B. Maynard, and P. J. Depetris, 2005, Mud and : Introduction and Overview: Verlag Berlin Heidelberg, Springer. Prasad, M., and M. H. Manghnani, 1997, Effects of pore and differential pressure on compressional wave velocity and quality factor in Berea and Michigan sandstones: Geophysics, v. 62, no. 4, p. 1163–1176. Riegl, B. M., and W. E. Piller, 2000, Reefs and coral carpets in the northern Red Sea as models for organism-environment feedback in coral communities and its reflection in growth fabrics: Journal of the Geological Society of London, v. 178, no. 1, p. 71– 88. Rowe, H., N. Hughes, and K. Robinson, 2012, The quantification and application of handheld energy-dispersive x-ray fluorescence (ED-XRF) in mudrock chemostratigraphy and geochemistry: Chemical Geology, v. 324-325, p. 122–131. Rullkotter, J., R. Marzi, and P. Meyers, 1992, Biological Markers in Paleozoic Sedimentary Rocks and Crude Oils from the Michigan Basin: Reassessment of Sources and Thermal History of Organic Matter, in Early Organic Evolution: Implications for Mineral and Energy Resources: Verlag Berlin Heidelberg, Springer, p. 324–335. Ruotsala, A.P., 1980, Mineralogy of Antrim Shale, Michigan: Report FE-2346-79, 85p. Ryder, R. T., 1996, Fracture patterns and their origin in the Upper Devonian Antrim Shale gas reservoir of the Michigan Basin: A review: Reston, Virginia. Sandberg, C.A., Ziegler, W., Dreesen, R., and Butler, J.L., 1988, Late Frasnian mass extinction; Conodont event stratigraphy, global changes, and possible causes: Courier Forschungsinstiut Senckenberg, v. 102, p. 263-307. Science Applications, 1979, Final Report on Assessment of Michigan Basin Shale Characterization Research and Results: Morgantown, . Shugar, N. A., and L. J. Mass, 2012, Introduction, in N. A. Shugar, and L. J. Mass, eds., Studies in Archaeological Sciences: Handheld XRF for Art and Archaeology: Leuven (Belgium), Leuven University Press, p. 17–37. Sloss, L. L., 1963, Sequences in the Cratonic Interior of North America: The Geological Society of America, v. 74, p. 93–114. Swezey, B. C. S., J. R. Hatch, J. A. East, D. O. Hayba, and J. E. Repetski, 2015, Total Petroleum Systems of the Michigan Basin — Petroleum Geology and Geochemistry and Assessment of Undiscovered Resources, in Geologic Assessment of Undiscovered Oil and Gas Resources of the U.S. Portion of the Michigan Basin: Reston, Virginia, USGS. Tarbell, E., 1941, Antrim-Ellsworth-Coldwater Shale Formations in Michigan: American Association of Petroleum Geologists, v. 25, no. 4, p. 724–733.

119 Wagenvelt, K. A., 2015, Thermal History of the Michigan Basin: Results from Thermal Maturation Data and Geodynamic Modelling: unpublished MS thesis, Western Michigan University. Wang, H. F., K. D. Crowley, and G. C. Nadon, 1994, Thermal History of the Michigan Basin from Apatite Fission-Track Analysis and Vitrinite Reflectance, in American Association of Petroleum Geologists Memoir 61; Basin Compartments and Seals: p. 167–177. Wells, N. A., A. H. Coogan, and J. J. Majoras, 1991, Field Guide to Berea Sandstone Outcrops in the Black River Valley at Elyria, Ohio: Slumps, Slides, Mud Diapirs, and Associated Fracturing in Mississippian Delta Deposits: The Ohio Journal of Science, v. 91, no. 1, p. 35–48. Wen, T., M. C. Castro, B. R. Ellis, C. M. Hall, and K. C. Lohmann, 2015, Assessing compositional variability and migration of natural gas in the Antrim Shale in the Michigan Basin using noble gas geochemistry: Chemical Geology, v. 417, p. 356– 370. Woodrow, D. L., F. W. Fletcher, and W. F. Ahrnsbrak, 1973, Paleogeography and Paleoclimate at the Deposition Sites of the Devonian Catskill and Old Red Facies: Geological Society of America Bulletin, v. 84, p. 3051–3064. Young, D. C., 1980, An investigation of the feasibility of recovering energy from Antrim oil shale by an in situ process, Final summary report: Report FE-2346-94, 25 p. and appendix.

120 APPENDIX A

Cross Sections

Figure A.1 Cross section network used in this study to determine the regional stratigraphy of the Upper Devonian-Lower Mississippian in the Michigan Basin.

121

The tops for each well were picked based upon the variation in GR, RHOB, NPHI, and PEF. GR is in the first track (left) for each well and is scaled from 0-450 API. The brighter the yellow the lower the API and the darker the shading the higher the API. RHOB, NPHI, and PEF are all included on track two (right) for each well. RHOB is scaled from 2-3 (g/cm3) and is the green curve. NPHI is scaled from 45% to -15% from left to right and is the blue curve. The shading in track two is the crossover between NPHI and RHOB. The shaded area increases as NPHI and RHOB increase. The dotted red line is PEF.

Figure A.2 Cross section A-A’. Northern most cross section running EW. Westward pinchout of the Upper Antrim, Bedford, Berea, and Sunbury. The sharp Upper Antrim Bedford contact noted in the east diminishes to the west. The low GR spike in the Lachine Member in the eastern most well (PN 48538) is most likely attributed to a calcareous commonly noted in the Antrim Shale (Cohee, 1979).

Figure A.3 Cross section B-B’. Northern EW cross section demonstrating the pinchout of the Upper Antrim, Bedford, Berea, and possibly the Sunbury moving towards the western margin. The contact between the Lachine and Ellsworth is much more gradational towards the west (PN 55068 and 40210) and the contact becomes less clear.

122 Figure A.4 Cross section C-C’. The central EW cross section which best represents the regional stratigraphic relationships between the Upper Devonian-Lower Mississippian formations. The Sunbury Shale is thickest towards the eastern margin in Huron and Sanilac Counties. The Berea Sandstone thickens in areas the distributary channels incised down into the Bedford Shale (PN 36363 and 38565). There is a pinchout of the Upper Antrim and Berea sandstone to the west. Although this cross section displays the Bedford extending to the western margin, the lithology noted by the substantial drop in GR in the western most well (PN 58249) is not Bedford-like lithology. The Lauber 1A-6 (PN 39073) is just north of well 58249 in the NW corner of Oceana County and was cored in the low GR interval ( ̴750-780 ft.). The fabric is primarily a partially dolomitized oolitic packstone common in shoal environments. Asseez (1969) made note of similar beds suggesting the dolomite and limestone beds are part of the upper Ellsworth and represent the contact between the delta front and prodelta similar to the eastern Bedford and Berea contact. The carbonate body could also be part of the overlying Coldwater Shale.

Figure A.5 Cross section D-D’. Southern EW cross section demonstrating another example of the Berea Sandstone incising into the Bedford. The Bedford Shale has a thinning and apparent thickening towards the western basin margin. The distinction between the Norwood and Lachine becomes difficult when the Paxton is thin or not present. The Upper Antrim, Berea, and Sunbury pinch out to the west.

123

Figure A.6 Cross section E-E’. The southernmost EW cross-section. The southern depocenters of the Paxton Member is demonstrated towards the middle of the cross-section. The pinchout of the Paxton to the west makes it difficult to decipher the Norwood from the Lachine. There is a significant drop in gamma-ray at the top of the Ellsworth which is most likely associated with the Coldwater “Red Rock”.

Figure A.7 Cross section F-F’. The westernmost NS cross-section. The Ellsworth Shale makes up a majority of the rock body. The gradational transition from the Lachine to the Ellsworth is consistent throughout the cross-section. The Upper Antrim, Bedford, and Sunbury Shales pinches out in the north and south.

124

Figure A.8 Cross section G-G’. The central NS cross-section demonstrating a significant thinning towards the south. This can also been seen in the Upper Devonian-Lower Mississippian isopach map (Figure 1.6). The north and south depocenters for the Paxton are seen clearly in this cross section.

Figure A.9 Cross section H-H’. The central NS cross-section similar to GG’, with the exception of the Bedford-Berea sequence getting thicker towards the eastern source of sediment. The thinning of the Paxton is demonstrated in the middle of the cross-section.

125 Figure A.10 Cross section I-I’. Easterly NS cross-section. The Berea Sandstone pinches out to the north and demonstrates another example of a major channel removing sediment from the underlying Bedford Shale.

Figure A.11 Cross section J-J’. Easternmost NS cross-section. This cross-section demonstrates a fairly uniform distribution of sediment. The Sunbury Shale is thick relative to areas further west in the basin. The Ellsworth becomes difficult to decipher between the Lachine and Upper Antrim.

126 APPENDIX B

Cross Section Well Information

UWI/API LABE OPERATOR WELLNAME WEL TWNS RN SECT L L # H G N 21001485380000 48538 H R F EXPLORATION AND PRODUCTION TRADER D2-25 27N 5E 25 INC 21001540050000 54005 STURGEON POINT DEVELOPMENT CO DEWAR C1-6 27N 9E 6 21005387320000 38732 BRUCE ALAN ENTERPRISES INC SMITH & GERARD 1 4N 13W 15 UNIT 21005404870000 40487 ALLEXCO LYNEMA 1-23 3N 16W 23 21005408820000 40882 ALLEXCO BREWER DOCK 1-11 2N 14W 11 21005554650000 55465 WEST BAY EXPLORATION CO MCALLISTER 1-6 2N 11W 6 21005583970000 58397 NORTHSTAR ENERGY LLC JACKSON D2-1 3N 11W 1 21007285830000 28583 UNION OIL CO OF CALIFORNIA SMITH, GEORGE C 1 31N 5E 32 21007399660000 39966 BROWN H L JR SNOWPLOW 6-9 29N 5E 9 21009576680000 57668 JORDAN DEVELOPMENT CO LLC BARGY 16-14 30N 9W 14 21011366090000 36609 SUMMIT PETROLEUM CORP COLLINS 4-22 19N 3E 22 21011381230000 38123 DART OIL AND GAS CORP BILACIC 1-11 19N 6E 11 21015284730000 28473 HALLWELL INC LOPEZ, BYRDIE ET 1 3N 9W 8 AL 21015389240000 38924 A N R PRODUCTION CO OCONNOR 1-16 3N 8W 16 21015405060000 40506 KULKA AND SCHMIDT INC HERRINGTON 1-31 1N 7W 31 21015564700000 56470 WEST BAY EXPLORATION CO CARTER FAMILY 1-3 2N 10W 3 L.L.C. 21017343890000 34389 ELEK AARON III ESTATE SCHWAB 2-28 15N 4E 28 21017348860000 34886 STOCKER AND SITLER OIL CO PFUND 1 14N 3E 25 21017363290000 36329 SUMMIT ESSEXVILLE INC WAZBINSKI 1-17 14N 6E 17 21017385650000 38565 SIBLEY LEWIS MEYER 1-21 13N 6E 21

127 UWI/API LABE OPERATOR WELLNAME WEL TWNS RN SECT L L # H G N 21017475300000 47530 FAIR ENERGIES INC KINNANE 1-7 14N 5E 7 21019556220000 55622 PRESIDIUM ENERGY LC STATE INLAND C1-33 26N 13W 33 21023346880000 34688 NATURAL RESOURCE MANAGEMENT RCOPELAND 1-1 5S 6W 1 CORP 21025454640000 45464 IN AMERICA PETROLEUM PROPERTIES BUCK 1-26 3S 7W 26 21035359630000 35963 MUSKEGON DEVELOPMENT CO SINCIC 1-36 20N 3W 36 21035376590000 37659 B W A B INC GARDNER 18-32 17N 4W 18 21035409840000 40984 H AND K RESOURCES INC DAVIS 1-12 19N 5W 12 21035409870000 40987 DART OIL AND GAS CORP STATE 1-31 20N 6W 31 WINTERFIELD 21035601340000 60134 CONTINENTAL RESOURCES INC KROCKER 1-17 18N 4W 17 21037223480000 22348 LAKELAND OIL CORP ZISCHKE 1 5N 3W 14 21037278110000 27811 MCCLURE OIL CO FOX 1 7N 1W 6 21037353830000 35383 PENINSULAR OIL AND GAS CO FARLEY 1-10 7N 4W 10 21039357260000 35726 MILLER BROTHERS STATE LOVELLS 1-33 27N 1W 33 21039359170000 35917 MILLER BROTHERS STATE GRAYLING 1-33 26N 3W 33 21045397080000 39708 C M S OIL AND GAS CO KEELEY 1-11 2N 4W 11 21045414230000 41423 B P AMERICA PRODUCTION CO ARNOLD UNIT 1-8 4N 5W 8 21045445050000 44505 COBRA OIL AND GAS CORP RIPLEY 1-16 2N 6W 16 21045488690000 48869 B P AMERICA PRODUCTION CO PARSONS C4-11 3N 3W 11 21049351720000 35172 CRYSTAL EXPLORATION CO STORY 1-3 9N 8E 3 21049359990000 35999 TRAVERSE OIL CO IUNG 1-7 8N 6E 7 21049487460000 48746 B P AMERICA PRODUCTION CO SWANEBECK D4-34 6N 6E 34 21051358460000 35846 MUSKEGON DEVELOPMENT CO AND ODULL 3-6 17N 2W 6 SUMMIT PETR 21051386480000 38648 ORYX ENERGY CO STATE SHERIDAN 1-15 19N 2E 15 21051389800000 38980 OXY USA INC STATE BUTMAN 1 20N 1W 36 21051586760000 58676 DART OIL AND GAS CORP BOREN 1-25 17N 2W 25 128 UWI/API LABE OPERATOR WELLNAME WEL TWNS RN SECT L L # H G N 21055409010000 40901 NORTHERN PROCESSORS INC AND STATE 1-13 27N 9W 13 ORYX ENERGY WHITEWATER 21055590190000 59019 TRENDWELL ENERGY CORP AMOS C1-4 28N 9W 4 21057242700000 24270 STRAKE G W DARVISHIAN, 1 11N 3W 17 HARRY 21057297390000 29739 MCCLURE OIL CO McClure SPARKS 1-8 10N 2W 8 21057333820000 33382 HUNT ENERGY CORP FABUS, JOHN 1-22 9N 2W 22 21057373770000 37377 JENNINGS PETROLEUM CORP SMITH 1-17 10N 4W 17 21057399910000 39991 UNION OIL CO OF CALIFORNIA WILES 1-34 10N 4W 34 21057409940000 40994 BAUMANN RESOURCES INC FROST 1-1 12N 1W 1 21057435230000 43523 COASTAL OIL AND GAS CORP SHERWOOD 1-27 12N 2W 27 21057605040000 60504 SILVER OAK ENERGY LLC SKARYD 9N 1W 23 21059396310000 39631 PATRICK PETROLEUM CORPORATION AWEIDNER 1-1 5S 2W 1 OF MICHIG 21059461290000 46129 TELESIS INC ZAIOCZKOWSKI 1-2 8S 1W 2 21063299260000 29926 EXXONMOBIL OIL CORP REIBLING, ARNOLD 1 17N 11E 35 UNIT 21063335500000 33550 SOUTHEASTERN EXPLORATION CO CLANCY, FELIX & 2-12 18N 12E 12 DORIS 21063408490000 40849 SWEPI LP ELENBAUM & 1-5 15N 10E 5 GETTEL ET AL 21065489920000 48992 B P AMERICA PRODUCTION CO BOTSFORD A2-8 4N 2E 8 21065498250000 49825 B P AMERICA PRODUCTION CO STONE D1-28 3N 1W 28 21067235740000 23574 MCCLURE OIL CO WILDMAN, VIRGIL 1 5N 7W 15 21067246190000 24619 AMBASSADOR OIL CORP TEN CATE, SAMUEL 1 7N 8W 34 ET AL 21067253750100 27271 SIBLEY DRILLING CO ALBRIGHT, EDWIN 1 7N 5W 8 & EVA 21067364350000 36435 PENINSULAR OIL AND GAS CO STOWE & LUMBERT 1-11 8N 6W 11

129

UWI/API LABE OPERATOR WELLNAME WEL TWNS RN SECT L L # H G N 21067367040000 36704 PETROLEUM CORP OF TEXAS FERRIS & STATE 1-28 6N 6W 28 ORANGE ET AL 21067416560000 41656 B P AMERICA PRODUCTION CO BEHRENWALD 1-34 6N 8W 34 UNIT 21069426380000 42638 HOWARD ENERGY CO INC STATE RENO 1-33 22N 5E 33 21069499280000 49928 MACK OIL CORP STATE OSCODA 1A-32 24N 6E 32 21073242560000 24256 BAUER BROTHERS RICHEY, BURDON 1 13N 5W 27 ET AL COMM 21073341660000 34166 MUSKEGON DEVELOPMENT CO POHL 1-36 15N 5W 36 21073357110000 35711 TOLAS OIL AND GAS EXPLORATION CO BURROWS 1 15N 3W 32 21073372430000 37243 PETRO HUNT INC OF KENTUCKY BEUTLER CARL & 1-34 15N 6W 34 HOWARD ET AL 21073379090000 37909 KNIGHT ROYALTY CORP MCDONALD 1-16 14N 5W 16 21075389460000 38946 DOMINION MIDWEST ENERGY INC BENN ET AL 1-17 2S 2W 17 21075392210000 39221 B W A B INC BAUM 24-43 1S 3W 24 21079402620000 40262 SWEPI LP STATE BLUE LAKE 1-24 28N 5W 24 21079412510000 41251 E P S RESOURCES CORP INGERSOLL 31-31 25N 8W 31 21079449670000 44967 MILLER OIL CORP STATE BEAR LAKE 1-9 26N 5W 9 21079459650000 45965 FAIRWAY LTD INC STATE GARFIELD 2-19 25N 5W 19 21079462800000 46280 DON YOHE ENTERPRISES INC LEE 1-20 28N 6W 20 21081246270000 24627 AMBASSADOR OIL CORP FRANCISCO, 1 8N 9W 35 GEORGE 21081248260000 24826 AMBASSADOR OIL CORP TEN, JOHN & HAVE 1 8N 9W 6 21081391620000 39162 PENINSULAR OIL AND GAS CO CARLSON ET AL 1-7 9N 9W 7 21081405530000 40553 WEST HOPKINS PETROLEUM CO LP REUHS 1-19 5N 10W 19 21081420890000 42089 UNION OIL CO OF CALIFORNIA TUINSTRA 1-13 5N 11W 13 21081488650000 48865 B P AMERICA PRODUCTION CO VANDENBERG B1-17 7N 9W 17 21081554770000 55477 WEST BAY EXPLORATION CO SWITZER 1-13 9N 11W 13

130

UWI/API LABE OPERATOR WELLNAME WEL TWNS RN SECT L L # H G N 21085268500000 26850 COOK BROTHERS AND C M S OIL AND STATE PEACOCK 2 19N 13W 9 GAS CO 21085560110000 56011 SPRUCE ENERGY LLC STATE YATES 1-27 17N 12W 27 21087239460000 23946 GOOD AND GOOD DRILLING THOM, IRVING GG-1 8N 9E 10 21093351990000 35199 DART OIL AND GAS CORP AND CLAVEY & STATE 1-7 1N 5E 7 MICHIGAN OIL HAMBURG 21093361320000 36132 DART OIL AND GAS CORP AND CLAVEY & STATE 1-6 1N 5E 6 MICHIGAN OIL HAM 21093378930100 37893 DON YOHE ENTERPRISES INC LAIER 1-23 4N 5E 23 21093454920000 45492 MERCURY EXPLORATION CO DEY A1-15 3N 3E 15 21093487480000 48748 B P AMERICA PRODUCTION CO GRAMER D1-11 4N 5E 11 21099520640000 52064 WEST BAY EXPLORATION CO BEREGASSI 1-5 2N 12E 5 21101389780000 38978 OMIMEX ENERGY INC LAKELAND 11-32 21N 17W 33 ASSOCIATION & STATE ET AL 21101539810000 53981 JORDAN DEVELOPMENT CO LLC LECKRONE 4-9 23N 15W 9 21105239560000 23956 MILLER BROTHERS JACOBSON UNIT 1 18N 17W 14 21105321110000 32111 CONOCO INC AND THE WISER OIL CO STATE HAMLIN 2-24 19N 18W 24 21105550680000 55068 LAVANWAY EXPLORATION CO LLC FREESOIL A3-22 20N 16W 22 21107354260000 35426 MIDWEST OIL PRODUCTIONS ROUSSEAU, 1-12 16N 8W 12 THELMA 21107377230000 37723 MERRILL DRILLING CO MAY 29-23 15N 7W 29 21107412670000 41267 COASTAL OIL AND GAS CORP ORR 1-9 13N 7W 9 21107414400000 41440 SAMSON NATURAL GAS CO AND BASSETT 1-16 16N 10W 16 MILLER ENERGY 21107422640000 42264 WOLVERINE ENVIRONMENTAL STATE COLFAX & 1-31 15N 9W 31 PRODUCTION INC KNIGHT 21111369310000 36931 ORYX ENERGY CO PLONA UNIT 1 14N 2W 13 21111413900000 41390 DART OIL AND GAS CORP WHITNEY 1-33 16N 2W 33 21111415300000 41530 PASADENA OIL AND GAS CORP BADOUR 1-31 14N 2E 31 131 UWI/API LABE OPERATOR WELLNAME WEL TWNS RN SECT L L # H G N 21113274990000 27499 MCCLURE OIL CO STATE WEST 1 23N 6W 10 BRANCH 21113321810000 32181 DART OIL AND GAS CORP TANIS, GLEN ET AL 1-11 21N 8W 11 21113370620000 37062 DART OIL AND GAS CORP EDWARDS 5-30 22N 6W 30 21117245940000 24594 HUTCHINSON MELVIN J BUCHHOLZ, 1 12N 8W 23 KENNETH 21117297900000 29790 MCCLURE OIL CO FANCETT 1-28 9N 6W 28 21117300270000 30027 MCCLURE OIL CO BETHAM & CLARK 1-26 10N 7W 26 UNIT 21117301170000 30117 MCCLURE OIL CO THORLUND FARMS 1-6 9N 7W 6 21117380810000 38081 SUMMIT PETROLEUM CORP TIESNER 1-27 11N 7W 27 21117394830000 39483 MANITOU EXPLORATION CO INC WALDRON 2-30 11N 5W 30 21117414800000 41480 CHARTREUSE EXPLORATION INC ROCHA & BUCK 1-15 10N 6W 15 21117500470000 50047 CRONUS EXPLORATION CO LLC TOW 1-3 10N 5W 3 HD-1 21119364780000 36478 A N R PRODUCTION CO DAVIS ET AL 1-19 30N 1E 19 21119502040000 50204 ATLAS GAS & OIL COMPANY LLC STATE HILLMAN 1-21 31N 3E 21 21121346330000 34633 ANCHOR OIL EXPLORATION CO INC MELTON 1 10N 16W 5 21121413970000 41397 TENNECO OIL CO HIBBS 1-22 9N 15W 22 21121432170000 43217 NORTHWESTERN NATURAL GAS EILERS 1-13 12N 18W 13 21123228490000 22849 WHITEHALL OIL CORP SHERIDAN 1 12N 14W 26 TOWNSHIP 21123266620000 26662 THUNDER HOLLOW OIL AND GAS CO THOMPSON, 1 15N 14W 20 WALTER F & ROSILEA 21123384370000 38437 PENINSULAR OIL AND GAS CO MCDONALD 1-12 11N 11W 12 21123385610100 38561 ENSOURCE INC ROSS 1-9 13N 12W 9 21123391660000 39166 SAVOY ENERGY LP ALTMAN 1-20 15N 11W 20 21123399520000 39952 DOMINION MIDWEST ENERGY INC WISE, ET AT 1-3 13N 11W 3

132

UWI/API LABE OPERATOR WELLNAME WEL TWNS RN SECT L L # H G N 21123406190000 40619 LEEDE OIL AND GAS INC STATE GARFIELD & 1-14 12N 13W 14 ANDERSON 21123411620000 41162 AMOCO PRODUCTION CO AND AMASON UNIT 1-24 11N 11W 24 PENINSULAR OIL 21123418920000 41892 DART OIL AND GAS CORP CROTON 1-30 12N 11W 30 21123462180000 46218 MERRILL ENERGY INC U S A MERRILL 1-18 15N 13W 18 21123580810000 58081 WELLMASTER PRODUCTION CO LLC WELLMASTER 1-1 14N 11W 1 21123591900000 59190 BELDEN & BLAKE CORP DBA WARD YSTATE MONROE C1-6 15N 12W 6 LAKE ENERG 21125492820000 49282 WHITING OIL AND GAS CORP INNOVATIVE LAND 22-29 3N 8E 29 DEV 21127344900000 34490 WENNER PETROLEUM CORP ADAMS 1-5 16N 17W 5 21127353930000 35393 C M S OIL AND GAS CO FEDO 1-27 15N 16W 27 21127390730000 39073 WENNER PETROLEUM CORP LAUBER, JACK 1A-6 16N 17W 6 21127392910000 39291 H AND H STAR ENERGY INC DBA NBOONE 1-10 16N 17W 10 PETROSTAR E 21127404300000 40430 SWEPI LP STATE COLFAX 1-4 16N 15W 4 21127520010000 52001 WELLMASTER PRODUCTION CO LLC ROOD 1-23 14N 16W 23 21127582490000 58249 OMIMEX ENERGY INC STATE HART & 4-30 15N 17W 30 FUEHRING 21129452540000 45254 TAURUS EXPLORATION USA INC BECK (167) 8-30 23N 3E 30 21133345360000 34536 DART OIL AND GAS CORP MCCORMICK ET AL 2-27 18N 8W 27 21133391860000 39186 HOWARD ENERGY CO INC BOYCE 1-19 20N 10W 19 21133396150200 41354 HOWARD ENERGY CO INC SHERMAN 1-20 20N 9W 21 21133405560000 40556 WHITING OIL AND GAS CORP LOWE 1-27 19N 10W 27 21133415400000 41540 WOLVERINE ENVIRONMENTAL MCCASEY 1-7 17N 9W 7 PRODUCTION INC 21133417880000 41788 LEEDE OIL AND GAS INC EISENGA 11-24 20N 8W 24 BROTHERS

133

UWI/API LABE OPERATOR WELLNAME WEL TWNS RN SECT L L # H G N 21133429450000 42945 EXXONMOBIL OIL CORP STATE HARTWICK 1-27 19N 8W 27 & YARHOUSE 21133564150000 56415 TRENDWELL ENERGY CORP EISENGA C2-17 20N 8W 17 21135473550000 47355 MARATHON OIL CO U S A & STATE BIG 1-35 26N 1E 35 CREEK 21135479200000 47920 FORCE ANTRIM DEVELOPMENT INC LE MIEUX D4-28 28N 2E 28 21135480120000 48012 SWEPI LP SHIRODA D2-22 26N 4E 22 21137474170000 47417 H R F ANTRIM LP WOLF 4-33 30N 3W 33 21137547890000 54789 BREITBURN OPERATING LIMITED WEST CORWITH A2-25 32N 3W 25 PARTNERSHIP 21139348850000 34885 CHEVRON U S A INC UMLOR ROBERT ET 1-3 8N 13W 3 AL 21139395270000 39527 C M S OIL AND GAS CO VAN RYSWYCK 1-34 8N 14W 34 21139414340000 41434 UNION OIL CO OF CALIFORNIA VANSLOOTEN 1-25 7N 15W 25 21143360150000 36015 SUMMIT PETROLEUM CORP SAINT HELEN UNIT 53 24N 1W 30 TRACT 21143374090000 37409 NEWPORT ROSEVILLE GUN 1-17 21N 1W 17 CLUB 21143391130000 39113 LADD PETROLEUM CORP STATE RICHFIELD 1-26 23N 1W 26 21143398260000 39826 BEARD OIL CO BGHL & SCHMALTZ 1-23 22N 3W 23 21145327450000 32745 STOCKER AND SITLER OIL CO WIROSTEK, ERNEST 1 9N 2E 29 21145363630000 36363 STOCKER AND SITLER OIL CO UNION TEXAS & 1-27 13N 4E 27 THOMS, C F, UNIT 21145397160000 39716 DANIEL MCGUIRE INC MURPHY 1-16 12N 3E 16 21145525080000 52508 BURLINGTON RESOURCES OIL AND RIVERCREST 13-20 10N 3E 20 GAS CO LP 21147266720000 26672 A N R PIPELINE CO KLUNG, FRANK 1 8N 13E 22 UNIT 21147389640000 38964 MILLER BROTHERS AND ATLANTIC ARCO & SENYK 1-30 6N 13E 30 RICHFIELD 21147406980000 40698 CONOCO INC LOHR 1-16 8N 15E 16 134 UWI/API LABE OPERATOR WELLNAME WEL TWNS RN SECT L L # H G N 21151240470000 24047 PHILLIPS PETROLEUM CO CLEARY, RAY J 1 14N 15E 21 21151357790000 35779 TRAVERSE OIL CO FROSTIC 1-30 11N 15E 30 21155275490000 27549 PETROLEUM PROMOTIONS INC BIRCHMEIER 1 8N 4E 12 21155279070000 27907 EXXONMOBIL OIL CORP JELINEK & FERRIS 1 5N 2E 5 UNIT 21155307270000 30727 MICHIGAN OIL CO HASSELBRING ET 1-5 6N 1E 5 AL 21155487720000 48772 B P AMERICA PRODUCTION CO CONN D2-8 6N 3E 8 21157234850000 23485 DANCEY R V DANCEY, R. V. 1 14N 9E 12 21157330640000 33064 BREHM E EDWIN WALAT FARMS, INC A-3-26 14N 7E 26 ET AL 21157353690000 35369 WIXIE AKIN DRILLING CO NIXON & NIXON 2-24 14N 7E 24 21157402840000 40284 TRUE OIL CO CLOTHIER 43-28 11N 11E 28 21157402850000 40285 TRUE OIL CO BINKOWSKI 21-11 11N 9E 11 21161599200000 59920 PAXTON RESOURCES LLC COOK 1-27 4S 5W 27 21165228900000 22890 HILLIARD B G LEESON, ETHELYN 1 22N 9W 32 & SOURS, RUBERTA 21165402100000 40210 WOLVERINE ENVIRONMENTAL PROD EDOSTAL, ET AL 1-27 21N 12W 27 AND TERRA 21165559680000 55968 MUSKEGON DEVELOPMENT CO HOESL 1-9 24N 12W 9 M5305 624- Dow/ERDA 100 9N 15E 8 771- 474 M5308 687- Dow Chemical Company Dow/ERDA 103 9N 15E 8 771- 474 M5309 686- Dow Chemical Company Dow/ERDA 104 9N 15E 8 771- 474

135 APPENDIX C

Formation Top Picks

API Permi Well Name Well Sunb Bere Bedf U Ellsw Lach Paxt Norw Trave t # ury a ord Antri orth ine on ood rse m 210014853 48538 TRADER D2- 1045 1073 1143 1232 1382 1472 1546 1575 1606 80000 25 210015400 54005 DEWAR C1-6 519 587 605 630 50000 210053873 38732 SMITH & GERARD UNIT 1 832 844 845 853 854 1375 1447 1451 1474 20000 210054048 40487 LYNEMA 1-23 612 613 614 615 616 1156 1244 1247 1256 70000 210054088 40882 BREWER DOCK 1-11 770 771 772 773 774 1278 1345 1360 1366 20000 210055546 55465 MCALLISTER 1-6 1079 1080 1081 1082 1083 1546 1614 1622 1631 50000 210055839 58397 JACKSON D2-1 1246 1247 1248 1250 1249 1683 1751 1760 1772 70000 210072858 28583 SMITH, GEORGE C 1 391 455 476 500 30000 210073996 39966 SNOWPLOW 6-9 567 636 654 685 60000 210095766 57668 BARGY 16- 653 712 740 752 80000 14 210113660 36609 COLLINS 4-22 1567 1594 1669 1727 1844 1928 2008 2044 2071 90000 210113812 38123 BILACIC 1-11 1428 1464 1579 1678 1844 1879 1950 1984 2014 30000 210152847 28473 LOPEZ, BYRDIE ET AL 1 1348 1351 1352 1402 1408 1692 1754 1760 1780 30000

136

API Permi Well Name Well Sunb Bere Bedf U Ellsw Lach Paxt Norw Trave t # ury a ord Antri orth ine on ood rse m 210153892 38924 OCONNOR 1-16 1499 1506 1507 1540 1585 1776 1834 1840 1856 40000 210154050 40506 HERRINGTON 1-31 1379 1387 1388 1431 1452 1620 1668 1685 1700 60000 210155647 56470 CARTER FAMILY L.L.C. 1-3 1358 1359 1360 1361 1362 1770 1834 1838 1852 00000 210173438 34389 SCHWAB 2-28 1571 1595 1681 1785 1934 1969 2038 2073 2101 90000 210173488 34886 PFUND 1 2183 2204 2277 2370 2480 2567 2626 2664 2687 60000 210173632 36329 WAZBINSKI 1-17 1457 1485 1582 1655 1813 1846 1914 1946 1967 90000 210173856 38565 MEYER 1-21 2031 2058 2228 2261 2384 2428 2499 2521 2539 50000 210174753 47530 KINNANE 1-7 1515 1538 1676 1725 1880 1910 1978 2017 2038 00000 210195562 55622 STATE INLAND C1- 1193 1280 1320 1340 20000 33 210233468 34688 COPELAND 1-1 1070 1075 1076 1107 1129 1217 1260 1284 1299 80000 210254546 45464 BUCK 1-26 1109 1117 1118 1159 1174 1305 1348 1365 1381 40000 210353596 35963 SINCIC 1-36 2511 2539 2541 2551 2651 2856 2946 3003 3037 30000 210353765 37659 GARDNER 18- 2575 2608 2619 2726 2946 3039 3088 3124 90000 32 210354098 40984 DAVIS 1-12 2739 2755 2756 2770 2874 3160 3253 3324 3354 40000 210354098 40987 STATE WINTERFIELD 1-31 2375 2382 2383 2392 2462 2836 2922 2999 3029 70000 210356013 60134 KROCKER 1-17 2863 2894 2895 2926 3016 3258 3350 3410 3442 40000

137 API Permi Well Name Well Sunb Bere Bedf U Ellsw Lach Paxt Norw Trave t # ury a ord Antri orth ine on ood rse m 210372234 22348 ZISCHKE, ERVIN W & 1 2046 2060 2079 2133 2196 2297 2344 2373 2389 80000 ARLIN E 210372781 27811 FOX 1 1999 2013 2032 2099 2179 2276 2333 2365 2374 10000 210373538 35383 FARLEY 1-10 2123 2134 2135 2207 2274 2381 2452 2470 2487 30000 210393572 35726 STATE LOVELLS 1-33 1506 1531 1532 1557 1688 1888 1964 2009 2036 60000 210393591 35917 STATE GRAYLING 1-33 1817 1843 1844 1872 1958 2269 2344 2398 2423 70000 210453970 39708 KEELEY 1-11 1818 1835 1836 1870 1913 2008 2062 2099 2115 80000 210454142 41423 ARNOLD UNIT 1-8 1911 1919 1920 1942 2020 2115 2189 2215 2231 30000 210454450 44505 RIPLEY 1-16 1612 1621 1622 1658 1684 1828 1884 1900 1918 50000 210454886 48869 PARSONS C4- 1916 1943 1944 1978 2029 2125 2170 2212 2226 90000 11 210493517 35172 STORY 1-3 1478 1519 1605 1742 1849 1875 1909 1930 1949 20000 210493599 35999 IUNG 1-7 1568 1594 1772 1812 1892 1912 1976 1994 2016 90000 210494874 48746 SWANEBECK D4- 1332 1351 1466 1556 1629 1645 1697 1716 1738 60000 34 210513584 35846 DULL 3-6 2609 2642 2645 2680 2802 2968 3059 3114 3149 60000 210513864 38648 STATE SHERIDAN 1-15 1839 1865 1885 1970 2086 2198 2279 2321 2351 80000 210513898 38980 STATE BUTMAN A 1 2137 2164 2166 2239 2347 2509 2593 2653 2682 00000 210515867 58676 BOREN 1-25 2155 2180 2256 2320 2453 2548 2629 2675 2712 60000

138 API Permi Well Name Well Sunb Bere Bedf U Ellsw Lach Paxt Norw Trave t # ury a ord Antri orth ine on ood rse m 210554090 40901 STATE WHITEWATER 1-13 877 1521 1588 1620 1646 10000 210555901 59019 AMOS C1-4 1031 1098 1123 1143 90000 210572427 24270 DARVISHIAN, HARRY ET 1 2289 2311 2311 2347 2446 2604 2676 2694 2716 00000 UX 210572973 29739 McClure SPARKS 1-8 2206 2220 2250 2286 2387 2517 2582 2596 2617 90000 210573338 33382 FABUS, JOHN 1-22 2038 2059 2061 2126 2220 2349 2416 2428 2451 20000 210573737 37377 SMITH 1-17 2284 2302 2302 2319 2394 2576 2648 2657 2682 70000 210573999 39991 WILES 1-34 2293 2307 2308 2327 2418 2576 2650 2667 2683 10000 210574099 40994 FROST 1-1 2281 2298 2342 2403 2505 2577 2620 2638 2650 40000 210574352 43523 SHERWOOD 1-27 2314 2330 2377 2419 2529 2655 2723 2745 2767 30000 210576050 60504 SKARYD 2018 2034 2038 2108 2215 2318 2380 2396 2416 40000 210593963 39631 WEIDNER 1-1 1220 1231 1232 1289 1343 1383 1419 1460 1472 10000 210594612 46129 ZAIOCZKOWSKI 1-2 679 688 689 745 780 821 860 895 912 90000 210632992 29926 REIBLING, ARNOLD 1 1540 1626 1706 1918 2030 2065 2112 2142 2171 60000 UNIT 210633355 33550 CLANCY, FELIX & DORIS 2-12 1024 1134 1237 1427 1533 1567 1627 1651 1678 00000 210634084 40849 ELENBAUM & GETTEL 1-5 1394 1451 1556 1719 1818 1869 1912 1938 1964 90000 ET AL 210654899 48992 BOTSFORD A2-8 1929 1943 1969 2075 2162 2201 2245 2286 2298 20000

139 API Permi Well Name Well Sunb Bere Bedf U Ellsw Lach Paxt Norw Trave t # ury a ord Antri orth ine on ood rse m 210654982 49825 STONE D1- 1895 1910 1914 1985 2054 2112 2157 2199 2214 50000 28 210672357 23574 WILDMAN, VIRGIL 1 1849 1857 1858 1898 1919 2102 2160 2171 2185 40000 210672461 24619 TEN CATE, SAMUEL ET 1 1782 1797 1798 1833 1892 2120 2172 2181 2200 90000 AL 210672537 27271 ALBRIGHT, EDWIN & 1 2079 2093 2094 2100 2156 2341 2409 2425 2440 50100 EVA 210673643 36435 STOWE & LUMBERT 1-11 2177 2191 2192 2198 2269 2456 2526 2535 2555 50000 210673670 36704 FERRIS & STATE 1-28 1948 1959 1959 1961 2016 2181 2263 2272 2289 40000 ORANGE ET AL 210674165 41656 BEHRENWALD UNIT 1-34 1789 1803 1804 1840 1901 2119 2184 2191 2202 60000 210694263 42638 STATE RENO 1-33 1583 1611 1679 1775 1919 1995 2074 2115 2147 80000 210694992 49928 STATE OSCODA 1A- 1664 1697 1748 1855 2007 2083 2157 2190 2221 80000 32 210732425 24256 RICHEY, BURDON ET AL 1 2502 2528 2528 2547 2626 2818 2895 2915 2936 60000 COMM 210733416 34166 POHL 1-36 2592 2610 2611 2620 2752 2924 3002 3042 3063 60000 210733571 35711 BURROWS 1 2494 2516 2516 2555 2678 2827 2910 2941 2970 10000 210733724 37243 BEUTLER CARL & 1-34 2546 2575 2576 2584 2670 2901 2981 3014 3039 30000 HOWARD ET AL 210733790 37909 MCDONALD 1-16 2557 2570 2571 2585 2711 2896 2975 2999 3026 90000 210753894 38946 BENN ET AL 1-17 1441 1456 1457 1520 1560 1618 1658 1706 1718 60000 210753922 39221 BAUM 24- 1612 1626 1627 1655 1712 1781 1819 1861 1872 10000 43

140 API Permi Well Name Well Sunb Bere Bedf U Ellsw Lach Paxt Norw Trave t # ury a ord Antri orth ine on ood rse m 210794026 40262 STATE BLUE LAKE 1-24 1192 1197 1198 1222 1319 1677 1764 1807 1833 20000 210794125 41251 INGERSOLL 31- 1569 1575 1575 1594 1616 2228 2308 2375 2403 10000 31 210794496 44967 STATE BEAR LAKE 1-9 1362 1368 1369 1387 1484 1859 1953 1993 2020 70000 210794596 45965 STATE GARFIELD 2-19 1891 1905 1906 1946 2001 2389 2472 2512 2534 50000 210794628 46280 LEE 1-20 1044 1054 1055 1061 1113 1626 1680 1717 1742 00000 210812462 24627 FRANCISCO, GEORGE 1 1928 1958 1959 2002 2050 2309 2349 2353 2369 70000 210812482 24826 TEN, JOHN & HAVE 1 1915 1925 1926 1987 2040 2328 2395 2399 2421 60000 210813916 39162 CARLSON ET AL 1-7 2024 2032 2033 2092 2141 2453 2529 2534 2549 20000 210814055 40553 REUHS 1-19 1351 1355 1356 1377 1378 1791 1851 1862 1872 30000 210814208 42089 TUINSTRA 1-13 1344 1349 1350 1355 1360 1795 1861 1867 1878 90000 210814886 48865 VANDENBERG B1- 1800 1822 1823 1887 1932 2208 2276 2279 2297 50000 17 210815547 55477 SWITZER 1-13 1634 1644 1645 1730 1764 2106 2182 2189 2204 70000 210852685 26850 STATE PEACOCK 2 1371 1402 1403 1421 1422 2063 2155 2187 2205 00000 210855601 56011 STATE YATES 1-27 1918 1962 1963 1969 1970 2517 2635 2659 2672 10000 210872394 23946 THOM, IRVING GG- 1403 1449 1572 1684 1781 1807 1843 1869 1891 60000 1 210933519 35199 LAVEY & STATE 1-7 1297 1309 1423 1497 1557 1586 1611 1654 1665 90000 HAMBURG

141 API Permi Well Name Well Sunb Bere Bedf U Ellsw Lach Paxt Norw Trave t # ury a ord Antri orth ine on ood rse m 210933613 36132 LAVEY & STATE 1-6 1306 1318 1431 1507 1569 1600 1623 1665 1681 20000 HAMBURG 210933789 37893 LAIER 1-23 769 785 885 964 1044 1074 1094 1128 1141 30100 210934549 45492 DEY A1- 1336 1347 1449 1503 1576 1610 1648 1687 1699 20000 15 210934874 48748 GRAMER D1- 854 870 978 1049 1130 1139 1184 1218 1232 80000 11 210995206 52064 BEREGASSI 1-5 358 457 463 497 513 524 40000 211013897 38978 LAKELAND 1-32 981 1003 1016 80000 ASSOCIATION & STATE ET AL 1 211015398 53981 LECKRONE 4-9 1000 1082 1119 1136 10000 211052395 23956 JACOBSON UNIT 1 766 782 783 802 803 1505 1573 1594 1603 60000 211053211 32111 STATE HAMLIN 2-24 1276 1366 1374 1392 10000 211055506 55068 FREESOIL A3- 577 601 602 603 604 1212 1308 1324 1339 80000 22 211073542 35426 ROUSSEAU, THELMA 1-12 2631 2651 2652 2658 2749 3095 3176 3211 3244 60000 211073772 37723 MAY 29- 2610 2636 2637 2655 2729 3037 3114 3148 3168 30000 23 211074126 41267 ORR 1-9 2745 2766 2767 2785 2856 3125 3206 3220 3247 70000 211074144 41440 BASSETT 1-16 2145 2169 2170 2189 2214 2716 2796 2818 2833 00000 211074226 42264 STATE COLFAX & 1-31 2345 2367 2368 2380 2458 2839 2906 2934 2953 40000 KNIGHT

142 API Permi Well Name Well Sunb Bere Bedf U Ellsw Lach Paxt Norw Trave t # ury a ord Antri orth ine on ood rse m 211113693 36931 PLONA UNIT 1 2385 2419 2444 2497 2592 2741 2818 2858 2884 10000 211114139 41390 WHITNEY 1-33 2471 2499 2523 2568 2693 2841 2933 2981 3011 00000 211114153 41530 BADOUR 1-31 2388 2406 2471 2551 2674 2763 2830 2869 2893 00000 211132749 27499 STATE WEST BRANCH 1 1981 2004 2005 2041 2110 2510 2624 2651 2685 90000 211133218 32181 TANIS, GLEN ET AL 1-11 2500 2515 2516 2543 2558 3083 3171 3242 3267 10000 211133706 37062 EDWARDS 5-30 2398 2420 2421 2443 2505 2917 3002 3069 3100 20000 211172459 24594 BUCHHOLZ, KENNETH 1 2402 2421 2422 2448 2514 2784 2857 2878 2899 40000 211172979 29790 FANCETT 1-28 2237 2245 2246 2263 2342 2527 2598 2610 2629 00000 211173002 30027 BETHAM & CLARK UNIT 1-26 2275 2281 2282 2317 2416 2605 2678 2688 2706 70000 211173011 30117 THORLUND FARMS 1-6 2214 2224 2225 2263 2356 2560 2632 2640 2659 70000 211173808 38081 TIESNER 1-27 2385 2393 2394 2420 2504 2720 2792 2799 2822 10000 211173948 39483 WALDRON 2-30 2379 2384 2385 2415 2509 2678 2749 2761 2781 30000 211174148 41480 ROCHA & BUCK 1-15 2330 2335 2336 2365 2455 2635 2708 2717 2736 00000 211175004 50047 TOW 1-3 2280 2286 2287 2317 2397 2575 2647 2661 2678 70000 HD- 1 211193647 36478 DAVIS ET AL 1-19 1111 1124 1123 1176 1261 1484 1557 1594 1617 80000

143 API Permi Well Name Well Sunb Bere Bedf U Ellsw Lach Paxt Norw Trave t # ury a ord Antri orth ine on ood rse m 211195020 50204 STATE HILLMAN 1-21 532 612 635 658 40000 211213463 34633 MELTON 1 879 905 906 923 924 1464 1560 1566 1585 30000 211214139 41397 HIBBS 1-22 1197 1199 1200 1201 1202 1768 1858 1862 1876 70000 211214321 43217 EILERS 1-13 843 896 1472 1566 1582 1605 70000 211232284 22849 SHERIDAN TOWNSHIP 1 1404 1448 1449 1468 1469 2021 2107 2113 2123 90000 211232666 26662 THOMPSON, WALTER F 1 1363 1399 1400 1412 1413 1990 2081 2087 2092 20000 & ROSILEA 211233843 38437 MCDONALD 1-12 2010 2050 2509 2594 2600 2612 70000 211233856 38561 ROSS 1-9 1775 1817 1818 1857 1866 2377 2468 2475 2481 10100 211233916 39166 ALTMAN 1-20 2081 2116 2117 2143 2155 2660 2749 2764 2773 60000 211233995 39952 WISE, ET AT 1-3 2009 2051 2052 2071 2111 2559 2641 2652 2665 20000 211234061 40619 STATE GARFIELD & 1-14 1591 1637 1638 1688 1697 2191 2290 2295 2303 90000 ANDERSON 211234116 41162 MASON UNIT 1-24 1950 1985 1986 2042 2090 2463 2548 2553 2564 20000 211234189 41892 CROTON 1-30 1796 1838 1839 1867 1912 2354 2439 2444 2454 20000 211234621 46218 U S A MERRILL 1-18 1530 1574 1577 1590 1591 2125 2241 2255 2263 80000 211235808 58081 WELLMASTER 1-1 2145 2185 2186 2195 2237 2684 2771 2784 2799 10000 211235919 59190 STATE MONROE C1-6 1804 1848 1849 1859 1860 2415 2529 2548 2559 00000

144

API Permi Well Name Well Sunb Bere Bedf U Ellsw Lach Paxt Norw Trave t # ury a ord Antri orth ine on ood rse m 211254928 49282 INNOVATIVE LAND 2-29 645 671 773 940 997 1004 1015 1028 1037 20000 DEVELOPMENTS INC 2-2 211273449 34490 ADAMS 1-5 749 760 789 789 1394 1477 1489 1495 00000 211273539 35393 FEDO 1-27 1322 1349 1350 1366 1367 1932 2029 2036 2041 30000 211273907 39073 LAUBER, JACK 1A-6 745 753 755 782 782 30000 211273929 39291 BOONE 1-10 1470 1553 1563 1570 10000 211274043 40430 STATE COLFAX 1-4 1210 1236 1237 1256 1253 1848 1944 1957 1966 00000 211275200 52001 ROOD 1-23 1273 1301 1302 1318 1319 1916 1991 1996 2018 10000 211275824 58249 STATE HART & 4-30 1148 1161 1162 1178 1179 1806 1901 1906 1911 90000 FUEHRING 211294525 45254 BECK (167) 8-30 1359 1382 1412 1473 1624 1737 1827 1865 1902 40000 211333453 34536 MCCORMICK ET AL 2-27 2437 2448 2449 2455 2516 2915 2988 3027 3050 60000 211333918 39186 BOYCE 1-19 2134 2159 2160 2161 2162 2809 2884 2944 2965 60000 211333961 41354 SHERMAN 1-20 2451 2470 2471 2506 2513 3068 3169 3233 3263 50200 211334055 40556 LOWE 1-27 2264 2276 2277 2280 2299 2848 2949 2996 3014 60000 211334154 41540 MCCASEY 1-7 2379 2396 2397 2413 2440 2931 3010 3046 3066 00000 211334178 41788 EISENGA BROTHERS 11- 2644 2662 2663 2688 2715 3184 3265 3329 3358 80000 24

145 API Permi Well Name Well Sunb Bere Bedf U Ellsw Lach Paxt Norw Trave t # ury a ord Antri orth ine on ood rse m 211334294 42945 STATE HARTWICK & 1-27 2758 2775 2776 2785 2841 3258 3338 3390 3420 50000 YARHOUSE 211335641 56415 EISENGA C2- 2641 2656 2657 2690 2699 3205 3293 3359 3385 50000 17 211354735 47355 U S A & STATE BIG 1-35 1885 1911 1912 1929 2079 2253 2340 2384 2411 50000 CREEK 211354792 47920 LE MIEUX D4- 1447 1480 1481 1507 1648 1796 1881 1924 1955 00000 28 211354801 48012 SHIRODA D2- 1472 1501 1546 1615 1729 1867 1950 1983 2014 20000 22 211374741 47417 WOLF 4-33 998 1006 1007 1059 1149 1429 1513 1556 1582 70000 211375478 54789 WEST CORWITH A2- 715 795 827 848 90000 25 211393488 34885 UMLOR ROBERT ET AL 1-3 1480 1510 1511 1587 1600 2060 2120 2128 2152 50000 211393952 39527 VAN RYSWYCK 1-34 1105 1151 1152 1177 1180 1754 1825 1830 1852 70000 211394143 41434 VANSLOOTEN 1-25 893 927 928 946 947 1474 1570 1574 1581 40000 211433601 36015 SAINT HELEN UNIT 53 1585 1595 1596 1612 1722 1945 2027 2081 2114 50000 TRACT 15 211433740 37409 ROSEVILLE GUN CLUB 1-17 2549 2574 2574 2613 2713 2921 3008 3067 3095 90000 B 211433911 39113 STATE RICHFIELD 1-26 1972 1990 1993 2038 2162 2331 2420 2470 2500 30000 211433982 39826 BGHL & SCHMALTZ 1-23 2358 2386 2388 2436 2538 2769 2856 2926 2952 60000 211453274 32745 WIROSTEK, ERNEST 1 1946 1961 1972 2056 2150 2255 2311 2328 2352 50000 211453636 36363 UNION TEXAS & THOMS, 1-27 1969 1990 2126 2170 2293 2354 2409 2443 2464 30000 C F, UNIT

146 API Permi Well Name Well Sunb Bere Bedf U Ellsw Lach Paxt Norw Trave t # ury a ord Antri orth ine on ood rse m 211453971 39716 MURPHY 1-16 2166 2186 2242 2346 2475 2544 2609 2635 2653 60000 211455250 52508 RIVERCREST 13- 1889 1906 1958 2038 2147 2235 2291 2305 2327 80000 20 211472667 26672 KLUNG, FRANK UNIT 1 909 1004 1082 1290 1388 1407 1431 1456 1473 20000 211473896 38964 ARCO & SENYK 1-30 696 768 847 1054 1145 1156 1179 1200 1213 40000 211474069 40698 LOHR 1-16 580 706 779 976 1085 1104 1145 1162 1178 80000 211512404 24047 CLEARY, RAY J 1 795 918 1011 1229 1331 1353 1414 1431 1456 70000 211513577 35779 FROSTIC 1-30 493 529 796 889 912 957 986 1002 90000 211552754 27549 BIRCHMEIER 1 1653 1672 1794 1902 1983 2039 2072 2093 2115 90000 211552790 27907 JELINEK & FERRIS UNIT 1 1525 1538 1610 1687 1748 1824 1861 1895 1909 70000 211553072 30727 HASSELBRING ET AL 1-5 1903 1914 1951 2029 2094 2188 2233 2257 2272 70000 211554877 48772 CONN D2-8 1451 1465 1608 1644 1721 1755 1802 1839 1857 20000 211572348 23485 DANCEY, R. V. 1 1446 1501 1588 1771 1867 1914 1960 1985 2010 50000 211573306 33064 WALAT FARMS, INC ET A-3- 1507 1547 1628 1785 1881 1913 1995 2017 2043 40000 AL 26 211573536 35369 NIXON & NIXON 2-24 1402 1440 1527 1676 1791 1824 1884 1917 1937 90000 211574028 40284 CLOTHIER 43- 1538 1612 1752 1873 1981 2003 2047 2072 2092 40000 28 211574028 40285 BINKOWSKI 21- 1733 1784 1899 2044 2154 2184 2233 2256 2277 50000 11

147

API Permi Well Name Well Sunb Bere Bedf U Ellsw Lach Paxt Norw Trave t # ury a ord Antri orth ine on ood rse m 211615992 59920 COOK 1-27 189 152 189 221 238 266 283 00000 211652289 22890 LEESON, ETHELYN & 1 2232 2248 2249 2268 2283 2896 2970 3041 3065 00000 SOURS, RUBERTA J 211654021 40210 DOSTAL, ET AL 1-27 1554 1582 1583 1592 1593 2231 2316 2360 2380 00000 211655596 55968 HOESL 1-9 1063 1064 1065 1066 1067 1711 1787 1821 1838 80000 M5305 624- Dow/ERDA 100 820 956 1012 1205 1328 1344 1389 1406 1429 771- 474 M5308 687- Dow/ERDA 103 935 990 1194 1310 1327 771- 474 M5309 686- Dow/ERDA 104 796 930 986 1190 1311 1324 1364 771- 474

148