THESE TERMS GOVERN YOUR USE OF THIS DOCUMENT

Your use of this Ontario Geological Survey document (the “Content”) is governed by the terms set out on this page (“Terms of Use”). By downloading this Content, you (the “User”) have accepted, and have agreed to be bound by, the Terms of Use.

Content: This Content is offered by the Province of Ontario’s Ministry of Northern Development and Mines (MNDM) as a public service, on an “as-is” basis. Recommendations and statements of opinion expressed in the Content are those of the author or authors and are not to be construed as statement of government policy. You are solely responsible for your use of the Content. You should not rely on the Content for legal advice nor as authoritative in your particular circumstances. Users should verify the accuracy and applicability of any Content before acting on it. MNDM does not guarantee, or make any warranty express or implied, that the Content is current, accurate, complete or reliable. MNDM is not responsible for any damage however caused, which results, directly or indirectly, from your use of the Content. MNDM assumes no legal liability or responsibility for the Content whatsoever.

Links to Other Web Sites: This Content may contain links, to Web sites that are not operated by MNDM. Linked Web sites may not be available in French. MNDM neither endorses nor assumes any responsibility for the safety, accuracy or availability of linked Web sites or the information contained on them. The linked Web sites, their operation and content are the responsibility of the person or entity for which they were created or maintained (the “Owner”). Both your use of a linked Web site, and your right to use or reproduce information or materials from a linked Web site, are subject to the terms of use governing that particular Web site. Any comments or inquiries regarding a linked Web site must be directed to its Owner.

Copyright: Canadian and international intellectual property laws protect the Content. Unless otherwise indicated, copyright is held by the Queen’s Printer for Ontario.

It is recommended that reference to the Content be made in the following form: Brigham, R.J., 1971. Structural geology of Southwestern Ontario and Southeastern Michigan; Mines and Northern Affairs, Petroleum Resources Section, Paper 71-2, 120p.

Use and Reproduction of Content: The Content may be used and reproduced only in accordance with applicable intellectual property laws. Non-commercial use of unsubstantial excerpts of the Content is permitted provided that appropriate credit is given and Crown copyright is acknowledged. Any substantial reproduction of the Content or any commercial use of all or part of the Content is prohibited without the prior written permission of MNDM. Substantial reproduction includes the reproduction of any illustration or figure, such as, but not limited to graphs, charts and maps. Commercial use includes commercial distribution of the Content, the reproduction of multiple copies of the Content for any purpose whether or not commercial, use of the Content in commercial publications, and the creation of value-added products using the Content.

Contact:

FOR FURTHER PLEASE CONTACT: BY TELEPHONE: BY E-MAIL: INFORMATION ON The Reproduction of MNDM Publication Local: (705) 670-5691 Content Services Toll Free: 1-888-415-9845, ext. [email protected] 5691 (inside Canada, United States) The Purchase of MNDM Publication Local: (705) 670-5691 MNDM Publications Sales Toll Free: 1-888-415-9845, ext. [email protected] 5691 (inside Canada, United States) Crown Copyright Queen’s Printer Local: (416) 326-2678 Toll Free: 1-800-668-9938 [email protected] (inside Canada, United States)

cm cm cm 1 ram by Robert J. Brigham

mines and northern affairs petroleum resources section

paper 71-2

structural geology of southwestern Ontario and southeastern michigan

robert j. brigham

toronto canada

foreword

The Department of Mines and Northern Affairs, in cooperation with the Department of Geology, University of Western Ontario, is pleased to have the opportunity to publish the thesis of Robert J. Brigham, submitted in partial fulfillment of the requirements for the degree of Doctor of Philoso­ phy. We are hopeful that the results of this work will assist persons in both the public and private sectors to better understand the structural geology of southwestern Ontario and to assist in the evaluation and exploitation of Ontario's petroleum resources. With the consent of the author, several changes have been made in the original text to take into account recent modifications in the Ontario Well Data System. The stratigraphic nomenclature presently in use for southern Ontario has also been revised from that proposed by Beards (1967), and used throughout this paper. For reference purposes, the current rock stratigraphic succession is included as an appendix.

D.A.S. Toronto 22.12.71

iii table of contents

Page

FOREWORD iii TABLE OF CONTENTS iv LIST OF FIGURES * vi ABSTRACT ix ACKNOWLEDGMENT x

CHAPTER 1 - INTRODUCTION 1 Description of Study Area 3

CHAPTER 2 - DEVELOPMENT OF DATA FILE AND COMPUTER PROGRAMS 5

Data Sources 5 Data Gathering and Reduction 5 File Maintenance, Updating and Access 8 File Usage Programs 9 Map Errors 14 Preparation of Illustrations 14 Contribution of the Data File and Programs to the Study 14 CHAPTER 3 - STRATIGRAPHY 15 Introduction 15 Cambrian 15 Ordovician 16 Silurian 21 Devonian 31 Discussion and Summary 35 CHAPTER 4 - REGIONAL STRUCTURE 37 Precambrian Surface 37 Trenton Surface 39 Queenston Surface 39 Clinton Surface 41 Guelph Surface 41 Bass Islands Surface 42 Dundee Surface 44 Pre-Pleistocene Erosional Surface 44 Discussion and Summary 45 CHAPTER 5 - HISTORICAL DEVELOPMENT OF REGIONAL STRUCTURE. . 47 Cambrian 48 Ordovician 48 Silurian 49 Devonian 49 Discussion 50

iv Page

CHAPTER 6 - DEVELOPMENT OF SUBREGIONAL STRUCTURES 57 Trend Surface Analysis 57 Clinton Structure 59 A-2 Carbonate Structure 63 Bass Islands Structure 67 Detroit River Structure 67 Discussion and Summary 67 CHAPTER 7 - STRUCTURAL MANIFESTATIONS OF SALT REMOVAL 69 Regional Structural Effect 69 Time of Salt Dissolution 71 Lateral Variation in Time of Salt Leaching 71 Generation of Localized Structure due to Variations in Time of Leaching 75 Modification of Existing Structures by Salt Leaching 79 Mechanism of Salt Leaching 80 Discussion and Summary 81 CHAPTER 8 - REEFS 83 Definition 83 Reef Composition 84 Reefs in the Study Area 84 Porosity 86 Reef Structure 86 Time of Reef Growth 90 Relationship of Reefs to Regional Structure 92 Discussion and Summary 100 CHAPTER 9 - SUMMARY OF CONCLUSIONS 101 REFERENCES 103 APPENDIX I Card Formats for Ontario Well Data System 107 APPENDIX II Rock-Stratigraphic Succession Units Toronto-Windsor Area 110

v list of figures

Figure Page 1 — 1 Regional basement structure 2 1—2 Subcrop map on base of Pleistocene 4 2—1 Distribution of present production 6 2—2 ODMNA well history summary card 7 2—3 Preprinted well printout form 7 2—4 Example map sheet 10 2—5 Cross section index map 12 2—6 Cross section 12 3—1 Index map of study area 16 3—2 Composite stratigraphic section, Lambton County, Ontario 17 3—3 Paleozoic nomenclature chart for southern Ontario 17 3—4 Index map of southwestern Ontario and southeastern Michigan 18 3—5 Isopachous map of Cambrian Formations 19 3—6 Isopachous map of Trenton — Black River Groups 19 3—7 Isopachous map of Meaford-Dundas-Collingwood Formations 20 3—8 Isopachous map of Queenston Formation 20 3—9 Isopachous map of Queenston-Collingwood Formations 22 3—10 Isopachous map of Cataract Group 22 3—11 Isopachous map of Clinton Group 24 3—12 Isopachous map of Guelph-Lockport Formations 24 3-13 West-east Salina cross section 26 3—14 Isopachous map of the total Salina 26 3—15 Isopachous map of A-1 Evaporite 28 3—16 Isopachous map of A-l Carbonate 28 3—17 Isopachous map of A-2 Evaporite 29 3—18 Isopachous map of A-2 Carbonate 29 3—19 Isopachous map of B Salt 30 3—20 Isopachous map of Bass Islands Formation 30 3—21 Isopachous map of Bois Blanc Formation 32 3—22 Isopachous map of Sylvania Formation 32 3-23 Isopachous map of Detroit River Group 33 3-24 Isopachous map of Dundee Formation 33 3-25 Isopachous map of Hamilton Formation 34 3—26 Isopachous map of Kettle Point Formation 34 4—1 Structure contours on Precambrian 38 4-2 Index of Control Wells 38 Table 4-1 Comparison of Rates of Dip on Structural Features of Southwestern Ontario 40 vi Figure Page 4—3 Structure contours on Trenton Group 41 4—4 Structure contours on Clinton Group 42 4—5 Structure contours on Guelph Formation 43 4—6 Structure contours on Bass Islands Formation 43 4—7 Structure contours on Dundee Formation 44 4—8 Topography on base of Pleistocence 45 5—1 Cross section across Algonquin Arch at right angles to strike of Precambrian 51 5—2 SW—NE cross section along Findlay-Algonquin Arches 52 5—3 Isopachous map of Queenston to Trenton 53 5—4 Isopachous map of Guelph-Lockport to Trenton 54 5—5 Isopachous map of A-2 Carbonate to Trenton 54 5—6 Isopachous map of Bass Islands to Trenton 55 5—7 Isopachous map of Detroit River to Trenton 55 6—1 Structure contours on Trenton Group 58 6—2 Structure contours on Clinton Group 58 6—3 First through fourth order Clinton trend surfaces 60 6—4 Second order residual from Clinton structure 62 6—5 Structure contours on A-2 Carbonate 62 6—6 Second order residual from A-2 Carbonate structure 64 6—7 Structure contours on Bass Islands Formation 64 6—8 Structure contours on Detroit River Group 66 6—9 Isopachous map of B Salt 66 7—1 Structure contours on A-2 Carbonate 70 7—2 Structure contours on C Shale 70 7—3 Gamma Ray cross section showing salt collapse 72 7—4 Salt leaching effects along Dawn structure 74 7—5 Salt leaching effects along Dawn structure 74 7—6 SE—NW cross section showing multiple stages of salt leaching 75 7—7 Reconstruction of salt collapse 76 7—8 West-east cross section across Clearville Field 77 7—9 Isopachous map of Bass Islands to Salina 77 7—10 North-south cross section across Dawn structure and Electric Fault 78 7-11 SW—NE cross section across Dawn 156 Reef Field 79 7—12 North-south cross section through Wardsville Silurian Field 81 7—13 Isopachous map of B Salt showing location of Devonian oil fields 82 8—1 Reef height and distribution 85 8—2 Structure contours on Guelph Formation 87 8-3 Seckerton North and Seckerton Reefs 88

vii Figure 8—4 Structure contours on A-2 Carbonate 8—5 Isopachous map of A-2 Carbonate to Guelph 8—6 Isopachous map of B Salt 8—7 Structure contours on B Salt 8—8 Index Map of reef study area 8—9 Structure contours on Guelph Formation 8—10 Structure contours on Clinton Group . . . 8—11 First order Clinton residual 8—12 Structure contours on non-reef Guelph Formation 8—13 Guelph residuals from non-reef surface

viii abstract

The subtle but definite structure of Paleozoic sediments in southwestern Ontario has received little attention and, due to the low regional dip and scarcity of outcrops, most geologic information is available only from the records of wells drilled for oil and gas. Prior to the advent of the Ontario Well Data file, regional studies required time-consuming searches of the files maintained by the Ontario Department of Mines and Northern Affairs, the Geological Survey of Canada, and the various oil companies. The Ontario Well Data file and associated computer programs made possible both regional and detailed study of the structure of a selected area of southwestern Ontario. The study area was extended across the international boundary into southeastern Michigan by utilizing data from a comparable file made available by Petroleum Information of Denver, Colorado. The structural attitude of the formations in this region has been affected by at least three factors: tectonic deformation, collapse as a result of dissolution of underlying salt strata, and drape over organic reefs. The major regional structural features of southwestern Ontario and southeastern Michigan are the Algonquin and Findlay Arches; the Chatham Sag, formed by the mutual plunge of the arches; and the Michigan and Appalachian Basins. The Algonquin and Findlay Arches, although considered by some as separate entities, are in reality two ends of the same feature, the less negative area between the Michigan and Appalachian Basins. The Upper Cambrian sediments provide evidence of an initial arch prior to that time. Dip into the Appalachian Basin was steepened mainly during two separate times, one in the Late Ordovician and the second in Late Devonian time. Dip into the Michigan Basin was created largely during Late Silurian and Middle Devonian times. The Chatham Sag is a fairly late feature which did not have a significant influence on sedimentation until late Silurian time. Salt dissolution occurred concurrently with sedimentation over long periods of geologic time. Frequently the focus of dissolution is a structural feature and collapse commonly -masks the history of the structural deformation. Structural "highs", as well as "lows", can be formed by collapse due to salt dissolution. The highs can be formed by draping over salt remnants or by draping over insoluble masses which were the result of an earlier period of dissolution. Salt dissolution appears to be the controlling factor for the traps of the shallow Devonian oil fields in the study area. Draping has also occurred over the margin of the reef complex which surrounds the Michigan Basin and over the pinnacle reefs found in a band parallel to and basinward from the reef complex.

ix acknowledgment

Many individuals and groups have contributed to the success of this study and the writer wishes to express his sincere appreciation. He wishes to thank his advisory committee — Drs. C. G. Winder, J. F. Hart and R. W. Hutchinson — for their considerable support and encouragement. The assistance, patience and friendship of D. A. Sharp, who supervised data preparation at the Ontario Department of Mines and Northern Affairs and has been a staunch supporter of the project, have been greatly appreciated. The writer is pleased to acknowledge the contribution of his co-workers on the Well Data project, Barry Arnett, Lorelei Konchak and Barbara Mills. The financial support of the University, the ODMNA, the participating companies, and the Geological Survey of Canada is gratefully acknowledged. The contributions by the management and staff of Amoco Production Company of computer time and assistance in drafting and reproduction have been considerable and are very much appreciated. Special thanks go to Marthann David for her critical reading of the manuscript. Finally, the writer acknowledges with affection the cooperation and sacrifices of his wife, Mary.

x chapter 1

introduction

Southwestern Ontario, a part of the Interior Lowlands (King, 1959) near the southern margin of the Canadian Shield, is underlain by essentially flat-lying Paleozoic sediments which range in age from Upper Cambrian to Upper Devonian. Because of the gentle and regular structure of the rocks, most previous geologic investigations of the area have been devoted primarily to stratigraphy. Little attention has been given to the historical development of the subtle, but definite, structural configuration of the region. Due to the low regional dip and scarcity of outcrops in this drift-covered region, most of the structural and stratigraphic information is available only from the records of wells drilled in the area for oil and gas. Prior to 1963, such records were maintained mainly by various oil companies engaged in exploration and production of oil and gas, and by the Ontario Department of Mines and Northern Affairs (formerly the Ontario Department of Energy and Resources Management). There was no readily accessible store of infor­ mation for either academic or industrial use. Geological studies by oil com­ panies for exploration purposes or by university scholars were, of necessity, prefaced by time-consuming and costly searches of files maintained by the Ontario Department of Mines and Northern Affairs (ODMNA) in Toronto and by the Geological Survey of Canada in Ottawa. In 1963, Dr. J. F. Hart, Department of Computer Science of the University of Western Ontario, initiated a pilot study to determine the feasibility of recording on punched cards the data from wells drilled in Ontario for oil and gas. After Hart successfully demonstrated the practicality of applying electronic data processing techniques to the Ontario well data, funds were obtained from ODMNA, the University of Western Ontario, and several of the companies directly involved in the Ontario oil and gas industry to undertake a full-scale project to prepare a computer file of the data. The project was designated the Ontario Well Data Project and was a three-year cooperative effort of the contributing organizations. ODMNA was respon­ sible for data preparation, the University was responsible for data reduction and program development, and the participating oil companies provided personnel for an advisory committee to determine objectives and evaluate results. Mr. D. A. Sharp of ODMNA supervised data preparation and the writer was project supervisor at the University.

1

Description of Study Area The area selected for this investigation (Figure 1-1) includes portions of southwestern Ontario and southeastern Michigan and is defined by latitudes 43° 35' N and 41° 40' N and longitudes 84° 00' W and 80° 00' W, except that portion southeast of the centre line of Lake Erie. Across the area, the southwest plunging Algonquin Arch and the northeast plunging Findlay Arch separate the Michigan Basin from the Appalachian Basin. The trends of these structures, are reflected in the subcrop patterns at the base of the Pleistocene (Figure 1-2). A broad syncline resulting from the mutual plunge of the two arches, named the Chatham Sag, has a northwest-southeast trend. The Michigan Basin to the west is nearly circular and has a centrally located maximum depth to basement of about 14,000 feet (Cohee, 1948, p. 1441). In contrast, the Appalachian Basin is elongate, has a postulated maximum depth of about 35,000 feet (A.A.P.G. basement map of North America, 1967), and is sharply truncated on the southeastern margin by the Appalachian Structural Front (Fettke, 1948, p. 1488). However, in the study area, basinward dips are similar on both sides of the axis of the Findlay-Algonquin arches. The broad regional structure of the study area has been modified by movements of a local nature, which have resulted in several monoclines and/or minor faults, and by organic activity. During Middle Silurian time, a reef bank encircled the Michigan Basin (Cummings and Shrock, 1928) and isolated bioherms or pinnacles (Sanford, 1968) developed on the Michigan Basin side of the bank. Some of these salients in the study area are as much as 500 feet thick and have influenced the local attitude or structure of the beds which were deposited over them. Further modification of the regional structure has resulted from leaching of salt members within the Upper Silurian Salina Formation and accompanying collapse of the overlying formations. Over distances of less than one-half mile, some stratigraphic levels vary in elevation by 200 feet or more as a result of such salt leaching and collapse.

3

chapter 2

development of data file and computer programs

Data Sources The Ontario oil industry began in 1858 when wells were dug at Oil Springs in Lambton County to produce shallow accumulations associated with observed seepages at the surface. In the ensuing years, several thousand wells were drilled in Lambton County to depths of 300 to 400 feet to probe the "Big Lime", now known as the Devonian Dundee Formation. Since 1889, many thousands of wells have been drilled in the counties north of the eastern part of Lake Erie to produce gas from shallow sands of Silurian age. More recent exploration targets in Ontario are the Middle Silurian Guelph reefs of Lambton, Middlesex and Huron counties; stratigraphic and structural traps in the Cambrian; and an extension into Lake Erie of the Silurian gas trend (Figure 2-1). Over 50,000 wells are estimated to have been drilled for oil and gas in Ontario; however, few adequate records exist for those drilled prior to 1930. ODMNA has compiled all available data from both industry and government files to establish a central file of approximately 20,000 wells. Operators are required by regulation to submit both records and a set of drill cutting samples on all wells drilled in Ontario. Stratigraphic data from examination of the chip or core samples have been incorporated into the well records maintained by ODMNA. The format for the document on which the data are recorded is shown in Figure 2-2. The range of data recorded is indicated by the headings. These documents are the source of data for the Ontario Well Data File (OWD).

Data Gathering and Reduction In the compilation of the OWD file, only those wells for which a valid location was known and which had a reported depth for at least one geologic horizon were included. About 8,300 of the 20,000 wells in the ODMNA file met these minimum requirements. Only data which are common to most of these wells were recorded; i.e. well location, operator, well name, well elevation, formation contact depths, casing record, production, etc. Drill stem test data, of which there are less than 200, and various miscellaneous items of information were not recorded so that data reduction costs could be kept at a minimum and the OWD file could become operational as quickly as possible. Such information can be added to the file at a later time if the need arises.

5

side-tracking. A disadvantage of ordering a file in this manner is that adjoining counties can be widely separated in the file, depending on their names. As searches often cross county boundaries, an identifying number based on coordinates could make file reading more efficient. For this reason, latitude and longitude, in degrees to five decimal places, were used for control in the OWD file. A series of 16 card formats (Appendix I) were used for the OWD file. Each card has a unique number or series.of numbers for identification. The data were key punched directly from the source documents by operators familiar with both the documents and the card formats. In this way, the laborious intermediate step of data transcription was avoided. The file was checked for completeness and accuracy by printing out the contents of each well record and proof-reading the printout against a microfilm of the source document. For this purpose, a preprinted form was designed which matched as closely as possible the format of the source document (Figure 2-3). Corrections of errors or omissions were made directly to the file of punched cards. The printout of each well record was proofread against a microfilm copy of the original document and corrections were made to the file. A second set of printouts was generated and checked. Then, as a final step, every tenth well was proofread for the third time. A check of latitude- longitude values was made by generating a well location map at the same scale as the original base map and superimposing the two on a light table. A program of machine logic checks was used for additional assurance of accuracy. This program provided checking of specific items against accept­ able values and sequential order of depth fields, plus verification for required entries. A printed message from the program indicated all errors detected by the logic checks. As examples: if a well does not have a recorded total depth, the message "WELL REQ TD" is printed; if the code for the final status of the well does not match one of a list of acceptable codes, "INVALID FINAL STATUS" is printed along with the offending entry; if depths do not increase for successively older tops, the conflicting data are printed as part of the message "FORMATION TOP FOR (formation code) AT (depth) SHOULD NOT BE ABOVE (formation code) AT (depth)." Location information necessary to identify and correct the well record is printed following the message or messages. While proofreading assured that the computer file was as accurate as the source file, the logic checks detected errors and inconsistencies not only in the computer file but also in the source documents.

File Maintenance, Updating and Access The OWD file is maintained on magnetic tape in both latitude-longitude sort, and county-township sort for rapid access and ease of duplication. The file contents are not static and must be repeatedly amended. As new wells are drilled and as errors are discovered through usage, the file must be periodically updated. When changes in data or new wells are reported, they are recorded in the hard copy file and from these a series of spread sheets are coded. These are keypunched and submitted to update the file. Sometime after its transfer from the University of Western Ontario to the Petroleum Resources Section of the Department of Mines and Northern Affairs, the OWD was converted from the IBM 7044 in use at that time to the new IBM/360-65 installed at the Department of Highways Data Centre at Downsview. The conversion programming was done by Paul Chapman of the

8 Department of Highways staff. At that time a number of changes made possible by the machine capabilities of the /360 as well as some other desired changes were incorporated in the system. The original update programs provided three basic capabilities: addition of a card, deletion of a card, and exchange of a card, the last being a combination of the first two. A fourth capability of the program permitted the latitude-longitude control field to be changed. Inasmuch as this parameter is common for all cards of a single well, the field can be changed by a single card, thus avoiding re-creation of each card with the corrected control field. This has a limitation in that no other changes may be made to the well on this update run. To these, two other capabilities were added to the system. The first of these lists all the cards having the specified latitude-longitude code. The second deletes all cards in a specified well location. Where several changes are necessary in a well in addition to the latitude-longitude changes, it is convenient to delete and resubmit the entire well. A code digit (1 through 6) punched in the last column of a correction card indicates the update action required of the computer as follows: a "1" indicates addition of a card; a "2" indicates deletion of a card; a "3" indicates exchange (a deletion and an addition) a "4" indicates a change in the latitude-longitude control; a "5" lists all cards for the specified location; and a "6" deletes the entire well at the specified location. The tape file is updated by the computer reading the current file, incorporating the coded changes, and generating a new tape. Prior to executing the update program, correction cards are placed on tape and sorted into the same order as the master tape file, i.e., ascending longitude coordinate, then latitude, and finally card number. During update, the computer merges the correction cards with the master file and eliminates the unwanted cards. If any latitude and/or longitude changes are made by the update, the file must be re-sorted by longitude and latitude to restore the proper file order. During the update, any well added or changed is flagged with a code number to permit machine editing and printing of a well sheet for each well added or corrected. The output can be compared to the hard-copy file for a complete check of the update. It is anticipated that updates will be made at three-month intervals.

File Usage Programs One of the most common uses of a hard-copy well data file is for construction of structure and isopach maps. Development of a program for computer generation of these maps from the OWD file was a primary objective of the project. Hand-posted data are usually plotted on printed base maps which include land survey lines, rivers, and municipal boundaries; however, it was considered impractical to attempt automatic posting of the OWD file data on printed base maps. As an alternative,' data are posted in their proper geographic position on blank transparent paper which can be superimposed over a base map. The data are posted with a high-speed IBM 1403 printer at 10 lines per inch on blank forms 13 inches wide with cross perforations at 17-inch intervals (Figure 2-4). A perforated half-inch margin along each side, containing the printer guide holes, can be removed so that plotted sheets of adjacent areas can be fitted together. One of the primary aims of the computer mapping program was to produce usable maps at a low cost. Since one of the slowest mapping

9 Figure 2—4. Example map sheet

operations is the passing of the input data tape, an extract of the master tape is used which contains only data required by the mapping program: (1) latitude and longitude as five-place decimal degrees, (2) elevation, (3) total depth, (4) an array for drilled depth of each formation reported, (5) legal description of location, (6) formation at total depth, and (7) sequence number. Further efficiency is gained by "map batching", i.e. five areas with up to 20 maps per area may be printed by a single pass of the tape if the areas do not have overlapping longitudes. This is made possible by the longitude-latitude sort sequence of the file. Two types of control cards are required to plot the maps. Each area is designated by a card, listing geographic limits, area name, scale, and number of maps of the area. A card for each map in the area lists the formation code number or numbers, map name, and map-type code. The formation code number represents the position of the formation in the formation depth array. A structure map is identified as map-type 1 and an isopach map as map-type 2. In the case of an isopach map, upper and lower formation boundaries are specified. The program provides the ability to specify up to five alternatives for the boundaries of formations. This provision is useful where a top is included under either a group name or a formation name or where, through pinchout, a formation may not everywhere overlie the same formation. Maps can be posted at any required scale. As a function of the scale, the desired area defined by latitude-longitude boundaries is divided into a matrix of segments the size of the printer sheets. These segments are

10 plotted by the computer as sur>maps which can be joined together to produce the full map. Row and column numbers, plus an alphabetic character which identifies the map within the area, are posted on the tear-off margin (Figure 2-4, A) for convenience in assembling the completed maps. After the control cards have been read and the sub-map areas assigned, the data tape is read to extract wells which fall in the first column of sub-maps. The first well read which has a longitude greater than the maximum of the sub-map column signals the start of the output phase. As each sub-map or map segment is processed, it is represented in the memory of the machine by a matrix of 136 rows by 132 columns of characters or print positions. Each well which has been extracted is tested for inclusion in the map segment. If a well falls within the latitude-longitude limits of the map segment, its location is converted by a scale factor to a print position. A print position is one-tenth of an inch square; therefore, the maximum error which can result would be seven-hundredths of an inch, the distance from a corner to the centre. As a guide to determining the reliability of the subsurface data, a unique symbol is printed at the well location to indicate tops picked (1) by government geologists, (2) from mechanical logs, (3) by company geologists, and (4) by drillers. The value which is to be printed out is stored in the memory matrix'below the well location. If the value is missing and the presence of deeper tops indicate that it is either missing or not recorded, the location symbol is replaced by an asterisk (Figure 2-4, B). For isopach maps, partial penetration of a section is flagged by printing a plus sign (+) following the thickness data. In areas of closely spaced data, several different wells may often require the same print position. Prior to storing a well symbol and the datum, a check is made to determine if the required positions are vacant. If the position for the well spot is vacant but spaces for the datum are not, an alphabetic character is plotted at the well spot (Figure 2-4, C) and the character is listed on a tailsheet at the bottom of the column of map segments along with the well location and datum information. If the well spot is occupied, the datum is printed without the character (Figure 2-4, D). In this way, "crowded-out" data is not lost and can be fitted in by hand later if desired. For large scale maps, printing of asterisks and crowded-out data can be suppressed. After all wells have been checked for inclusion in a map segment, fiducial marks (+) (Figure 2-4, E) are placed at each intersection of 5 minutes of latitude and longitude for guidance "in superimposing the segment over a base map. The degrees and minutes of these marks are printed at the margins (Figure 2-4, F). The segment, as represented in memory, is then produced on the printer. After each segment of the column has been completed for the first map, the crowded-out wells are printed. When each horizon has been completed, the program returns to the input phase and wells for the next column are read. In contrast to several weeks that might be necessary for manual plotting, the computer plotting of the data requires only a few minutes of machine time. The mapping program serves to display the relative depths of a single horizon as well as the interval between two horizons. However, visualization of the vertical relationships of more than two horizons requires another type of display — the cross section or fence diagram. Two types of sections are commonly constructed, structure sections which show attitude of the formation boundaries with respect to sea level and stratigraphic sections

11 to

Figure 2—5. Cross section index map Figure 2—6. Cross section which emphasize variations in formation thickness by using a formation surface as a datum. These two types of section are somewhat analogous to structure and isopach maps. The program to generate cross sections on the printer is similar to the mapping routine. Ten sections containing up to 20 wells each can be generated from a single pass of the mapping input tape. Control information for each section includes: identification of wells and their spacing; designated formation tops; vertical and horizontal scales; and depth interval with respect to the datum. Wells are specified in order of appearance by their sequence number, which is part of the mapping input tape record. Distances between wells can be specified in inches; however, if distances are not specified, they will be computed from the relative latitude-longitude values according to the horizontal scale. A reference sheet is printed with each section (Figure 2-5) which contains an index map and list of the wells with their sequence numbers and legal descriptions. Wells are depicted on the cross section (Figure 2-6) by vertical lines and the tops are indicated by printing a four-character abbreviation for the formation name at the appropriate position relative to the datum. Reference depths are printed at one-inch intervals along the left margin of the section. Correlation lines are drawn by hand rather than by the program as interpretation may be involved. The cross-section program as described, while still available, has not received much use in that it can not select the wells to be used in the section. As such, the saving in time of preparation is not of the same order as the posted maps. The increasing tendency to take mechanical logs on wells in the province has made a section composed of log tracings more desirable and has obsoleted the stick-section program. At the time of conversion to IBM /360 a generalized retrieval program was added. This program supplies eight standardized listings:

Location — latitude, longitude, county, township, permit no., well name, completion date, total depth, elevation, datum (GR, KB) and sample tray number. Formation — structure elevation, isopach value, depth of top Oil — oil intervals, formation at interval, flow Gas — as above for gas Water — water intervals, type of water, flow Core — core intervals, recovery and analysis Logs — log intervals, type of log and logging company Status — status and change dates, final status, initial and final lahees, completion date, cummulative oil and gas pro­ duction, abandonment date, status comments.

These listings may be specified in any- combination for any group of wells. Location data is always printed with any group of other data selected. Selection of wells can be made on the basis of latitude, longitude, county, township, completion date, formation tops, formation bases, geologist, or source of tops, in any combination. In addition to the eight standard listings provision is made to add a FORTRAN routine where the data required is not available from the standard forms. Furthermore, should the need for this new routine occur with sufficient regularity it may be easily incorporated permanently.

13 On each of the above lists a Well Index-Number is printed for each well. This number is used to cross-reference the lists to allow all well data to be correlated. The location list is always printed whether specified or not, when one or more of the other lists have been requested since it provides the identification of the well.

Map Errors In spite of the intensive cross-checking, to insure accurate OWD data, about twenty Ontario wells were misplotted due to errors of latitude or longitude and a somewhat greater number of errors in formation depth were found. The only errors which could be recognized as such were those which were not compatible with the regional grain of structure or thickness. For those which affected areas of sparse control, a correction was made by hand; in less critical areas, the erroneous values were ignored in contouring. On some of the machine-contoured maps, a different procedure was followed as the computer could not differentiate between good and bad values. A deck of cards containing the well number, latitude-longitude and the data values was punched to serve as input to the contouring program. An initial map with a coarse contour spacing was made from which corrections or deletions could be made to the card deck. The final map was then created from the corrected deck.

Preparation of Illustrations Presentation of the considerable number of large maps used in this study at anything approaching full scale was clearly impractical. Therefore, a county outline base map was superimposed on each computer-generated work map and the contours were traced in ink. A six-fold photographic reduction was then made. Inability to present the original data on which the work maps were based is an unfortunate, but necessary sacrifice.

Contribution of the Data File and Programs to the Study The computerized data file has made several contributions to this study. An indirect benefit is the upgrading of the data which was a consequence of file development. Of a more direct nature, the data file made feasible the utilization of far more data than would have been possible by hand. While all of the structure and isopach maps could have been posted by hand, as a practical matter they would not have been. The convenient availability of any map or cross section at any scale made it possible to follow avenues of speculation which arose in the course of the study. This, however, is not an unmixed blessing as-it is possible to become inundated with data. Contouring programs also fall into the category of operations which could be done by hand if sufficient man-time were allocated. Here, however, there is a difference in the product. Contouring programs are of necessity mechanical and, therefore, totally objective; whereas in the process of contouring by hand, the geologist interprets his data. Mathematical and statistical treatment of the data, such as trend surface analysis, is a procedure which could not possibly have been made by hand.

14 chapter 3

stratigraphy

Introduction The Precambrian basement rocks which underlie the Paleozoic sediments of southwestern Ontario and southeastern Michigan are exposed at the northern margin of the Paleozoic cover between Georgian Bay and the eastern end of Lake Ontario (Figure 3-1). These rocks form part of the Grenville Structural Province (Stockwell, 1965) from which the majority of radiometric age determinations are on the order of 900 to 1000 MY (Tilton et al, 1960, p. 4176). The rocks are probably as old as Archean (>2.5 BY) in this province and ages of up to 1.75 BY have been measured near the southern end of the Grenville Front. Nwachukwu et al, (1965) suggested that differences in magnetic anomaly trends could be used to extend the front from the outcrop area on the northern shore of Georgian Bay southwestward across Lake Huron into Michigan at Saginaw Bay. According to these authors, magnetic intensities and sample examination indicate a "homogeneous" basement consisting primarily of grantic rocks and metasediments. The Paleozoic sedimentary cover is a sequence of carbonates and shale with minor amounts of evaporites and sandstone (Figure 3-2). The maximum drilled thickness in Ontario is 4727 feet in a well near Sarnia. The stratigraphic nomenclature followed in this study (Figure 3-3) is that of Beards (1967). This nomenclature was subsequently revised and appears as Appendix II. Maps appearing in this chapter will have county outlines for reference. County names are shown on an index map (Figure 3-4).

Cambrian Sediments of presumed Upper Cambrian age in southwestern Ontario and southeastern Michigan can be subdivided into a lower unit of white, fine to medium grained sandstone; a middle unit of interbedded sandstones and dolostones containing minor amounts of glauconite; and an upper unit of buff to white, medium crystalline dolostone. In the area west of 81° W longitude (Figure 3-5), Sanford and Quillian (1959) applied Wisconsin stratigraphic nomenclature to these units in ascending order: Mount Simon, Eau Claire, and Trempeleau; east of the arbitrary longitude, they utilized nomenclature from New York in ascending order; Potsdam, Theresa and Little Falls. Within the study area, thickness of the combined Upper Cambrian units

15

DRIFT (PLEISTOCENE AND RECENT)

PORT LAMBTON SUNBURY PORT LAMBTON BEREA KETTLE POINT BEDFORD KETTLE POINT

HAMILTON MARCELLUS DUNDEE DETROIT RIVER LUCAS DETROIT RIVER GROUP AMHERSTBURG — SYLVANIA BOIS BLANC BASS ISLANDS BOIS BLANC SPRINGVALE ORISKANY

BASS ISLANDS G UNIT F UNIT (UP TO 3 SALT SUBUNITSI SALINA GROUP E UNIT D UNIT (SALT) C UNIT B UNIT (SALT AND / OR ANHYDRITE) • A-2 UNIT (CARBONATE) A-2 UNIT (EVAPORITE) GUELPH-LOCKPORT A-1 UNIT (CARBONATE) A-1 UNIT (EVAPORITE) CLINTON GROUP CATARACT GROUP GUELPH-LOCKPORT — (UNDIVIDED) CLINTON — DECEW ©UEENSTON — ROCHESTER — IRONDEQUOIT — REYNALES MEAFORD-DUNDAS- — NEAHGA COLLINGWOOD — THOROLD

CATARACT GRIMSBY CABOT HEAD mm TRENTON GROUP MANITOULIN WHIRLPOOL

QUEENSTON MEAFORD-DUNDAS BLACK RIVER GROUP COLLINGWOOD

COBOURG SHERMAN FALL FRAGMENTAL SHERMAN FALL ARGILLACEOUS KIRKFIELD BLACK RIVER - COBOCONK GULL RIVER : INTER BEDDED |;V;: ;| SANDSTONE fgggj SALT SHADOW LAKE : SAND a CARB. PH CARBONATE |VVQJ SHALE CAMBRIAN — (UNDIVIDED)

PRECAMBRIAN

Figure 3—2. Composite stratigraphic section, Lambton Figure 3—3. Paleozoic nomenclature chart for southern County, Ontario Ontario Figure 3—4. Index map of southwestern Ontario and southeastern Michigan

1957). The Gull River is a brown and cream coloured, lithographic limestone containing interbeds of fine crystalline dolostone (especially near the base) and scattered chert nodules. The thickness of the Gull River ranges from 75 feet to 400 feet (Sanford, 1961). The Coboconk is a buff, fine grained crystalline to crypto-crystalline limestone containing scattered chert nodules and varying in thickness from 25 to 100 feet. The overlying Trenton Group, composed of the Kirkfield, Sherman Fall, and Cobourg Formations, has a total thickness range of 300 to 500 feet (Sanford, 1961). The Kirkfield is a fragmental limestone, interbedded with shaly and, in some places, carbona­ ceous limestone. The Sherman Fall is a clastic unit of fragmental limestone with abundant shale partings and interbeds, particularly in the lower part. The Cobourg is mainly a brown to dark brown, fine crystalline limestone; in the extreme western part of the Ontario peninsula and in much of the southern peninsula of Michigan, the upper 5 to 30 feet of the Cobourg is brown sucrosic dolostone. The Upper Ordovician Collingwood, Meaford-Dundas, and Queenston Formations are primarily shales and have a total thickness of 600 to 1500 feet. The lowermost Collingwood, a dark grey and black, fissile, bituminous shale, is in sharp well-defined contact with the underlying Cobourg limestone. Reported thicknesses for the Collingwood in southwestern Ontario are in the range of 100 to 200 feet; however, as a result of the gradational upper contact with the Meaford-Dundas, the reported thickness of the Collingwood exhibits considerable local variation. The Meaford- Dundas is a grey shale with local carbonate, and sandstone interbeds. Regionally, the carbonates increase to the west and the sandstones increase to the east. For the purpose of this report, the Meaford-Dundas and

18

Collingwood Formations are treated as a single unit. The unit thickens across the study area (Figure 3-7) from 350 feet on the west to about 900 feet on the east. An apparent reversal to the trend of southeastward thickening occurs at the eastern edge of the map. The uppermost Ordovician Queenston Formation is mainly a brick red shale with some green shale partings or mottlings and scattered interbeds of limestone. The number of limestone beds increases westward, and toward the base. The Queenston, like the Meaford-Dundas-Collingwood, thickens markedly from west to east (Figure 3-8) with the direction of major thinning about N 60° W. The thickness varies from less than 200 feet on the west to over 900 feet in the eastern part of the study area. In contrast to the lower part of the Upper Ordovician, the Queenston thins at a more rapid rate to the west. Although the isopach of the Meaford-Dundas-Collingwood and the Queenston both exhibit considerable local variability, the isopach of the total Upper Ordovician (Figure 3-9) is strikingly regular, thus perhaps reflecting the inherent difficulty in determining the boundary between the Queenston and the Meaford-Dundas. The rate of thinning of the total unit is about 9 feet per mile in a N 60° W direction. West of the 700-foot isopach, the uniform thinning east to west appears to reverse; the scattered control in southeastern Michigan indicates slight thickening into the Michigan Basin.

Silurian The Silurian includes the Cataract, Clinton, Guelph-Lockport and Salina Groups and the Bass Islands Formation. The total thickness of the Cataract Group increases from around 75 feet in the northeastern part of the study area to over 150 feet to the south and southwest (Figure 3-10). The Lower Silurian Cataract Group in the study area is a carbonate and shale sequence with minor sandstone beds. Along the Niagara escarpment of the outcrop area the lowest Silurian formation is the white, fine-grained Whirlpool Sandstone which is up to 25 feet thick and lies in sharp contact on the Ordovician Queenston red shales. In the area where the Whirlpool is present, the overlying Manitoulin Dolostone is 10 to 25 feet thick. Farther west this carbonate bed lies directly on the Queenston and thickens to as much as 80 feet in western Essex County. The Cabot Head, which overlies the Manitoulin, is green, red and grey shales with shaly dolostone interbeds. It is 50 to 70 feet thick in the northeast and thickens to over 100 feet southward and westward. In the southeastern part of the study area, the Cabot Head is overlain by up to 50 feet of maroon sandstones and shales of the Grimsby Formation. The Middle Silurian includes shales, carbonates, and minor sandstones of the Clinton Group and the overlying Guelph-Lockport carbonates. The lowest formation of the Clinton Group, the thin Thorold Sandstone, is confined mainly to the extreme eastern part of the study area. Farther to the east, in the Niagara Peninsula, it is succeeded by the thin Neahga Shale. In most of the area, the basal Middle Silurian is the Reynales Formation, a buff to grey dolostone which varies in thickness from a few feet to about 25 feet. At the outcrop on the Niagara Peninsula, the Reynales is overlain disconformably by the Irondequoit Dolostone (Bolton, 1957). Westward the Irondequoit pinches out (Figure 3-11) and gives place to the overlying Rochester (Sanford, 1969a). Within the study area, the uppermost formation of the Clinton Group is the grey dolomitic Rochester Shale. This unit is over

21

60 feet thick in the eastern part of the study area and thins to 10 to 20 feet to the west and north. A salient of thick Rochester extends westward into Lambton County where thicknesses of over 50 feet are reported. The combined thickness of the Clinton Group is illustrated in Figure 3-11. East of the study area, argillaceous carbonates of the Decew Formation overlie the Rochester. North of the study area, the Clinton changes laterally to dolostones and shales of the Dyer Bay, Wingfield, St. Edmund, and Fossil Hill Formations (Bolton, 1957). The Guelph-Lockport Formation overlying the Clinton is the cliff-forming dolostone of the Niagara escarpment. This is the oldest formation exposed at the surface in the study area (Figure 1-2) and at the extreme northeastern part is overlain by Pleistocene glacial sediments. In outcrop, the Guelph and Lockport formations can be differentiated, and along the Niagara escarp­ ment, the Lockport is subdivided into the Gasport, Goat Island and Eramosa members (Bolton, 1957). The Gasport is a massive blue-grey to white, coarse to fine crystalline, crinoidal, dolomitic limestone. The overlying Goat Island is dark to light grey, dense to fine crystalline, locally argillaceous dolostone; in the Niagara peninsula the lower Goat Island contains lenses and nodules of chert and is termed the Ancaster Chert member. The Eramosa, a brown dolostone, is recognized by its high bitumin content and petroliferous odour. The Guelph Formation overlying the Lockport is a massive, buff to light grey, sucrosic dolostone. In the subsurface, workers have had difficulty recognizing the Guelph and Lockport and the Lockport subdivisions. As a result, the entire section is often grouped together as the Guelph-Lockport (Caley, 1940, p. 72). A three-part subdivision of the combined Guelph- Lockport was recognized by Evans (1950, p. 60), and many workers now utilize a three-fold breakdown (Pounder, 1963a, and Alguire, 1962). The combined Guelph-Lockport section exhibits a considerable regional variation in thickness (Figure 3-12). Across the southern and eastern parts of the study area, the section ranges in thickness from 200 to 575 feet. Within this belt, the greatest thickness is attained toward the southwest and the northeast. This belt is part of a broad regional reef complex which is thought to have surrounded the Michigan Basin during Late Middle Silurian time (Cumings and Schrock, 1928; Krumbein, Sloss and Dapples, 1949; Low- enstam, 1950). The 200-foot isopach is defined here as the margin of the reef complex. The section thins basinward to the north and northwest to a range of 75-150 feet as well as outward from the Michigan Basin (Cumings and Schrock, 1928). Southwest of Lake St. Clair, this trend reverses and the section begins to thin in the southwestern part of the study area. In the areas of thin Guelph-Lockport, pinnacle reefs are found within the upper units. Pinnacle reefs are isolated bioherms defined by Shouldice (1955, p. 500) as "structures having a height of over 250 feet and a length of less than two miles". Pinnacle reefs are discussed in more detail in a subsequent chapter. Sanford (1969a), in an extensive study of the Silurian of southwestern Ontario, has proposed that the Gasport and Goat Island be raised from member to formation status and included as part of a group which he calls Amabel. The Amabel Group is similar to the Lockport Formation of Bolton (op. cit), except that it includes the underlying Rochester and Irondequoit and excludes the overlying Eramosa member which is made a member of the Guelph. In the area of thin Guelph-Lockport, the Gasport, Goat Island, and Guelph equate to the units of the three-part subdivision. The Upper Silurian is represented by the Salina and Bass Islands

23

Formations. The entire section was subdivided by Landes (1945) into eight units A through H in ascending order, with the uppermost H unit representing the Bass Islands. The letter designation for the Bass Islands is seldom used, however. Evans (1950, p. 59) modified Landes' classification by further subdividing the lower A unit (see Figure 3-3). A west-east section (Figure 3-13) illustrates the lithologies of these subdivisions of the Salina. The section consists of carbonates, halite, anhydrite, and dolomitic shale. The Salina is best known from studies of well samples, for there are almost no surface exposures. The section is absent in the northeastern corner of the study area, presumably due to erosion, but is about 350 feet thick along the eastern edge of the area (Figure 3-14). From the 500-foot isopach, the formation thickens markedly into the Michigan Basin to over 2200 feet on the west. The greatest thickening is in the salt members.

Distribution and Thickness of Salina Units The A-1 Evaporite unit includes the A-1 Salt and the overlying A-1 Anhydrite. Often a few inches to a few feet of anhydrite are found between the Guelph and the A-1 Salt; however, this bed is not reported as a separate zone. The A-1 Evaporite unit is generally restricted to the basinward area north and west from the Guelph-Lockport reef complex (Figure 3-15). Along the basin shelf only the A-1 Anhydrite is present, varying, without observable regional pattern, from a few feet to as much as 30 feet in thickness. Marked thickening of the unit at the rate of about 4 feet per mile in the northwestern part of the study area is attributed to the thickness of the A-1 Salt. Beyond the margin of the salt, the overlying anhydrite thins rapidly to a few feet. The A-1 Carbonate unit (Figure 3-16) changes laterally from a dark grey, thin bedded, argillaceous limestone to the north and west to light brown sucrosic dolostone changing to dark grey argillaceous limestone or dolostone toward the base. On the margin of the Michigan Basin it overlaps the A-1 Anhydrite and to the southeast rests directly on the Guelph. Over the Guelph reef complex (Figure 3-16) the A-1 Carbonate ranges in thickness from about 20 to 50 feet. Over the pinnacle reefs the unit can be much thinner or missing. Toward the Michigan Basin from the reef complex, the A-1 Carbonate thickens rapidly into a broad belt where thicknesses range from 100 to 130 feet. Farther to the northwest, the unit thins again to as little as 50 feet.

The A:2 Evaporite (Figure 3-17), like the A-1 Evaporite, is primarily salt with minor anhydrite beds (0 to 20 feet) at the top and bottom which coalesce and extend slielfward beyond the edge of the salt. Both of these anhydrite beds thin into the Michigan Basin and are apparently absent in the northwestern part of the study area. The A-2 Evaporite exhibits marked thickening to the northwest. The rate of thickening is not uniform but shows an increase between the 150 and 350'foot isopachs to 10 feet per mile as compared to about 4 feet per mile on either side. Because of nomenclatural differences across the international border, the individual beds cannot be mapped. In Ontario, the lower anhydrite bed is designated the A-2 Anhydrite and the upper bed is included with the A-2 Salt, while in Michigan the upper bed is called the A-2 Anhydrite. Figure 3-17 was obtained by mapping from the top of the A-2 Salt to the A-1 Carbonate in Ontario and from the top of the correlative A-2 Anhydrite in Michigan. The thickness of the A-2 Salt has been affected by leaching along its margin in southern

25

Lambton County northeast of Lake St. Clair (Sanford, 1965, p. 3). It is also absent over the pinnacle reefs. Where the salt is absent over the pinnacles, the combined thickness of the anhydrite beds increases to as much as 50 feet. The A-2 Carbonate (Figure 3-18) resembles the A-1 Carbonate in that it is commonly a brown limestone and/or dolostone in the upper portion, changing to grey or brown argillaceous limestone toward the base. The A-2 Carbonate is also thinnest over the reef complex. Thicknesses range from about 50 to 75 feet over the reef complex in the northeast part of the map area and from 60 to 100 feet in the southwest part. Into the Michigan Basin, thicknesses increase to between 150 to 160 feet; then, unlike the A-1 Carbonate, remain at this level northwestward. The B Unit (Figure 3-19) is predominantly salt with some thin interbeds of anhydrite and dolostone; it is the only salt which extends over the reef complex. Along the southeastern margin of the unit, the distribution of the salt is irregular, presumably due to solution (Evans, 1950, p. 71; Grieve, 1955; Sanford, 1965, p. 4). Although these irregularities tend to mask the regional picture, the B Unit does thin over the reef complex. Beyond the margin of the reef complex, the unit, where present, ranges between 150 to 200 feet in thickness. Northwestward in the Michigan Basin, it thickens irregularly to around 300 feet and then, unlike the A-2 and A-1 Salts, ceases to thicken. In Essex County and the western part of Kent County in Ontario, isolated occurrences of B Salt are reported over the reef complex. A similar situation is observed in Wayne, Monroe, and Washtenaw counties in Michigan. The C Unit is a grey to greenish grey, dolomitic shale containing rounded quartz sand grains. It is present throughout the map area with the exception of the extreme northeast part. Reported thicknesses vary locally between 60 and 120 feet, but no regional trends are evident. The D Unit is composed of two salt beds separated by a thin but markedly persistent dolostone. Its areal distribution is very similar to that of the A-2 Salt and, where unleached, the unit varies in thickness locally from 30 to 60 feet. Unlike the A-2 Salt, the D Unit does not exhibit any regional thickening. The E Unit carbonate, present throughout the map area, has a uniform regional thickness which varies locally between 80 and 120 feet. Throughout much of the study area, the F Unit is shale and shaly dolostone with a uniform regional thickness of about 100 feet. In the northwest, up to six salt beds occur within the section which increase the total thickness of the F Unit in that area to over 600 feet (see Figure 3-13). The G Unit, the uppermost unit of the Salina, is composed of 20 to 30 feet of buff, dense dolostone overlain by 15 to 30 feet of grey argillaceous dolostone. The Bass Islands, the topmost formation of the Silurian, overlies the Salina and is commonly a buff to grey-buff, dense dolostone. In most of the study area, the unconformable upper surface of the Bass Islands is overlain by the uppermost Lower Devonian Bois Blanc, although in other areas, the Lower Devonian Oriskany or the Middle Devonian Detroit River may overlie. There is lack of agreement concerning the lower contact of the Bass Islands with the Salina; some geologists have included the G Unit beds with the Bass Islands. As a result, discrepancies of as much as 60 feet in the thickness can be assigned to the Bass Islands.

27

The Bass Islands Formation, like the Salina, thickens into the Michigan Basin, increasing from zero at its erosional edge in the northeastern part of the study area to over 350 feet in the northwestern part (Figure 3-20). To the northeast, the thickness ranges from 50 to 100 feet over the Algonquin Arch; the unit thickens northwestward to as much as 200 feet at the southern end of Lake Huron. In Essex and Kent counties in Ontario, the thickness is extremely variable, generally between 150 and 200 feet, but locally as great as 450 feet. The areas of local thickening, where sufficiently defined by drilling to establish a trend, usually trend east-west, north-south, and northwest-southeast. Along the margin of the Michigan Basin, the rate of thickening appears to increase markedly in the area of the 200 and 300 foot isopachs.

Devonian The Lower Devonian in Ontario is represented by the Oriskany and Bois Blanc Formations. The Oriskany sandstone occurs only in the Niagara Peninsula as scattered outliers within the Appalachian Basin. The sand is friable, white to yellowish in colour, and contains rounded quartz grains, up to 1/8 inch in diameter (Winder, 1961, p. 86). The Oriskany disconformably overlies the Bass Islands and is disconformably overlain by the Bois Blanc. Nowhere in Ontario is the Oriskany more than 20 feet thick (Sanford, 1967, p. 979). The Bois Blanc Formation (Figure 3-21) is composed of interbedded limestones and/or dolostones, cherts and cherty limestones. In the Niagara Peninsula, the Springvale basal sandstone member is differentiated from the underlying Oriskany on the basis of faunal evidence (Best, 1953). The thickness trends of the Bois Blanc are similar to those of the Bass Islands; throughout much of southwestern Ontario, the average thickness is 100 to 150 feet. Areas of greater thickness occur locally; the most pronounced of these is an irregular, east-west trending belt along the Lambton-Kent county line where thicknesses range from 200 to 300 feet. In the westernmost part of Lake Erie and the extreme southeastern part of Michigan, the Bois Blanc is missing; according to Sanford (1967, p. 980), it has "been overlapped disconformably by early Middle Devonian dolomites or sandstones of the Amherstburg Formation". The position of the contact between the Bois Blanc and overlying Amherstburg, except where the Sylvania Sandstone intervenes, is difficult to determine from well cuttings. As a result, there is considerable variability of reported thicknesses. The Middle Devonian consists of the Detroit River Group and the Dundee and Hamilton Formations. "The Detroit River Group is subdivided into the Amherstburg and Lucas Formations. Locally, the basal Sylvania sandstone member of the Amherstburg is composed of large, rounded, frosted quartz grains (Landes, 1951; Reavely and Winder, 1961). This unit is restricted to the western part of the study area (Figure 3-22) and is generally less than 150 feet thick except in isolated areas, such as Wayne and Washtenaw counties in Michigan where thicknesses as great as 360 feet have been observed. The Amherstburg is characteristically a "dark brown, organic- looking carbonate (dolomite and/or limestone) with bituminous partings" (Beards, 1967, p. 10). In outcrop along the east shore of Lake Huron, numerous bioherms have been assigned by Best (1953, p. 103) to the Amherstburg and designated as the Formosa Reef Fades, although Fager- strom (1961, p. 348) states that this correlation is "quite uncertain". The

31

overlying Lucas Formation is lighter in colour and includes interbedded anhydrite. Evaporites within the Lucas become more abundant toward the Michigan Basin where as much as 1000 feet of cumulative salt and anhydrite are reported in north-central Michigan (Briggs, 1959, p. 46). The necessary restriction for an evaporite basin may have been marginal banks of the Amherstburg carbonates (Briggs, 1959, p. 46). The thickness for the combined Detroit River Group (Figure 3-23) varies from zero at the southeast to over 1000 feet in the extreme northwestern part of the study area. The rate of thickening is 5 to 6 feet per mile around the basin margin and about 10 feet per mile in northern St. Clair County, Michigan. The Dundee, formerly named Delaware or Dundee-Delaware (Beards, 1967), is buff to grey, often fragmental, fine grained limestone with scattered amber spore cases and occasional chert nodules. The Dundee lies immediately below the glacial drift in a wide band which extends across southwestern Ontario from Grand Bend to Long Point. This band makes up the bedrock in much of Essex County at the northeast end of the Findlay Arch. In the central part of the study area, where the Dundee is uneroded, reported thicknesses vary from 60 to 120 feet (Figure 3-24). The formation thickens southeastward to as much as 190 feet and northwestward to around 300 feet. Considerable variability in reported thicknesses of the Dundee is due mainly to difficulty in picking the Dundee-Detroit River contact. The lower part of the Hamilton is grey calcareous shale and limestone and does not occur in outcrop; the upper part is limestone interbedded with a grey calcareous shale. In Ontario, a northwest-trending outcrop belt of the Hamilton runs through western Kent County from Lake Erie to Lake St. Clair; a second belt extends through western Elgin and Middlesex counties into northern Lambton County from Lake Erie to Lake Huron, northeast of Kettle Point. Where a complete section can be observed, the Hamilton varies from about 150 feet on the south to as much as 650 feet at the north edge of the study area (Figure 3-25). The Upper Devonian (Winder, 1966) is represented by the Kettle Point and Port Lambton Formations. The Kettle Point unconformably overlies the Hamilton and forms the bedrock throughout much of Kent and Lambton counties. It is a thin-bedded, dark brown to black shale, with some hard greenish grey silty shale, commonly containing amber-coloured spore cases and small pyrite concretions. Spherical limestone concretions, varying in size from 8 inches to nearly 4 feet, occur near the base of the formation at Kettle Point, the type locality on Lake Huron. The name of the type locality stems from the resemblance of the partially exposed concretions to overturned kettles. An uneroded section of Kettle Point thickens northwestward in southeastern Michigan from about 160 to over 300 feet (Figure 3-26). The Port Lambton is the youngest Paleozoic formation in southwestern Ontario and occurs only in a limited portion of Moore Township in Lambton County (see Figure 1-2). The formation is equivalent to the Bedford Shale, Berea Sandstone, and Sunbury Shale of Michigan (Beards, 1967, Chart 1). Although younger formations occur within the Michigan portion of the thesis area, no provision was made for them in the mapping format; therefore they were not mapped.

Discussion and Summary Apart from the veneer of Pleistocene glacial deposits, the sedimentary cover in southwestern Ontario and southeastern Michigan is Lower Paleozoic in

35 age. The formations mapped range in age from Upper Cambrian through Upper Devonian. Sediments representing the early part of each period are either absent (Lower and Middle Cambrian, Lower Ordovician) or constitute a small percentage of the section (Silurian, Devonian). The rocks are essentially an alternating sequence of blanket-like carbonates and shales. Exceptions to this generalization are: the distribution of the Upper Cambrian sediments around the Algonquin Arch, the salients in the Guelph-Lockport and Bass Islands, the minor occurrences of sandstones and the Salina evaporites. The most pronounced trends of thickening are to the southeast in Upper Ordovician rocks and to the northwest from Upper Silurian through Devonian.

36 chapter 4

regional structure

Precambrian Surface The four major structural features of southwestern Ontario — the Michigan Basin, Appalachian Basin, Algonquin Arch, and Findlay Arch (Figure 4-1) — are readily apparent from contours on the Precambrian basement. The definition of the arches and basins is perhaps a matter of comprehension as the features are, in fact, contiguous. The structure at the level of this major unconformity is quite regular with a gentle gradient ranging from 17 to 85 feet per mile. The maximum regional dip is less than one degree. The broad Algonquin Arch gains definition southwestward, particularly where the nose is deflected westward by the east-west trending Electric Fault just past the northeastern end of the more sharply defined Findlay Arch. Control points on the Precambrian surface, although scattered, define a lack of symmetry between the Michigan Basin and Appalachian Basin sides of the Algonquin-Findlay axis. The structure contours along the margin of the Michigan Basin have a smooth arcuate pattern, broken only by the Electric Fault, in contrast to a triangular pattern for those along the Appalachian Basin, with the apex at the centre of the Chatham Sag, which is not apparent northwest of the fault. Dips on the various sediment surfaces over the structural features of southwestern Ontario all increase basinward. To compare the rate of dip into the Michigan and Appalachian Basins, key deep wells were selected for control, as shown on Figure 4-2. Due to lack of deep control, dips were calculated only for the Michigan Basin flank of the Findlay Arch (C-C). To illustrate dips into the Chatham Sag, two pairs of wells were selected along the Algonquin-Findlay Axis (D-D' and E-E').The slopes between the control wells (in feet per mile) for the Precambrian, Trenton, Queenston, Clinton, Guelph-Lockport, Bass Islands, and Detroit River surfaces are summarized in Table 4-1. The Precambrian surface dips uniformly from the Algonquin Arch into the Michigan Basin (31 feet per mile along A-A') and into the Appalachian Basin (32 feet per mile along B-B'). Within the sag the gradient is much steeper toward the Michigan Basin than toward the Appalachian Basin. Dips into the Michigan Basin are considerably steeper on the northwestern flank of the Findlay Arch (85 feet per mile along C-C). However, they are steeper only in a relative sense. The plunges of the Findlay and Algonquin Arches to

37 the Chatham Sag are much less in comparison (17 feet per mile along D-D' and 19 feet per mile along E-E'). Several displacements of the Precambrian surface suggest faulting. The most obvious are the Electric Fault with a maximum observed vertical displacement of 305 feet and the smaller scale displacement of the -3400 foot contour (see Figure 4-1) observed in the area of the Clearville Field (Cambrian oil production). A similar displacement of the -2900 foot contour occurs in the Willey Field (also Cambrian oil). A northwest displacement of the -3500 and -3600 foot contours can be observed south of the Kimball-Colinville Field (Silurian reef).

Trenton Surface The structural configuration of the Middle Ordovician Trenton surface (Figure 4-3) is quite similar to that of the Precambrian; however, dips on the Trenton are reduced due to increasing thicknesses of Upper Cambrian sections into both basins and of the Middle Ordovician toward the Michigan Basin. The slope of the Algonquin Arch into the Appalachian Basin (B-B', Figure 4-2) is only 24 feet per mile between the control wells because of an additional 216 feet of Upper Cambrian section. Similarly, the slope of the arch into the Michigan Basin is somewhat less (28 feet per mile in contrast to 31 feet per mile on the Precambrian) due to an increase of 84 feet in the thickness of the Trenton-Black River and of 12 feet of Upper Cambrian. The configuration of the Findlay Arch is broader on the Trenton surface than on the Precambrian because of increased thickness of the Upper Cambrian to the southwest. More structural features are apparent on the Trenton contours due to an increased number of control points. An east-west, elongate, closed low is evident in the area of the Dover Field (longitude 82° 20' and latitude 42° 20') which produces both oil and gas from the Trenton-Black River. This synclinal feature has negative relief of about 150 feet on the Trenton. West of the Tilbury Field (Silurian gas production) there is a closure of the -2100 foot contour (longitude 82° 18' and latitude 42° 09'). At the far western edge of the study area in Michigan, the pronounced Howell-Northville anticline is apparent on the Trenton surface. This feature has a maximum relief of about 1000 feet, and is sharply asymmetrical on its southwest flank due to normal faulting (Kilbourne, 1948). Gas and oil are obtained from the Trenton-Black River, where it has been dolomitized along the Howell-Northville anticline. West of Lake Huron the Trenton appears to be displaced downward on the west side of a north-south trending fault or monocline by about 300 feet. This feature is herein referred to as the Peck Fault.

Queenston Surface As a result of the southeastward thickening of the Upper Ordovician, the dip on the Queenston surface from the • Algonquin Arch into the Appalachian Basin is reduced from 24 feet per mile on the Trenton surface to 21 feet per mile (B-B', Figure 4-2). The dip into the Michigan Basin (A-A') is increased from 28 feet per mile to 34 feet per mile. The thickness of the Upper Ordovician (Figure 3-9) exerts little to no influence on the dip of the Queenston surface from the Findlay Arch into the Michigan Basin (C-C) as it is fairly uniform in this area. The trend of the Upper Ordovician isopachs is slightly counterclockwise to the alignment of the control wells on the Algonquin and Findlay Arches and this affects the dips of the arches into the

39

Dundee Surface Structure contours on the top of the Middle Devonian Dundee (Figure 4-7) are more regular within the Chatham Sag than those on the Bass Islands with fewer closed lows observable at a regional scale. Dips along the Algonquin and Findlay Arches into the Chatham Sag appear to be very slight; however, at the Dundee-Detroit River contact, the dips are only 1 feet per mile less between the control wells than on the Bass Islands. The base of the Dundee was used for the slope measurement because of the limited extent of uneroded Dundee on the arches. Dip into the Michigan Basin is 3 feet per mile less from the Algonquin Arch. The rates of dip could not be determined

Figure 4—7. Structure contours on Dundee Formation

for the Findlay Arch or for the Appalachian Basin side of the Algonquin Arch due to erosion of the Detroit River.

Pre-Pleistocene Erosional Surface The surface at the base of the Pleistocene Drift (Figure 4-8) exhibits the most pronounced erosional unconformity except for the Precambrian. At no other level have so many surfaces been truncated. There are two large-scale positive features, one to the northeast which approximates the position of the Algonquin Arch and another to the west which trends northeast. The

44 Figure 4—8. Topography on base of Pleistocence present course of the St. Clair River is in the low area between these highs. Along the north shore of Lake Erie are two elongate low areas. The one to the east is more pronounced and has apparently influenced the subcrop pattern (Figure 1-2). South of Norfolk County this low has cut through the Hamilton and has displaced the Dundee subcrop southward. In western Elgin County, it has exposed an outlier of Dundee within the Hamilton. At the western end of the low, erosion has caused a finger of Hamilton subcrop to extend into the Kettle Point.

Discussion and Summary The regional features of the Precambrian surface are extremely subtle, with most dips in the order of a third of a degree or less. The very broad Algonquin Arch has comparable dips into both the Michigan Basin and the Appalachian Basin. The more sharply defined Findlay Arch dips more steeply into the Michigan Basin. The Electric, Clearville, and Willey Faults as well as the Kimball-Colinville monocline are apparent on the Precambrian surface. On the Trenton, the additional features which become apparent because of the increased number of'control points are the Dover syncline, the Peck Fault, and the Howell-Northville anticline. The rates of dip into the basins are less, due to the increasing thickness of Cambrian sediments into each basin and of Middle Ordovician rocks into the Michigan Basin. The southeast thickening wedge of Upper Ordovician sediments has the effect of decreasing east dips and increasing west dips on the top of the Queenston. As a result, the axes of the arches, as expressed by contours on the Clinton, are shifted to the southeast of the Trenton axes.

45 Although not directly related to tectonics, the thick Guelph-Lockport on the arches has steepened the dips as indicated by the control wells. In addition, the contours show that the central point of the Chatham Sag is southeast of its position on the Clinton. Addition of the wedge of Salina sediments has reduced dips into the Michigan Basin on the Bass Islands. The arches appear somewhat broader due to the lessened dips. The limited extent of uneroded Dundee on the arches makes comparison with previous surfaces difficult; however, the regional structural features appear to be even more gentle. Examination of the structure of the various surfaces from older to younger indicates that initially steep dips become progressively less steep on younger horizons; however, this is inaccurate in an historical sense, because the contours of each horizon represent the present structural configuration. Thus, the present attitude of each horizon is the result of its initial attitude plus the net effect of all movement subsequent to its deposition. The development of these present structures, both regional and subregional, is the subject of the following two chapters.

46 chapter 5

historical development of regional structure

The regional structural framework of southwestern Ontario and southeastern Michigan is basically quite simple. This simplicity was described by William Logan (1854) who wrote that ". .. between the Michigan and Appalachian coalfields there is a flat anticlinal arch, the axis of which runs from Lake Ontario to the western end of Lake Erie, and that between the town of Chatham and the village of Zone there is in it a slight transverse depression". However, despite the simplicity, opinions on the historical development of the main features of the area have remained quite diverse. Lockett (1947) considered both the Algonquin and Findlay Arches to be a northeast continuation of the Cincinnati Arch. He postulated that the arch is a buried remnant of an early mountain chain which extended from the Canadian Shield through western Ontario and southward into Ohio. According to his hypothesis, the Chatham Sag was pulled down by the weight of sediments which had accumulated in the Appalachian and Michigan Basins. Lockett stated that the weight exerted where the two basins were separated only by the old positive arch would have been sufficient to initiate and sustain down warp across the axis. Green (1957) considered the Findlay and Algonquin Arches to be genetically related. He stated (p. 637), "It seems permissible to think of the rims of a subsiding basin as having been at lines which marked the outer influence of the down-pull toward that basin. The relatively high structure between the three basins has resulted from resistance to the subsidence in the bordering basins rather than from uplift in the area between the basins." Kay (1942, p. 1621) considered the arches to be tectonically related. He stated that "the Cincinnati Arch is first clearly indicated in the Clinton. This event is about synchronous with that in Ohio, where Albion sediments persist but are thin on the arch whereas the Clinton offlaps (Foerst, 1935). That the arch was not present in the Ordovician has been indicated from the outcrop and isopachs; the Chatham Sag was suggested by Mohawkian isopachs, however." B. V. Sanford, personal communication 1971, discounts the belief that the arches are tectonically related. He feels that the Findlay Arch was raised independently of the Algonquin Arch during the Late Middle and Upper Devonian times.

47 Thus, the views expressed on the tectonic origins of the structure of southwestern Ontario and southeastern Michigan vary widely, both in time and in form, i.e. (1) from early to late, (2) one arch or two, (3) pushed up or left "up" by adjoining subsidence, and (4) a suggestion that the sagging preceded the arching. The history of tectonic movements in this area is recorded both in the rocks and by absence of the rocks. Barrell (1917) pointed out that the accumulation of sediment is related to subsidence below base level. Accordingly, the thickening of a marine rock unit is interpreted to indicate greater relative subsidence in the direction of thickening. One can argue that the same thickness distribution could result from uplift and thinning by erosion. However, to account in this way for the regional thicknesses of various units mapped in the study area (Chapter 3), invokes the postulation of repeated regional subsidence with intervening uplift of the arches.

Cambrian The thickness distribution of the Upper Cambrian (Figure 3-5), the earliest sedimentary record in southwestern Ontario, cannot be accounted for solely in terms of depositional thickening. Kay and Colbert (1965, p. 148) maintain that intense erosion took place at the end of Early Ordovician time. The absence of Upper Cambrian sediments from the northeastern portion of the Algonquin-Findlay arches, whether by erosion or non-deposition, indicates a topographic structure had formed by Middle Ordovician time (at the latest) with dips to the northwest into the Michigan Basin and to the southeast into the Appalachian Basin. Sanford and Quillian (1959, p. 10) report a transgressive overlap of the Upper Cambrian units onto the present Algonquin Arch, indicating presence of the arch during Late Cambrian time. Evidence of an even earlier age for the arch is provided by the northeast- southwest alignment of local highs on the Precambrian surface (Sutterlin and Brigham, 1967). Thinning of the Upper Cambrian over these highs suggests that they existed as erosional features prior to deposition of the Cambrian. If the distribution of Cambrian sediments reflects the position of the early Algonquin-Findlay arches, there has been migration of the axis to the southeast to its present position (Figure 4-1). No influence on the thickness of the Cambrian sediments by the Chatham Sag can be observed.

Ordovician Thickening of the Middle Ordovician Trenton-Black River toward the northwestern part of the study area (Figure 3-6) suggests mild downwarping of the Michigan Basin during this time. The relatively uniform thickness of the section over the Algonquin-Findlay arches indicates a subsidence in response to the broadened downwarping of the Michigan Basin. The slight thickening of the section in the area of the Chatham Sag may, indeed, be evidence of the early development of the sag. The marked thickening of the wedge of Upper Ordovician terrigenous and marine sediments to the southeast (Figure 3-9) is indicative of considerable subsidence in the Appalachian Basin in Late Ordovician time. The wedge thickens to a much lesser degree in the Michigan Basin to the northwest. The axis between these two areas of subsidence lies northwest of the present Algonquin-Findlay arches, trending north 20 degrees. The isopachs of the

48 Upper Ordovician reveal no bowing to suggest greater subsidence in the area of the Chatham Sag.

Silurian Facies changes in the Lower Silurian and lower part of the Middle Silurian have led numerous workers (Schuchert, 1911; Cumings, 1939; Kay, 1942, Freeman, 1951) to postulate formation of an arch along the northwestern margin of the Appalachian Basin during Early Silurian time. However, the word "arch" denotes an elongate structure with slopes downward in two directions from a central axis; but, the evidence of uniform formation thicknesses for both the Lower Silurian Cataract Group and the Middle Silurian Clinton Group does not suggest such a structure. Rather, distance from source seems a more plausible explanation for the observed facies changes in these sections. Thinning of the Middle Silurian Guelph-Lockport into the Michigan Basin (Figure 3-12) led Cohee (1948) to postulate an uplift of the central part of the Michigan Basin, either during the time that carbonates were being deposited in the surrounding area, resulting in nondeposition over the uplift, or at a later time which would have resulted in erosion. Ehlers and Kesling (1962, p. 16) proposed the name mid-Michigan Ridge for a "conspicuous" positive area across the southern peninsula of Michigan which "only during the latter part of Guelph time, subsided to receive marine deposits (Guelph) upon its surface". However, the rapid thinning of the Guelph-Lockport over a distance of a few miles in the study area is inconsistent with thinning due to "uplift" on a regional scale. Within the map area, the Upper Silurian Salina thickens into the Michigan Basin from about 350 feet to at least 2286 feet (Figure 3-14). Assuming that deposition kept pace with subsidence, the basin area subsided at least 1900 feet during deposition of the Salina. As the cross section shows (Figure 3-9), nearly all of the basinward thickening of the Salina is in the salt sections. Sloss (1953, p. 154) stated, "it may be shown that the basinal subsidence of the province concerned tends to vary directly as the degree of environmental restriction." The reason given (p. 158) is that "the greater the tectonic differentiation the greater the effectiveness of the sill and the more extreme is the evaporitic environment." While the cross section supports the inter-relationship of subsidence and restriction, it does not illustrate conclusively why more rapid subsidence in the basin than in the sill area should bring about "greater effectiveness of the sill." Subsidence during Late Silurian time contributed a dip of approximately 20 feet per mile to the slope of the Precambrian surface into the Michigan Basin (Table 4-1). Across most of the map area, thickness trends for the Upper Silurian are parallel but opposite those of the Upper Ordovician, i.e. the Upper Silurian thickens into the Michigan Basin whereas the Upper Ordovician thickens into the Appalachian Basin. Thus, Late Silurian subsidence depressed the northwestern flank of the Algonquin-Findlay arches and contributed to the present axis of the arching.

Devonian The sparse representation of Lower Devonian sediments in southwestern Ontario and southeastern Michigan indicates uplift of the entire area following the Silurian. Commencing with the latest Early Devonian (Bois Blanc) time and accelerating in early mid-Devonian (Detroit River) time, the

49 Michigan Basin flank of the Algonquin-Findlay arches became steepened. The Dundee sediments overlying the Detroit River thicken into the Appalachian Basin as well as the Michigan Basin (Figure 3-24). The Dundee isopachs show that the position of the Algonquin Arch axis was about 25 miles northwest of the present position indicated on the top of the Dundee (Figure 4-7). Subsidence of the Michigan Basin continued through Middle Devonian time. An eastward shift of the centre of subsidence is suggested by the northerly shift in direction of thickening of the uppermost Middle Devonian Hamilton (Figure 3-25). The thickness of the overlying Kettle Point (Figure 3-26) indicates a decrease in subsidence in the Michigan Basin during Late Devonian time. There was, however, increased subsidence to the southeast in the Appalachian Basin. Sanford (1967, p. 993) states that the thickness of the Kettle Point increased to over 1000 feet at the southern shore of Lake Erie.

Discussion From the foregoing, it is apparent that the present-day Algonquin-Findlay arch structure is the result of alternating subsidence on opposite sides of the arches. This relationship is illustrated by a cross section constructed at right angles to the strike of the Precambrian surface (Figure 5-1). The thinning of the Upper Ordovician extends to the northwest beyond the present axis of the Algonquin Arch; the marked eastward thinning of the Upper Silurian stops short of the axis. The influence of subsidence during the Middle Devonian cannot be demonstrated due to erosion. Dips of 17 to 19 feet per mile on the Detroit River (Table 4-1) indicate that more than half of the present dip on the Precambrian is the result of tectonic activity which occurred after Middle Devonian time. Subsidence on either side of a medial area is not sufficient explanation of the development of the Chatham Sag. As discussed above, the Sag may have been initiated during the Middle Ordovician. A southwest-northeast cross section (Figure 5-2) along the Algonquin-Findlay arches shows little sedimentary thickening within the Sag except for the Middle Ordovician section and the Upper Silurian sediments. This relatively late development of the Chatham Sag can be illustrated by a series of cumulative isopachs. The Upper Ordovician Queenston to Trenton isopach interval (Figure 5-3) indicates increased subsidence into the Appalachian Basin and, to a lesser extent, to the northwest into the Michigan Basin. The medial area is in southeastern Michigan, -northwest of the present-day axis of the Algonquin- Findlay arches. As previously pointed out, the isopachs do not bow to suggest greater subsidence in the Chatham Sag area. Extension of the cumulative isopach to the Guelph-Lockport to include the Lower and Middle Silurian (Figure 5-4) shows some change in the regional gradient at the margin of the Guelph-Lockport barrier reef, but the regional picture remains essentially the same. This map shows most of the subsidence is to the southeast toward the Appalachian Basin and the arch, if one can be said to exist, lies in southeastern Michigan.

50 NW SE 10 12

LZn M. ORD.

u 0RD INDEX MAP u. SIL. E3 -

Figure 5-1. Cross section across Algonquin Arch at right angles to strike of Precambrian

Inclusion of the Upper Silurian Salina A-2 and A-l units effects a marked change in the cumulative isopach (Figure 5-5). This map reflects the increased subsidence of the Michigan Basin during the Upper Silurian. The apparent axis between the two areas of subsidence has shifted to the southeast. The increased gradient caused by the Guelph-Lockport reefing has been erased by compensation with the A-2 and A-l units. The thin area shown by the 13 00-foot isopach and a similar configuration to the northeast, slightly beyond the map boundary, are the first indications of increased sediment accumulation in the area of the present Chatham Sag. Extension of the cumulative isopach to the top of the uppermost Silurian Bass Islands Formation (Figure 5-6) reveals the continued subsidence of the Michigan Basin. The apparent Algonquin-Findlay Arch has moved still farther southeast toward its present-day position. Bowing of the isopachs on both the Michigan Basin and Appalachian Basin sides of the arch suggests the presence of the Chatham Sag at that point in time. However, sedimentary thickening along the arches toward the centre of the Sag is only 100 to 200 feet. The thin areas within the Sag are due to the absence of the Salina B Unit salt. Two other features are illustrated by this cumulative isopach. The first is in the Howell-Northville structure where the trend of thickening changes sharply to the northeast and the gradient is greatly steepened. The second is the thinning on the southeastern and northeastern extremities of the arching.

Figure 5—3. Isopachous map of Queenston to Trenton

53

Due to truncation by erosion, the top of the Middle Devonian Detroit River Formation is the highest level to which the cumulative isopach can be extended on a meaningful regional basis. The Detroit River to Trenton isopach (Figure 5-7) shows that the apparent axis between the two basins has moved still farther southeast to a position nearly coincident with the present-day structural axis. There is still only about 200 feet of thickening into the Chatham Sag along the axis. The structure and cumulative isopach maps of the study area lead to the conclusion that the present-day Algonquin-Findlay arched structure is the passive result of subsidence on either side. It has been shown that the opposing flanks were pulled down during differing periods of time. Subsidence to the southeast predominated during Upper Ordovician and to the northwest during Upper Silurian and Middle Devonian. The axis of the intervening arch as expressed in the cumulative isopachs is seen to migrate southeastward in response to encroachment of the Michigan Basin. The Chatham Sag, although first indicated by the Trenton-Black River interval, does not appear to have had a significant influence on sedimentation until Upper Silurian, and most of the development of this feature occurred later than Silurian. The Algonquin and Findlay Arches are, in reality, two ends of the same feature; the less negative area between the Michigan and Appalachian Basins.

56 chapter 6

development of subregional structures

Investigations of the structural attitudes and thicknesses of the formations on a regional scale have been reported in previous chapters. The scale of the regional maps (4 miles = 1 inch) adequately reveals the presence of faults and folds in the study area, but the resolution is insufficient for analysis of the development of these sub-regional structures. Accordingly, a limited section of the study area was selected for mapping at a scale of 1 mile = 1 inch. The index area (outlined on Figure 6-1) which is bounded by 42° 25' to 42° 55' latitude and 81° 35' to 82° 35' longitude includes the southwest end of the Algonquin Arch and a portion of the Chatham Sag. The significant sub-regional structural features in the index area are the Electric Fault, the Dawn structure and the Kimball-Colinville monocline. Inasmuch as this area is entirely within Ontario, it was decided not to edit the data as an illustration of the effectiveness of initial editing of the Ontario well file. Three types of maps were constructed: (1) structure maps to outline the subregional structures, (2) trend-surface residual maps to enhance subtle structures, and (3) successive isopach maps to date the structural develop­ ment. Here again, the assumption was made that a change in the elevation of one area of a oppositional surface relative to another will result in a difference in the amount of sediment accumulated over a given interval of time. Structure and isopach maps, both familiar tools, have been adequately illustrated in earlier chapters. Trend surface analysis is a somewhat more complex means of- illustrating regional variations and merits some explana­ tion.

Trend Surface Analysis Trend surface analysis is a mathematical method of analyzing map data to distinguish between regional and local effects. A basic assumption of trend analysis is that a regional trend exists which can be approximated by a smooth surface; and, further, that this surface can be approximated by a mathematical equation. With the equation, each datum or observation can be separated into two components, a trend value and a residual value. The trend value is the expected height of the surface at the location of the observation and the residual value is the difference between the expected and observed values. Because of the extreme number of computations involved, trend

57

surface analysis is practical only with the aid of a computer. The equations used to represent the smooth regional surfaces are known as polynomial equations. Polynomial equations are identified by the highest power, or order, of their exponents; in other words, a first-order equation contains no term with an exponent greater than 1, a second-order equation has no exponent greater than 2, and so on. The maximum number of gradient reversals which can be accommodated by a polynomial surface will be one less than the order of the equation; the surface which corresponds to a first-order polynomial will be a plane; the second-order surface can contain a single curvature, either a high or a low; the third-order, a high and a low, et cetera. The positions and shapes of the highs and lows on the computed surface will be a function of the order of the polynomial equation and the distribution of the data. According to Krumbein (1956, p. 2165), a contour-type map represents the simultaneous effect of widespread and local controls. If a polynomial equation can provide a suitable model for the widespread effects, these can be removed and the local effects may be examined free of masking by large scale variations. Before a person can make valid use of trend surface analysis, he must determine which, if any, of the orders of polynomial equations provides a reasonable geologic model of the regional trend. This decision may be based on experience with the trend surface technique and with the data; also there are statistical tests which help to indicate the appropriateness of the surface. In this study, first- through fourth-order surfaces were fitted to the data and the decision of which model to use was based on a statistical summary provided by the program. The "goodness of fit" of each surface is indicated by summing the squares of the differences between each observed data point and the computed value of the surface at the location of the point. This value is called the "residual sum of squares". The square of the difference is used since the surface is positioned through the data in such a way that there is as much positive departure as there is negative departure and the sum of the residuals is zero. As a starting point of comparison, the mean value of all data points is determined and the sum of squares of deviation about that mean is computed. For successive surface orders the residual sum of squares is calculated and is compared as a percentage reduction from the initial sum of squares about the mean and from that of the previous order. The percent reduction will increase as additional features can be fitted by the increasing complexity of the surfaces. However, after the major features have been accounted for, the increase in percentage reduction will commonly become much less. In most cases, the best residual surface to map for interpretation is the highest one which shows a significant increase in percentage reduction. In summary, the value of trend analysis is to expose local anomalies which might otherwise remain masked by-the regional structure. Conversely, it may suggest regional trends which have been hidden by a complexity of local anomalies.

Clinton Structure The Clinton Formation (Figure 6-2) is the lowest horizon for which control is adequate to define the structures. The overlying Guelph-Lockport is an oil-producing horizon and its thickness is considered an indication of the proximity of reef structures; therefore, operators in the area often drill

59 exploratory wells through the Guelph-Lockport into the Clinton. Clinton tops have been reported for 668 wells in the index area. The predominant regional feature of the area is the northwest dip into the Michigan Basin. Also evident is the westward displacement of the Algonquin Arch. Less obvious is the northeast end of the Findlay Arch at the southern edge of the map. Although the Electric Fault has a pronounced effect on the contours, the control points are insufficient for precise definition of its location. Displacement along the fault is about 275 feet, downthrown to the south. The Clearville Field (Cambrian oil) is at the southeast corner of the index area. The field is situated on the upthrown side of a north-south fault, downthrown to the west. The difference in elevation from the crest of the structure to the lowest observed top is about 170 feet. The Dawn structure at the centre of the area, which extends east-west for about nine miles, has a maximum indicated displacement of 155 feet down to the south. At the northwest corner of the index area, the Kimball-Colinville monocline trending northwest-southeast has a maximum displacement of about 140 feet down to the southwest; the displacement decreases to less than 100 feet to the southeast. Other indicated anomalies (A) in the index area are mostly due to single control points, so interpretation of their significance is difficult; further­ more, they may represent errors in identification of the top of the Clinton or in recording of the top. To enhance subtle structures, the Clinton data were processed for trend analysis. The results of fitting first- through fourth-order trend surfaces to the top of the Clinton are tabulated below.

General Mean value for Clinton Elevation —1528 Sum of squares about mean 4.1836 x 107

Fit of order 1 Residual sum of squares 1.1591 x 107 Percent reduction from original 7.2295 x 101 Percent reduction from previous not applicable

Fit of order 2 Residual sum of squares 1.6552 x 106 Percent reduction from original 9.6066 x 101 Percent reduction from previous 8.5720 x 101

Fit of order 3 Residual sum of squares 1.4785 x 106 Percent reduction from original 9.6466 x 101 Percent reduction from previous 1.0674 x 101

Fit of order 4 Residual sum of squares 1.2175 x 106 Percent reduction from original 9.7590 x 101 Percent reduction from previous 1.7650 x 101

The sum of the squares of the differences between the data values and the computed first-order surface is 72.3% less than that between the data values

61

and the mean. The strong first-order component is due to the northwest dip into the Michigan Basin. There is a 96.1% reduction with a second-order surface. Application of the second order has reduced the residual sum of squares by 85.7%. The improved fit stems from the ability of the second-order surface to account for southward dip of the surface along and south of the Electric Fault. The third- and fourth-order surfaces account for 96.5% and 97.1% of the original sum of the squares; there is, of course, little improvement, as the second-order surface accounts for such a large part of the initial variability. The first- through fourth-order Clinton trend surfaces are illustrated in Figure 6-3. The relevance of the first- and second-order trend surfaces to the Clinton structure is obvious. The third-order trend surface shows little change. The added flexure in the fourth-order surface reflects the north- south strike of the contours in the area of the Clearville Field. The second-order trend surface, the highest order surface which provides a significant improvement in fit, was selected for removal from the observed values. The resulting residual map (Figure 6-4) shows the differences between observed tops of the Clinton and the computed values of the second-order surface. Subtraction of the regional trend emphasizes features not related to the trend. The regional gradient is reduced by the extent of the fit of the surface while there is no apparent reduction in the magnitude of the anomalies. In the area of uniform northwest dip into the Michigan Basin, the 25-foot contours are spread considerably, except along the Kimball-Colinville monocline. Removal of the regional trend gives that feature the appearance of a northwest-trending asymmetrical anticline. The character of the Dawn structure and an eastward extension of the apparent discontinuity of the feature are enhanced by the second-order residual. The Electric Fault and Clearville Field features appear to be essentially unchanged due to the low degree of fit by the second-order surface. The high percentage of reduction of the sum of the squares (96%) would seem to dispute this. However, there is a much greater proportion of control in the northwest part of the map where the fit is good, so the high statistical fit is somewhat misleading. Three features which are largely masked by the strong northwest dip are numbered on Figure 6-4. Number 1 is a feature similar to the Kimball- Colinville monocline, although reduced in scale. Number 2 is a northwest- southeast trending low defined by the zero contour. Number 3 is a more extensive low with a similar trend defined by the-25 foot contour. In summary, trend analysis of the Clinton structure reveals few anomalies not already apparent from the structure contours.

A-2 Carbonate Structure The A-2 Carbonate structure (Figure 6-5) above the Guelph-Lockport oil-producing horizons was examined to obtain better resolution of the subregional structures. The number of control points for this surface (1560) is more than double that for the Clinton surface (668) which enhances the structural definition of all subregional features, except the Clearville Field; at Clearville, the Clinton is also penetrated to reach the Cambrian producing horizon. Displacement along the Kimball-Colinville monocline is essentially un­ changed; the apparent steepening of the feature is due to additional control points. Displacements on the Dawn structure and the Electric Fault are also

63

unchanged; however, the added control more effectively delineates these structures. The monoclinal feature (1) suggested by the Clinton residual is confirmed by control in the West Becher Field. This surface also reveals a number of highs north of the Kimball-Colinville feature with about 100 feet of relief. These highs of limited areal extent are due to underlying Guelph pinnacle reefs (see Chapter 9). A number of apparent anomalies on this surface (indicated by open triangles) reflect single-well data and are apparently the result of erroneous values for the top of the A-2 Carbonate; an extreme example is the "wrapping" of contours north of the west end of the Dawn structure. Trend surfaces through the fourth order were fitted to the 1560 A-2 Carbonate control points. The trend data, summarized below, is quite similar to that for the Clinton.

General Mean value for A-2 Carbonate Elevation — 1022 Sum of squares 8.5738 x 107

Fit of order 1 Residual sum of squares 2.1243 x 107 Percent reduction from original 7.5223 x 101 Percent reduction from previous not applicable

Fit of order 2 Residual sum of squares 6.6852 x 10°" Percent reduction from original 9.2203 x 101 Percent reduction from previous 6.8530 x 101

Fit of order 3 Residual sum of squares 6.0721 x 10^ Percent reduction from original 9.2918 x 101 Percent reduction from previous 9.1720 x 10^

Fit of order 4 Residual sum of squares 5.8527 x 106 Percent reduction from original 9.3174 x 101 Percent reduction from previous 3.6128 x 10^

The original sum of squares is reduced 75.2% by the first-order surface and 92.2% by the second-order. As with the Clinton data, third and fourth orders provide only very slight improvements to the fit. The second-order residual improves the definition of the subregional features considerably (Figure 6-6). The linear extent of the Kimball-Colin­ ville monocline is greater than expected from the Clinton second-order residual. The influence of this feature extends to the southwest, almost intersecting the Dawn structure, which is also more extensive than previously suspected. Somewhat diminished in magnitude, the Dawn feature swings southwest to west and connects with the subtle monoclinal feature (1) revealed on the Clinton second-order residual. The Dawn structure, as extended, runs parallel to the Electric Fault. There is a swing to the southwest in the Electric Fault and a return to an east-west trend as well. Lack of control prevents determining if the parallelism of the Dawn

65 structure and the Electric Fault extends to the northwest. The residual highs in the northwest part of the area which are due to the underlying Guelph reefs are quite prominent on the A-2 Carbonate second-order residual.

Bass Islands Structure The rather straightforward characteristics of the sub-regional structures in the Clinton and A-2 Carbonate surfaces are lost when the uppermost Silurian Bass Islands is considered (Figure 6-7). The monoclinal Dawn feature becomes a low trough about 200 feet deep which extends farther to the east and west. The southwestward extension of the Dawn structure, which could be seen on the A-2 Carbonate surface only with the aid of trend surface techniques, is prominent on the Bass Islands surface. The Electric Fault, while still observable in places, does not everywhere conform to the trend established on the A-2 Carbonate surface. Relief on the Kimball-Colinville monocline is increased. Two control points which differed by 137 feet on the Clinton differ by 210 feet on the Bass Islands. Numerous isolated closed lows in the northern part of this area exhibit up to 300 feet of negative relief. South of the Electric Fault there are several large north-south trending lows and highs, e.g. one is west and north of the Clearville Field. The drop of the Clearville structure to the west has been reversed so that there is instead a drop of similar magnitude to the east. The change from the simplicity of the Clinton and A-2 Carbonate horizons is also reflected in the trend surface data. The reduction for the first order is only 46%, and for the second through the fourth orders the reductions are 53%, 55%, and 57%. Thus, the first order fits, to a lesser degree, the northwest regional dip; but the second through fourth orders can do nearly nothing with the multiplicity of highs and lows. The second-order Bass Islands residual is almost identical to the Bass Islands structure.

Detroit River Structure The Detroit River structure for the index area (Figure 6-8) is strikingly different. First, the relief on the Kimball-Colinville monocline is greater. Second, the Electric Fault is not observable; in fact, along the western end the dip has reversed to the north. Furthermore, the dip along the west side of the Clearville Field is reversed from that of the Bass Islands and is to the west as on the A-2 Carbonate surface. Modification of the structures on the upper horizons independently of the lower horizons suggests emplacement or removal of material in the intervening section. The correlation between the areas of structural anomalies and absence or thinning of the B Salt is quite distinct on the isopach map of the B Salt (Figure 6-9). It appears that the presence or absence of B Salt has a considerably greater effect on the attitude of the upper beds than tectonics. Isopach maps for individual formations and groups of formations generated for this study also showed that in the index area of subregional structures variations of sediment thickness above the B Salt coincide with variations of thickness of the B Salt. Chapter 8 includes additional discussion of the variations in the salt thickness.

Discussion and Summary The large-scale regional maps discussed in the preceding chapter have inadequate resolution for analysis of development of subregional structures. Therefore, a control area which contains several significant subregional

67 features was selected for mapping at the more suitable scale of 1 inch = 1 mile. First, structure maps were generated to outline the structures and trend surface maps were created to reveal subtle structures masked by the gradients of the regional structures. Finally, isopach maps to date structural development of the subtle features were generated on the assumption that changes in sediment thickness which cross the strike of a structure would date times of movement. These detailed maps of the control area effectively give better definition to the subregional structures; furthermore, the trend surface maps reveal a connection between the Dawn structure and the monoclinal structure south of the West Becher Field. However, interpretation of the historical development of the subregional structures is not feasible with the method employed, due to structural modifications independent of tectonics. More detailed study is suggested as a possibility for future work.

68 chapter 7

structural manifestations of salt removal

Drilling of Devonian structural "highs" in anticipation of corresponding highs in the Silurian revealed significant variations in thicknesses of the Salina salt members. Caley (1945), Roliff (1949), and Evans (1950) postulated that removal by dissolution could account for these variations. Grieve (1955) demonstrated that dissolution or leaching of the salt beds had resulted in increased thickness of the overlying Upper Silurian and Devonian sediments and that these increases could be used to establish the history of leaching in time and space. The B Salt member of the Salina is irregularly distributed in southwestern Ontario and thins markedly near the margins of occurrence (Figures 3-19 and 6-9). The thickness of the A-2 Salt member also varies, but to a lesser extent (Figure 3-17), so that removal of the B Salt is the principal cause of change in thickness of overlying sediments. The other salt members are regionally absent where significant thickness variations take place. The distribution pattern of the B Salt (Figure 7-1) exhibits four elongate areas trending generally east-west where the member is absent. The largest of these areas (1) extends over 45 miles along the trend and north of the Electric Fault; the maximum north-south width is about 10 miles at the west end. Connecting with this end and extending south of the Electric Fault is a second area (2) 30 miles long by 6 miles wide. A third area which is barren of B Salt (3) lies north of the east end of the Electric Fault; this area is about 15 miles long by 8 miles wide. The fourth area (4) lies north of the Electric Fault along the Dawn structural trend; this area of 15 miles long by a half mile to a mile wide extends southward at each end to connect with the saltless area along the Electric Fault. Several other isolated areas, which are barren of B Salt, are 1 to 3 miles across; some are associated with underlying Guelph-Lockport reefs.

Regional Structural Effect The regional structural effects of dissolution of the B Salt are apparent from comparison of the underlying A-2 Carbonate structure with the overlying C Shale structure (Figures 7-1 and 7-2). Apart from the relationship between salt absence and the Electric Fault and Dawn structural trend, the distribution of the salt apparently is not affected by the A-2 Carbonate structure. However, the salt-barren areas correspond closely with structural

69 lows on the top of the overlying C Shale. As the C Shale has a fairly uniform regional thickness and the corresponding structural lows are not present on the A-2 Carbonate, the structural configuration of the C Shale must have resulted from lowering due to removal of considerable quantities of the intervening B Salt.

Time of Salt Dissolution The structural lows on top of the C Shale over areas of presumed dissolution or leaching of the underlying salt are not always reflected upward to the base of the formation at the surface; they become less evident in compensating for increased sediment thickness as shown on a cross section of four wells in Macomb County, Michigan where the B Salt is present (Figure 7-3). Well #3 in this cross section has a normal regional sequence. In well #4, the uppermost formations (Kettle Point and Hamilton) are thicker than normal and successively deeper formations are progressively lower with respect to well #3. The cumulative thickness of the formations from the top of the Kettle Point to the top of the Salina is greater in well #4 by 380 feet. The gamma ray-neutron log traces illustrate that the compensation is distributed throughout the expanded interval. This is most evident in the Hamilton Formation, within which the interbedded shale and carbonate exhibit good correlation points on the mechanical log. Within the Salina interval of well #4, the F Salt, the D Salt, and most of the B Salt are absent; the aggregate total thickness of salt is 315 feet less than in well #3 and the total Salina is thinner by the same amount. However, removal of 315 feet of salt is insufficient to account for the observed increases in sediment thickness. This suggests that the original thickness of salt was greater and that leaching occurred in well #3 during a later period of time than is represented by the present sequence. In well #2, all salts (totalling 600 feet in thickness) are absent. Additional section, mainly in the Bois Blanc and Hamilton, totals 325 feet, which is 275 feet short of compensating for the salt supposedly removed. The Kettle Point is 250 feet lower structurally than in the reference well which suggests that leaching occurred subsequent to deposition of the Kettle Point. In well #1, only a half mile west of well #2, the time of leaching was quite different. The Dundee and Detroit River are 381 feet thicker in contrast to only 80 feet of increase in well #2. The cumulative section from the Kettle Point to the Salina is 560 feet thicker than in well #3, that is, nearly equal to the amount of salt originally present. Thus, the cross section of wells in the local area shown in Figure 7-3 reveals that salt removal and surface lowering occurred concurrently with sedimentation over a long period of geologic time. The sediments were deposited in greater thickness as regional subsidence was augmented by local subsidence due to salt leaching. The variation in the time of leaching over short distances indicates that the controlling factors are local.

Lateral Variation in Time of Salt Leaching The course of salt dissolution or leaching through time can be demonstrated by structure and isopach maps. The Guelph structure for a local area in the vicinity of the Dawn structure is shown in Figure 7-4, A. The northeast- southwest regional trend is broken by the east-west trending, southward dipping Dawn structure which has a maximum relief of about 160 feet. The Dawn Salina gas fields lie along the crest of the Dawn structure. Four Guelph

71 pinnacle reefs are known within this local area. Two small reefs are at the west end of the Dawn structure, Dawn 47-49 to the south and Dawn 59-85 to the north. Two larger reefs, Dawn 156 and 167, are about 3 miles north of the Dawn structure. These reefs have a maximum relief on the Guelph of about 350 to 400 feet. The structural configuration on the Salina A-2 Carbonate at the base of the B Salt is essentially the same as on the Guelph with the exception that the relief over the four reefs has been reduced to about 125 feet. The B Salt which is 200 to 250 feet thick over most of this local area (Figure 7-4, B) is absent along most of the Dawn trend and over three of the four reefs. The well control indicates that the thinning of the B Salt takes place within a distance of about one-half mile to a mile. The elongate area of absent B Salt is narrowest in an east-west trend along the crest of the Dawn structure and widens at each end toward the south where it connects with the saltless area along the Electric Fault (see Figure 7-1). The B Salt is absent over the Dawn 156 and Dawn 167 reefs and is thin in a surrounding area of about a half mile to a mile. The Dawn 47-49 reef lies within the saltless area at the west end of the Dawn structure and the relationship of B Salt to the reef cannot be demonstrated. The B Salt is thinner over the Dawn 59-85 reef by 100 to 125 feet which is equivalent to the relief at its base on the A-2 Carbonate. This suggests that the thinning is due to compensation for a pre-existing high rather than to salt dissolution. The structure on the top of the Bass Islands (Figure 7-4, C) corresponds approximately to the thinning of the B Salt. The surface of the Bass Islands is lower by an amount approximately equal to the reduced thickness of the B Salt. Structural highs occur within closed lows over the Dawn 156 and Dawn 167 reefs. These highs show a maximum reversal of about 125 feet. A comparable high over the Dawn 47-49 reef is not within a closed low. No low is associated with the Dawn 59-85 reef, apparently because the salt has not been dissolved. Thinning of the B Salt and increased thickness of the Detroit River-Bois Blanc section correspond closely (Figure 7-5, A); along the Dawn structure, an increased thickness of 150 to 200 feet coincides with the area of thin or absent salt. In the area of the Dawn 156 reef, increased thickness of this section is limited to the northeast end of the reef. In the area of the Dawn 167 reef, thickening coincides closely with thinning of the B Salt. The thickening within the Dundee is generally less than 50 feet and would appear to be related to salt thinning. Local thickening of the Hamilton is from 50 to 75 feet (Figure 7-5, B). primarily along the central part of the saltless area. At the north end of the Dawn 156 reef, the Hamilton is thicker by 60 to 90 feet. The youngest formation in this local area is the Kettle Point (Figure 7-5, C) which forms bedrock below the glacial drift, except in the southeast part of the area where it is absent. Because the erosional surface at the base of the drift is essentially horizontal, the increased thickness of the Kettle Point probably resulted from subsidence which took place either during or after deposition. Along the north and west margins of the saltless area adjacent to the area of thick Detroit River-Bois Blanc, the Kettle Point apparently thickens by as much as 100 feet. In summary, two major periods of salt leaching in the Dawn area have occurred. The first was during deposition of the Detroit River-Bois Blanc sequence, and the second was during deposition of the Kettle Point

73

B

A

1 2 3 4 5 6 7 8 9 10 11 12 13 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Figure 7—7. Reconstruction of salt collapse migration of salt leaching. Figure 7-8 is a section across the Clearville Field. It crosses an area 3 miles wide by 8 miles long where the Bass Islands far exceeds the regional thickness (see Figure 7-9). The Bass Islands is thickest along the trend of the north-south fault on the west side of the Clearville Field. In well 3, the Bass Islands is 464 feet thick, about 300 feet more than normal. To the east in wells 4, 5 and 6, a lens of Oriskany sand about 120 feet thick overlies the Bass Islands which is thinner by a comparable amount. At the east end of the section, the Bass Islands has diminished to about 131 feet, the Oriskany is absent, and the overlying Detroit River-Bois Blanc has increased from 200 feet to around 416 feet. If formation thickening is assumed to indicate increased sedimentation due to collapse over salt leaching, the following history can be proposed. Solution was initiated during deposition of the Bass Islands in the area of well 3 and later spread to the east, allowing a considerable amount of Oriskany to accumulate in wells 4, 5 and 6. The pocket of Oriskany within the resulting sag was apparently protected from stripping during subsequent erosion. Still later solution at both ends of the section, dated in part by thickening of the Detroit River-Bois Blanc section, caused the present structural high at the base of the Bois Blanc'The Detroit River-Bois Blanc interval over the high is as much as 250 feet thinner than would be expected regionally. This suggests that a topographic feature in the area prior to Lower Devonian Bois Blanc sedimentation which caused compensation during a period not represented by sedimentation could account for this; in this case, during the period represented by the hiatus which followed deposition of the Oriskany. To postulate salt dissolution as solely responsible for approximately 300 feet of thickening in the Bass Islands and Oriskany would require that at least that much salt had been present originally; however, the salt thickness now present northeast and southwest of Clearville is 120 feet less than required for this interpretation.

76

In addition to faults and folds, pinnacle reefs can act as centres of salt leaching as illustrated in Figure 7-11, a cross section across the Dawn 156 reef. The A-1 Carbonate and A-2 Salt are apparently truncated against the flanks of the reef, while the A-2 Carbonate, considerably thinned, extends over the reef ,with a relief of about 125 feet. It is assumed that this relief was erased by deposition of the B Salt and that the B Salt and subsequent forma­ tions up to and including the Bass Islands were deposited horizontally over the reef. At some time, presumably during the interval represented by the Detroit River-Bois Blanc, solution of the B Salt began over the northeast end of the Dawn 156 reef resulting in subsidence of the overlying formations. With a lowering of the sea floor, additional sediment was permitted to accumulate over the site of salt solution. At the time of initial deposition of the Dundee, the top of the Detroit River was horizontal but its base was lower over the northeast end of the reef. No significant salt solution seems to have occurred in this area during deposition of the Dundee and Hamilton. Then, during or after deposition of the Kettle Point, leaching commenced on the southwest side of the reef. Since the top of the Kettle Point has been eroded, it is not possible to establish an upper limit for the time of leaching. Figure 7-11 illustrates the high within a low noted on the Bass Islands structure (Figure 7-4, C). The formations overlying the B Salt, up to and including the Bass Islands, were affected by both periods of solution and were lowered a distance equal to the total thickness of salt removed, resulting in a low over and around the reef. However, since there was initially less salt deposited over the reef due to its topography on the bottom, the result has been the creation of a structural high within a low. A final, and perhaps the simplest illustration, of salt collapse structure is drape over residual salt as illustrated in a north-south cross section through the Wardsville (Silurian) Field (Figure 7-12). Drape of this type over a salt remnant is an excellent way of obtaining a trap for hydrocarbons. It is not necessary that the salt be an isolated remnant to achieve a trap. Dip reversal, and hence a trap, could occur anywhere along the downdip margin of the trends of absent B Salt. In fact, draping over salt remnants is probably the trapping mechanism for most of the Devonian oil fields in Ontario. Figure 7-13 illustrates the relationship between the position of these Devonian fields and the distribution of B Salt. The B Salt contours are machine generated, and it should be pointed out that a better apparent correlation could be obtained by interpretive contouring.

Mechanism of Salt Leaching It has been shown that salt leaching has taken place over long periods of geologic time and that the controls are local in nature. The fluids which dissolved the salt apparently came from above, since it is the upper salts which are removed first (see Figure 7-3, well 4). Leaching appears to have taken place under submarine conditions; however, whether the fluids which dissolved the salt came from the ocean floor or from a porous formation above the salt is unclear. The close correlation between the Electric Fault and salt leaching indicates that the fluids moved along faults. The presence of salt leaching over pinnacle reefs suggests that fractures resulting from differential compaction might also be responsible. Regional joints too may have played a

80

chapter 8

reefs

Subsurface carbonate mounds with up to 540 feet of relief have been found in a 10- to 20-mile-wide band in Michigan and Ontario near the southern end of Lake Huron. These features are commonly called reefs, pinnacle reefs, or simply pinnacles. Their bases are in the upper part of the Middle Silurian Guelph Formation at subsea depths of approximately 500 to 3000 feet. Although the pinnacle reefs are not structures in the same sense as folds or faults, they are considered here because of their effect on the structure of the overlying formations and because of their importance as reservoirs for oil and gas.

Definition The most commonly cited definition of a reef is that of Lowenstam (1950, p. 433) who said, "a reef, in terms of ecologic principles, is the product of the actively building and sediment-binding biotic constituents, which because of their potential wave resistance, have the ability to erect rigid, wave-resistant topographic structures." According to this definition, the presence of actively building and sediment binding organisms must be demonstrated. Felber (1964, p. 59) reporting on the Michigan reefs states that "Algal forms are present and become particularly important in volume as the upper portion of the coral biosome is approached. The algae are predominantly encrusting types that bind fragmented reef material into a cohesive mass. From the jumbled and fragmented nature of many of the corals and brachiopods in the coral biosome it is evident that the organic structure had raised itself to within the range of wave action and during the latter stage of the coral growth the potential for a wave-resistant structure was estab­ lished." According to Felber, the lower parts of the reefs he studied did not possess the potential for wave resistance. Gill and Briggs (1970) refer to the lower 150 feet of the Belle River Mills reef (see Figure 8-9) as a "mound" which grew in quiet, relatively deep water. Thus, in the strictest application of Lowenstam's definition, only the upper parts of the structures in southwestern Ontario and southeastern Michigan are reefs. The semantic problem is here avoided by classifying the whole of the structures as stratigraphic reefs in the broader sense of Dunham (1970, p. 1931). Dunham applies the term "stratigraphic reef" to "thick,

83 laterally restricted masses of pure or largely pure carbonate rock". Stratigraphic reefs may, but not necessarily, include stages which meet the ecologic criteria of Lowenstam.

Reef Composition Several workers have differentiated the formation of pinnacle reefs into growth phases. Gill and Briggs (op. cit.), working with the Belle River Mills reef, recognized three phases: "(1) bioherrhal, consisting of skeletal (crinoid, bryozoan, coral, and tabular stromatoporoid) wackestone, and packstone rudites and arenites; (2) organic reef, consisting of a reef core (massive stromatoporoids, corals, and algae? ) and associated interbedded and interfingering lithofacies of "backreef' skeletal wackestone rudite, burrowed mudstone and laminite, and coarse skeletal forereef talus; and (3) supratidal cover complex, composed of stratified algal stromatolites, flat pebble conglomerates, oncolites, and burrowed pelletal mud". To Phase 1, they attribute 150 feet of growth in "quiet", relatively deep water. Phase 2 is attributed to a height of 300 feet in turbulent water. While Gill and Briggs do not make the correlation, Phase 3 appears to equate to the supra-reef carbonates, ascribed by many workers to the A-1 Carbonate. Hill (1966), studying a drill core from the Payne reef seven miles northeast of Belle River Mills (see Figure 8-9),, identified five phases of growth. The initial phase was one of accelerated crinoid accumulation, resulting in a buildup with 30 to 50 feet of relief. Phase 2 is represented by 50 feet of skeletal carbonate containing considerable debris of colonial algae and algal pellets. Phase 3 was the main stage of reef growth in which vertical growth kept pace with basin subsidence. The 307-foot section of organic lattice ascribed to this phase was identified as being primarily of algal origin with about 15% solitary and colonial corals. A 4-foot zone of stroma­ toporoids represents the fourth phase of reef growth. Hill interprets this phase as resulting from a temporary halt in basin subsidence which permitted the reef to grow up into turbulent water thereby favouring the wave-resistant "stroms". The fifth phase consists of scattered reef detritus and bedded, fine-grained carbonates which, according to Hill, may represent lagoonal deposits at the top of the reef. Felber (op. cit.) studied cores from several southeastern Michigan reefs and concluded that there were three biosomes (biofacies?): a lower echinoderm or crinoid biosome which grades upward into a coral biosome which is, in turn, overlain by an algal biosome. Algae are present in all three zones and increase in importance upward. Common to the three studies by Gill & Briggs, Hill, and Felber are (1) a crinoid-rich base, (2) an intermediate phase or biosome characterized by corals, and (3) an algal cap.

Reefs in the Study Area The 72 pinnacle reefs within the study area are plotted on Figure 8-1. The nonproductive reefs, usually penetrated only by a single test, are indicated by a standard dry hole symbol. Generally, the reefs are elongate, trending north-south to northeast-southwest. The average lateral dimensions for 34 of the reefs are 4950 feet long by 2150 feet wide. These figures would be too large for the total population since they include only those reefs for which there are sufficient penetrations to make an estimate feasible. The estimated height of each reef above the regional Guelph shows a maximum

84

development of about 400 feet, parallel to the trend of reef occurrence. Along and perpendicular to the trend, some gradient in height is apparent. To the northeast, the reefs are higher by about 50 feet; the Bayfield reef near the northern margin of the study area is considerably higher and has 540 feet of relief. The trend of pinnacle reefs is approximately parallel to the margin of the reef complex (arbitrarily taken to be the 200-foot isopach of the Guelph to Rochester interval). Recent data from current exploration activity in the lower peninsula of Michigan indicate that the reef trend may be continuous around the Michigan Basin.

Porosity Production of oil and gas from within the pinnacle reefs has prompted continual search for these features. Of the 72 known reefs in the study area, 37 are classed as gas producers and 14 are oil producers. The oil-bearing reefs usually have a considerable gas cap. Five of the reefs contain salt water and 16 have salt-plugged porosity or are nonporous. There are no obviously predictable trends of occurrences of oil, gas, or water within the reefs; however, trends of salt occurrences can be defined with some degree of reliability. North of the dash-dot line on Figure 8-1 the porosity of the reefs is almost completely salt-plugged; in contrast, salt plugging does not destroy a significant amount of porosity in the reefs south of this line.

Reef Structure A cluster of reefs within the general study area (outlined on Figure 8-8 at the southern tip of Lake Huron) was selected for more detailed study of reef structures. A computer-generated structure map on the top of the Guelph is shown in Figure 8-2. Two distinct chains of reefs and the elongate north-south trends of each pinnacle are quite evident. The western chain includes the Seckerton, Seckerton North, and Corunna reefs; the Seckerton and Seckerton North features may actually be parts of a single reef. The eastern Kimball-Colinville reef chain was first defined by two separate discoveries, but development drilling revealed the structure as a single entity. The structural low at the south end, which suggests separation into another field, was discovered in a still later stage of development. Each of the reefs in this restricted area has relief on the order of 400 feet. Because of steep slopes on the reef margins, few wells penetrate the flanks. As a result, the machine-generated contours do not necessarily reflect the true slopes since the computer program, or a geologist contouring manually, will interpolate between points of control. Actually, the amount of slope indicated is governed by the proximity of the wells to the reef flanks. For example, 12 slopes calculated from the contours on the top of the Guelph range from 5° to 35° with an average of 16°. A 35° slope calculated for the west flank of the Seckerton reef results from a well which penetrated the lower part of the west flank. In contrast, the nearest off-reef well on the east side of the Kimball-Colinville reef is nearly a mile distant from the reef and the calculated slope is then only 5°. The true slope of most of the reefs in this limited area is probably on the order of 30°. This observation is based on contouring where drilling has been on a closely spaced, fairly regular pattern and on measurements of dips in cores from flank wells. The regional northeast trend of the Guelph surface in this study area is broken by a number of low features or depressions as well as the reef highs.

86

The A-l Carbonate thins considerably on the flanks of the reef and there is lack of agreement regarding its presence over the reef crest. The writer has examined cores of the reef crest and found at the top of the massive reef section about a 5-foot stromotoporoidal section which is overlain by bedded calcarenites ranging in thickness from 15 to 60 feet. The calcarenites are overlain by 20 to 50 feet of A-2 Anhydrite. Some workers, especially those in Michigan, report the A-l Carbonate top at the base of the A-2 Anhydrite and the Guelph or reef top at the stromatoporoidal "cap". Others report the Guelph top at the base of the anhydrite. These differences may be observed in the cross section where the A-l Carbonate is shown in some reef wells, but not in others. In either event, the A-l Carbonate is considerably thinned due to the reef. The A-2 Salt is not present over the crest of the Seckerton-Seckerton North reef, nor, so far as is known, over any other pinnacle reef in southwestern Ontario and southeastern Michigan. A 10-foot section of salt is found in the saddle between the pinnacles (well 7). Where the A-2 Salt thins against the elevated A-l Carbonate, the A-2 Anhydrite thickens (wells 3 and 7). Anhydrite beds are encountered at both the top and base of the A-2 Salt (see Chapter 3). Normally, the lower unit thickens at the reef margin. At the crest of reefs where there is no intervening salt between the two anhydrite units, the combined thickness of the two units is as much as 50 feet. A thick A-2 Anhydrite is not found over all pinnacles. Where the A-2 Salt is less than 75 feet thick, only a few feet of anhydrite, or none at all, are encountered. The latter is true especially of reefs in Huron County east of Lake Huron. Regionally, the A-2 Carbonate is about 150 feet thick, but over the crest of the Seckerton-Seckerton North reef the thickness is only about 90 feet. As a result of this thinning and the thinning or pinching out of the A-2 Salt, the A-l Carbonate and the A-l Anhydrite, the structural effect of the under­ lying reef is considerably diminished when viewed on the top of the A-2 Carbonate (Figure 8-4). The positions of the reefs are still readily apparent, but the relief has been reduced to about 100 feet. The configuration of the off-reef contours is essentially the same as that for the top of the Guelph except for the reduction of the lows around the reefs which were imposed by the steeper Guelph gradients. Thinning of the lower Salina units, which is responsible for the reduced gradient, is well illustrated by an isopach map of the A-2 Carbonate to Guelph interval (Figure 8-5). The regional thickness of the interval is relatively uniform, ranging from about 390 feet on the southeast to 420 feet on the northwest except over the reef crests where the section is less than 125 feet thick. It should be noted that the 425-foot contours at the reef margins enclose areas of no control. Here again, these areas are in response to the steep gradients of the reef flanks. The regional thickness of the B Salt (Figure 8-6) ranges from 275 to 300 feet. Over the reefs the salt is thinned to as little as 159 feet. There are two wells (*) on the west margin of Seckerton, one west of Seckerton North, and two at the margins of Kimball which are thinner than would be expected from their position relative to the reef. This may represent an early stage of the salt solution phenomenon which is associated with numerous reefs, especially those in the northeast part of the reef trend. There is also a well north of Colinville and east of Corunna which is 44 feet thinner than its nearest neighbor. A structure map on the top of the B Salt (Figure 8-7) readily distinguishes between salt thinned by the reefs and thinning (*) due probably to solution. Little discernible structure variation of the top of the salt occurs in wells over reefs whereas the top of the salt is low in the thin off-reef wells. 89

can be made that structural features present prior to Guelph time may have had a localizing effect on the reefs. In addition to the pinnacle reefs, there are similar structures with relief in the order of 30 to 100 feet. The more prominent ones are marked with a star in Figure 8-9. They are commonly referred to as incipient reefs (Shouldice, 1955) or subdued reefs (Pounder, 1963a). It is generally assumed that the features represent potential pinnacle reefs which ceased to grow at an early stage, presumably as a result of drowning by the rapid subsidence or from being "killed off by the A-1 Anhydrite, depending on one's concept of the time of reef development. In the horizontal dimensions, these buildups are similar to the pinnacles. Eleven such fields average 2100 feet wide by 4400 feet long. However, the Guelph contours cast some doubt on the assumption that the so-called incipient reefs and the initial stages of the pinnacle reefs are one and the same. In virtually every case where there are sufficient wells to establish a trend, the trend of the incipient features is approximately east-west. This contrasts strongly with the north-south trend of the pinnacles. Much of the regional structure was removed by surface fitting in an effort to determine if the apparent contrast of the pinnacle reefs and incipient features in trends is the result of exaggeration due to regional dip. It would be misleading to fit a surface to all values for the Guelph, because the values for the reef tops are not a direct function of regional structure or, in other words, not a part of the model, which is a slightly concave slope into the Michigan Basin. If all Guelph tops were included in the computation of the trend surface, the resulting surface would be above the regional and below the reef crests and representative of neither. Wells with apparent reef buildup were excluded to remove the local effect of reefing from computation of the regional trend. It was necessary to use indirect evidence of buildup since most wells do not completely penetrate the Guelph, thereby preventing a direct thickness calculation. As previously noted, the A-1 Carbonate and the A-1 Evaporite thin on the flanks of the reefs (see Figure 8-3). Wells with less than a normal thickness of A-1 were excluded from the computations. This was accomplished by first removing wells with less than 80 feet of section; then, first-order through fourth-order surfaces were fitted to the remainder. A second-order surface provided the best fit and 246 wells in which the A-1 section was more than 10 feet thinner than the regional trend were also removed. The Guelph structure of the remaining "non-reef wells (Figure 8-12) shows the outlines of the pinnacles (shaded) and it can be seen that the reef influence has been essentially removed. Coefficients were computed for a second-order trend surface using the "non-reef Guelph tops. A simple program was written to evaluate the equation of the surface at each of the original control points, both reef and non-reef. The differences in feet between the computed non-reef surface and the actual or observed surface are'shown contoured in Figure 8-13. The Kimball-Colinville monocline trends at a right angle to the regional strike and is not approximated by the second-order polynomial equation. Removing the component contributed by basin subsidence gives the Kimball-Colinville monocline the appearance of an asymmetrical anticline. The east-west lineation of the incipient reefs is clearly enhanced by removal of the regional component. That the orientation is different from that of the pinnacle reefs is evident. There appear to be three modes of

95

occurrence: as tails on known reefs (1, 2j & 3, Figure 8-13); as separated features a short distance east of known reefs (4, 5, 6 & 7); and as features not associated with known reefs (8, 9, 10 & 11). There are several possible reasons for the differences in orientation; however, a discussion of these reasons based solely on geometry must be considered speculative. One possibility is that the incipient reefs are the result of an earlier stage of reef development which died out before the true reefs got started. Acceptance of this idea would require that some change- in ecologic conditions took place. A second possibility is that the incipient reefs are debris piles in the lee of pinnacles. If this is true, several areas are enhanced for reef exploration. If the incipient reefs are neither separated in time from the pinnacles nor an erosional product of the pinnacles, the constituent organisms may have differed and could not respond to increased water depth. This latter idea, however, still does not explain the difference in orientation.

Discussion and Summary The pinnacle reefs are some of the most prominent structural features in the area and affect the structure of overlying horizons up to at least the base of the C Shale. The A-2 Salt, the A-1 Anhydrite and, perhaps, the A-1 Carbonate pinch out against the flanks of reefs while the B Salt and A-2 Carbonate are thinned and elevated over the reefs. How much of, or whether, the overlying structure is original or is the result of differential compaction cannot be determined from the present geometry. The average dimensions of the reefs are slightly less than a mile long by half a mile wide. Individual reefs trend north-south to northeast-southwest while the trend of occurrence is approximately parallel to the margin of the reef complex. Along the central part of the trend, the reefs have a relief of about 400 feet and show a decrease in height along each margin. The distribution of reef occurrence and height suggests that a band-like area developed where conditions for reef initiation and growth were optimal and that these conditions changed from favourable to unfavourable over a distance of a few miles on either side of the band. The exact nature of the controls for these conditions cannot be demonstrated, but they were probably depth related. The controlling factors for the positions of individual reefs are also indeterminate from the structure maps presented, but there are indications that pre-existing highs may have had an effect. The lower-lying "incipient" or "subdued" reefs have been assumed, by previous investigators who have mentioned them, to be pinnacle reefs which died at an early stage. However, the distinct difference in orientation demonstrated by this study makes this interpretation unlikely.

100 chapter 9

summary of conclusions

The following conclusions are drawn from this study: 1. Development of the Ontario Well Data file, which forms the major source of data for this study, and of associated computer application programs has provided a valid and convenient means of studying the structure and distribution of the sediments in southwestern Ontario. 2. The structural attitude of the formations in the study area has been affected by at least three factors: tectonic deformation, collapse as a result of salt dissolution, and drape over organic reefs. 3. The major positive features of the study area, the Algonquin and Findlay Arches, are in reality two ends of the same feature (the less negative area between the Michigan and Appalachian Basins) which remained positive in a relative sense as the opposing flanks were pulled down during differing periods of time. Evidence provided by the Upper Cambrian sediments indicates there was an initial arch present prior to that time. Dip into the Appalachian Basin was steepened mainly during the Late Ordovician and Late Devonian. Dip into the Michigan Basin was created largely during Late Silurian and Middle Devonian. 4. The Chatham Sag is a fairly late feature which did not have a significant influence on sedimentation until Late Silurian time, and most of the development occurred later than the Silurian. 5. The history of development of subregional features, such as the Electric Fault and the Dawn structure, is masked by collapse due to salt leaching. 6. Salt dissolution occurred concurrently with sedimentation over long periods of geologic time in response to local controls. 7. In addition to lows resulting from collapse, highs have been formed due to drape over salt remnants or over centres of earlier collapse. 8. From limited deeper control, salt collapse structures appear to have provided the traps for shallow Devonian oil fields in the study area. 9. The trend of pinnacle reef occurrence is approximately parallel with the margin of the reef complex which surrounds the Michigan Basin. Maximum reef height decreases on either side of the trend, suggesting that conditions favourable for reef initiation and growth changed over a distance of a few miles.

101 10. The distinct difference in orientation of the lower lying subdued reefs makes unlikely the interpretation that they are pinnacle reefs which died at an early stage.

102 references

Alguire, S. L., 1962, Some geologic and economic aspects of Niagaran reefs in eastern Michigan; in Silurian rocks of the southern Lake Michigan area: Guidebook of Mich. Basin Geol. Soc, p. 30-38.

Badgley, P. C, 1965, "Structural and Tectonic Principles": New York, Harper and Row, 521 p.

Barrell, J., 1917, Rhythms and the measurement of geologic time: Geol. Soc. America Bull., v. 28, p. 745-904.

Beards, R. J., 1967, Guide to the subsurface Palaeozoic stratigraphy of southern Ontario: Ont. Dept. of Energy and Resources Management, Paper 67-2.

Best, E. W., 1953, Pre-Hamilton Devonian stratigraphy, southwestern Ontario, Canada: Ph.D. thesis, Univ. of Wisconsin.

Bolton, T. E., 1957, Silurian stratigraphy and palaeontology of the Niagara Escarpment in Ontario: Canada, Geol. Survey, Mem. 289.

Briggs, L. I., 1959, Physical stratigraphy of Lower Middle Devonian rocks in the Michigan Basin: Guidebook of Mich. Basin Geol. Soc, p. 39-58.

Caley, J. F., 1940, Palaeozoic geology of the Toronto-Hamilton area, Ontario: Geol. Surv., Canada, Mem. 224.

Caley, J. F., 1945, Palaeozoic geology of the Windsor-Sarnia area, Ontario: Geol. Surv., Canada, Mem. 240.

Caley, J. F., and Liberty, B. A., 1957, The St. Lawrence and Hudson Bay and Palaeozoic outliers, Chapt. 4 of Stockwell, C.H., ed., Geology and economic minerals of Canada: Canada Geol. Survey Econ. Geology ser., no. 1, 4th ed., p. 207-246.

Cohee, G. V., 1948, Cambrian and Ordovician rocks in the Michigan Basin and adjoining areas: Am. Assoc. Petroleum Geologists Bull., v. 32, p. 1417-1448.

103 Cumings, E. R., 1939, Silurian system in Ontario, pt. 3 of Canadian extension of the interior basin of the United States: in Geologie der erde, Erich Krenkel, ed., North America vol. 1, p. 594-600.

Cumings, E. R., and Shrock, R. R., 1928,Niagaran coral reefs of Indiana and adjacent states and their stratigraphic relations: Geol. Soc. America Bull., v.39, p. 579-620. de Sitter, L. U., 1964, "Structural Geology": New York, McGraw-Hill, 551 p.

Dunham, R. J., 1970, Stratigraphic reefs versus ecologic reefs: Am. Assoc. Petroleum Geologists Bull., v. 54, p. 1931-1932.

Ehlers, G. M. and Resting, R. V., 1962, Silurian rocks of Michigan and their correlation; in Silurian rocks of the southern Lake Michigan area: Guidebook of Mich. Basin Geol. Soc, p. 1-20.

Evans, C. S., 1950, Underground hunting in the Silurian of southwestern Ontario: Geol. Assoc. Canada Proc, v. 3, p. 55-85.

Fagerstrom, J. A., 1961, Age and stratigraphic relations of the Formosa reef limestone (Middle Devonian) of southwestern Ontario, Canada: Geol. Soc. America Bull., v. 72, p. 341-349.

Felber, B. E., 1964, Silurian, reefs of southeastern Michigan: unpublished Ph.D. thesis, Northwestern Univ.

Fettke, C. R., 1948, Subsurface Trenton and sub-Trenton rocks in Ohio, New York, Pennsylvania and West Virginia: Am. Assoc. Petroleum Geologists Bull., v. 32, p. 1457-1492.

Flawn, P. T., ed., 1967, Basement map of North America: Am. Assoc. Petroleum Geologists and U.S. Geol. Survey.

Freeman, L. B., 1951, Regional aspects of Silurian and Devonian subsurface stratigraphy in Kentucky: Am. Assoc. Petroleum Geologists Bull., v. 35, p. 1-37.

Gill, D., and Briggs, L. I., 1970, Silurian reef in Michigan Basin — stratigraphic-, facial-, and reservoir-properties analysis: Am. Assoc. Petroleum Geologists Bull., v. 54, p. 848-849, abst.

Green, D. A., 1957, Trenton structure in Ohio, Indiana and northern Illinois: Am. Assoc. Petroleum Geologists Bull., v. 41, p. 627-642.

Grieve, R. O., 1955, Leaching (?) of Silurian salt beds in southwestern Ontario as evidence in wells drilled for oil and gas: Canadian Mining Metall. Bull., v. 48, p. 10-16.

Hadley, C. J., 1970, Subsurface studies of reef debris will help reveal the presence and quality of pinnacle reefs in southwestern Ontario: Ont. Petrol. Inst., Proc, Ninth Ann. Conf.

Hill, J. V., 1966, Silurian reef carbonates: Ont. Petrol. Inst., Proc, Fifth Ann. Conf.

Jodry, R. L., 1969, Growth and dolomitization of Silurian reefs, St. Clair County, Michigan: Am. Assoc. Petroleum Geologists Bull., v. 53, p. 957-981.

104 Kay, G. M., 1942, Development of the Northern Allegheny Synclinorium and adjoining regions: Geol. Soc. America Bull., v. 53, p. 1601-1658.

Kay, G. M., and Colbert, E. H., 1965, "Stratigraphy and Life History": New York, John Wiley and Sons, Inc., 736 p.

Kilbourne, D. E., 1948, Origin of the Howell anticline: Compass, v. 25, no. 3, p. 132-138.

King, P. B., 1959, "The Evolution of North America": New Jersey, Princeton Univ. Press, 190 p.

Krumbein, W. C, Sloss, L. L., and Dapples, E. C, 1949, Sedimentary tectonics and sedimentary environments: Am. Assoc. Petroleum Geologists Bull., v. 33, p. 1859-1891.

Krumbein, W. C, 1956, Regional and local components in facies maps: Am. Assoc. Petroleum Geologists Bull., v. 40, p. 2163-2194.

Landes, K. K., 1945, The Salina and Bass Islands rocks in the Michigan Basin: U.S. Geol. Surv. Oil and Gas Invs., prelim, map no. 28.

Landes, K. K., 1951, Detroit River Group in the Michigan Basin: U.S. Geol. Survey Circ. 133, 23 p.

Lockett, J. R., 1947, Development of structures in basin areas of northeastern United States: Am. Assoc. Petroleum Geologists Bull., v. 31, p. 429-446.

Logan, W. E., 1854, On the physical structure of the western district of Upper Canada: Can. Journal, v. 3, no. 1, p. 1-2.

Lowenstam, H. A., 1950, Niagaran reefs of the Great Lakes area: Jour. Geol., v. 58, p. 430-487.

Nwachukwu, S. O., Beck, A. E., and Currie, J. B., 1965, Magnetic provinces of Lake Huron and adjoining areas and their geologic significance: Can. Jour. Earth Sci., v. 2, no. 3, p. 227-236.

Pounder, J. A., 1963a, Guelph-Lockport drilling should reveal more reefs: Oil & Gas Jour., Feb. 4, 1963, p. 144-148.

Pounder, J. A., 1963b, Structure, economics play key roles in Guelph-Lockport search: Oil & Gas Jour., Feb. 11, 1963, p. 162-164.

Reavely, G. H., and Winder, C. G., 1961, The Sylvania Sandstone in southwestern Ontario: Canadian Mining Metall. Bull., v. 54, p. 139-142.

Roliff, W. A., 1949, Salina Guelph fields of southwestern Ontario: Am. Assoc. Petroleum Geologists Bull., v. 33, p. 153-188.

Sanford, B. V., 1961, Subsurface stratigraphy of Ordovician rocks in southwestern Ontario: Canada, Geol. Survey, Paper 60-26, 54 p.

Sanford, B. V., 1962, Sources and occurrences of oil and gas in the sedimentary basins of Ontario: Geol. Assoc. Canada Proc, v. 14, p. 59-89.

105 Sanford, B. V., 1965, Salina salt beds southwestern Ontario: Canada, Geol. Surv., Paper 65-9, 7 p.

Sanford, B. V., 1967, Devonian in Ontario and Michigan in Internat. Symposium on the Devonian system, Calgary, Proc. v. 1: Alberta Soc. Petroleum Geologists, p. 973-999.

Sanford, B. V., 1968, Oil and gas in southwestern Ontario in Natural gases in rocks of Paleozoic age: Am. Assoc. Petroleum Geologists Bull., Mem. 9, v. 2, p. 1798-1818.

Sanford, B. V., 1969a, Silurian of southwestern Ontario: Ont. Petrol. Inst. Proc, Eighth Ann. Conf.

Sanford, B. V., 1969b, Geology Toronto-Windsor area, Ontario: Canada Geol. Surv. map 1263A.

Sanford, B. V., and Quillian, R. G., 1959, Subsurface stratigraphy of Upper Cambrian rocks in southwestern Ontario: Canada, Geol. Survey, Paper 58-12, 29 p.

Schuchert, Charles, 1914, Medina and Cataract formations of the Siluric of New York and Ontario: Geol. Soc. America Bull., vol. 25, p. 277-320.

Sharma, G. D., 1966, Geology of Peters reef, St. Clair County, Michigan: Am. Assoc. Petroleum Geologists Bull., v. 50, p. 308-326.

Shouldice, J. R., 1955, Silurian reefs of southwestern Ontario: Canadian Inst. Mining Metall. Bull., v. 48, p. 500-503.

Sloss, L. L., 1953, The significance of evaporites: Jour. Sed. Pet., v. 23, p. 143-161.

Stockwell, C. H., 1965, Structural trends in Canadian Shield: Am. Assoc. Petroleum Geologists Bull., v. 49, p. 887-893.

Sutterlin, P. G. and Brigham, R. J., 1967, Trend surface analysis — a new look at old data: Ont. Petrol. Inst., Proc, Sixth Ann. Conf.

Tilton, et al, 1960, 1000-million-year-old minerals from eastern United States and Canada: Jour. Geophys. Research, v. 65, p. 4173-4179.

Winder, C. G., 1961, "Lexicon of Paleozoic Names in Southwestern Ontario": Toronto, University of Toronto Press, 121 p.

Winder, C. G., 1966, Conodont zones and stratigraphic variability in Upper Devonian rocks, Ontario: Jour. Paleo., v.40, p. 1275-1293.

106