Western Michigan University ScholarWorks at WMU
Dissertations Graduate College
6-2010
Glaciotectonic Deformation along the Valparaiso Upland in Southwest Michigan, USA
Brian C. Bird Western Michigan University
Follow this and additional works at: https://scholarworks.wmich.edu/dissertations
Part of the Geology Commons, Geomorphology Commons, and the Glaciology Commons
Recommended Citation Bird, Brian C., "Glaciotectonic Deformation along the Valparaiso Upland in Southwest Michigan, USA" (2010). Dissertations. 349. https://scholarworks.wmich.edu/dissertations/349
This Dissertation-Open Access is brought to you for free and open access by the Graduate College at ScholarWorks at WMU. It has been accepted for inclusion in Dissertations by an authorized administrator of ScholarWorks at WMU. For more information, please contact [email protected]. GLACIOTECTONIC DEFORMATION ALONG THE VALPARAISO UPLAND IN SOUTHWEST MICHIGAN, USA
by
Brian C. Bird
A Dissertation Submitted to the Faculty of The Graduate College in partial fulfillment of the requirements for the Degree of Doctor of Philosophy Department of Geoscience Advisor: Alan E. Kehew, Ph.D.
Western Michigan University Kalamazoo, Michigan June 2010 UMI Number: 3470399
All rights reserved
INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. UMI' Dissertation Publishing
UMI 3470399 Copyright 2010 by ProQuest LLC, All rights reserved. This edition of the work is protected against unauthorized copying under Title 17, United States Code.
uestA ®
ProQuest LLC 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106-1346 GLACIOTECTONIC DEFORMATION ALONG THE VALPARAISO UPLAND IN SOUTHWEST MICHIGAN, USA
Brian C. Bird, Ph.D.
Western Michigan University, 2010
Glaciotectonic deformation has been observed and analyzed across parts of the Valparaiso, Kendall, and Inner Kalamazoo Moraines in southwest Michigan,
USA. Deformation structures such as folds and faults have been analyzed using techniques typically used by structural geologist along with the fabric of elongated clasts in the surficial diamicton. The structures are consistent with simple shear, horizontal compressional stresses, and pure shear. A series of drumlin fields exist along the western boundary of the study area. Direct investigation of the deformed sediments was conducted at a gravel pit in one of these drumlins. Lacustrine derived sediments interbedded with coarse sand and gravel dominate the stratigraphy of the study area. The rheology of the sediments is responsible for the deformation observed as the Lake Michigan Lobe (LML) readvanced across the area to the
Kalamazoo Moraine. The LML readvanced into a proglacial lake system impounded between the ice margin and the Kalamazoo Moraine. The lacustrine sediments, restricting the flow of subglacial water, increases the pore water pressure, which in turn decreased the shear strength of the sediments promoting the deformation of these sediments. Advancing ice deformed sediments in two stages, proglacially along a decollement at the margin, then subglacially as ice overrode the sediments. Copyright by Brian C. Bird 2010 ACKNOWLEDGMENTS
I would like to acknowledge the Department of Geosciences of Western Michigan University. The department offered a challenging yet enjoyable experience. The knowledge and experience gained is invaluable.
I would like to first acknowledge my advisor Dr. Alan Kehew for his patience and help during this process. I truly appreciate it. I also would like to acknowledge
Dr. Ronald Chase and Dr. William Sauck for their help being on my committee. I appreciate their assistance and thoughts. Lastly, for the committee I must acknowledge my best friend Dr. Andrew L. Kozlowski. Having been friends for many years there is little we have not been through. I am haunted by the near death experiences of drilling wells while at the same time amused at the never-ending comedy. I would not be here without him.
Though not on my committee and never had taken one of her classes I must acknowledge Dr. Carla Koretsky for her tremendous help with my program of study and the graduate college.
Finally, I need to acknowledge my loving wife Kelly McHugh. Thanks for the help, love and understanding.
Brian C. Bird
ii TABLE OF CONTENTS
ACKNOWLEDGMENTS ii
LIST OF TABLES vii
LIST OF FIGURES viii
CHAPTER
1 INTRODUCTION 1
Location 1
Nature of the Problem 6
Background of Glaciotectonic Deformation 9
Till Fabric 12
Folds 14
Faults 17
2 PREVIOUS WORK IB
Surficial Mapping 18
Geologic Setting 20
Glacial Landforms 22
Southwest Michigan Chronostratigraphy 25
Deformation of Glacial Sediments 30
3 METHODS OF INQUIRY 34
Data Collection 34
Fold and Fault Data 35
iii Table of Contents - Continued
CHAPTER
Fabric Data 37
Data Analysis 39
Analysis Procedure 40
Folds 40
Faults 42
Till Fabric 43
4 RESULTS 46
Tacy Pit 48
Stratigraphy 48
Deformation 51
White Pit 56
Stratigraphy 57
Deformation 57
Wyoming Pit 58
Stratigraphy 58
Deformation 62
Michigan Paving Pit 65
Stratigraphy 66
Deformation 68
iv Table of Contents - Continued
CHAPTER
Kusmack Pit 69
Stratigraphy 69
Deformation 71
Klett Pit 72
Stratigraphy 73
Deformation 74
Cleveland Pit 76
Stratigraphy 76
Deformation 80
Remington Pit 82
Stratigraphy 83
Deformation 84
Smith Pit 87
Stratigraphy 87
Deformation 88
Brewster Pit 94
Stratigraphy 94
Deformation 94
JD Pit 98
v Table of Contents - Continued
CHAPTER
Stratigraphy 98
Deformation 98
Compilation Data 103
5 DISCUSSION AND CONCLUSIONS 107
Ice Melt-Out 107
Ductile Deformation of Glacial Sediments 108
Sediment Grain-Size Ill
Pore Water Pressure 111
Topography 112
Stresses Under the LML 115
Deformation in Drumlins 120
Sequence of Events 126
Conclusions 130
APPENDIX 132
BIBLIOGRAPHY 140
vi LIST OF TABLES
4.1: Stratigraphic Code for Glacial Sediments Listed in Stratigraphic Logs ....47
vii LIST OF FIGURES
1.1: Extent of Laurentide Ice Sheet and Later Ice Lobes 2
1.2: Location of Aggregate Mining Operations within the Study Area 3
1.3: Traditionally Mapped Moraines in Southwest Michigan 5
1.4: Digital Elevation Model of Southwest Michigan 6
1.5: Fold Nomenclature 8
1.6: Deformation Style 11
1.7: Deformation of the Saugatuck Till across the Study Area 13
1.8: Folds Influenced by Progressively Increasing Shear 16
1.9: Conjugate Fault Pair 17
2.1: Aggregate Mining Operation Locations along the Traditionally Mapped
Valparaiso, Kendall, and Inner Kalamazoo Moraines 19
2.2: Bedrock Geology of Michigan 21
2.3: Features A and B on the Inner Kalamazoo Moraine within the Study Area 24
2.4: Time-Distance Plot of Glacial Chronology for Western Lake Michigan Lobe... 27
2.5: Time-Distance Plot of Glacial Chronology Including the Eastern and
Northern Great Lakes 28
2.6: Approximate Age of Major Advances by the Lake Michigan Lobe 29
3.1: Gravel Pit Locations in the Study Area 35
3.33.2: Fol71- Diagrad Axism Measuremen Constructiot nb y Excavation of the Hinge Area 4318
viii List of Figures - Continued
3.4: Relationship of Applied Stress to Fold Axis Orientation 42
3.5: Orientation of Principle Stress Directions for a Faulted Block 44
3.6: Trend and Plunge Stereonet Plot of Elongated Clasts in Diamicton 45
4.1: Aggregate Mining Operations in the Study Area 46
4.2: Key to Values Presented with Fabric and Fold Axes Data 48
4.3: Tacy Pit Location 49
4.4: Tacy Pit Stratigraphic Log 50
4.5: Photographs of Diamicton in Tacy Pit 52
4.6: Clast Fabric Plot of Diamicton in Tacy Pit 53
4.7: Folds in the Tacy Pit 54
4.8: White Pit Location 56
4.9: White Pit Stratigraphic Log 57
4.10: Folds in the White Pit 59
4.11: Wyoming Pit Location 60
4.12: Wyoming Pit Stratigraphic Log 61
4.13: Deformation Observed in the Wyoming Pit 63
4.14: Wyoming Pit Structural Data 64
4.15: Michigan Paving Location 65
4.16: Michigan Paving Pit Stratigraphic Log 66
ix List of Figures - Continued
4.17: Deformation Observed in the Michigan Paving Pit 67
4.18: Stereonet Plots of Folds in the Michigan Paving Pit 68
4.19: Kusmack Pit Location 70
4.20: Kusmack Pit Stratigraphic Log 71
4.21: Deformation Observed in the Kusmack Pit 71
4.22: Klett Pit Location 72
4.23: Klett Pit Stratigraphic Log 74
4.24: Structural Data of Folds at the Base of the Diamicton in the Klett Pit 75
4.25: Clast Fabric Analysis of the Diamicton in the Klett Pit 76
4.26: Cleveland Pit Location 77
4.27: Cleveland Pit Stratigraphic Log 78
4.28: Deformation Observed in the Cleveland Pit 79
4.29: Deformation Observed in the Cleveland Pit 81
4.30: Clast Fabric Analysis of the Diamicton in the Cleveland Pit 82
4.31: Remington Pit Stratigraphic Log 84
4.32: Deformation Observed in the Remington Pit 85
4.33: Structural Data of Deformation in the Remington Pit 86
4.34: Clast Fabric Analysis of Diamicton in the Remington Pit 87
4.35: Smith Pit Location 89
4.36: Smith Pit Stratigraphic Log 90
x List of Figures - Continued
4.37: Deformation Observed in the Smith Pit 91
4.38: Clast Fabric Analysis of Diamicton in the Smith Pit 93
4.39: Clay Partings Perpendicular to Bedding found in the Hinge Area of a Fold .... 93
4.40: Brewster Pit Location 95
4.41: Brewster Pit Stratigraphic Log 96
4.42: Folds Observed in the Brewster Pit 97
4.43: JD Pit Location 99
4.44: JD Pit Stratigraphic Log 100
4.45: Deformation Observed in the JD Pit 101
4.46: Conjugate Reverse Faults in the JD Pit 102
4.47: Compilation of Data for the Cleveland Pit 103
4.48: Compilation of Data for the Remington Pit 104
4.49: Compilation of Data for the Klett Pit 104
4.50: Compilation of Data for the Smith Pit 105
4.51: Compilation of Data for the Tacy Pit 105
4.52: All Fold Axes Measured Defining a Shear Plane with a Strike of 40°
and a Dip of 20° to the Southeast 106
5.1: Regional Map of Specific Gravel Pit Exposures 109
5.2: Brittle Deformation Displayed in the Kusmack Pit 110
XI List of Figures - Continued
5.3: Sequence of Ice Melt-Out Causing Brittle Deformation 110
5.4: Examples of Till in the Study Area Displaying Evidence of being Coupled
and Decoupled to the Bed 114
5.5: Evidence of Shear in Glacial Sediments across the Study Area 116
5.6: Evidence of Proglacial Deformation of Glacial Sediments in Eastern
England 117
5.7: Drumlin Zone 118
5.8: Net Stress Exerted on the Substrate by Loading of a Glacier 119
5.9: Map of Principal Stress Directions, Inferred Ice Flow Directions, and
Drumlin Axes Orientations 121
5.10: Alignment of Topography with Principal Stress Directions and Clast Fabric. 122
5.11: Axial Planar Partings Perpendicular to Bedding Found in the Smith Pit 123
5.12: Divergence of Ice Flow Path around a Rigid Gravel Core 124
5.13: Clastic Dike Cross-Cutting Horizontal Beds in the Smith Pit 124
5.14: Sequence of Glacial Advance of the Gleann Einich Outlet Glacier 125
5.165.15: ModeScotlanSequencl odfe thofe DeformatioSubglacial nDeforming/Slidin during the Loch gLomon Bed d Readvance in Central 121298
xii CHAPTER 1
INTRODUCTION
Location
The state of Michigan owes its diverse landscape to the tremendous forces of glacial ice and water. As evident from the map in Figure 1.1 the entire state of
Michigan was covered by glaciers during the Pleistocene. The interplay of ice and water, along with the underlying bedrock surface, built topographic highs, sculpted out low areas, and flooded other areas with lakes and rivers. The chronicles of these events are preserved in the sediments and landforms the glaciers left behind.
The shape of the Lower Peninsula of Michigan reveals a clue to shape of the glacial ice that influenced the state. In its waning phase, the Laurentide Ice Sheet developed series of distinct ice streams or lobes. These lobes were dynamically similar in flow and behavior. The modern shoreline of Lower Michigan basically delineates the different lobes of the Laurentide Ice Sheet that once occupied the present lake basins. Named for the bodies of water that remain behind, the Lake
Michigan Lobe, the Saginaw Lobe, the Huron Lobe, and Erie Lobe were bodies of glacial ice that moved generally from the north spreading out across the land toward the south. Each lobe possessed unique characteristics resulting from a myriad of causes including but not limited to, ice flow velocity, thickness of the ice, the substrate beneath the ice, slope of the bed, and meltwater influences.
1 86WW 84WW 82WW 80WW
Figure 1.1: Extent of Laurentide Ice Sheet and Later Ice Lobes. (Modified from Fullerton, 1986)
Spreading laterally out of the Lake Michigan Basin, into western Michigan, the Lake Michigan Lobe of the Laurentide Ice Sheet occupied the study area during the final stages of glaciation (Leverett and Taylor, 1915).
For this investigation, the area of interest is the southwestern portion of
Michigan. The bulk of this area is contained within Van Buren County but extends
2 into the southeastern portion of Allegan County to the north and the northern portion of Berrien County to the south and west. The arcuate study area extends from a gravel pit location northeast of the city of Allegan, Michigan, 62 kilometers southwest to a sand and gravel pit located south of the town of Coloma, Michigan.
This trend follows a broad, dissected ridge that is about 20 kilometers wide. Within this area, eleven aggregate operations were investigated. The study area is delineated in Figure 1.2 and displays the aggregate operation locations.
Figure 1.2: Location of Aggregate Mining Operations within the Study Area.
3 The modern landscape of southwest Michigan is the result of retreating glacial ice punctuated by stagnation and readvance. As the Lake Michigan Lobe flowed along the modern day Lake Michigan basin, it spread laterally. The resulting landscape in the study area indicates that during the late Wisconsin Stage, the ice was flowing generally from northwest to southeast (Fullerton, 1986; Leverett and Taylor
1915). Leverett and Taylor (1915) investigated this region of Michigan and mapped a series of moraines. These moraines trend generally northeast to southwest and were interpreted by Leverett and Taylor (1915) as recessional moraines that are progressively younger from east to west (Fig. 1.3). For this investigation, the focus of inquiry is on what Leverett and Taylor (1915) refer to as the Valparaiso Moraine, which is situated between the older Kalamazoo Moraine to the east and the younger
Lake Border Moraine to the west.
The topography of the area ranges from a high more than 300 meters ASL on the Kalamazoo Moraine to the modern Lake Michigan level of approximately 175 meters. Figure 1.4 shows the digital elevation model of the study area. On this elevation map, the trend of the upland ridges, generally coinciding with the moraines mapped by Leverett and Taylor (1915), is apparent. Valleys between the ridges have been interpreted as meltwater channels sculpted by water from not only the Lake
Michigan Lobe but also the Saginaw lobe (Bird, 2005; Kozlowski, 2004; Kehew et al., 2005)
4 Figure 1.3: Traditionally Mapped Moraines in Southwest Michigan. (Modified from Leverett and Taylor, 1915)
5 M-^ ••• -^'"Allegan
f Cf^V
/ .4 i • * 4 •* i Van Buren " A ../ . Vf vV r • > • | fV i / e. ft ' ' Berrien ' ...A * 7 « > "V" "mi Figure 1.4: Digital Elevation Model of Southwest Michigan. Nature of the Problem Bird (2005) investigated the surficial geology of the Paw Paw, Decatur, and Lawrence 7.5-minute USGS quadrangles, which contained gravel pits displaying deformed sediments, at times extending to greater than five meters in depth below the surface. The deformation observed included faulted and folded sediment. Folds in the study area ranged from simple upright folds to more complex sheath folds (Fig 1.5). In some pits, shear planes not unlike thrust faults were observed. Reconnaissance observations in the surrounding counties yielded examples of similar deformed sediments. An explanation of the deformation as the result of melting of buried ice and at one site slumping at the margin of a kame was presented on a Friends of the Pleistocene field trip led by Kevin Kincaire and Byron Stone (2003). One stop at a pit near Watervliet, MI displayed deformed sediment that did not appear to be consistent with melt-out of buried ice. Discussion led to the realization that a detailed structural analysis was never completed in this area and such an analysis was needed to determine the true cause of the deformation. The primary objective of this research is to collect and analyze structural data of the deformed features found in the glacial sediment throughout the area. Structural analysis uses the orientation of the strain features to ascertain the stress direction responsible for creating the strain. From the analysis, a hypothetical model of the cause of the deformation can be made. 7 Open Isoclinal Tight Sheath Upright r\T\ Symmetrical Recumbent Asymmetrical Over turned Figure 1.5: Fold Nomenclature. (Modified from Benn and Evans, 1998) The problem to be addressed is to determine if the observed deformation was the result of melting of buried ice as proposed by Stone et al. (2003) or direct interaction with active glacial ice. If the analysis suggests direct interaction with dynamic glacial ice, a further question is whether the deformation was the result of subglacial deformation or proglacial deformation. The structural data analysis will yield a model of deformation as well as implications of the glacial landscape development for the area. Background of Glaciotectonic Deformation The deformation of glacial sediment consist of two basic types (Fig. 1.6a), brittle or ductile. Structures such as folds, fault and shear planes, and clast fabrics are the products of deformation. The orientation of the principal stress direction can be used to reconstruct stress fields or patterns responsible for the observed strain (Fig. 1.6b). In order to assess the cause of a particular deformation structure, the cause must be based on the geometry of structures as well as the reconstruction of the type of stress field necessary to form them (McCarroll and Rijsdijk, 2003). Brittle deformation occurs when materials fail along a fault or fracture. Faulting observed in the study area includes both normal faults and reverse faults. Normal faults are the result of extensional forces that will generally elongate the overall package of deformed material. Reverse faulting is the opposite with the forces involved being compressional and causing an overall shortening of the deformed package. Ductile deformation can be divided into two sub-categories: uniform or non- uniform. Non-uniform deformation as a whole is complex, but can be broken down into fundamental parts that are uniform. Uniform ductile deformation occurs in two forms, pure shear and simple shear. The difference between pure and simple shear 9 is based upon the strain ellipse. The strain ellipse is elongated in the direction of maximum strain (Fig. 1.6). Pure shear is the flattening of the strain ellipse without rotation. In this case, the principal stress direction is perpendicular to the long axis of the strain ellipse. An example of pure shear is loading. The force of gravity acting vertically on a glacier will result in compression of the underlying material and cause vertical flattening and horizontal extension. In the case of loading, the principal stress direction is vertical. Pure shear can also be present in the horizontal sense as compressional forces are translated to the margin of the glacier. In this case, folding and reverse faulting will take place. Simple shear results in a strain ellipse that has been rotated. The principal stress direction will be oblique to the strain ellipse. Simple shear structures include shear laminae, folds progressing from upright to overturned to recumbent, as well as isoclinal folds and boudins. A brittle response to simple shear can include faults, thrust wedges, and extensional clastic dykes (McCarroll and Rijsdijk, 2003). Beneath the interior of the glacier the principal stress direction is caused by loading and is straight down. Toward the ice margin, the stress equalizes to a hydrostatic state with the stresses translating to lateral and compressional nearer the margin. In areas where the ice margin becomes pinned, or at the limits of a readvance, however, sequences are likely to be dominated by large-scale compressional deformation structures with the principal stress axes nearly parallel to the direction of ice movement (McCarroll and Rijsdijk, 2003). 10 Simple Shear <$=> !ce ftcw ^reckon flow djreetcn SBtgsS&sheajpsaSsJ© Sarpto shear pratde WitftdlKfete Ductile Deformation o Pure Shear Ml M&rix reaSgnmsnl Bstta ccfnptessnjnai ftactmes Horizontal /•Ai Fc > J \ M glacsar fcreW-d Vertical 30 Pertstfy erw&r graves diantci dc!ormaSon \f sag nepasOi&ia) as in a ketSe ho:o Figure 1.6: Deformation Style. (Modified from Benn and Evans, 1996 and McCarroll and Rijsdijk, 2003) Till Fabric The diamicton observed at or near the land surface across the study area was defined by Monaghan et al. (1986) as the Saugatuck Till. This till is between 1.0 and 2.0 meters thick, matrix supported, reddish-brown in color, and contains occasional boulder-sized clasts greater than 0.5 m in diameter. The Saugatuck Till in the study area is interpreted to be a subglacial till. Across the area, the till varies from a matrix supported massive diamicton as described above to a sheared till that incorporates layers of deformed sediment and rotated boulders. Deformation till is defined as sediment that has been completely or largely homogenized by shearing in a subglacial deforming layer, while a glacitectonite is sediment that has been deformed by subglacial shearing but retains some of the structural characteristics of the parent material (Benn and Evans, 1996). The Saugatuck Till fits the definition of deformation till in some areas while elsewhere it may be defined as a glacitectonite as it clearly displays deformation in terms of folding, incorporation of underlying material, as well as streaking and shearing of sand lenses within the diamicton (Hart and Boulton, 1991). Figure 1.7 includes a series of photographs of the diamicton in the study area displaying deformation structures. 12 During subglacial deformation, elongated clasts will align with their long axes parallel to the applied total stress; thus, the fabric of these clasts reflects the strain of the sediment (Carr and Rose, 2003). The direction of shear is closely related to the ice movement; however, Benn (1994) asserts that the direction of shear in the deforming layer varies spatially and may diverge from the flow of the overlying ice. For the study area, the divergence of the shear direction preserved in the fabric does not deviate significantly from the interpreted ice flow direction. Many have debated the value of using the statistical clustering strength by way of eigenvalues and eigenvectors of till fabric to determine how the observed till was formed (Bennett et al., 1999; Larsen and Piotrowski, 2003). The statistical determination of the origin of a till is limited (Bennett et al., 1999). For this reason the analysis of the till fabric is best used as an indicator of relative strain direction and not the till origin. Classification of the till origin is based upon the deformation structures within the till and not the statistical analysis of eigenvectors and eigenvalues (Bennett et al., 1999; Hicock et al., 1996; Benn and Evans, 1996). Folds Folds of one type or another were present in each gravel pit in the study area. Folds can provide information about strain history and mechanical conditions of the substrate during folding, as well as the mechanics of folding itself (Hudleston, 1986). The plot of fold axes on a stereonet can give an indication of the stresses that created the fold. As shearing increases, the fold can become more asymmetrical. The asymmetry, being a product of the shearing, can help define the shear surface (Hudleston, 1986). In this study, the shear surface is usually the till. In one 14 exposure, there is a shear surface that propagates through the underlying sediments. This shear surface is separate from the till but is similar in attitude. When plotted, the trends of the minor fold axes spread across the stereonet and define a great circle or girdle, which will in turn define the shear plane (Mies, 1991; Hansen, 1971). The trend of the girdle can approximate the direction of principal stress responsible for the fold. Figure 1.8 shows folds influenced by progressively increasing shear. For each scenario, the resulting stereonet plot of the fold axes is also shown. For ease of reference to the diagram, the stereonet plot letter designations will be used to describe folds as categories. Fold Category A is a typical cylindrical fold, in which the fold axes will group tightly and plot orthogonal to the shear direction of the fold. Category B is a noncylindrical fold, where the fold axes are still nearly orthogonal to the direction of shear yet are starting to disperse along the shear plane attitude. Category C represents a further stretching of the hinge of the fold and the axes measured will spread along the attitude of the shear plane. Finally, the greatest amount of shear, Category D, is a sheath fold where the fold axes will concentrate in an area that is aligned with the direction of the shear. As demonstrated by this diagram, the plot of fold axes can be used to help define the shear plane that created the folds. In this study the plot of the fold axes was used in conjunction with the attitude of the till bed and the fabric of the till to help define the stress directions. 15 • Fold axis Shear Plane a. Fold Axes Plot of 1 a. Fold Axes Plot of 2 a. Fold Axes Plot of 3 a. Fold Axes Plot of 4 Figure 1.8: Folds Influenced by Progressively Increasing Shear (Modified from Hudleston, 1986) 16 Faults Two conjugate faults were observed and measured. A pair of conjugate faults is useful to determine orientation of the principal stresses involved in deforming the material (Fig. 1.9). There were many faults observed in the pits; however, only two were conjugate. Other faults can be used to determine the principal stresses but a lineation and a displacement marker must be present on the fault surface, of which none were found. When the conjugate faults are plotted as great circles on the equal area stereonet the point of intersection represents the median value of stress (a2). The maximum (al) and minimum (a3) stress directions are on a plane 90° from this intersection. The primary stress direction (al) will bisect the acute angle (near 60°) of the fault planes while a3 will bisect the obtuse angle. The relative movement along the fault planes must be noted to assure the correct alignment of al and a3. i Gl Block Diagram of Conjugate Faults Stereonet Plot of Conjugate Faults Figure 1.9: Conjugate Fault Pair. 17 CHAPTER 2 PREVIOUS WORK Surficial Mapping Leverett and Taylor (1915) mapped the glacial geology of Indiana and Michigan in an effort to interpret the geologic history of the Great Lakes. In the area of this study in southwestern Michigan, their efforts produced the first detailed account of the topography, framework, and distribution of morainic systems, outwash plains and lakes. The moraines mapped by Leverett and Taylor, as well as the aggregate mining operations investigated, are shown in Figure 2.1. The bulk of the investigation occurs along the traditionally mapped Valparaiso Moraine with two sites in the Kalamazoo Moraine and one in the Kendall Moraine. Martin (1955) revised Leverett and Taylor's map to produce a new state map of the glacial geology of Michigan. Farrand and Bell (1982) further revised the state map to include the surface texture of the sediment, based on soils mapping, with the glacial landforms. 18 Figure 2.1: Aggregate Mining Operation Locations along the Traditionally Mapped Valparaiso, Kendall, and Inner Kalamazoo Moraines. Surficial glacial deposit maps have been produced for Berrien, Van Buren, and Allegan Counties. Stone (2003) created a 1:200,000 scale surficial geology map of Berrien County. This map shows the distribution of morphosequences throughout the county, which are bodies of stratified meltwater sediments contained in a 19 continuum of landforms, grading from ice proximal forms to distal forms, and that were deposited as a single event graded to a specific base level (Stone et al., 2003). Terwilliger (1954) produced a surface geology map of Van Buren County in an assessment of the groundwater resources of the county. Results of this project were not only the distribution of sediment across the area but also an interpretation of ice retreat across Van Buren County. Kehew (2002) produced a glacial terrain map of Van Buren County at the 1:62,5000 scale. Gephart and Larson (1982) mapped Allegan County. Various glacial terrain maps on the 1:24,000 scale have been completed in Allegan County (Kozlowski, 2000, Durham, 2004; Kehew, unpublished;). Numerous investigators have studied the glacial stratigraphy of southwestern Michigan (Monaghan et al, 1986; Monaghan and Larson, 1986, Kozlowski et al., 2005; Beukema, 2003; Bird, 2005; Kehew et al., 2005). The correlation and chronology of glacial landforms of the Great Lakes area has been investigated by many (Leverett and Taylor, 1915; Zumberge, 1960; Farrand and Eschmann, 1974; Fullerton 1986; Mickelson et al., 1982; Hansel et al., 1985; Monaghan and Larson, 1986; Beukema, 2003; Bird, 2005; Kehew et al., 2005; Kozlowski et al., 2005). Although analysis of deformation of glacial sediment has not been published within the defined study area, deformation has been observed and studied in Wisconsin (Stanford & Mickelson 1985), southern Ontario (Hicock and Dreimanis 1985; Albino and Dreimanis 1988; Menzies 1990), Ilinois (Higuera-Diaz et al., 2008), and western Michigan (Rieck et al., 1991; Larson et al., 2003) Geologic Setting No bedrock is exposed at the surface in the study area. The glacial drift ranges in thickness from 25 m to nearly 170 m. The bedrock immediately below the 20 drift is the Coldwater Shale. The Coldwater Shale is early Mississippian in age and varies in color from gray, blue-gray, to red, is fossiliferous and locally in the western part of the state contains limestone and dolomite (Catacosinos et al., 2001). The basic structure of the underlying Paleozoic sedimentary rocks is a large intracratonic basin known as the Michigan Basin (Fig. 2.2). Legend RED BEDS GRAND RIVER FORMATION SAGINAW FORMATION ; BAYPORT LIMESTONE MICHIGAN FORMATION j MARSHALL FORMATION ! COLDWATER SHALE SUNBURY SHALE 1 BEREA SANDSTONE & BEDFORO L ] BEDFORD SHALE : ELLSWORTH SHALE "j ANTRIM SHALE Hi TRAVERSE GROUP IHfl BELL SHALE DUNDEE LIMESTONE •I DETROIT RIVER GROUP ~j SYLVANIA SANDSTONE MACKINAC BRECCIA BOIS BLANC FORMATION GARDEN ISLAND FORMATION i BASS ISLAND GROUP SALINA GROUP 0 SO Km 11 I I 111 11 Figure 2.2: Bedrock Geology of Michigan. (Modified from MDEQ 1987 Bedrock Geology) 21 Glacial Landforms The landforms created by the Lake Michigan Lobe (LML) in the study area reflect the direction of the ice flow. The northeast-southwest trending ridges have traditionally been interpreted as a series of end moraines that preserve the marginal position of the LML as it paused or readvanced in its general retreat from the area (Leverett and Taylor, 1915). The moraines encompassed within the study area include the Kalamazoo, Kendall, and Valparaiso morainal systems. These systems are inferred to be marginal positions as ice retreated from east (oldest) to west (youngest) across the area, The nature of the traditionally mapped moraines has been called into question (Kehew and Kozlowski, 2001; Kehew et al., 2001, Kehew et al., 2005; Bird, 2005). Amongst these moraines are low relief areas that have been mapped as outwash and lacustrine deposits (Leverett and Taylor, 1915; Terwilliger, 1954; Martin, 1955; Farrand and Bell, 1982; Kehew et al. 2002, 2005; Kozlowski, 2004). The easternmost moraine of the Lake Michigan Lobe in the study area is the Kalamazoo Moraine. The Kalamazoo Moraine reaches an elevation of more than 300 m and a maximum drift thickness of 180 m. Leverett and Taylor (1915) describe the Kalamazoo Moraine in this area as two well-defined ridges, two miles in width, separated up to four miles apart. The Kalamazoo morainic system of the LML extends from Prairieville, Michigan southwestward into Indiana where its traceable extent is truncated near Warren, Indiana (Leverett and Taylor 1915). The two ridges making up the Kalamazoo morainic system are the Inner Kalamazoo (western ridge) and the Outer Kalamazoo (eastern ridge). In the study area, the Outer Kalamazoo Moraine is characterized by deformed masses of diamicton, bedded sand and gravel, and bedded silt and clay (Kehew et al., 2002). Leverett and Taylor (1915) interpret 22 the Outer Kalamazoo Moraine as a terminal position of the LML. Kehew (2005) cites evidence supporting the terminal nature of the Kalamazoo Moraine including inferred tunnel valleys, distal outwash fans and glaciotectonic features. The Inner Kalamazoo Moraine in Van Buren County has also been interpreted as an ice marginal position (Leverett and Taylor 1915, Terwilliger 1950). This origin for the Inner Kalamazoo Moraine in Van Buren County is called into question by this investigation. The prominent features of the Inner Kalamazoo Moraine in the study area are an upland (A on Figure 2.3), extending from Decatur northeastward about 8 km and Prospect Hill (B on Figure 2.3), which is just south of the town of Paw Paw. These landforms are capped with a discontinuous brown diamicton. Below this diamicton are beds of medium to coarse sand and gravel. The sand and gravel units in contact with the diamicton are deformed. Extending from the northeastern portion of Van Buren County to the southeastern portion of Allegan County is an upland that has been mapped as the Kendall Moraine (Leverett and Taylor, 1915). This is a narrow ridge, with a width of less than 4 km, and a north-south length of approximately 15 km. Leverett and Taylor (1915) originally grouped this moraine with the Kalamazoo morainic system, qualifying that it could also be associated with the Valparaiso Moraine immediately to the west (Leverett and Taylor, 1915). Kozlowski et al. (2005) refer to the Kendall Moraine as the Kendall upland. Along the upland surface are fine-grained silts, clays and to a lesser extent diamicton overlying outwash deposits (Kozlowski et al., 2005). Deformation of the glaciofluvial deposits in exposures along the Kendall Moraine has also been observed (Kozlowski, 2005). Sedimentologic and stratigraphic data from exposures along the Kendall upland indicate the presence of a deep, proglacial lake to the west of the Kalamazoo Moraine (Kozlowski et al, 2005). 23 Allegan Van Buren —-T7? / - j WvH^rrten -T ^ ^ i v.•. , ..» vc* . 'if- fc -. A 3- ' J " - i . , . ' , jr. * ' 'Iff- -J Figure 2.3: Features A and B on the Inner Kalamazoo Moraine within the Study Area. 24 West of the Inner Kalamazoo Moraine is the Valparaiso Moraine. Attaining an elevation of 250 m above mean sea level within the study area, the Valparaiso Moraine is characterized by moderate to low relief with a rolling surface. In this area Leverett and Taylor (1915) describe the Valparaiso Moraine as a broad band of weak ridges, in which sand and gravel predominate over till. In Indiana and Illinois, there are several distinct ridges of the Valparaiso morainic system (Zumberge, 1960). The Valparaiso Moraine as mapped by Leverett and Taylor (1915) in the study area consists of an upland surface discontinuously capped with a brownish red diamicton and bedded sand and gravel with some areas of silts and clay (Kehew et al., 2005). Terwilliger (1954) explains that any assertion made about the recessional or terminal nature of the moraine must be made outside Van Buren County for he discovered no feature indicating retreat and readvance. There are some areas that contain sediment assemblages suggesting an ice-marginal position (Kehew et al. 2001). Drumlins along the western edge of this feature give a definite indication that the area was overridden and is not in its full extent a terminal landform assemblage. West of the Valparaiso Moraine is a broad, gently sloping plane of low relief underlain by plane-to-cross bedded sand and gravel to plane-to-ripple bedded silt to laminated silt and clay (Kehew, 2002). This lowland is interrupted by some higher areas that are capped with diamicton and covered with a veneer of eolian sand in places (Kehew, 2002). This sediment assemblage along with the topographic expression of the area is consistent with glaciofluvial and glaciolacustrine deposition. Southwest Michigan Chronostratigraphy According to Winters and Rieck (1991) glacial ice advanced into the Michigan basin around 26 ka calendar years BP. This assertion is based upon the 25 presence of organic deposits in southern Michigan that date to the mid-Wisconsin episode. The ice continued to advance to its maximum extent between 20 to 19 ka ca BP in east-central Illinois (Hansel and Johnson, 1992). Figure 2.4 illustrates the chronology of glacial events for the Lake Michigan lobe in a transect from south of Peoria, Illinois, to north of the Straits of Mackinac in Michigan as constructed by Johnson et al. (1997). Karrow (2000) modified this time-distance plot in order to bridge between the LML and the eastern-northern Great Lakes (Fig. 2.5). After advancing to its maximum, the LML began to retreat to the north. The general retreat into the Great Lakes watershed was interrupted by several major readvances about 15.5, 13.0, 11.8, and 10.0 ka calendar years BP (Larson and Schaetzl, 2001) (Fig. 2.6). The Kalamazoo Moraine delineates the readvance at 15.5 ka in this area. This assertion is made based on the clay mineralogy, till stratigraphy, and moraine correlation (Monaghan et al., 1986; Monaghan and Larson, 1986; Monaghan, 1990). The Saugatuck Till is a discontinuous, surficial diamicton that can be traced from the shore of Lake Michigan eastward to the Kalamazoo Moraine and is associated with the Lake Border Moraine, and Valparaiso Moraine (Monaghan and Larson, 1986; Kehew et al., 2005; Beukema, 2003; Bird, 2005). Along the shore of Lake Michigan, the Lake Border Moraine in Illinois has been reported to have formed approximately 14 ka (Hansel et al., 1985), however the age of the Lake Border Moraine in Michigan is less constrained. The deformation of the glacial sediments observed in the study area would have occurred between the Milwaukee Phase and the Crown Point Phase about 15 ka 14C BP. as described by Hansel et al. (1997) and Karrow (2000). 26 Gcochronologic Time-distance diagram for a transect from south of Peoria, Illinois, to the Straits of Mackinac and Chronostrati- in Michigan, showing diachronic units and their material referents straits of graphic Units Peoria Chicago Milwaukee Mackinac Figure 2.4: Time-Distance Plot of Glacial Chronology for Western Lake MichigaLobe. (Modified from Johnson et al., 1997) 27 LAKE MICHIGAN LOBE EASTERN-NORTHERN GREAT LAKES to oo ILLINOIS ILLINOIS EPISODE EPISODE Figure 2.5: Time-Distance Plot of Glacial Chronology Including the Eastern and Northern Great Lakes. (Modified from Karrow, 2000) Advance of 15,5 ka Advance of 13 ka Advance of 11.8 k® — — — Advance of 10 ka Figure 2.6: Approximate Age of Major Advances by the Lake Michigan Lobe. (Modified from Larson and Schaetzl, 2001) 29 Deformation of Glacial Sediments "Among all glacial environments, the subglacial systems are probably least understood (Piotrowski, 1997)". The study and subsequent interpretation of folds, faults, and till fabric in glacigenic sediments can provide clues to the stresses induced by glaciers (Benn and Evans, 1996; Boulton, 1996; Hart, 1997; Van der Wateren, 1999; Bennett et al., 2000; Hart and Rose, 2001; van der Meer et al., 2003; Moller, 2006; Evans et al., 2006; Piotrowski et al., 2004, 2006; Phillips et al. 2008). The analysis of the structural data can assess the stresses required to produce the observed deformation. The integration of the fold, fault and till fabric data can yield the direction of the glacial flow as well as an indication of ice dynamics (Benn and Evans, 1993; Phillips et al., 2002; Hiemstra et al., 2005). The dynamics of the ice flow may be the driving force behind the deformation but the role of water in the system is of significance also (Boulton et al., 2001; Boulton and Caban, 1995; Lee and Phillips, 2008; Piotrowski et al., 2009). Water in the deforming substrate will lead to an increase of pore pressure, which can lead to the development of structures usually associated with profound ductile shearing at much lower levels of shear (van der Wateren et al., 2000; Lee and Phillips, 2008). Increased pore pressure beneath the ice can even produce catastrophic failure or water-escape features (Dreimanis and Rappol, 1997; van der Meer et al., 1999; Rijsdijk et al., 1999). The deformation observed in the glacigenic sediment has also been used as evidence for the deforming bed model to explain till formation. In the deforming bed model the saturated, unfrozen, high pore pressure, unconsolidated bed of the glacier deforms and can lead to an increase in velocity of the glacier (Johnson and Hansel, 30 1999; Boulton, 1986; Hart and Boulton, 1991; Roberts and Hart, 2005; Piotrowski et al., 2004; Evans et al., 2006; Lee and Phillips, 2008 ). The deformation of subglacial sediments in relation to ice flow, till formation and sediment transport has been a focus of debate (Hart and Rose, 2001; Piotrowski et al., 2001; van der Meer et al., 2003). Deformation observed in glacial sediments typically involves folding and thrusting. These are comparable to structures found in foreland fold-and thrust belts (Aber et al., 1989; Andersen et al., 2005). The structural analysis of these features has been used to provide evidence for the development of the deformation proglacially (Hart, 1990; Boulton et al., 1999; Phillips et al., 2002; Benediktsson et al., 2008) or subglacially (Benn and Evans, 1996; Boulton et al., 1996; Hart and Rose, 2001; Phillips et al., 2007; Lee and Phillips, 2008). Structures believed to have formed during subglacial deformation have been used to support the deforming bed model (Boulton, 1986; Boulton and Hindmarsh, 1987; Hart and Boulton, 1991; Roberts and Hart, 2005). The presence of deformation structures, especially on the scale of several meters, is used to support the intense disruption of subglacial sediment as a result of glacial shear stresses (Larsen et al., 2006). Recent investigations provide evidence that the subglacial bed is a mosaic of areas of deformation interspersed with undeformed zones that change configuration in time and space (Piotrowski and Kraus, 1997; Piotrowski et al., 2004; Evans et al., 2006; Larsen et al., 2006; Lee and Phillips, 2008). Analysis of the deformation observed has recently led some to the conclusion that the deformation is polyphase or diachronic in that deformation at the margin or even proglacial is then overridden and enhanced by subglacial shearing (Phillips et al., 2002; Benn and Prave, 2006; Phillips et al., 2008; Rijsdijk et al., 2010). 31 Focusing on the region in proximity to the study area, little has been published on the occurrence and analysis of deformed glacigenic sediments. Approximately 200 km north of the study area along the shore of Lake Michigan, deformation has been observed and analyzed (Larson et al., 2003). At this location a 7.5-km stretch of shore was mapped, 15 sites were selected, and the deformation was analyzed. The structures analyzed included both large-scale and small-scale deformation structures such as domes, basins, folds, faults and diapirs (Larson et al. 2003). Rieck et al. (1991) documented numerous faults and folds, some of which have an amplitude of 5 m, in a fen in northern Michigan. Geologic relationships at the site and a generally accepted chronology of the region indicate that deformation was most likely produced by the LML as it advanced across the area. Deformation of glacial sediments has also been observed and analyzed in northeastern Illinois (Higuera-Diaz et al. 2008). The deformation documented occurs in an aggregate mining operation near Spring Grove, IL. The aggregate mine is situated within the LML influence on the Fox Lake Moraine which is in close proximity to the Valparaiso Moraine in Illinois (Stumpf et al., 2004). The nature of the deformation includes various types of folds and faulting along an exposure about one kilometer long. Within the study area, others have observed deformation, although a detailed structural analysis has not been completed. Kincare (1999) observed deformation in an aggregate mining operation in northern Berrien County and commented on the need to conduct a study of the structural deformation. Stone (2003) described deformation in aggregate mines in a feature he described as a kame deposit near Coloma, MI in Berrien County. He attributed the deformation to collapse of the sediments as the ice walls of the kame melted. Kehew et al., (2005) and Bird (2005) observed deformation in two aggregate mining operations, one near Coloma, MI in 32 Berrien County and another in Bloomingdale, MI in Van Buren County. Some preliminary investigation into the deformation commenced but a detailed structural analysis was not conducted at the time. 33 CHAPTER 3 METHODS OF INQUIRY Data Collection Exposures throughout southwest Michigan show deformation of the unconsolidated glacigenic sediment. The focus of this study was the deformation observed along the Valparaiso Moraine. Figure 3.1 shows the location of gravel pit operations where the deformation was observed and measured. These operations were discovered through previous research of the author, various geologic field trips, other researchers working in the area, general reconnaissance of the area using USGS 7.5-minute quadrangles and conversing with aggregate operators and landowners. Two major criteria were used to select an area to investigate. The first criterion was if deformed sediment was present in the aggregate mining excavation. The second criterion was the relative age of the excavation face exposure. Fresh surfaces within the excavation were less likely to be influenced by slumping and also the amount of exposure is far greater in extent, which can aid in the evaluation of how pervasive the deformation is. Notwithstanding, permission to access the excavation was also a concern as two aggregate operations with deformation denied entrance. 34 Figure 3.1: Gravel Pit Locations in the Study Area. Fold and Fault Data Deformation structures throughout this area include various types of folds. These folds range in size, style and complexity from relatively simple, upright folds to increasingly more complex folds such as sheath folds. Faults in the glacial sediments are often associated with these folds. Like the folds, the complexity of the faults exhibits a broad range from simple dip-slip faults to a complex undulating 35 shear surface that behaves similar to a thrust fault. The structures were investigated using conventional structural geology techniques with the use of a Brunton type pocket transit (Brunton). The Brunton was used to find the horizontal strike of the bedding within the fold limbs or in the case of a fault, the strike of the slip plane. The clinometer of the Brunton was used to find the angle of the dip of these planar surfaces. Folds observed and measured in the study area are described as large folds and minor folds. All folds possessing amplitudes greater than 0.25 m are referred to as large folds, whereas folds with amplitudes less than 0.25 m are referred to as minor folds. Generally, the scale of the large folds was such that the bedding planes of the fold limbs were easily identified and the strike and dip could be measured directly with the Brunton. These data could then be plotted on an equal area stereonet to determine the fold axis trace. If possible, the fold axis trace was also measured directly in the field. The limbs of minor folds were often too small to measure directly with the Brunton and maintain a required level of accuracy. For minor folds, the fold axis and/or axial plane were measured through careful excavation of the fold hinge area. The nature of the fold and the cohesiveness of the sediment determined how the fold was excavated. Figure 3.2 illustrates the various ways a minor fold was excavated to expose the hinge area thus allowing for the accurate measurement of the fold axis trace. The simplest method of excavation was to locate a folded bed of sediment that was cohesive enough to remain intact (typically clay or silt) with an adjacent bed that was weak enough to easily remove with a trowel (typically sand or gravel). As the weak bed is removed the cohesive bed remains and the hinge of the fold is exposed in all three dimensions. From this three-dimensional fold, a wooden dowel was inserted parallel to the fold axis and the trend and plunge of the axis could be measured (Figure 3.2a). Another 36 method of fold axis trace measurement was to first etch a reference line in the sediment to bisect the fold at the hinge. This line marks where the axial plane intersects the wall of the pit excavation. By following the hinge of a bed, the fold axis could then be measured with the Brunton by measuring the azimuth of the trend of the axis and then the plunge of the axis. After the measurement was taken, the entire fold was excavated in order to confirm that the attitude of the axis was correct (Figure 3.2b). Occasionally the fold was excavated along a horizontal plane if the sand was too dry to be cohesive. The same vertical reference line was drawn; however, this time a horizontal line, determined by the Brunton, is also traced. The resulting pattern of the beds allowed for a measurement of the fold axis (Figure 3.2c). Fabric Data Some of the aggregate pits contained diamicton. Typically, this diamicton was at the surface of the pit excavation; however, at times there were other diamicton beds at depth. At some locations, the diamicton at the surface was scraped away as overburden. If the diamicton was exposed in the pit, the fabric of the elongated pebbles was measured with a Brunton and wooden dowel. The face to be measured was scrutinized to determine if it was naturally occurring and in place as deposited by the glacier. At two locations, the surface material was a diamicton that was removed from one area of the pit and transported to another. The resulting bed of diamicton appeared to be in place. At one of these locations, the pit operator made the author aware of the transported material. At the other location, human transported material was discovered after a piece of tire was unearthed within the bed. The diamicton measured was also examined to be sure the face of the excavation had not slumped after the face was exposed. 37 Fold Axis Vertical Reference Line Brunton Compass Figure 3.2: Fold Axis Measurement by Excavation of the Hinge Area. 38 Fabric data were collected by measuring the azimuth or bearing direction of the dip of the elongated clast as well as the magnitude of that dip. The long axis of the clast (a) was at least twice the length of the short axis (c) (Larsen and Piotrowski, 2003). Typically, the ratio far exceeded the 2:1 ratio of a/c. The clasts measured for the fabric generally range from 3 to 16 cm in diameter. Carr and Goddard (2007), working at the Vestari-Hagafellsjokull, in Iceland, found that there is a correlation between particle size and divergence of the fabric from the ice flow direction. Of the size fractions measured, the largest, 1.6 to 3.0 cm, proved to have the strongest alignment to the ice flow direction. Benn and Evans (1996), using 3 to 100 cm clasts, obtained similar results. The fabric data were collected by first removing the outermost, exposed sediment of the pit face to encounter clasts that were in intimate contact with sound, unweathered, matrix material. The long axis of the elongated clast was noted and the clast was then carefully removed in order to expose the true orientation of the long axis. A sharpened wooden dowel was then inserted in to the void left by the removal of the clast such that the orientation of the dowel parallels the long axis (Andrews, 1970). The Brunton was used to measure the azimuth of the plunge of the axis as well as the magnitude of that plunge. The diamicton face was examined vertically and laterally for elongated clasts. The fabric data collected represents an area at least 20 m2 of the exposed diamicton. Typically, about 50 clasts were sampled with two sites having over 100 clasts in the sample set. Data Analysis After collecting the structural data from the folds, faults, and fabric in the field, the data were plotted and processed with Stereonet for Windows 1.2© by Richard Allmendinger (2003) and Rockworks99©. Data were plotted on equal-area, lower-hemisphere Schmidt nets (stereonets). Some pits contain faulted sediments. In this case, the attitudes of the slip surfaces were measured and also plotted on equal- area, lower-hemisphere Schmidt nets. The attitude of the slip surface was measured in a similar way to the folds in that the sediment was carefully excavated away to reveal the true trend of the dipping surface. Analysis Procedure Folds The strike and dip data gathered from the limbs of large folds were plotted on the stereonet as great circles. These great circles were converted to poles by finding the line perpendicular to each surface. The scatter plot of points representing the poles on the stereonet plot can be connected by a great circle. This great circle defines the n- girdle of the fold. The great circle created from the poles to bedding yields the fold axis, also known as the Ti-axis, which is the line perpendicular to the girdle. The plot of the 71-girdle of the fold is referred to as a 71 diagram (Marshak and Mitra, 1988). Figure 3.3 illustrates a fold with the great circle, poles to bedding and n diagram stereonet plots. For the minor folds encountered, the fold axis or axial plane was plotted directly on to the stereonet. The fold axis is significant in that not only can the plunge of the fold be determined but also the stresses involved in the creation the fold can be assessed. With the girdle being perpendicular to n, or the fold axis, its strike defines the direction of the stresses. Figure 3.4 demonstrates this concept. 40 Poles to Bedding Fold axis Id axis Stereonet Plot of Stereonet Plot of Bedding Planes of Poles to Bedding Fold Planes of Fold Figure 3.3: n- Diagram Construction. 41 Fold Figure 3.4: Relationship of Applied Stress to Fold Axis Orientation. Faults In a few of the excavations reverse faults were encountered. Without slip lineations preserved in the unconsolidated sediment, the most useful types of faults to determine stress directions are conjugate faults. Following Anderson's theory of faulting (Anderson 1942), a set of conjugate faults can be used to determine the stress directions involved in creating the fault. With a tri-axial set of stress directions the principal stress al is the greatest stress and oriented 30° to 45° to the fault, c2 is 42 intermediate and lies in the fault plane, and a3 is the minimum stress and is oriented 45° to 60° to the fault. This concept is illustrated in Figure 3.5. Noting the style of fault is important so that the sense of motion can be transferred to the stereonet; this is crucial to the determination of the ol and a3 directions. With a pair of conjugate faults, the a2 direction would be parallel to the intersection of the two faults. To determine the al and a3 stress directions the fault surfaces are plotted as great circles on the stereonet. A great circle is drawn 90° from the intersection of the conjugate faults. The ol and a3 directions lie upon this great circle. The acute bisectrix between the faults defines the al direction while the obtuse bisectrix defines the a3 direction. Due to the lack of slip lineations along the fault surfaces, the stress directions from the faults alone lack much confidence. For this reason, additional data from the folds and fabric were examined. Till Fabric The trend and plunge of till fabric data were plotted on the stereonet as a scatter plot (Fig. 3.6). The data were then contoured using the Kamb method option in Stereonet for Windows 1.2©. The Kamb method smoothes irregularities that are of little statistical significance (ie. "bullseye" contours around single data points) without altering the positions of contours of larger populations (Marshak and Mitra 1988) . The data points were contoured at a 2a interval with a 3a significance level. The area of the counting circle used to count the points and the expected number of points within the area varied from plot to plot. The area of the counting circle and expected number of points were calculated by Stereonet for Windows 1.2© and are listed with the plot. Along with the contouring information, the till fabrics were analyzed by calculating the eigenvalues using the Rockworks99© computer program. 43 Figure 3.5: Orientation of Principle Stress Directions for a Faulted Block. As described by Mark (1973) the orientation of the elongated clast is treated as a vector in x, y, and z coordinate space. Eigenvalues SI, S2, and S3 calculated from the eigenvectors represent the degree of clustering of the data points and as the difference in value increases the strength of the clustering increases. Eigenvalues SI and S3 are displayed on the fabric diagram to give an indication of the degree of clustering of the data. The eigenvector (VI) associated with the SI eigenvalue gives the mean lineation direction for the data set, which is also displayed on the stereonet. 44 Idealized diamicton with elongated clasts Trend = bearing from North in the direction of Plunge Plunge = Angle from horizontal Long Axis of Clast Stereonet plot of Elongated Clasts Axes Figure 3.6: Trend and Plunge Stereonet Plot of Elongated Clasts in Diamicton. 45 CHAPTER 4 RESULTS Figure 4.1 is a map of the study area that shows the name and location of all the aggregate mining operations that were investigated for this study. BrewsterPit JD Pit X Smith Pit X Allegan Cleveland Pit Remington Pit Michgan Paving Pit X Klett Pit X X Wyoming Pit x Kusmack Pit Van Buren Figure 4.1: Aggregate Mining Operations in the Study Area. 46 Table 1 displays the acronym and explanation for the sediment facies code given for the sediment unit in each stratigraphic log. This designation is modeled after Benn and Evens (1998). Data presented on stereonets depicting fabric data or fold axes will be accompanied with statistical data. Figure 4.2 is a representation of the type of statistical analysis present on the stereonet plots. Table 4.1: Stratigraphic Code for Glacial Sediments Listed in Stratigraphic Logs (Modified from Benn and Evans, 1998) D Diamicton Dmm Matrix supported, massive Dms(s) Matrix supported stratified and sheared Dm-cmd Matrix to clast supported massive and deformed Dmmd Matrix supported massive and deformed G Gravels Gm Clast supported, massive Gmd Massive, deformed Gomd Openwork gravels, massive, deformed Gop Openwork gravels, planar bedded GRpd Coarse sand, planar cross bedded Sh Fine to coarse sand, planar bedded Sp Medium to coarse sand, planar bedded Shd Fine to coarse sand, planar bedded, deformed Spd Medium to coarse sand, planar bedded, deformed F1 Silts and clays, laminated Fid Silts and clays, laminated, deformed 47 Mean Vector Trend and Plunge (95% confidence Radius) Eigenvalue SI Eigenvalue S3 n Number of samples CCA Counting circle area for Kamb contour En Expected number of points within counting circle Figure 4.2: Key to Values Presented with Fabric and Fold Axes Data. Tacy Pit The Tacy pit is located just off Red Arrow Highway between the towns of Coloma and Watervliet. It is in section 28, Coloma Township in Berrien County, Michigan as shown in Figure 4.3. The Tacy pit is located just off Red Arrow Highway between the towns of Coloma and Watervliet. It is in section 28, Coloma Township in Berrien County, Michigan as shown in Figure 4.3. This is the westernmost area studied for this study. The pit is located on an upland that attains an elevation of 242 meters AMSL and is 2.8 km long and 1 km wide at its widest. Referred to as the Hennesey Kame Complex, this upland has been interpreted as an assemblage of coalesced ice-hole deposits (Stone et al., 2003). Stratigraphy The Tacy pit contains four observable stratigraphic units (Fig.4.4). At the southern edge of the pit the surface diamicton was exposed, albeit very poorly.The surface diamicton was scraped from the pit as it was undesirable as aggregate. The small, 3-m wide location where the diamicton was exposed was too inaccessible to do 48 42,19N 86.30W 42.18N88.27W Red Arrow Highway Figure 4.3: Tacy Pit Location. a detailed study of the nature of the contact with the material directly below and also a fabric analysis. A sample of diamicton was available for study as portions of the diamicton slumped to the pit floor. This 1.0 to 1.5 m of diamicton is brown, sandy, 49 matrix supported, mostly massive but with some sand stringers within the diamicton. Clasts in the brown diamicton ranged from gravel to boulder size with none observed over 0.5 m. Immediately below the brown diamicton is a unit composed of bedded, medium sand. This unit is approximately 3m thick and is comprised of fine-to medium-bedded sand with lenses of gravel and silt dispersed throughout. Bedding is about 0.02 to 0.25 m thick. Some minor deformation structures, namely folds, were Scale Sediment in Meters Log Designation Description Sandy brown diamicton, matrix supported, massive Dmm with some areas of laminated sand Bedded fine to medium sand with lenses of coarse sand and gravel Sandy, red diamicton, matrix supported with sand stringers incorporated within and sheared 10 Bedded medium to coarse sand with lenses of gravel and silty clay 15 Figure 4.4: Tacy Pit Stratigraphic Log. observed and measured in this unit. Below this sand unit is a 0.5 to 1.5 m thick silty, matrix-supported diamicton. This diamicton is red in color and highly deformed. The diamicton displays folding and shearing. Thin (centimeter scale) layers of medium sand occur within the diamicton (Fig. 4.5). Below the diamicton is a large sequence 50 of coarse sand and gravel with some lenses and beds of silty sand. Bedding in this unit is thicker than the upper sand unit with beds ranging from 0.1 to 0.3 m. The beds directly beneath the red diamicton are deformed. The deformation extends as far as 6 m below the diamicton, but became less pronounced at depth. Deformation There are many deformation structures not only within the red diamicton but also above and below. The red diamicton contains many stringers of medium sand within, and also contains a rotated boulder as the sediments on either side of the boulder are disturbed (Fig 4.5). The average attitude of the bedding in and near this diamicton has a strike of 20° and a dip of 22° west. The trend and plunge of elongated pebble clasts in the diamicton were measured in two separate locations to evaluate the fabric of the diamicton. The mean trend and plunge is 306°, 29° as shown in Figure 4.6. Numerous folds occur in this pit and range in amplitude from 6.0 to 0.25 m. The limbs of four large folds were measured directly at this location. Structural data and a photograph of each major fold are displayed in Figure 4.7. Fold T1 is an overturned fold in the lower sand unit consisting of beds of coarse sand and gravel and has an amplitude of 6.0 m. The gravel in the core of the fold is cemented. The poles to bedding for fold T1 were plotted on a stereonet and the best fit girdle to the poles yields a strike of 149°. Fold T1 had a northwestward sense of vergence. As the vergence of a fold is dependent upon the sense of rotation when viewed in the direction of the plunge of an asymmetrical fold, only folds that are asymmetrical, plunging and display rotation will be described. Fold T2 is adjacent to Fold T1 and is an overturned fold with an amplitude of 3 m, consisting of the red diamicton. 51 Close-up of red diamicton with deformed sand within (between arrows) Close-up of rotated boulder in red diamicton Figure 4.5: Photographs of Diamicton in Tacy Pit. 52 MVT 306, 22 (7.9°) SI 0.72 S3 0.062 n 121 CCA 0.069 En 8.38 Figure 4.6: Clast Fabric Plot of Diamicton in Tacy Pit. The poles to bedding for fold T2 were plotted on a stereonet and the best fit girdle to the poles yields a strike of 137°. Fold T2 also had a northwestward sense of vergence. Fold T3, in the lower sand unit, is nearly isoclinal with sandy gravel limbs and a disharmonically folded sand core, and has an amplitude of 0.5 m. The poles to bedding for fold T3 were plotted on a stereonet and the best fit girdle to the poles yields a strike of 115°. Fold T4 is a Z type fold in bedded sand directly underlying the diamicton with an amplitude of 1.0 m. The poles to bedding for fold T4 were plotted on a stereonet and the best fit girdle to the poles yields a strike of 187°. The average of the girdle of these four folds large folds is 147 . Fold T4 has a southward sense of vergence. Along with these larger folds, the fold axes of 10 minor folds were measured. These folds occur within layers of the upper sand unit. The mean vector trend (MVT) of these minor axes is 241°, plunging 23°. The girdle to the MVT of these minor fold axes has a strike of 152°. 53 Figure 4.7: Folds in the Tacy Pit. Figure 4.7 - Continued. White Pit The White pit is located just off County Route 140, 0.2 km south of Interstate 94, south of the town of Watervliet. It is in section 27, Coloma Township, Berrien County, Michigan, as shown in Figure 4.8. The pit is 231 m AMSL and is located 1.4 km to the east and 0.7 km south on the same Hennesey Kame Complex (Stone et al., 2003) upland feature as the Tacy pit. 42.18N 86.28W 42.17N86.25W Figure 4.8: White Pit Location. 56 The White pit displays two stratigraphic units (Fig.4.9). The uppermost unit is a 2.0 m thick sandy, brown diamicton, which is matrix supported, massive, with some areas of laminated sand, and is highly deformed. Below the diamicton unit is a 14 m thick unit of fine sand and silt with occasional lenses and beds of coarser sand and gravel. The bedding of this unit is on the centimeter scale and is planar when not deformed. Scale Sediment in Meters Log Designation Description Sandy, brown diamicton, matrix supported, mas- Dmm sive, some laminated sand, highly deformed Fine sand and silt with lenses and beds of Fid/ Slid coarser sand and gravel deformed 10 Fl. xSh Fine sand and silt with lenses and beds of coarse sand and gravel, undeformed 15 Figure 4.9: White Pit Stratigraphic Log. Deformation This pit shows deformation not only in the surficial diamicton but also in the underlying sediment to a depth of about 8 m (Figure 4.10). Folds in this pit range in amplitude from 0.1 to 5 m. Fold W1 is a recumbent fold in the surface diamicton that 57 is interbedded with fine to medium sand with an amplitude of 1.0 m. The girdle trend of the poles to bedding is 132°. The fold hinge has a vergence to the northwest. Fold W2 is an upright fold. It is comprised of 0.1 to 0.3 m thick beds of fine to medium sand and silt with some coarser sand and gravel in the core of the fold. This fold is directly under the contact with the diamicton. Poles to the bedding yield a girdle trend of 127°. The average of the girdle of these two folds is 130 . Along the base of the diamicton are three minor folds. The girdle to these fold axes is 103°. The folds verge to the west-northwest. The average strike and dip of the base of the diamicton in this pit is 26°, 8° east. Wyoming Pit The Wyoming pit is 22 km east of the White pit on 46th Street, 0.18 km south of Interstate 94. The Wyoming pit is located in Section 24, Lawrence Township in Van Buren County (Fig. 4.11). This pit is located on an upland feature at about 277 meters AMSL. The upland is the traditionally mapped Valparaiso Moraine. Bird (2005) described the area of this pit as an ice stagnation ridge. Stratigraphy Limited areas of this pit are still active. Only a very basic stratigraphic section can be drawn, as the observable areas are not continuous. This pit has many areas of screened and mixed material dumped back over the pit face. The Wyoming pit has three distinct units (Fig. 4.12). The uppermost unit is a brown diamicton as evident from the numerous piles as well as discussion with the pit operator 58 1' Recumbent fold (Wl) in Deformed diamicton Diamicton and fold W2 at X in Photo B Figure 4.10: Folds in the White Pit. Deformation exposed IQ-im beSow surface 42.20N 86.01W 42.18N85.98to Figure 4.26: Cleveland Pit Location. 60 This brown diamicton has been removed from the surface and none remained in place to perform a fabric analysis. The next unit below the diamicton is a package of bedded fine sand and silt. This unit was about 2.3 m thick and is highly deformed (Fig. 4.13). The base of this unit becomes obscured by transported and slumped material. About 2 m lower in elevation and 20 m to the west, a fresh Scale Sediment in Meters Log Designation Description Hi t / j Brown diamicton (removed) Fl Bedded medium to course sand with lenses of VS h /s gravel and silty clay, deformed Bedded fine sand and silt, deformed /Sh Dms(s) Reddish-brown diamicton, matrix supported with sand stringers, sheared Sp Bedded fine to medium sand with lenses of coarse sand and gravel, deformed 15 U Figure 4.12: Wyoming Pit Stratigraphic Log. exposure crops out. This exposure displays a similar fine sand and silt package but is thinly bedded and not as pervasively deformed, but does show some deformation at 61 the contact with a lower reddish-brown diamicton. This reddish-brown diamicton is only 4 cm thick. The diamicton contains stringers of fine sand and is deformed. A few gravel-sized clasts were present but not abundant in the diamicton. The attitude of the diamicton bed has a strike of 200°,dipping 16° west. The lowermost unit observed in the Wyoming pit is 1.2 m of bedded silt and fine sand that is also deformed near the reddish-brown diamicton with small folds (Fig. 4.13). The beds of sand and silt are thinly bedded, on the centimeter scale, and appear to cyclically alternate from sand to silt. Deformation Deformation is observed in some areas of the pit. Two large folds, WY1 and WY2, were measured, along with eight minor. The folds occur in fine to medium sand with beds of silty sand. The large folds, WY1 and WY2, are upright in nature and about 0.5 m in amplitude. The average trend of the girdles of these two folds is 65 (Fig. 4.14). Minor folds are present throughout the pit. The fold axes of these minor folds yield a girdle trend of 70°. 62 Figure 4.13: Deformation Observed in the Wyoming Pit. Fold WY1 n 8 Figure 4.14: Wyoming Pit Structural Data. Michigan Paving Pit The Michigan Paving pit is located 3 km due north of the Wyoming pit at the intersection of 56th Avenue and Red Arrow Highway, in Section 12, of Lawrence Township in Van Buren County. Situated at 246 m AMSL, it is located on the Valparaiso Moraine 2.4 km from the Paw Paw River Valley (Fig. 4.15). This is part of the same ice stagnation ridge as the Wyoming pit described by Bird (2005). 42.22N 86.01W 42.21N 85.93W Figure 4.15: Michigan Paving Location. 65 Stratigraphy This pit is limited in size at only 0.2 km2, and only about 20 m of exposure is active. Diamicton in this pit has been completely removed. This pit has two distinct units aside from the removed diamicton (Fig. 4.16). The uppermost unit in this pit is 6.3 m thick, weakly cemented, openwork, cobble gravel, which is locally known as crag. The openwork cobble gravel appears to have a slight indication of bedding with the lower 4 m of the unit having meter scale beds dipping to the Scale Sediment in Meters Log Designation Description Openwork cobble gravel, slight indication of bedding dipping southeast coarse sand to boulders Fine to medium sand with silt, centimeter scale planar bedding,, deformed Figure 4.16: Michigan Paving Pit Stratigraphic Log. southeast. Grain size ranges from coarse sand to boulders with no boulders over 0.5 m observed. The unit underlying the cobble gravel is an exposure of fine to medium sand and silt at least 3 m thick (Fig. 4.17). Bedding thickness is on the centimeter 66 scale and displays alternating beds of sand and silt. The upper meter of this unit is intensely deformed with minor folds and faults (Fig. 4.17) but becomes less intense with depth. Exposure of Michigan Paving operation Figure 4.17: Deformation Observed in the Michigan Paving Pit. 67 Deformation A large fold with a 0.6 m amplitude is observed in the sand at the base of the cobble gravel. This fold, MP1, is an upright fold with the core of the hinge faulted by numerous reverse faults. The girdle trend of this fold is 115 (Fig. 4.18). Due to the lack of contrast between layers of the fine to medium sand in the core of this fold, the faults are too faint to excavate and measure with any degree of confidence. The whole lower fine to medium sand unit is an upright fold (Fold MP2) with at least 2.0 m amplitude and a girdle trend of 138°. At first glance this unit appears to consist of uniformly dipping beds however, the exposure has been excavated oblique to the fold axis and the northwest limb has been removed. GT 138 Fold MP2 n 6 Figure 4.18: Stereonet Plots of Folds in the Michigan Paving Pit. 68 Kusmack Pit This pit is located on 39th Street, 1.3 km south of 72nd Avenue in Section 34 of Paw Paw Township in Van Buren County (Fig. 4.19). The Kusmack pit is 9 km southeast of the Michigan Paving pit. This pit is 240 meters AMSL and located on an upland that is 4.9 km long and 2.2 km wide. The upland, which is an extension of the Inner Kalamazoo Moraine, is discontinuously capped by a brown, sandy diamicton. Below the diamicton is evidence of deformation in the sand and gravel beneath (Bird, 2005). Stratigraphy This pit was investigated as part of the surficial mapping of the Paw Paw U.S.G.S. 7.5 minute quadrangle (Bird, 2000). At that time, a brown sandy diamicton capped the pit. When revisited in 2005 the pit was overgrown and mostly abandoned. In 2006 the pit was freshly reopened and had expanded somewhat. The surface diamicton has been completely stripped from the surface. Exposed in the pit is basically one, 6-m thick unit in place after the removal of the diamicton (Fig.4.20). The unit is composed of planar-bedded fine sand and silt directly beneath the diamicton. These finer sediments extend to a depth of 2.0 m. A 0.5 m faulted lens of bedded gravel and coarse sand is in contact with the base of a silt layer. At 2.0 m, the sediments coarsen to cross-bedded gravel. Underlying the gravel bed are planar- bedded, fine- to medium-grained sand beds. This fine to medium-grained sand gives way to coarser sand and gravel which is planar-bedded. 69 42.17N 85.94W 42.15N 85.91W Figure 4.19: Kusmack Pit Location. 70 Scale Sediment in Meters Log Designation Description Brown diamicton (removed) Planar-bedded fine sand and silt, deformed Faulted lens of bedded gravel and coarse sand 5 - Sh Planar-bedded fine to medium sand Planar-bedded, coarse sand and Sh gravel Figure 4.20: Kusmack Pit Stratigraphic Log. Deformation The measurable deformation is limited to one minor recumbent fold in the upper portion of the unit with a girdle trend of 105° (Fig. 4.21). T 105 Fold K1 n 4 • Minor Fold Axis Figure 4.21: Deformation Observed in the Kusmack Pit. 71 Klett Pit The Klett pit is located on County Route 665, 0.7 km north of 64th Street, in Section 24 of Paw Paw Township in Van Buren County (Fig.4.22). At an elevation of 246 m AMSL, the Klett pit is on an isolated upland surface 5 km northeast of the Kusmack pit. This upland, referred to as Prospect Hill, is 3.8 km long and 0.9 km wide. Prospect Hill is part of the Inner Kalamazoo Moraine. 42.20N 85.90W Figure 4.22: Klett Pit Location. 72 Stratigraphy Activity in the pit has basically ended and the pit is in the process of being reclaimed. An exposure 10 m long exposed two distinct units (Fig. 4.23). At the surface of this pit, the upper unit is a 2.0 m thick, brown diamicton which is matrix supported and massive in the upper 1.0 to 1.5 m. The lower 0.5 to 0.7 m of the diamicton is clast supported and is intensely deformed. Generally, the diamicton becomes coarser with depth. The clasts range in size from gravel to cobbles with an occasional boulder about 0.5 m in diameter with gravel and sand stringers present. The lower unit extends below the diamicton for at least 4 m. This unit is comprised of bedded fine sand and silt with one bed of coarse sand and gravel. At the contact with the diamicton, the fine sand and silt are intensely deformed and penetrates to a depth of 1.2 m below the diamicton. In the undeformed layers at depth, the bedding is a few centimeters thick in the fine sand and silt and 0.4 m thick in the coarse gravel and sand. 73 Scale Sediment in Meters Log Designation Description Sandy, brown diamicton, matrix supported, Dm-cmd massive, grading into clast supported and coarser with depth showing deformation at the base Sh Thinly bedded fine sand and silt /(F l with a 0.12-m bed of coarse sand and gravel Gm Figure 4.23: Klett Pit Stratigraphic Log. Deformation The focus in this pit is an area in which the contact of the diamicton with the underlying medium sand is highly deformed (Fig.4.24). The base of the diamicton is folded with the core of these folds composed of medium sand from the underlying sediments. This deformation extends down into the sand to a depth of 2.4 m. Three large folds, KL1-KL3 and five minor fold axes were measured. Fold KL1 is an open, upright fold with an amplitude of 0.4 m. The girdle trend of KL1 is 113°. Fold KL2 is an overturned fold with an amplitude of 0.5 m , verges to the southeast, and has a 74 girdle trend of 134°. Fold KL3 is an overturned fold with an amplitude of 0.55 m and also has a vergence to the southeast. The girdle trend of KL3 is 122°. The average girdle trend of these folds is 123° (Structural Data Fig). The fabric of elongated pebble clasts in the surface diamicton yield a mean of 270° and a plunge of 18°, (Structural Data Fig.). Figure 4.24: Structural Data of Folds at the Base of the Diamicton in the Klett Pit. 75 270, 28 (12.0°) 0.726 0.068 55 0.15 7.7 Figure 4.25: Clast Fabric Analysis of the Diamicton in the Klett Pit. Cleveland Pit The Cleveland pit is located on County Route 665, 0.3 km south of 18th Avenue, in Section 29, Bloomingdale Township, Van Buren County (Fig. 4.26). The Cleveland pit is 18-km north-northwest of the Klett pit. At 250 m AMSL, this pit is located on a subtle, low relief hill, which is part of the Valparaiso Moraine. This pit was the site used by Monaghan et al. (1986) to show a contact between the Saugatuck Till and the Ganges Till. Stratigraphy The Cleveland pit is composed of three stratigraphic units (Fig. 4.27). The uppermost unit consists of 2.0 m of a reddish-brown matrix supported diamicton. This diamicton shows deformation as well as incorporation of sand lenses into the diamicton (Fig. 4.28). The underlying unit is an extensive sequence of fine sand with some silt, clay and gravel. A 0.75 m thick bed of clay is exposed in one part of the pit; however, it is not continuous across the exposure. The bedding of this unit 76 appears planar, although it is obscured by the deformation. Bedding ranges from centimeter scale beds in the fine sands and silts to the 0.25 m scale for the coarse sand and gravel beds. The lower-most unit in the pit is a gray, matrix-supported diamicton that has been correlated to the Ganges till by Monaghan et al. (1986). 42.36N 85.97W 42.34N 85.95UV Figure 4.26: Cleveland Pit Location. 77 Scale Sediment in Meters Designation Description Reddish-brown sandy diamicton, matrix sup- Dmmd SLti '^J ported, deformed with sand laminae Fid Discontinuous clay bed Gmd Planar bedded fine sand, inter-bedded 'Sh with silt, medium sand and gravel, deformed Shd Gray diamicton, massive, matrix sup- Dmm ported Figure 4.27: Cleveland Pit Stratigraphic Log. 78 7' Extensive deformation VO B Sand stringers in diamicton Figure 4.28: Deformation Observed in the Cleveland P it. Deformation Portions of this pit show extensive deformation in the form of folded sediments. The folds are primarily recumbent with areas of disharmonic folding. Evidence of shear is observed throughout the eastern face of the pit. The attitude of the dominant shear plane strikes with an azimuth of 46° and dips 18° northwest. Three large folds were measured directly (Fig. 4.29). Fold CI, is a synform-antiform complex with an amplitude of 2.0 m and a girdle trend of 207 . Figure 4.29a is a photograph of the fold. Fold C2 is a recumbent fold with an amplitude of 1.0 m. The girdle trend of this fold is 127°. Fold C3 is another antiform-synform complex with an amplitude of 0.3 m and a girdle trend of 128°, which when averaged with C2 yields an average of 128 . The axes of 11 minor folds were also measured and plotted. The long axes of elongated pebble clasts in the diamicton were measured at two locations (Fig. 4.30) and yield an average of 138 , plunging 25°. 80 Figure 4.29: Deformation Observed in the Cleveland Pit. 138, 25 (1.0°) 0.817 0.042 111 0.076 8.31 Figure 4.30: Clast Fabric Analysis of the Diamicton in the Cleveland Pit. Remington Pit The Remington pit is located on County Route 665, 0.5 km south of 18th Avenue. At 243-m AMSL, this pit is on the same low hill as the Cleveland pit approximately 0.5 km east, in Section 28, Bloomingdale Township in Van Buren County (Fig. 4.26). 82 Stratigraphy The Remington pit shares much of its stratigraphy with the adjacent Cleveland pit. This pit contains four units (Fig. 4.31). The uppermost unit is the same 2.0 m thick, brown, sandy diamicton as in the Cleveland pit. This diamicton is matrix supported and displays deformation and shearing as well as the incorporation of beds of sand (Fig. 4.32). Underlying the diamicton unit is a 2.7 m thick unit of fine sand and silt. This unit is thinly bedded; however, the deformation and shearing make determining the true style of bedding difficult. Differing from the Cleveland pit, the next unit stratigraphically below is a 2.2-m thick unit of coarse sand and gravel, which is massive and clast supported. The lowermost exposed unit in this pit is a 4.75 m thick fine- to medium-grained, bedded sand and silt, which is deformed. At one location on the east wall of the Remington pit at a depth of 11.3 m from the surface, a package of climbing ripples is preserved. The crests of these ripples were carefully excavated to measure the trend of the ripple crest. The average trend was 22° indicating a flow direction of 112°. 83 Scale Sediment in Meters Log Designation Description W3 Dmmd Reddish-brown sandy diamicton, matrix sup- ported, deformed with sand laminae Fid Planar bedded fine sand, inter-bedded with silt, deformed Gmd Clast supported massive gravel, deformed % 10 Planar bedded fine sand, inter-bedded Shd with silt and medium sand, deformed 15 Figure 4.31: Remington Pit Stratigraphic Log. Deformation Deformation in this pit is extensive on the west face with some deformation exposed on the east face. Deformation includes isoclinal folds, open folds, overturned folds, thrusted sediment, as well as faulting (Fig.4.33). The limbs of a large overturned fold, Rl, with an amplitude of 1.0 m, were measured as well as the fold axes and planes of ten minor folds. The girdle of Fold Rl has a trend of 123 . Two conjugate, reverse fault planes were observed and measured to determine the al direction that trends 129 and plunges at 28 . The mean trend of the fabric of the surface is 317 (Fig. 4.34). 84 00 Figure 4.32: Deformation Observed in the Remington Pit. Figure 4.33: Structural Data of Deformation in the Remington Pit. MVT 317, 22 (1.0°) Si 0.848 S3 0.033 n 50 CCA 0.153 En 7.63 Figure 4.34: Clast Fabric Analysis of Diamicton in the Remington Pit. Smith Pit The Smith pit is located on 44th Street, 0.5 km north of 110th Avenue. This pit is 16 kilometers north of the Remington pit in Section 4 of Cheshire Township in Allegan County (Fig. 4.35). At 231 m AMSL, the Smith pit is located in a drumlin field in the Valparaiso Upland (Kehew et al., 2005). In fact, this pit is located on the southwestern edge of a drumlin in the middle drumlin field of this field. Stratigraphy Capped by a 2.2 m thick, brown, sandy diamicton, the Smith pit has four stratigraphic units (Fig.4.36). The diamicton is matrix supported and mostly massive. There is indication of deformation and incorporation of the underlying sand into the diamicton. The sediments in this pit are so intensively deformed that the true stratigraphy is difficult to determine as the lithologies in contact with the diamicton 87 change from one face to another. Stratigraphy will be described starting at the northern edge of an east-west face and continuing toward the south along a north- south face and finishing on a north-south aligned face to the west of the former. Underlying the diamicton in this section is a unit that is actually isoclinally folded. It consists of a package of bedded silt, fine to medium sand and occasional beds of coarse sand and gravel. Bedding is relatively uniform at 0.25 m thick with some beds at the centimeter scale. Moving around the exposure to the south and east, the sequence changes to a centimeter-scale bedded fine sand and silt package in contact with the diamicton. This unit is 3.4 m thick at the thickest and highly deformed. Below this unit is a 0.7 to 1.5 m thick unit of massive, clast-supported coarse sand and gravel. On the third face of the pit the fine sands and silts unit is much thinner (~0.5 m). Here the coarse sand and gravel unit is interbedded with fine to medium sand. The lowermost unit is a 4.2 m sequence of interbedded fine and medium sand, planar bedded at the 0.25 m scale and is also deformed. Deformation Deformation in this pit is extensive and is observed throughout all units. Large folds in this pit are generally overturned to recumbent and have an amplitude of 1-3 m. Five large folds, S1-S5, were measured (Fig. 4.37). Fold SI is overturned with a fold amplitude of 1.7 m. SI is synform and the axial surface dips east with a girdle trend of 96°. Fold S2 is an upright synform with an amplitude of 0.75 m and a girdle trend of 149°. Fold S3 is at the base of the surface diamicton and is upright with an amplitude of 0.3 m. This fold is also synform and the girdle trends 60°. Fold S4 is an overturned synform with an amplitude of 3.0 m and a girdle trend of 43°. The axial surface for fold S4 dips southwest. The amplitude of fold S5 is difficult to determine because the hinge area has been removed. 42.50N 85.99W V////////////M 200 m I 1 Clrh A'.'e Figure 4.26: Cleveland Pit Location. 89 Scale Sediment in Meters Log Designation Description Dmmd Reddish-brown sandy diamicton, matrix sup- ported, deformed with sand laminae Fid Planar bedded fine sand, inter-bedded with silt, deformed Gmd Clast supported massive gravel, deformed _C Fl Sh Planar bedded fine sand, inter-bedded with silt and medium sand, and gravel, deformed Shd 15 Figure 4.36: Smith Pit Stratigraphic Log. The sediments that form this fold suggest that the amplitude is more than 3.0-m. This fold is isoclinal and synform with a girdle trend of 148°. The average girdle trend of the five folds is 99°. In addition, 14 minor folds were measured. The stereonet results of these folds are shown in Figure 4.37e. The clast fabric was also measured and has a mean of 333° and a plunge of 15°. Two cobble-sized limestone clasts in the diamicton were heavily striated. The average trend of these striae, 103°, was plotted on the stereonet of the diamicton fabric (Fig.4.38). Within an eight-centimeter bed of silty clay, parallel partings were observed. These partings are oriented transverse to the bedding plane and were a few millimeters thick. The poles to these surfaces are 90 coincident with the direction of the clast fabric trend. Figure 4.39 shows the stereonet of the partings with the mean vector (137 , 17 ) of the poles to the partings. / r 1 - Fold SI Fold S2 GT 96 149 Fold SI Fold S2 h 11 7 Fold S3 Fold S3 Figure 4.37: Deformation Observed in the Smith Pit. 91 Figure 4.37 - Continued. f MVT 333, 15 (1.0°) SI 0.697 • J 1LJ > S3 0.097 CCA 0.153 En 7.63 in 50 Figure 4.38: Clast Fabric Analysis of Diamicton in the Smith Pit. 93 Brewster Pit The Brewster Pit is 1.5 km west of State Route 40 on 121st Avenue. This pit is located in Section 11 of Allegan Township in Allegan County (Fig. 4.40). At 231 m AMSL the Brewster pit is 9 km northeast of the Smith pit. This pit is also in the drumlin zone (Kehew et al. 2005) and is between two of the drumlin fields. Stratigraphy Fresh exposure is limited in this pit and no diamicton was observed in place. The Brewster pit has two stratigraphic units (Fig. 4.41). The uppermost unit is a fine to medium sand with bedding ranging in thickness from 0.1 to 0.6 m. This unit is at least 4.0 m thick but no direct contact between the two units was observed. The underlying unit is a sequence of coarse sand to cobble gravel. This unit is clast supported with bedding ranging from 0.1 to 0.3 m. This unit is at least 4 m thick. Deformation Deformation in this pit consists of two large folds with amplitudes ranging from 0.75 to 2.0 m. The limbs of these folds were measured directly. Fold B1 is an upright antiform, in a sequence of gravel and has a girdle trend of 84°. Fold B2 is observed in the fine to medium sand unit as an upright synform and has a girdle trend of 89°. The average of the girdle for these folds is 87 (Fig. 4.42). 94 42.58N 85.94W 42.57N 85.92W Figure 4.40: Brewster Pit Location. 95 Scale Sediment in Meters Log Designation Description Planar bedded fine to medium sand, Shd deformed Gravel, openwork, clast supported, massive, Gomd deformed Figure 4.41: Brewster Pit Stratigraphic Log. 96 Fold B2 Figure 4.42: Folds Observed in the Brewster Pit. 97 JD Pit The JD Pit is the northernmost pit of this study. This pit is located west of 23rd Street, 0.15 km south of 122nd Avenue in Section 7 of Watson Township in Allegan County (Fig. 4.43). The JD pit is located 13.5 km east-northeast of the Brewster pit. At 240 m AMSL, this pit is on the traditionally mapped Kendall Moraine (Leverett and Taylor, 1915), specifically on a high relief, hummocky upland (Kozlowski, 2005). Stratigraphy The JD pit has three distinct units (Fig. 4.44). The uppermost unit is a brown, sandy diamicton at least 1 m thick; however, most has been scraped away. The diamicton is matrix supported and contained a higher percentage of cobble-sized and larger clasts than other diamictons observed across the study area. Beneath this diamicton is a 2.4-m thick unit of planar bedded fine to medium sand and silt. Bedding thickness ranges from 0.05 to 0.3 m. Underlying this unit is a unit at least 6- m thick composed of bedded coarse sand and gravel that is interbedded with medium sand. The coarse beds are generally between 0.25 and 0.75 m while the finer sand beds are 0.5 and 0.2 m thick. Deformation Two large folds and one minor fold were measured in this pit (Fig. 4.45). Fold JD1 is a synform, upright fold with an amplitude of 1.5 m. It is composed of bedded sand and gravel and has a girdle trend of 87o. Fold JD2 is an antiform, upright fold with an amplitude of 1.5 meters. This fold is composed of bedded fine sand and silty sand and had a girdle trend of 84°. The fold axis of one minor fold 98 42.58N 85.78W 42.57N S5.76W Figure 4.43: JD Pit Location. 99 Scale Sediment in Meters Log Designation Description Brown, gravelly diamicton, matrix supported, ,cy. • » » J» massive Planar bedded fine to medium sand Shd with some silt, deformed Planar bedded, coarse sand and gravel, inter- Spd bedded with medium sand and silt, deformed Figure 4.44: JD Pit Stratigraphic Log. with an attitude of 180°, 15°was measured and plotted on the stereonet of Fold JD2. The average of the girdle of these two folds has an azimuth of 86°. Two reverse faults occur west of Fold JD1 (Fig. 4.46). The attitude of the fault planes was recorded and the stress directions were calculated. The principal stress vector trends 110°and plunges 0.5°. 100 Fold JD1 Figure 4.45: Deformation Observed in the JD Pit. Figure 4.46: Conjugate Reverse Faults in the JD Pit. 102 Compilation Data In order to better visualize the interplay of glacier motion and applied stresses the clast fabric data can be layered onto the stereonet with the large fold data, minor fold data and any other structural details. This compilation of data can be used to refine the analysis by having the clast fabric and fold axes superimposed. Figures 4.47 through 4.51 are the compilation data for each pit that contained a measured diamicton. In an effort to compare the structural data across the study area all fold axes were plotted together on one stereonet. This plot yields the overall shear plane for the area (Fig. 4.52). Attitude of diamicton Attitude of Shear Plane Mean Fold Axis (n=13) Inferred ice flow from fabric Average Girdle Trend of Large Folds (n=3) Symmetrical Rose Diagram of Clast Fabric (n=109) Symmetrical Rose Diagram of Fold Axes fend (n=13) 46°, 10° E Cleveland Pit Compilation Figure 4.47: Compilation of Data for the Cleveland Pit. 103 Attitude of diamicton A Mean Fold Axis (n=10) * % Inferred ice flow from fabric Average Girdle Trend of Large Folds (n=l) CT1 Direction 129, 28 Symmetrical Rose Diagram of Clast Fabric (n=50) 123° Symmetrical Rose Diagram of Fold Axes Trend (n=10) Remington Pit Compilation Figure 4.48: Compilation of Data for the Remington Pit. Attitude of diamicton A Mean Fold Axis (n=8) % Inferred Ice flow •A from fabric v V\ Average Girdle fend of .90° Large Folds (n=3) \\ Symmetrical Rose Diagram V> of Clast fabric (n=55) V\ Symmetrical Rose Diagram \\ of Fold Axes Trend (n=8) 123° Klett Pit Compilation Figure 4.49: Compilation of Data for the Klett Pit. 104 Attitude of diamicton 17°, 28° A Mean Fold Axis (n=16) % v Inferred ice flow from fabric Average Girdle Trend of Large Folds (n=4) 990 Symmetrical Rose Diagram • of Clast Fabric (n=121) Symmetrical Rose Diagram V, of Fold Axes Ti-end (n=16) -?6< Average Pole to Clay tertlngs 141° Striae Smith Pit Compilation Figure 4.50: Compilation of Data for the Smith Pit. Attitude of diamicton Mean Fold Axis (n=13) % Inferred ice flow ^ from fabric Average Girdle Ti-end of Large Folds (n=3) Symmetrical Rose Diagram of Clast Fabric (n-109) Symmetrical Rose Diagram of Fold Axes Trend (n=13) Tcicy Pit Compilation Figure 4.51: Compilation of Data for the Tacy Pit. 105 Figure 4.52: All Fold Axes Measured Defining a Shear Plane with a Strike of 40° and a Dip of 20° to the Southeast. 106 CHAPTER 5 DISCUSSION AND CONCLUSIONS Ice Melt-Out The impetus to analyze the deformation observed in gravel pit exposures across the Valparaiso Upland was the observation of deformation structures that were not consistent with a previous model of formation by melting of buried ice. The major objective of this study is to determine if the deformation was the result of melting of buried ice or direct interaction with active glacial ice. Only a single location, the Kusmack pit (Fig 5.1), displayed deformation that is consistent with the melting of buried ice. The brittle, extensional nature of the deformation in that pit is similar to the normal faulting found in eskers (Sharpe, 1988), kames (Brodzikowski, 1991), and kettles (Maizels, 1992). It is this brittle, extensional style deformation, caused by the melting of ice supporting a package of sediments that will distinguish it from the ductile, compressional style of deformation from flowing ice actively shearing the sediment package. This 1.5 m wide area near the surface of the pit shows brittle deformation that extends to a depth of 2.5 m from the upper most portion of deformed material (Fig. 5.2). The stratigraphy of the exposure, along with the normal faulting can be used to invoke the ice melt-out. The stratigraphy represents what is to be expected with the retreat and subsequent readvance of an ice margin. Starting at the base of the exposure, the sediment grain size changes from coarse sand with gravel to fine sand toward the top. This sequence can be explained by the sediment source, the ice margin, moving farther away. The planar bedded silt and fine sand at the upper 107 portion of the exposure is consistent with a lacustrine sequence and this unit is capped with a diamicton, which represents the readvance of ice over the location. The faulted, well-sorted gravel lens most likely represents a cohesionless subaqueous debris flow deposited by the advancing ice into the lacustrine environment. Figure 5.3 shows the inferred sequence of motion that would account for the observed brittle deformation. Evidence of this sequence is based upon the undeformed lacustrine sediment stratigraphically above the deformed section. This undeformed unit must have been deposited after the faulting occurred. The silt layer that defines the uppermost bed of deformed sediment is peculiar in that the thickness is inconsistent across the faults (Fig. 5.2). An inferred scenario is that during deformation, the faulted sediment moved differentially. As movement occurred along the fault planes, the silt layer was disturbed thus initiating the transport of the loose material away from the bed. Ductile Deformation of Glacial Sediments All other deformation observed across the study area is interpreted to be the result of the direct interaction with active, advancing glacial ice. The congruency of the fold and fault stress directions and the direction of ice movement obtained from the clast fabric supports this conclusion. Another focus of this study is to assess whether the deformation was the result of subglacial deformation or proglacial deformation. This problem is challenging, as the deformation is most likely the result of both. To assess the nature of the deformation, the style of folding, thrusting, and landforms need to be examined. Each can be addressed as a function of meltwater drainage, topography, and sediment grain-size. 108 Tacy Pit / i ... , White Pit Kusmack Pit a 0 Km 9 Berrien Van Buren Figure 5.1: Regional Map of Specific Gravel Pit Exposures. 109 Figure 5.2: Brittle Deformation Displayed in the Kusmack Pit. (Arrows show silt layer) Figure 5.3: Sequence of Ice Melt-Out Causing Brittle Deformation 110 Sediment Grain-Size The regional stratigraphic framework of the area consist of surficial till, the Saugatuck Till, which caps a sequence of lacustrine sediments, which in turn, overlies the Ganges Till. Although the bulk of the sediments that comprise the lacustrine sequence are fine sand, silts and clay, there are units of coarse sand, gravel and cobbles interbedded with the fine-grained materials. The coarse sand, gravel and cobbles are moderately well sorted and typically contain an openwork, clast- supported fabric. In many locations the coarse sand, gravel and cobbles are cemented with a post depositional calcium carbonate cement (locally known as crag). The grain-size as well as the degree of sorting will affect the hydraulic conductivity. Meltwater filling the voids between sediment particles will flow preferentially through coarse, well-sorted material as compared to finer material. This contrast in hydraulic conductivity and grain-size will affect the how the sediments deform. The overall rheology of the material is dependent upon the individual components of the deforming mass. Beds and lenses of coarse sediments will act as rigid bodies within the more ductile, fine-grained matrix. An analogy can be made with contrasting minerals in a high-grade metamorphic rock. The softer, more ductile minerals will deform differentially as compared to harder, more rigid minerals. Pore Water Pressure The shear strength of sediment is influenced by pore water pressure. An increase in the pore water pressure tends to dilate the sediment, which will reduce the cohesion and internal friction of the bed. Pore Pressure is often inversely proportional to hydraulic conductivity, achieving its highest values in fine-grained, 111 poorly drained sediments. In this area of southwest Michigan, Kehew et al. (2005) gave four reasons why the pore water pressure would be high under the LML: (1) the fine-grained lacustrine sediment would impede subglacial water flow from beneath the glacier, (2) the glacier was advancing upslope to the east, (3) the proglacial lake impounded between the ice and higher topography to the east would oppose the subglacial meltwater hydraulic gradient toward the margin, and (4) once at the Kalamazoo Moraine, where meltwater would have drained away from the margin to the east, the glacier could have frozen to the bed, again impeding the drainage of subglacial meltwater. The fine-grained lacustrine sediments beneath the upper diamicton have a lower hydraulic conductivity that will retard pore water flow, increase pore water pressure, and facilitate deformation of the sediment because of the decrease in the shear strength. The relationship of increasing pore water pressure and decreasing strength will facilitate deformation as long as the ice stays in contact with the bed. If the pore water pressure exceeds the force exerted by the weight of the ice, the ice can decouple from the bed and deformation will stop (Iverson et al., 2003), as the transfer of stress to the substrate can no longer take place. The basal contact of the till can be used to infer whether or not the ice was coupled to the bed. Where the contact is sharp between the till and the substrate, decoupling caused by increased pore water pressure would be expected. Evidence of ice both coupled to and decoupled from the bed is present throughout the study area indicating variable pore water pressure (Fig. 5.4) Topography The readvance of ice across southwest Michigan encountered a preexisting 112 topography that influenced the deformation of the glacial sediments in the area. From the modern day shore of Lake Michigan, there is a general increase in elevation to the east. The extremes of this slope range from 176 m ASL at Lake Michigan to upwards of 315 m ASL on the Kalamazoo Moraine. This rise allows for the formation of a proglacial lake, bounded by the ice margin to the west and the general rise in topography to the east, culminating at the higher, Kalamazoo Moraine. This proglacial lake will contribute to the increase in pore water pressure as explained earlier. The fine-grained lacustrine sediments are not only responsible for folding but also the propagation of shear faults and thrusting. Ramped thrusts along weak lacustrine sediments are evident in several gravel pit exposures (Fig. 5.5). The sediments being thrusted proglacially across this area would not be unlike those described by Phillips (2008) (Fig. 5.6). The thrusting of proglacial sediment would create thickened packages of sediment In front of the glacier. Evidence of this stacking can be found north of the study area along the Lake Boarder Moraine, where preliminary findings demonstrate an older till lying stratigraphically above a younger till (Kehew, personal communication). Following Tulaczyk (1999) and also Eyles and Boyce (1998) the roughness of the sole of the glacier, in this case caused by the topography of the pre-existing landscape with the proglacially thrusted stacked sediments, allows for the transfer of stress several meters deep into the substrate. The areas of higher relief will impose higher stresses on the base of the glacier. The effect of topographic relief on the deformation of sediments is evident in what Kehew et al. (2005) describes as the drumlin zone. This is a band of moderate relief with streamlining only on the highest portions of the landscape (Fig. 5.7). 113 Figure 5.4: Examples of Till in the Study Area Displaying Evidence of Being Coupled and Decoupled to the Bed. 114 The extent and intensity of proglacial deformation will be dependent upon the ice marginal profile as well as the stratigraphy of the sediments beneath and ahead of the glacier (Phillips, 2008). In the study area, the lack of continuous exposures prevents the ability to track a single thrust or decollement beyond a specific gravel pit exposure. For this reason, the extent of proglacial deformation can not be determined with any confidence; however the interpreted proglacial stacking of the sediments is evident and ultimately will invoke a topographic rise at the margin of the glacier of which the advancing glacier must override. Stresses under the LML The strain pattern under the LML, which varied spatially from the interior to the margin, was dependent on several factors, including pore water pressure (Boulton and Caban, 1995; Piotrowski and Kraus, 1997; Lee and Phillips, 2008), ice flow direction (Kehew et al. 2005; Hart and Boulton, 1991), and decollement surfaces (Andersen et al., 2005). The stress exerted at the base of the glacier is a function of the thickness of the ice (Fig. 5.8). Benn and Evans (1998) describe the "gravity spreading model" which implies that the differential load caused by the sloping ice margin, or more specifically the decreasing ice thickness toward the margin, is an important cause of proglacial thrusting in the subsurface sediments (van der Wateren, 1985; Aber et al., 1989). The net stress of the diminishing loading near the margin translates to horizontal compressive stress. 115 Figure 5.5: Evidence of Shear in Glacial Sediments across the Study Area. shedding of detritus large till-prism potentially topographic features (moraines) out of sequence forming barrier to subglacial developed above 'till-prisms' thrusting melt-water drainage basal thrust or decollement largc-scalc folding developing till-prism soft-sediment deformation concentrated push-moraine thrusting deformation along margins of till-prism Figure 5.6: Evidence of Proglacial Deformation of Glacial Sediments in Eastern England. (Modified from Phillips, 2008) Figure 5.7: Brumlim Zone. (Modified from Kehew et al. 2005) Modeling by Anderson et al (2005) supports a gravity spreading model for the formation of proglacial thrusts and concludes that the combination of loading and depth to the decollement surface are primary controls on large-scale glaciotectonic thrusting. The thick package of fine-grained, saturated, lacustrine sediments on the Valparaiso Upland provide not only the high pore water pressure required for 118 Ot 1 •••••• • 'i M . * 1 L "a3 ct3 | C71 =<72=03 ' (Hydrostatic state of stress) c < Figure 5.8: Net Stress Exerted on the Substrate by Loading of a Glacier. (Modified from Andersen et al., 2005) deformation but also a weak layer for development of a decollement surface. In the area of horizontal compressive stress near the margin of the glacier, the deformation style observed has been used to interpret glacial environments (McCarroll and Rijsdijk, 2003; Phillips et al, 2002, 2008; Rijsdilk et al, 1999). Beneath the ice at the margin, the lateral forces will tend to shear the underlying sediments toward the glacier foreland. With a proglacial lake in this area, the pore water pressure will be elevated thus increasing the likelihood of deformation. The simple shear induced by the flow of ice coupled to its bed will begin to form the observed folds, faults, and shear zones. The principal stress directions along with the inferred ice flow determined from till fabric were plotted on a map in order to display the regional trend of the directions (Fig. 5.9). As evident from this map the overall direction of the ice changed across the study area from a more northwest to southeast direction near the Tacy Pit to a more west to east near the JD Pit. 119 Deformation in Drumlins Two streamlined features within the study area contained gravel pit exposures. The Tacy and White pits were located on a structure that displays what is interpreted as the initial stages of drumlin formation, while the Smith pit was located on a well defined drumlin (Fig. 5.1). The analysis of the deformation at these locations yields insight to the deformation process associated with drumlins in this region of southwest Michigan. The upland that contains the Tacy and White pits in Berrien County is partially streamlined in the direction of ice flow. Although the structure was not streamlined to the extent of other drumlins in the area, the topographic high areas of this upland are aligned with the measured stress directions (Fig. 5.10). The deformation in the Smith pit is the most intense and deepest of all exposures studied. Aside from the overturned, isoclinal folding throughout the exposure, an indication of the intensity of the deformation stresses is the occurrence of what are interpreted as cleavage planes in a bed of silty clay. This 12 cm thick bed contains millimeter scale axial planar partings aligned with the strike of the shear surface defined by the fold axis (Fig. 5.11). 120 \ TOIobntodsmato \ \ Orwnfal orientation PHndpte stress drec&mef \ observed deformation Figure 5.9: Map of Principal Stress Directions, Inferred Ice Flow Directions, and Drumlin Axes Orientations The attitude of these planes being perpendicular to the principal stress direction suggests that the partings were created by the reorientation of the sediment grains, 121 much the same way that rock cleavage forms. No documentation of cleavage in unconsolidated sediment could be found to help support the formation of these structures. Tacy Pit 151° 0 Km 0.5 Figure 5.10: Alignment of Topography with Principal Stress Directions and Clast Fabric. The orientation of the clast fabric and principal stress directions creating the folds for the Tacy pit and Smith pit are not perfectly aligned to each other. The oblique orientation of the girdle trends of the folds for these exposures is best explained by the divergence of ice flow around the protruding upland (Fig. 5.12). These areas of higher relief contain beds of coarse sand and gravel. Coarse sediment would enable rapid drainage and lower pore water pressure, increasing strength of the sediment and would behave much like boudins (Boulton, 1987), deforming at a slower rate than the surrounding fine sediments as the ice flow is diverted around the obstruction. 122 Figure 5.11: Axial Planar Partings Perpendicular to Bedding Found in the Smith Pit. When the trend of the principal stress is oblique to the clast fabric direction or drumlin trend, the exposure is on the flank of the drumlin. The White pit, which is nearer the top of the feature, shows a strong agreement between the stress direction and long axis of the drumlins. Cutting across beds in the Smith pit is a clastic dike (Fig 5.13). Clastic dikes represent the escape of over-pressurized water and can form proglacially (Boulton and Caban, 1995) or subglacially (Denis et al., 2009). The escape of water reduces the pore water pressure beneath the glacier, which can increase the strength of the sediments and can also lead to the ice being coupled to the bed. The association of the clastic dike in the drumlin implies the dike was the result of pore water pressure release during the formation of the drumlin. 123 Figure 5.12: Divergence of Ice Flow Path around a Rigid Gravel Core. (Modified from Boulton, 1987) Figure 5.13: Clastic Dike Cross-Cutting Horizontal Beds in the Smith Pit. 124 Stage 1: Proglacial folding - D1 phase Stage 2: Proglacial thrusting and dceollemcnt D£collemenr Direction of D1 Fold Sediment tip ice (low axis- slacking Icebergs Nonh Stage 3: Fold overturning and sediment fluidisation Stage 4: Overriding and D2 phase deformation Penetrative Zone of glaei-tectonite Fluidiscd sand sediment Deformation till | Non-penetrative mobilised along imbrication glacitecionitc thrust plane Overturned D1 fold axis | South South North Figure 5.14: Sequence of Glacial Advance of the Gleann Einich Outlet Glacier. (Modified from Golledge, 2002) Sequence of Events The structures observed in the study area may have initiated proglacially but were subsequently overridden and enhanced subglacially. Figure 5.14 shows the sequence of glacial advance of the Gleann Einich outlet glacier into a lacustrine environment in the Cairngorm Mountains after the last glacial maximum about 18 ka 14C BP (Golledge, 2002). Ice advancing into a proglacial lake propagated folds along a decollement at the glacier margin. Continued advance overrode these folds and resulted in continued deformation of the substrate. Phillips et al. (2002, 2008) demonstrate a similar pattern of deformation in Scotland and Eastern England (Fig. 5.15). The readvance of the LML to the Kalamazoo Moraine in southwestern Michigan most likely followed a very similar sequence of events. The LML advanced from the Lake Michigan basin up slope into a lacustrine environment impounded initially by the slope of the topography and later dammed by the Kalamazoo Moraine. Elevated pore water pressures as described by Kehew et al. (2005) facilitated the deformation of the sediments. The lateral forces exerted by the glacier created folded ridges of sediment directly ahead of the advancing glacier as evident by the existence of shear surfaces and fold along those surfaces preserved in some gravel pit exposures. As the ice approached the western edge of the Valparaiso Upland, it encountered a mix of fine-grained sediment with packages of coarse- grained sediments. These packages of coarse-grained sediments were capable of transmitting pore water away efficiently, thus becoming relatively rigid cores. As the glacier continued over this ridge the areas of highest elevation became streamlined as ice flowed over and around the rigid cores. With meltwater now impounded to the east by the Kalamazoo Moraine, the pore water pressure in the fine-grained sediments 126 was sufficiently high to deform the sediments and in some instances great enough to decouple the ice from its bed. Kehew et al. (2005) attributes the fast flow of the LML, among other factors, to decoupling of the ice from its bed. Once the advance reached the Kalamazoo Moraine, deformation in the form of proglacial thrusting is evident. Subglacial water was released through the Kalamazoo Moraine as evident by the fans along the eastern side of the moraine. Upon retreat of LML the landscape was influenced by reworking from meltwater outbursts that flooded and sculpted the region. Channels draining water from the north and east can easily be traced across the landscape. The deformation observed across the study area is similar to what has been described by Piotrowski (2004) as a mosaic of spatially separated deformed areas within areas that are undeformed. The major controls on the location of deformation at work in the study area are sediment grain-size, pore water, and topography. The model of deformation for this area takes into account a combination of ideas from Tulaczyk (1999), Eyles and Boyce (1998), Boulton (1987), and also Piotrowski (2004). The areas of deformation occur generally in the topographically higher areas as these areas generally contain beds of coarse gravel. Alternately, the topographically lower areas generally contain fine-grained materials. The fine lacustrine sediments would favor increase pore water pressures and decoupling, resulting in sliding. Figure 5.16 depicts patches being deformed when the shear stress is greater than shear strength used by Piotroski and Kraus (1997). This model fits the study area with the addition of the deforming spots are the areas of higher topography and a mixed rheology of deforming sediments. 127 NB-directed advance of ^ ridge into Lake Blane Loch Lomond glaciei coalesced proximal deltaic D1 deformation ouiwash fans s^uence ^ ^^ wWl well developed foresett & topsels ® inclined to recumbent large-scale F1 folds'* subglacial D1 deformation undefomred deltaic sediments equivalent to Unit I decreasing intensity of D1 deformauon variable ice-flow .fenosilion or subglacial directions due to surface diomicum (Unit ltl) NE-ditected advance of Loch Lomond glaciej no evidence for a D2 deformation 02 thrast-block ridge © P2 shear zones subglacial disruption Unit 11 Unit II pjo-glacial D2 deformation undefornted deltaic sediments subglacial D2 deformation equivalent to Unit I restricted to minor ttmisting leading to shearing and disrapuon of Unitll sediments decreasing intensity of D2 deformation ssssasssc-""" m 128 Figure 5.16: Model of the Subglacial Deforming/Sliding Bed. (Modified from Piotrowski and Kraus, 1997) Conclusions An analysis of deformed glacigenic sediment was conducted across southwest Michigan in order to determine the cause of the deformation. The structures observed were evaluated to determine the role of glacial ice in the formation of these structures. The stresses involved are either dynamic shear stresses exerted by the flowing ice upon the substrate or extensional stresses that develop as buried ice melts. The glacial history of the area can be reconstructed by evaluating not only the stratigraphy but also the deformation observed. Specific conclusions include: • Structural analysis has demonstrated that the deformation along the Valparaiso Upland, including portions of the Inner Kalamazoo Moraine and Kendall Moraine, is consistent with stresses imposed by the LML as it advanced across the region to the Outer Kalamazoo Moraine. • The compressional nature of the deformation observed, along with the consistency of the stress indicators, rejects ice block melt-out as a cause of the widespread deformation in the study area. • During the re-advance that deposited the Saugatuck Till, ice advanced into a lacustrine environment which facilitated deformation by increasing pore water pressure in the subglacial sediments. • Deformation was concentrated on the topographic highs along the Valparaiso Upland. Areas of significant relief are more subject to deformation than low- relief areas (Denis et al., 2009; Tulaczyk, 1999; Eyles and Boyce, 1998). The topographic highs contain coarse sediment. These well drained, coarse materials behave as rigid bodies within the poorly drained, weaker lacustrine sediments (Boulton, 1987). The resulting contrast of rheology leads to the formation of drumlins within a narrow zone perpendicular to ice flow. • The deformation is most likely the result of a varied stress regime. As the ice 130 advanced, the horizontal stresses deformed sediment along a decollement proglacially. The advancing ice then overrode these deformed sediments to induce a greater amount shear and deformation • The application of techniques used by a metamorphic petrologist in terms of fold, fault and fabric measurement can and should be applied to the field of glacial geology as the formation till and stresses invoked by moving ice is not unlike shear zones encountered in metamorphic terrains. The active or passive interaction of ice with the glacial sediment will leave a diagnostic pattern of deformation that may only be truly expressed after structural analysis. 131 APPENDIX 132 Tacy Pit Data Clast Fa jric Trenc and Plunge Fold Data (Right 260 32 295 24 305 20 290 19 Fold 1 Fold 2 320 22 340 40 250 34 10 16 170 44 60 49 325 30 322 25 345 20 30 34 145 44 43 25 342 7 270 16 140 15 318 25 185 29 60 36 281 19 340 27 310 9 200 36 135 12 270 4 320 37 350 36 224 50 328 17 300 12 315 20 234 62 339 12 340 8 290 32 230 85 150 14 260 18 298 32 55 72 330 6 286 17 350 10 Fold 3 Fold 4 130 15 300 20 205 33 38 80 95 38 235 24 310 22 250 4 222 83 265 43 316 8 300 28 270 16 25 82 265 52 223 5 332 7 265 5 198 77 275 68 262 12 254 34 280 50 75 50 277 76 10 16 306 25 335 5 65 54 110 25 284 5 284 10 311 20 75 48 95 90 300 15 275 12 255 25 65 50 110 68 303 32 297 24 280 18 185 75 310 25 293 19 320 10 204 90 120 8 358 18 330 7 20 80 10 22 0 31 318 5 335 20 68 25 270 14 Minor Fold Axes 0 26 318 17 320 28 (Trend and Plunge) 290 5 332 32 260 15 248 9 290 30 290 31 328 22 216 14 180 11 320 42 95 13 230 6 280 12 339 48 310 23 85 4 265 20 90 25 290 24 280 6 0 18 290 31 270 43 244 16 325 15 295 20 305 25 49 4 0 29 340 36 292 22 219 34 300 10 288 18 300 23 97 11 65 12 325 19 280 10 290 19 310 42 320 52 321 30 275 12 330 36 310 32 295 10 260 18 68 25 300 12 175 12 342 40 322 10 302 27 133 White Pit Data Fold Data (Right hand rule) Fold 1 Fold 2 Minor Fold Axes 245 15 210 15 (Trend and Plunge) 250 67 215 67 190 12 240 80 205 80 10 3 90 90 55 90 22 25 208 16 173 16 217 8 75 55 40 55 220 10 260 19 225 19 166 20 131 20 221 30 186 30 240 40 205 40 Wyoming Pit Data Fold Data (Right hand rule) Fold 1 Fold 2 Minor Fold Axes 137 29 345 26 (Trend and Plunge) 300 24 170 35 137 21 130 20 185 24 140 2 350 10 302 15 174 13 330 25 15 20 157 2 140 30 154 29 318 15 310 17 Till Plane (RHR) 180 16 Minor Fold Axial Plane 165 89 149 84 20 78 Michigan Paving Pit Data Fold Data (Right hand rule) Fold 1 Fold 2 254 32 254 32 231 54 231 54 236 15 236 15 265 30 265 30 243 48 250 53 286 16 286 16 134 Cleveland Pit Data Clast Fabric Trend and Plunge Fold Data (Right hand rule) 308 3 120 3 139 6 Fold 1 Fold 2 280 1 132 6 140 27 290 75 267 31 87 10 132 16 139 15 300 55 255 33 118 12 310 4 145 4 295 85 255 42 290 15 130 13 147 16 295 75 260 59 160 11 125 34 147 16 328 62 355 52 130 3 270 16 148 3 330 60 345 25 130 2 170 12 149 7 35 40 340 35 125 9 95 2 150 5 28 42 352 41 155 9 295 3 152 17 75 48 290 14 148 3 155 10 76 50 130 5 112 8 155 0 290 26 148 18 142 36 157 14 50 25 135 12 150 5 162 8 145 2 147 21 166 3 Minor Fold Axes 140 25 87 3 173 12 (Trend and Plunge) 135 32 85 8 270 9 25 15 102 2 86 11 284 26 110 10 155 2 87 10 282 5 115 28 323 6 89 6 282 0 76 25 75 10 95 0 286 15 30 30 295 6 102 2 288 6 40 16 85 10 112 8 290 14 108 38 285 3 90 12 292 5 80 0 150 5 124 17 295 9 40 8 150 14 119 12 308 3 106 39 353 3 120 22 310 4 80 8 325 2 122 4 320 7 155 15 125 7 322 3 137 10 126 26 328 8 127 6 127 6 322 6 118 21 129 8 335 6 129 0 120 26 130 10 118 15 130 19 310 4 131 8 275 35 132 22 288 20 135 19 173 14 137 29 135 Remington Pit Data Clast Fabric Trenc and Plunge 335 14 290 12 Fold Data (RHR) 338 16 105 6 Fold 1 320 13 150 13 230 38 325 19 290 17 225 72 312 20 325 2 210 70 317 9 265 16 205 42 285 8 160 5 205 71 310 5 295 12 35 78 320 21 305 22 30 62 335 18 315 11 218 53 340 8 100 7 332 21 318 13 Minor Fold Axes 345 4 (Trend and Plunge) 338 24 33 4 290 8 18 18 320 22 30 42 325 18 240 20 310 19 26 15 315 8 40 16 300 4 253 25 160 18 310 5 Till Bed (RHR) 305 11 90 2 150 13 105 3 331 16 335 26 Fault Planes 305 9 44 76 130 11 205 20 195 13 290 18 65 4 329 12 135 6 330 7 318 22 142 12 320 9 340 5 136 Klett Pit Data Clast Fa Dric Trenc and Plunge Fold Data (Right hand rule) 261 36 28 24 232 15 130 12 Fold 1 Fold 2 300 38 270 12 251 20 255 23 180 37 64 38 312 8 262 31 273 22 310 18 182 18 175 33 285 21 272 10 53 25 225 22 30 35 122 35 284 3 230 14 258 23 295 10 32 65 Minor Fo d Axes 282 32 243 7 306 12 300 12 Fold 3 (Trend and Plunge) 295 8 285 24 287 33 322 10 0 4 202 10 230 16 218 28 230 24 320 28 228 68 44 8 245 16 115 22 239 18 260 15 230 18 60 25 66 28 263 30 275 10 328 22 23 40 226 6 43 27 290 28 243 31 95 13 245 43 75 3 275 42 322 16 268 12 310 23 40 23 264 20 268 12 230 19 290 24 105 17 300 17 283 7 247 13 270 43 Till Bed (RHR) 291 30 290 32 288 18 305 25 25 25 225 28 275 9 210 33 292 22 61 26 300 23 280 10 320 52 330 36 37 13 260 18 175 12 302 27 39 21 Kusmack Pit Data Fold Data (Right hand rule) Fold 1 160 33 170 12 35 20 54 31 137 Smith Pit Data Clast Fa brie Trend and Plunge Fold Data (Right hand rule) 0 25 290 24 Fold 1 Fold 2 342 15 280 30 15 40 258 33 345 28 340 10 5 35 245 25 7 25 110 30 355 57 285 35 330 15 260 24 5 72 285 28 10 25 333 12 205 35 50 44 30 33 312 20 185 60 35 47 140 26 310 4 200 44 40 42 345 32 140 21 353 84 321 15 328 31 5 42 325 5 342 16 10 30 25 40 160 10 20 35 155 9 310 21 20 2 315 24 Fold 3 Fold 4 135 10 300 38 Minor Fold Axes 340 62 330 40 42 5 (Trend and Plunge) 342 40 315 63 75 15 170 40 50 25 335 38 100 12 180 33 40 18 150 35 325 11 190 22 60 25 90 30 190 30 100 35 45 29 280 25 0 28 120 25 335 30 120 48 135 35 100 45 354 75 128 42 140 30 90 15 128 42 320 15 80 12 Fold 5 325 28 92 18 244 66 10 14 7 2 239 80 5 25 58 16 235 85 30 29 133 6 230 90 330 29 151 20 245 72 350 36 59 5 59 80 320 21 58 75 135 9 60 55 322 15 Clay Par' ings (RHR) 320 31 Till Bed (RHR) 255 74 355 11 68 26 230 75 65 20 200 75 245 80 205 65 138 Brewster Pit Data Fold Data (Right hand rule) Fold 1 Fold 2 320 66 350 8 300 57 353 6 305 62 2 25 345 80 8 38 350 70 190 15 16 70 175 15 45 58 170 42 322 80 JD Pit Data Fold Data (Right hand rule) Fold 1 Fold 2 JD Faults 300 23 135 15 220 26 300 23 125 12 0 27 290 22 140 9 235 20 5 14 230 9 10 33 331 34 15 24 215 26 30 10 260 17 315 30 230 32 139 BIBLIOGRAPHY Aber JS, Croot DF, Fenton, MM., 1989. Glacitectonic Landforms and Structures. Kluwer: Dordrecht. Albino,K., Dreimanis, A., 1988. A time transgressive kinostratigraphic spanning 180o in a single section at Bradtvill, Onterio, Canada. In Croot, D.G., Glaciotectonics. Forms and Processes. Balkema, Rotterdam, 11-20. Andersen, L.T.,Hqansen, D.L., Huuse, M., 2005. Numerical modelling of thrust structures in unconsolidated sediments: Implications for glaciotectonic deformation Journal of Structural Geology 27, 587-596 Anderson, E. M., 1942. The dynamics of faulting and dyke formation with application to Britian: Oliver and Boyd, Edinburgh, 191p. Andrews, J.T., 1970. Techniques for till fabric analysis. Technical Bulletin of the British Geomorphological Group Number 6. Bird, B.C., Kozlowski, A.L., Kehew, A.E., 2000, Interpretation of 3 Dimensional Glacial Mapping of the Paw Paw and Decatur 7.5 minute Quadrangles Michigan, Geological Society of America, Abstracts with Programs, Annual Meeting, v. 32, no 7 p. A-19. Bird, B. C., 2005. Glacial stratigraphy and Surficial Geology of the Decatur, Lawrence, and Paw Paw U.S.G.S. 7.5 Minute Quadrangles in van Buren County Michigan. Unpublished M.S. Thesis, Western Michigan University, USA. Benediktsson, I.O., Moller, P., Ingolfsson, O., van der Meer, J.J.M., Kjaer, K.H., Kriiger, J., 2008. Instantaneous end moraine and sediment wedge formation during the 1890 surge of Bruarjokull, Iceland. Quaternary Science Reviews 27, 209-234 Benn, D.I., 1994. Fluted moraine formation and till genesis below a temperate valley glacier: Slettmarkbreen, Jotunheimen, Southern Norway. Sedimentology 41, 279-292. Benn, D.I., Evans, D.A., 1993. Glaciomarine deltaic deposition and ice-marginal tectonics: The Loch Don Sand Moraine, Isle of Mull, Scotland. Journal of Quaternary Science. 8, Pages 279-291 Benn, D.I., Evans, D.A., 1996. The interpretation and classification of subglacially- deformed materials. Quaternary Science Reviews 15: 23-52. Benn, D.I., Evans, D.A., 1998. Glaciers and Glaciation. Arnold: London 734pp. 140 Benn, D.I., Prave, A.R., 2006. Subglacial and Proglacial glaciaotechtonic deformation in the Neoproterozoic Port Askaig Formation Scotland. Geomorphology 75, 266- 280. Bennett, M. R., Waller, R. I., Glasser, N. F., Hambrey, M. J. and Huddart, D. 1999. Glacigenic clast fabrics: genetic fingerprint or wishful thinking? J. Quaternary Sci., Vol. 14, pp. 125-135. Beukema, S. P., 2003. Stratigraphy of Lake Michigan Lobe Deposits in Van Buren County, Michigan. M.S. Thesis. Western Michigan University, Kalamazoo, Boulton, G.S., 1986. Push-moraines and glacier-contact fans in marine and terrestrial environments. Sedimentology 33 (5), 677-698. doi:10.111 l/j.1365- 3091.1986.tb01969.x. Boulton, G.S., 1987. A theory of drumlin formation by subglacial deformation. In: Menzies, J., Rose, J. (Eds.), Drumlin Symposium. Balkema, Rotterdam, pp. 25-80Boulton, G.S., Dobbie, K.E., Zatsepin, S., 2001. Sediment deformation beneath glaciers and its coupling to the subglacial hydraulic system. Quaternary International 86, 3-28. Boulton, G.S., 1996. Theory of glacial erosion, transport and deposition as a consequence of subglacial sediment deformation. Journal of Glaciology 42, 43-62. Boulton, G. S., Caban, P. E. 1995. Groundwater flow beneath ice sheets: Part II its impact on glacier tectonic structures and moraine formation. Quaternary Science Reviews 14, 563-587. Boulton, G.S. and Hindmarsh, R.A., 1987. Sediment deformation beneath glaciers: Rheology and geological consequences. Journal of Geophysical Research, 92, p. 9059-9082. Boulton, G.S., van der Meer, J.J.M., Hart, J.K., Beets, D., Ruegg, G.H.J., van der Wateren, F.M., Jarvis, J., 1996. Till and moraine emplacement in a deforming bed surge an example from a marine environment. Quaternary Science Reviews 15, 961-988. Boulton, G.S., Van der Meer, J.J.M., Beer, D.J., Hart, J.K., Ruegg, G.H.J., 1999. The sedimentary and structural evolution of a recent push moraine complex: Holmstromsbreen, Spitsbergen. Quaternary Science Reviews 18, 339-371. Brodzikowski, K. and van Loon, A.J. 1991. Glacigenic Sediments. Elsevier, Amsterdam 141 Carr, S. J. & Goddard, M. A. 2007 (October): Role of particle size in till-fabric characteristics: systematic variation in till fabric from Vestari-Hagafellsjokull, Iceland. Boreas, Vol. 36, 371-385 Carr, S.J., Rose, J., 2003. Till Fabric Patterns and significance: particle response to subglacial stresses. Quaternary Science Reviews 22, 1415-1426. Catacosinos, P.A., Harrison, W.B., Reynolds, R.F., Westjohn, D.B., Wollensak, M.S. 2001. Stratigraphic Lexicon for Michigan. Mi DEQ Geology division and Mi Basin Geologic Society. Bulletin 8, Lansing, Michigan, p. 18-19. Denis, M., Buoncristiani,J.F., Guiraud, m., Fluid-pressure controlled soft-bed deformation sequence beneath the surging Brei6amerkurjokull (Iceland, Little Ice Age), Sediment. Geol. (2009), doi:10.1016/j.sedgeo.2009.07.013 Dreimanis, A., Rappol, M., 1997. Late Wisconsinan sub-glacial clastic intrusive sheets along Lake Erie bluffs, at Bradtville, Ontario, Canada. Sedimentary Geology 111,225-248 Evans, D.J.A., Phillips, E.R., Hiemstra, J.F., Auton, C.A., 2006. Subglacial till: formation, sedimentary characteristics and classification. Earth Science Reviews 78, 115-176. Farrand, W.R. and Bell, D.L. 1982. Quaternary Geology of Michigan 1:500,000 (map), University of Michigan Department of Geological Sciences; Michigan Department of Natural Resources. Farrand, W. R. and Eschmann, D.F., 1974. Glaciation of the southern peninsula of Michigan: a review. Michigan Academician, v7, P.31-56 Fullerton, D.S., 1980, Stratigraphy and correlation of glacial deposits from Indiana to New York and New Jersey. Quaternary Science Reviews, v5, 23-37. Golledge, N., 2002. Glaci-tectonic deformation of proglacial lake sediments in the Cairngorm Mountains. Scottish Journal of Geology 38, 127-136. Hansel, A.K., Johnson, W.H.,1992. Fluctuations of the Lake Michigan lobe during the late Wisconsin subepisode. Sveriges Geologiska Undersokning Ser. Ca 81, 133-144. Hansel, A.K., Mickelson, D.M., Schneider, A.F., and Larsen, C.E. 1985. Late Wisconsin andHolocene history of the Lake Michigan basin. In Quaternary Evolution of the Great Lakes, eds. P.F. Karrow and P.E. Calkin, pp. 39-53. Geol. Assoc. of Canada Special Paper 30.et al., 1985; Hansen, K., 1971. Tunnel valleys in Denmark and Northern Germany, Bulletin of the Geological Society of Denmark 20 (1971), pp. 295-306. 142 Hart, J.K., 1990. Proglacial glaciotectonic deformation and the origin of Cromer Ridge push moraine complex, North Norfolk, England. Boreas 19, 165-180. Hart, J.K., 1997. The relationship between drumlins and other forms of subglacial deformation. Quaternary Science Reviews 16, 93-108. Hart, J.K. and Boulton, G.S. (1991). The interrelationship between glaciotectonic deformation and glaciodeposition within the glacial environment. Quaternary Science Reviews, 10, 335-350. Hart, J., Rose J., 2001. Approaches to the study of glacier bed deformation. Quaternary International 86, 45-58. Hicock, S. R. & Dreimanis, A. 1985. Glaciotectonic structures as useful ice movement indicators in glacial deposits: four Canadian case studies. Canadian Journal Earth Sciences 22, 339-346. Hicock, S.R., Goff, J.R., Lian, O.B., Little, E.C., 1996. On the interpretation of subglacial till fabric. Journal of Sedimentary Research 66, 928-934. Hiemstra, J. F., Rijsdijk, K. F., Evans, D. J. A. & van der Meer, J. J. M. 2005 (February): Integrated micro- and macro-scale analyses of Last Glacial Maximum Irish Sea Diamicts from Abermawr and Traeth y Mwnt, Wales, UK. Boreas, Vol. 34, pp. 61-74. Higuera-Diaz, I.C., Stravers, J., D Kulczycki, D., 2008. Glaciotectonic deformation of Wisconsin Episode sediment: Shallow decollments, sheath folding and corroborating strain indicators in overlying diamicton. In Curry, B., ed., Deglacial history and paleoenvironments of northeastern Illinois, 54th Midwest Friends of the Pleistocene Field Conference; Illinois State Geological Survey, Open File 2008-1, p. 152-163 Hudleston, P. J., 1986. Extracting information from folds in Rock. Journal of Geological Education 34, 237-245. Iverson, N.R., Cohen, D., Hooyer, T.S., Fischer, U.H., Jackson, M.,Moore, P.L., Lappegard, G., Kohler, J., 2003. Effects of basal debris on glacier flow. Science 301, 81-84. Johnson, W.H., Hansel, A.K., 1999. Wisconsin Episode glacial landscape of central Illinois: A product of subglacial deformation processes. In: Mickelsen, D.M., Attig, J.W. (Eds.), Glacial Processes Past and Present. Geological Society of America Special Paper 337, pp. 121-135. 143 Johnson, W.H., Hansel, A.K., Bettis III, E.A., Karrow, P.F., Larson, G.J., Lowell, T.V., Schneider, A.F., 1997. Late Quaternary temporal and event classifications, Great Lakes Region, North America. Quaternary Research 47, 1-12 Karrow, P.F., Dreimannis, A., Barnet, P.J., 2000. A Proposed Diachronic Revision of Late Quaternary Time-Stratigraphic Classification in the Eastern and Northern Great Lakes Area. Quaternary Research 54, 1-12 Kozlowski, A. L., 2004. The origin of the central Kalamazoo RiverValley: relationships between ice dynamics, the Lake Michigan lobe, and theSaginaw lobe of the Laurentide Ice Sheet. Ph.D. dissertation, WesternMichigan University, Kalamazoo. Kozlowski, A.L., Kehew, A.E., Bird, B.C., 2005. Outburst flood origin of the central Kalamazoo River valley, Michigan. Quaternary Science Reviews, 24, p. 2354- 2374. Kozlowski, A.L., Kehew, A.E,. Bird, B.C. 2001. An outburst hypothesis for the origin of the Kalamazoo river valley, Kalamazoo and Allegan counties, Michigan, Geological Society of America Abstracts with Programs 33, p. 269. Kehew A.E., Beukema S.P., Bird B.C., Kozlowski A.L.,2005. Fast flow of the Lake Michigan Lobe: Evidence from sediment-landform assemblages in southwestern Michigan, USA, Quaternary Science Reviews, 24, p. 2335-2353 Kehew, A. E, Bird, B. C., Beukema, S. P., Kozlowski, A. L., Lee, J., 2002. Glacial terrain map of Van Buren County, Michigan. Michigan Department ofEnvironmental Quality. Kehew, A. E., Bird, B. C., Kozlowski, A. L., Wong, S. A., and Beukema, S. P., 2001. Origin of the Valparaiso moraine, Van Buren County, Michigan. Geological Society of America Abstracts with Programs, Vol. 33, No. 6, p. 268. Kincare, K., 1999. Landforms and Process on the Valparaiso and Inner Kalamazoo Moraines, Berrien County, Michigan. AIPG Berrien County Glacial Geology Field Trip Guide Larsen, N.K., Piotrowski, J.A., 2003. Fabric pattern in a basal till succession and its significance for reconstructing subglacial processes. Journalof Sedimentary Research 73, 725-734. Larsen NK, Piotrowski JA, Christoffersen P, Menzies J. 2006a. Formation and deformation of basal till during a glacier surge; Elisebreen, Svalbard. Geomorphology 81: 217-234 144 Larson, G., and Schaetzl, R. 2001. Origin and evolutionof the Great Lakes. J. Great Lakes Res. 27:518-546. Lee, J.R., Phillips, E.R., 2008. Progressive soft sediment deformation within a subglacial shear zone - a hybrid mosaic-pervasive deformation model for Middle Pleistocene glaciotectonised sediments from eastern England. Quaternary Science Reviews 27, p. 1350-1362. Leverett, F., and Taylor, F., 1915. The Pleistocene of Indiana and Michigan and theHistory of the Great Lakes. U.S. Geological Survey Monograph 53, p. 527. Maizels, J.K. 1992. Boulder ring structures produced during jokulhlaup flows: origin and hydraulic significance. Geografiska Annaler 74A, 21-33. Mark, D.M., 1973. Analysis of axial orientation data, including till fabrics. Geological Society of America Bulletin 84, 1369-1374. Marshak, S. and Mitra, G., 1988. Basic Methods of Structural Geology. Prentice Hall. 448p. Martin, H. M. 1955. Map of the Surface Formations of the Southern Peninsula of Michigan 1:500,000. Geological Survey Division, Michigan Dept. of Natural Resources. McCarroll, D. and Rijsdijk, K. F. 2003. Deformation styles as a key for interpreting glacial depositional environments. J. Quaternary Sci., Vol. 18 pp. 473-489. Mies, J.W., 1991. Planar dispersion of folds in ductile shear zones and kinematic interpretation of fold-hinge girdles. Journal of Structural Geology 13, 281- 297. Menzies, J., 1990. Sand Intraclasts within a diamicton melange, southern Niagara Peninsula, Ontario, Canada. Journal of Quaternary Science, 5: 189-206. Monaghan, G.W., 1990. Systematic variation in the clay-mineral composition of till sheets: Evidence for the Erie Interstade in the Lake Michigan basin. In: Schneider, A.F., Fraser, G.S. (Eds.), Late Quaternary History of the Lake Michigan Basin. Geological Society of America Special Paper 251, pp. 43-50 Monaghan, G. W., Larson, G. J., and Gephart, G. D., 1986. Late Wisconsinan driftstratigraphy of the Lake Michigan Lobe in southwestern Michigan. Geological Society of America Bulletin, Vol. 97, p. 329-334. Moller, P., 2006. Rogen moraine: an example of glacial reshapingof pre-existing landforms. Quaternary Science Reviews 25, 362-389. 145 Phillips, E.R., Evans, D.J.A., Auton, C.A., 2002. Polyphase deformation at an oscillating ice margin following the Loch Lomond Readvance, central Scotland, UK. Sedimentary Geology 149, 157-182. Phillips,E., Lee J.R., Burke, H., 2008. Progressive proglacial to subglacial deformation and syntectonic sedimentation at the margins of the Mid-Pleistocene British Ice Sheet: evidence from north Norfolk, UK. Quaternary Science Reviews 27, 1848-1871 Phillips, E., Merritt, J., Auton, C., Golledge, N., 2007. Microstructures in subglacial and proglacial sediments: understanding faults, folds and fabrics, and the influence of water on the style of deformation. Quaternary Science Reviews 26, p. 1499-1528. Piotrowski, J. A.. 1997. Subglacial Environments- an introduction. Sedimentary Geology 111, 1-5 Piotrowski, J. A., Hermanowski, P., Piechota, A.M., 2009. Meltwater discharge through the subglacial bed and its land-forming consequences from numerical experiments in the Polish lowland during the last glaciations. Earth Surf. Process. Landforms 34, 481-492 Piotrowski, J.A., Larsen, N.K., Junge, F.W., 2004. Reflections on soft subglacial beds as a mosaic of deforming and stable spots.Quaternary Science Reviews 23, 993- 1000. Piotrowski, J.A., Larsen, N.K., Menzies, J., Wysota, W., 2006. Formation of subglacial till under transient bed conditions: deposition, deformation, and basal decoupling under a Weichselian ice sheet lobe, central Poland. Sedimentology 53, 83-106. Piotrowski, J.A., Kraus, A.M., 1997. Response of sediment to ice-sheet loadingin northwestern Germany: effective stresses and glacier-bed stability. Journal of Glaciology 43, 495-502. Piotrowski, J.A., Mickelson, D.M., Tulaczyk, S., Krzyszkowski, D., Junge, F.W., 2001. Were deforming subglacial beds beneath past ice sheets really widespread? Quaternary International 86, 139-150. Rijsdijk KF, Owen G, Warren WP, McCarroll D, Van der Meer JJM., 1999. Clastic dykes in over-consolidated tills: evidence for subglacial hydrofracturing at Killiney Bay, eastern Ireland. Sedimentary Geology 129: 111-126. Rijsdijk, K.F., Warren, W.P., Van der Meer, J.M., 2010. The glacial sequence at Killney, SE Ireland: terrestrial deglaciation and polyphase glaciotectonic deformation. Quaternary Science Reviews 29, 696-719. 146 Roberts D.H., Hart J,. 2005. The deforming bed characteristics of a stratified till assemblage in north East Anglia, UK: investigating controls on sediment rheology and strain signature. Quaternary Science Reviews 24: 123-140. Rieck,R.L., Klassner, J.S., Winters, H.A., Marlette, P.A. 1991. Glaciotectonic effects on a Middle Wisconsin boreal fenland peat in Michigan, USA. Boreas, Vol.20, 155-167. Sharpe, D.R. 1988. Late glacial landforms of Wollaston Pennisula, Victoria Island, NWT: product of ice-marginal retreat, surge, and mass stagnation. Canadian Journal of Earth Sciences 25, 262-279. Stone, B.D. 2001. Surficial geologic map of Berrien County, Michigan; U.S. Geological Survey Digital Open-file Report 01-256, scale 1:100,000, explanatory text. Stone, B.D., Kincare, K.A., O'Leary, D.W., Lundstrom, S.C., Taylor, E.M., Brown, S.E., 2003. Glacial and postglacial geology of the Berrien County Region of Michigan. Guidebook of the 49th Midwest Friends of the Pleistocene Field Conference. Stumpf, A.J., Barnhardt, M.L., Hansel, A.K., 2004. Surficial Geology of Wauconda Quadrangle. Illinois State Geologic Survey. Terwilliger, 1954. The glacial geology and groundwater resources of Van Buren County, Michigan. Occasional papers for 1954, Michigan State Geology Survey, p. 95. Van der Wateren, F.M., 1999. Structural geology and sedimentology of the Heiligenhafen till section, Northern Germany. Quaternary Science Reviews 18, December 1999, Pages 1625-1639 van der Meer, J.J.M., Kjser, K., Kruger, J., 1999. Subglacial waterescape structures and till structure, Slettjokull, Iceland. Journal of Quaternary Science 14, 191-205. van der Meer, J.J.M., Menzies, J., Rose, J., 2003. Subglacial till: the deforming glacier bed. Quaternary Science Reviews 22, 1659-1685. Winters, H. A., Rieck, R.L., 1991. Late glacial terrain transformation in Michigan. Michigan Academician, 23, 137-148. Zumberge, J. H., 1960. Correlation of Wisconsinan drifts in Illinois, Indiana, Michigan, and Ohio. Geological Society of America Bulletin, Vol. 71, p. 1177-1188. 147