Ao.‘

 

l
Rodi

 

LIBRARY
Michigan State
University

 

This is to certify that the
thesis entitled

A GEOPHYSICAL INVESTIGATION OF LARGE-SCALE
GLACIOTECTONIC DEFORMATION, LUDINGTON RIDGE,
MICHIGAN

presented by

ROBERT L. AYLSWORTH JR.

has been accepted towards fulfillment
of the requirements for the

Master’s degree in Geolgical Sciences

 

 

fl”? 2 MW perm

Major Professor’s Signature

QCEVMAU/ [0‘1 20295)

Date

MSU is an Al‘i‘innative Action/Equal Opportunity Employer

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 KIIProj/AccaPres/CIRC/DateDue indd

 

A GEOPHYSICAL INVESTIGATION OF LARGE-SCALE GLACIOTECTONIC
DEFORMATION, LUDINGTON RIDGE, MICHIGAN

By

Robert L. Aylsworth Jr.

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Geological Sciences

2008

ABSTRACT

A GEOPHYSICAL INVESTIGATION OF LARGE-SCALE GLACIOTECTONIC
DEFORMATION, LUDINGTON RIDGE, MICHIGAN

By
Robert L. Aylsworth Jr.

Late Wisconsin glaciotectonic deformation structures are visible in a 1.5km long
section of cliff face along the eastern shore of Lake Michigan south of Ludington,
Michigan. Several apparent clay diapirs rise from below beach level to near the top of the
~50m high cliff. The sediments exposed in the cliff face in order from oldest to youngest
are a lower diamict, a lower stratified sand, an upper diamict, an upper stratified sand,
and finally an eolian sand cap. The lower stratified sand is characterized by laminated
clay and silt which was mobilized and deformed by glaciotectonic processes, and for the
deformation structures.

Springs and gullies on the surface indicate a preferred pattern of groundwater
drainage and past landslides. This project has concluded that at least one of these
deformation structures is a continuous feature that does trend toward the northeast. Such
information is vital in order to determine ice flow direction and preferred drainage
patterns. Little is known about the local stratigraphy near the deformation structures, as
the public borehole information is sparse. This study characterizes the inland extent of
these deformation structures using geophysical methods, and improves understanding of
the glacial history of the Ludington Ridge, specifically the related glaciotectonic

pI'OCCSSCS.

ACKNOWLEDGEMENTS
First and foremost I would like to thank my advisor, Dr. Remke van Dam, for all his help
with the field work, background literature, and helping with my writing. He pushed me
harder than I thought I could go, and I became a better scientist because of it. Also, thank
you to Dr. Dave Hyndman and Dr. Grahame Larson for serving on my committee, and
offering great insights and comments on my research. This project was made possible
through generous support from the Warren W. and Anneliese C. Wood fellowship.
Without their support, summer field work would not have been possible. I also have to
give a huge thanks to Michael S. Morse, who spent the entire summer with me in
Ludington, camping on a cliff overlooking Lake Michigan and walking through poison
ivy for two months. We had a great time and I could never have done this without him.
Thank you to the Anderson family, Dow Chemical, Consumers Energy, and all the other
landowners, for providing permission to work on their land and historical accounts.
Finally, thank you to my parents for their all of their support, both financially and

emotionally, even when things looks tough and the end was apparently nowhere in sight.

TABLE OF CONTENTS

LIST OF TABLES .................................................................................. v
LIST OF FIGURES ................................................................................. vi
1. INTRODUCTION AND LITERATURE REVIEW .......................................... 1
I. a. Background and History of Glacial Deformation ................................. 1
- I. b. Glaciotectonic deformation structures ............................................. 2
I. c. Characterization of Glaciotectonic Deformation Structures .................... 10
I. c. i. Traditional Methods ....................................................... 10
I. 0. ii. Geophysical Methods .................................................... 12
I. d. Field Site Description and Objective ............................................... 15
I. e. Hypotheses ............................................................................ 18
I. e. i ................................................................................. 1 8
I. e. ii ............................................................................... 18
I. e. iii ............................................................................... 19
I. e. iv. .............................................................................. 19
II. METHODOLOGY .............................................................................. 20
II. a. Background ........................................................................... 20
II. b. Electrical Resistivity Surveys ...................................................... 22
II. b. i. Vertical Electrical Sounding (V ES) .................................... 23
II. b. ii. Constant Spread Traverse (CST) ..................................... 24
II. b. iii. Multi-electrode Survey ................................................ 25
II. c. Other Methods ........................................................................ 26
II. c. i. Soil Box Methods ......................................................... 27
II. c. ii. Grain Size Analysis ...................................................... 27
II. 0. iii. Soil Moisture Curve .................................................... 30
II. 0. iv. Inversion and Forward Modeling ..................................... 30
III. RESULTS ........................................................................................ 31
IV. DISCUSSION ................................................................................... 53
V. CONCLUSIONS ................................................................................. 56
VI. REFERENCES .................................................................................. 58

iv

LIST OF TABLES

TABLE 1: Factors considered important for genesis of glaciotectonic
phenomena9

TABLE 2: Measurement details for surveys shown in figure 9 ..................................... .18
TABLE 3: Resistivity values for various materla1522
TABLE 4: Grain size distribution data and Rosetta modeling results ............................ 33

TABLE 5: Calculated values of resistivity using equation for best fit line from soil box
resistivity curves ......................................................................................... .34

LIST OF FIGURES
(Images in this thesis are presented in color.)

FIGURE 1: Triad of effects created by glaciation, on which modern glacial theory is
based ................................................................................. 1

FIGURE 2: Chart of common glaciotectonic structures arranged according to their
typical horizontal scales ........................................................... 3

FIGURE 3: Cartoon illustrating how large intrusive structures in the subsurface may
have no morphologic expression at the surface ................................ 5

FIGURE 4: Landscape zones of the Laurentide Ice Sheet: 1= zone of extensive end
moraines, hummocky moraines and ice-shoved hills; 2= zone with long
eskers and ice-flow lineaments; 3= zone of extensive ribbed moraine,
drumlins, and eskers ................................................................ 8

FIGURE 5: Glacial features of central Brookhaven, CT .................................. 11

FIGURE 6: GPR image showing a till layer boundary and hyperbolas interpreted as
cobbles ............................................................................. 13

FIGURE 7: 50 MHz GPR data with the interpretation below. Evidence of
glaciotectonic thrusting is interpreted in the glaciofluvial-lacustrine

sediments ........................................................................... 14

FIGURE 8: A cross-section showing the sediments exposed in the western end of the
Ludington Ridge .................................................................. 15

FIGURE 9: Aerial photo of the field site showing all transects, data collection sites,
and outcrops ........................................................................ 17

FIGURE 10: Grain size distribution curves for each of the four distinct layers in the
Ludington ridge ...................................................................................... 32

FIGURE 11: Results of the soil box resistivity test ..................................................... 33
FIGURE 12: Soil box resistivity curves with residual moisture contents indicated....34

FIGURE 13: Results of the DCINV modeling to determine the stratigraphy near the
intersection of Iris Rd. and S. Lakeshore Rd ......................................... 36

FIGURE 14: Results of the DCINV modeling to determine the stratigraphy near the
intersection of Chauvez Rd. and S. Lakeshore Rd ................................ 37

vi

FIGURE 15:

FIGURE 16:

FIGURE 17:

FIGURE 18:

FIGURE 19:

FIGURE 20:

FIGURE 21:

FIGURE 22:

FIGURE 23:

FIGURE 24:

FIGURE 25:

FIGURE 26:

FIGURE 27:

FIGURE 28:

Results of the DCINV modeling to determine the stratigraphy on the

beach between the two southern most diapirs ....................................... 38
Results of the DCINV modeling to determine the stratigraphy on the

Haul Rd. 720m north of Chauvez Rd .................................................... 39
Stratigraphic information from borehole data ........................................ 40

Effect of change in the thickness of the sand layer in a two-layer,
stratified sand over red clay model ........................................................ 41

Results of the 30 and 90 meter CST transect as well as the topography
of the transect along the CST 5 line ....................................................... 42

Results of the 25 and 75 meter CST transect as well as the topography
of the transect along the CST 1 line ....................................................... 43

Results of the 30 and 90 meter CST transect as well as the topography
along CST 6 line .................................................................................... 43

Results of the 30 and 90 meter CST transect as well as the topography
along the CST 2 line .............................................................................. 45

Results of the 30 and 90 meter CST transect as well as the topography
along the CST 4 line .............................................................................. 45

Results of the 30 and 90 meter CST transect as well as the topography
along the CST 3 line .............................................................................. 46

The measured apparent resistivity pseudosection at the top, the
calculated apparent resistivity pseudosection in the middle, and an
inverted resistivity section after 2 iterations in the bottom image ......... 48

The measured apparent resistivity pseudosection at the top, the
calculated apparent resistivity pseudosection in the middle, and an
inverted resistivity section afier 2 iterations in the bottom image ......... 59

The measured apparent resistivity pseudosection at the top, the
calculated apparent resistivity pseudosection in the middle, and an

inverted resistivity section after 1 iteration in the bottom image .......... 50

The location and trend of the low resistivity anomaly is plotted on both
an aerial photo and a digital elevation model (DEM) ............................ 52

vii

I. INTRODUCTION AND LITERATURE REVIEW

1. a. Background and History of Glacial Deformation

Glacial theory was first introduced by Jean de Charpentier in the mid 18303 and
the concept of former extension of glaciers and ice sheets was furthered by Louis Agassiz
(Teller 1983). Not until Charles Lyell (1863) did anyone suggest that glaciers could
deform rocks and sediment, which he observed in deformed strata at Norfolk, England,
the Italian Alps, and in other areas. Lyell suggested that there are three possible
mechanisms for deformation. These include pushing by stranding icebergs, melting of
buried ice masses, and pushing before advancing glacier ice (Lyell 1863). Today,
modern glacial theory is based on a triad of effects created by glaciation, which includes

erosional, deformational, and depositional evidence (figure 1).

   
     

DEPOSITION EROSION

  

ANCIENT
GLACIATION

 

DEFORMATION

Figure 1. Triad of effects created by glaciation, on which modern glacial theory is based (Aber et a1.
1989)

The first person to use the term glacial tectonics was (Slater 1926), which is now

shortened to glaciotectonics in America and glaciatectonics in Britain (Aber et al. 1989).

Glaciotectonics became a very popular area of research in The Netherlands during the
19505 due to the ice-shoved ridges being the most dominant topographic features in an
otherwise flat country. During this time period, glaciotectonics also became a popular
area of research in Poland (Jahn 1956; Dylik 1961; Galon 1961) and Denmark
(Berthelsen 1973; Berthelsen 1978; Petersen 1978).

Little glaciotectonic research was performed in North America until the latter part
of the 19508, due in part to the relatively low number of glacial scientists and the
significantly larger and less developed area to study (Aber et al. 1989). Beginning in the
19503, large-scale topographic maps and aerial photographs became widely available,
which allowed for much easier research of such a large area. The large area on which to
focus North American glacial research is a result of the Laurentide Ice Sheet, which is
approximately twice the size of the Fennoscandian/British Ice Sheet coverage (Flint
1971; Aber et al. 1989). After years of county or state-scale glaciotectonic
investigations, the first attempt at continent-scale synthesis concerning ice-sheet
dynamics, distribution of glacial landforms, and genesis of glaciotectonic phenomena in
the entire glaciated Great Plains region of the United States and Canada was published in

Moran et al. (1980).

I. b. Glaciotectonic deformation structures

Since a glacier or ice-sheet can deform substratum by both forward movement
(dynamic) and vertical loading (static) and both typically occur simultaneously, the
effects of each cannot usually be separated (Aber et al. 1989). Due to this fact, the
resulting structures are considered joint manifestations of secondary deformations

produced during glaciation. Such secondary structures within the substratum are termed

exodiamict (Aber et al. 1989). Aber et a1. (1989) defines glaciotectonism as glacially
induced structural deformation of bedrock and/or masses of glacial drift, as a direct
result of glacier-ice movement or loading. Glacial drift refers to any sediment which had
been deposited by glacial ice.Aber et al. (1989) considers glaciotectonics to exclude
deformed structures within glacier ice, structures classified as glaciodynamic,
glaciokarstic, iceberg drifting, and other crustal structures not created by active glacier
ice. Therefore, identification of glaciotectonic structures and landforms is based on the
presence of recognizable masses of pre-existing bedrock (or any pre-Quaternary rock or
sediment material against or over which the glacier moved) and/or drift and the presence
of glacially induced deformations within those masses (Aber et al. 1989). This
deformation, in the form of folds, faults, breccia, slickensides, or other disturbances may
be produced either in situ, or during the transportation and deposition of a detached mass
(Aber et al. 1989). Glaciotectonic features represent situations where glacier ice
deformed pre-existing strata without completely removing or destroying the rock or

sediment beyond recognition (Aber et al. 1989).

 

 

 

 

 

 

 

 

 

 

 

 

 

Horizontal Scale
1000km 100km 10km 1 km 100m 10m 1m
More .
Brittle Composite Small Faults
A Thrusts
Moraines & Ridges Fractures
Glaciotectonic
Dia irs & Intrusions
V Arc & Belts p
More CNStal Small Folds
Ductile Depression

 

 

Figure 2. Chart of common glaciotectonic structures arranged according to their typical horizontal
scales (logarithmic). Ductile structures toward bottom; brittle structures toward top (Occhietti I973;
Aber et al. 1989).

Glaciotectonic structures have a very large range in scale, from microscopic to
continental size structures. Most commonly, deformed materials include sedimentary
strata which are unconsolidated to poorly or moderately consolidated. Less commonly,
well consolidated sedimentary strata and crystalline rocks are deformed by glaciotectonic
processes (Kupsch 1955; Babcock et al. 1978). A comprehensive list of all
glaciotectonic structures is not possible due to the variation in type, style and size.
Figure 2 from Aber et al. (1989) based on Occhietti (1973) list several common
glaciotectonic structures. These structures can be separated into two broad categories,
ductile and brittle, the classification of which depends on the nature of deformation (Aber
et al. 1989).

Ductile deformation is caused by internal flow of material in a fluid-like or plastic
manner. This type of deformation represents essentially no internal strength in the rock
mass, so that substantial changes in the size or shape of the mass are a result of very
small pressure differences (Aber et al. 1989). Ductile glaciotectonic deformation is most
common in unconsolidated or fine-grained strata such as clay or silt, which deform under
high confining pressures (Aber et al. 1989). Folds, intrusions, and diapirs often represent
ductile deformation.

Brittle deformation is a result of a rock mass failing by fracturing along a plane,
and the resulting movement or adjustment along the fracture plane, while the internal
fabric of the rock is not permanently deformed. These structures are characteristic of
consolidated or coarse-grained strata such as sand, gravel, sandstone or limestone, and
are deformed under low confining pressure (Aber et al. 1989). Joints, faults, breccia,

fissures and other fractured structures are examples of brittle deformation.

Both ductile and brittle structures can be found in the same sequence of
glaciotectonically deformed strata due to changes in lithology and thickness of the layers.
Layers of differing lithology and thickness respond to deforming pressure differently,
resulting in some layers developing ductile deformation, while others exhibit brittle
structures (Aber et al. 1989). Finally, it is also possible to observe both brittle and ductile
deformation in the same rock or strata due to variations in pressure, temperature, or fluid
content during deformation. Glaciotectonic structures ofien mimic those seen in igneous
and metamorphic rocks. Croot (1987) acknowledges that, "Ice-shoved hills may be

regarded as natural scale-models of mountains; the only significant difference is size."

 

.. .... TWO POINTS 'OF

"n u-' VIEW

0
a a a O o as .

 

 

 

Figure 3. Cartoon illustrating how large intrusive structures in the subsurface may have no
morphologic expression at the surface (Aber et al. 1989)

One typical structure observed in glaciotectonic sequences is the intrusive
structure. Intrusive structures include any structure in which one material has been
mobilized, then squeezed or injected into the body of another material. Typical intrusive

materials are clay- or silt-rich sediments, while the host material is typically some type of

unconsolidated sediment or possibly even soft bedrock. Intrusive structures may be in
the form of diapirs, intrusions and wedges, and may range in lateral size from only a few
cm to > 100m (figure 2) and may vertically extend up to 103 of meters (Aber et al. 1989).
Intrusive structures do not necessarily result in distinctive landforms and topography, but
may in fact be quite remarkable in cross-sectional view (figure 3). The injected sediment
behaves as a fluid due to pore water trapped under high pressure, and the host material
behaves in a less ductile or even brittle manner (Aber et al. 1989). Such intrusions
generally form in subglacial, water-saturated conditions with intergranular movement as
the main mechanism of deformation (Brodzikowski et al. 1985).

Intrusive structures can be further classified into two groups, originating from
either below or above the host material. Structures originating below the host material
include diapirs, stocks, plugs, dikes, and sills, and are similar to igneous intrusions of low
viscosity magma or diapirs formed by salt mobilization. Diapirs form as a result of fine-
grained sediments being compacted and mobilized by high pressure water. This sediment
then behaves as a fluid and migrates toward zones of lower pressure, until a density
equilibrium is achieved or the intrusive sediment supply is exhausted (Aber et al. 1989).

Structures with the intrusive sediment originating from above the host material
generally penetrate along cracks in the brittle host material. These structures include
wedges, veins, and fissures. Wedge sediments are typically homogeneous till to
laminated sand, silt or clay, and may be oriented from vertical to nearly horizontal. The
strike of the structure may be related to the direction of ice movement, or a subglacial
pressure gradient, or merely zones of weakness within the host material (Aber et al.

1989). Aber (1989) gives several case examples of diapir and wedge structures that are

glaciotectonic in origin. These examples are located in the Kansas Drift, Kansas; Herdla
Morains, Norway; and Systofte, Denmark (Aber et al. 1989). In each of these cases, clay
or silt-rich sediment is injected as a result of glacier overriding, and none of them result
in any surface expression.

Another important quality of glaciotectonic deformation is the distribution of such
phenomena. When first recognized, glaciotectonic features were considered unusual or
rare, yet today are thought to be widespread in the outer portions of glaciated regions
(Sauer 1978). Aber and Lundqvist (1988) provide us with a highly generalized model for
continent-scale distribution of glaciotectonic phenomena, which includes three primary

ZOIICSZ

1. Inner zone — zone in which widespread, small- and moderate-sized
glaciotectonic features are developed in older drifi.

2. Intermediate zone — zone where small, isolated glaciotectonic features are
found mainly in locally thick drift of the last glaciation.

3. Outer zone — zone in which all manner of large and small glaciotectonic
phenomena are present in drift and soft sedimentary bedrock both onshore and

offshore.

The boundaries of these zones are quite sharp in some cases and more transitional in
others. This three-zone model can be best demonstrated for the last major glaciation with
the Laurentide Ice Sheet in North America (figure 4) and the Fennoscandian Ice Sheet in

northern Europe (Aber et al. 1989).

 

 

 

 

 

 

 

 

Figure 4. Landscape zones of the Laurentide Ice Sheet: l= zone of extensive end moraines,
hummocky moraines and ice-shoved hills; 2= zone with long eskers and ice-flow
lineaments; 3= zone of extensive ribbed moraine, drumlins, and eskers (Dyke et al. 1987a).

Aber er al. (1989) also considers the regional distribution of glaciotectonic
features for the outer zone, where their abundance and pattern of distribution can been
compared with other glacial phenomena. Table 1 shows the factors that are considered
important for genesis of glaciotectonic phenomena. Aber et al. (1989) recognized two

regional distribution patterns for glaciotectonic features:

1. Random, sporadic distribution of megablocks, rafts, diapirs, and other features
that have little or no morphologic expressions, along with small cupola-hills.

These features were presumably created in subglacial settings far behind ice

margins. Their locations are primarily related to conditions of substratum
materials.

2. Ice-marginal distribution of morphologically prominent hill-hole pairs,
composite ridges, and large cupola-hills. These features were created at or
near active ice margins and their locations are closely related to the

development of ice lobes or tongues.

Table 1. Factors considered important for genesis of
glaciotectonic phenomena (Aber et al. 1989).

 

Subglacial Proglacial

1. Lateral pressure gradient S P
2. Elevated ground-water pressure S P
3. Ice advance over permafrost S P
4. Ice advance against topographic obstacle S P
5. Lithologic boundaries in substratum S P
6. Surging of ice lobes S P
7. Subglacial melt-water erosion S P
8. Damming of proglacial lakes P
9. Thrusting 1n front of ice P
10. Compressive flow with basal drag S

l 1. Shearing fault blocks up into ice S

While documenting the distribution of patterns for glaciotectonic phenomena
across the Great Plains of western Canada and the north-central United States, Moran et
al. (1980) recognized three features with which glaciotectonic features are often found in
conjunction. These features include bedrock escarpments, subsurface aquifers, and
identifiable ice-margin positions. Moran et al. (1980) and Bluemle and Clayton (1984)
interpreted this pattern as simultaneous creation of streamlined and ice-thrust features

beneath and behind a stationary ice margin.

I. c. Characterization of Glaciotectonic Deformation structures
I. c. i. Traditional Methods

Traditional methods for characterizing glaciotectonic deformation structures
include outcrop and aerial mapping, as well as borehole analysis, borehole logging, and
core penetration tests. Outcrop mapping is extremely common and provides excellent
two dimensional information on grain size, stratigraphy, and exposed structures. The
limitations of outcrop methods lie in the availability or unavailability of exposures.
Many times there are no adequate outcroppings of glaciotectonic deformation structures,
or the outcrops have been overgrown as noted by Bakker (2004).

Aerial mapping is another common method to characterize glaciotectonic
deformation structures. Aerial mapping can also include digital elevation models
(DEMs) in order to more clearly identify topographic features. In central Brookhaven,
Connecticut, Tvelia (2007) uses DEMs to map and characterize glacial deformation

Stl'llCtlll'CS.

 

Figure 5. Glacial features of central Brookhaven, CT (Tvelia 2007).

In some areas, there may be no available outcrops, and no apparent surface
expressions of glacially deformed structures in the subsurface. For these situations it is
often useful to gather borehole information. Boreholes can provide stratigraphic, grain
size, density, water content, and other information that can be used alone, or in addition
to other data to deduce various interfaces or to confirm interpretations. One useful
application is continuous core which preserves the stratigraphy and the fabric of the
layers, which can allow for grain-size analysis and gathering of stratigraphic data.
Borehole logging uses more technical methods which gather information including water
content, density, and gamma radiation which is an indicator of organic content. Finally,
an Electric Cone Penetration Test (CPT) can be used for shallow exploration. The CPT
provides information on the mechanical composition of the shallow subsurface. A CPT
survey involves the penetration of a metal cone with a surface of 10cm2 into the

subsurface. A truck inserts the cone at a constant rate while a number of different

11

parameters are measured, including cone resistance and sleeve friction. Borehole
methods are quite useful for gathering relatively high-resolution data in one dimension.
Unfortunately, many boreholes are required to gather information in two or three
dimensions, and the intrusiveness of drilling a borehole can be a limiting factor.
Borehole analysis, logging, and the CPT are all used by Bakker (2004) in order to
provide information on stratigraphic layers and reflection interfaces while performing

ground penetrating radar surveys.

I. c. ii. Geophysical Methods

Ground-penetrating radar is a very common, and well suited, method of imaging
and characterizing glaciotectonic structures. GPR allows for adequate penetration and
high enough resolution to characterize even small faults and folds in glacial sediments.
Recent efforts have been made to characterize glaciotectonic features in areas where there
are no exposed outcrops. Various studies have employed the use of ground-penetrating
radar in order to investigate glaciotectonic structures (Hansen et al. 1997; Tvelia 2007).
For example, 100, 200, and 500 MHz antennae have been used to examine deformation
structures in a proglacial environment in central Brookhaven, Suffolk County (Tvelia
2007). Both the 100 and 200 MHz data sets indicate large-scale glaciotectonic
deformation structures in the form of fault and fold structures, while results from the 500
MHz data indicate many faults and folds, but are highly affected by the presence of
cobbles in the glacial till (Tvelia 2007). The 500 MHz data does penetrate deep enough

to locate the bottom of the till layer.

Distance (meters)
135 140 145 150 155 160 165

 

Figure 6. GPR image showing a till layer boundary and hyperbolas interpreted as cobbles (Tvelia
2007).

Ground penetrating radar was also used to characterize the glaciotectonic
deformation structures in the Falsterselv area, Jameson Land, East Greenland (Hansen et
al. 1997). This study acknowledges that the 50 MHz antenna is used for maximum
penetration (10-1 5 m) with low vertical resolution (2-2.5 m), while the 200 MHz antenna
is used for lower penetration (2-2.5 m) with high vertical resolution (0.5 m) (Hansen et
al. 1997). Ground penetrating radar was chosen in order to image large-scale
sedimentary and glaciotectonic structures. Hansen, Jensen et al. (1997) indicate that
glaciotectonic structures and sandy sediments overlain by till are discemable to a depth of
about 10 m (figure 6) provided that the clay content of the sediment is low. This study
was able to reach their expected penetration in the coarse-grained sections, but reached a

shallower depth in the fine-grained areas (Hansen et al. 1997).

13

NW SE

\'A$—.QF' !% “in T. ‘ I" .

Q

‘O.’ '_ _"l se_..\ -'1 s
' . ‘ y. ‘ .4, 0.1.? - N. we. ‘_“’“:

6O cisterns-w .* '
6 . ‘ . , . ' v1 -- '-.'
m ”at?” 9;?" wade":"'WJ-nw ‘
Elf}- ‘ ‘1‘ t ‘ ’ ~ ‘ ,AI

- V... ‘ ' ' . ., n IA
“ “a“ (“4" A" '“Innuumamh'Pw 4’ ”u
"fi-‘ 0 ‘ ' ~‘\.- the ‘fi‘ 'imlwf‘h-W-"P‘J.5J
s . . _ . ‘ ‘ , ‘ ‘ 1 . «~Aw' . > i I

    

50

 

 

Figure 7. 50 MHz GPR data with the interpretation below. Evidence of glaciotectonic thrusting is
interpreted in the glaciofluviaI-Iacustrine sediments (yellow) (Hansen et al. 1997).

Another common geophysical method for characterizing glacially deformed
sediments is a seismic survey. A seismic survey is useful in recording the reflections of
seismic waves from bedrock, and from the internal velocity and density variations within
sediments. A simple, common seismic source can be multiple sledgeharnmer blows, but
can include explosives or even vibrating trucks. The reflection data is then recorded
using a series of geophones. Seismic surveys gather data which is similar to GPR data,
and can provide quite detailed information on the stratigraphy and structure of the
subsurface.

Both GPR and seismic surveys gather data based on reflection or refraction of
waves at an interface or reflector. In this study area, the focus is on a large deformed clay

layer. This study was not concerned with smaller scale structures or faults, which might

l4

have been observed in GPR or seismic data, and is not likely to be observed in electrical
resistivity data.

Unfortunately, at this site, the deformation structures that we are looking for are
often more than 10 meters deep. Also, since they are composed of very fine grain
material, GPR would most likely not result in any data below the top of the deformed

clay layer.

1. d. Field Site Description and Objective

This study focuses on an area known as the Ludington Ridge, near the town of
Ludington, Michigan. The Ludington Ridge is an elongated dome extending NW—SE,
and is almost 11 km long and up to 6 km wide (Larson et al. 2003). Due to the locally
severe disruption of sediments, this project investigated areas along the western edge of
the Ludington Ridge which is characterized by a steep cliff dropping up to 70 meters to

Lake Michigan, as well as numerous washout gullies.

 

Figure 8. A cross-section showing the sediments exposed in the western end of the Ludington Ridge
(Larson et al. 2003). The distance between horizontal tick marks is 100 m. The vertical scale
indicates height above Lake Michigan.

Figure 8 shows the exposed edge of the Ludington Ridge, directly beneath the
field area in this study. Larson et al., (2003) identify these sediments as the lower

stratified sand, which contains massive to laminated mud. The lower stratified sand is 7

15

to more than 35 meters thick and consists of horizontally bedded, planar to trough cross-
bedded medium to coarse sand (Larson et al. 2003). The massive to laminated mud form
a cluster of four diapirs which are shown in figure 8. These diapirs are derived from a
bed of massive to faintly stratified mud within the lower stratified sand and can rise more
than 40 meters in height. These diapirs are caused by the weight of the overlying ice
creating a large vertical pressure gradient, which fluidizes the clay in the stratified sand
which then migrates up to zones of lower pressure. The result of the migration is an
undulating clay layer of which the most dramatic portion is shown in figure 8 and is the
focus of this study.

Throughout the Ludington ridge, springs and gullies on the surface indicate a
preferred pattern of groundwater drainage and past landslides. However, currently it is
not known whether the deformation structures are isolated or continuous features, and if
the latter, what is their orientation. Such information may help further understand the
glacial history of the area. Little is known about the local stratigraphy near the
deformation structures, as the public borehole information is sparse. The overall objective
of this study is to characterize the inland extent of these deformation structures using
geophysical methods, and to improve understanding of the glacial history of the
Ludington Ridge, specifically related to glaciotectonic processes.

Figure 9 provides an aerial view of the field area with locations of survey
transects and other data collection points noted, while table 2 provides more detailed

information on specific measurements.

16

\ = Constant Spread Traverse
\ - Haiti-electrode Survey

Q =- Vertical Electrode
Sounding

. = Borehole Information

‘ = Diapir Outcrop:

f-

Bradshaw Rd.

CST 4

Washout -
Gullv
‘9

Chauvez Rd.

’

A

 

.8

Figure 9. Aerial photo of the field site showing all transects, data collection locations, and outcrops.

l7

Table 2. Measurement details for surveys shown in figure 7.

 

Measurement Array Array Smallest A- Largest A- Step Size Starting Location
Name Direction Length (mL gaming (m) Spacing (m) (m) (Lat/Long)

CST I N 975 25 75 45 N43.907 W86.446
CST 2 N 405 30 9O 45 N43.913 W86.445
CST 3 N 120 30 9O 45 N43.909 W86.444
CST 4 N 315 30 9O 45* N43.909 W86.443
CST 5 N 1710 30 9O 45 N43.906 W86.443
CST 6 N 2700 30 90 45 N43.913 W86.434
M 1 N 498 6 166 n/a N43.910 W86.445
M 2 S 1356 6 166 n/a N43.921 W86.444
M 3 E 888 6 166 n/a N43.9l4 W86.446
S 1 N-S 284.7 0.3 94.9 Log N43.911 W86.446
S 2 N-S 284.7 3 94.9 Log N43.906 W86.443
S 3 N-S 284.7 3 94.9 Log N43.921 W86.444
S 4 N-S 284.7 3 94.9 Log N43.913 W86.436

 

*15 m between 675-780m from Chauvez

I. e. Hypotheses
11. e. i. Due to the high contrast in electrical properties between the stratified sand and
the clay in the Ludington Ridge, electrical resistivity is a useful geophysical
method to gather information on the subsurface deformation.
0 1D sounding surveys can be used to characterize the subsurface stratigraphy.
o Constant spread traverses will be an effective tool used to locate
glaciotectonic structures inland from the cliff face.
11. e. ii. The deformation structures in the Ludington Ridge are elongated diapirs which
trend to the NE as sub-parallel ridges, as proposed by Larson et al. (2003).
0 Geophysical tools will be used to determine if the deformation structures are
elongated diapirs, or individual intrusions, and what the orientation of any

elongated structures is.

18

11. e. iii. Because of the low borehole density in the Ludington Ridge, electrical
resistivity field surveys, coupled with lab measurements and forward modeling,
will allow for interpolation of the stratigraphy between boreholes.

11. e. iv. Multi-electrode surveys will allow for imaging the spatial distribution of

deformation features at higher resolution. Major topographic features at the
surface of the Ludington Ridge are expressions of the subsurface deformation

structures.

0 Topographic features and geophysical anomalies will be correlated using GIS

software.

19

II. METHODOLOGY

II. a. Background

The defining characteristic of a material includes that materials resistance to the
flow of electrical current. Sediments and other geological materials are no different. A
very useful, non-invasive technique for identifying subsurface materials is to perform an
electrical resistivity survey. Electrical resistivity methods have been in use since the
early 19003, but have become much more widely used since the 19703 and the

development of computer assisted data gathering and analysis (Reynolds 1997).

The equation for resistance is given in Reynolds (1997) as:

R=V/I (with the units expressed in (0); (Equation 1)
where:

V= voltage (volts)
I= current (amps)

When measuring electrical resistivity, a geometric factor must also be taken into account.
This depends on the geometry of the array, and the electrode spacings. The equation for
the geometric factor is:

1 .
K = 27: 1 l 1 1 , (Equation 2)

— — +
AM AN BM BN

 

 

where;

20

AM, AN, BM, BN = Distance between respective electrodes

There are a number of possible configurations of the electrodes for electrical
resistivity surveys, but the one we will discuss in this section will be the Wenner array
(Wenner 1912a; Wenner 1912b). The Wenner array is quite simple in that all of the
electrodes are exactly the same distance apart. Since, for the Wenner array all of the

electrodes have the same spacing, the result is a fairly simple geometric factor (equation
4) and resulting apparent resistivity (pa) equation (equation 5) which has been adapted

from Reynolds (1997). Since, in the Wenner array, all the electrode spacings, or a-

spacings, are equal, we can simplify this equation to:

 

K = 27: ; (Equation 3)

where;
a = a—spacing
This equation then simplifies even further to:

K=2na; (Equation 4)

The geometric factor, K, is then multiplied to the resistance, R, resulting in the equation

for apparent electrical resistivity using a Wenner arrray:

pa: 21taR; (Equation 5)

2l

Reynolds (1997) also provides a table of resistivity values for common geologic
materials. For the sedimentary materials encountered in this study, as well as similar
materials, an amended table is included.

Table 3. Resistivity values for various materials (Reynolds 1997).

 

 

 

 

 

Material Nominal resistivity (9m)
Soil (40% clay) 8
Soil (20% clay) 33
Top soil 250-1700
Boulder clay 15-35
Clay (very dry) 50-150
Mercia mudstone 20-60
Coal measures clay 50
Middle coal measures >100
Chalk 50-150
Coke .2-8
Gravel (dry) 1400
Gravel (saturated) 100
Quatemary/Recent sands 50-100
Sand clay/clayey sand 30-215
Sand and gravel 30-225

 

DC electrical resistivity methods measure the ability, or inability, of electrons to flow
through sediments. In the case of the Ludington Ridge they exploit the large contrast in

electrical conductivity between the two dominant texture types (sand and clay).

II. b. Electrical resistivity surveys

For the surveys discussed in this thesis, there were three types of array
deployment. The first type, known as a vertical electrical sounding (VES), is used for
depth sounding, allowing us to determine the vertical variations in resistivity. As with all
types of electrical resistivity surveys, the depth of penetration increases with an increase

in electrode spacing. Also, with an increase in electrode spacing, a larger volume of the

22

subsurface is contributing to the apparent resistivity. With the VES a horizontal, layered
earth is assumed. In addition, inversion and interpretations use a horizontal, layered earth
model. The other type used is known as a constant separation traverse which is used to
traverse across the horizontal variations. The final method employed in this study is the
multi-electrode survey, which allows for computer controlled data collection of many
points over a relatively large area. This multi-electrode array may also be, “rolled-

along,” to gather cover transects longer than the available cable sections.

11. b. i. Vertical electrical sounding (VES)

From Reynolds (1997) we know that as the distance between the electrodes
increases, so increases the depth to which the current is able to penetrate. The data
gathered from a given array is obtained from a position at some depth, located at the
midpoint of the array. While performing the VES, we begin with the electrodes close
together in order to gather information from the shallow subsurface. We then expand the
array, maintaining the geometry of a Wenner array, with the spacing between the
electrodes increasing logarithmically. A logarithmic rate of expansion is used because
when plotting the results on a logarithmic scale, the data points are equally spaced. The
values obtained are then plotted on a graph with electrode spacing as the independent
variable and the apparent resistivity as the dependent variable. Sources of error include
inaccurate topographic information leading to inaccurate data inversions. The
stratigraphic information is gathered from well log data, which was likely interpreted by
individuals whose focus was on the hydrologic properties, not the electrical properties of

the sediment. In addition, we do not have stratigraphic information for each point along

23

our transects, so there is little control over the data inversion in some areas. Error in the
inversion results can be reduced by improving constraints using laboratory measurements

and additional borehole information.

11. b. ii. Constant spread traverse

The other method used in this study is known as the constant spread traverse.
This is performed by manually installing a Wenner array, obtaining a single data point,
and then moving the entire array along a profile. For example, in this field study we
gathered data with both a 90m separation and a 30m separation, and progressed along the
profile at 45m increments. These values were then plotted on a linear graph as a function
of distance along the profile. Variations in the apparent resistivity values indicate
anomalies along the traverse (Reynolds 1997). Sources of error are most likely to arise in
the calculation of the apparent resistivity values. That value is calculated based on the
geometric factor, which is a relatively simple calculation for the Wenner array type.
However, having exact a-spacings and a perfectly linear array is nearly impossible;
therefore corrections to the geometric factor must be considered. For our situation, we
were able to maintain a reasonably straight transect, never exceeding more than 10
degrees of offset for any electrode. Assuming the worst geometric error of 10 degrees
between any two electrodes, the maximum error in the apparent resistivity calculation is

less than 0.6%, which is quite acceptable.

24

 

 

 

 

 

\

 

p = Ja[(cos 0')"2 + (sin 0')"2]

Equation 7. Error analysis for two possible array errors with resulting geometric factor.

II. b. iii. Multi-electrode survey

When using a multi-core cable, an automated earth resistivity meter, and the
accompanying software, VES measurements can be automated in order to gather larger
amounts of data with less physical demands (Barker 1981). This multi-core cable, and
the accompanying setup can also allow us to gather automated CST data (Griffiths et al.
1990)

Another advantage of using a multi-core cable is the ability to perform a multi-
electrode survey. A multi-electrode survey is limited in the number of electrodes and the
maximum electrode spacing by the length of the multi-core cable, as well as the number
of electrical take-outs and their spacing. For this study, the six multi-core cables had 14

take-outs each with a maximum spacing of six meters. Therefore, we were able to deploy

25

84 electrodes with a total survey length of 498 meters. The instruction manual that
accompanies the earth resistivity meter provides the optimal measurement settings for a
multi-electrode survey. While this survey is relatively more labor intensive than the CST
to set up, a computer-generated command file controls the acquisition of hundreds of data
points with no human interaction. The measurement unit controls the combinations of
the current and potential electrodes in order to gather a very high-resolution data set
spanning the length of the survey line getting progressively deeper toward the midpoint
of the line. Additionally, if the transect is longer than the cables will allow, a roll-along
survey can be performed. In this method, the measurement unit is told to, “Roll-along,” a
section of cable. In order to do this, the first cable section is disconnected and moved to
the end of the line in the forward direction. Errors primarily arise if the array is not
completely straight, if topographic information is not precise or accurate, and during the
data inversion process. Careful array setup and survey information can reduce the effect
of the lateral variations on the apparent resistivity values. High-resolution topographic
information can be obtained, and when coupled with better constraints on resistivity and
stratigraphic information, will allow for more accurate data inversion and forward

modeling.

II. c. Other Methods

In order to improve the interpretation of the electrical resistivity data, I used a suite of
laboratory tests which aided in the creation of accurate geologic models and in the
interpolation of the subsurface where borehole information was unavailable. From
(Larson et al. 2003) we know that this specific section of the Ludington ridge is primarily

two distinct layers, sand over clay. There is a thin (~1 m) gray clay layer found

26

intermittently, as well as a thin (~1 m) layer of eolian sand near the cliff, but inland these
layers appear to be non-existent. This project aims to determine the locations and
thicknesses of the two primary layers inland, away from their exposure in the cliff face.
In order to best interpret the sounding, CST, and multi-electrode data, an accurate
geologic model must be created. Since the resistivity of a sedimentary unit is a function
of the material, as well as the water content, we use laboratory measurements to constrain

that information.

II. c. i. Soil box method

The soil box test was performed following the procedure described in (ASTM-
GS7-95). Initially, representative samples were gathered for each of the four sedimentary
layers. Each of these samples was oven dried and placed in a soil resistivity box. A soil
box is a small box in which sediment is placed to determine the resistivity of the sample.
There is a plate at each end of the rectangular box which acts as the current electrode, and
there are two pins toward the middle section of the box which serve as the potential
electrodes. For each sample, volumetric water content vs. resistivity curve was developed
by gradually increasing the volumetric water content from 5% to 25% and monitoring the
decrease in apparent resistivity. Error may occur in the laboratory soil box measurement
by losing a small volume of sediment when wetting, drying, or transporting the sediment

in the soil box.

II. c. ii. Grain-size analysis
Now that the range of resistivity values is known for each sample, we needed to

further constrain the sediments in the sample. This is accomplished by sieve analysis for

27

the sand-size fraction, and a hydrometer test for the silt- and clay-size fractions. These
analyses were performed according to procedures described by Liu et al.( 2003). For the
sieve analysis, the sample was thoroughly dried, mixed, and separated into fourths using
a sediment divider. Next, the sample is placed in a stack of sieves ranging from #5 to
#230, fining downward. This stack is placed on a mechanical shake table and agitated for
15 minutes. The amount retained in each sieve is then weighed and compared with the
total weight to determine the percent retained by each sieve. For the silt-and clay-size
fractions, 50 grams of the sample must be mixed with 125 mL of deflocculant and
allowed to soak. A hydrometer reading is determined by the specific gravity of the
solution. Since the specific gravity is dependent on temperature and is affected by the
deflocculant, a composite correction factor must be determined. The composite
correction is the sum of the specific gravity correction (Fz), the meniscus correction (FM),
and the temperature correction (FT). The specific gravity correction is determined by
plotting the hydrometer readings for the upper and lower range of temperatures, and
assuming a linear relationship between them. The meniscus correction accounts for the
difference between the actual level of the liquid and the meniscus formed on the
hydrometer. Finally, the temperature correction can be determined by the following
equation:

FT = -4.85 + 0.25T (for T between 15° and 28°); (Equation 8)

Once the composite correction is determined, the deflocculated solution is placed in a
1000 mL sedimentation cylinder which is filled to 1000 mL by adding distilled water. At

this point the hydrometer measurements are taken at very specific time intervals over a 24

28

hour period. Upon completion the percent passing for each hydrometer reading can be

found by

P = 5M6 x 100 ; (Equation 9)

P = percentage of soil remaining in suspension at time reading was taken
M = Mass of dry soil (g)
a = Correction factor for specific gravity
Assume specific gravity = 2.7.
R = Corrected hydrometer reading = reading- composite correction

The corresponding diameter for each hydrometer reading is

D = K J; ; (Equation 10)

D = Diameter (mm) corresponding to the percent passing found in the first step

K = Constant depending on specific gravity of particles and temperature of
suspension (Table 9-3 Text, ASTM standard)

L = Distance from top of mixture to the level where the density of the mixture is
being measured (cm), can be found from Table 9-4 in book or ASTM

standard.
T = Time elapsed since start of test (min)
At this point, the laboratory tests have allowed us to determine the range of resistivity
values as well as the specific grain size distribution for each sediment sample. The most
common source of error is the loss of sample as dust, or the inability to completely
remove the entire sample fiom the sieve. Error in the hydrometer analysis occurs when

the temperature fluctuates from 20°C, as well as when the hydrometer is inserted into the

sedimentation cylinder and the sampling in unavoidably disturbed. A

29

II. c. iii. Soil moisture curve

Using the grain-size distribution data, coupled with a computer program used to
determine soil moisture content for specific grain size distributions, we can determine the
moisture content for each sample under gravitational drainage pressure (Schaap et al.
2001). This allows us to know the precise amount of moisture we can expect for each of
our sediments. That amount of moisture can be compared with the volumetric water
content vs. resistivity curve to determine a precise value for the electrical resistivity of

each sample under field drainage conditions.

11. c. iv. Inversion and Forward modeling

At this point, we can input our known VES data into a modeling program such as
DCINV which will develop a possible model for the subsurface by combining the field
sounding data and holding our laboratory determined resistivity values constant, and
allowing the thickness of each layer to change (Pirttijarvi 2001). We can also forward
mode] by creating a reasonable geologic model and observe the resulting resistivity

curve.

30

III. RESULTS

Grain-size distribution curves (figure 10) indicate a very high contrast between
the sand and clay layers. Both the stratified sand and the eolian sand are 99.6% and
95.4% sand-size particles respectively. The gray clay and red clay are in fact silt loams
with 74.3% and 69.9% silt respectively, and both have 0% sand. The stratigraphy shown
in figure 8 shows the stratified sand and the thick, red clay, as well as thinner clay layers,
which are identified as the gray clay. The eolian sand is not shown in figure 8 but occurs
in a thin layer along the beach and along the top of the cliff. Each of the samples was
collected from the cliff face at approximately the same latitude as the S 1 sounding

location (figure 9).

31

Particle Size (mm)

 

 

 

 

 

 

 

10 1 0.1 0.01 0.001
100 +4—— . . . -,-____ _ LEE; ,_,_,_1
90 ~
80 - _. \
70 ~ \
A 60 d \
.\°
or 50 -
.E = g a 3
3 4o - +Eolian
If Sand
30 _ +Stratified
Sand
20 _ .--Gray Clay
10 1 +Red Clay
I _
0 AL I 2 I ’ j ' ' * 4 4r e c e e A

Figure 10. Grain size distribution curves for each of the four distinct layers in the Ludington ridge.

The grain-size distribution data was then used to determine the residual water
content under gravitational drainage. The computer program, “Rosetta,” was used to
determine the residual water content, or 0,, by entering the percentage of sand, silt, and
clay particles (Schaap et al. 2001). The percentages, along with the resulting residual
water content, Theta_r, and saturated water content, Theta_s, values are shown in table 4.

In addition, each sediment sample was placed in a soil box resistivity unit and the
resistance of the sample was measured as the volumetric water content was increased.
Figure 11 shows the exponential decay of the resistance curve as the volumetric water

content increases from approximately 5% to just over 20%

32

2500

 

75 2000 ~
E
.C
3 1500 ~
Q)
0
C
:3 1000 -«
.2
8
a: 500 —
0

 

     
    
 
  

y = 4.5860x"~8027

y = 2.1968x'1-8384

y = 98.462x'1'1097

 

 

. Eolian Sand

I Stratified Sand

A Gray Clay
0 Red Clay

 

 

 

y = 90.076x"677

 

 

l

0 0.05

Volumetric Water Content (cm‘3lcm‘3)

0.1

0.2

Figure 11. Results of the soil box resistivity test.

Table 4. Grain size distribution data and Rosetta modeling results.

Eolian Sand Stratified Sand Gray Clay Red Clay

 

 

% Sand 99.606 95.424
°/o Silt 0.000 1.101
% Clay 1 0.394 3.475
Theta_r 0.052 0.055
Theta_s 0.376 0.373

0.000
74.333
25.667

0.086

0.474

0.000
69.984
30.016

0.090

0.481

0.25

Using the data from Rosetta (Schaap et al. 2001), we then use the results from the

soil box resistivity test (ASTM-G57-95) to determine the resistivity value for the specific

material under specific water content. The resistance curves have been changed slightly

due to the fact that the data point for approximately 5% volumetric water content has

been disregarded for the two clay samples. The error in this measurement was quite high

as a result of the inability to properly disperse the moisture throughout the sample. The

33

equations of the best fit lines have been adjusted to reflect this change, and the residual

moisture content has been noted for each sample (figure 12).

 

  
 
   
 

 

 

 

 

 

 

 

 

3000

2500 . y = 98.462x’1'1097
g o Eolian Sand
.5 2000 — u ---------- » , I Stratified Sand
E1500 - A Gray Clay
5 14606 . Red Clay
E: 1000 ,y z 8 8712* _____________________________ XTheta_r values
to -
a" y = 9 . ‘

500 -

y = 2-41498‘79‘M‘fi5
0 . . . ~

 

 

0 0.05 0.1 0.15 0.2 0.25
Volumetric Water Content (cm‘3/cmA3)

Figure 12. Soil box resistivity curves with residual moisture contents indicated.
Next, using the equation of the best fit line, we input the residual water content to

determine precise values for the resistivity of each material (table 5).

Table 5. Calculated values of resistivity using the equation for the best fit
line from the soil box resistivity curves.

 

Eolian Stratified Gray Red
Theta_r 0.052 0.055 0.086 0.090
Resistivity 670.989 2446.057 196.540 295.842
Theta_s 0.376 0.373 0.474 0.481
Resistivity 174.791 293.783 9.202 25.670

After determining the apparent resistivity values for each layer under gravitational

drainage, we input those values into a forward modeling program called, “DCINV,”

(Pirttijarvi 2001). This program develops a geologic model, containing parameters we

34

can selectively hold constant at known values, or allow the computer to vary, that would
yield similar VES data to the data collected in the field. The results near the intersection
of Iris Rd. with S. Lakeshore Rd. were compared with well-log data from a residential
water well. The results of this modeling are shown in Figures 13, 14, 15, and 16.
Figure 13 is the results from data collected at the intersection of Iris Rd. and S.
Lakeshore Rd (figure 9). The resistivity value used for the top layer is 88.9 Ohmm
which was allowed to change in the model and reflects a reasonable value for the gray
clay. The value for the second layer is the resistivity calculated using the residual
moisture content for the stratified sand, and that value was held constant at 2446.0
Ohmm. The third layer was also allowed to change and the resulting value of 25.7
Ohmm is very close to the calculated resistivity for the red clay using the saturated
moisture content. The RMS-error value is 0.066, which indicates a very good correlation

between the calculated values with the measured values.

35

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A
10 3 3 Wenner Model description
: 9 Com Resist (Ohmm) Thickness (m)
t O 1. 88.9 4.3
E - 2. 2446.0 24.9
E . 3. 25.7 --
g RMS-error: 0.066197
b. .
E 0
1:3 0 o
‘93 IOAO
ii .
h i-
*5 : r
2 . :Depth (m)
3 - E
2‘ * 10A]
10A] I I I I I I l I I I f T I I I I ' ' ' :
Distance (“1) Resrst. (Ohmm)

Figure 13. Results of the DCINV modeling to determine the stratigraphy near the intersection of Iris
Rd. and S. Lakeshore Rd.

Figure 14 is the results from data collected at the intersection of Chauvez Rd. and S.
Lakeshore Rd (figure 9). The resistivity value used for the top layer is 138.3 Ohmm
which was allowed to change in the model and also reflects a reasonable value for the
gray clay. The value for the second layer is the resistivity calculated using the residual
moisture content for the stratified sand, and that value was held constant at 2446.0
Ohmm. The third layer was also held constant and at a value of 25.7 Ohmm which is the
calculated resistivity for the red clay using the saturated moisture content. The RMS-
error is 0.056, which again shows a very good correlation between the calculated values

and the measured values.

36

 

 

10"3 e nun-O\C Model description

0 Measured Resist. (Ohmm) Thickness

  

 

llllll

Computed
(m)
= 1. 138.3 4.2
q 2. 2446.1 39.1
0 3. 25.7 --

- RMS-error: 0.055809

 

 

 

 

 

1000

[1111

 

 

Depth (m)

IIIIIIT I I

1001

Apparent resistivity (Ohmm)

 

WIII

 

 

 

 

 

 

 

 

 

10A] 1 I I TI j T I I I I I I I I I I ' '
IOAO 10A] IOAZ 10A] .IOA2 10A3
Distance (“1) Resrst. (Ohmm)

Figure 14. Results of the DCINV modeling to determine the stratigraphy near the intersection of
Chauvez Rd. and S. Lakeshore Rd.

Figure 15 is the results from data collected on the beach half way between the
two southern diapirs (figure 8). The resistivity value used for the top layer is 671.0
Ohmm which is the calculated resistivity value for eolian sand using a residual water
content of 5.2%. The model suggests this layer is quite thin, which was the case as
observed in the field. The resistivity value for the second layer was allowed to change
with the model and was calculated to be 177.3 Ohmm which is a reasonable value for the
red clay since it is on the shoreline at or below lake level. The best model results were
obtained when a third layer below the red clay was introduced. This layer resulted in a
value of 46.7 Ohmm, which I interpret is an increase in the moisture content of the red
clay. Again, the RMS-error is .041, which indicates a very good correlation between the

calculated values and the measured values.

37

 

 

1003

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I Wenner Model description
‘ 0 Measured Resist. (Ohmm) Thickness
" Computed (m)
A j 1. 671.0 0.2070
El 2. 177.3 7.6
f}, ‘ 3.46.7 --
9 q i O RMS-error: 0.040807
3? NEW
.’>
'3 \
g, - lOA-l
L. ‘ O '
+1 : :
5 o 5
E ‘ Q g 1000
3‘, i
<1: Depth
‘ (m)
[ : 10"]
10M . .......i . r ......i . ...n , I
lOA-l 10"0 1001 10"2 10,? 10,3
Distance (111) Resist. (Ohmm)

Figure 15. Results of the DCINV modeling to determine the stratigraphy on the beach between the
two southern most diapirs.

Figure 16 is the results from data collected on the Haul Rd. (S 4), 720 meters
north of Chauvez Rd. (figure 9). The resistivity value used for the top layer is 53.8
Ohmm which was determined by the model to be a very thin layer and reflects values
similar to that of the clays. The resistivity value for the second layer was allowed to
change with the model and was calculated to be 965.5 Ohmm which is a reasonable value
for the stratified sand with significant water content. The best model results were
obtained when a third layer below the stratified sand was introduced. This layer resulted

in a value of 47.5 Ohmm, which I interpret as the red clay layer. The RMS-error is .062

38

which again shows a very good correlation between the calculated values and the

measured values.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1003 _ . .
. Model description
: Resist. (Ohmm) Thickness
_ (m)
A g I. 53.8 0.6885
El 2. 965.5 32.1
E ' 0 3.47.5 --
9 ‘ RMS-error: 0.061986
b.
3
>
.3
.E’. -
n .
a, . _
" C
+’ I :
a ‘ 5
m —
h -r
a _
o. . .
Do _
4 2
1 0A] I I I I l I TT I I I I I l I I I I I E
10"0 10"] IO"2 1002 10"3

Distance (In)

Resist. (Ohmm)

Figure 16. Results of the DCINV modeling to determine the stratigraphy on the Haul Rd

north of Chauvez Rd.

lOA-l

Depth
(111)
1 0’0

10"]

. 720m

Additional information on the stratigraphy in the Ludington Ridge was gathered

from public borehole information. Figure 17 shows the borehole stratigraphy from a

series of four wells that were drilled along Iris Rd.

39

 

Location of Iris Rd.
Sounding

V

 

 

 

‘—»--—._._.— fl,

 

 

West - East
¢——.

 

.: - Gravel

 

 

:25: Coarse
=Sand

 

 

 

 

 

 

--------- '* 35333535532333: Med.

 

 

1%; 1...; gagsgssisizizizisia ' Sand

 

 

 

 

 

 

 

 

 

Figure 17. Stratigraphic information from borehole data. For locations see figure 9.

The same program used to invert our VES data was also used to investigate the
effect changing the depth to a clay layer would be on VES data. In order to forward
model the results of a change in the depth to the clay layer, a simple two-layer model was
used. The resistivity value for the top layer is 2446.1 Qm, which is that calculated for the
stratified sand with residual moisture content. The resistivity value for the second layer
is 295.8 Qm, which is that calculated for the red clay with residual moisture content.
Figure 18 shows the expected results of the VES when the depth to the clay layer is

changed from 1 to 80 meters.

40

1 0000

 

 

 

 

 

 

 

 

 

E t
g l —1 m Thick Stratified Sand
5 ’ —2 m Thick Stratified Sand
.5 5 m Thick Stratified Sand
g 1000 ~: —10 m Thick Stratified Sand
0: ’ —20 m Thick Stratified Sand
'5 ---40 m Thick Stratified Sand
3 -—80 m Thick Stratified Sand
2

100 m 1 1 4 1 1 1 1 ; 1 r 1 4 1 x m

1 10 100

Wenner Array A-Spacing (m)

Figure 18. Effect of change in the thickness of the sand layer in a two-layer, stratified sand over red
clay model.

In addition to allowing us to interpret the stratigraphy in areas with poor or absent
borehole information, we can also use the DCINV models to interpret our multi-electrode
data. Unfortunately, due to the relative labor intensity of a large, multi-electrode, roll-
along survey, we needed to first determine where the best localities to image a
deformation structure would be. As discussed in the previous chapter, the CST was used
to gather low resolution data in a very efficient manner which allowed us to cover large
distances with relative case. In figure 8 the lines labeled, CST I, CST 5, and CST 6 were
performed in order to determine if there was reason to gather high resolution, multi-
electrode data. Figure 19, 20, and 21 are the results of those transects. Data was
gathered using the Wenner array at 30 and 90 meter spacing, 25 and 75 meters for the

beach. The 90 and 75 meter spacing allowed us to notice any deep variation in the depth

4]

beach. The 90 and 75 meter spacing allowed us to notice any deep variation in the depth
to the clay layer. The 30 and 25 meter spacing provided us with much shallower data and

indicated whether or not the clay layer had risen significantly toward the surface.

 

 

 

 

 

 

 

 

3000 270
+90 meter
+
2500 . . 30meter _____________ ii 260
+Transect
.. Topography
g 2000‘ -250 ‘g
3 E
" 3
E1500 ~24o ,.
.2. o
35- ‘2
3 i 3
E 1000 y r 230 2
I:
e ‘V. 1 t z
a . ca
a '5
g 5004 § . .. 220 I
v
o . ' 4- 210
-500 I I I I I I m I 2m
0 200 400 600 800 1000 1200 1400 1600 1800

Distance from Chauvez Rd. intersection (m)

Figure 19. Results of the 30 and 90 meter CST transect as well as the topography of the transect
along the CST 5 line.

42

8

 

CD
0

N
O

O)
O

01
O

 

 

+75 meter
20” """ """ ' """" """ " ' """ +25meter"

Apparent Resistivity (Ohmm)

 

 

 

 

 

O l l T T
0 200 400 600 800 1000

Distance starting 300 meters South of the Gully Drainage (m)

 

Figure 20 . Results of the 25 and 75 meter CST transect as well as the topography of the transect
along the CST 1 line. Topography along the beach was flat.

 

 

 

 

 

 

 

 

 

 

1400 250
31200» - ------------------ ~~ -+9°m°‘e' mi 240 g
E +30meter :
2100M """"" ' +Topography 230 a
a . -'
E 300i ~- 220 3
'3 (I)
i 600i ~- 210 3
m ‘ "l. i g
u <
E 400i " v, i 200;
a l ' 9

O
f;- 2001 . 4 190 z
I “1
0 r I r I I "180
500 1000 1500 2000 2500 3000 3500 4000

Distance from Chauvez Rd. (m)

Figure 21. Results of the 30 and 90 meter CST transect as well as the topography along CST 6 line.

43

The CST data gathered along Haul Rd. (CST 6) (figure 21) shows a distinct drop
in apparent resistivity values as we move north, away from Chauvez Rd. The topography
dropped steadily toward the Pere Marquette River to the north. Along this transect we
noticed numerous springs which indicated that the clay layer was very near the surface,
and water was flowing along the layer and reaching the surface. This is shown in the data
starting near 3000 meters. This is likely not a glaciotectonic structure. The evidence for
a possible structure lies in the CST transect along S. Lakeshore Rd. (Figure 19). The
results show only a single decrease in apparent resistivity values which occur near the
intersection of Bradshaw Rd. with S. Lakeshore Rd. If, in fact, the four observed diapirs
observed in the cliff face extend as sub-parallel ridges to the NE, we would expect to see
them in the data gathered along S. Lakeshore Rd. (CST 5). Near the 700 meter mark we
see a noticeable drop in apparent resistivity values. In order to investigate this possible
anomaly further, we proceeded to gather three additional CST data sets were gathered 50
east from the cliff edge, 200 meters east from the cliff edge (CST 2), roughly halfway
between the cliff and the S. Lakeshore Rd. (CST 3), and a final line 50 meters west from
S. Lakeshore Rd. (CST 4), making sure to intersect any elongated structure that would lie

between the Bradshaw Rd. - S. Lakeshore Rd. intersection, and the cliff edge (figure 9).

44

 

 

 

 

 

 

3500 265
EM" ”260?
E :
925004 «255;
a 3
E2000~ 4503
.3 a:
815001 +2452
a: 3
§1000- 1-2405
g f.
2500‘ ~235§
O r n 230

 

 

 

§

800 900 1000 1100 1200 1300 1400 1500 1600
Distance from Chauvez Rd. (m)

Figure 22. Results of the 30 and 90 meter CST transect as well as the topography along the CST 2
line.

1200 260

 

§
CD

—~ 250

800 ~ . -— 240
+ 90 meter

1, +30 meter

 

 

 

 

Apparent Resistivity (Ohmm)
8
O

 

 

 

/ ------- -— 230

+Topograph
400 - 7 , -— 220
200 ~ -~ 210
0 I l l 1 fl 1 I l 200

400 450 500 550 600 650 700 750 800 850
Distance from Chauvez Rd. (m)

Figure 23. Results of the 30 and 90 meter CST transect as well as the topography along the CST 4
line.

45

Height Above Sea Level (m)

 

   

 

 

 

 

 

 

1600 250

1400‘ ‘r 245
g 1200« *240 E
5‘2. E
E1000 - .. 235 3
5 8
.5 800 ’230 '2
.2 35
E 600 ‘ + 225 <
2 z
a .9
2 400 .- —-«-O—90 meter 220 :g

+30meter
200 ‘ *************** via—Topography 215
0 I V l l I l I 210

 

450 500 550 600 650 700 750 800 850
Distance from Chauvez Rd. (m)

Figure 24. Results of the 30 and 90 meter CST transect as well as the topography along the CST 3
line.

We can see from figure 22 that the apparent resistivity values remain much
higher than those in figures 23 and 24, which fall below 100 Ohmm. Also, according to
this data, we see the lowest values along the CST 4 transect occurring between 700 and
750 meters, while the lowest values along the CST 3 line occur between 650 and 700
meters. As the transects are located closer to the cliff edge, the low-resistivity anomaly is
detected further to the south.

Finally, we have developed reasonable geologic models for areas in the Ludington
ridge. We have also used the constant spread traverse to locate the most likely locations
for the multi—electrode survey to image a low-resistivity anomaly. The next step is to use
the models, along with topographic information, and input these data, along with the
results of the multi-electrode, roll-along surveys, in Earthlmager 2D. This program can
then invert the field data and develop a model which will yield similar results as seen in

the field. Figures 25, 26, and 27 are the results from the multi-electrode surveys that

46

were performed in areas that were determined to most likely intersect a large-scale
glaciotectonic deformation structure. A first multi-electrode survey was completed
approximately 50 meters east of the cliff which extended 498 meters north from the
northern edge of the washout gully (M 1) (Figure 25). This survey yielded only one low-
resistivity anomaly.

We can see in the inverted resistivity section (Figure 25) that this survey was able
to image a single deformation structure. Additionally, a multi-electrode, roll-along survey
was completed along S. Lakeshore Rd.(M2) along the same transect as CST 5, and again
indicated only a single low-apparent resistivity anomaly near the intersection of

Bradshaw Rd. with S. Lakeshore Rd. (Figure 26).

47

dud—E E325 2: 5 39.580: N cuts 558.». @3338.— ..832: :a can 6:5...— 2: E 5338—58:

33530.. 2.9.25." toga—=23 2: £3 2: an 3338—503 33:38.. 2.9.235 59:53:. 2: .mn 959.,—

ooe:_
5-220

0:

   

,3.

afl

2255

Sennaseieoz $312.2 “1:822. sexegsxuflpé

 

 

 

 

 

 

_
Fe 3+ m3.

=oavoac§§m .9533. .3523. 3.5.5.5

_
«Mm 1w... 3H ma.— ..3 00 5 c

 

 

 

 

 

 

55882.8; .95an 52.254. “352.62

 

 

 

 

 

 

(m) untimely

(m) mdea

(m) qidea

48

dun—E 882x. 2: 3 2.33.3: N not: .3398 @3569. eotoZ: as was 6.2::— 2: 3

5333—52.... @3533. Eouanaa 6833.3 2: £8 2: «s 533330....— b_>_unm8._ Bahama 69:33:. 2:. .3 953%

 

awn
0G
n3.
mag m
v.34. a

51:30

G. .0“
to 1
man. 1

m—fi?

5.5:. u

E b u macaw ova coo—m .m. on n ma sown in u wZM n u sonic: :ocoom bfiiom C85»:—

 

 

 

 

 

 

 

_ _ q
03.n— nm: MSG— ...ow On... .Orr. mm». EH 3.— C

:ozueaoEEma .CScflmoM 50.8224. tea—BED

 

 

 

 

 

 

C

. .4 _ _ a 4 A
can. mm: was «new cup 05. u»... an 3.—

:ozooa2252m bran—mom 292x54. 53:28:

 

 

 

 

 

 

V@

I «1.0

I Mn.
1

I pH

_ _ . _ . _ _ 1 q a
0.....N— Hméq max: ”1.0% On.“ Our. Mm». MM“ 3— C

(111) 110921913

([11) lpdeq

(m) Indeq

49

.0225 53::— 05 :_ 539.0: — not: .5393 3.55.39. ear—9:: :a as: 6.2::— a...
5 5:933:02. brim??— Eouaama noun—=23 2: £3 a... «a nervous—.8; 3538.. 2.9.2:? 69598:. 2:. SN PEEK

o ._ E c n «Eu—Em ucobuflm ww mm n mg .33 Lu. M £41 _ H £2.50: 5.53% £3233. c382:
N I!

 

 

 

 

coo:— _ . A _ _ _ _
27:35. ace. Sm On- CFO 0:. av... 00m ct“ cm: co

 

C

#533025...“ bran—mom 25897.. rfihgfifib

 

 

 

 

Eiéiiéjii.

 

 

 

 

 

 

 

5.5.5 .30. c _ m an... and :3. .7... 00». Gym CM: 00. o
:ozoongona bar—350M 50.2%; cub—nau— ‘
n 9. 3
f
3: 1 mm
03. I Mn.
r
X Z I «n
N d N": 4 4 d d d d d d 4 d c
SEE—n . 25. 2 _ m any. and 03. 9.... 0.0». Op n 9.: ca c

 

(m) uonenra

(m) mdoo

(m) mdoq

50

A final multi-electrode, roll-along survey was perfomed along Bradshaw Rd.
extending from the edge of the cliff, to the intersections of Bradshaw Rd. with Inman Rd
(Figure 27)

Topographic data was gathered along each of the CST and multi-electrode transects. The
topographic data, coupled with both the CST and multi-electrode data indicate some
correlation between subsurface deformation structures and the surface topography.
However, not all of the topography is a result of subsurface deformation structures.
Figure 28 is a digital elevation model (DEM) of the field area, on which locations of
diapirs in outcrop, and low resistivity anomalies are noted. Results from both the CST
and multi-electrode survey indicated only one continuous low resistivity structure; the

remaining three were not imaged.

51

I

 

 

Figure 28. The location and trend of the low resistivity anomaly is plotted on both an aerial photo
and a digital elevation model (DEM).

52

IV. DISCUSSION

This project focused on four main hypotheses; 1) that electrical resistivity will
exploit the large contrast in electrical properties between the sand and clay in the
Ludington ridge, 2) that we will use this to support the idea that there are four sub-
parallel, northeast trending diapirs, 3) that electrical resistivity field surveys, lab work,
and forward modeling will allow us to interpolate the stratigraphy between boreholes, 4)
and that major topographic features at the surface are expressions of subsurface
deformation structures.

Hypothesis 1) states that due to the high contrast in electrical properties between
the stratified sand and the clay in the Ludington Ridge, electrical resistivity is a useful
geophysical method to gather information on the subsurface deformation. We were able
to determine the precise grain-size distribution for each sample, as well as determine the
most probable amount of natural water retention for each sample under gravitational
drainage. According to the soil box resistivity data (figure 11), we can see that the two
fine grained-sediments have much lower resistance values than the two sand dominated
samples. The forward modeling (figures 13, 14, 15, 16) was also able to provide
reasonable geologic models that were calculated to have very similar VES data to the
data we measured in the field. Figures 19-24 indicate that the CST method was
successful in covering large areas of the Ludington ridge, and was also able to locate a
low resistivity anomaly.

Hypothesis 2) states that the deformation structures in the Ludington Ridge are
elongated diapirs which trend to the NE as sub-parallel ridges. Unfortunately, we were

only able to trace a single deformation structure continuously from the cliff face,

53

approximately 500 meters inland. The remaining three structures were not able to be
imaged by our survey. Perhaps we might be able to image the other three structures by
changing out survey geometry, or expanding the electrode separation. Our electrode
separation was six meters, and by using 84 electrodes, our total line length was 498
meters. The maximum separation for a Wenner survey, as controlled by the SuperSting
computer, uses a 166 meter separation. This separation was able to provide data as deep
as 84 meters. As observed in the cliff face, the clay layer approaches an approximate
maximum depth of 20 meters at the highest point of each diapir, and a maximum depth of
over 60 meters at the deepest part of the troughs. This raises the possibility that all four
diapirs observed in the cliff face do not in fact continue inland toward the northeast as
sub-parallel ridges. The remaining three structures may be more localized similar to a
true diapir, or they may be ridge-like structures that plunge away from the cliff edge.

Hypothesis 3) states that because of the low borehole density in the Ludington
Ridge, electrical resistivity field surveys, coupled with lab measurements and forward
modeling, will allow for interpolation of the stratigraphy between boreholes. Coupling
the laboratory grain size and water retention data with resistance curves developed in the
laboratory, we were able to tightly constrain the input values used in the DCINV
modeling program. This program was used to develop reasonable geologic
interpretations which would provide us with similar resistivity sounding curves to the
data we collected at three sites.

Hypothesis 4) states that multi-electrode surveys will allow for imaging the
spatial distribution of deformation features at higher resolution. Major topographic

features at the surface of the Ludington Ridge are expressions of the subsurface

54

deformation structures. By examining the multi-electrode data sets, we see in each set
that there is a clearly defined, low-resistivity anomaly. When these surveys are plotted
on a digital elevation model (figure 28), with the location of the anomaly clearly marked,
we can see that not only do they indicate a single, northeast trending structure, but that

structure occurs along the northern edge of a northeast trending valley.

55

V. CONCLUSIONS

This study was able to exploit the large contrast in electrical properties between
the clay deformation structures surrounded by sand in the Ludington ridge. We were able
to image the deformation structures, and determine that we only observe a single,
continuous deformation structure extending from the edge of the cliff, inland
approximately 500 meters, trending toward the northeast. This is in accordance to Larson
et. a], (2003), who suggests that the deformation structures trend as four sub-parallel
ridges toward the northeast. Unfortunately, we were unable to confirm the continuation
of the remaining three structures beyond the cliff face.

In addition, the topography of the Ludington ridge is largely driven by the
subsurface deformation structures. Since the structures are made of clay, there is a
preferred pattern of groundwater flow in the Ludington ridge. This flow has created a
large valley extending from Lake Michigan, northeast toward the Pere Marquette River.
At the Lake Michigan mouth of the valley, there is a large washout gully which formed
as a result of years of preferential groundwater flow, coupled with episodes of intense
precipitation and flooding. Interestingly, the single deformation structure we were able to
image continuously, is located along the northern edge of this long valley.

Finally, laboratory methods were able to allow us to gather precise data on
electrical resistivity properties of each type of sediment, as well as water content and
grain size information. This allowed us to create a well constrained and reasonable
geologic model for each of the three areas where we performed a VES. The model was
checked with well data from the comer of Iris Rd. and S. Lakeshore Rd., and was used to

interpret data gathered at the intersection of Chauvez Rd. and S. Lakeshore Rd., as well

56

 

as on the beach between two large deformation structures, where there is no available

well information.

57

VI. REFERENCES

Aber, J. S. (1985a). "The character of glaciotectonism." Geolcgie en Mijnbouw 64: 389-
395.

Aber, J. S., D. T. Croot, et al. (1989). Glaciotectonic Landforms and Structures.
Dordrecht, Kluwer Acedemic Publishers.

ASTM-G57-95 A Standard Test Method for Field Measurement of Soil Resistivity Using
the Wenner Four-Electrode Method.

Babcock, E. A., M. M. F enton, et al. (1978). "Shear phenomena in ice-thrust gravels,
central Alberta." Canadian Journal Earth Sciences 15: 277-283.

Barker, R. D. (1981). "Offset system of electrical resistivity sounding and its use with a
multicore cable." Geophysical Prospecting 29(1): 128-143.

Berthelsen, A. (1973). "Weichselian ice advances and drift successions in Denmark."
Bulletin Geol. Instit. Univ. Uppsala n.s.(5): 21-29.

Berthelsen, A. (1978). "The methodology of keneto-stratigraphy as applied to glacial
geology." Geological Sociatv Denmark Bulletin 27(Special Issue): 25-38.

Bluemle, J. P. and L. Clayton (1984). "Large-scale glacial thrusting and related processes
in North Dakota." Boreas 13: 279-299.

 

Brodzikowski, K. and A. J. van Loon (1985). "Inventory of deformational structures as a
tool for unraveling the Quaternary geology of glaciated areas." Boreas 14: 175-
188.

Croot, D. G. (1987). "Glacio-tectonic structures: A mesoscale model of thin-skinned
thrust sheets?" Journal Structural Geology 9: 797-808.

Dylik, J. (1961). The Lodz region. VI INOUA Congress. Guidebook of exursion. Lodz.
Flint, R. F. (1971). Glacial and Quatemm geology. New York, J. Wiley and Sons.

Galon, R. (1961). North Poland, area of the last galciation. VI INOUA Congress.
Guidebook of excursion. Lodz.

 

Griffiths, D. H., J. Tumbull, et al. (1990). "Two-dimensional resistivity mapping with a
computer-controlled array." First Break 8(4): 121-129.

58

Hansen, L. A., J. B. Jensen, et al. (1997). "Pleistocene sedimentary record of the
Falsterselv area, Jameson Land, East Greenland." Geolgg/ of Greenland Survey
Bulletin 176: 84-88.

Jahn, A. (1956). Wyzyna Lubelska, rzezba i czwartorzed. (Geomorphology and
Quaternary history of Lublin Plateau.) Warszawa, Panstw. Wyd. Nauk.: 453.

Kupsch, W. O. (1955). "Drumlins with jointed boulders near Dollard, Saskatchewan."
Geological Society America Bulletin(66): 327-338.

Larson, G. J ., J. Ehlers, et al. (2003). "Large-scale glaciotectonic deformation in the Great
Lakes basin, USA-Canada." Boreas 32: 370-385.

 

Liu, C. and J. B. Evett (2003). Soil Properties: Testing. Measurement, and fialuation.
Upper Saddle River, Pearson Education.

Lyell, C. (1863). The geological evidences of the antiquity of man. London, John
Murray.

Moran, S. R. (1971). Glaciotectonic structures in drift. Till/a symposium. R. P.
Goldthwait. Columbus, Ohio State University Press: 127-148.

Moran, S. R., L. Clayton, et al. (1980). "Glacier-bed landforms of the Prairie region of
North America." Journal Glaciology 25: 457-476.

Occhietti, S. (1973). "Les structures et deformations engendrees par les glaciers--Essai de
mise au point." Revue Gegraphique de Montreal 27 : 356-3 80.

Petersen, K. S. (1978). Applications of glaciotectonic analysis in the geological mapping
of Denmark. Danmarks Geologiske Undersogelse. Arbog. 1977: 53-61.

Pirttijarvi, M. (2001). DCINV. University of Oulu.

Reynolds, J. M. (1997). An Introduction to @plied and Environmental Geophysics.
Chichester, John Wiley & Sons.

Sauer, E. K. (1978). "The engineering significance of glacier ice-thrusting." Canadian
Geotechnical Joumgl 15(457-472).

Schaap, M. G., F. J. Leij, et al. (2001). "ROSETTA: a computer program for estimating
soil hydraulic parameters with hierarchical pedotransfer functions." Journal of

Hydrology 251: 163-176.

Slater, G. (1926). Glacial tectonics as reflected in disturbed drifi deposits. Geologists'
Association Proceedings.

59

Teller, J. T. (1983). Jean de Charpentier 1786-1855. Biobibliographical Studies. London,
Mansell Publ. Co. 7: 17-22.

Tvelia, S. (2007). Characterization of the Glaciotectonic Development of the Selden Hill
through Digital Elevation Models and Ground Penetrating Radar: 11.

Wenner, F. (1912a). "The four-terminal conductor and the Thompson Bridge." Q
Bureau of Standards Bulletin 8: 559-610.

Wenner, F. (1912b). "A method of measuring earth resistivity." US Bureau of Standards
Bulletin 12: 469-478.

Williams, G. D., P. J. Brabham, et al. (2001). "Late Devensian glaciotectonic deformation

at St Bees, Cumbria: a critical wedge model." Journ_al of the Geological Society,
London 158: 125-135.

60

 

MICHIGAN STATE

llllll ||

3 1293

NIVERSITY LIBRARIES

Ill lllll
0

3062 6430