THESIS I 1.1:;- ztaar
$133523?“ lHl”IllIIIIILIIIIIIIIIIIJIIIIIIIIQIIIIW

 

 

This is to certify that the

dissertation entitled
THE ORIGIN AND MORPHOGENETIC SIGNIFICANCE OF
PATTERNED GROUND IN THE SAGINAW LOVE.“ OF
MICHIGAN
presented by

David Paul Lusch

has been accepted towards fulfillment
of the requirements for

Ph . D . degree in Geography

W//é7:4¢¢Jo4fiI»am/[/L——~_.

Major professor

 

 

Date November 10, 1982

 

MS U i: an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

RETURNING MATERIALS:
lvlfSlflf Place in book drop to
LIB R remove this checkout from
5:2 your record. FINES will

 

 

 

be charged if book is
returned after the date
stamped below.

 

 

"}

Iggy 71;], 3 Leona

-_.,f«?ie§§;9°5

30x A 001. meme

5 m 1P?“

3.9

V

EriY'EVO .II

APR 67 1999

 

 

 

n.
5

.2

—-‘

”,4

.:1

'V
-
-1»J~

_L .

THE ORIGIN AND MDRPHOGENETIC SIGNIFICANCE OF PATTERNED GROUND

IN THE SAGINAW LOWLAND OF MICHIGAN

By

David Paul Lusch

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Geography

1982

rte
.. o. .I c
H w.
.l r.
— 0.”.
a...

.I}
I a
L ,. 3

fl G,

ABSTRACT

THE ORIGIN AND MORPHOGENETIC SIGNIFICANCE OF PATTERNED GROUND
IN THE SAGINAW LOWLAND OF MICHIGAN

By

David Paul Lusch

More than 350 mi2 of patterned ground have been delimited in the
Saginaw Lowland of east-central Michigan on the basis of the interpreta-
tion of aerial photography of varying scale, film type and date of ac-
quisition. This phenomenon consists of an ordered network of broad,
shallow troughs enclosing slightly higher centers. Individual forms
range from circular features (79 feet in diameter) to elongated cells
(1,289 x 480 feet). Most of the patterns, which can be classified as
nonsorted nets, formed in somewhat poorly drained loam or silt loam
drift on low-relief (§_15 ft/miz) surfaces which slepe less than 32.

The morphostratigraphic relationships of the Saginaw patterned
ground to the Port Huron Moraine and proglacial Lake warren I suggest
that the nets formed not earlier than 13,000 yrs B.P., but are probably
younger than 12,730 yrs B.P. The lower elevational limit of the nets
is coincident with the shoreline of proglacial Lake Elkton (Lundy) in
both the horizontal and isostatically uplifted terrain, indicating that
active pattern formation was contemporaneous with, and bounded by, Lake
Elkton. Early Lake Algonquin, the next proglacial lake in the sequence,
came into existence about 12,400 yrs B.P. and provides a minimum date for

the cessation of active pattern growth.

The Saginaw patterned ground is interpreted to be the result of
thermal contractionecracking and ice wedge development in periglacial
permafrost. An ice-wedge cast, underlying the mesh of one of the
patterns, provided supporting evidence for this conclusion. This
pseudomorph is composed of a well-sorted, medium sand surrounded by a
sandy clay loam to clay loam till. The abrupt boundary and marked
textural difference between these two sediments suggest that the sand
was deposited from above into a pre—existing wedge-shaped void in the
till.

Ice-wedge fossilization was rare in the Saginaw Lowland, however.
The somewhat poorly drained, fine-textured drift on which the patterns
occur must have been conducive to the formation of ice-rich permafrost.
Upon thaw, this material would have been prone to liquefaction, thereby
severely restricting the opportunity for the replacement of wedge ice
with allochthonous sediments.

The nonsorted netsin the study area provide geomorphic evidence
of at least local discontinuous permafrost conditions during a brief
interval (350-600 years) of the Late Wisconsinan Substage. This mor-
phogenetic conclusion implies a mean annual air temperature of -1° to
-6°C in the Saginaw Lowland of Michigan during the waning Port Huron

Stadial.

To Claudia, Peter and Ann

ii

ACKNOWLEDGMENTS

The author wishes to gratefully acknowledge the many hours of patient
discussion and skillful editing which Dr. Dieter Brunnschweiler provided
during the six-year evolution of this dissertation. Special thanks are
also extended to the members of my guidance committee, Dr. Harold A. Winters
and Dr. Jay R. Harman of the Department of Geography and Dr. Delbert L. Mbkma
of the Department of Crap and Soil Sciences. Their helpful advice was
greatly appreciated. Dr. Gary A. Manson, Chairperson, Department of Geo-
graphy, provided valuable logistical assistance which helped to ensure the
completion of the degree program.

A special acknowledgment must be made of the superlative efforts of
Miss Elizabeth Bartels, whose outstanding typing skills and attention to
editorial detail were invaluable during the preparation of this text.
Additionally, the support received from Mr. William R. Enslin, Manager,

Center for Remote Sensing, cannot be overlooked.

Finally, the essential contributions of my family to this endeavor must
be recognized. The support, encouragement and understanding of my wife,
Claudia, went far beyond the norm. Her acceptance of countless lonely evenings
and weekends cannot be repaid; her love will always be treasured. My children,
Peter and Ann, have had to sacrifice the fatherly attention they deserve.

Their smiles and hugs at the end of a long day helped tremendously.

iii

TABLE

LIST OF TABLES. . . . . .

LIST OF FIGURES . . . . .

LIST OF PLATES. . . . . .

CHAPTER I

CHAPTER II

CHAPTER III

INTRODUCTION
Background .

Statement of

Methods and Techniques .

OF CONTENTS

Problem

LATE WISCONSINAN GEOMDRPHOLOGY

Introduction

Moraine Boundary Mapping .

Topography of the Port Huron Moraine .

Composition of the Port Huron Moraine.

waterlain Moraines .

OF THE SAGINAW’LOWLAND

Proglacial Lake Sequence and Chronology.

Eolian Landforms

CHARACTERISTICS OF THE PATTERNED GROUND IN

LOWLAND. . .

11

ll

24

30

35

37

48

THE SAGINAW

Distribution and Extent of the Patterned Ground.

Classification of the Patterns .

Surface Mbrphometry of the Saginaw Nonsorted Nets. .

Edaphic Characteristics of the Nonsorted Nets.

iv

51

51

53

55

55

Chapter III (Cont'd.)

CHAPTER IV

CHAPTER V

The Topographic Setting. . . . . . . . . . . . . .
Age of the Nonsorted Nets. . . . . . . . . . . . . .
PROPOSED ORIGIN OF THE SAGINAW NONSORTED NETS. .

Review of Suggested Mechanisms Not Requiring Regolith
C raCking O I O O O O O O O O O O O O I I O O O O O O

Erosional Processes . . . . . . . . . . . . . .
Pedologic Processes . . . . . . . . . . . . .

Bedrock Control . . . . . . . . . . . . . . . .
Glacial Stagnation Processes. . . . . . . . . .

Review of Suggested Mechanisms Requiring Regolith
cracung I O O O O O O O O O O O O I O O O O O O O O

Thawing . . . . . . . . . . . . . . . . . . . .
Synaeresis. . . . . . . . . . . . . . . . . . .
Gilgai Development. . . . . . . . . . . . . . .
Partial Wetting . . . . . . . . . . . . . . . .
Desiccation . . . . . . . . . . . . . . . . . .
Dilation. . . . . . . . . . . . . . . . . . . .
Salt Cracking . . . . . . . . . . . . . . . . .
Seasonal Frost-Cracking . . . . . . . . . . . .
Permafrost-Cracking . . . . . . . . . . . . .

Evidence Supporting An Ice-Wedge Origin for the
Saginaw Nonsorted Nets . . . . . . . . . . . . . . .

Thermokarst Modification of the Saginaw Patterned
Gro ‘md O O O O O O O O O O O O O O O O O O O O O O O

PALEOENVIRONMENTAL SIGNIFICANCE OF THE PATTERNED
GROUND IN THE SAGINAW LOWLAND. . . . . . . . . . . .

IntrOduction O O O I O O O O O O O O O O O O O O O O .

76

81

87

87

87

88

89

89

9O

90

90

9O

91

91'

92

92

92

93

97

103

108

108

Chapter V

APPENDIX A

APPENDIX B

BIBLIOGRAPHY.

(Cont'd.)

Paleoclimatic Implications of the Saginaw Patterned
Gr Quad 0 O O O O O O C O O O O O O O O O O O I O O O

Geomorphic Evidence of Former Periglacial Conditions
and Permafrost From Elsewhere in the Great Lakes
Region I O O O O O O O O O O O O O O O O O O O O O O

Palynological Evidence of Late Wisconsinan Tundra
Environments in Michigan . . . . . . . . . . . . .

Regional Synopsis of Late Wisconsinan Tundra and
Permafrost Indicators. . . . . . . . . . . . . . . .

Summary and Conclusions. . . . . . . . . . . . . . .

Recommendations for Further Research . . . . . . . .

SOIL MANAGEMENT GROUPS . . . . . . . . . . . . . . .

ELECTRICAL RESISTIVITY SURVEYING . . . . . . . . . .

vi

110

112

119

124

127

130

133

136

140

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

Table 7.

Table 8.

Table 9.

LIST OF TABLES

Page
Predominant soil types associated with the waterlain
segments of the Port Huron Moraine in the Saginaw
LWland I I I I I I I I I I I I I I I I I I I I I I I I I 31
Predominant soil types associated with the subaerially-
emplaced segments of the Port Huron Moraine in the
sagmw Wland I I I I I I I I I I I I I I I I I I I I I 3 2
Radiocarbon ages of materials associated with Lake
Whittlesey or the Lake Arkona-Lake Whittlesey
t rans it ion I I I I I I I I I I I I I I I I I I I I I I 40
Radiocarbon ages of materials associatedeith Early Lake
Erie and Lake Iroquois . . . . . . . . . . . . . . . . . 45
Frequency of occurrence of nonsorted nets by soil manage-
ment groups (SMG) in the Saginaw Lowland . . . . . . . . 60
Frequency of patterned ground by local relief category . 79
Dominant topographic and pedologic conditions associated
with the patterned ground in the Saginaw Lowland . . . . 81
Reported sizes of active, inactive and fossil tundra
palygons I I I I I I I I I I I I I I I I I I I I I I I 9 8
Geomorphic indicators of former permafrost in the Great
Lakes Region I I I I I I I I I I I I I I I I I I I I I I 115

vii

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

10.

11.

12.

LIST OF FIGURES

Page
Extent and classification of the Port Huron Moraine
in the study area, as shown by Martin, 1955 . . . . 12
Revised extent and classification of the Port Huron
Moraine in the study area . . . . . . . . . . . . . . l4
Topographic profiles and soil types across the "Bay
City Moraine" in north-central Bay County . . . . . . 15
Topographic profiles and soil types across the "Bay
City Moraine" in Tuscola County . . . . . . . . . . . 16
Locations of the topographic profiles shown in Figures
3 and 4 I I I I I I I I I I I I I I I I I I I I I I I 17
Comparison between the revised delineation of the Port
Huron Mbraine and the boundaries shown by Martin,
1955I I I I I I I I I I I I I I I I I I I I I I I I I 18
Topographic profiles across the Port Huron Moraine in
western Tuscola County and eastern Saginaw County . . l9
Topographic profiles across the Port Huron Moraine in
northeastern Tuscola County . . . . . . . . . . . . . 22

Topographic profiles across the Port Huron Moraine in
Larkin and Midland townships, Midland County and Beaver
and Williams townships, Bay County. . . . . . . . . .

Portion of the Midland North 7.5-minute topographic
quadrangle showing the sand dunes along the distal
slope of the Port Huron Moraine in central Larkin
Township, Midland County. . . . . . . . . . . . . . .

Topographic profiles across the Port Huron Moraine in
Grim Township, Gladwin County and Gibson Township,
Bay county I I I I I I I I I I I I I I I I I I I I I I

Parent materials of the soils in Larkin and Midland
townships, eastern Midland County . . . . . . . . . .

viii

23

25

27

33

Figure

Figure
Figure
Figure
Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure-

Figure

Figure

Figure

13I

14.

15.

l6.

17.

18.

19I

20.

21.

22.

23.

24.

25.

26I

27.

28I

29.

30.

Page

An expanded topographic profile along the crest of the
Port Huron Moraine and the deformed water planes of
selected proglacial lakes in the Saginaw Lowland. . .

Lake Saginaw during the maximum Port Huron Stadial. .
Lake Warren I during the waning Port Huron Stadial. .
Lake Grassmere during the waning Port Huron Stadial .
Lake Elkton during the waning Port Huron Stadial. . .

Late Wisconsinan glaciolacustrine shorelines in the
study area. . . . . . . .

Distribution and extent of patterned ground in the
888 MW Mland I I I I I I I I I I I I I I I I I I I

Mesh of the nonsorted nets at the Lawndale site as in-
terpreted from.the airphoto in Plate III. . . . . . .

Delineation of nonsorted nets in the vicinity of the
Bridgeport wedge. . . . .

Relationship between soil management groups and the
distribution of the Saginaw nonsorted nets. . . . . .

Soil associatiOns and patterned ground distribution
in Bay County and eastern Midland County. . . . . . .

Site and situation of the Bridgeport wedge. . . . . .

Composite cumulative grain-size frequency curve showing
the range of particle sizes associated with both wedge-
infill and host sediments .

Cumulative grain-size distribution of the sediment from
two beach ridges associated with glacial Lake Grass-

mere I I I I I I I I I I I I I I I I I I I I I I I I I

Electrical resistivity profile showing the conductivity
anomaly associated with the subsurface continuation of
the Bridgeport wedge. . . . . . . . . . . . . . . . .

Site and situation of the Lawndale study area . . . .

Dominant soil profile textures associated with the mesh
of the Lawndale nonsorted net cell. . . . . . . . . .

Soil profiles of the three pit excavations at the Lawn-

dale site showing the stratigraphic positions of the
samples taken for particle size analyses. . . . . . .

ix

36
38
41
43

44

47

52

54

57

59’

61

63

67

69

72

73

74

75

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

Figure

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

BlI

BZI

BBI

Cumulative grain-size frequency curves showing the
contrasting partent materials at the Lawndale site. .

Electrical resistivity profile across the mesh zone
of the Lawndale net compared to subsurface textural

data I I I I I I I I I I I I I I I I I I I I I I I I I

Local relief of patterned terrain in the Saginaw
L ow1 and I I I I I I I I I I I I I I I I I I I I I I I

Spatial distribution of the Port Huron Moraine and
the nonsorted nets in the Saginaw Lowland . . . . .

Distribution of the Saginaw patterned ground and the
shorelines of Lake Elkton and Lake Warren . . . . . .

Thermal contraction-crack origin for ice wedges in
permafrost. . . . . . . . . . . . . . . . . . . . .

Relationship of ice wedges to ground patterning in
active permafrost . . . . . . . . . . . . . . . . . .

Suggested sequential development of thermokarst
topography in the Saginaw Lowland . . . . . . . . . .

Geomorphic indicators of permafrost in the Great Lakes
Region I I I I I I I I I I I I I I I I I I I I I I I I

Pollen diagram of the Cheboygan bryophyte bed and
associated sediments. . . . . . . . . . . . . . . . .

Time—distance diagram of ice-marginal fluctuations
along the axes of the three major glacial lobes in the
Great Lakes Region, showing the relationships between
the paleobotanical evidence of tundra and the geo-
morphic indicators of Late Wisconsinan permafrost . .

Schematic diagram of electrical resistivity apparatus

Linear electrode configurations commonly used in
resistivity surveying . . . . . . . . . . . . . . . .

Detection of near-surface resistivity anomaly . . .

78

8O

84

85

94

96

106

114

121

125

136

137

138

Plate

Plate

Plate

Plate

Plate

Plate

Plate

Plate

Plate

Plate

Plate

II.

III.

IV.

VI.

VII.

VIII.

IX.

XI.

LIST OF PLATES

High—altitude, color-infrared airphoto of the Vassar,
Michigan area showing the well-drained soils (white
to light cyan tones) associated with the Port Huron
Moraine. . . . . . . . . . . . . . . . . . . . . . .

High-altitude aerial photographs of the sand dunes
on the distal slope of the Port Huron Moraine in
eaa tern Midland county I I I I I I I I I I I I I I I

Color-infrared airphoto showing the network of none
sorted patterned ground at the Lawndale site . . . .

Variations in form, size and marking of nonsorted
flats in the sagmaw LWlando o o o o o o o o o o 0

View of the Bridgeport wedge exposure, looking south-

east I I I I I I I I I I I I I I I I I I I I I I I I

The Bridgeport exposure showing the irregularly
downward. tapering wedge o o o o o o o o o o o o o o 0

Lower portion of the Bridgeport wedge showing its
irregular, but abrupt perimeter and its basal width
exposed by groundwater piping away the sandy wedge-
infill material. . . . . . . . . . . . . . . . . . .
Bridgeport wedge exposed in the ditch wall and the
continuation of its truncated basal section across
the floor of the ditch . . . . . . . . . . . . . . .

Aerial view showing the lack of patterns below the
615 ft contour and conspicuous forms occurring only
above 620 ft (Lake Elkton shoreline) . . . . . . . .

Basal section of the Bridgeport wedge exposed in the
ditch floor.showing the orientation of two elongated
pebbles I I I I I I I I I I I I I I I I I I I I I I I

Conspicuous patterns confined to terrain above the
Elkton beach (620 ft) near Reese, Michigan; faint
nets (arrows) are associated with an Elkton offshore

barI I I I I I I I I I I I I I I I I I I I I I I I I

xi

Page

21

26

54

56

64

65

66

7O

82

101

104

CHAPTER I

INTRODUCTION

Background

Patterned ground is the general term.used to describe any regolith
surface which exhibits a discernible, more or less orderly and symmet-
rical micro-physiographic pattern (Brown and Kupsch, 1974). It was
originally proposed by Washburn (1950) as the only satisfactory expres-
sion in English which embodies the meanings of a number of German words
for the phenomenon, including Strukturboden (structured soil), Polygon-
bgdgg (polygonal soil), Zellenboden (cellular soil), Spaltennetze
(fracture nets), Netzrissboden (fracture-net soil), Wabenboden (honeycomb

ground) and Rautenboden (rhomboid soil). Patterned ground is character-

 

istic of, and best developed in, regions of intensive frost action (Flint,
1971; Embleton and King, 1975; French, 1976). It is not, however,
restricted to these morphoclimatic areas and, as noted by washburn (1956,
1970, 1973, 1980), may have a variety of origins; for instance, nonsorted
polygons resulting from thermal contraction-cracking of permafrost on the
arctic coastal plain of Alaska (Black, 1952) are nearly identical in size,
shape and marking to desiccation-crack polygons on some of the playa sur-
faces in the Mojave Desert of California (Chico, 1968). Correctly inter-
preting the genesis of patterned ground, therefore, requires careful
geomorphic analysis.

Patterned ground in Michigan was recognized more than a decade ago
(Brunnschweiler, 1969), but has since received little attention from.geo-
morphologists. Some of the patterns in southwestern Bay County and north-
central Saginaw County were studied by Tillema (1972) who concluded that

they were fossil ice-wedge polygons which formed in periglacial permafrost.

This study, however, is not fully convincing, both in the presentation of
the field evidence for ice-wedge casts and in the discussion of criteria
by which such relict structures can be identified (Johnsson, 1959; Black,
1976).

The location of the patterned ground in the glaciated terrain of the
Saginaw Lowland suggests either a glacial or periglacial origin for this
phenomenon. Although the Saginaw patterns are somewhat similar in appear-
ance to certain types of ice-stagnation terrain, their morphometry and
pedologic structure point toward a mode of origin determined by the
presence of ice wedges in permafrost. It is known that during the waxing
phase of the Wisconsinan glaciation, permafrost formed in front of the ex-
panding ice sheet and may have persisted beneath the glacier in localized
areas where subglacial temperatures were below 0°C (Pewe: 1973). As the
continental ice mass waned, recently uncovered drift became perennially
frozen and ice wedges formed in locales where especially rigorous condi-
tions existed.

Although the frost-debris tundra zone1 in the Midwest had a more re-
stricted latitudinal extent than elsewhere (Brunnschweiler, 1962; 1964),
the impact of the climatic conditions in this realm on Late Wisconsinan
morphogenesis cannot be discounted. Paleobotanical evidence suggestive
of this migrating tundra zone in Michigan is limited, but the overall
paucity of organic remains associated with the interstadials following
the Late Wisconsinan maximum (19,000-21,500 yrs B.P.) has been inter-
preted as an indication of cold, dry conditions which were too rigorous
to allow plant migration from the south (Goldthwait, 1968; Wayne, 1968;

Dreimanis, 1977).

 

1This term, translated from the German Frostschuttzone, was suggested by
Bfidel (1951) to differentiate the cryomorphologically more active plant-
1ess tundra zone frmm the forest tundra zone along its equatorward mar-
gins.

At issue is the existence of a morphoclimatic zone fringing the
waning continental ice sheet within which permafrost cracking and ice
wedge development occurred. In a recent review article, Black (1976)
critically evaluated numerous published reports of fossil ice-wedge
casts and patterned ground in temperate NOrth America. In most cases,
he found that the data presented in these studies were inconclusive or
misinterpreted and, as a result, an ice wedge origin for most of these
features was rejected. It was also speculated in this review that the
only favorable location for ice wedge development in the Great Lakes Re-
gion during the Late Wisconsinan was the large interlobate re-entrant in
western Wisconsin (the so-called "Driftless Area"). Of particular rele-
vance to the present study is a statement by the same author that ice-
wedge casts were not to be expected inside of the Late Wisconsinan ice
border in the United States, yet he cited two such occurrences (Black, 1965).

Indisputable evidence of Late Wisconsinan ice wedges and associated
patterned ground have been reported from southwestern Ontario (Morgan, 1972;
Greenhouse and Morgan, 1977). These features are especially germane because
their appearance and form are very similar to the patterns in the Saginaw

Lowland and they occupy a comparable morphostratigraphic position.

Statement of Problem
The Port Bruce glacial readvance in the eastern Great Lakes Region
culminated about 14,000 yrs B.P. and tundra conditions probably existed
in the periglacial zone at this time (Dreimanis, 1977). Within a rela-
tively short time thereafter, the ice margin began to retreat rapidly, yet
plant fossils from two sites of Mackinaw Interstadial age in Michigan docu-
ment tundra conditions during this interval (Miller and Benninghoff, 1969;

Burgis, 1970). The subsequent readvance during the Port Huron Stadial

deposited tills which differ in composition and fabric from those of the
preceeding Port Bruce Stadial, suggesting a rearrangement of the glacial
flow patterns following the Mackinaw Interstadial (Dreimanis, 1977). These
variations can be interpreted to indicate either a significant climatic
change or alterations of the thermal-dynamic regime of the Laurentide ice
sheet, but the fossil ice-wedge polygons in southwestern Ontario (Morgan,
1972; 1982) provide evidence of a substantial lowering of temperature.

This dissertation will focus on the origin and morphogenetic signifi-
cance of patterned ground in the Saginaw Lowland of Michigan. In parti-
cular, the hypothesis that these features are fossil ice-wedge polygons
resulting from.permafrost cracking will be tested. Acceptance of this
hypothesis is contingent on several factors. First, the morphometric
attributes of the Saginaw patterns must be comparable to the geometry,
mesh size and surficial marking of active, inactive or fossil icedwedge
polygons which have been reported elsewhere in North America and northern
Eurasia. Second, the edaphic characteristics of the patterned terrain in
the study area must be consistent with the textural and natural drainage
conditions which are known to promote permafrost and ice wedge develop- -
ment. Lastly, fossil ice-wedge casts must be present underlying the mesh
of the Saginaw patterns. Field criteria by which ice-wedge pseudomorphs
can be identified have been reported by Johnsson (1959) and Black (1976)
and will be utilized in this study.

Given the striking appearance of the Saginaw patterned ground and
considering the paucity of geomorphic documentation of periglacial perma-
frost in the Great Lakes Region during the ice-marginal fluctuations of
the Late Wisconsinan, the following research objectives were pursued and

are discussed in the following chapters:

1) Documentation of the distribution of the patterned terrain in
the Saginaw Lowland and analysis of its geomorphic and edaphic
character;

2) Description of the morphostratigraphic relationships between
the patterns and the known or re-interpreted glacial and glacio—
lacustrine landform.sequences in order to determine the time-
stratigraphic position of the patterns;

3) Investigation of the geomorphic agents and associated processes
which may have been active during pattern formation, to be ac-
complished through an analysis of data on the surface morphom-
etry and subsurface structure of the patterns, as well as a
critical evaluation of theories suggesting alternative origins
for such features; and

4) Discussion of the paleoenvironmental significance of the Saginaw
patterned ground, with special emphasis on the reconstruction of
the thermal regime during and immediately after ice retreat in
the study area.

As mentioned earlier, patterned ground is a general term for a
variety of geometrically arranged surface phenomena which may result from
numerous processes. Although this study will specifically test the hypo-
thesis that the patterns in the Saginaw Bay area resulted from.ice wedge
development in permafrost, other mechanisms which are reported in the
literature to form similar features will also be given full consideration.
Conclusions concerning the origin of the patterned ground in the Saginaw
Lowland of Michigan will allow their morphogenetic significance to be
assessed in both the local and regional context and should contribute to
a better understanding of the Late Wisconsinan environment in east-central

Michigan.

Methods and Techniques

The interpretation of aerial photographs was a major research tech-
nique used in this study. Mapping the distribution and extent of the
patterned ground in the Saginaw Lowland, as well as studying its morpho—
metric characteristics, is only feasible by analyzing airphotos because,
in spite of their sharp definition in the aerial view, these features are
virtually imperceptible to the ground-level observer. During the pre—
liminary phase of this investigation, inspection of various types and
scales of aerial imagery of the counties bordering Saginaw Bay revealed
that the patterns are well defined on even small-scale photography,
particularly if it was acquired in the spring when soil moisture varia-
tions in the barren fields accentuate the network mesh.

Small-scale, color infrared (CIR) imagery acquired by NASA in the
spring of 19752 was very useful for delimiting the areal extent of the
patterned ground, but detailed analysis of individual patterns was more
accurately accomplished from larger-scale photography, such as the
1:24,000 CIR photos from the Michigan Department of Natural Resources,
or the county coverage from the Agricultural Stabilization and Conserva-
tion Service (A.S.C.S.).

Since agricultural activities and other changes in land cover can
mask the patterned ground on airphotos, a multitemporal analysis was
conducted with blackvanddwhite panchromatic imagery acquired for the
A.S.C.S. in 1955, 1963, 1969 and 1970. Both individual 9" x 9" prints
and county photomosaics were used to map the areal limits of the patterns.
In total, aerial photography from six different acquisition dates spanning

23 years was interpreted, providing the highest possible accuracy for

 

2NASA-JSC Mission 309. All of the high-altitude imagery of Michigan ac-

quired by NASA's Manned Spaceflight Center and used in this study is
archived at the Center for Remote Sensing, Michigan State University.

visual pattern delimitation.

Areas of especially prominent patterns in the Saginaw Lowland were
delineated from l:120,000 CIR aerial photographs. Compared to other
available imagery of the study area, these small-scale photos revealed
the patterns remarkably well, probably because they were acquired the day
after the Saginaw Valley received an average of 0.32" of precipitation.

In addition, this photo mission provided total coverage of the study area
for the same day on a consistent film type and photo scale.

In order to recognize the morphostratigraphic context of the patterns,
and as a means of establishing their relative age, the boundaries of the
significant ice-marginal landforms and the locations of several proglacial
lake shorelines had to be determined. Previous delineations of these sur-
face formations were consulted (Leverett and Taylor, 1915; Martin, 1955),
but it was necessary to redefine some of these features. These re-
interpretations were accomplished by the combined analyses of aerial photo-
graphs, topographic maps and soil surveys coupled with field verification.

Special emphasis was placed on tapographic profiles constructed across
the previously delineated trends of the moraines in the study area because
such landforms are defined primarily on the basis of their constructional
topography (American Geological Institute, 1962). These profiles were com-
piled at a scale of l:24,000 from U.S. Geological Survey tapographic quad—
rangles which have a contour interval of 5 ft. Vertical exaggerations of
either 40 or 50 times were chosen according to the amount of local relief
present. The profile transects were extended several miles beyond the
previously mapped boundaries of the moraines in order to assess the ac-
curacy of these delimitations. Since the Port Huron drift is known to be
fine-textured (Leverett and Taylor, 1915), soil data aided the interpre-

tation of dune complexes, beach ridges and lacustrine bars and spits whose

topographic expression, in some instances, could be confused with morainic
terrain.

The shorelines of selected proglacial lakes were mapped by combining
the interpretations of aerial imagery with pedologic and topographic data
from existing maps. Many of these littoral features, both erosional and
depositional, are obvious on medium and small-scale aerial photography
as discontinuous, curvilinear trends in the landscape. These were mapped
on acetate photo-overlays and subsequently transferred to l/2":1 mile
county base maps. Shore features in the zone of horizontality for Lakes
Grassmere and Elkton were identified on the basis of their topographic
elevation (640 ft. and 620 ft., respectively); the beaches were mapped
across the uplifted zone north of the zero isobase by means of their
spatial continuity and relative position. Because the study area is north
of the Warren hinge line, all of the shorezone features associated with
Lake warren have been uplifted relative to its horizontal elevation of
680 ft. Nevertheless, the conspicuous shoreline along the proximal flank
of the Port Huron Moraine in Tuscola County has been accepted as the
Warren beach in previous studies (Leverett and Taylor, 1915; Martin, 1955)
and this served as the point of reference for the mapping of the remainder
of the Warren shore.

Although airphotos are used as a mapping base for modern soil surveys,
the patterned ground in the Saginaw Lowland has not been recognized in
these reports primarily because of the larger size of the soil mapping
units compared to the pattern mesh and due to the fact that the patterns
are not associated with a single soil series. Nevertheless, soil maps do
provide valuable information on the dominant soil texture and natural
drainage conditions characteristic of the patterned terrain. Other edaphic

data essential to this research, such as evidence of subsurface soil

structures or textural discontinuities underlying the surface patterns,
were gathered from on-site field examinations. Most of the arable land
in the study area is drained by road-side ditches and more than 100 miles
of these drains, many of which expose five or six feet of drift, were
inspected in the field. Sites where a pattern mesh intersected a ditch,
as located by airphoto interpretation, were given special scrutiny in an
effort to discern any sedimentological manifestations of the surface
pattern.

At the Bridgeport site, five paired samples of the material in a
soil wedge and the adjacent host regolith, each weighing about 200g, were
extracted from.the cleaned face of a drainage ditch exposure at depths of
0.2m, 0.55m, 0.8m, 1.1m and 1.4m. These sediments were subjected to par-
ticle size analyses using standard sieve and hydrometer techniques. As
a means of assessing the three-dimensional character of the soil wedge,
electrical resistivity surveys3 were conducted. This geophysical tech-
nique has been reported to be useful for mapping patterned ground
(Greenhouse and Morgan, 1977). The resistivity transects at the Bridge-
port site utilized a Wenner probe configuration with 0.5m electrode
spacings in order to concentrate the measurements in the upper 1.5m of
regolith which contained the bulk of the soil wedge.

Due to the lack of other exposures, it was necessary to excavate
three pits at the Lawndale site. The central and eastern excavations
were approximately 2.5 x 3 ft., while the western pit was about 3 x 5 ft.
All were dug to a depth of 6 ft. and their long axes were oriented east-
west, transverse to the local trend of a prominent mesh zone. An elec-

trical resistivity survey was conducted along a west-to-east transect

 

3Appendix B describes the electrical resistivity surveying technique
and illustrates how resistivity anomalies can be detected.

10

immediately north of these pits. The Wenner array configuration was also
used at this site, but the electrodes were spaced only 14" apart in order
to provide maximum resolution of subsurface resistivity anomalies in the
upper 3.5 ft. of the regolith. Samples from the C horizons of the two-
storied soils exposed in these excavations were also subjected to particle
size analyses. Other edaphic data, particularly dominant profile texture,
were collected from 25 auger borings which sampled the cell mesh to a
depth of 5 ft. These auger sites were spaced at approximately 50 ft. in-
tervals, except where observed changes in surface texture or color sug—
gested that additional samples were necessary.

An analysis of the morphometry of the Saginaw patterns was conducted
using photometric techniques. The lengths of the major (a) and minor (c)
axes of 200 selected patterns in the study area were measured to the
nearest 0.1mm on 1:60.000 CIR imagery. The average scale of each photo
was calculated with reference to U.S. Geological Survey 7.5' topographic
maps. These airphotos were utilized in an effort to standardize the
analysis of the patterns from synoptic imagery which was acquired under
nearly constant conditions on the same day.

The local relief of the patterned terrain was determined by calculating
the arithmetic difference between the highest and lowest contour lines in
each survey section as shown on U.S. Geological Survey 7.5' topographic
maps. Since the contour interval on these quadrangles is 5 ft., these
data have an inherent precision of.i 10 ft. Additionally, the areal unit
of measure, nominally one square mile, varies somewhat due to imperfections

in the U.S. Public Land Survey System.

CHAPTER II
LATE WISCONSINAN GEOMORPHOLOGY OF THE SAGINAW LOWLAND

Introduction

The reticulate network of patterned ground in the Saginaw Lowland
which is the focus of this research occurs within a broadly arcuate
tract across parts of Bay, Midland, Saginaw and Tuscola counties. The
study area is part of the Saginaw Lake-Border Plain physiographic region
of the state which is, in general, bounded by the shoreline of Lake
Saginaw, the highest Late Wisconsinan proglacial lake, and the present
shore (Brunnschweiler and Lusch, 1977). In order to correctly interpret
the significance of the patterned ground, its geomorphic context, par-
ticularly with respect to the major glacial and glaciolacustrine land-

forms, must be established.

Moraine Boundary Mapping

The Saginaw Lowland contains a large segment of the Port Huron
Moraine which, as preliminary investigations (Lusch, 1977) suggest, is
spatially related to the distribution of the patterned ground. This
hypothesis prompted a re—examination of the extent of the Port Huron
and Bay City moraines in the study area. Previous attempts to delineate
the moraines in this part of Michigan are limited to the reconnaissance
mapping of Leverett and Taylor (1915) and several manuscript maps com—
piled at some later date by Leverett. The most recent graphic summary

of these previous investigations is the "Surface Formations' map by
Martin (1955) which depicts these moraines as shown in Figure 1.

Little is known about the specific criteria and methods used by
Leverett and Taylor to delineate these ice-marginal landforms, but they

were unable to utilize aerial photographs and detailed soils and

11

12

o a
mmma sauna: an asozm mo .mouo mason on» :« unwouoz nous: uuom onu mo oofiuoowwfimmmao use unouxm .H ouswwm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.oo 33......

.00 3.505

ngasm

.oo 00:02 .63.. 53:...
60 23.22 «‘5

60 52680 Spun

 

 

 

 

 

 

 

 

 

 

Op

d

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l3

topographic maps since these were unavailable at the time of their work.
Differences, perhaps resulting from the lack of these detailed data sources,
are apparent between the present interpretation (Figure 2) and Martin's
1955 map. Most notably, no evidence could be found to support the delinea—
tion of the Bay City Moraine proximal to the Port Huron system in the study
area. The only conspicuous, positive topographic feature in the vicinity
of this "moraine" in Bay County is the fairly continuous Nipissing beach
ridge (Figure 3). In Tuscola County, the "Bay City Moraine" appears to
correspond to a rather complex series of discontinuous beach ridges
associated with proglacial Lake Grassmere and Lake Elkton (Figure 4).
The location of these topographic transects are shown in Figure 5.

With respect to the general trend of the Port Huron Moraine across
the study area, the present delineation and previous maps agree very well.
There are, nevertheless, several places where major differences in the
mapped extent of the moraine are obvious (Figure 6). The first of these
is in the vicinity of the Saginawauscola county line in T. 11 N., R. 6
and 7 E., and T. 12 N., R. 7 E. Here the present interpretation, pri—
marily on the basis of topographic data (Figure 7), established the
proximal margin of the moraine several miles north of the previously ac-

cepted boundary. The extent of waterlain1 moraine in this area has also

 

1In a recent review of waterlain tills, Dreimanis (1979) concluded that
tremendous confusion exists regarding the terminology used to describe
'these sediments. He agreed with Francis (1975) who pointed out that
"waterlain" (i.e., lying in water) was linguistically more appropriate
in this context than the often-used term."waterlai " which means laid
down by water. Consequently, Dreimanis (1979: 167) modified his earlier
definition of "waterlaid" till (Dreimanis, 1969) to read, "waterlain
till: Glacial drift deposited as till in glaciolacustrine, glaciomarine
or glaciofluvial environments." It seems reasonable to adopt a similar
terminology for landforms which were emplaced in an aquatic environment,
as well. In this report, the term."waterlain" moraine describes a type
of moraine or a moraine segment which was deposited along an ice margin
standing in proglacial meltwater or which was subsequently modified by
glaciolacustrine processes.

.oouo human onu ma oawouoz nous: uuom oSu mo sowuoowmwmmmao moo acouxo vomw>om .N ouowam

 

14

 

 

 

can

 

 

80:00 -I:O_.Obm

 

2"}.
:4’:
I6 )l
5:1” '
21,3.
9/
’53,? 0’
«22
1 44‘

 

 

\ n"... inseam

w. . 2....3 .. (Kr W

 

 

 

 

 

 

 

 

 

 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.‘“( OCH“ 0. .0.
2:30! 52.3.3 palm/WW /\I|\ ..........
.u .. mm a... 0......0..\
2.. .2
o. . . . \
.. w. . 3.0 :0 ._.. .,. . _ _
. .. 0.”... .o..oo..o.o.o 17m 3.3.?“ \. .wru. ,
..M.o..u...h.o .o.. oo ”ma-.4. ”mm-«4.. dial
.- .o.o cinch...” V. ....o . “(have °=.-‘-: p
...........0. 2
2. 3o .2: w “T...
s \ km.
. 2 w ham:
A 5....
5s 3M1...“
1. t c.

 

 

 

 

003:.
o— O

 

 

 

.oo :33...
60 3.50.0

\

 

.3. sum

60 02.92 .63... 502.3
.8. 2.22: a: u .u can...

.8 533.0 3. mo .. .

vi

 

 

05
v'

 

 

 

 

 

 

 

 

 

 

 

 

15

.%ua=oo mom Hmuuooolauuo: a. :oofiouoz huwo hum: on» mmouom mom». Hwom mam mmafimoue oanmmuwomoa .m ouawwm

 

 

 

 

 

 

 

   

         

 

 

 

     

 

 

 

 

 

    

 

 

  
     

  

   

    

 

 

  

 

 

 

 

     

 

  
 

 

 

 

.95.. 2.3.0.2.... :0. 3 5.3.3.... .2 < .3232? com 3003 5...: an :30... .. 05...: 3.0 aom .. .- ..o:.o.m
0...... ham;
.3... 2.. P 33.6... . M . 4o
2.. J .73.... 2 2..
m ausnlanfllood
. out... so. a 23......- ..o . . ll
3000.“
2... it... own—.1, 02.3.... m on.
522...... .3. co.
2.00 o g aflsfl 00.
3...... no. .9... on. .0 one... so... 05.83.: 33 0 m .iménomfiruo
2.. .1 3.72... ad...
2?; CQIoolao
co. .0 . 0.9.2.. 0...... z 2.03.0.0 aflolamnmfl do.
one. - III! a“.
O
a... .H 0.3.6... on.
ooo— 1’. I 3.3:... 83......«uoo... ,3.
.«L .m 3.... so... afsassz 4.3.6.. 33-3.9.8...

 

050 n ousvloofi

 

.10..
Q anal?

_ )1

ausolnufiloo.“

 

 

 

 

2.. j (I;
.«o . 3: o .

   

o... a... a...
3323......2. M

 

 

on... so... 9.3.3.:
2

 

find—.mvé

I”. I. .DNO
/<u_oo.
.mvé _.uw.¢ .-

 

 

16

£35.00 .3953. 5” :25.qu >30 .33.. a... mmouom won... .30. on. modwoun owsamumomom.

.q ouswfih

 

$030.0 «60500666:— =°¢ no :O_~n_..OOOU uO..< IUCOQQ< 00¢

Goo—v 5...: 3 :32... .a 05.35. 3.0 .30 11 coidd

 

 

 

 

 

 

   

 

 

  

 

  

  

 

 
  

 

 

  

   

 
    

 

 

 

 

 

 

 

 

 

 

 

 

co. . o... 2..
0°.“ 0°.“ 0' 0‘.“ no N
a... . . . . w .c: c a on.
.v. a... v.
on. 1 2... :3... =2... 1 3o
3.02.0.0
000
000 .
a. on.“ c. on a
nun 0N0
... 8.... a _. 1 8o
.1. ..
on. ...o:.o.m out... so... u. ..o 2. e2...u 0:: on «
2.5.3.0 _ . coo
ONO 000 0' do.“
on.“ I
3. can
\..\2\LL11. 2...? x I
0.. 2.0.9.2.. e333 .2: on... I on.
8.. _ IL 1\|1 8.
000 .N3 on..«.—o\1\llJ\/\I\.\
1 / p. I ONO
.o A
$02.0.m (J 0! 600
23.22.. 5.5m
can... so... 22:39.0 on.“
000 |||\. COO
. a
on. a...» 3.. «A... \mL1\../lu.\l1ll\‘ a...
r .22 o m J .38.... n.
can 23.22.. eo.x.u 2.0. it... I one
.. 5.3.33... :9. ..
10F 32...... no.» 2...
hw<mIPDO¢ p O 4...? hwmgrhmoz

 

 

 

 

 

17

.e cam m mmuawwm cw nsonm mmawwoun oasamuwoaou wan mo mnowuaooa .m unawam

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3:...
q d d
09 O
.o .
302.... 2:0... ® 0 .
2.3360305. Q m
o
coo-.9... 2.3.03.6 _.
\ s

so: 6:3: 633
059.02 >20 awn

 

 

 

 

.00 2003;. .0

 

 

 

 

.o
.00 32.3.0 .o .0
.oo :0 .v .1.
.00 02.32 .63... £35.. .0 .e

 

.3 22...: a: m .u
.8 533.0 2. mo ..

 

 

 

 

 

 

 

 

 

 

 

18

.3 .30... 33.25... m... .38 05...: .55... an... a... mo sowummafidmv .03.... m... .003... confine—50

.mmm. .n..uuz

 

 

 

.0 «Human
3...... .m a... .m a... .m 5. .m o... .m o... .m c... .m a... .m 2.4.. .L on...
_ . J
pp 9... o .o .s .955... o. a O .n. . Q E

5 5...... 3......
553...... .o 5......

>2... 5... o. 33......
.23.... .3...

a. .3... 5...... .3
5.3.5.... .58....
o... 2...... .....<

3.... 5...: a.
3.2.5.. 252......
2.... .. >2... 5...

an .......5... .53....
o... 2...! ....<

.2... a...

.3 a... 3....
5...: .5 53.2....
05.50.... 05 6003...-

3.5.2.. .o .....< ,

 

 

 

-.fi.. .H.

.qmut
0.:
.r.

 

 

.m wwé .m op...—

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

0
.n1‘..

 

05.33. :95: ton.

 

.00 509......

60 .55..»

.oo :0

.oo 0...; .63... 5.05..
60 .3552 «2 u

.00 5.650 .3 ma

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

huaaoo awafiwmm :u.....
can mucsoo .Hoomaa auoumoa :« .....Qz noun: a... on. ..ouo. .maamoum ufizmmuwomos .s unawum

 

l9

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

o .52“. 5 :30... 23...... 5...... .33.... 2.5.02.2... .... .0 2.5.5.... ... < 5.2.2:? ...
.30....
.53.: :95... ...... .... .0 3.3.3:... 0...»... u > .........oo.... _ ....5
......2. .3. p
.0.

on. . no.“ oo.«1 can

can V 1' fill ll 000

.3 L 2..
.0

co. - ...... a... o...

nu. . . can

one - one
.. _.
0N
suitatlau: and D
co. _ no... . . . 331.313: cc.
Duo 0. «In rials—0.... p » 00.... 0N0
K on.“ ....

o... W , on.
.2 33...: 2
’h _ . . . N

000 fitlklbloo «...-o a loan; . coo

d o. «1......

one one

:2
uni-bod

O00 W I.” m 00.“

Duo
.4

a.
.....02

 

 

20

been changed in that all of the moraine below the Warren shoreline (ele-
vation 700 feet), which occurs one mile northwest of Vassar, is inter-
preted as having been deposited in standing water. In Tuscola County,
the portions of the Port Huron Moraine which sustained little modifica-
tion by lacustrine processes exhibit strong local relief and surficial
soil mottling and tend to be naturally well drained. As observed on
color infrared airphotos (Plate I), the bright, more well-drained soils
of the subaerially emplaced moraine contrast vividly with the darker,
more poorly-drained soils of the segment modified by lacustrine action.
Additionally, the surficial mottling west of (below) the Warren shorezone
tends to be subdued, providing corroborative evidence for the waterlain
character of this section of the moraine. No evidence was found to sup-
port the waterlain moraine boundary in the vicinity of Frankenmuth as
shown on Martin's (1955) map.

The differences in the extent of the moraine in north-central Tuscola
County shown in Figure 6 are the result of a re-interpretation of the
waterlain portions of the Port Huron Moraine which will be discussed more
fully in a later section. The platform on the northwestern flank of the
moraine (Figure 8) was previously delineated as lake plain (inundated
till plain), but topographic data suggest that it may in fact be part of
the moraine. This area would certainly have been inundated by Lake
Saginaw during its initial formation and subsequently was subjected to
both fluvial and lacustrine erosion.

A third locality where major differences in the interpretation of
the glacial morphology can be seen is along the eastern border of Midland
County. Topographic data for this area indicate that the Port Huron Mo-
raine extends westward into Midland County (Ts. l4 and 15 N., R. 2 E.)

farther than previously recognized (Figure 9). Although the distal margin

21

 

(NASA-JSC 309-21-102; May 13, 1975)

Plate I. High-altitude, color-infrared airphoto of the Vassar,
Michigan area showing the well-drained soils (white to
light cyan tones) associated with the Port Huron Moraine.

22

.muosoo «accuse auoummonuuoa Ga moaouoz noun: uuom o:u mmouum mmafimoun canamuwonoa .w ouowfim

 

 

o 952“. ... :32? 2.0.38. 2:9...

2.30.0 2.2.3923... ..oo .o 2.2.2.33 ..o. < 523...; one

 

 

 

 

 

 

 

2.... :ooo.«
2.222 .35: ...... 2.. .o 3.3253 3...... u > ‘ ... 522.838
.3
I 'o 0“.
one no a 9 ill“.— 8..
. 8..
005 a? CD.“ _
2:.
on» 00
.00

 

o3 , .. P
3.. 3 . .8: 3 _ \l
'0.“ \E’

”...“, / \.\H,\

. . _ao.NIn«sQ
on «non N one

_ OOh
_ 005

 

 

no-
. and .72;
—

p

‘ 000
mm

ou~¢ cast

.....«3. _S.«uoo.« Tau...

 

 

 

.‘NII‘

 

 

 

Saussom

JV

L

_ cos

- 605
<< :

3035.02

 

 

23

$3.500 .23 £323.53 2.8.33.3 ...... .5235 com .3550 2.33:
£332.33 33.2.“: 25 5..qu a.“ 038.32 no.3. uuom a... amouoo madman.“ canaouwomou. .m mun—mum

 

 

3.5.0 2.2.3523... ..o- .0 2.2.9.0.... .8 < 52.022 0: o 2.6.“. ... .52.. 2.2.2.0. 2:0...

one
one
one

can
050

050
3

2.3.0: 5...... to... o... .0 3.3252. 93...... u p

ensign...“ 8.... 32.3:
If . Lanna? . . kl I

_ Z _ 3.9.3.... 8.3.2. 11 one
_... it P <3 . ......”

 

 

 

 

 

 

 

 

 

    
    

 

 

 

 

 

 

 

 

00.9.8.“ p o
. ..«3..ao.«aoo.u
zoo...“
3.3.0323
.. o. 9 :00. p .80
_ ll /.mlf an no on a 0‘0
H... F <\—\ ...fl 2b

nasalnofllon.“

_ i 1E 8.9.3.... 33-..”...8; one
mm x/éi

. use

 

 

 

 

oodlaolnusv

      

 

3......8... £35976...
_ 3-3-3 _

 

 

 

can
one
2

 

.m

 

 

 

.aom 2.03

 

 

24

of the moraine in this region is obscured by sand dunes (Figure 10 and
Plate II), published pedologic information, to be discussed later,

corroborates the more westerly position of this moraine border.

Topography of the Port Huron Moraine

The Port Huron Moraine exhibits two distinct topographic expressions
in the Saginaw Lowland. Across most of Bay, Midland and Saginaw counties,
where the moraine was waterlain, it has a very subtle topographic form,
with gentle slopes and low relief. In northernmost Bay County and in
Tuscola County east of vassar, on the other hand, the moraine becomes much
more pronounced and, in general, local relief along its trend increases to
the north and northeast.

In Gibson Township (T. 18 N., R. 3 E.), Bay County, the Port Huron
Moraine attains an elevation of over 820 feet, rising some 50 to 70 feet
above the surrounding countryside (Figure 11). This three-mile-wide sec-
tion of the moraine is quite rugged with local relief averaging 45-50
ft/mi2 and occasionally becoming as much as 85 ft/miz. Along the southern
fringe of this hummocky moraine is a narrow (l-2.5 mi.) lacustrine-
modified morainic zone which is only 5-15 feet above the adjacent land-
scape and has 20—25 feet of local relief per section.

The large morainic segment in eastern Midland, southwestern Bay and
northwestern Saginaw counties, interpreted as being entirely waterlain,
is 2.5-6 miles wide and has a relatively smooth and regular proximal
slope. Its crest decreases in elevation toward the southeast, from 690-
695 feet in Larkin Township (T. 15 N., R. 2 E.), Midland County to 615
feet in Saginaw Township ( T. 12 N., R. 5 E.), Saginaw County. As men-
tioned earlier, the distal slope of this part of the Port Huron Moraine

is covered in many places with sand dunes and these eolian landforms

25

   

    
 
    

. 4.....-3 ------

{r

m

 

 

LS

 

 

 

 

 

 

 

 

 

 

\.\ M In

   

 

1 i >\ \ f E
I ~Townhall V |

12".-xx ‘“ w V ,C'V\

./

 

Figure 10. Portion of the Midland North 7.5-minute topographic quad-
rangle, showing the sand dunes along the distal slope of the
Port Huron Moraine in central Larkin Twp., Midland County.

26

    
   

J'

K. k’ "' i ' ‘7. *3' 2
. 0v , .. .-

(NASA-JSC 309-21—67; May 13, 1975

   

; Original in color)

Plate II. High-altitude aerial photograph of the sand dunes on the
distal slope of the Port Huron Moraine in eastern Midland
County.

27

.5550 ham 5.23.50... 000.35 2.0
.3250 0.22.0.5 52.0030... afiuu ....“ 055.0: no.5: uuom w... 009.00 00.3.0.3 0.3025000... .HH 9.0m...

 

 

050
can
can

own

was
005.
Obs.

con

00..

can
obs

coo

02.0.0 2.2.5023... :00 .0 2.2.2.002... .0. < 50000.2 000 o 0.00.... ... 030:0 2.2.2.0. 2:0...

    

35:00.05“... 00.0.02 00...... ..00 2.. .0 02.00.52. 00030. n)

N ((F/ o 00. F
. -..... 8.72.3-3...

undiagmlnwxv lam;

 

 

     

   

    

  

modlao
Ian

nN\MIaD.NIaN\v
f N

am.Nlno.NlnN\v ION;
\/.\/ aux? _ OSIOD _ wodlnmlub
I E 00h
. > / ' >L
1W Obs.
.500... <

QN\0IDQ.NIDN\V co_.a.ououxo

_fl0=._0>
.8? no...

.3...
7 caduceus“? .52; "E0 .3650-.. 0....
/ \.‘ fl

.x </ \ . L...“
é

 

 

 

   

=00

 

 

 

 

 

 

andlnmdlnmxv 99.53?

t 1) In“: as)...
/ _ 3-8-2.? on.
2..
.2. /./\I \ r .s

{ 1 000
.03 .025 ..

1\

 

 

 

 

 

 

 

28

increase the local relief along its western and southwestern flanks to as
much as 35—40 ft/mi2 in contrast to the 5-15 ft/mi2 that is typical along
its proximal slope.

The l-4.5 mile-wide waterlain portion of the moraine in Saginaw and
Tuscola counties, from Bridgeport eastward to the Vassar area, presents
a similar topographic appearance. Here, too, the relatively smooth crest
decreases in elevation toward the Saginaw River from 680-690 feet just
west of Vassar to about 625 feet in the area north of Bridgeport. The
low relief (5-10 ft/miz) of this part of the moraine is typical of its
waterlain character.

The largest hummocky segment of the Port Huron Moraine in the study
area trends northeastward across Tuscola County from Vassar to Gagetown.
Like its subaerially-emplaced counterpart in northern Bay County, this
part of the moraine exhibits a strong topographic expression, attaining
elevations more than 100 feet above the adjacent plains in distances of
one or two miles. Although relatively narrow, only 1-3 miles wide in
most places, it becomes more extensive in Elkland Township (T. 14 N.,

R. 11 E.) where it reaches a width of 6 miles. Throughout its length,
this portion of the moraine is relatively hummocky, with local relief
averaging approximately 40-60 ft/miz. Its crest is much more broken comr
pared to the waterlain parts and, in general, rises in elevation from
about 710 feet near Vassar to over 800 feet south of Gagetown.

Across most of the study area, the Port Huron Moraine displays an
asymmetrical form, its distal slope being significantly steeper than its
proximal counterpart (see Figures 9, p. 23 and 11, p. 27). The ratio of
the distal slope to the proximal slope varies between 1.1:1 and 6.8:1
with an average of 2.6:1. This asymmetry may result from three factors.

Primary asymmetry could occur due to the mechanics of moraine deposition

29

directly from the ice mass. Debris in an ice lobe is usually concentrated
on the surface of the marginal zone and along thrust fault planes and di-

'minishes up-ice. Such a distribution of supra— and englacial drift, upon

deglaciation, could result in an asymmetric moraine. Deposits of outwash

in the form of fans or aprons along the steeper distal flank of a moraine

would tend to reduce this primary asymmetry, but such deposits are notably
lacking along the Port Huron Moraine (Leverett and Taylor, 1915: 298).

The impact of proglacial lakes on the Port Huron Moraine is the
second factor. In many places, the distal slope of the moraine was sub-
jected to erosion along several proglacial lake shorelines, including
those of Lakes Warren, Grassmere and Elkton. In Midland and Saginaw
counties, these distal slope beach zones faced into the prevailing west-
erly winds and must have been subjected to stronger wave action than the
proximal slopes to the leeward. This situation was reversed in Tuscola
County, however, and the prominent bench cut into the proximal margin of
the moraine there owes its origin, at least in part, to enhanced wave
action along a windward shore. According to Taylor (Leverett and Taylor,
1915: 300), the unusual strength of this feature may be the result of
fluvial erosion by a meltwater stream draining to the southwest between
the ice front and the moraine.

The influence of post-glacial erosion is the third factor. On both
sides of the Saginaw Lowland, the Port Huron Moraine acts as a modest
topographic barrier to surface drainage. The courses of both the Titta—
bawassee and Cass rivers are controlled to some extent by this moraine.

As a result, the toe of the distal slope has been subjected to fluvial
erosion which may have maintained a steeper slope through time than would
otherwise have been the case. These three factors, acting in concert, may

explain the asymmetrical form of the Port Huron Moraine in the study area.

30

Composition of the Port Huron Moraine

With the exception of the eolian and lacustrine sands scattered
across its surface, the Port Huron Moraine is composed of a clay-rich
drift throughout the Saginaw Lowland, as noted by Taylor (Leverett and
Taylor, 1915: 297), although loam is the most frequently mapped parent
material texture.

The pedologic distinction between the waterlain portions of the
moraine and its more rugged counterparts is primarily a difference in
slope class and soil drainage characteristics. In general, the sub-
aerially deposited parts of the moraine are composed of well, moderately
well, or somewhat poorly drained soils having slopes of up to 12 percent.
The waterlain segments, on the other hand, consist of poorly and somewhat
poorly drained soils, with slopes of no more than 3 percent. Tables 1
and 2 summarize the predominant soil types associated with these two
moraine categories.

The distal border of the Port Huron Moraine in Midland County, as
delimited in this study, is two to four miles west of its previously
recognized location, an increase in area of about 34 square miles.

Figure 12 depicts the soils of Larkin (T. 15 N., R. 2 E.) and Midland
(T. 14 N., R. 2 E.) townships as generalized from Hutchison (1979). The
genetic implications of these parent material distributions support the
inclusion of this area within the moraine.

0n the west side of Larkin and Midland townships, glaciofluvial
sediments and more recent alluvium cover the Tittabawassee River outwash
channel and have, to varying degrees, been modified by wind action. These
deposits consist primarily of Belleville loamy sand, Covert sand, Kings-
ville loamy fine sand, Kinross mucky sand, Oakville fine sand, Pipestone

sand, Sloan loam and Wixom loamy sand.

31

Table l

Predominant soil types associated with the waterlain segments

of the Port Huron Moraine in the Saginaw Lowland

 

 

 

Soil Slope
County Soil Mapping Unit Name Management Group* Class (2)
Bay Londo-Poseyville complex 2.5b & 3/2b 0-3
Tappan-Poseyville complex 2.5c-c & 3/2b 0-3
Poseyville loamy sand 3/2b 0-3
Tappan loam 2.5c-c 0-3
Londo loam 2.5b 0.1
Saginaw Capac sandy loam-Brookston 2.5b & 2.5c 0-3
loam
Capac-Brookston loams 2.5b & 2.5c 0-3
Wixom—Thetford-Arenac loamy 4/2b, 4b & 5/2b 0-3
sands
Arenac - Kingsfield sands 5/2b & Sc 0-3
Capac-Shebeon-Parkhill loams 2.5b, 2.5b & 2.5c 0-3
Parkhill-Kilmanaugh loams 2.50 & 2.50-d 0-3
Sims clay loameParkhill loam 1.5c & 2.5c 0-3
Tuscola Parkhill—Londo loams 2.5c & 2.5b 0-2

 

*See Appendix A.

32

Table 2

Predominant soil types associated with the subaerially-emplaced
.gggments of the Port Huron Moraine in the Saginaw Lowland

 

 

 

Soil Slope
County Soil Mapping Unit Name Management Group* Class (2)
Bay Londo-Wixom complex 2.5b & 4/2b 0-4
Guelph-Menominee complex 2.5a & 4/2a' 2-8
Wixom loamy sand 4/2b 0-3
Tuscola Guelph loam 2.5a 0-12
Londo loam 2.5b 0-2
Boyer-Manceolona loamy sands 4a & 4a 0-6
Marlette-Menominee associa- 2.5a & 4/2a 0-6

tion

 

*See Appendix A.

33

 

*—

man "955]
M

Larkin Township
1115 N. 3.2 E.

Midland Townshlp

T.14 N. R.2 E.

 

- Tlll

Sand. 20' - 60' overlying till

00-:
\

80nd>eo' (mostly dunes)

 

fins-textured Iscuslrlno sediments

Alluvlnl sediments
m

Urban land

 

       
 

 

Figure 12. Parent materials of the soils in Larkin and Midland
townships, eastern Midland County.

 

34

Consistent with its waterlain character, lacustrine sediments are
found in many places on or near the Port Huron Moraine in this part of
Midland County. The largest areas of such deposits are outside of the
distal border of the moraine in northwestern Larkin Township. The
finer-textured lacustrine materials, consisting of Bowers silty clay
loam, Ingersol silt loam and Lenawee silty clay loam, are often juxta-
posed with coarse-textured deposits such as the Lenawee-Wixom—Belleville
complexes. The complicated, interspersed distribution of these lacus-
trine facies supports the interpretation of this locale as a shallow,
littoral environment in the proglacial lakes which inundated this part
of the study area.

The occurrence of many tracts of till (Londo and Parkhill loans) at
the surface in northeastern Larkin Township indicates isolated pockets
of the morainic surface which were not buried by lacustrine or eolian
sediments. Many other areas in eastern Larkin and Midland townships con-
sist of sand of varying thickness, overlying till. These two-storied
soils include Belleville loamy sand, Covert sand (loamy substratum),
Pipestone sand (loamy substratum), Poseyville-Londo complex and Wixom
loamy sand.

Large portions of the distal slope of the moraine in Midland County
are covered by deep, hummocky sand deposits associated with dunes or
beach ridges. These include Covert sand, Oakville fine sand, Pipestone
sand, Pipestone-Oakville complex and Plainfield sand. Most of the re-
maining surface sediments in this area are lowbrelief deep sands, such as
Abscota loamy sand and Kingsville loamy fine sand. These sandy soils re-
present deltaic materials which were deposited in the littoral zones of
several progressively lower proglacial lakes, particularly those of the

warren and Grassmere phases.

35

Waterlain Moraines

Even during the reconnaissance mapping of the surficial formations
of Michigan by Leverett and Taylor (1915), it was recognized that certain
moraines, in part of in total, had been deposited in standing proglacial
meltwater. As mentioned previously, extensive segments of the Port Huron
Moraine within the study area were emplaced in this manner. Among the
criteria used to distinguish waterlain moraines from their subaerially
deposited counterparts are their comparatively more subtle topographic
expression and muted local relief, although the mechanism(s) of waterlain
till deposition are not well understood (Dreimanis, 1979). On the basis
of these criteria, most portions of the Port Huron Moraine above the Warren
shorezone can be classified as a typical, subaerially-emplaced moraine
since they rise abruptly to as much as 100 feet or more above the adjacent
landscape and have high-relief surfaces averaging 40-60 ft/miz. However,
the ice marginal position marked by this moraine-dammed proglacial Lake
Saginaw and the level of this water body determined how much of the mo-
raine was waterlain. The shorezone of this meltwater impoundment has been
deformed by postglacial rebound and in the modern landscape it rises to
the north-northeast. The Lake Saginaw shoreline increases in elevation
from about 715 feet near the central axis of the study area to 750 feet
in the vicinity of Cass City. As shown in Figure 13, the Lake Saginaw
water plane intersects the Port Huron Moraine at about 725 feet a little
southwest of Watrousville and at approximately 750 feet near Cass City.
The portions of the moraine above this place were deposited in a subaerial
environment whereas those below it were waterlain.

A conjugate situation can be observed in northern Bay County and
southwestern Gladwin County where the rugged portions of the Port Huron

Moraine are also restricted to elevations above 750 feet, flanked on the

36

.000H300 Boofiwom 0:0 :« 0030a H0H00Hwoua 00000H00 mo moaoan 00003

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

00500000 030 000 000000: 000:: 000m 0:0 00 00000 0:0 wooao magmoua vanaouwoaou 00000000 0< .ma ouawam
000 000 000 00....m 3.0.. m
000 000 000 0.00.0005 3.0.. 0
.50
:0—~..—.°OIX. , QC. 000 ace 9 0C).; .8.1- 5.;
_¢0-:0> =00? .
0001. 02. 000 000 . 00:03 3.0. .s
00> 0:. 000 300.000 3.0.. 0
50.0.00... 0.0.55... 3: 00:03.0
00 2: 00:03.0 00.0 >030 0:00 .0...00..0..
00° .
.00 0.0000... _ .00 300.000 _ >00 .00 0.3.0.: .00 5300.0 .00 >00
l. 000
m ll .llhum
.. 2. 3 0 g z. 00>
I- I
M0ufl H'U ‘Vfflanvs. 0 on.
1 V» 30
(S
.00 x0
2
*\ Q 0 O k. Q ‘8 b? b A.
\0 0%“ £90 \\\A . \é 0° 0 as? O\
0 ¢ 00 .0 $¢ \Q \00 \b\ \Q

 

 

 

37

south by a waterlain apron between 725 and 750 feet which was deposited

in a subaquatic environment.

Proglacial Lake Sequence and Chronology

During the Late Wisconsinan, the Saginaw Lowland was the site of
several proglacial lakes. These meltwater impoundments affected the type
and distribution of several landforms, particularly waterlain moraines.
They influenced the texture of soil parent material and left as their geo-
morphic legacy a series of erosional and depositional shorezone features.
Although numerous lakes of similar origin may have existed in the Saginaw
Lowland at various times throughout the Pleistocene, direct surface evi-
dence can be found for only those which came into existence during the
Late Wisconsinan.

The Saginaw Lowland, and probably all of the Lower Peninsula of Michi-
gan, was ice-free about 13,300 yrs B.P. during the Mackinaw Interstadial
(Farrand, Zehner and Benninghoff, 1969; Dreimanis, 1977). Any lake which
may have occupied the Huron Basin at that time was below the present lake
level of 580 feet (Hough, 1963; Dreimanis and Goldthwait, 1973).

During the subsequent Port Huron Stadial, the ice readvanced and
built the Port Huron Moraine. This major ice advance blocked lower drain:
age outlets to the east and resulted in an independent, high-level, pro-
glacial lake in the Saginaw Lowland which drained westward through the
Maple-Grand outlet. Lake Saginaw, shown in Figure 14, received meltwater
from both Lake Whittlesey, impounded in the Huron basin to the east, via
the Tyre-Ubly channels, as well as from numerous outwash streams. In
addition, some meltwater from the interlobate area of Oscoda and Alcona
counties to the north was channeled southward along the ice margin and

entered Lake Saginaw near West Branch in southwestern Ogemaw County.

38

40.003 00.5.. 000.. aaafixoa 05

0.50:0 30.5w0m 00.0...

.0. 0000.0

 

 

 

 

 

 

 

 

 

 

 

 

 

  

0.05.0..0
23:23. 2:
0.0.. 00.0—30.0

\/
3.2:...3 8.0... \/

\

 

 

 

 

 

 

 

 

 

 

 

 

 

39

As shown in Table 3, five radiocarbon dates have been determined for
wood samples associated with Lake Whittlesey sediments or those that are
transitional between Lake Arkona and Lake Whittlesey. The average of
these dates is 12,976;: 524 yrs B.P. suggesting an age of about 13,000 yrs
B.P. for the correlative Lake Saginaw and Port Huron Moraine.

As discussed earlier, Lake Saginaw had an elevation of about 750 feet
in the vicinity of Cass City near the mouth of the Lake Whittlesey dis-
charge. The entire study area is north of the Saginaw/Whittlesey zero
isobase so the 695 foot elevation of the lake in the horizontal zone does
not apply (Leverett and Taylor, 1915). Instead, evidence of the Lake
Saginaw shoreline throughout the Saginaw Lowland occurs at or above 715
feet.

Nearly 300 years later, the tip of Michigan's "thumb" was uncovered
and the proglacial lake in the Saginaw Lowland once again became an em-
bayed extension of the Huron basin. At this time, the water plane fell
to a new level of 680 feet (in the zone of horizontality) and Lake Warren,
which also discharged to the west via the Maple-Grand system, came into
existence (Figure 15). Within the study area, most of which is north of
the Warren hinge line (Leverett and Taylor, 1915), the Warren shoreline
has been deformed by postglacial rebound to its present elevational range
of 700 to 725 feet. The Warren I stage has been assigned an age of
12,730;i 230 yrs B.P. (I-3665) on the basis of wood samples from a site
on the edge of the Erie Lowland in western New York (Calkin and MoAndrews,
1969; Buckley, 1976). This sample (I-3665) came from a peat zone which
was deposited either during a late stage of Lake Whittlesey or during the
highest Lake Warren stage (Warren I) which followed. According to Calkin
(1970), the elevation of the dated sample, only five feet below the pro-

jected Warren I water plane, and its association with the remains of

40

Table 3

Radiocarbon ages of materials associated with Lake Whittlesey
or the Lake Arkona-Lake Whittlesey transition

C14 Age

(yrs B.P.)

Sample No.

Remarks

Reference

 

12, 660 i 440

13,600 i 500

12,920 i 400

12,800 i 250

12, 900 i 200

8-31

W-33

W—430

Y-240

I-3l75

Sample from a waterworn Larix

log 2.5" x 2.5' from 32'-42'

beneath the surface of an Ar-
kona-Whittlesey bar at Ridge-
town, Ontario. Elevation -
703'-713' A.T. Highest War-
ren beach (705') is 1 mile
SE.

Wood from a stratigraphic
position between deposits of
Lake Arkona and Lake Whittle-
sey near Cleveland, Ohio.

Wood from peaty zone below
Lake Whittlesey beach gravels
at Ohio Turnpike-Ohio Rte. 4
intersection in Erie County.
Underlying the peat is sandy
alluvium and lake clay over
calcareous till. Near Par-
kerstown, Ohio.

Spruce wood fragments imbedded

in beach sediments of glacial
Lake Whittlesey from a well
4.5 miles SE of Bellevue,
Ohio.

Wood in Whittlesey beach near
Elyria, Ohio.

Dreimanis, 1966

Suess, 1954

Rubin & Alexander,
1958

Barendsen et al.,
1957

Calkin, 1970

 

41

.H0H000m 000:: 000m wowoma 0:0

mowua0 H 000003 0304

. 3 00:0...

 

 

 

 

 

 

 

 

 

 

 

 

 

 

    

0
000.0000.

 

   

...-gnu:

AAA

 

  

IDDFIbDID}
1111 I
Gui-00.00.
0 0
0

 

 

I
lplpbh

 

I'lII

   

 
 

A

 

      

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

42

shallow-water flora favor its correlation with the Warren I stage.

According to Hough (1963), after Lake Warren had been lowered to
670 feet (Warren II), it fell temporarily to the level of Lake wayne
(660 feet) and then rose to the Warren III elevation of 665 feet. Little
evidence can be found in the study area for any post-Warren I levels above
the Grassmere shoreline. Nevertheless, there are some landscape features
which suggest a shorezone on top of the waterlain Port Huron Moraine in
western Bay and eastern Midland counties. These consist of relatively
narrow, curvilinear zones of sandy material which frequently contain low
ridge forms. These minor topographic features are usually faint but they
occur in many places and, in association with the sandy soils, seem to
define a former shorezone at 655-660 feet which may be associated with
Lake Wayne or some other minor halt in the progressive lowering of Lake
warren.

As the Twocreekan Interstadial continued, the levels of the glacial
Great Lakes in the Huron and Erie basins fell to even lower elevations
as new outlets became operative. Progressive ice margin retreat in the
vicinity of Syracuse, New York allowed meltwater to flow eastward via
the Mohawk Valley and, as a result, Lake Grassmere came into existence.
This proglacial impoindment (Figure 16) had an elevation of 640 feet in
the zone of horizontality which includes most of the study area. NOrth
of its zero isobase, in northern Bay and Tuscola counties, the Grassmere
shoreline rises to about 650-655 feet.

Lake Elkton (Lundy) was the next proglacial lake in this short-
lived series and its horizontal shoreline segments have an elevation of
aboug 620 feet (Figure 17). Elkton shorelines deformed by rebound in
northern Bay County are found at 630 feet while those in northeastern

Tuscola County reach the 640 foot level.

400003 00.00.. 0.00.. 9.0003 05 $5.000 000600000 00.0.. .eH 000w“...

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

43

 

 

 

 

 

 

 

 

 

 

 

 

 

....... \IL
o.
000000000000000000000000
000000000000000000000000000000000
0000000000000000000000000000000000000
0000000000000000000000000000000000000
000000000000000000000000000000000000000
nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn
000000000000000000000000000000000000000
oooooooooooooooooooooooooooooooooooooooo
000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000
000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000
000000000000000000000000000000000000000000000
000000000000000000000000000000000000000
0000000000000000000000000
00000000000000000000000000000000000000000000
000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000
000000000000000000000000000000000000000000
000000000000000000000000000000000000000000
0000000000000000000000000000000000000
0000000000 1000000000 00000.0 0000000
00000000000000000000000000000000
oooooooooo r..-...... o, 0......
000000000000000000000000000
000000000000000000000000000
cccccccccccccccccccccccc
000000000000000000000000
0000000000000000000000000
0000000000000000000000000
0000000000000000000000000000
...... v.......... ......u .....u..
000000000000000000000000000000000
ooooooooooooooooooooooooooo 000000000
"l’ ?’ -’ l.’ .IIDbII Lb,
144114 44: 44 k ‘4
000000000000000000000000000000000000000
0000000000000000000000000000000000000000000
000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000
000000000000000000000000000000000000000000
000000000000000000000000000000000000000000000
000000000000000000000000000000000000000000
0000000000000000000000000 Assoc-0000000000000
00000000000000000000000000000000000000000
000000000000000 v0.0.0......00-.0000-A00000-c
000000000000000000000000000000000000000000
000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000
000000000000000000000000000000000000000000000
...... .............................................
00000000000000000000000000000000000000000000000
... .............................................
. .
. 000000000000 0000000 000000000000000000000000000000000000000000000000000000000000000000000000000000
000000000000000000000000000000000000000
00000000000000000000000000000000000000
0000000000000000000000000000000000000000000000000
000000000000000000000000000000000000000000000000
OOOOOOOOOO . ... ...ICIOOIfi ICIOIIII IIOIOIC IIICQOII.
\ IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII
00000000000000000000000000000000000000000000000
000000000000000000000000000000000000000000000000
00000000000 . .0000 0000000000 00000... 000.0.0000a0-0-
00000000000000000000000000000000000000000000000
0000000000000000000000000000000
000000000000000000000000000000
00000000000000000000000000000000
00000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000
000000000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000000000
IIIIIIIII I IOIQODIDOI ......CIUIWIII'IOII. ‘CICOOIIII.
000000000000000000000000000000000000000000
00000000000
00000000000000000000
000000000000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000000000
000000000000000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000000000000
000000000000000000000000000000000000000000000000000000000
000000000000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000000
00000000000000000000000000000000000000000000000
0000000000000000000000000000000000000000000000
000000000000000000000000000000000000000000000
IIIIIII 1'7le .lll ....-
lllll lqu 1d‘d1ll111 2‘
000000000000000000000000000000000000000000
00000000000000000000000000000000000000000
000000000000000000000000000000000000000
0000000000000000000000000000000000
00000000000000000000000000000000
00000000000000000000000000000000
0000000000000000000000000000000
0000000000000000000000000000000
0000000000000000000000000000
0000000000000000000000000000
00000000000000000000000000
0000000000000000000000
0000000000000000
00000000000000000
0000000000000000
000000000000
000000000000
00000000000
00000000
000000000
I
00000000
00000000
00000000
00000000
00000000
000000000
00000000
000000000
0000000000
0000000000
0000000000
00000000000
0000000000
0000000000
000000000
0000000000
000000000
0000000000
000000000
0000000000
b o
000000
000000000
sssssssss
000000000
000000000
000000000
000000000
000000000
000000000
00000000
000000000
000000000
........ C
0000000
00000000
0000000
00000000
00000000
000000000
000000000
000000000
butt
0

 

 

 

 

 

 

44

.H0ac0um nous: uuom waaa03 0:0 wcwuav souxam 0304

.NH 0u=wwm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

\

 

 

 

 

 

 

 

 

 

 

45

Following the Elkton stage, continued downcutting of the Chicago

outlet eventually formed Early Lake Algonquin at an elevation of about

605 feet, but no shorelines or other features associated with this pro-
glacial lake are knwon (Karrow et a1, 1975). The oldest radiocarbon dates
on plant detritus and wood correlated with Early Lake Erie and Lake Iro-
quois, contemporaries of Early Lake Algonquin, average 12,373 i 466 yrs
B.P. and are summarized in Table 4. Thus, Early Lake Algonquin probably
came into existence about 12,400 years ago. Continuing ice margin retreat

into Ontario opened the Kirkfield-Trend Valley outlet to the east and,

Table 4

Radiocarbon ages of materials associated with Early Lake Erie
and Lake Iroquois

 

 

14
C Age

(yrs B.P.) Sample No. Remarks Reference

12,650 i 170 I-4040 Basal plant detritus asso- Lewis, 1969
ciated with Early Lake Erie
from the Pelee Basin of
western Lake Erie.

12,660 i 400 We86l Sample from Picea mariana Rubin & Alexander,
log in laminated silt and
clay deposits of Lake Iro-
quois at 340' near Lewis-
ton, New York.

12,080 :_300 We883 Sample of another Picea Rubin & Berthold,
mariana log collected from 1961
the same place at the same Muller, 1965
time as W¥861.

12,100;: 400 I-838 Spruce wood in Lake Iro- Buckley et al.,

quois sediment 4.5 miles 1968
N of Lockport, New York.

 

46

beginning about 12,000 yrs B.P., the lake level in the Huron basin was
once again below its present level of 580 feet (Karrow et al., 1975). It
is unclear whether crustal rebound or a minor ice advance caused the
abandonment of the Kirkfield outlet, but whatever the cause, Karrow et a1.
(1975) concluded that as a result of its closure, Main Lake Algonquin
formed at an elevation of 605 feet approximately 11,200 yrs B.P.

Discontinuous segments of the relict shorelines of several Late Wis-
consinan proglacial lakes can be traced across the study area as shown in
Figure 18. Shorelines at or above the Elkton level, excluding the pre-
Warren stages, were delineated because of their spatial relationship to
either the Port Huron Moraine or the patterned ground, or both (Lusch,
1977). The proglacial lakeshore features below the Elkton beach were
not investigated, except for those portions of the Algonquin and Nipissing
shorelines in Bay County which were relevant in the search for the Bay
City Moraine.

The depositional landforms depicted in Figure 18 consist mainly of
low, sandy ridges representing either beach ridges or offshore bars. The
stippled shorezone areas define landscapes which are dominated by ero-
sional features, such as wave-cut nicks and benches, but also include
some depositional forms. The shoreline segments shown by the dotted lines
are probably not as accurately located as the other types, since they are
based on incomplete or less convincing evidence. Nevertheless, their
trends are clearly indicated and the zone of uncertainty containing their
true locations is limited to perhaps a few thousand feet on either side
of the plotted lines.

The morphologically most distinct shoreline erosion was developed
along the southeastern margins of the warren and Elkton basins, consistent

with the presumption of prevailing westerly winds. In southwestern Bay

47

 

 

:00 30:00 >00 ... 00000....

 

. 02.03 .5

o 0.303 03

 

 

 

0.20.020 a

 

 

 

 

 

coax-m N

00.00000? <

 

 

 

 

 

 

.2333... x

 

 

 

   

 

 

 

 

”NOS—.0 mx<4 450(40033 . .
0 .. ..
.0? m .. a. .. u” . 3 m
3 m ........ up....pm0 .. .. ..
... ... a! 0 .‘V.’ .H o. o1
00 an .041 00.;
0.00 2.30.0003 2.0 =00 2020500.. :0 0000.. 05.0.0... 30.5.3.2? . .. .L .. 03.
d 000 A 0”
00.300. .0..o...0oa00 0...... w .. ...- \ 0. 3
000 06.2.0... 30200.0 000.05 go...) 000.0 00000.20 ... .. .fi 3.x s
x . .. .. .. \0.
0.0.. 0.2.0.... 000 0000.. c0000 00.020... 60.300. 00:03... .00050230 I. ... z < m o \ _
Q
30.

 

 

 

 

 

 

 

 

 

 

 

48

County, several spits, attributed to both the Grassmere and Elkton lake
levels, were identified and mapped. Other depositional shoreline features
of Lake Elkton in western Saginaw County have been significantly modified
by wind action and, as a result, the location of this segment of its

shore may be shown too far to the east. The generally weak character of
most of these glaciolacustrine shorelines is probably a function of the
short time interval during which any particular lake level was maintained.

As will be discussed later, the lower elevational limit of the pat-
terned ground is coincident with the shoreline of proglacial Lake Elkton
in the central and eastern part of the study area where numerous deposi-
tional and erosional landforms mark this former coastline. This morpho-
logic relationship continues northeastward across the hinge line into the
uplifted zone where Elkton beachridges rise from their horizontal eleva-
tion of 620 feet to 640 feet. This spatial association was utilized to
map the Elkton shorezone in Bay and north-central Saginaw Counties where
other geomorphic evidence of the Elkton beach is lacking.

In comparison with previous interpretations, as summarized by Martin
(1955), the present delineation of the proglacial lake shorelines in the
Saginaw Lowland is of considerably greater accuracy in Bay and north-central
Saginaw Counties. Additionally, the location of the Grassmere beach within
the study area was previously undefined. Although these glaciolacustrine
features are not the principal focus of this research, their locations
throughout the Saginaw Lowland had to be determined in order to establish

the relative time-stratigraphic position of the patterned ground.

Eolian Landforms
In light of the Late Wisconsinan history of the area, it is not sur-

prising that sand dunes are abundant in the Saginaw Lowland. Each of the

49

proglacial lakes which inundated this area had the potential of depositing
sandy material along its shoreline; additional sediments which were sus-
ceptible to subsequent deflation were provided by the numerous prograding
deltas that were extending into these meltwater impoundments. Perhaps

even more important than these littoral and deltaic sediments, however,
were the glaciofluvial deposits. Particularly in the western part of the
study area, the dense network of lowbgradient rivers and creeks undoubtedly
provided expansive areas of sandy material after each flooding.

The largest concentrations of sand dunes in the Saginaw Lowland are
located in eastern Midland, western Bay and southwestern Saginaw counties.
Additionally, many tracts of land south of the Cass River in Tuscola
County are covered by dune forms. Dune formation was favored in these
areas because of the abundant availability of sandy materials derived
primarily from.the many sediment-laden streams.

Thus, the Tittabawassee, Cedar and Tobacco rivers in southwestern
Gladwin and north-central Midland counties, as well as the upper Cass
River and the headward reaches of many of its modern tributaries in Tuscola
County, all deposited large quantities of sediment, not only on their flood-
plains, but also into Lake Warren (see Figure 15, p. 41). With the lowering
of the water plane to the Grassmere and then Elkton levels (see Figures 16
and 17, pp. 43 and 44), the Cass River became fully established in Tuscola
County and emptied into the proglacial impoundments in southeastern Saginaw
County. At the same time, the headward reaches of the Shiawassee River and
many of its tributaries provided copious amounts of alluvium to the pro-
glacial lakes in southwestern Saginaw County. In eastern Midland County,
the Chippewa and Pine rivers contributed their sediment loads to that of

the Tittabawassee River.

50

All across the Saginaw Lowland, the parabolic dune form predominates
(see Figure 10 and Plate II, pp. 25 and 26), although in many places, par-
ticularly in the western part of the study area, the dune ridges have been
drawn out downwind to form hairpin dunes. Virtually everywhere in the
Saginaw Lowland the axes of these inland dunes trend west-northwest to
east-southeast. In this part of Michigan, eolian landforms developed as
the landscape emerged from beneath the progressively lowering proglacial
lakes of the Late Wisconsinan. They document a late-glacial paleoenviron—
ment in which eolian processes were dominant and indicate 1) that vegeta-
tion was probably scarce in the imediate periglacial zone of the Saginaw
Lowland at this time and 2) that the prevailing wind direction during the

Late Wisconsinan was essentially the same as today, namely westerly.

CHAPTER III
CHARACTERISTICS OF THE PATTERNED GROUND IN THE SAGINAW LOWLAND

Distribution and Extent of the Patterned Ground

In the only other study of the patterned ground in the Saginaw Lowe
land, Tillema (1972) analyzed approximately 135 mi2 of terrain in south-
western Bay and north-central Saginaw counties. Although it was stated
that patterned ground could be observed in Midland County and as far north
as northern Bay County, no attempt was made to map its extent beyond the
boundaries of the aforementioned study area.

The present study expanded the search area considerably by including
all of Bay, Saginaw and Tuscola counties, as well as the eastern two-
thirds of Midland County, the southeastern quarter of Gladwin County and
the southwestern margin of Huron County. Figure 19 depicts in a general
way the distribution and extent of patterned ground in the study area,
which encompasses nearly 2,500 miz. The boundary lines shown on this map
represent the outermost limits beyond which no patterns could be observed
on aerial photographs; they enclosed approximately 360 mi2 of patterned
terrain within which areas of especially prominent patterned ground are
shown. Most of these patterns occur in Bay, Saginaw and Tuscola counties.
Since conspicuous patterns were not observed on airphotos of southwestern
Huron County, the boundary lines were halted at the northern border of
Tuscola County although the zone of pattern formation probably continues

northeastward for one or two miles.1

 

1Although Huron County appears to be devoid of patterned ground, another

area of distinct ground patterns is obvious on aerial photographs of
eastern Sanilac County.

51

52

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

000:5... B000w0m 0.00 .... 000o0w 000000000 .0 0000.00 000 .50000000000 .3 00..me
E O I O I O O C O I O I O O O O O
.._. .00... 00 0 0.. 0 mo .... 00 0 m. 0 00 0 00 0 m. 0 .
q . 269....
0.. 0
00.0200 200.03... .0 000.< .03.: .00.... .25...
.zapfi
.... 7
«3.5.00. 0.0.... 000000. . ...
0000... 000.020.. 0... .0 0...... \ . .. .. . .
x . . . z... ..
\ ... . an“...
I ' ‘ 2 5‘ C I 0 . 0
x .. ... .... 0.0....
s . . .... .23...
. .. ......
s s u .
b .
. .... .. .20:
«.... 0. ”0..
V .
H... @020:
.00 0.0000... .0 ... .
.00 305°C” .0 .. \m .0 .. . ...sn. . III
.0 ... ,/ o 0. . .
.00 .00 ... \ ,, . ...... z... ..
.00 0000.< :03... 0.00.... .0 .0. a
x .a
.8 05...... «:0 .0 M .0 .20...
.p
.00 0.2.00.0 02mm .p

 

 

 

 

 

 

 

 

 

 

 

 

S3

Variations in soil moisture accentuate the mesh of many of the pat-
terns in the study area and are particularly obvious on CIR photos taken
in the spring. Dry mineral soils exhibit increased infrared reflectivity
and are rendered in hues of white to light cyan on CIR film. MOist soils,
on the other hand, absorb infrared radiation and appear as medium to dark
cyan hues (Stoner etal., 1980). Typically, the mesh of the Saginaw Lowland
patterns is of a darker cyan color than the surrounding ground on CIR

imagery, indicating its higher moisture content (Plate III; Figure 20).

Classification of the Patterns

Many schemes to classify patterned ground have been developed, but the
one most widely accepted employs a purely descriptive terminology, based on
geometric form and the presence or absence of sorting (Washburn, 1956).
This system recognizes circles, nets, polygons, steps and stripes as the
principal geometric forms, each of which can be either sorted or nonsorted.
Nicholson (1969) has proposed an adaptation of this classification based on
three criteria: pattern form, pattern grouping and the manner in which the
pattern is marked. Four configurations (equiform, elongate, stripe or step)
and three classes of grouping (isolate, grouped or contiguous) are recog-
nized. Additionally, variations in either relief, particle size, or vege-
tation are recorded as the mode of pattern marking.

The patterned ground in the Saginaw Lowland falls into the class of
nonsorted nets in Washburn's (1956) scheme because its geometric unit is
primarily ovoid and rather than surficial sorting, microrelief is the
major variable marking the patterns. In Nicholson's (1969) terminology,

these features could be classified as "grouped, relief elongates."

54

 

(NASAPJSC 309-22-143; May 13, 1975)

Plate III. Color-infrared airphoto showing the network of nonsorted
patterned ground at the Lawndale site.

 

 

 

 

 

 

 

 

 

Figure 20. Mesh of the nonsorted nets at the Lawndale site as inter-
preted from the airphoto in Plate III.

 

55

Surface Morphometry of the Saginaw Nonsorted Nets

The mean dimensions of a net cell in the Saginaw Lowland are 380 x
223 feet (rms 127 x 69 ft) and ranged from very large, elongated features
measuring 1,289 x 480 feet to smaller, circular forms 79 feet in diameter.
Most of the mesh cells in the study area are somewhat elongated, as indi-
cated by the 1.7:1 average a/c ratio (range - 1:1 to 3:1). The ratio of
major to minor axis length (a/c) is a measure of the elongation of the
form. A ratio of 1 indicates a more or less circular feature, while
larger ratios are associated with oval-shaped patterns.

Plate IV illustrates some of the variations in the form, size and
marking of the Saginaw nets. The patterns range from large, elongated
features (a) to small, circular ones (b). Although frequently marked by
soil moisture differences (as at a and b), the pattern mesh can also be
outlined by differential crop vigor (c and d), as well as changes in the
natural vegetative cover (e and f). As mentioned earlier, due to temporal
variations in agricultural activities and soil moisture conditions, seg-
ments of the reticulate pattern mesh may not be obvious on a single date
of aerial photographs. As a result, multi-temporal photo-interpretation
is the only practical way to assess the true extent of these features.
Figure 21 presents the results of an analysis of four different dates of
blackvand-white airphotos spanning 18 years for the vicinity of the Bridge-

port wedge site in east-central Saginaw County.

Edaphic Characteristics of the Nonsorted Nets
There are nearly 100 different soil series recognized within the
Saginaw Lowland. They were formed in diverse parent materials ranging
from clay to sand, as well as in organic deposits. The natural drainage

conditions of these soils vary from.well-drained to very poorly drained.

56

 

    

,_ . ,- '7 -» ... Fig.0:
. 1 . .19. .11.: ..-‘ > Mr '
(NASAFJSC 309-22-143; May 13, 1975; Original in color)

Plate IV. Variations in form, size and marking of nonsorted nets in
the Saginaw Lowland.

S7

 

 

1 0-30-37

6-26-4 1

\'..

‘.

i
\

I",
w.’.:"-.'"..~. fi'fi. .

fi)?‘\
.000! \/

  

".:. e°'°.
0 :0. O

 

0-25-50

Reimer Roed a

 

 

... .:‘ .' '. ‘ .I \. ‘4
. ':- ' °°-. .\ .’.-'3
u"; ."-. ~'\’ .
-. . I... .0 . .' "\. z
‘ ....o .
x..v’=.-u., ! z:
. ’.'- ' ' ‘- P.
(4.“ ‘. 1 / \
."' . I I
i "’

Pattern mesh observed on:

A 0| tour sets oi photography
--00 fly“ 0 e 0

two: 0 0

single set only

 

 

W Bridgeport Wedge Site

Interpreted from black end white panchromatic aerial photographs dated:

10-22-55

 

Figure 21.

Delineation of nonsorted nets in the vicinity of the Bridge-

port wedge.

 

58

Soil series differ from one another in the thickness and arrangement of
their horizons, as well as in other physical, chemical and biological
properties. For the purposes of geomorphic mapping, however, the two most
important soil attributes are dominant profile texture, indicative of the
depositional environment (e.g. outwash sand and gravel compared to clay
loam glacial till), and the natural drainage which is suggestive of the
topographic setting (e.g. well to somewhat poorly drained morainic soils
compared to somewhat poorly to poorly drained drift of a lacustrine plain).

Soil management groups, the system of categorizing soil texture and
drainage used in this study, designate these edaphic parameters by a set
of numbers and letters.2 The generalized soil management groups within
the study area are shown in Figure 22; it reveals that most of the pat—
terned terrain is underlain by somewhat poorly drained loam and silt loam
drift or two-storied soils in which a fine-textured substrate is overlain
by coarser materials.

A more detailed analysis of the soil types associated with the promi-
nent patterns shown in Figure 19 was made by plotting their locations on
the available soil maps.3 As shown in Table 5, the majority of these
patterns groups occur on somewhat poorly drained loam and silt loam.soils.
A lesser, but still significant, concentration is found on two-storied
soils in which sandy loam or sand to loamy sand overlies loam to clay loam.

An example of the spatial association between the nonsorted nets and these

particular soil texture/drainage classes is shown in Figure 23.

 

2Summarized in Appendix A.

3Of the soil surveys available for the study area, only Bay County (Weesies,

1980) and Midland County (Hutchinson, 1979) were recently mapped. The re-
ports for Saginaw County (Mahjoory and Whiteside, 1976) and Tuscola County
(Mokma and Whiteside, 1974) are revisions of older data (1933 and 1926,
respectively) wherein various mapping units were renamed.

59

.0000 0000o00o0 300Hw0m 000 mo 0o0000000000 000 000 000o0m 000E0w000a HHom 0003000 00:00oa00a0m

.NN 000w00

 

00:...
c— o
o! .0
on.“ 0200
.00.; .00; ..
00.0 0 00.0 .2 u
a“; a
.00..» .020 ...:v
.020 ...u; ...—.0 .0 a
an ....
00 .m 0
00.... .o
3.. .o 0.. ..
00." .n n
n— .<

000. ...o.00> 0.0.. 00.0000

 

 

 

 

 

0000.0 000500000! :00

 

.00 0.0000...
.00 300.000

.00 >00

.00 0000.2 ....ih 0.000...

.00 000.32 «xv m

.00 0.30.0 0: mo

 

0520 0.5.0...

7%

 

 

 

 

 

 

 

 

 

 

   
 

§

00006 003020..

 

 

 

 

 

 

 

 

 

 

.0000 000000000 300aw0m 000 HO 00H000H000H0 000 000 mmaouw 0008000005 HHom 0003000 000000H00H0m .NN 0uswam

 

59

 

000. £300.. 0.0.. 00.0000

 

 

 

00......
q‘ 1
00
02 ...
00.0. 00.00

.3: .33 ..
no.» 0 no.» .0

00.0 0

.003 .33 .020
60.0 .33 .23 . .0

 

 

a0 ...
00 .m
00.0. .0
00.« .0
00.« .0
00 .<

 

 

 

0030.0 «00.000000! ..00

 

.00 0.000.... .0

 

.00 300.000 .0

 

 

.00 :00 .0

 

 

.00 0000.0 :03... 0.000... .0

 

.00 000.0... a: m .«

 

 

 

 

 

 

.00 0.30.0 0: m0 .0

 

 

 

 

 

 

60

Tab 10 5

Frequency of occurrence of nonsorted nets by soil management

.groups (SMG) in the Saginaw Lowland

 

Z Frequency

 

 

 

 

 

 

.5533 ‘SEQ Soil Series
Study Area 38 2.5b-2.5c
12 2.5b-3/2b
8 2.50
2.5b
4/2b
5 2.5b-4/2b
23 others
Midland County 29 1.5c Lenawee silty clay loam
21 4/2b Wixom loamy sand
17 2.5b-s Ingersoll silt loam
12 1.5b Bowers silt loam
21 others
Bay County 39 2.5b-3/2b Londo-Poseyville
16 2.5c-c—3/2b Tappan-Poseyville
16 4/2b Wixom loamy sand
10 2.5b Londo loam
19 others
Saginaw County 38 2.5b-2.5c Capac-Brookston loans
19 2.5b-2.Sc Capac sandy loam-Brookston
loam
14 2.5c-2.5c-c Parkhill-Kilmanaugh loans
29 others
Tuscola County 54 2.5c-2.5b Parkhill-Londo loams
15 4/2b-2.5b Iosco-Londo association
11 4/2b-5b Iosco-Au Gres sands
20 others

61

$00000 000.03. 0.000000 000 000000 >00 0." 000000000000 0000...» 000000000 000 000000300000 :00

.mN 005000

 

 

00.0500 000.305....
3.100000 .0 000.0

) 0.50.0 000.0300

0... .o 0:5:

00 0 0.4.00 .0100.N 60‘»
00.0 a ouxv .00.p .00 .0I00.u
ca; 0 “5.0 .00 .00

00 d 00 .00.“ ...«s*

00.“ 0 00.“ .nu~Q

00.“ a on.“ 5.0%? ...«00

no a ou~v .00.“

0N\v 0 no.“ .00.“

0.1;? 0 00..“ .00.“

no.“ 0 00.“

0000.0 0008000002 ..00

00:...
O p

 

 

 

 

 

0//<
0 ”HHHMwHHWIAWMHH
A.

 

 

 

 

 

 

0000.0 20.000000... ..00 .0
00223000 .0. < 0.00000< 000

 

 

 

 

 

 

 

 

 

.00 >00

    

'00 “WW"!

 

 

 

9: 0o «:3...
52:90 0030300

£683.00 00 89>
030209 303.305

 

 

61

.000000 000000: 0000000 000 000000 >00 00 000000000000 00000w 000000000 000 000000000000 000m

.mN 000w0m

 

3 a 3:. ....3... .33
3.. a on: .3.. .3 6-3...
3.. a 3:. .3 .3

3 a 3 .3... .33

3... a 3.0 .33

3.0 a 3.0 .3: .33

3 a 33 .30

3:. a 3... .3...

33 .0 3.0 .30

3.0 a 3.0

0000.0 30.300000... =00

 

...
.-
....

 

....

 

 

 

7
.41
a.

 

 

 

 

,0...

 

 

 

 

 

 

 

00:30 30.000000... :00 .0
02.2.0000 .0. < 50:00.; 000

 

 

/H‘Je Wei . r w \x N: ‘

 

   

 

 

 

 

.00 >00

 

(Zita)

'90 ”NIP!"

 

 

 

 

 

62

As a rule, patterned ground is not found on well-drained sites in
the Saginaw Lowland, nor is it associated with coarse-textured or organic
soils. Patterns which occur in areas of coarse surface texture are usually
underlain by finer-textured sediments within less than three feet of the
surface. Most of these two-storied soils consist of lacustrine or eolian
sands deposited over till. Conspicuous patterning in the study area is
rare on clay loam or clay soils.

In east-central Saginaw County, a soil wedge, associated with one of
the nonsorted nets, was observed in cross-section in a ditch along Reimer
Road 3.5 mi east-northeast of Bridgeport (Figure 24). The site is located
on a waterlain segment of the Port Huron Moraine at an elevation of 630 ft
in nearly level terrain with slopes of 11 or less. The local soil type is
Kawkawlin loam.according to the 1933 soils map (Moon et al., 1938); in a
recent revision, Mahjoory and Whiteside (1976) classified it as Capac-
Parkhill loams.

Plate V shows the Bridgeport wedge exposed in the wall of this road-
side ditch adjacent to an agricultural field. Its axis is oriented nearly
vertical and the wedge-form extends to a depth of about 2.1m, but tapers-
irregularly (Plate VI). During excavation, the ground water table was en-
countered 20cm beneath the ditch floor and prevented the exposure of the
basal apex of the wedge. The maximum depth of the wedge (2.1m) was deter-
mined by soil auger borings in the bottom of the pit and is subject to an
error of not more than :8-10cm.

The Bridgeport wedge is differentiated from the surrounding material
primarily on the basis of its texture. Although the wedge perimeter is
irregular in outline, it nevertheless defines an abrupt textural change
from the coarse sediments within the wedge to the finer material of the

host (Plate VII). As Figure 25 indicates, the wedge is composed of a

63

.00003 0000000000 000 00 000000000 000 0000

.00 00:000

 

 

.mhi

010000;

£30009

£30000

.0 _ v. .
J7 .. 0000002000

ad

_
.00.: ...—00.... .0600 _
.000000 32 .0 0.000 ... 00: .

_
2.0 33>. .3302... V.

\ .00...

2.0.3.0... 02.000: 20...... 000.. ‘C

00.00.00
2.0.0.000
00000

O

O

30.000000 0
00:..m m

.3000... :2... 3:3.
002300000
0...: 00.050000

. .9... ......

.22....

 

 

64

 

(tape measure length - 1m)

Plate V. View of the Bridgeport wedge exposure, looking southeast.

65

 

:(tape measure length - 1m)

Plate VI. The Bridgeport exposure showing the irregularly downward-
tapering wedge.

66

 

f ’.

(scale 8 150m)

Plate VII. Lower portion of the Bridgeport wedge showing its irre-
gular, but abrupt perimeter and its basal width exposed
by groundwater piping away the sandy wedge-infill
material.

67

000000000 000... 000 00000000003 0.000 0003 0000000000

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

00000 00000000 .00 00000 00.0 0003000 00000 0000:0000 0000I00000 0>00000800 000000600 .00 000000
A
.0.
A K
n m o o
.0 o o o
no u m u u
I n 0:. 03000
n . w _ u . ... .u . _
><..o ......0 oz<0 ..m><ca 0000000
:55 00.0 0.05
000.0 000.0 00.0 00.0 0.0 0.0 p 0 00 00 000 000
.0 . .
A.
00
.0
o
J
0
o
3 w
.8
m.
u
. a
00 ..
3:000: 0000? m
m.
m
00
2.0.5.000 ~00... .. .
m _ ......Hm.
, 000

 

 

 

 

 

 

 

 

 

68

medium sand (median diameter - 0.4700), in stark textural contrast to
the adjacent sandy clay loam to clay 10am.host material (median diameter
- 0.017mm). Less than 52 of the wedge sediment is finer than 0.5mm, but
the surrounding regolith consists of 50 to 752 silt and clay.

The clay loam parent material of the host contains 12 or less gravel-
size particles, but is, nevertheless, poorly sorted (sorting coefficient
- 8.3)4 and appears to be a glacial till. By comparison, up to nearly 20%
of the well-sorted wedge material (sorting coefficient - 1.8) is composed
of coarse fragments (3 2mm). Neither of these sediments display any bed-
ding or stratification.

Since the wedge site is located downwind (east) from two Grassmere
beach ridges, the possibility existed that these beach deposits were the
source areas of the wedge infill material. To investigate this hypothesis,
particle size analyses were conducted on two samples taken at a depth of
three feet from these littoral accumulations (see Figure 24, p. 63). As
shown in Figure 26, the beach material is a very well-sorted fine sand
(sorting coefficient 8 1.26; median diameter - 0.15mm). Although some of
this fine beach sand could have been transported to the wedge site by
eolian activity, it does not constitute the majority of the infilling sedi-
ment .

The three-dimensional nature of the Bridgeport wedge was revealed by
the continuation of its truncated basal segment for a distance of 1.5m
across the floor of the drainage ditch (Plate VIII). The local strike of

the wedge axis is S 79° W. The wedge pinched out abruptly prior to

 

4The sorting coefficient (So), developed by Trask (1932), is a measure of
the spread of a grain-size distribution; the coefficient becomes larger

as the sorting becomes poorer. It is defined as the square root of the
ratio of the 252 quartile value to the 75% quartile value. Well sorted
sediments have So values less than 2.5, moderately sorted materials have

So values ranging from 2.5 to 4.0 and poorly sorted deposits have So values
larger than 4.0 (Krumbein and $1088, 1963).

69

.000000000 0000 0000000
0003 0000000000 000000 00000 030 0000 00060000 000 m0 000000000000 0000I00000 0>00000a0o .0N 000000

 

 

A
u
A K
O
u m u a
u u m. u u o
u I C 0.530.
w _ o _ 0 _ o T... = _
>30 5.0 . 023 0033.0 0000000
«EB—v .N-O B-Cba
30... 30.0.0.0 00.0 0.0 0.0 0 0 a. 00 2: 000
Z .
8

 

 

CV

,
,
, .
/ 00
,
,
,

 

 

 

 

mogom 59 10mg weaned

 

 

 

 

 

 

 

 

 

 

 

 

 

 

r

 

Plate VIII.

70

 

"H“; '. X‘JV

(tape measure length - 10; scale 8 15cm)

Bridgeport wedge exposed in the ditch wall and the
continuation of its truncated basal section across

the floor of the ditch.

71

reaching the opposite ditch wall and did not crop out in the drainage
ditch on the west side of Reimer Road; it did, however, extend back into
the slope behind its exposed face for at least 3 to 4m, as determined by
auger borings and electrical resistivity surveys. The partial apparent
resistivity curve shown in Figure 27 reveals the presence of the soil
wedge at station 6 and shows it to be a zone of relatively high conduc-
tivity.

In north-central Saginaw County, well-expressed nonsorted nets near
Lawndale, Michigan occur on the same soil types (Capac-Parkhill loans),
with similar topographic conditions (elevation - 630 ft; slope < 20 ft/mi)
and in the same geomorphic setting (proximal slope of a waterlain segment
of the Port Huron Mbraine) as those at the Bridgeport site (Figure 28).
The soil conditions underlying the mesh of one of the largest and most
conspicuous of these patterns (shown at "A" in Figure 20, p. 54) were
studied. As shown in Figure 29, the auger borings made at this site re-
vealed that two-storied soils are dominant within the relatively narrow
mesh zone along the north and west margins of the nonsorted net. The
majority of these samples consist of 30" to 48" of sand overlying 10am to
sandy clay loam till. In several places, the sand layer exceeds five feet
in thickness. 0n the south and east borders of this net cell, as well as
within the area enclosed by the mesh, the soil parent material is loam to
sandy clay loam till.

Figure 30 depicts the soil profiles exposed in the three pit excava-
tions. The thickness of the superjacent sand deposit decreases toward the
west, away from the center of the mesh zone, and its texture grades from a
well sorted, medium sand (sorting coefficient a 1.68; median diameter -
0.28mm) to a moderately sorted, very fine sand (sorting coefficient -

2.5; median diameter - 0.09mm) in the same direction. As exposed in Pit 3,

72

 

Appstsm Rsslsiivity —0-

 

 

tssistivity

 

 

 

 

 

2 a 4 a s 7 "m"
l l L 1 J l l ststion
\ fimf \ \ \‘ x \ .
\
sppsrsnt trscs ‘.‘
9%. oi tits wsdgs‘x
\
«70°: \
‘ -‘— “‘
E
'2
p

 

 

 

N<

 

 

Figure 27. Electrical resistivity profile showing the conductivity
anomaly associated with the subsurface continuation of
the Bridgeport wedge.

 

73

 

 

 

 

. ,G H.3E.
,Ws 2
J" .
. . . . ‘ Bay County
“a . ' Ssolnsv County

 

    

~ Lawndale
' Study Area
NEt/4.NEt/4.. .

PROGLAOIAL LAKE
SHORELINES
(dottsd whsrs tsntstlvs)

A Algonquin E Elkton

 

G Grsssmsrs Ws Wsyns

PORT HURON MORAINE
(wststlsln)

 

 

 

Figure 28. Site and situation of the Lawndale study area.

74

 

 

------

 

 

Dominant profile texture

or sh
grsss I u o... ssnd

   

 

 

 

 

t Iosnt - ssndy clay iosm
. U - D
on cli ssnd 30 48 over
Iosrn - ssndy clay loam
. °®N
I
011 °’
Of
I
I
i r
' |
c s ‘
so ‘ f
. t \
=2 . 1
1:1; 0.
thros i? , o
5:! 0 '0"
P“ I; slsotricsi rssistivity '
excavation” - trsnssot .’
u If - ”My
. 0 <5: ‘ ssntpis 8)
hi
5 s'
=' a so 100
I—d—l—J
§ L toot

 

 

 

Dominant soil profile textures associated with the mesh of

Figure 29.
the Lawndale nonsorted net cell.

 

.00000000 0000 00000000 000 00000 0000800 0:0 00 00000 0
I0000 00:0000000000 000 0003000 0000 00000300 000 00 000000>00x0 000 00050 0:0 00 00000000 0000 on 000 00

 

7S

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

“Wain r «en 0 .w utu 0
l.h
sac. \. .\.|\
0 10.0.0 00000 1 .000. ii.) «0
0 I 00000 \co 0.00. 00.0 0: 0.00. 00.0
nlaL. .. 000001000. 00000 \00 00000 ‘0 0 0
0 I 0-73111 0.0.: 0.. 0. 4
o / o.
P 0 1 v .u
n v I 0000 es. 0.... 0000 m.
.i .anse. \ho 0 \&\is «a
u. 000. 2... 8/ 00: 1 0 )
n” 0 / VIWKIMI 30> % l\
( 0.00m: ll.
0 l
... r 00 00 / .0 1:0! 0
ii; #1 2
ii .‘I’J,liui L p
p I 2
0< 0< 0< 0<
.. 0
o I
...r .0. ".1; .0 . mm .0 [1.1.0.01

~09,
«000

 

 

76

this sand layer pinched-out abruptly along its western edge. Nowhere
within the three pit exposures did the sand deposit exhibit any structures,
bedding or stratitification. The underlying till varied in texture from
loam.to sandy clay loam.and the boundary between it and the overlying

sand was smooth and abrupt, occurring within a vertical distance of no more
than 3cm. Figure 31 presents the results of the particle size analyses of
the eight soil samples whose stratigraphic positions are shown in Figure 30
(p. 75).

Figure 32 shows the resistivity profile along the transect located 3
ft north of the excavations in comparison with the textural data from the
three pits and auger-sample 8. These resistivity data can be interpreted
to indicate: 1) a positive resistivity anomaly is associated with the
sand layer overlying the till; 2) the two-storied soil exposed in the three
excavations and detected in anger sample 8 appears to be continuous between
these locations; and 3) the width of the sand deposit, which is correlated
with the mesh zone of the net, is about 50 ft. The abrupt change in elec-
trical resistivity associated with both the eastern and western mesh mar-
gins is probably indicative of the rapid thinning of the sand deposit at
these locations. The lower conductivity of this material compared to the
adjacent finer-textured till can be attributed to its increased porosity

and decreased water holding capacity.

The Topographic Setting
As a physiographic region, the Saginaw Lowland is characterized by
gentle slopes and subtle local relief. NOtable exceptions to this general-
ization are located in northern Bay and central Tuscola counties where more
rugged topography is associated with the subaerially deposited segments of

the Port Huron Moraine. Other localized areas of steeply sloping terrain

77

.0000 00000300
000 00 000000000 000000 00000000000 000 0003000 00>000 000000000 0000:00000 0300000000 .0m 000000

A
O
a M
u m .0 o
A s o o
.. .. m ... u
w . w _ m . o _ s 00.. _ 00.000
><00 him nz<¢ 0m><¢0 wwamnco
. «0:5 00.0 0.0.0
000.6 000.0226 00.0 0.0 0.0 p D or 00 an: 000

 

 

 

1, 00
VI... XIV-“

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a
O
m
O
m
Orifl. Vi} u...
’ 00 w
Ir 0
.M
00 “I
m
wiry
1V 1 rlllnnlu 00
/ 0.4..
I
0: 00.

N00

78

 

 

 

 

 

 

 

 

 

 

5000
4.6-
4.04
Eiaotrioai toaiativity ttanaoot
aa“
A
s
_ 300‘
N
2’ .
2.5%
2.04
105‘
resistivity promo stations
1.0 I i I .-.-.= T =: I :s T 0
1o 20 so 40,-}- s '- so
a. -‘
'.'.'.'o:.:;: S‘fld ”2
//// Loan to sandy '3
////
_/_/_/_/_l clay loan
-4
Wanner configuration uaod .. 5
for tho raaiativity survey-
aiacttooa apaoing : 14' --6
total array length: 3.5' Auger
Pit 3 Fit 1 Fit 2 sample 8
Av- voltage dtop between potantiai alocttodaa
is total current

 

Depth (toot)

 

 

Figure 32. Electrical resistivity profile across the mesh zone of the
Lawndale net compared to'subsurface textural data.

79

and increased relief are found along the valleys of both the Tittabawassee
and Cass rivers and on the distal margin of the Port Huron Moraine in the
western part of the study area where large dunes are common.

The local relief in the general area of the patterned ground is shown
in Figure 33. The nonsorted nets occur primarily on surfaces of low re-
lief, as summarized in Table 6. 0f the areas exhibiting conspicuous pat-
terns, 832 have no more than 15 ft/mi2 local relief. All of these sites
slope less than 32 and a large proportion of them have less than 12 slope.
Although 17% of the nets are located in survey sections having 20 ft or
more of relief, these patterns usually occur on sites within these areas

which slope less than 32 and are generally flat.

Table 6

Frequency of patterned ground by local relief category

 

 

Local Relief (ft/miz) Frequency (2)
‘ < 10 23
10-15 60
20-30 16
> 30 1

More than half of the patterned ground sites are found on terrain
having 10-15 ft/miZ relief. These areas tend to be composed of somewhat
poorly drained drift, whereas poorly drained tracts usually have less
than 10 ft/mi2 local relief and well-drained sites typically attain
20 ft/mi2 or more elevational change.

Elevations in the study area vary from 580 feet near the shore of
Saginaw Bay to more than 980 feet in the morainic terrain of south-central

Tuscola County. The highest point within the Port Huron Moraine in the

80

.00 V

.flflp i 0—

U 90 I 00
t

......X

A .52:

. luau: 4(000

..00 0.002....
.00 005000

.00 non

...-3b 0.005.—

.00 08838 a: m
figi-g O: 00

0000300 3000000 000 00 0000000 000000000 00 000000 00000

.00 000000

 

11!. 000

  

 

 

 

 

 

0b”

80

.0000300 3000000 000 00 0000000 000000000 .00 000000 0000.0 .00 000000

WIJIJ
0— 0

.32.... .2. @

00A

A .82:
0m_._mm ..(OO._

.00 0.0000...
.00 300.000

.00 >00

.00 0000.2 .030 0.00:...
.00 0020.: w: m

.00 0.3020 0: mm

 

60:90 003030.".

 

81

Saginaw Lowland is in Gibson Township, Bay County (T. 18 N, R. 3 E) where
elevations of 845 feet occur. The nonsorted nets are confined to a much
more restricted elevational range. Patterned ground in the study area is
rare at elevations below 615 feet (Plate IX), nor has it been observed on
terrain above 725 feet. Table 7 summarizes the topographic and pedologic
conditions which appear to have been most favorable for the formation and/or

preservation of the Saginaw Lowland nonsorted nets.

Table 7
Dominant topographic andgpedologic conditions associated
with the patterned_ground in the Saginaw Lowland
Soil texture: loam and silt loam or loam to clay overlain
by sandy loam, loamy sand or sand

Natural drainage: somewhat poorly drained

Slope: less than 32
Local relief: 10-15 ft/miz
Elevation: 615-725 feet

.55; of the Nonsorted Nets

0n the basis of observations in southwestern Bay and north-central
Saginaw counties, Tillema (1972) suggested that the patterned ground was
confined to the proximal slope of the Port Huron Moraine. More recently,
Lusch (1977) indicated that the nonsorted nets may be associated with the
waterlain segments of this moraine. Detailed mapping of the distribution
of both the patterned ground and the moraines throughout the Saginaw Low-
land has shown that both of these hypotheses are only partly correct.
Patterned ground does occur on the waterlain portions of the Port Huron
Moraine in the central part of the study area, but most of the nets are

on the inundated till plain flanking the proximal margin of the moraine

82

   
 

(NASA-JSC 309-22-143; May 13, 1975; Original in color)

   

Plate IX. Aerial view showing the lack of patterns below the 615 ft
contour and conspicuous forms occurring only above 620 ft
(Lake Elkton shoreline).

83

(Figure 34). Furthermore, two smaller areas of patterned ground are loca-
ted beyond the distal edge of the Port Huron Moraine in northeastern Mid-
land County. This spatial distribution indicates that the patterned ground
post-dates the Port Huron readvance, i.e., it is younger than 13,000 yrs
B.P. (see Table 3, p. 40). The similarities between the distal and proxi-
mal nonsorted nets and those which occur on the Port Huron Moraine suggest
that all the patterns formed penecontanporaneously.

A more precise estimate of the age of the patterned ground may be de-
rived from an analysis of the spatial relationship between the limits of
patterning and the shorelines of the proglacial lakes, a method which was
successfully used in southwestern Sweden (Svensson, 1973). thing the
conformity between the lower limit of patterning and the elevation of Lake
Elkton and the occurrence of nonsorted nets above and below the Lake Grass-
mere shoreline, Tillema (1972) concluded that the nonsorted nets were post-
Lake Grassmere in age and had developed during the presumed BOO-year
existence of Lake Elkton.

Recent studies of the Late Wisconsinan chronology of the Great Lakes
suggest that less than 350 years elapsed between the formation of Lake
warren I and the establishment of Early Lake Algonquin (Calkin and McAndrews,
1969; Clakin, 1970; Karrow et al., 1975). Six separate proglacial lake
levels have been identified within this time interval: warren I and II,
Wayne, warren III, Grassmere and Elkton. Obviously, the post-Grassmere
period must have been considerably shorter than 350 years which brings
into question such an age for the Saginaw Lowland nonsorted nets.

The congruence of the lower limit of patterned ground and the Elkton
shoreline is unmistakable (Figure 35). The shoreline of Lake Elkton has
been deformed by isostatic recovery in the northern parts of the study

area from its horizontal elevation of 620 feet upwards to 640 feet. The

84

0000300 3000w0m 0:0 00 0000 000000000 0:0 000 000000: 00005 0000 0:0 .00 0000000000000 0000005

.00 ousmfim

 

050.0! :05: tom

.o:=:ou=.oto=3
02.000 300.503

0000.0 00:00:00

 

or

 

 

 

 

 

 

 

 

.00 0.002....

.00 300.000

.00 >00

.00 00:02 :03... 0.00:...
.00 0020.2 0: u

.00 0.30020 tsp ma

:0!

.0

.w

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

84

 

 

as 114 aladW'" °°'

1.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

10

 

 

. 13 13"
o :1 19 0
0 3 .3}:
o O ”.5
0 (9-‘0 0
o 0 'U
0 I. .0 fi :
‘0 - ° 5 :2 :n 0
c o o '0
:2 r- . a: .2 5: :34:
2 0 ° ' O .n 3
" 0 .5 O N V'
u 9 a C O-
\ O ’ . a
' ‘ ' o F
3 u
I“ » °
I . o
a Q
.
%%%%¢
990000
009909
0000 o.
‘pooooo.
,‘ggapooo‘
« ‘ O O . 0
«909990.
”‘90...
. O 6 O 6 o
oo¢%%
. 0 §
\ . . . O \
‘0 O O O O 0 s
. 9090
J 90999
,‘%%&&p .
.g gaapo ‘9‘.
00 9d?“
‘ 6 o o o 9’
‘0 006’
\
A ' v ‘
ysfifi
odv&
od§$$Q
‘od%%9
QQQ’
.0.
u&&
gc
: "O
«3:...»
, 9099
.v 0990
$00 990.,
I - v V . . ,’
‘ .9 0 ¢
{6:03. ‘ . . 9'09 0 s o“ ’6‘... O, ’
...cn.‘%%u..‘awvvvw /”’”"
3ggugg¢§§$§QQQ§ono €%%%&$’
‘gggggg‘g%~%&&podV5és d%%%%%’
-oooooo ......avamnaw~0¢‘0¢0”
009999090990 ‘0\oooo “““Q
\ 60......9999 9.. O O ( ‘ ‘ O O 0 0.0.0.6.‘ "A ‘
‘gga’ ‘Nfifis aeearaaw'
.g¢39 aflufib’
“1'4 '
$$$§8

 

 

 

 

 

 

 

Lowland'
Saginaw

1n the

ted net8

I e 1101130r

H ron “braine d th
I: u

1but1°n °

1 d1Btr

85

.000003 00—00 000 0000.0... 00.00 .00 0000000050 0:0 000 00000w 000000000 3000mmm 0:0 00 0000000000009

.mm 000000

 

if.

830 3.0.: m
00:03 9.00”?
i
9.0.. 38.02..

30.0.0000. 0.003
02.000 3000000...

0000.0 00000000.".

 

Op

0

 

T3

 

 

 

 

 

 

.00 0.00000 .0

.00 305000 .0

.00 000 .0

.00 0000.2 :03... 0.0003 .0
.00 000.32 a: m d

.00 5300.0 0: ma .0

 

 

 

 

 

 

 

 

 

85

 

 

SE 114 Giadwin Co.

1.

E 112 Midland Co.

‘Linooin 109.. Avenue Co.

4. Day Go.

 

 

 

 

Saginaw Go.

a. 'l'uoooia 0o.

 

 

Patterned Ground

(boundary dashed

where indeiinite)

 

 

 

 

 

 

 

 

 

 

10

 

 

 

 

 

 

 

 

 

 

 

Distribution of the Saginaw patterned ground and the shorelines of Lake Elkton and Lake Warren.

Figure 35.

86

lower elevational limit of the patterned ground follows this deformed line,
indicating that the nonsorted nets were actively forming during the exis-
tence of Lake Elkton and that the patterns were uplifted along with the
Elkton beach by post-glacial rebound.

The upper elevational limit of the patterned ground, marking the be-
ginning of its formation, is more complex to interpret. Virtually the
entire length of the upper limit shown in Figure 35 is strongly influenced
by topographic (slope and relief) and/or edaphic (texture and drainage)
factors, as discussed earlier. One spatial constraint stands out, however:
patterned ground in the study area is extremely rare above the Warren I
shoreline which, due to isostatic uplift, ranges in elevation from 685
to 730 feet.

0n the basis of this evidence, it can be said that the nonsorted nets
in the Saginaw Lowland are younger than 13,000 yrs B.P. (i.e., post-Port
Huron readvance), but the majority of the patterns may post-date the
warren I proglacial lake stage, which has a radiocarbon age of 12,730 i
220 yrs B.P. (I-3665). The cessation of patterned ground formation appears
to have been contemporaneous with the fall of the proglacial lake in the
Saginaw basin from.the Elkton level. Since the next lower lake level,
that of Early Lake Algonquin, came into existence about 12,400 yrs B.P.,
this age may be taken as a minimum date for the termination of patterning
in the study area. Hence, the nonsorted nets in the Saginaw Lowland
appear to have formed during a period not longer than six centuries imme-
diately following the Port Huron readvance, but many of them may have
formed in less than 350 years after Lake warren fell from its highest

level and before Lake Elkton drained.

CHAPTER IV
PROPOSED ORIGIN OF THE SAGINAW LOWLAND NONSORIED NETS

Examples of patterned ground have been reported from many different
environments of widely varying thermal and moisture regimes. The iden-
tification of inactive or fossil patterned ground can be of great value
in environmental reconstructions and paleoclimatic studies (Washburn,
1973). As summarized by Washburn (1956, 1970), the genesis of most pat-
terned ground is problematical because 1) similar forms of patterned
ground can result from different processes; 2) dissimilar forms can be
produced by a single process; and 3) there are more proposed genetic
processes than there are recognized types of patterned ground. The fol-
lowing review separates the numerous published hypotheses of patterned
ground formation into processes in which cracking of the surface regolith
is the essential and immediate cause of the pattern and those in which
cracking is non-essential.

Review of Suggested Mechanisms Not Requiring
‘gggolith Cracking
Erosional Processes

Presant and Protz (1967) suggested that some types of nonsorted pat-
terned ground may be caused by depressions in the B horizon of the soil
which were related to channel scour in braided streams. They found that
trends in the micro-relief of the B horizon surface were coincident with
major landscape trends and attributed these coincidental trend directions
to fluvial erosion. An undulating B horizon could affect the water
holding capacity of the soil and produce nonsorted patterned ground as a

result. This hypothesis must be rejected as the origin of the nonsorted

87

88

nets in the study area because 1) the trends of the patterns are not co—
incident with topographic trends; 2) the depth of the B horizon surface
is not consistently depressed along the net mesh; and 3) this mechanism
cannot produce unstratified soil wedges such as the one at the Bridgeport
site.

Morgan (1972) considered surficial drainage along fractures in till
as a possible origin for nonsorted polygons. These fractures were related
to the ice movement and some of the observed patterns were aligned nearly
parallel to this direction. In the Saginaw Lowland, this origin is un-
tenable because trends in the net mesh are unrelated to either ice flow
direction or surface drainage. Additionally, fluvial processes could not
account for the narrow, unstratified soil wedge such as that exposed at

the Bridgeport site.

Pedologic Processes

The similarity between soil tongues and certain types of patterned
ground was the first reported by Yehle (1954). More recently, White
(1971) and Byrne (1975) have investigated similar phenomena in South
Dakota and Ontario, respectively. These pedologic features result from
differential solution by percolating soil water and are common in sandy
gravel deposits. They are also well developed in the Fox soil series
which is composed of loamy sediments overlying sand and gravel. The
Saginaw patterns, on the other hand, formed on soils having much finer
dominant textures. Soil tongues and related features can have a surface
expression, but their mesh dimensions are usually less than 10 feet
across, in contrast to the significantly larger cell diameters of the
nets in the study area. Additionally, these in situ edaphic processes are

incapable of producing the significant textural differences between the

89

boat material and the wedge infilling which were observed at the Bridge-

port site.

Bedrock Control

Striking examples of nonsorted patterned ground have been reported
from southern England (Perrin, 1963). Here, polygonal and elongated pat-
terned ground was formed on thin drift underlain by chalk. The surficial
patterns were associated with a ridge-and-trough micro-topography on the
upper surface of the bedrock which was attributed to periglacial distur-
bances. MOst of the drift in the Saginaw Lowland is more than 100 feet
thick and it is extremely unlikely that bedrock micro-topography or
jointing would be revealed as surficial nonsorted nets through such a

drift thickness.

Glacial Stagnation Processes

The geometry, size and marking of the Saginaw pattern network show a
certain similarity to ice stagnation landforms such as those reported by
Gravenor and Kupsch (1959) and, even more so, to the photo pattern of hum-
mocky ground moraine discussed by Parizek (1969). That the patterned
ground in the study area is unrelated to glacial stagnation processes is
demonstrated by its distribution distal to, on the surface of, and proximal
to the Port Huron Moraine, as well as the lack of any other landform indi-
cators of ice stagnation in the region. The low relief of the Saginaw
patterned ground and the conformity of its lower elevational limit with a
proglacial lakeshore also argues against a stagnation origin for these

features.

90

Review of Suggested Mechanisms Requiring
Regolith Cracking

 

Thawing

Thawing of regolith, whether seasonally or perennially frozen, has not
been sufficiently investigated as a potential cause of nonsorted patterned
ground but, according to Washburn (1973), this mechanism lacks convincing
evidence. Black (1976) has suggested that thawing of buried ice blocks
could produce nonsorted patterns. These "kettle cracks" form in the out-
wash which may inundate stagnant ice blocks. All the patterns observed in
the Saginaw Lowland occur on much finer drift and their extent and geo-

metric regularity are inconsistent with this proposed origin.

Synaeresis

According to Kostyaev (1969), synaeresis cracks can form polygons
which are tens of meters in diameter. These fissures, resembling mud
cracks, develop as water is expelled from a clay-rich suspension by in-
ternal forces. Synaeresis cracks probably determine the location of desic-
cation fissures but, as noted by Washburn (1973), they are relatively un-
important in forming nonsorted patterned ground. The larger size of the
pattern mesh in the study area also precludes this mechanism as a possible

origin.

Gilgai Development

Costin (1955) noted the striking similarity of surface patterns pro-
duced by gilgai development and classic frost-soil patterned ground.
Gilgai soils form.in response to volume changes resulting from desiccation
and rehydration of regolith having a very coefficient of linear extensibi-
lity (Hallsworth et al., 1955). The swell potential of most soil series

in the study area is usually less than 52, well below the 30% swell

91

capacity of typical gilgai soils. As a result, these Michigan soils are
incapable of the shrinkrswell volume changes necessary to produce large

nonsorted nets.

Partial Wetting

 

Cracks formed by partially wetting regolith samples during laboratory
experiments were reported by Corte and Higashi (1964) who suggested sur-
face tension as their origin. The crude patterned ground formed by this
unusual phenomenon is, however, of small mesh diameter, unlike the patterns
in the study area. As with synaeresis cracks, these “wetting cracks"
probably determine the location of subsequent desiccation fissures, but are

unrelated to the Saginaw nonsorted nets.

Desiccation

 

One of the most common causes for the evolution of patterned ground is
regolith contraction due to drying (Washburn, 1973). In fact, most well
developed nonsorted polygons less than three feet in diameter are probably
due to desiccation cracking (Tricart, 1969). There is also ample evidence,
however, that desiccation can produce large-diameter patterned ground
(Land, 1943; Wilden and Mabey, 1961; Chico, 1963, 1968; Neal and Motts,
1967; Neal, Langer and Kerr, 1968; Neal, 1972).

The similarity between the surface patterns produced by desiccation
cracking and thermal contraction in permafrost has been noted by Knechtel
(1951) and Black (1952a), but evidence of the long term, severe drought
conditions necessary to form large dehydration polygons is lacking in
the Saginaw Lowland. Quite to the contrary, most of the patterned ground
in the study area occurs on somewhat poorly drained soils and it is likely

that similar or more poorly drained conditions existed during the waning

92

phase of the Late Wisconsinan, when several proglacial lakes occupied the

Saginaw basin.

Dilation

Dilation cracking, resulting from local differential heaving or sub-
sidence of the ground, is known to produce patterned ground (Benedict,
1970; Washburn, 1973), but crack networks of this kind have limited spatial
extent and small mesh dimensions. Clearly, the large size and number of
the Saginaw nets preclude dilation cracking as a cause for the patterns in

the study area.

Salt Cracking

Large, nonsorted polygonal fissures in hard rock salt and desert salt
crusts have been reported by Hunt and Washburn (1966). Although not com—
pletely understood, the genesis of these salt cracks seems to be thermal
contraction. Considering the lithologic material involved, this process
is obviously inapplicable to explain the reticulate patterns of the Saginaw

Lowland.

Seasonal Frost-Cracking

Seasonal frost-crack polygons result from thermally induced contrac-
tion fissuring of seasonally frozen ground (Washburn, 1973). Active exr
amples of this type of patterned ground have been reported from both arctic
(Hopkins et al., 1955; Danilova, 1956; Friedman et al., 1971; Katasonov,
1973) and mid-latitude sites (washburn, Smith and Goddard, 1963; Washburn,
1973; Black, 1976). There is some question, however, whether frost cracks
in a seasonally frozen layer can be preserved in the geologic record, but
Dylik (1966) and Pissart (1970) concluded that certain types of very nar-

row soil wedges are the pseudomorphs of seasonal frost fissues. Washburn

93

(1973) contends that fossil forms of these seasonal features would docu-
ment such deep seasonal freezing that permafrost conditions were likely.
It is doubtful that the large-diameter reticulate pattern of the
Saginaw nonsorted nets could have been formed by seasonal frost cracking,
because the polygons produced by this process are usually less than 50
feet across (Shumskiy and Vtyurin, 1966). The Bridgeport wedge lacks
the vertical bedding which Pissart (1970) considers to be characteristic
of fossilized seasonal frost cracks and it is also much larger than the

seasonal frost-fissure pseudomorphs reported from.Europe.

Permafrost Cracking

 

NOnsorted patterned ground, formed by thermal contraction-fissuring
in regions of perennially frozen ground, is one of the most common peri-
glacial surface forms. According to Washburn (1973), this type of pat—
terned ground is the most likely to be preserved and its fossil forms are
of considerable significance because of their temperature implications.
Thermally induced permafrost-cracking has been recognized since the early
19th century, but it was another hundred years before Leffingwell (1915,
1919) was able to scientifically explain the details of the process.
Lachenbruch (1960, 1961, 1962, 1966) demonstrated that Leffingwell's
hypothesis was theoretically sound and it has also been supported by field
evidence (Kerfoot, 1972; Mackay, 1974).

Leffingwell proposed that, during the winter, thermal contraction of
the regolith produces vertical fractures several millimeters wide which
penetrate a few meters into the permafrost (Figure 36a). By the subse-
quent fall (Figure 36b), these cracks may be filled with meltwater, snow
and/or hoarfrost, resulting in an ice vein (Mackay, 1975). Horizontal re—

expansion of the warming permafrost during the summer can deform the

Permairoat table >1 .. . .

OPEN cancx— 3

94

 

" < Permatroet table

 

 

 

 

 

005

5 .~dianiicton

 

b.

 

 

Pematroat table >. ,

open GRAOK‘ . '

C.

tat Winter

frozen

. - ' Active
’ f“ lav-t» * .

 

 

”WOO! tONO

 

 

 

 

Figure 36.

 

 

/ '. “ICE

 

d.

 

tron! Lachenbruch. 1062

Thermal contraction-crack Origin for ice wedges in permafrost.

95

frozen ground strata adjacent to these ice veins. These incipient ice
wedges are zones of weakness in the permafrost and during subsequent
freezing cycles renewed thermal tension will re-open the vertical, ice-
cemented cracks (Figure 36c). Mature ice wedges result from a repetition
of such cycles over a long period (Figure 36d). The natural surface form
of tension cracks such as these is almost always polygonal and in active
permafrost areas ice wedges underlie the polygonal mesh (Figure 37).

During the thermal degradation of permafrost, the enclosed ice wedges
thaw or sublimate and the space they occupied can be replaced with adja-
cent, overlying or transported material (Pébé, Church and Andresen, 1969).
Ice-wedge casts formed in this manner are often composed of sedimentary
material which differs in texture, color and/or fabric from.the sur-
rounding host material. Although ice wedges are best developed in satu-
rated (i.e., high ice content), fine-grained soil having little primary
structure, these same edaphic conditions make it very difficult to pre-
serve any evidence of their former presence (Black, 1976). The melting
ice in the saturated permafrost promotes the rapid flowage and slumping
of the fine-textured host sediment which can obliterate the wedge form if
the ice wedge has not been replaced by transported material.

Johnsson (1959) presented the following criteria by which true ice-
wedge casts could be identified:

1) The infill material of the supposed cast must have been deposited

from above.
2) The host material surrounding the wedge case should display dis-
torted strata due to pressure effects.
3) Elongated clasts in the host material should be preferentially

aligned parallel to the axis of the pseudomorph.

96

‘r
1

 
 
  
   
     

:: ‘ § A . U ’ ' _
0 . : -— ‘5": ._ —_‘—-
Q .0 fl — .
~g o f...
.2; _/- NEW-NH - . /
\
~ -r ,..~. -.-.~- 0,, 3.». ,
‘3 f A .06.”... .d'." a. '0 C I
O ‘ ~ . ‘e
I

.. '5'}: M'-'-;'f'.*.’-f‘:'??tf§:w‘—~ —— ’ /' “I.
e‘ ' ‘23 I /'/

0'
a

e \
_‘ C. . .._ £5é”.”"'/'"!Il . ‘“\!w.-.'- - My {.1 . . /. i.
\‘ : .e- ..R‘.‘ L- d.‘-lfl<$<1£«d{-‘2.W ’.——\ ‘. ../‘/. .' /

=0,—

 

  

 

 

 

 

 

 

 

Figure 37. Relationship of ice wedges to ground patterning in
active permafrost.

97

4) The ice-wedge cast should be triangular in shape, widest at the

top and tapering with depth.

5) True ice-wedge pseudomorphs stand more or less vertically.

Other useful criteria for correctly identifying ice-wedge casts were
suggested by Black (1976). He called attention to the need for supporting
evidence of perennially frozen ground, such as a fossil permafrost table,
and stated that biological indicators of frozen ground were only sugges-
tive, not diagnostic. Black also stressed that the host material should
be of a type that would be supersaturated under normal permafrost condi-
tions and that multiple wedge casts should be present which form polygons
in the plan view. Black reiterated the views of Johnsson (1959) with re-
gard to the preservation of pressure effects and realigned clasts in the
host material. Additionally, he proposed that slump fabrics in true ice—
wedge pseudomorphs should show arcuate stratification (concave upward),

in contrast to the near-vertical fabric of sand wedges.

Evidence Supporting An Ice-Wedge Origin for the Saginaw Nonsorted Nets
Data from the preceeding chapter strongly suggest that the Saginaw
Lowland patterned ground was initially formed by permafrost cracking and
ice-wedge development. The geometry of the reticulate nets in the study
area, for instance, is analogous to the surface form of modern permafrost
fissures (Aartolahti, 1977). The average size of the Saginaw patterns
(380 x 223 ft) is similar to those of ice-wedge polygons reported from
elsewhere in North America and from.northern Eurasia. As shown in Table 8,
active and inactive tundra polygons average 34 to 767 feet in diameter, inr
eluding both small features (7-10 ft in diameter) and very large examples
(>3,000 ft across). Most of the fossil polygons discussed in the litera-

ture are smaller than the latter, with average diameters of 37-112 feet.

98

Table 8

Reported sizes of active, inactive and fossil tundra_polygons

 

Location

Diameter (ft)

Reference

 

a) Active and Inactive Structures

b)

south of Barrow, Alaska

 

Vorkuta region, USSR

Bolshezyemielskay tundra,
USSR

 

near Barrow, Alaska
initial polygons
secondary polygons
tertiary polygons

 

northern Sweden

Fossil Structures
southern Finland
northern Great Plains, USA
southern Sweden
Quebec, Canada
northern Norway
Jutland, Denmark
northern Germany
central New Jersey
west-central Indiana
west-central Indiana
central Iowa
Lappland

eastern England
northern Norway

southern Alaska

southern England

10-200
164-656

82-3,280
10-3,280

10-100

33-328
13-26
3-10

10-328
7-100
33-131

7-43
50-200
115-213
66-197
13-108
16-33
7-49
10-100
50-100
49-197
4-20
33-131
7-35
131—213
7-98

25-50

Hussey and Michelson, 1966
Tricart, 1969

Dylik, 1966

Dylik, 1966

Brown and Kupsch, 1974
Black, 1952a

Lachenbruch, 1966
Pebez 1966a
Rapp and Clark, 1971

Aartholati, 1970
Clayton and Bailey, 1970
Johnsson, 1963; 1981
Lagarec, 1973
Svensson, 1964a
Svensson, 1972
Svensson, 1976
Walters, 1975; 1978
Wayne, 1963

Wayne, 1967

Wilson, 1958

Rapp and Rudberg, 1964
Williams, 1964
0hrengren, 1967

Pebefl Church and Andresen,
1969

Gruhn and Bryan, 1969

 

99

However, large fossilized structures, approximately 200 feet in diameter,
have been reported from the northern Great Plains, Indiana and Quebec,
as well as from southern Sweden and northern Norway.

The relief-marking of the Saginaw patterned ground (low mesh, high
center) is consistent with the hypothesis that it resulted from the ther-
mal degradation of a network of ice wedges (Christensen, 1978; Washburn,
1978). This aspect of thermokarst modification of the patterns during
their waning development will be discussed in more detail in a subsequent
section.

On the basis of its structural and sedimentological characteristics,
I interpret the soil wedge at the Bridgeport site to be an.icerwedge
pseudomorph which documents the former existence of permafrost in the
Saginaw Lowland. The medium-sand texture of its infilling material
clearly shows that it could not have formed pedogenically from the sur-
rounding sandy clay loam to clay loam till and the abrupt boundary between
these two sediments suggests that the sand was deposited from above into a
pre-existing wedge-shaped void in the drift. Although it lacks stratifi-
cation, this sandy infill material is presumed to have been deposited by
running water in a lowbenergy environment, augmented by eolian processes.

Evidence of pressure-effect deformation structures is lacking at the
Bridgeport site, but this is undoubtedly the result of the massive, un-
stratified nature of the till. A similar situation was reported by Morgan
(1982) for ice-wedge casts near Muir, Ontario. The relative paucity of
pebbles and cobbles in the drift surrounding the Bridgeport wedge made
it difficult to assess whether any realignment of elongated clasts existed.
The long axis of one roller-shaped pebble in the till adjacent to the basal

section of the pseudomorph was oriented parallel to the strike of the wedge

100

cast, but another elongated clast not far away was oriented transverse
to the wedge axis (Plate X).

The downward-tapering wedge form and near-vertical axis orientation
of the Bridgeport wedge are both characteristic of true ice-wedge casts.
No conclusive evidence of a relict permafrost table was observed in the
study area, but palynological data from elsewhere in Michigan (to be dis-
cussed in a subsequent chapter) indicate that tundra conditions, and
therefore possibly permafrost, were penecontemporaneous with the onset of
pattern formation in the Saginaw Lowland. -

The somewhat poorly drained loam and silt loam soils underlying most
of the patterned ground in the study area would have been saturated with
ice when they were perennially frozen during the Late Wisconsinan. Mul-
tiple wedge casts are lacking in the Saginaw valley, but the Bridgeport
wedge is directly associated with a surficial nonsorted net. The overall
paucity of fossilized ice-wedge casts in the study area is, nevertheless,
in accordance with the textures of the host sediments which are prone to
flowage upon thaw, an edaphic condition which greatly reduces the opportu-
nity for ice-wedge replacement.

The large mesh dimensions (crack spacings) and irregular net form of
the Saginaw patterned ground are indicative of permafrost cracking which
was primarily controlled by randomly distributed flaws in the regolith.
This type of cracking pattern is typical of recently drained areas of
permafrost terrain (Lachenbruch, 1966) and is, therefore, consistent with
the late-glacial drainage history of the study area. Oriented, orthogonal
patterned ground, with one set of cracks parallel to a shoreline and a
second set perpendicular to it, is often associated with slowly draining
lakes on permafrost. With rapid shoreline recession, on the other hand,

no such orientation would be expected (Lachenbruch, 1966). The short time

101

 

(scale - 15cm)

Plate X. Basal section of the Bridgeport wedge exposed in the ditch
floor showing the orientation of two elongated pebbles.

102

duration for the lake level reduction from Warren I to Elkton is corro-
borated by the lack of oriented patterned ground in the Saginaw Lowland.

Based on the Late Wisconsinan chronology of the proglacial lakes in
the study area, on the one hand, and the spatial agreement of some of
their shorelines with the elevational limits of the patterned ground, on
the other, I suggest that the formation of the nets occurred within a
time span of 350 to 600 years (refer to discussion on p. 37). The large
size of the nets in the Saginaw Lowland may be explained by this rela-
tively short period available for their formation, since smaller patterns
are usually the result of repetitive subdivisions of primary fissures
during a protracted period of permafrost cracking (Dostovalov and Popov,
1966; Dylik, 1966). The width and depth of the ice-wedge cast at the
Bridgeport site also suggest short-lived permafrost conditions. Black
(1952b; 1974) determined that ice-wedges on the arctic coastal plain of
Alaska were growing horizontally at the annual rate of l-3mm/year and
similar rates were measured by Mackay (1974) for ice-wedge cracks on
Garry Island, Northwest Territories. Although these rates cannot be
directly extrapolated to the Saginaw Lowland during the Late Wisconsinan,
they nevertheless indicate that the ice wedge documented by the Bridgeport
pseudomorph (ca. 15cm wide) could have formed in the 350 year interval be-
tween Warren I and Elkton. The dominant occurrence of the nonsorted nets
on the finer-textured soils of the study area also suggests a relatively
mild thermal regime in the permafrost (Dylik, 1966).

The spatial correlation between the Elkton shoreline and the lower
elevational limit of patterning indicates that most of the nonsorted nets
formed subaerially. Patterned ground, resulting from permafrost cracking,
can form beneath shallow water as well and has been reported from various

places (Mackay, 1967; Kerfoot, 1972; Danilov, 1973; Shilts and Dean, 1975).

103

At several locations in the study area, faint patterned ground, which may
have formed subaqueously, occurs at elevations slightly below the Elkton
level on sites interpreted as spits or offshore bars formed in Lake Elkton
(Plate XI). If these nonsorted nets did form underwater, the cracking
must have initiated within subaqueous permafrost rather than at the lake-
ice surface because the latter situation could not allow the recurrence of
the identical geometric pattern from one year to the next (Mackay,
Konischev and Popov, 1979).
Thermokarst Modification
of the Saginaw Patterned Ground

The subtle morphology and sedimentological characteristics of the
nonsorted nets throughout the Saginaw Lowland suggest that thermokarst
erosion,1 rather than ice-wedge replacement, was the dominant geomorphic
process associated with the degradation of the permafrost in the study
area. The morphologic importance of thermokarst in areas of perennially
frozen ground has been emphasized by Kachurin (1962), Svensson (1970),
Aleshinskaya, Bondarev and Gorbunov (1972) and Soloviev (1973).

Thermokarst develops in response to the disruption of the thermal
equilibrium of permafrost by either climatic (e.g. increase in termpera-
ture, precipitation and/or continentality) or nonclimatic (e.g. forest
fires, agricultural activities, construction, etc.) factors (Czudek and
Demek, 1970). Because of the greater ice content in perennially frozen
soils having impaired drainage, thermokarst is typically best developed
in lowland environments. French (1974) attributed regional thermokarst
to a climatic amelioration which furthurs a progressive lowering of the

permafrost table; this appears to have been the case in the Saginaw

 

1Thermokarst is the process of melting ground ice accompanied by local
collapse of the surface and the formation of depressions (Muller, 1947;
Washburn, 1973; Brown and Kupsch, 1974).

104

 

“=2qu :

.1." ..»

(NASAFJSC 309—22—156; May 13, 1975; Original in color)

Plate XI. Conspicuous patterns confined to terrain above the Elkton
beach (620 ft) near Reese, Michigan; faint nets (arrows)
are associated with an Elkton offshore bar.

105

Lowland as proglacial Lake Elkton drained during the Late Wisconsinan.

The thawing of ice-wedge polygons within permafrost can produce cir-
cular or ovoid mounds2 separated by curvilinear depressions which are
situated over the melting ice wedges. This type of thermokarst topo-
graphy develops in stages, as the active layer thickens (Schumskiy and
Vtyurin, 1966; Czudek and Demek, 1970; Soloviev, 1973). In the initial
phase (Figure 38a), high—centered polygons form as the reticulate network
of ice veins begins to thaw. These trough—like depressions grow succes-
sively deeper as the thermal erosion of the wedge ice continues. The
polygon cores usually retain their initial elevations and their flat tops
until the encircling troughs reach a depth of about Lm (Figure 38b). This
amount of local relief is generally sufficient to induce slumpage and/or
flowage of the unstable active layer which tends to smooth-out the topo-
graphy (Figure 38c). In addition, the melting of interstitial ice in the
permafrost can obliterate the deformation structures frequently associated
with ice-wedge development (Dylik, 1968).

According to French (1975), the amount of ground ice in saturated
permafrost usually exceeds the liquid limit3 of the regolith, making it
prone to liquefaction upon thaw. This type of rapid gelifluction probably
had a significant geomorphic impact on the landscape of the Saginaw Low-
land as the climate ameliorated during the late Port Huron Stadial (early
Twocreekan Interstadial ?). For example, the liquid limits of the parent

materials of the four soil series commonly associated with the Saginaw

 

2These inter-trough mounds are called baydjarakhs or cemetery mounds in
Siberia (Brown, 1967) and thermokarst mounds in Alaska (Pewe, 1954).

3The liquid limit is the percent water content of unconsolidated material

at the point where it passes from a plastic solid to a turbid liquid
(American Geological Institute, 1962).

106

 

 

 

Ice wedges

Permstrost tsbl

 

 

 

 

.v".
.. e..'.:.'“o
... e . ... . .. 0.
. e" . . . 'd’... ' .....se”...
.'-‘--.'" '.- . -,::'.-'.=-'--.?:-‘.'..'- -. , ,v'.
.'.' ..e- 0. .l O. .. ... ‘e.".. .' .... e . '0 ....' . e0.
- '. cs '. ... . 1:” .....'e ‘- .... .... ..."... 7......; O . . ' 'gf’... ." . .- ....s. 3.. '.e" be. ....
e.'.. .'. "3 I: O "s’I‘I... . . ‘v.:'e...":‘..“v...'- ' .. Q. ..... e "" e’. ...'o..' ' 00...... " . ... .eO
_ . . _ e . e . g. ‘ s s . . I . .s . . ' ' .

.'.':.:.;‘e'.".-.. ..;;....e '..'I. ......19. : ....r;.ee...;.. '..' k"...‘.' ’f '. a... 't’- .;":' 3.... . .... 'o..sflr.';e-..'e..e.....‘v‘. (...-"1.33““
0; ... 'e-‘ . .' o-0".".'.3s.' °" ‘s'..'e 'e-‘. e. ""-‘-"-"'eo"l '--. ..' .0 " " "... H "- ' """
‘ " O.’ :". ' . .... .' . .' Q .0 0' . e. ... .3, .... ..; 0' O . ' '. n ... O'... ...e. ... '0‘. 0.... H 00" .... 00's .' .
-~'.--.--‘--'.~s-.. ..'-"-.-..':' -- "‘.-¢"'.-2 -'.'-- 'I‘m - ‘.'.‘--..'-~'.'.'.".-' 1’? '- -’ ' --' ' "f-
, "o" ...-.053. . "" '0'}. ‘3. .: 2'55”: . t;:’ '.".°.'O-'.:.." - ‘.:‘-.|'!..': -".. 3." «Zoo-'3'. .‘ol..'., :b.‘ :e'... '...'.'.-.':.'...' ?':~.-.'.“-'. '..‘.'e'.
..5 . .;.‘. L .. ... ......0 . . ... ...fs. .‘.-'. I... ..e...; ‘..‘.... .r. ...‘.l'.."..t.. ....“ ...... .."...’. . ..M...'..I.‘ .. II...s.e..: .... .....sv 2..
-:.'o..‘. “..‘l . ‘ :5. ° .:.-.-‘ ”:"a. ‘°.‘v.’."-'.-.'."-.'-'-."."’--~..‘°~o I.‘ . °.-'-'-'."'o .'°.'-‘--v

 

 

 

   
     
 

'0
I ' .
' I . . I

e . . e e ' . _ s. . _

I e g . ;. . '. e. E. .0.. 0 ‘i . .. - ' ' ' .... . 9.. .. . 0 t
- -"-' . - '.-'r"°‘e' -;-":-. . . o"-"-‘-. «r. ~'-'-' - M12"...

e' -. 0 ." ' u'l .u' ' ," ..I .'-' "' ‘ eCIII_ '. "-s U .Is 0.
. ...D. ... .... ... .. ... g. e, .....',‘.s_. ... W... ..o.".. a", {... ... "’9' '. ' .. . - . ........'..0. .. ...: 0' . .0.“. ...?

. . . - - 0 . - ‘ . . - . - , - . - ' ‘ H 0 .
...-.'AAA'..!A'..';'”;$ 1A.......v..1'l:'.'.00".-. ..‘“.'... ... ‘....-'L.'AIULL'. '..Jl' . A‘:-..:'L" I.-. ..s' ..-

Permstrost tsble

Fluvlsl sndlor sellsn sediments

 
 

 

 

10m

 

Figure

38.

 

Suggested sequential development of thermokarst topography in
the Saginaw Lowland.

107

patterned ground (Capac, Londo, Parkhill and Tappan) range from 15 to 35%
(Soil Conservation Service, 1978a; 1978b; 1979a; 1979b). In comparison
to these limits, Czudek and Demek (1970) reported that perennially frozen
loam soils under saturated conditions can contain 30 to 80% segregated
and/or vein ice. The pattern mesh in the study area was initially formed
by thermal contraction cracking and ice-wedge development in perennially
frozen ground, as revealed by the ice-wedge cast at the Bridgeport site.
The final morphological expression of these features, in particular their
reticulate rather than polygonal form, may be attributed to thermokarst
processes operating in an ice-rich permafrost environment. As illustrated
by the sedimentological data from the Lawndale site, the last development
phase of the Saginaw patterned ground involved the localized infilling of
some of these curvilinear thermokarst depressions with fluvial or eolian

sediments (Figure 38d).

CHAPTER V

PALEOENVIRONMENTAL SIGNIFICANCE OF THE PATTERNED GROUND

IN THE SAGINAW LOWLAND

Introduction

 

The concept of interpreting the morphology of the earth's surface
in terms of past climatic events is one of the major historical under-
pinnings of geomorphology. Since the development of the Glacial Theory
in the early part of the 19th century, it has been recognized that the
formation, expansion and eventual disappearance of glaciers occurred in
response to climatic change. Concurrent with the birth of this idea of
an "ice age" were numerous observations by earth scientists of nonglacial
evidence of former cold climates--remains of the woolly mammoth, fossil
shells of cold-water snails, frost-riven bedrock, solifluction mantles and
the like. These data were used, individually and collectively, to infer
paleotenperatures colder than those prevailing today at particular loca-
tions. In this respect, periglacial geomorphology can make unique contri-
butions to the study of the Quaternary because several periglacial land-
forms form only in areas underlain by permafrost, which have mean annual
ground temperatures 5 0°C.

The distribution of permafrost in arctic and subarctic North America
has changed very little since the end of Pleistocene time, but its distri-
bution in the temperate latitudes underwent major dislocations during the
late Quaternary (Pewe: 1973). Fossil biological evidence can suggest that
more rigorous climatic conditions once existed in temperate North America,
but it cannot document the former presence of permafrost. Other peri-

glacial indicators, such as involutions, block accumulations, talus fields

108

109

and asymmetrical valleys, are often best developed in permafrost areas,
but they do not require perennially frozen ground to form. The past exis-
tence of permafrost can only be ascertained from fossil examples of ice
and sand wedges and their associated patterned ground, pingos, rock glaciers
and perhaps altiplanation terraces (Brown and Pébefl 1973; Pewez 1973).
French (1976) takes a more conservative view by stating that the only re-
liable indicators of former permafrost are the casts of ice or sand wedges.
Thermal contraction cracking is governed, to a major extent, by the
temperature regime of the upper thirty feet or so of permafrost. While a
temperature change of only 4°C is sufficient to initiate cracking in super-
saturated perennially frozen ground, less saturated or thermally less con-
ductive permafrost may require a flux of 8°-10°C to begin cracking (Black,
1976). Studies in Siberia have established that ground temperatures of
only -2° to -4°C promoted ice wedge development in fine-grained materials,
but temperatures of -7° to -8°C were necessary to maintain active ice
wedges in gravel deposits (Romanovskij, 1973). The significant temperature
decreases necessary for contraction cracking typically occur in areas of
continuous permafrost with thin active layers and a temperature below -5°C
at the depth of zero annual amplitude. The microclimate of local areas of
discontinuous permafrost may also allow thermal contraction-cracking, but
ice-wedge growth in such environments is slower and more aperiodic, com-
pared to regions of continuous permafrost (Black, 1976). Ice wedges rarely
form in sporadic permafrost. Ice wedge growth is typical in areas having
minimum winter temperatures of less than -15° to -20°C at the permafrost
table if the perennially frozen ground has the mechanical prOperties of
polycrystalline ice (Lachenbruch, 1966; French, 1976). Pewe'(l975) re-

ported minimum.winter ground temperatures at the permafrost table to be

110

<-ll°C for actively growing ice wedges and between -4° and -10° or -15°C
for inactive wedges in Alaska.

The relationship between mean annual air temperature (MAAT) and a
thermal regime of the ground which promotes contraction cracking and ice
wedge growth is complex. For example, average annual air temperatures
can be 2° to 6°C cooler than mean annual ground temperatures (Gold and
Lachenbruch, 1973; Pewe: 1975). Additionally, as Black (1976) has pointed
out, thermal contraction-cracking is often controlled more by local sur-
face conditions and short-lived cold spells than by annual temperature
regimes.

thwithstanding the difficulties involved in the quantification of
the relationship between ice wedge growth and mean annual air temperature,
several general observations can be made. On the basis of extensive re-
search in Alaska, presently active ice wedges appear to be restricted to
areas having MAAT of -6° to -8°C or colder (Pehe: 1964, 1966a, 1966b,
1975; Pewez Church and Andresen, 1969; Washburn, 1973). Inactive ice
wedges occur in areas of Alaska where the MAAT varies from.-2° to -6° or
-8°C. Gold and Lachenbruch (1973) concluded that the discontinuous perma-
frost region, where ice-wedge formation is frequently inactive, was bounded
by the -1° and -6°C MAAT isotherms and that continuous permafrost (asso-
ciated with active ice wedges) occurred in areas having MAAT of less than
-6°C. Brown and Pewe'(l973) cite the -4°C MAAT isotherm as the southerly
limit of discontinuous permafrost and the -8.5°C MAAT isotherm as the
southern boundary of continuous permafrost.

Paleoclimatic Implications
of the Saginaw Patterned Ground
The nonsorted nets in the Saginaw Lowland are the result of permafrost

cracking and ice-wedge development during a time span beginning about 12,730

111

years ago with the fall of proglacial meltwaters from the Warren I level
and ending with the draining of Lake Elkton (not later than 12,400 yrs
B.P.). This patterned ground documents a climate which was considerably
colder and more rigorous than at present. The mean annual air temperature
which may have prevailed during this period when ice wedges were actively
growing in the study area can be deduced from the climatic conditions re-
quisite for ice wedge formation today.

The mean annual air temperature in the Saginaw Bay area today is
8.3°C and the average minimum winter air temperature is -8.3°C (Michigan
Weather Service, 1974). If the Saginaw patterned-ground complex formed
in continuous permafrost, the MAAT may have been l4°-l6°C colder than at
present. Undoubtedly, this estimate represents the maximum temperature
depression for this brief period (12,730-12,400 yrs B.P.). Considering
the fine-textured soils on which most of the net cells occur and the
width and depth of the ice-wedge cast at the Bridgeport site, it is more
likely that the ice wedges in the Saginaw Lowland formed in discontinuous
permafrost having an average ground temperature of -2° to -4°C. The MAAT
may have been on the order of 9°-l4°C colder than today (i.e., -1° to
-6°C). As noted by Morgan (1982), katabatic winds flowing off the con-
tinental ice mass may have locally depressed temperatures in the peri-
glacial zone, particularly near places where favorable ice-surface con-
figurations concentrated such cold air drainage. During the Late
Wisconsinan, the Saginaw Lowland may have been cooled by katabatic winds
which were focused on the area as a result of the more rapid dissipation
of the Saginaw Lobe compared to the neighboring ice lobes in the Michigan
and Huron basins.

The occurrence of discontinuous permafrost which supported ice wedge

growth in the Saginaw Lowland during the Late Wisconsinan indicates that

112

precipitation, particularly snow, may have been much lower then than at
present. Studies conducted over the last decade andaihalf at the McGill
Sub-Arctic Research Laboratory, Schefferville, Quebec (approximate MAAT
-5°C) have documented that snow cover is the most important single factor
controlling the distribution of perennially frozen ground in this area of
discontinuous permafrost (Gold and Lachenbruch, 1973; Nicholson and
Granberg, 1973). Earlier studies indicated that snow depths greater than
40cm were sufficient to prevent permafrost formation (Annersten, 1966),
but more recent investigations have revised this figure to 70-75cm
(Nicholson and Granberg, 1973).

Geomorphic Evidence of Former Periglacial Conditions and Permafrost From
Elsewhere in the Great Lakes Region

 

Fossil periglacial features have been reported from many places in
the Great Lakes Region. For example, talus accumulations, block streams,
”block fields and asymmetrical valleys indicative of a former "frost
climate" have been studied in the Driftless Area of southwestern Wisconsin
(Smith, 1949, 1962; Judson and Andrews, 1955). Wayne (1967) investigated
asymmetrical valleys in central and southern Indiana and attributed them
to periglacial frost action. Involutions, possibly resulting from freeze-
thaw activity in the active layer above permafrost, have been described
from northeastern and north-central Illinois (Sharp, 1942; Ekblaw and
Willman, 1955; Frey and Willman, 1958), central and southern Indiana
(Wayne, 1956, 1963a; Gamble, 1958) and northwestern Ontario (Sutton, 1963).
Leighton and Brophy (1961) interpreted enlarged bedrock joints south of
Carbondale, Illinois to have resulted from periglacial frost action.

Straw (1966) attributed the mass movement of dolomite blocks in the Nia-
garan Escarpment south of Meaford, Ontario to periglacial conditions asso-

ciated with the Valders ice advance. In contrast to these areas, the

113

scarcity of periglacial phenomena in Ohio has been noted by Goldthwait
(1959). As mentioned previously, however, none of these fossil periglacial
features are reliable indicators of former permafrost.

Péwé (1973) assumed that Wisconsinan permafrost formed in front of the
expanding continental glaciers and persisted beneath the ice in selected
areas where the bottom temperature was less than 0°C. subglacial tempera-
tures below freezing can occur in pressure melting situations (-2° to -1°C)
or where the basal ice is frozen to the underlying ground. In areas south
of the maximum extent of Wisconsinan ice, periglacial permafrost is thought
to have formed 22,000 to 14,000 years ago or earlier (Péhé, 1973). As the
ice margin withdrew to the north, recently uncovered drift became peren-
nially frozen and, in those areas where especially rigorous conditions
existed, ice wedges formed (Brown and Péwé, 1973).

Figure 39 shows the locations of all reported permafrost indicators in
the Great Lakes Region which have been accepted by most researchers. Fur-
ther documentation of these sites is given in Table 9. By far the greatest
number of ice-wedge casts have been observed in Wisconsin (Black, 1965).
The majority of these pseudomorphs occur along or south of the maximum
extent of Wisconsinan ice (locations 1-19, Figure 39) and are thought to
have formed 32,000 to 29,000 years ago during the late Altonian (locally
termed Rockian).

The four ice-wedge casts in Columbia County, Wisconsin (location 20)
are interpreted as having formed during the late Woodfordian 16,000-12,500
yrs B.P., but their location almost 50 miles beyond the local correlative
of the Port Huron Moraine (ca 13,000 yrs B.P.) suggests their age is prob-
ably closer to the 16,000 yrs B.P. estimate. The two wedge casts at site
21 (Outagamie County, Wisconsin), on the other hand, are located several

tens of miles proximally to the same Port Huron correlative moraine and

114

.oofiwom maxed uoouo onu :H umoumoauoa mo muoumoavow uenmuoaooc .mm enough

 

 

_ .hsop 6:35.20 so: eeaeo nee 233.com 359.360. @

32.22.30 nee ..

23203 :95... ton 7.

323.230 use ..

2.23: :6 5.5.. J

59.2: 02 5530023
2a.. 825va

38:... ...-8. ©

 

2520
asp—3:2. 02.3232

 

 

 

 

 

 

 

 

7230 3.33.80. e

 

 

 

 

ee..E
. 4 a
2: o
so.
3»
r
a. a.
a . .0005».
o
p 3 orer so
”so as
D

 

 

 

 

115

ooo.a~-ooo.~m
ooo.a~uooo.~m
ooo.a~-ooo.~m
ooo.a~uooo.~m
ooo.m~uooo.~m
ooo.a~uooo.~m
ooo.a~uooo.Nm
ooo.a~uooo.~m
ooo.a~uooo.~m
ooo.a~uooo.~m
ooo.a~.ooo.um
ooo.a~uooo.~m

.ooo.s~.ooo.~m

ooo.a~uooo.mm
ooo.a~uooo.~m
ooo.a~-ooo.~m
coo.s~.ooo.~m
ooo.a~uooo.~m
ooo.a~ucoo.~m

mumoo owvoanoow Houm>om
momma owvmzloow e

undo owvoalmoq H

momma owvoslooa Houo>om
mumou owvoslooa N

dunno omvoaloow +c~
mumoo mwvosloow +mN
momma owvoslmow m

muons owvoaloow Houo>om
momma owvosloow Houo>mm
momma owuoaloofi Houo>om
muons owvosloofl «H
mumou ammosloow Houo>mm
momma ownosloofl NH

umoo owvoalooa H

undo owooaloofi H

mumao owuoslmoa m

mumoo omwosloow 9

women ownmslooa +oM

owmooomwz ..oo xoom
ofimooomwz ..oo ououu
camaoomuz ..oo sumac
camaoomfiz ..oo mama
owmooomuz ..ou vomasoam
:Hmooomaz ..ou mouse:
camsoomfiz ..oo owmuuom
owmooomwz ..oo oomxoofi
nwmooumaz ..oo sumac
oamaoomfiz ..ou xuoflo
camcoomfis ..oo oufimao sou
camooooaz ..ou ouwoao 2mm
camooomwB ..oo ouqoau sou
ofimooomfiz ..oo ouwmau sou
owmooomfiz ..oo noon
camaooqu ..oo coon
camooomfiz ..ou mouoam
aeoooomaz ..oo nacho .um

camooomwz ..oo xfiouo .um

moms .xomam
moms .somam
mesa .xuasm
moss .sumam
moms .xumam
moss .somam
moms .xuaflm
moms .somam
moss .somHm
moms .somfim
moms .sumflm
moms .sumflm
moss .soaam
moms .xomHm
moms .somflm
meafi .somflm
moms .xumam
moms .sosam
moss .xosflm

HNMQfiONQO‘

 

A.m.m mu%v
sowudauom
«0 mafia

Amvounuoom

sussmuoq

OUGQHQMOK

 

m manna

seamed magma uoouu onu ea umoumoauom‘uoauom mo muoumowvofi ounmuoaoou

nonesz
nowuoooq

as:

116

 

oom.¢Hv AHV oonm HHmmou oHnHumoe H sooHvoH ouoummonuuoo momH .o:%m3 “.1 mm
man .uono: use
coo.HHuooo.mH womaHa HHmmoH +oom «HoaHHHH .anMon HuHxaHm .HmamHm Hm
sow
IHnon ..oo mHoomsa moor.
oo¢.~Hlooo.mH mum: wouuomooo swanom .vooHsz .zom
ooe.~HIooo.mH ammo owvoalooH H oannon ..oo smeHwom. . renew mHmh cu
coo.mHlooo.mH vooouw vodumuuoe wouuomooo OHHouoo auoumosnuaom -. 2H,... , cw
ooc.MHIooo.mH momma omvoslooH s oHuouso .uHsz «waH .oomuoz
coo.mH|oom.mH nasoum voououumn vouuomoo: oHumuoo .uooosouHx ..1
coo.mHuoom.HH some unassumUH H oHusuao .HoameouHe m, NHmH .ammuoz mm
coo.NH|ooo.H~ vooouw voououuoe vouuomooo oomHvoH Houueoolumoalx
oco.~HIooo.wH mumoo owneknooH N mooHvoH ..oo muoaowuooz ncmH .oomoz .Hwiem
mmaH
ooo.- ammo owsoznmoH H mHoaHHHH ..oo euoHuooz .amaHHHz was scum mm
coo.eH some «museumUH H «HoaHHHH ..ou ammuam mqu .mumnuom NH
oom.NHIooo.oH mumoo omnoanooH N onsoomH3 ..oo oHaomouso mooH .xuon Hm
oom.~H.ooo.eH sumac «mamasooH s :HmaoomHz ..oo .aHsasHoo momH .sumHm oH
A.m.m muhv Amvounuooh huHHoUOH monouomom nonasz
oOHunauom oOHuoUOH
mo oaHs an:

as.uaoov m mHan

117

must have formed sometime later than 13,000 yrs B.P.

A single ice-wedge cast was reported from.location 22 (Horberg, 1949)
and also at site 23 (Frey and Willman, 1958), but in both cases the de-
scriptive data provided in the literature are too meager to allow a proper
assessment of their significance. Wayne (1967) discovered two ice-wedge
casts in Montgomery County, Indiana (site 24) associated with nonsorted
patterned ground which probably formed between 18,000 and 17,000 years
ago.1 He also reported isolated groups of patterned ground from several
other locations in west-central Indiana which may have formed at about the
same time.

ansorted patterned ground and associated ice-wedge pseudomorphs near
Kitchener, Ontario (locality 25) were investigated by Morgan (1972) and
Greenhouse and Morgan (1977) who concluded that they resulted from thermal
contraction cracking of permafrost and suggested a Late Wisconsinan MAAT
l3°-l4°C colder than at present. Because these features were located dis-
tal to the Paris Moraine (a Port Huron Moraine correlative) on Port Stanley
Till, Morgan (1972) assigned them an age of 13,500-13,000 yrs B.P. However,
a reinterpretation of glacial Lake Whittlesey shoreline data from.south—
western Ontario (Barnett, 1979) suggests that the Paris Moraine (and,
therefore, the Kitchener patterned ground) may be of Port Bruce Stadial
age (14,500-13,500 yrs B.P.). Mbre recently, Morgan (1982) reported numerous

other occurrences of patterned ground in southwestern Ontario and described

 

1In two previous review articles (Brown and Pehe: 1973; Pehez 1973) these
ice-wedge casts were erroneously listed as younger than 14,500 yrs B.P.
Wayne never included a date for these features in his published descrip-
tions, but the pseudomorphs occurred in the Cartersburg Till Member of
the Trafalgar Formation which is younger than the underlying Vertigo
alpestris oughtoni bed, the average age of which is 20,083 i 831 yrs B.P.
(Wayne, 1963b). The wedge casts are located on the distal flank of the
inner Crawfordsville Moraine, a possible correlative of the Farmersville
and Reesville Moraines farther to the east which formed about 17,000
years ago (wayne, 1965; Dreimanis and Goldthwait, 1973).

118

several ice-wedge casts which were excavated near Muir, Ontario. Consi-
dering the rockrstratigraphic units underlying these polygons, it was con-
cluded that the patterns formed in the period from about 15,000 to 13,000
yrs B.P. Morgan suggested that these features indicated a MAAT of -3°C to
-4°c, at this time.

More than 500 fossil pingos occur in an area of approximately 115
square miles (site 27) near DeKalb, Illinois (Flemal, Hinkley and Hesler,
1973). Since modern pingos develop only in areas of permafrost, pingo
scars are indisputable evidence of former perennially frozen ground
(Flemal, 1976). These "DeKalb Mounds" were presumed to have formed 16,500-
15,000 yrs B.P. on the basis of their stratigraphic position between the
underlying Tiskilwa Till and the overlying Richland Loess, but more recent
estimates place this time interval at 18,000-17,000 yrs B.P. (Dreimanis,
1977). Another circular soil feature, also interpreted as a pingo scar,
was reported from near Corunna, Indiana in the northeastern part of the
state (Wayne, 1967). This feature (location 28) was identified from
aerial photographs but never investigated in the field. Without further
substantiating field evidence, the pingo origin of this feature remains
speculative.

On the basis of high quality acoustic profiling records, Bowlby
(1975) reported an extensive network of fossil ice-wedge casts on the
'floor of the Kingston Basin of eastern Lake Ontario beneath almost 100
feet of water. Nearly 500 presumed pseudomorphs were detected in an area
of about 0.6 mi2 and were assigned an age of about 11,000 yrs B.P. Al-
though an ice wedge origin for these patterns is within the realm of pos-
sibility, given the difficulty of correctly identifying true ice-wedge

casts by normal visual observations and field sampling (Johnsson, 1959;

119

Black, 1976), such a genesis for these underwater features is questionable
when based solely on acoustic data.
Palynological Evidence of Late Wisconsinan Tundra
Environments in Michigan

After decades of research, a detailed paleobotanical record of late
Pleistocene vegetation has been compiled in Europe. From the outset in the
late 1940's, palynologists in this country anticipated that their studies
would yield pollen spectra similar to those of their European colleagues,
including a basal herb assemblage (particularly grasses, sedges and compo-
sites) representing late—glacial tundra conditions, overlain by several
arboreal pollen zones indicative of postglacial vegetational succession.
In this respect, the early pollen diagrams from the Midwest (e.g. Potzger,
1946, 1948, 1951) were disappointing; the basal organic sediments were
usually dominated by Pigga pollen, contained high percentages of Piggg pol-
len and oftentimes displayed significant amounts of pollen from thermo-
philous species such as Quercus. Unfortunately, most of these early pollen
counts did not include herbaceous species, making it impossible to identify
any of the common nonarboreal tundra indicators. In fact, Potzger's exclu-
sion of nonarboreal pollen was vigorously criticized by Deevey (1951) who
contended that evidence of the tundra zone was to be found in the under-
lying mineral sediments directly beneath the lowermost organic deposits.
It is now recognized that many of these lacustrine and bog sample sites may
not have become depositional basins until as much as one or two millenia
after deglaciation as a consequence of the persistence of buried stagnant
ice (Florin and wright, 1969; Wright, 1971; Odgen, 1977).

Meager evidence of open tree stands, possibly representing park tundra,
is available, however. The pollen record from George Reserve in Livingston

County, for instance, although dominated by Picea at its base, included

120

30-352 herb pollen (Andersen, 1954). No radiocarbon dates were determined
from this core, but the basal sediments were correlated with the Port Huron
advance (13,000 yrs B.P.). Stoutamire and Benninghoff (1964) analyzed the
organic detritus associated with a mastadon skull found at Pontiac, Michi-
gan which was dated at 11,900 i 350 yrs B.P. This material was dominated
by gigga pollen (82.81 of total AP), but also contained large amounts of
nonarboreal pollen (852 of total AP). Although such a high proportion of
NAP can be indicative of incomplete forest cover, in this case more than
half of the nonarboreal species were representative of shallow ponds, marshes
or shrub swamps.

The best palynological evidence of Late Wisconsinan tundra conditions
in Michigan comes from two widely separated but time-stratigraphically
penecontemporaneous sites (Cheboygan and Lapeer counties). The well known
Cheboygan bryophyte bed, of such importance as an interstadial marker,
yielded the first solid data indicating tundra conditions in the state.
This one-centimeter-thick moss layer occurred at a depth of just over four
meters beneath red, clayey till capped by stratified sand and gravel. Four
radiocarbon dates, ranging from 12,500 to 13,300 yrs B.P., were determined
from the organic materials (Farrand, Zahner and Benninghoff, 1969), but ac-
cording to Dreimanis and Goldthwait (1973) the most reliable of these dates,
as a result of an alkali-acid pretreatment, is 13,300 i 400 yrs B.P.
(L-1064).

The pollen diagram shown in Figure 40 is based on the analysis of four
samples: one from within the bryophyte bed, one from the subjacent sand
and two from.the overlying silty clay (Miller and Benninghoff, 1969). Of
particular interest is the large amount of herb and shrub pollen identified
in these samples. Cyperacea yielded the most abundant nonarboreal pollen,

varying from 852 in the sand beneath the moss layer to 382 in the overlying

121

 

NMONMNO

 

 

HVI'IIWVJNH
VHOHVLIUOV

 

SBUOdS BflDNfld

 

838063 880"
WOHLSVIOSd
NOIHOAHLOG

 

 

 

 

SVSOWOOdOOA'I
SVSOVIOOdA‘IOd

 

 

 

BVBOVOILHH + —
NOHLOI'IVHJ. + ‘—
‘SOddO VDVEHIXVS + -
SVOV'I'lM-IdOAHVO + -

 

 

 

'SWV-ONSHO Ill—
BVLISOJHOO Hal-I10

 

V's'"a I 8' (.JA.L.A'A.A.A--'lv"A A n a . A s e s s‘ I 1 s e e s A A e s A J
'C
I

 

VISOHGNV
3'3".“vaa Within-.-...-“u, .

BVBOVHSJAO

 

 

 

XI'IVS

 

 

SflN'IV

 

Sfl'lAHOO

 

 

 

 

 

 

 

VAHlso-SNNIdUVO
V'fl'llaa W

Sfl'lfldOd
SflNlXVt-H
Sflfl'lfl

 

 

 

 

 

 

 

snoaano
fl:

 

VABVO
SNV'IDOI‘

 

 

 

 

8|.L130 +

 

330V 4-
SSIBV

SflNld

 

 

 

 

V30“

 

 

 

 

”09"?“ bed ' _..:

 

inunlwmwthIhmmmmNMtjooo

50%

one or two orsln occurrences)

(+

Pollen diagram of the Cheboygan bryophyte bed and associated sediments.

Figure 40.

122

lacustrine sediments. Smaller, but still significant amounts of 51325,
'§ali§, Gramineae, Ambrosia and Artemesia were also detected. 'Of the ar-
boreal pollen, Pigga was the most common, ranging from 82 in the lowermost
sample to 312 in the silty clay 3.5cm above the bryophyte bed. .Piggg pol-
len accounted for less than 12 of the total AP in the organic zone and
underlying sand and reached a maximum of only 132 in the younger lacustrine
material. Betula pollen occurred only in the upper two samples and a small
percentage of Abigg pollen was detected in the three non-organic sediments,
but not in the bryophyte bed.

In common with most late-glacial pollen spectra from.Michigan, small
amounts of pollen from thermophilous deciduous trees, such as 95523 and
92ercus, were identified in the upper three samples. Additionally, one or
two grains of Acer, Celtis, Fraxinus, ggglans, Populus, Ulmus or Cagpinus-
Ostgza occurred sporadically throughout the samples. This pollen from.more
temperate arboreal vegetation may have been redeposited from older sediments,
but more likely was transported to the site from great distances (Miller and
Benninghoff, 1969).2

In addition to pollen, macroscopic plant fossils were preserved in the
bryophyte bed; they consisted mainly of mosses, mixed with a few liverworts
and some leaves, twigs and seeds of several vascular plants. Eight dif-
ferent species of mosses were identified, including Bryum cryophilum which
is considered to be a truly arctic species (Steere, 1947). Macrofossils of
six vascular plant species were recognized and among the three sedges clas-
sified was Carex supine, which is presently restricted to the Arctic and

Subarctic.

 

2Modern long-distance transport of pollen has been documented by Ritchie
and Litchi-Federovich (1967) who identified small amounts of Carya, Fraxi-
lggg,‘ggglans, Quercus and Ulmus in the pollen rain at Churchill, Manitoba
which is nearly 400 miles beyond the range of these genera.

 

123

The most abundant flowering plant in the bryophyte bed was Dryas inte—

 

grifolia. According to Porsild (1957), this dwarf shrub is a ubiquitous
pioneer species in the arctic or alpine tundra. The fossil remains of two
other arctic-alpine indicator plants also occurred in the organic zone.
Twigs and well-preserved leaves of Salix herbacea were identified, as well
as fossil leaves of vacinium uliginosum var. alpinum.

Both the pollen spectra and the macrofossil assemblage document open
vegetation which was dominated by sedges and other herbaceous plants.
Spruce and possibly other trees were locally present, but scattered, as
indicated by the presence of the several tundra heliophytes, including
Dryas integrifolia. The lack of a closed forest is also suggested by the
relatively small total amount of spruce pollen in the spectra. An addi-
tional indicator of open tundra conditions is the larger-thanrunity NAP/AP
ratio in the bryophyte bed and in two other samples (Livingstone, 1955).

As pointed out by Miller and Benninghoff (1969), it is highly improb-
able that the many fragile structures (leaves, seeds, etc.) which were
preserved in the organic debris could have been transported any significant
distance and still remain identifiable. Thus, these macrofossils provide
conclusive evidence that the plants which produced them were growing on or
near the depositional site. Similar conclusions cannot be drawn from the
pollen record, however, since both local and distant vegetation contribute
to the pollen rain at any site. Taken together, the pollen spectra and
macrofossil assemblage in the Cheboygan bryophyte bed document the exis-
tence of an open vegetation community at this northern Michigan site some 3
13,000 years ago which was floristically similar to present-day tundra com- bl
munities in arctic or subarctic North America.

Additional evidence of late-glacial tundra conditions in Michigan

comes from the Weaver Drain site in northeastern Lapeer County (Burgis,

124

1970; Eschman, 1978). This drainageway, located in a linear depression be-
tween two moraines, formed part of the Imlay Outlet of glacial Lake Maumee.
Organic remains recovered from this locale yielded a radiometric age of
13,770 i 210 yrs B.P. (I-4899).3 Although the Weaver Drain material was
not subjected to pollen analysis, the fragile macrofossils clearly document
in situ tundra vegetation in the southeastern part of the state during the
Mackinaw Interstadial. Paleoentomological data from the Weaver Drain site
also indicate tundra-like conditions, although the presence of one fossil-
ized bark beetle (Pglygraphic rufipennis) suggests that trees were not far
away (Morgan, Elias and Morgan, 1981). According to Morgan (1982), these
data document an MAAT in the -1° to -4°C range.
‘Rggional Synopsis of Late Wisconsin Tundra
and Permafrost Indicators

The temporal and spatial inter-relationships between the paleobotanical
'tundra indicators and the previously discussed geomorphic evidence for peri-
glacial permafrost are summarized in Figure 41. Among the oldest features
shown are the DeKalb Mounds located proximally to the Bloomington Moraine
on the Tiskilwa Till. According to Frey and Willman (1973), the most sig-
nificant Woodfordian glacial retreat of the Lake Michigan lobe occurred
between the deposition of the Tiskilwa and Malden Till Members of the
Wedron Formation. Because no organic deposits have ever been found between
these two tills, Dreimanis (1977) suggested that this phase of the ice re-
treat approximately 17,500 years ago may have represented a cold, dry
episode rather than one of warming. The pingo origin for the DeKalb Mounds

would be in agreement with this hypothesis.

 

3This deposit has been overlooked by even the most recent reviews of Michi-

gan's late-glacial flora (e.g. Kapp, 1977), yet it contained well-preserved
macrofossils of Dryas integrifolia, Salix herbacea and Vaccinium uliginosum.

125

.umouumauon 52525033 33 no 93:39: 3:93:50» on”. new muons... no menu...

1.26 HooHoouoaooHon can. costume maHsoGOHuoHou on... 93593 $5.33. moo—3 uoouu on”. :H non—0H

HoHoon no he...

owns”. 0:”. mo none 05 was: mooHuonuoon HoonuoalooH «0 gauge mucouoHoloaHH

3.0. 5.57.: assoaiEoS: tonnes}. ...-ml... .301.» o...>oo..o.320u..o 5.3.33.0...» 32.3.32

.Ho ouome

 

:2: 3.56.20 E2. 2.2.3.2... 3592.73:

 

 

.38 .5332

 

 

‘11 \
x .
9.0.2.3.... N ...... 2322:.» N
— - H 1
rl 00* co» co“ oo—

    

L.»

mmo. o.¢<»zo-m_mm/~./\....fiuvuufluu:

 

.o [Egon

 

 

.33.... 2.5

 

 

.93» so...» ...on.

...—3:. 32....an
.93» :95: ...on.

 

     

 

  
    

  

 

 

 

   

 

..2e.=. 3.005 2:...
.3.» 3.2;

 

..e.o.:. an» ....o:

 

/\</\/

2.3.2....» 2:...
sexed :95
53:»

 

 

 

@—
.\ “moo. 20...... m
l
2.3; ”Human“ ..F :32... 5.5 ... MM
[ID no.5.» ...om .233 1
{Lb
D
. Ivan . 4 :4.
3:. cacao:
9‘! o 1 ...—:35 00/5032. ieciv «H
\\
>mm04 z<o_:o_.2 mi...
.91» a;
flop

126

The contemporaneous patterned ground and associated ice—wedge casts
in Montgomery County (Indiana) are located up-ice from.the outer Craw-
fordsville Moraine, an ice marginal position from which the Huron lobe
retreated as the Erie Interstadial began. This retreat probably occurred
in the form of large-scale ice stagnation during another cold but dry
episode (Goldthwait, 1968; Wayne, 1968). The overall paucity of organic
remains from the Erie Interstadial may indicate that conditions were too
rigorous for any significant plant migrations from the south and a thin
cover of tundra vegetation may have dominated the periglacial zone at
this time (Dreimanis, 1977). The patterned ground in Montgomery County
was formed by permafrost cracking and corroborates Dreimanis' conclusions
about the climate in the periglacial zone during this interval.

The organic deposits in the Cheboygan County bryophyte bed and at
the Weaver Drain site are associated with the Mackinaw Interstadial and
both document tundra vegetation. The icedwedge casts and nonsorted pat-
terned ground near Kitchener (Ontario) are restricted to the Port Stanley
Till of Port Bruce age and they provide morphologic evidence of a cold
periglacial zone in which permafrost existed during the Mackinaw Inter-
stadial or the waning Port Bruce Stadial.

The youngest geomorphic indicator of periglacial permafrost in the
Great Lakes Region is the patterned ground in the Saginaw Lowland described
in this study. These nonsorted nets, only slightly younger than the Che-
boygan bryophyte bed which conclusively documented a local Late Wiscon-
sinan vegetation community which was floristically similar to those of the
present North American tundra, provide fossilized evidence of a short-
lived, rigorous climate associated with the waning Port Huron Stadial.

It is significant that all four of these geomorphic indicators of

permafrost from Illinois, Indiana, Michigan and Ontario are associated

127

with waning ice margins. Taken together, they indicate that at least
localized areas of periglacial permafrost existed along the margins of
the retreating Lake Michigan, Huron and Ontario lobes during the Late

Wisconsinan.

Summary and Conclusions

As a result of interpreting numerous aerial photographs of varying

scales, film type and date of acquisition, the true extent of patterned
ground in Michigan's Saginaw Lowland has been delimited in this study.
(More than 350 mi2 of patterned terrain occurs within a 3-8 mile-wide
arcuate tract across parts of Arenac, Bay, Gladwin, Midland, Saginaw and
Tuscola counties. These patterns consist of a reticulated network of
broad, shallow troughs which encircle slightly higher centers. Indivi-
dual pattern cells are typically oval in plan view, as indicated by the
1.7:1 average ratio of their major to minor axis lengths. They range in
size from small, circular forms 79 feet in diameter to very large, elonr
gated features measuring 1,289 x 480 feet. The average cell mesh is 380
x 223 feet across. On the basis of their geometry and relief marking,
they are classified as nonsorted nets.

Morphostratigraphically, the Saginaw patterns occur upon, distal to
and proximal from the Port Huron Moraine, the boundaries of which were re-
interpreted during this research. It has been shown, for instance, that
the extent of the Port Huron Moraine, particularly in Midland County, is
greater than depicted on a map of the surface formations of the Lower
Peninsula (Martin, 1955). Additionally, the waterlain portions of this
ice-marginal feature have been more accurately delimited compared to pre-
vious interpretations.

On the basis of their juxtaposed boundaries, Tillema (1972) correlated

128

the lower elevational limit of the Saginaw patterned ground with the level
of glacial Lake Elkton. This relationship held true along the entire
length (lo-15 miles) of his study area in southwestern Bay County and
north-central Saginaw County which is south of the Elkton zero-isobase.

In order to assess the validity of this hypothesis, all proglacial lake
shorelines in the study area, from the Warren I elevation down to the
Elkton level, were mapped. For more than 100 miles, from the northern
border_of Bay County around to the Huron-Tuscola county line, the lower
limit of the nonsorted nets corresponds remarkably well with the Elkton
shoreline. Significantly, this correlation continues across the deformed
zone north of the zero isobase and indicates that active pattern formation
was contemporaneous with and bounded by Lake Elkton. Recent investiga-
tions from elsewhere in the Great Lakes basin (see Table 4, p. 45), have
concluded that Early Lake Algonquin, the first stable water plane after
Lake Elkton drained, formed by 12,400 yrs B.P. This represents the minimum
date for the cessation of patterned ground formation in the study area.

The upper elevational limit of the nonsorted nets is more difficult
to interpret, but appears to be related to the highest level 0f proglacial
Lake Warren. A single radiocarbon date on wood samples from.western New
York established an age of 12,730 i 230 yrs B.P. for this lake stage
(Calkin, 1970). On the basis of their morphostratigraphic position alone,
the patterns must be younger than 13,000 yrs B.P. (the age of the Port
Huron Moraine and proglacial Lake Saginaw).

Although the Saginaw nets are topographically bounded by the Warren
and Elkton shorelines, other landscape factors influenced their local dis-
tribution. Most of the reticulate patterns in the Saginaw Lowland formed
in somewhat poorly drained loam or silt loam materials on surfaces with

less than 32 slope and only 10-15 ft/mi2 local relief.

129

Of the 14 proposed origins for nonsorted nets which were reviewed,
only permafrost cracking and ice wedge development is supported by the
data from the study area. The geometry and morphometry of the Saginaw
patterns are very similar to known active, inactive and/or fossil tundra
polygons from numerous locations. The structural and sedimentological
attributes of the Bridgeport wedge clearly document that it is a true ice-
wedge cast. As illustrated by the edaphic conditions at the Lawndale site,
however, thermokarst erosion, rather than ice-wedge fossilization, was one
of the more important geomorphic processes associated with the thawing of
the permafrost in the Saginaw Lowland.

Studies from Alaska and elsewhere indicate that presently active ice-
wedge formation is rare above certain threshold temperatures. Based on
these relationships, the patterned ground in the study area, which formed
by thermal contraction-cracking of permafrost, documents a brief interval
within the Late Wisconsinan Substage (12,730-12,400 yrs B.P.) when the
MAAT was on the order of -1° to -6°C, i.e., 9°-l4°C colder than at present.
In addition, the existence of this perennially frozen ground suggests that
diminished winter precipitation and/or enhanced eolian activity prevailed
in east-central Michigan at this time since snow depths greater than
50-75cm are known to prevent permafrost formation (Nicholson and Granberg,
1973).

Finally, the nonsorted nets in the Saginaw Lowland are the youngest
of the four Late Wisconsinan permafrost indicators from Illinois, Indiana,
Michigan and Ontario. All of these periglacial landforms are associated
with waning glacial stadials. Taken together, they indicate that at
least localized zones of rigorous climate existed during ice-marginal re-

cessions and they suggest that some interstadial periods during the Late

130

Wisconsinan may have been initiated by cold, dry conditions rather than

by warming.

Recommendations for Further Research

 

A largely qualitative approach has been taken in this study to de-
scribe the characteristics of the patterned ground in the Saginaw Lowland
and its relationships to other environmental factors. Future research
should concentrate on quantitative evaluations of these data. The comp
parisons of the spatial distribution of the nonsorted nets with respect
to soil texture and drainage, local relief and elevation were based on
interpretations of small-scale maps. By necessity, the topographic and
edaphic details of the landscape in these graphic presentations had to
be generalized. The specific correlation between patterned ground occur-
rences and soil management groups, for instance, is a problem which re-
quires further investigation. Using appropriate sampling techniques,
data could be generated and statistically analyzed so that confidence
limits could be assigned to the_co—occurrence of the nets by soil texture/
drainage classes. A potentially more powerful analytical technique would
be to quantitatively test the hypothesis that patterned ground in the
study area is restricted to low-relief (i.e., :15 ft/miz) sites between
the Elkton and Warren shorelines which are underlain by somewhat poorly
drained loam and silt loam drift. Such a multivariate analysis would
undoubtedly enhance our understanding of the environmental conditions
optimal for the formatiOn and/or fossilization of the Saginaw patterns
and, at the same time, provide a more objective data base for the evalua-
tion of other hypotheses which may be proposed in the future concerning

the genesis of these features.

131

As mentioned earlier, the deformation structures which are common
in the host sediment adjacent to ice-wedge casts were lacking at the
Bridgeport site, probably because the till is massive and unstratified.
Nevertheless, compressional forces must have been exerted on this material
during the growth-cycle of the Bridgeport wedge when the saturated perma-
frost below the active layer and above the level of zero annual amplitude
expanded as it warmed somewhat in the summer months. Evidence of these
stresses may be preserved as a micro-fabric in the host material at the
Bridgeport site and may be obvious in thin-sections cut from.samp1es
extracted from.near the wedge perimeter after being impregnated with a
plastic resin.

1 Another potentially valuable technique would be to measure the linear
coefficient of thermal expansion of soil samples taken from.the C horizons
of various soil management groups in the study area. In the laboratory,
samples of the same soil type but differing saturation levels could be
tested in order to model the impact of soil water on thermal contraction.
Data from such experiments could provide quantitative information by which
the thermal contraction-crack origin of the Saginaw patterns could be
evaluated.

The existence of permafrost in the Saginaw Lowland during the interval
between 12,730 and 12,400 yrs B.P. presents the obvious challenge to search
for evidence of contemporaneous perennially frozen ground elsewhere in
Michigan. As a guide, the terrain between the Warren I shoreline and the
level of Lake Elkton in other parts of the state should be given special
scrutiny. A cursory perusal of published soil surveys and selected aerial
photographs has revealed that numerous patches of patterned ground, similar
in form and marking to the Saginaw nonsorted nets but of less extent, occur

in parts of Monroe, Wayne, Macomb and St. Clair counties. Previous work by

132

the author in Sanilac County concluded that the patterned ground in that
vicinity may not be related to permafrost, but the impact of severe ther—
mokarst erosion was not considered and this area should be re-examined
with this hypothesis in mind.

The edaphic characteristics of the nonsorted nets in the Saginaw
Lowland suggest that well-developed patterned ground in Michigan may be
restricted to the somewhat poorly drained, finer-textured soils which are
typical of the lake bed areas. However, small, scattered patches of pat-
terned ground have been observed on airphotos of Branch, Hillsdale, St.
Joseph, Cass and Calhoun counties where coarse-textured drift is common.
These features need to be examined in terms of their morphometric, topo-
graphic, pedologic and stratigraphic characteristics in order to discern
similarities and differences compared to the Saginaw Lowland patterns in-
vestigated in this study. 3

It appears that periglacial permafrost, as documented by fossil pat-
terned ground, may have been more extensive in Michigan than was previously
thought possible. Undoubtedly, this perennially frozen ground was discon-
tinuous and probably formed only in relatively restricted areas where local
conditions provided a conducive environment. Nevertheless, its impact on
the sculpturing of the modern landscape has yet to be fully understood and

awaits further research.

APPENDICES

APPENDIX A
SOIL MANAGEMENT GROUPS

The basic mapping unit shown on soil surveys is the soil series which
is named for the town or other geographic feature nearest to the place
where it was first recognized. Soils within a given series are similar
to one another in their depth, thickness and arrangement of horizons and
other biological, chemical and physical properties, but the series name
indicates none of these characteristics. According to Mokma (1978), two
of the most significant physical parameters of soils are their dominant
profile texture and natural drainage conditions. Associations of soils
based on these two factors are called Soil Management Groups and are de-

noted by the systematic arrangement of numbers and letters listed below.

A. Mineral soils developed from uniform parent material

 

 

Dominant profile texture S 01
>602 clay 0
40-602 clay ‘ 1
Clay loam and silty clay loam 1.5
Loam and silt loam 2.5
Sandy loam 3
Loamy sand 4
Sand (strong subsoil development) 5.0
Sand (medium subsoil development) 5.3
Sand (weak or no subsoil development) 5.7
Gravelly or stony loamy sand to loam G
Bedrock at less than 20" depth R

133

C.

134

Dominant profile texture

 

Alluvial or lowland soils

loamy

sandy

Mineral soils developed from parent materials of contrasting,

 

juxtaposed textures

 

Dominant profile texture

Clay, 20—40 inches, over gravelly sand

Sandy loam, 14—40 inches, over clay

Sandy loam, 20-40 inches, over loam to clay loam

Sandy loam, 20-40 inches, over gravelly sand

Loamy sand, 14-40 inches, over clay

Sand to loamy sand, 20—40 inches, over loan to

clay loam

Sandy to loamy sand, 40-60 inches, over loam to clay

Loam, 20—40 inches, over bedrock

Sandy loam, 20—40 inches, over bedrock

Sand to loamy sand, 20-40 inches, over bedrock

Organic soils

 

Depph of organic material
>51 inches

16-51 inches

16-51 inches

16-51 inches

16-51 inches

16-51 inches

Underlying material

Clay

Sandy loam to clay loam
Loamy sand to sand

Marl

Bedrock

4/2
5/2
2/R
3/R

4/R

M/ 1c
M/ 3c
M/4c
M/mc

M/Rc

135

Natural drainage is indicated by lower case letters:
a - well and moderatley well-drained
b - somewhat poorly drained

c - poorly and very poorly drained

APPENDIX B

ELECTRICAL RESISTIVITY SURVEYING

A schematic diagram of the apparatus used for electrical resistivity
measurements is shown in Figure Bl. In operation, an electrical current
of about 100 vdc is passed through the soil body between the two end
(current) electrodes while the voltage differences between the middle two
(potential) probes is measured. Apparent resistivity is proportional to
the ratio of voltage change across the potential electrodes to the total

current .

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

, sdlustsble
reslstsnce
better es
mllllsmmeter
voltmeter
01 P1 P2 \62
Current 90'0”." Current
electrode probes electrode
}< x x XX W“_7XT‘—7X)‘
l l J

Figure Bl. Schematic diagram of electrical resistivity apparatus.

A large variety of electrode configurations have been proposed for
resistivity surveying, but the linear arrays are the most common. Figure

BZ shows three of the frequently used linear arrays. The Lee and the

136

137

Wenner configurations, which are symmetrically arranged, are similar to
one another with the exception of the additional potential probe in the
middle of the Lee array. In both cases, however, the entire arrangement

is moved laterally in fixed increments, but the Wenner array, having

fewer electrodes, is less cumbersome in this respect. In the Schlumberger
array configuration, on the other hand, the closely spaced potential probes
are moved as a pair between the widely separated stationary current elec-

trodes.

 

 

 

‘_c1_L P11, TLLPZ '02 Lee
_O_‘li P14, lP2 I C2’nng'
_—:1_L P14, .|,P2 lc um .[0692'

Figure B2. Linear electrode configurations commonly used in

resistivity surveying.

The penetration depth of a resistivity profile is primarily deter-
mined by the distance between the current electrodes. According to Paragsis
(1973), less than 30% of the total current penetrates below a depth equal to
the array length. The large separation distance between the current elec-
trodes in the Schlumberger array, for instance, makes this configuration
inappropriate for detecting shallow resistivity anomalies.

Figure B3 illustrates how near-surface resistivity anomalies can be
detected using this instrumentation. According to Jakosky (1950), equi-

potential lines are displaced away from a zone of high conductivity and as

138

 

4 3 2 1
l l l l .
zone of relstlvely

nlgb conductlvlty

413$ -3

1111 .

 

lll .

 

 

 

 

 

 

 

Appsrent Reslstlvlty

Trsverse Dlstsnce 01 Number 1 Electrode

trom Jsltoslty. 1 050

 

 

 

Figure B3. Detection of a near-surface resistivity anomaly.

139

the electrode array approaches such a conductor, the potential difference
between probes 2 and 3 is decreased somewhat causing a slight lowering of
the apparent resistivity at sample point A. At station B, the electrical
current is short-circuited between electrodes 1 and 2 by the conductive A
zone, making the potential between probes 2 and 3 large and producing a.
maximum apparent resistivity value.

The opposite situation occurs at array location C. Here the potential
difference, as measured across the conductor, is very small and apparent
resistivity falls to a minimum. The relative electrode positions at station
D are the same as were encountered at B and apparent resistivity once again
Sis maximum. A Wenner array configuration passing over a zone of relatively
low conductivity would produce an inverted resistivity curve having the

same general characteristics.

B IBLI OGRAPHY

BIBLIOGRAPHY

Aartolahti, T. (1970) Fossil Ice-wedges, Tundra Polygons and Recent
Frost Cracks in Southern Finland. Annalee Academiae Scientiarum
Fennicae, Ser. A, III, Nr 107. 26p.

. (1977) Personal communication.

Aleshinskaya, Z. V., Bondarev, L. G. and Gorbunov, A. P. (1972) Peri-
glacial Phenomena and Some Paleogeographical Problems of Central
Tian-Shan. Biul. Peryglac., Vol. 21: 5-13.

American Geological Institute. (1962) Dictionary of Geolggical Terms.
Garden City, N.Y.: Dolphin Books. 545p.

Andersen, S. T. (1954) A late-glacial pollen diagram from southern Michi-

gan, U.S.A. Danmarks Geologiske Undersogelse, Vol. 2, Nr. 80: 140-
155.

Annersten, L. J. (1966) Interaction between surface cover and permafrost.
Biul. Peryglac., Vol. 15: 27-33.

Barendsen, G. et a1. (1957) Yale natural radiocarbon measurements II.
Science, Vol. 126: 908-919.

Barnett, P. J. (1979) Glacial Lake Whittlesey: the probable ice frontal
position in the eastern end of the Erie basin. Can. J. Earth Sci.,
Vol. 16: 568-574.

 

Benedict, J. B. (1970) Frost Cracking In The Colorado Front Range. Geo-
grafiska Annaler, Ser. A, Vol. 52: 87-93.

Black, R. F. (1952a) Polygonal Patterns and Ground Conditions From Aerial
Photographs. Photogram. Eng., Vol. 18: 123-134.

(1952b) Growth of Ice-wedge Polygons In Permafrost Near Barrow,
Alaska. Geol. Soc. Am. Bull., Vol. 63: 1235-1236.

. (1965) Ice-wedge Casts Of Wisconsin. Wis. Acad. Sci., Arts, and
Lets, V01. 54: 187-222e

 

. (1976) Periglacial Features Indicative of Permafrost: Ice and
Soil Wedges. Quat. Res., Vol. 6: 3-26.

Bowlby, J. R. (1975) Late Glacial Ice Wedge Casts in the Kingston Basin
of Eastern Lake Ontario. XVIII Conference on Great Lakes Research.
International Assoc. for Great Lakes Research. May, 1975. Albany,
N. Y.

 

140

141

Brown, R. J. E. (1967) Comparison of permafrost conditions in Canada and
the U.S.S.R. Polar Record, Vol. 13: 741-751.

 

Brown, R. J. E. and Kupsch, W. D. (1974) Permafrost Terminolggy. Ottawa:
Nat. Res. Council of Can., Tech. Memo. No. 111. 62p.

Brown, R. J. E. and Peme: T. L. (1973) Distribution of Permafrost in North
America and Its Relationship to the Environment: A Review, 1963-1973.
North American Contribution; Permafrost, Second International Con-
ference. Washington: National Academy of Sciences: 71-100.

Brunnschweiler, D. (1962) The Periglacial Realm in North America During
the Wisconsin Glaciation. Biul. Peryglac., Vol. 11: 15-27.

. (1964) Der Pleistozane Periglazialbereich in Nord Amerika. Zeit-
schrift Ffir Geomorph., Bd. 8: 223-231.

. (1969) Periglacial Relics In The Great Lakes Area. 73rd Meeting,
Michigan Acad. Science, Arts and Letters. Univ. of Michigan, Ann
Arbor.

Brunnschweiler, D. and Lusch, D. P. (1977) Physiographic Regions. '13
Sommers, L. M. (ed.). Atlas of Michigan. East Lansing: Michigan
State Univ. Press: 24-25.

Buckley, J. (1976) Isotopes' Radiocarbon Measurements XI. Radiocarbon,
Vol. 18, No. 2: 172-189.

Buckley, J. et al. (1968) Isotopes' Radiocarbon Measurements VI. Radio-
carbon, Vol. 10, No. 2: 246-294.

Bfidel, J. (1951) Die klimazonen des Eiszeitalters. Eiszeitalter und
Gegenwart, Vol. 1: 16-26.

Burgis, W. A. (1970) The Imlay Outlet of Glacial Lake Maumee, Imlay City,
Michigan. Unpublished M.S. thesis, Univ. of Michigan. 74p.

Byrne, J. (1975) Subsurface wedge Structures in the Leamington Moraine,
Essex County, Ontario. Unpublished M.A. thesis, Univ. of Windsor.
87p.

Calkin, P. E. (1970) Strand lines and chronology of the glacial Great
Lakes in northwestern New York. Ohio J. Sci., Vol. 70: 78-96.

Calkin, P. E. and McAndrews, J. H. (1969) Dating late glacial recession
and vegetation in the Erie basin, northwestern New York. Geol. Soc.
Am. AbstraCts with Programs for 1969. Part 1: 5.

Chico, R. J. (1963) Playa Mudcracks: Regular and King Size. Geol. Soc.
Am.p§pec. Paper No. 76: 306.

. (1968) Soil Cracks--Giant. ‘I2_Fairbridge, R. W. (ed.). Encyclo-
pedia of Geomoppholggz, N. W.: Reinhold Book Corp.: 1028-1029.

142

Christensen, L. (1978) Personal communication.

Clayton, L. and Bailey, P. K. (1970) Tundra Polygons In The Northern Great
Plains. Geol. Soc. Am. Abstracts with Programs, Vol. 2, No. 6: 382.

Corte, A. E. and Higashi, A. (1964) Experimental Research On Desiccation
Cracks In Soil. U.S. Army Cold Reg. Res. Eng. Lab., Res. Rpt. 66.
72p.

Costin, A. B. (1955) A Note on Gilgaies and Frost Soils. J. Soil Sci.,

 

Czudek, T. and Demek, J. (1970) Thermokarst in Siberia and Its Influence
on the Development of Lowland Relief. Quat. Res., Vol. 1: 103-120.

Danilov, I. D. (1973) Subaqueous Pseudomorphs in Pleistocene Deposits.
Biul. Peryglac., Vol. 22: 339-345.

Danilova, N. S. (1956) Soil Wedges and Their Origin. .Ip Meister, L. A.
(ed.). Data on the Principles of the Study of Frozen Zones in the
Earth's Crust. Ottawa: National Research Council of Canada, Tech.
Translation 1088 (1964): 90-99.

Deevey, E. 8., Jr. (1951) Late-glacial and postglacial pollen diagrams
from.Maine. Am. J. Sci., Vol. 249: 177-207.

Dostovalov, B. N. and Popov, A. I. (1966) Polygonal Systems of Ice-wedges
and Conditions of Their Development. Proc., Permafrost International
Conf. Washington: National Academy Sci., National Research Council,
Pub. 1287: 102-105.

Dreimanis, A. (1966) Lake ArkonaeWhittlesey and post-Warren radiocarbon
dates from "Ridgetown Island" in southwestern Ontario. Ohio J. Sci.,
Vol. 66: 582-586.

 

. (1977) Late Wisconsin glacial retreat in the Great Lakes region,
North America. Ann. N. Y. Acad. Sci., Vol. 288: 70-89.

. (1979) The Problems of Waterlain Tills. lg Schluchter, Ch. (ed.).
Moraines and Varves. Rotterdam: A. A. Balkema: 167-177.

Dreimanis, A. and Goldthwait, R. P. (1973) Wisconsin glaciation in the
Huron, Erie and Ontario lobes. .93 Black, R. F., Goldthwait, R. P.
and Willman, H. B. (eds.). The Wisconsinan Stag . Geol. Soc. Am.
Men. 136: 71-106.

Dylik, J. (1966) Problems of ice wedge structures and frost fissure poly-
gons. Biul. Peryglac., Vol. 15: 241-291.

. (1968) Thermokarst. .In Fairbridge, R. W. (ed.). The Engyclopedia
of Geomorphology. N. Y.: Reinhold Book Corp.: 1149-1151.

 

143

Ekblaw, G. E. and Willman, H. B. (1955) Farmdale drift near Danville,
Illinois. Illinois Acad. Sci. Trans., Vol. 47: 129-138.

Embleton, C. and King, C. A. M. (1975) Periglacial Geomorphology. N. Y.:
John Wiley and Sons, Inc. 203p.

Eschman, D. F. (1978) Pleistocene Geology of the Thumb Area of Michigan.
'13 Keeling, R. V. (ed.). Field Excursions, The North-Central Section,
Geological Society of America meeting, Ann Arbor, Michigan: 35-62.

Farrand, W. R., Zahner, R. and Benninghoff, W. S. (1969) Cary-Port Huron
Interstade: evidence from a buried bryophyte bed, Cheboygan County,
Michigan. Geol. Soc. Am. Spec. Paper 123: 249-262.

Flemal, R. C. (1976) Pingo and Pingo Scars: Their Characteristics, Dis-
tribution and Utility in Reconstructing Former Permafrost Environments.

gust. Res., Vol. 6: 37-53.

Flemal, R. C., Hinkley, K. C. and Hesler, J. L. (1973) DeKalb Mounds: A
Possible Pleistocene (Woodfordian) Pingo Field in North-Central Illinois.
.23 Black, R. F., Goldthwait, R. P. and Willman, H. B. (eds.). The Wis-
consinan Stage. Geol. Soc. Am. Memoir 136: 229-250.

 

Flint, R. F. (1971) Glacial and Quaternary Geology. N. Y.: John Wiley and
Sons, Inc. 892p.

Florin, Maj-Britt and Wright, H. E., Jr. (1969) Diatom evidence of the per-
sistence of stagnant glacial ice in Minnesota. Geol. Soc. Am. Bull.,
vol. 80: 695-704.

Francis, E. (1975) Glacial sediments: a selective review. '12 Wright, A. E.
and Moseley, F. (eds.). Ice Ages: Ancient and Modern. Geol. Jour.
Spec. Issue 6: 43-68.

French, H. M. (1974) Active Thermokarst Processes, Eastern Banks Island,
Western Canadian Arctic. Can. J. Earth Sci., Vol. 11: 785-794.

. (1975) Man-Induced Thermokarst, Sachs Harbour Airstrip, Banks Island,
Nerthwest Territories. Can. J. Earth Sci., Vol. 12: 132-144.

 

. (1976) The Periglacial Environment. N. Y.: Longman, Inc. 309p.

Friedman, J. D. et a1. (1971) Observations on Icelandic Polygon Surfaces
and Palsa Areas. Photo Interpretation and Field Studies. Geografiska
Annaler, Vol. 53A: 115-145.

Frye, J. C. and Willman, H. B. (1958) Permafrost Features Near The Wis-
consin Glacial Margin In Illinois. Am. J. Sci., Vol. 256: 518-524.

. (1973) Wisconsinan climatic history interpreted from.Lake Michigan
Lobe deposits and soils. .EE Black, R. F., Goldthwaite, R. P. and
Willman, H. B. (eds.). The Wisconsinan Stag_. Geol. Soc. Am. Memoir
136: 135-152.

144

Gamble, E. (1958) Descriptions and interpretations of some Pleistocene
sections in Wayne County, Indiana. Earlham College Sci. Bull. 3.
41p.

Gold, L. W. and Lachenbruch, A. H. (1973) Thermal Conditions In Permafrost-
A Review of North American Literature. Permafrost: North American
Contributions to the Second International Conference. Washington:
National Academy of Sciences: 3-23.

Goldthwait, R. P. (1959) Scenes in Ohio During the Last Ice Age. Ohio J.
Sci., Vol. 59, No. 4: 193-216.

. (1968) Two loesses in central southwest Ohio. .15 Bergstrom, E. R.
(ed.). The Quarternary of Illinois. Univ. of Illinois College Agr.
Spec. Pub. 14: 41-47.

Gravenor, C. P. and Kupsch, W. O. (1959) Ice-disintegration features in
Western Canada. J. Geol., Vol. 67: 48-64.

Greenhouse, J. P. and Morgan, A. V. (1977) Resistivity mapping of fossil
permafrost patterns in southwestern Ontario. Can. J. Earth Sci.,
Vol. 14: 496-500.

Gruhn, R. and Bryan, A. L. '(1969) Fossil Ice-Wedge Polygons In Southeast
Essex, England. E Pewe, T. L. (ed.). The Perijlacial Environment:
Past and Present. Montreal: McGill Univ. Press: 351-363.

 

Hallsworth, E. C., Robertson, G. K. and Gibbons, F. R. (1955) Studies in
Pedogenesis in New South Wales. VII The "Gilgai" soils. J. Soil
Sci., Vol. 6: 1-31.

Hopkins, D. M. et a1. (1955) Permafrost and Ground Water In Alaska. U.S.
Geol. Surv. Prof. Paper 264-F: 113-146.

Horberg, L. (1949) A Possible Fossil Ice Wedge In Bureau County, Illinois.
J. Geol., Vol. 57: 132-136.

Hough, J. L. (1963) The Prehistoric Great Lakes of North America. Am.
Scientist, Vol. 51: 84-109.

Hunt, C. B. and washburn, A. L. (1966) Patterned Ground. U.S. Geol. Surv.
Prof. Paper 494-B: 104-125.

Hussey, K. M. and Michelson, R. W. (1966) Tundra relief features near
Point Barrow, Alaska. Arctic, Vol. 19: 162-184.

Hutchison, D. E. (1979) Soil Survey of Midland County, Michigan. Washing-
ton: U.S.D.A., Soil Conservation Service. 98p.

Jakosky, J. J. (1950) Exploration Geophysics. Newport Beach, CA: Trija
Pub. Co. 1195p.

Johnsson, G. (1959) True and False Ice-wedges In Southern Sweden. Geo-
fiska Annaler, Vol. 41: 15-33.

145

Johnsson, G. (1963) Periglacial Ice-wedge Polygons At Hasselholm, Southern-
most Sweden. Svensk Geogr. Arsb., Vol. 39: 173-176.

. (1981) Fossil patterned ground in southern Sweden. Geologiska
Foreningen: Stockholm Forhandlingar, Vol. 103: 79-89.

Judson, S. and Andrews, G. W. (1955) Pattern and form of some valleys in
the Driftless Area, Wisconsin. J. Geol., Vol 63: 328-336.

Kachurin, S. P. (1962) Thermokarst Within the Territory of the U.S.S.R.
Biul. Peryglac., Vol. 11: 49-55.

Kapp, R. O. (1977) Late Pleistocene and postglacial plant communities of
the Great Lakes Region. ‘Ig_Roman, R. C. (ed.). Geobotany. N. Y.:
Plenum Pub. Co.: 1-27.

Karrow, P. F. et a1. (1975) Stratigraphy, Paleontology and Age of Lake
Algonquin Sediments in Southwestern Ontario, Canada. ant. Res.,

Katasonov, E. M. (1973) Classification of frost-caused phenomena with
references to the genesis of the sediments in Central Yakutia.
Biul. Peryglac., Vol. 23: 71-80.

Kerfoot, D. E. (1972) Thermal contraction cracks in an arctic tundra
environment. Arctic, Vol. 25: 142-150.

Knechtel, M. M. (1951) Giant Playa Crack Polygons in New Mexico Compared
With Arctic Tundra Crack Polygons. Geol. Soc. Am. Bull., vol. 62:
1455.

Kostyaev, A. G. (1969) Wedge and Fold-like Diagenic Disturbances in
Quaternary Sediments and Their Paleogeographic Significance. Biul.
Peryglac., Vol. 19: 231-270.

Krumbein, W. C. and 31033, L. L. (1963) Stratigraphy and Sedimentation.
San Francisco: W. H. Freeman and Co. 660p.

Lachenbruch, A. (1960) Thermal contraction cracks and ice wedges in
permafrost. U. S. Geol. Surv. Prof. Paper 400B: 404-406.

. (1961) Depth and spacing of tension cracks. J. Geophys. Res.,
Vol. 66, No. 12: 4273-4291.

. (1962) Mechanics of thermal contraction cracks and ice-wedge
polygons in permafrost. Geol. Soc. Am. Spec. Paper 70: 69p.

. (1966) Contraction theory of ice-wedge polygons: A qualitative
discussion. In Proceedings, Permafrost International Conference.
Washington: National Academy Sciences, National Res. Council Pub.
1287: 63-70.

Lagarec, D. (1973) Postglacial Permafrost Features in Eastern Canada.
Nerth American Contribution; Permafrost, Second International
Conference. Washington: National Academy of Sciences: 126-131.

 

146

Lang, W. T. B. (1943) Gigantic drying cracks in Animas Valley, New Mexico.
Science, Vol. 98: 583-584.

Leffingwell, E. de K. (1915) Ground-ice wedges; the dominant form of
ground ice on the north coast of Alaska. J. Geol., Vol. 23: 635-654.

. (1919) The Canning River Region, North Alaska. U.S. Geol. Surv.
Prof. Paper 109. 251p.

Leighton, M. M. and Brophy, J. A. (1961) Illinoian glaciation in Illinois.
Jo G601. , v01. 69: 1-31e

Leverett, F. and Taylor, F. B. (1915) The Pleistocene of Indiana and
Michigan and the History of the Great Lakes. U.S. Geol. Surv. Mono-

ggaph 53. 529p.

Lewis, C. F. M. (1969) Late Quaternary History of Lake Levels in the
Huron and Erie Basins. Proc. 12th Conf. Great Lakes Res.: 250-270.

Livingstone, D. A. (1955) Some pollen profiles from arctic Alaska. Ecology,
Vol. 36: 587-600. -

Lusch, D. P. (1977) The origin and paleoclimatic significance of patterned
ground in the Saginaw basin. A proposal for dissertation research sub-
mitted to the faculty of the Department of Geography, Michigan State
University. 23p.

Mackay, J. R. (1967) Underwater patterned ground in artificially drained
lakes, Garry Island, N. W. T. Geogr. Bull. 9: 33-44.

. (1974) Ice-wedge cracks, Garry Island, Northwest Territories. Can.
J. Earth Sci., Vol. 11: 1366-1383.

. (1975) The closing of ice-wedge cracks in permafrost, Garry Island,
Northwest Territories. Can. J. Earth Sci., Vol. 12: 1668-1674.

Mackay, J. R., Konishchev, V. N. and Popov, A. I. (1979) Geologic controls
of the origin, characteristics and distribution of ground ice. Proc.
Third International Conf. on Permafrost, Vol. 2: l-18.

Mahjoory, R. and Whiteside, E. P. (1976) Soils of Saginaw County, Michigan,
1976. East Lansing: Michigan State University, Department of Crop
and Soil Sciences. 379p.

Martin, H. M. (1955) Map of the Surface Formations of the Southern Penin-
sula of Michigan. Michigan Dept. of Conserv., Geol. Surv. Div.,
Pub. 49.

Michigan Weather Service. (1974) The Climate of Michigan. Lansing:
Michigan Dept. of Agriculture. 275p.

Miller, N. G. and Benninghoff, W. S. (1969) Plant fossils from a Cary-
Port Huron Interstade deposit and their paleoecological interpretation.
Geol. Soc. Am. Spec. Paper 123: 225-248.

147

Mokma, D. L. (1978) Soil Management Units and Land Use Plannimg. East
Lansing: Michigan State University, Cooperative Extension Service,

Mokma, D. L. and Whiteside, E. P. (1974) Soils of Tuscola Counpy, Michi-
gan, 1974. East Lansing: Michigan State University, Department of
Crop and Soil Sciences. 388p.

 

Moon, J. W. et a1. (1938) Soil Survey of Saginaw Countyerichigan.
Washington: U.S.D.A., Bureau of Chemistry and Soils. 53p.

Morgan, A. V. (1972) Late Wisconsinan ice-wedge polygons near Kitchener,
Ontario, Canada. Can. J. Earth Sci., Vol. 9, No. 6: 607-617.

. (1982) Distribution and probable age of relict permafrost features
in south-western Ontario. Proc. Fourth Canadian Permafrost Conference.
Ottawa: National Research Council: 91-100.

Morgan, A. V., Elias, S. A. and Morgan A. (1981) Paleoenvironmental impli-
cations of a late-glacial insect assemblage from south-east Michigan.
Geol. Assoc. Can. Annual Meeting, Calgary.

Muller, E. H. (1965) Quaternary geology of New York. ‘Ig_Wright, H. E.,
Jr. and Frey, D. G. (eds.). The Quaternary of the United States.
Princeton, N. J.: Princeton Univ. Press: 99-112.

Muller, S. W. (1947) Permafrost or permanent1y_frozen;ground and related
engineering problems. Ann Arbor, Michigan: J. W. Edwards. 231p.

 

Neal, J. T. (1972) Playa surface features as indicators of environment.
'13 Neal, J. T. (ed.). Playas and Dried Lakes. Stroudsburg, PA:
Dowden, Hutchinson and Ross, Inc.: 363-388.

Neal, J. T. and Motts, W. S. (1967) Recent Geomorphic Changes in Playas
of Western United States. J. Geol., Vol. 75, No. 5: 511-525.

Neal, J. T., Langer, A. M. and Kerr, P. F. (1968) Giant Desiccation Poly-
gons of Great Basin Playas. Geol. Soc. Am. Bull., Vol. 79: 69-90.

Nicholson, F. H. (1969) An investigation of patterned ground. Unpublished
Ph.D. dissertation, Univ. of Bristol, England.

Nicholson, F. H. and Granberg, H. B. (1973) Permafrost and snowcover rela-
tionships near Schefferville. Permafrost: North American Contribution
to the Second International Conference. Washington: National Academy
of Sciences: 151-158.

Ogden, J. C. III. (1977) The late Quaternary paleoenvironmental record of
northeastern North America. Ann. N. Y. Acad. Sci., Vol. 288: 16-34.

Ohrngren, S. (1967) Polygon fields on the Laksefjord Finnmark. Lund
Studies In Geography, Ser. A., Physical Geog., No. 40: 58-67.

 

148

Parasnis, D. S. (1973) Mining Geophysics. N. Y.: Elsevier Scientific
Publishing Co. 395p.

Parizek, R. R. (1969) Glacial ice-contact rings and ridges. Geol. Soc.
Am. Spec. Paper 123: 49-102.

Perrin, R. M. S. (1963) The use of aerial photographs in the study of
patterned ground in East Anglia. Trans. Symp. on Photo Interpre-
tation (Archives Internationales De Photogrammetrie): 183-188.

Pewe: T. L. (1954) Effect of permafrost on cultivated fields, Fairbanks
area, Alaska. U. S. Geol. Surv. Bull. 989-F: 315-349.

. (1964) Ice wedges in Alaska - Classification, distribution and
climatic significance. Geol. Soc. Am. Spec. Paper 76. 129p.

. (1966a) Ice wedges in Alaska: Classification, distribution and
climatic significance. Proc., International Permafrost Conference.
Washington: National Academy of Sciences, National Research Council,
Pub. 1287: 76-81.

. (1966b) Paleoclimatic significance of fossil ice wedges. Biul.
Peryglac., Vol. 15: 65-73.

. (1973) Ice-wedge casts and past permafrost distribution in Nerth
America. Geoforum, Vol. 15: 15-26.

. (1975) Quaternary Geology of Alaska. U. S. Geol. Surv. Prof.
Paper 835. 145p.

I I
Pewe, T. L., Church, R. E. and Andresen, M. J. (1969) Origin and paleo-
climatic significance of large-scale patterned ground in the Donnelly
Dome area, Alaska. Geol. Soc. Am. Spec. Paper 103. 87p.

Pissart, A. (1970 Les phemomenes physiflues essentielles liem au gel, les
structures periglaciaires qui en resultent et leur signification
climatique. Soc. geol. Belgique Annalee, Vol. 93: 7-49.

 

Porsild, A. E. (1957) Illustrated flora of the Canadian Arctic Archipelago.
National Museum of Canada Bull. 146. 209p.

Potzger, J. E. (1946) Phytosociology of the primeval forest in central-
northern Wisconsin and upper Michigan and a brief post-glacial history
of the Lake Forest formation. Ecological Monographs, Vol. 16, No. 3:
227-250.

. (1948) A pollen study in the tension zone of lower Michigan.
Butler Univ. Bot. Stud., Vol. 8: 161-177.

. (1951) The fossil record near the glacial border. Ohio J. Sci.,
Vol. 51: 126-133.

Presant, E. W. and Protz, R. (1967) The origin of soil tongues in some
Ontario soils. Geol. Mineral. Assoc. Can. and International Joint
Meetings, Kingston, Ontario.

149

Rapp, A. and Clark, C. M. (1971) Large nonsorted polygons in Padjelanta
National Park, Swedish Lappland. Geggrafiska Annaler, Vol. 53A:
71-85.

Rapp, A. and Rudberg, S. (1964) Studies on periglacial phenomena in
Scandinavia, 1960-1963. Biul. Peryglac., Vol. 14: 75-89.

Ritchie, J. C. and Litchi-Federovich, S. (1967) Pollen dispersal phenomena
in arctic-subarctic Canada. Rev. Paleobotany and Palynology, Vol. 3:
255-268.

Romanovskij, N. N. (1973) Regularities in formation of frost-fissures and
development of frost-fissure polygons. Biul. Pepyglac., Vol. 23:
237-277.

Rubin, M. and Alexander, C. (1958) U. 8. Geological Survey radiocarbon
dates IV. Science, Vol. 127: 1476-1487.

. (1960) U. 8. Geological Survey radiocarbon dates V. Am. J. Sci.,
Radiocarbon Supplement, Vol. 2: 129-185.

Rubin, M. and Berthold, B. (1961) U. S. Geological Survey radiocarbon
dates VI. Radiocarbon, Vol. 3: 86.

Sharp, R. P. (1942) Periglacial involutions in northeastern Illinois.
J. Geol., Vol. 50, No. 2: 113-133.

Shilts, W. W. and Dean, W. E. (1975) Permafrost features under arctic
lakes, District of Keewatin, N. W. T. Can. J. Earth Sci., Vol. 12,
No. 4: 649-662.

Shumskiy, P. A. and Vtyurin, V. A. (1966) Underground ice. Proc. Perma-
frost International Conference. Washington: National Academy of
Sciences, National Research Council Pub. 1287: 108-113.

Smith, H. T. U. (1949) Physical effects of Pleistocene climatic changes
in nonglaciated areas: eolian phenomena, frost action and stream
terracing. Geol. Soc. Am. Bull., Vol. 60: 1485-1516.

. (1962) Periglacial frost features and related phenomena in the
U. S. Biul. Peryglac., Vol. 11: 325-342.

Soil Conservation Service. (1978a) Londo Series; Soil Interpretations
Record. National Cooperative Soil Survey, Mimeo.

. (1978b) Parkhill Series; Soil Interpretations Record. National
Cooperative Soil Survey, Mimeo.

. (1979a) Capac Series; Soil Interpretations Record. National
Cooperative Soil Survey, Mimeo.

. (1979b) Tappan Series; Soil Interpretations Record. National
Cooperative Soil Survey, Mimeo.

150

Soloviev, P. A. (1973) Thermokarst phenomena and landforms due to frost
heaving in central Yakutia. Biul. Peryglac., Vol. 23: 135-155.

Steere, W. C. .(1947) Musci. .Ig Polunin, N. Botany Of The Canadian
Eastern Arctic, Pt. 11. Ottawa: National Mus. Can. Bull 97:

370-490.

Stoner, E. R. et a1. (1980) Extension of laboratory-measured soil spectra
to field conditions. Soil Sci. Soc. Am. J., Vol. 44: 572-574.

Stoutamire, W. P. and Benninghoff, W. S. (1964) Biotic assemblage asso-
ciated with a Mastodon skull from Oakland County, Michigan. Pap.
Mich. Acad. Sci. Arts Let., Vol. 46: 47-60.

Straw, A. (1966) Periglacial mass movement on the Niagara Escarpment
near Meaford, Grey County, Ontario. Geogr. Bull., Vol. 8: 369-376.

Suess, H. (1954) U. S. Geological Survey radiocarbon dates I. Science,

Sutton, R. F. (1963) Involutions in surficial deposits, NW Ontario.
Geol. Soc. Am. Bull., Vol. 74: 789-794.

Svensson, H. (1964) Aerial photographs for tracing and investigating fos-
sil tundra ground in Scandinavia. Biul. Pepyglac., vol. 14: 321-325.

. (1970) Termokarst. Svensk Geografisk Arsbok, Vol. 46: 114-126.

. (1972) The use of stress situations in vegetation for detecting
ground conditions on aerial photographs. Photogrammetria, Vol. 28:
75-87.

. (1973) Distribution and chronology of relict polygon patterns on
the Laholm Plain, the Swedish west coast. Geografiska Annaler,
Vol. 55A: 159-175.

. (1976) Relict ice-wedge polygons revealed on aerial photographs
from Kaltenkirchen, northern Germany. Geogpafisk Tidsskrift, Bd. 75:
8-12.

Tillema. G. A. (1972) Fossil ice-wedge polygons in the Saginaw basin.
Unpublished M.A. research paper, Michigan State University. 64p.

Trask, P. D. (1932) Origin and Environment of Source Sediments of Petro-
leum. Houston: Gulf Publishing Co. 67p.

Tricart, J. (1969) Geomorphology of Cold Environments. N. Y.: St.
Martin's Press. 320p.

Walters, J. C. (1975) Origin and paleoclimatic significance of fossil
periglacial phenomena in central and northern New Jersey. Unpublished
Ph.D. dissertation, Rutgers Univ. 147p.

. (1978) Polygonal patterned ground in central New Jersey. gust. Res.,

151

Washburn, A. L. (1950) Patterned ground. Rev. Can. de Gebgraphie, Vol. 4:
5-59e

. (1956) Classification of patterned ground and review of suggested
origins. Geol. Soc. Am. Bull., Vol. 67, No. 7: 823-865.

. (1970) An approach to a genetic classification of patterned ground.
Acta Geographica Lodziensia, Vol. 24: 437-446.

. (1973) Periglacial processes and environments. N. Y.: St. Martin's
Press. 320p.

(1978) Personal communication.
. (1980) Geocgyology. N. Y.: John Wiley and Sons, Inc. 406p.

Washburn, A. L., Smith, D. D. and Goddard, R. H. (1963) Frost cracking
in a middle-latitude climate. Biul. Peryglac., Vol. 12: 175-189.

Wayne, W. J. (1956) Pleistocene periglacial environment in Indiana.
Indiana Acad. Sci. Proc., Vol. 65: 164.

. (1963a) Pleistocene patterned ground and periglacial temperatures
in Indiana. Geol. Soc. Am. Spec. Paper 76: 176-177.

. (1963b) Pleistocene formations of Indiana. Indiana Geol. Surv.
Bull. 25. 85p.

. (1965) The Crawfordsville and Knightstown Moraines in Indiana.
Indiana Geol. Surv. Rept. Prog. 28. 15p.

. (1967) Periglacial features and climatic gradient in Illinois.
Indiana and western Ohio, east-central U. S. .23 Cushing, E. J. and
wright, H. E., Jr. (eds.). Quaternary Paleoecology. (Vol. 7 of the
V11 Cengress of INQUA). New Haven, CT: Yale Univ. Press: 393-414.

. (1968) The Erie lobe margin in east-central Indiana during the
Wisconsin glaciation. Ind. Acad. Sci. Proc., Vol. 77: 277-291.

Weesies, G. A. (1980) Soil Survey of Bay County, Michigan. Washington:
U.S.D.A., Soil Conservation Service. 105p.

White, E. M. (1971) Contemporary soil wedge formation in western South
Dakota. Soil Sci. Soc. Am. Proc., Vol. 35: 306-309.

Wilden, R. and Mabey, D. R. (1961) Giant desiccation fissures on the
Black Rock and Smoke Creek Deserts, Nevada. Science, Vol. 133:
1359-1360.

Williams, R. B. G. (1964) Fossil patterned ground in eastern England.
Biul. Peryglac., Vol. 14: 337-349.

 

152

Wilson, L. R. (1958) Polygonal structures in the soil of central Iowa.
Oklahoma Geology Notes, Oklahoma Geol. Surv., Vol. 18, No. 1: 4-6.

Wright, H. E., Jr. (1971) Late Quaternary vegetational history of Nerth
America. .12 Turekian, K. K. (ed.). The Late Cenozoic Glacial Age .
New Haven, CT: Yale Univ. Press: 425-464.

Yehle, L. A. (1954) Soil tongues and their confusion with certain indi-
cators of periglacial climate. Am. J. Sci., Vol. 252: 532-546.