...

 

4...,

u.

“vs-a...

 

~~.-

I'm.

Illllllllllllll

 

This is to certify that the

thesis entitled .

GLACIAL HISTORY OF SLEEPING BEAR DUNES NATIONAL
LAKESHORE, NORTHWESTERN LOWER MICHIGAN: EVIDENCE FOR
LATE PLEISTOCENE AND HOLOCENE LAKE LEVEL CHANGES

 

presented by

Todd Erick Wallbom

has been accepted towards fulfillment
.of the requirements for

M.~S. ‘ degreeinfifinlngical Sciences

M ' r professor

Datewzv; 2W5

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution 1?
’ l
r

 

 

 

LIBRARY

University

Michigan State

 

 

 

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

 

DATEjDUE I DATE DUE

DAEEIDUE

 

\T/

”“332: 6 i Z“)?

 

W 2 APR°04 may“
‘7 0023*?“ -

 

inr172m0

 

BE? ”12“im 1 0

 

 

 

 

 

 

 

 

 

 

11/00 Woes-p.14

 

 

GLAClAl
NORWI.‘

GLACIAL HISTORY OF SLEEPING BEAR DUNES NATIONAL LAKESHORE,
NORTHWESTERN LOWER MICHIGAN: EVIDENCE FOR LATE PLEISTOCENE
AND HOLOCENE LAKE LEVEL CHANGES

By

Todd Erick Wallbom

A THESIS

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

MASTER OF SCIENCE

Department of Geological Sciences

2000

GLACII‘:
NORTH.

IOIIOwj
rapres
Huron
Great
lewla
nage:

Lake

ABSTRACT

GLACIAL HISTORY OF SLEEPING BEAR DUNES NATIONAL LAKESHORE,
NORTHWESTERN LOWER MICHIGAN: EVIDENCE FOR LATE PLEISTOCENE
AND HOLOCENE LAKE LEVEL CHANGES

By

Todd Erick Wallbom

Despite several decades of research, fundamental questions have
challenged glacial geologists working in the Sleeping Bear Dunes National
Lakeshore (“the Lakeshore”) area located in northwestem lower Michigan.
These include: 1) what is the origin of the high, prominent, transverse ridges that
characterize the Lakeshore, 2) did Greatlakean ice extend into the region, and 3)
what glacial lake stage followed the ice.

Recent surficial mapping at the Lakeshore by the author suggests the
following: that the transverse ridges oriented roughly northwest-southeast likely
represent subaqueous crevasse-fill deposits formed within disintegrating Port
Huron ice over Port Huron-age till; that there is no conclusive evidence for the
Greatlakean ice advance but it probably did extend as thin lobes into the
lowlands; and a strand at about 222 to 225 m elevation extends along many
ridges at the Lakeshore and probably represents the Calumet stage of glacial
Lake Chicago.

This thesis is dedicated to Aggie.

aflofhe

 

overall s.I
motivate
commits
Providing

support a

 

“Wan
the hydrog
WI Arc/Ir
‘97-’93) th
very SUpp
Intemrete
think of a
Aggie, f0,

IOI going I

ACKNOWLEDGMENTS

I would like to extend my sincere thanks to the following people for
all of their help, support, and guidance: my advisor, Dr. Grahame Larson, for his
overall superb teaching, wisdom, and critical comment (as well as for those
motivational and inspirational field discussions), to the other members of my
committee, Dr. Randall Schaetzl, for his assistance with lake levels as well as for
providing much-needed feedback, and Dr. William Cambray, for his experience,
support, and understanding. In addition, I am grateful to those in the GIS
cartography lab (John Linker for doing a great digitizing job) as well as those in
the hydrogeology lab, especially Dave Bout and Chris Hoard, for their expertise
with Arc/Info and Archew. I would also like thank the “F.O.G. house” dwellers,
‘97-’98, the “Riv crew”, the “Matanuska Mafia” camp crew, summer ’97, and the
very supportive and knowledgeable park rangers (especially Roger and Max),
interpreters, and staff at Sleeping Bear Dunes National Lakeshore. I couldn’t
think of a better field area to do a thesis! But most of all, I’d like to thank my wife,
Aggie, for providing a secure environment in which to work, for showing interest,
for going over seemingly endless thesis revisions without complaint, for keeping
me focused and on track, and especially for her unselfishness, incredible

patience, caring, and compassion.

LIST OF

LIST OF

 

INTROD
Gene

Histor'
Th'

Ar:

 

(nTrnmw-IH

TABLE OF CONTENTS

Page
LIST OF TABLES ............................................... vii
LIST OF FIGURES .............................................. viii
INTRODUCTION
General Statement ........................................... 1
Purpose of this investigation ................................ 2
Historical Background
The Lake Michigan Lobe and its fluctuations ..................... 4
Extent of ice advances in northwestern lower Michigan .......... 6
Arguments on the extent of Greatlakean ice in the
Sleeping Bear Dunes National Lakeshore region ............ 9
Arguments on the origin of the ridges at Sleeping Bear
Dunes National Lakeshore ............................... 13
Glacial and post-glacial lake levels in the Lake Michigan basin ..... 16
Glacial Lake Chicago .................................... 16
Post-glacial lakes ....................................... 18
Arguments on glacial to—post-glacial lake levels in the
Sleeping Bear Dunes National Lakeshore region ............ 19
SECTION II: GEOLOGIC UNITS AND LANDFORMS
Methods
Field mapping ............................................ 23
Surficial deposits .......................................... 25
Well-log data ............................................. 25
Measurement of sections ................................... 26
Analysis of geomorphology .................................. 27
Sampling strategy ....................................... 28
Map assembly ............................................ 30
Data
Bedrock geology .......................................... 31
Drift thickness .......................................... 34
Surficial geology ........................................... 34
Leland formation ........................................ 34
Duck Lake diamict ...................................... 39
Pyramid Point deposit .................................... 41
Pearl Lake deposit ...................................... 44
Sand and gravel-undifferentiated ........................... 48
Honor deposit .......................................... 50
Stream terrace deposit ................................... 52

Glen Lake deposit ....................................... 53

Sleeping Bear deposit .................................... 54

Beach/dune complexes ................................... 54

Fan deposit ............................................ 56

Modern delta and terrace deposit ........................... 57

Marsh and swamp deposit ................................ 58

Modern beach deposit ................................... 59

Geomorphology ........................................... 6O

Drumlins .............................................. 60

Ridges and kames ...................................... 61

Lowlands, embayments, and channels ...................... 64

Bluffs and scarps ....................................... 66

Dunes and beaches ..................................... 67
SECTION III: GLACIAL AND POST-GLACIAL HISTORY

Deglaciation History ........................................... 70
Extent and tills of the Port Huron and Greatlakean

ice advances ........................................... 83

Origin of the ridges at Sleeping Bear Dunes National Lakeshore ........ 89

A proposed conceptual model of a subaqueous crevasse-fill deposit. . 93

Discussion ............................................... 97

History of crevasse-fills .................................. 100

The extent of ridges (crevasse—fills?) in the Lake Michigan basin.102

Lake Level History ........................................... 106

Discussion .............................................. 1 12

Conclusions ................................................ 12

LIST OF REFERENCES ......................................... 124

vi

Table 1.

Table 2.
IIIOTII Eyi

Table 3.
Inset) f0'

Table 4.

 

LIST OF TABLES

Page
Table 1. Glacial lithofacies codification (adapted from Eyles et al., 1983) . . . . 29

Table 2. Diamict lithofacies code and symbols in addition to those in Table 1

(from Eyles et al., 1983) .......................................... 29
Table 3. Scarp elevations at the Lakeshore and vicinity (in meters). Note map

(inset) for survey site locations ..................................... 69
Table 4. Lakeshore paleostrandline elevations and interpretations ........ 113

vii

LIST OF FIGURES

Page

Figure 1. Map showing location of study area (from “Geology of the Southern
Peninsula of Michigan”, Leverett and Taylor, 1915) ..................... 3

Figure 2. Map showing the southern margin of landforms associated with Port
Bruce, Port Huron, Greatlakean, and North Bay Stades (adapted from Larson et
al. 1994 and Rovey II and Borucki, 1995) ............................. 7

Figure 3. Glacial timelstratigraphy of the Lake Michigan basin (from Hansel et
al.1985) ...................................................... 11

Figure 4. Glacial map showing distribution of moraines in the northern part of the
lower peninsula of Michigan (Melhom, 1954) .......................... 12

Figure 5. Map of the Lakeshore and vicinity showing major landforms (from
Leverett and Taylor, 1915) ........................................ 15

Figure 6. Comparison of Clark et al.’s (1994) pr6dicted Glenwood
height/distance profiles to Evenson (1973) and Taylor (1990) ............. 22

Figure 7. Comparison of Clark et al.’s (1994) predicted Algonquin shorelines to
data of Larsen (1987) ............................................ 22

Figure 8. Comparison of Clark et al.’s (1994) predicted Nipissing shorelines to
data of Larsen (1985) ............................................ 22

Figure 9. U.S.G.S. 7.5 minute Quadrangles associated with the Lakeshore . . 24

Figure 10. Map of bedrock units at the Lakeshore and vicinity (from Milstein et
al., 1987) ...................................................... 32

Figure 11. Map showing elevation of bedrock surface and well locations with
spot thicknesses of glacial drift (in meters a.s.l.) at the Lakeshore and vicinity
(from the Hydrologic Atlas of Michigan, Geologic Survey Division, Michigan
Department of Natural Resources, 1981) ............................. 33

Figure 12. Surficial Geology of the Glen Haven, Glen Arbor, Good Harbor Bay,

Empire, Burdickville, and Beulah 7.5 minute Quadrangles, Leelanau and Benzie
Counties, Michigan (Wallbom and Larson, 1998) ....................... 36

viii

Figure 13. Stratified sand and gravel of the Leland formation overlain by Duck
Lake diamict at the Leelanau County gravel pit ........................ 38

Figure 14. Laminated clay and silt of the Leland formation with some small-scale
normal faulting at Pyramid Point. Note deformed beds near pencil ......... 38

Figure 15. Detail of massive Duck Lake diamict superimposed on Leland
formation deposits, Leelanau County gravel pit ......................... 42

Figure 16. View upsection of a major shear plane (dashed line) in deformed
Duck Lake diamict shown slightly offset by post depositional slumping at Pyramid
Point .......................................................... 42

Figure 17. View of the bluff at Pyramid Point from Lake Michigan. Direction of
view is to the south .............................................. 43

Figure 18. Closer view of glaciolacustrine sediments at Pyramid Point. The
general direction of flow was from left to right ......................... 43

Figure 19. Stratigraphic sections at Pyramid Point: A-E. Refer to Figure 28 for
facies descriptions .............................................. 47

Figure 20. Thick (~2 in) sets of stratified, coarse-grained Pearl Lake glaciofluvial
sediments at the Kasson Sand and Gravel Mine, Leelanau County ......... 49

Figure 21. Closer view of Pearl Lake glaciofluvial sediments at the Kasson Sand
and Gravel Mine. Well-developed clast imbrication indicates that the general
current flow was southward ....................................... 49

Figure 22. The gravel pit of the lower delta northwest of Honor, Benzie County . .

.............................................................. 51
Figure 23. Detail of laminated sediments in the Honor delta complex ....... 51
Figure 24. A ~5-meter high terrace in an abandoned channel near Empire.
Direction of flow was from left to right. View is towards the southwest ....... 55
Figure 25. A large blowout surrounded by stabilized dunes perched atop
Sleeping Bear bluffs. Note possible Algonquin-age terrace (arrow) developed
near the southem end of South Manitou Island ........................ 55
Figure 26. View of the bluffs occurring near Empire. The view is towards the
northeast ...................................................... 63

Figure 27. Hypothetical extent of the Greatlakean ice margin conforming mainly
to lowland areas along the Lakeshore (z11,800 yrs B.P.) ................ 88

ix

 

Figure 28.
Paint .....

Figure 29.
crevasse fi'

Figure 30.
shown are
al. (1994)
Solid lines
moody d

Figure 31.
data of Eve

Figure 32.
(1999) in p

Figure 33.
DIBdICIEd c

Flsure 34.
Sens of Ii
yrs B.P.).
Arrows 3hr

 

 

Figure 28. Description and distribution of the main lithologic units at Pyramid
Point ........................................................ 92

Figure 29. A conceptual model and general depositional environment for mega
crevasse fillings at the Lakeshore ................................. 96

Figure 30. Height/distance plot of lake level elevations at the Lakeshore. Also
shown are shoreline elevations from Evenson (1973), Taylor (1990), and Clark et
al. (1994). Numbers on x-axis refer to survey sites listed in Table 3 (see map).
Solid lines signify well-developed strandlines and dashed lines signify moderately
to-poorly developed strandlines ................................... 108

Figure 31. Comparison of the Lakeshore Calumet shoreline elevation (1999) to
data of Evenson (1973), and Taylor (1990) .......................... 116

Figure 32. Comparison of the Lakeshore main Algonquin shoreline elevation
(1999) to predicted data of Clark et al. (1994) and to data of Larsen (1987) . .116

Figure 33. Comparison of the Lakeshore Nipissing shoreline elevation (1999) to
predicted data of Clark et al. (1994) and to data of Larsen (1985) ......... 116

Figure 34. Mature stage of Greatlakean ice retreat/stagnation with formation of a
series of large cascading lakes partially or completely ice blocked (11,800-10,700
yrs B.P.). Main channels or drainageways are shown by a stippled pattern.
Arrows show probable direction of drainage .......................... 119

 

 

 

NIQN.R

 

amsmz
xsnhl;
more the
and SCI.
3 iew OI
orchard

theNP.

LQbe
mere

1938.

INTRODUCTION

General Statement

Sleeping Bear Dunes National Lakeshore (or “the Lakeshore”; T.27
to 32N.,R.13 to 16W.) is located in northwestern lower Michigan (Figures 1, 2)
and is managed by the National Park Service (NPS). It includes about 30 km of
scenic Lake Michigan shoreline in Leelanau and Benzie Counties and covers
more than 72,000 acres of land and deeded lake surface area, as well as North
and South Manitou Islands. The Lakeshore area is mostly open wilderness with
a few old-growth forest stands, and includes some undeveloped land, active
orchards, and abandoned fields being returned to pre-settlement conditions by
the NPS.

The Lakeshore is characterized by a series of 40 to 129 m high, prominent
ridges that traverse the region, are oriented roughly northwest to southeast, and
often form dramatic headland bluffs along the Lake Michigan shoreline. Between
the ridges are broad, open, arcuate-shaped, lowland plains that inland from
shoreline embayments between the bluffs. Capping many of the bluffs are
extensive dune fields that include Michigan’s most famous dune, the Sleeping
Bear Dune, from which the name of the Lakeshore is derived.

Although it has long been recognized that glaciers from the Lake Michigan
Lobe (LML) intermittently covered the Lakeshore and that most of the landforms
there are of glacial origin (Leverett and Taylor, 1915; Waterman, 1926; Dow,
1938; Calver, 1947; Melhom, 1954; Taylor, 1990), many questions still remain

 

 

regarding 4.

 

 

regarding the deglaciation history of the region. These include the following: 1)
how and when were the prominent ridges at the Lakeshore formed, 2) did
Greatlakean ice advance into the region, and 3) what were the elevations of lake

levels in the northern Lake Michigan basin pre- and post-deglaciation?

Purpose of this Investigation

Recent geologic investigations (of glacial deposits and water-plane
elevations at the Lakeshore) by the author have uncovered evidence regarding
deglaciation and relative position of the ice margin since about 13,000 yrs B.P.
(Schneider, 1990; Schneider and Hansel, 1990). These findings have resulted in
the identification and facies descriptions of fourteen glacial and post-glacial units
at the Lakeshore along with Iandform interpretations and relative time of
formation estimates. In addition, they have generated new elevation data and
interpretations for five strandlines in the region.

The key issues being addressed in this thesis are: 1) map and describe
the surficial geologic deposits at the Lakeshore and discuss the stratigraphy, time
of deposition, and origin of the ridges; 2) define the extent of the Greatlakean ice-
margin; and 3) outline the glacial and post-glacial lake history in the Lakeshore
area from glacial Lake Chicago to modern Lake Michigan with emphasis on
differentiating high lake levels of glacial Lake Chicago.

 

 

t.fl.:— .CC.C.C<¢ HI!

6:03:04 05.30300; ~ u _

 

 

.33.. 5:5 2m 3293 .2822:
.6 m_:m:_con_ Eofiaow oz. .0 .6280. EB: mofi .633 Co 5:30. @5265 no.2 .F 0.59m

    

mus

      

3......(4NITM .

......K

I...
8

2...... 352.... B
9.533 2.3263 E

 

2559.3 3:262
355 Sam 9.336

 

 

when the]
8P (Mei.I
1985'. H:
mumple
Wiscon:
informa
1915-. i
1973;;

1982;|

Historical Background

The Lake Mich'gan Lobe and its fluctuations

The State of Michigan was last glaciated during the Late Wisconsinan
when the Laurentide ice sheet was at its maximum around 20,000 to 12,000 yrs
B.P (Melhom, 1954; Hough, 1958; Evenson, 1973; Drexler, 1975; Eschman,
1985; Hansel et al., 1985). During this period, the LML ice-margin underwent
multiple oscillations leaving behind a complex sedimentological record in eastern
Wisconsin, Illinois, Ohio, and southern Michigan, for which a wealth of
information is available (Chamberlin, 1877; Alden, 1904; Leverett and Taylor,
1915; Alden, 1918; Thwaites, 1943; Thwaites and Bertrand, 1957; Evenson,
1973; Evenson et al., 1974; Mickelson and Evenson, 1975; Farrand and Bell,
1982; Eschman, 1985; Hansel et al., 1985).

During the Late Wisconsinan (Woodfordian Substage, or oxygen isotope
Stage 2) along the eastern side of the Lake Michigan basin, four major episodes
of LML ice advances (Figures 2, 3) were marked by a series of morainal ridges
(terminal, recessional) built along the ice front (Farrand and Eschman, 1974;
Eschman, 1985). These are, from oldest to youngest, the Tekonsha, Sturgis-
Kalamazoo, Lake Border, and the Port Huron morainal complex; part of which
includes the Manistee moraine (Figure 4), culminating roughly at about 21,000,
15,500, 14,100, and 13,300 to 12,200 and 12,000 to 11,700 yis B.P. (for the Port
Huron morainal complex) respectively, until the state was free of ice at around

10,000 yrs B.P. (Thwaites, 1943; Bretz, 1951; Thwaites and Bertrand, 1957;

 

Broedr
Eschm
Hanse'
Gross

andIE

YIS BI

IOMMM
Inlerst

rEéoorr

rough

IFlgur

OSCIII;
(LEVe

1990;

deep.x
MIChIg
Gran<
Meth

Broecker and Farrand, 1963; Evenson, 1973; Farrand and Eschman, 1974;
Eschman, 1985; Hansel et al., 1985; Monaghan et al., 1986; Monaghan and
Hansel, 1990; Taylor, 1990). Advances were rapid over the basin; Kempton and
Gross (1971) estimate an advance rate of 62 mlyear for the LML between 23,000
and 18,000 yrs B.P.

A period of retreat, named the Erie lnterstade, occurred at about 15,600
yrs B.P. (Dreimanis and Goldthwait, 1973; Monaghan et al., 1986) and was
followed by several more closely spaced ice advances. During the Erie
lnterstade, the Lake Michigan basin was, for the most part, ice-filled as no clear
record exists of a glacial lake occupying the basin (Eschman, 1985).

Following the Erie lnterstade, several more LML advances occurred in the
roughly two-thousand year period between about 15,500 and 13,000 yrs B.P.
(Figure 3). During this period, named the Port Bruce stade (Farrand and
Eschman, 1974), the front of the LML fluctuated repeatedly with only minor
oscillations in eastern Wisconsin, northeastern Illinois, and northwestern Indiana
(Leverett and Taylor, 1915; Monaghan et al., 1986; Monaghan and Hansel,
1990).

A major advance at about 14,100 yrs B.P. reworked lake sediment and
deposited it in the Lake Border moraine along the rim of the eastern Lake
Michigan basin. Following this, the ice rapidly retreated north of the Glacial
Grand Valley during the Mackinac lnterstadial (Leverett and Taylor, 1915;
Melhom, 1954; Larson et al., 1994), which lasted until about 13,300 yrs B.P.

 

(Ferrari:

1985).

 

Extent 0‘

 

Interstac
basin pt:

Port Hurt

 

(Farrand and Eschman, 1974; Fullerton, 1980; Eschman, 1985; Hansel et al.,
1985)

Extent of ice advances in northwestern lower Michigan, Following the Mackinac
lnterstadial, at least three main LML advances occurred in the Lake Michigan
basin prior to final retreat (Figure 3). These include at least two closely-spaced
Port Huron ice advances and the Greatlakean, or Two Rivers, ice advance.
These advances occurred between about 13,300 to 12,200 yrs B.P. and between
about 12,000 to 11,700 yrs. B.P. respectively (Broecker and Farrand, 1963;
Evenson, 1973; Farrand and Eschman, 1974; Eschman, 1985; Taylor, 1990;
Larson et al., 1994).

The Port Huron stade ice advances, between 13,300 and 12,200 years
B.P. (Farrand and Eschman, 1974; Fullerton, 1980; Hansel et al., 1985; Blewett
and Winters, 1995), occurred at least twice in southern Michigan. The first Port
Huron ice advance formed the outer Port Huron moraine in Michigan (Figures 2,
3) between 13,300 and 13,000 yrs B.P. (Farrand and Eschman, 1974; Hansel et
al., 1985; Blewett, 1990) and can be traced from upstate New York to eastern
Wisconsin. The ice margin then stagnated and retreated northward from the
outer Port Huron moraine (Bretz, 1959; Hough, 1963; Farrand and Eschman,
1974; Hansel et al., 1985; Taylor, 1990) and meltwater quickly built outwash of
the Mancelona Plain which now occupies the swale between the outer and the
inner Port Huron moraines (Leverett and Taylor, 1915; Martin, 1957; Evenson,

1973; Farrand and Eschman, 1974; Blewett, 1990; Blewett and Winters, 1995).

 

 

 

OSN

0N .

{IN ‘

Fig UL
With *
AISC
I0W1;
(adal

 

 

Port Huron
Marginq‘

43W "-

41‘N «-

 

 

     

l.

a). ‘5'"

- 43'N

 

Figure 2. Map showing the southern margin of landforms associated
with Port Bruce, Port Huron, Greatlakean, and North Bay stades.

Also shown are the locations of Sleeping Bear Dunes National

Lakeshore (star), Cheboygan bryophyte bed (CBB), Indian River
lowland, Two Creeks forest bed (T CFB), and St. Francis Bluffs (SFB).
(adapted from Larson et al. 1994 and Rovey II and Borucki, 1995).

The second Port Huron advance (Figures 2, 3) occurred when ice
advanced to the position of the Manistee moraine between 13,000 to 12,200 yrs
B.P. (Evenson, 1973; Farrand and Eschman, 1974; Eschman, 1985; Hansel et
al., 1985; Taylor, 1990; Blewett and Vlfinters, 1995). This moraine is the western
continuation of the inner Port Huron moraine in northeastern Michigan (Leverett
and Taylor, 1915; Melhom, 1954; Hough, 1958; Evenson, 1973; Farrand and
Eschman, 1974; Drexler, 1975; Eschman, 1985; Hansel et al., 1985; Taylor,
1990) and marks the final ice advance of the LML during the Port Huron stade in
northwestern lower Michigan (Melhom, 1954; Hansel et al., 1985; Taylor, 1990).
It is also the most conspicuous topographic feature within the Lakeshore and was
first described and mapped by Leverett and Taylor (1914; 1915). Since then, its
position has been only slightly modified on maps produced by Melhom (1954),
Martin (1957), and Farrand and Bell (1982).

The Manistee moraine, as mapped by Melhom in 1954 (Figure 4), trends
southwest to northeast and eventually parallels the Lake Michigan shoreline from
the city of Manistee (not shown) northward to Glen Lake, Leelanau County.

From Glen Lake, it bends eastward around the head of Grand Traverse Bay, and
then turns northward again near the south end of Elk and Torch Lakes where it
becomes subdued near Clam Lake in southwestern Antrim County. Near Rapid
City the Manistee moraine bends sharply to the northeast where it appears as a
series of isolated ridges. At this point, these ridges are no longer collectively
named as part of the Manistee moraine but instead as part of the inner Port

Huron morainal system within a small outwash plain located behind the inner

Port H
the Ma
comer

al.19‘.

Plain 3
east). 1
3) at al
Hansel

rl'Z‘main

Port Huron moraine. Despite the name change, many workers believe that both
the Manistee and the inner Port Huron moraines were formed
contemporaneously (Farrand and Eschman, 1974, Hansel et al., 1985, Larson et
al., 1994).

Following the Port Huron stade and construction of parts of the Mancelona
Plain and the Manistee moraine (including the inner Port Huron moraine to the
east), the ice margin retreated north during the Twocreekan lnterstadial (Figure
3) at about 11,800 yrs B.P. (Evenson, 1973; Farrand and Eschman, 1974;
Hansel et al., 1985). Port Huron meltwater probably completed building the
remainder of the Mancelona Plain during this time (Leverett and Taylor, 1915;
Blewett, 1990; Blewett and \Ninters, 1995).

The final advance of the LML was during the Greatlakean stade (often
referred to by earlier workers as the "Valders” glaciation) when ice advanced
southward into the basin (Figure 2) between about 11,800 and 11,200 yrs B.P.
(Thwaites and Bertrand, 1957; Evenson, 1973; Schneider, 1990; Schneider and
Hansel, 1990).

Arguments on the extent of Greatlakean ice in the Sleeping Bear Dunes National

Lakeshore QQIOH. The advance of Greatlakean-Valders ice is not marked by any
distincfive landforms in northwestern Michigan or in eastern Wisconsin and so its
ice margin extent, thickness, and correlative till have been debated for several
decades (Melhom, 1954, Thwaites, 1957, Evenson, 1973, Taylor, 1990, Larson

et al, 1994). Some researchers have pointed out that ice from the Greatlakean

 

 

 

state he
Strap)
older Po.
al., 1985
that ice 0
far as th
till over
his SIUdlt
deposit l
Superim
Based c
WIdeSpr
00Vere<
I
stratjgr
WISOOI
ChebC
nonhe
is sup
organi
”5 8.1
1990)

1994)

stade barely extended out of the Lake Michigan basin and then only for a few km
as tongues of ice into low areas leaving scattered till deposits superimposed on
older Port Huron drift (Thwaites and Bertrand, 1957; Evenson, 1973; Hansel et
al., 1985; Larson et al., 1994). On the other hand, Melhom (1954) suggested
that ice of the Greatlakean (Valders) stade was extensive, advanced at least as
far as the inner Port Huron (Manistee) moraine, and left a thin carpet of red clay
till over the browner, coarser Port Huron-age till. In addition, Taylor (1990), from
his studies in a few areas of the Lakeshore, identified a patchy red, silt clay till
deposit with the “same character” as Melhom’s 'Valders” till (pg. 106)
superimposed over older Port Huron-age deposits located several km inland.
Based on this evidence, he argued that the till is likely Greatlakean in age and is
widespread at the Lakeshore. Therefore, he concluded that Greatlakean ice
covered large regions proximal to the shoreline.

Elsewhere in the Great Lakes basin, Greatlakean ice is well recorded by
stratigraphic evidence at Two Creeks forest bed (Figures 2, 3) located in eastern
Wisconsin (Broecker and Farrand, 1963; Evenson, 1973; Schneider, 1990) and
Cheboygan bryophyte (moss) bed (Figure 2) located about 200 km to the
northeast in Cheboygan, Michigan (Larson et al., 1994). At both places, a red till
is superimposed on lacustrine deposits that in turn are superimposed on the
organic beds. The age of the Two Creeks forest bed is given as 11,850 i 100
yrs B.P (Broecker and Farrand, 1963), and 11,910 :I: 120 yrs B.P. (Schneider,

1990). The bryophyte bed is given as 12,000 to 11,700 yrs B.P. (Larson et al.,
1994).

10

 

LITHOSTRA'EISRAPHIC LAKE LEVELS INLETS OUTLETS
UNI

. . and
Wisconsm Lake

Michigan

LAKE
PHASES

Green Bay
U. Peninsula
channels
Indian River
lowland
Siralrs ro
Fenelon Falls
Straits to
North Bay
Strait: l0
Port Huron

Tolesron
(184ml
Grand Valley

North <————-———————’— South

Michigan

Algoma

Nipissing ll

Nipissing
Nipissing I

Lake
Forest
Mbr

E
u.
C
.3,
c
.9
2
I)
x
to
_.i

. Chippewa
Winnetka Chippewa low
Mbr

Present lake level

unconformitv
(red)
Marquette advance

"Post-Algonquin” Post-Algonquin

' | l .
South Haven Mbr Algonqum Kirkfield

Two Rivers advance Calumet
Two Creeks retreat Two Creeks low

Inner Port Huron
advances Glenwood ll

Kewaunee Fm

?

Outer
Chica o

Cary-Port Huron retreat g lntra-Glenwood low

Equality Fm

Lake Border Glenwood l

Wedron Frn

Wadsworth
Mbr
Tinley

Cary advances

Vilparaiso
Milwaukee

 

I . C t I I

ll

 

 

.369 .5052): 59:22 .0 «Ems—.2. .950. 05 .0 tan Emcee: on. 5 $520... 6 8:39:55 9.56% an... 58.0 .v 959....

Z < 0 _ I 0 _ 2
so «Snafu; 35:53 us»
to PR! EUIFKOZ

'sa

m<2 450430

 

12

Arguments on the origin of the ridges at Sleeping gar Dunes National

Lakeshore

The origin of the ridges at the Lakeshore have been described and/or
debated by various researchers for over 175 years (Schoolcraft, 1821; Desor,
1851; Leverett and Taylor, 1915; Dow, 1938; Calver, 1947; Melhom, 1954;
Monaghan et al., 1986; Taylor, 1990). Schoolcraft (1821) was the first to study
landforms at the Lakeshore and interpreted sandy deposits in the exposed bluff
at Sleeping Bear Point as ‘a bank of sand 200 feet high and 8 or 9 miles long”

(pg. 401). Desor (1851) also noted thick sand accumulations along with
laminated pebble-free clay occurring near the bases in the bluffs at Sbeping
Bear Point (Figure 5) and other bluffs in the Lakeshore area and basically agreed
with Schoolcraft that these bluffs were nothing more than large dunes.

Leverett and Taylor (1915) described similar laminated sand and clay
deposits in the Lakeshore bluffs and were the first to state that the deposits were
derived glacially yet expressed “much uncertainty” as to the origin (pg. 314).
They noted that the elevations of the clay are sometimes 60 m above present
lake level yet interpreted it as having been deposited in water and mapped the
bluffs and associative ridges as part of the Manistee moraine (Figure 4) formed
during the Late Pleistocene (Leverett and Taylor, 1914). Leverett (1914) also
added the suggestion that deposition occurred not in a large proglacial lake but

“in narrow strips of water between lobes of ice that occupied the great valley-like

13

 

 

 

lowiands

TOlChIig"

thick on
“...they (
Surface
C
nearby s
apparen
moraine
crevassr
(Part II,
by inclu
and des
advanc
thick d5
fiuunde
Provide

replay

lowlands now occupied by the arms of Grand Traverse Bay, Pine Lake,
Torchlight Lake, and lesser lakes of the Grand Traverse region.” (pg. 315).

Dow (1938), like Leverett and Taylor (1915), noted the outcroppings of a
thick pink laminated clay spread throughout the bluffs and speculated that
“...they (the ridge clays) indicate a widespread deposit, or (date) earlier than the
surface morainal drift, but rest upon older glacial deposits.” (pg. 429).

Calver (1947), working on the geomorphology in Benzie County and other
nearby sections of the Lakeshore, was more intrigued by the morphology and
apparent linearity of the local ridges and expressed doubt that they represent
moraines. Instead, he suggested that ‘these ridges may have formed as
crevasse fillings in a stagnant block of ice that once occupied the Platte Valley.”
(Part II, pg. 13). Melhom (1954), on the other hand, expanded on Calver‘s work
by including some sedimentologic and interpretive discussions about the ridges
and described them as composed mainly of red-clay till and formed by the final
advance of the LML (Valders/Greatlakean). Near Honor, Taylor (1990) noted
thick deposits of “outwash sands and gravels” overlain by red, sandy, pebble-rich
till underlying the crests of many of the ridges (pg. 106). While no explanation is
provided for the origin of the “outwash”, he suggests that the till probably

represents some of the “older sandy tills of probable Port Huron age” (pg. 106).

14

    
      

Manitou l

 

4.?yramld

. Pchtbluif:5.N.lflletn
4. amasmmis.
‘* M7.MlllarHIlIridga;8.

a-tt-- .

an...

4e . F‘s“.

 

 

 

Figure 5. Map of the Lakeshore and vicinity showing major
landforms (slightly modified from Leverett and Taylor’s map, 1915).

15

C-Tacai

Adrano
landiom
formed r
includec
Lake Ch

NIPtssing

 

Glacial and met-glacial lake levels in the Lake Mim'yan basin
The lake level history of the Lake Michigan basin is varied and complex.

Advances and retreats of the LML in the basin scoured, gullied, and dissected
landforms as well as produced a series of large glacial lakes (Figure 3) that
formed extensive lake plains and beach deposits within the region . These lakes
included glacial Lake Chicago, followed possibly by a low interphase lake named
Lake Chippewa, and a series of post-glacial interbasin lakes named Algonquin,
Nipissing, and Algoma (Figure 3).

Proper treatment of the entire chronology of glacial and post-glacial lake
levels in the Lake Michigan basin would require a full-length discussion and
therefore lies outside the scope of this study. Only those general facts about
levels that may have affected the northern basin, and therefore the Lakeshore,
are described here. For excellent summaries of glacial and post-glacial lakes,
please refer to Bretz (1951; 1959; 1963), Hough (1958; 1963), Farrand (1982),

and especially Hansel et al. (1985) and Larsen (1987).

Glacial Lake Chgg‘ o. Leverett and Taylor (1915) first coined the term ‘glacial
Lake Chicago’ to describe the proglacial lake that formed when ice sat in the
Lake Michigan basin. They recognized the existence of three main levels or
“stages” of this lake in the southwestern Lake Michigan basin: the Glenwood
level at 195 m above sea level (a.s.l), the Calumet level at 189 m, and the
Toleston (Algonquin?) level at 183 to 184 m (Figure 3).

16

Glenwood. The Glenwood phase consisted of two lake levels, Glenwood l
and II, both at 195 in, and controlled by an outlet at Chicago. These levels were
produced between 14,500 to 12,500 yrs B.P. as the LML retreated northward
from the Lake Border moraine (Leverett and Taylor, 1915; Hansel et al., 1985;
Taylor, 1 990).

Calumet. The main Calumet phase, at 189 m, was produced at the onset of
the Greatlakean stade approximately 11,800 yrs B.P. (Evenson, 1973; Farrand
and Eschman, 1974; Schneider and Hansel, 1990). According to Hansel et al.
(1985), the Calumet phase occurred when LML ice readvanced to the position of
the Manistee moraine in northwestem Michigan and the Two Rivers moraine in
northeastern Wisconsin. As previously stated, this advance of the lobe did not
build the Manistee moraine since the moraine was built during the Port Huron
stade. The advance of the lobe during the Calumet phase subsequently blocked
outlets at the Straits of Mackinac and Indian River thereby causing Lake Chicago
waters to rise to a high lake level but still several meters lower than the
Glenwood high lake level (Hansel et al., 1985).

The Calumet rise drowned the forest at Two Creeks, Wisconsin (Figure 2)
before the Two Rivers Till was deposited (Evenson, 1973) and buried interglacial
organic matter with lake sediment at Cheboygan, Michigan (Figure 2) before the
Munro Till was deposited (Larson et al., 1994) by Gmatlakean ice. The Calumet
level remained stable until the Straits of Mackinac were again deglaciated. After

which it dropped several meters to the Toleston/Algonquin level.

17

Toleston/Algonguin. The Toleston/Algonquin level, at 184 m, was produced
between 11,500 and 11,200 yrs B.P. (Larsen, 1987; Taylor, 1990) as the LML ice
margin retreated north during the Two Creekan and Opened the Indian River
lowland (Figure 2) and the Straits of Mackinac. At that time, water flowed out
from the Lake Michigan basin through the Indian River lowland, through the
Straits of Mackinac, and out into the Lake Huron basin as the two lakes became
confluent. This formed the main phase of glacial Lake Algonquin (Hansel et al.,
1985; Larsen, 1985).

M Alter 10,000 yrs B.P. Lake Algonquin levels dropped to the
Chippewa Low (140 to 160 m; or below present lake level) as the ice front
continued its retreat northward and opened an outlet at North Bay, Ontario
(Hough, 1958; 1963; Larsen, 1985; Larsen, 1987). This drop marked the
transition from glacial to post-glacial lakes as ice positions ceased to be a factor,
and isostatic rebound, climate changes, and outlet elevations became more

significant variables during the Early-to-Middle Holocene.

Post;glacial lakes. A series of post-glacial lakes also formed in the Lake
Michigan basin. These include:

Nipissing. The Nipissing phase was produced at 8,000 to 4,000 yrs B.P.
(Hansel et al., 1985; Larsen, 1985; 1987) with Nipissing l, at 184.5 m, and
Nipissing II, at 180.5 m (Larsen, 1985) as Lakes Superior, Michigan and Huron

became confluent. During this time, the lake level was becoming progressively

18

 

 

lass infi.

 

and Hun

ab0ut2.

less influenced by rising spillways to the north (at North Bay) and more controlled
by outlets to the south at Chicago and Port Huron (Hough, 1958; Larsen, 1985).
Algoma The Algoma level was produced at 181.5 in after 4,700 yrs B.P. as
the North Bay sill isostatically rose above the Port Huron sill resulting in the
downcutting of unconsolidated sediments at Port Huron which caused the lake
level to fall (Hansel et al., 1985; Larsen, 1985). This event forever abandoned
the bedrock-floored Chicago outlet (Hough, 1958) and forced the cut-off between
the Michigan-Huron basins with the Superior basin. Subsequently, the Michigan
and Huron basins formed one single lake similar to its present configuration at

about 2,500 yrs B.P. (Eschman, 1985).

Arguments on glacial to-post:glacial lake levels in the Sleeping Bear
Dunes National Lakeshore Egion. Leverett (1915) stated that the oldest

strandline that he found north of the Manistee moraine represents the Toleston
(Algonquin?) level. In addition, many researchers since him agree and it seems
to be likely that Glenwood l and II levels should not be present on slopes north of
the Manistee moraine because: a) they are likely buried beneath a mantle of Two
Rivers drift or were eroded away by advancing Greatlakean ice and; b) the bluffs
along the Lake Michigan shoreline are higher than the limit of Glenwood l and Il
transgression and bluff recession has no doubt eradicated much of the evidence
of any paleostrandlines (Evenson, 1973; Farrand and Eschman, 1974; Mickelson

and Evenson, 1975).

19

Co
the ice ad
225 to 23
developer
as former
Interprets
Calumet 1
Slade for
Using a d

Shoreline

modeling

Conversely, Taylor (1990) suggested that Glenwood strandlines survived
the ice advance during the Greatlakean in the Lakeshore area and interpreted
225 to 235 m high scarps, terraces, deltas, and other related phenomena
developed on some of the ridges of the Manistee moraine around Benzie County
as formed by the Glenwood ll level of Lake Chicago (Figure 6). Taylor next
interpreted 210 to 215 m high scarps, terraces, and deltas as indicative of the
Calumet stade which is 21 to 26 in higher than the 189 m horizontal Glenwood
stade for the southern Lake Michigan basin demonstrated by Evenson (1973).
Using a different approach, Clark et al., (1994) have argued for Glenwood
shorelines north of the Manistee moraine and based their argument on computer
modeling.

Clark and others (1994) calculated shorelines for Glenwood using a
numerical Earth model with varying elastic and viscosity structure. They ran the
algorithm based on two simulations (thin vs. thick ice) which took into account the
ice-sheet load and subsequent rebound of the land surface as the ice sheet is
removed. From this model, they drew predicted Glenwood and Algonquin lake
level curves which were compared to historic lake-level gauge data and timing of
known outlet activity.

In general, shoreline data from Clark and others (Figures 6-8) for a thick
(~2,000 m) ice sheet are probably high estimates, whereas their data for thin
(~700 m) ice conforms better to field observations made by Larsen (1985) and

Taylor (1990). Clark et al. concluded that the thin-ice model works better when

20

predicting lake level curves, even more so once the ice thickness was increased
by 30%. Therefore, using this thin-ice model, the shoreline curve for Glenwood
project into the Lakeshore area (~400 km on the transect line as indicated; Figure
6) and predict a Glenwood level there (Figure 6) at about 260 m, while the
shoreline curve for Algonquin (Figure 7) at the Lakeshore is predicted at about
191 m (1994).

21

 

 

 

 

.95.... 03 FEE III

 

.565 mm 0:: 53:9: 05 co
.5. 8.. .88 a 8.82 m. 92.0.9.3 9:

 

 

 

 

bb-I

 

as. 8: access...

t8: 523
co Emu 9 $5.205 539.09.... 3862a
$.85 m.._m Lo {£0 .6 comtmano K 9:9”.

Soc .83... wee? 8:8...5

89.3.3308":—

Phi—PI

bth

bib—PPPb—bbbh—thb_bbb

o
F

 

(UWOWOIH
g a
L. A It. alw;

 

  

5385.. It
.fim8.xsfi ...:
scanssfil

 
 

'IIWWWIYVUI

I
!§!§‘°
fl

 

 

(“1) W913

aswm.v_

5&3 Co fine 2 35.205 mc_mm_a_z 3669a

€83 was a {so .0 c8538 .m 9:9“.

(33) “099593
i! 8
JLLI l I l l [J] l [J l I l

 

Acne .3mqflh wee? 3535

b

hh_b bb hI—IP DDIP— - -

   
   

 

F P

  

 

8 8
on n

§
(In) “OWN-"l8

   

amassed .
x .85 833E. . i -
.85 8.5.... ..Ii

.
III!
N

   
 

 

.68: 533 ecu .83: 829m

o. $an oocmfifibzgoc voozcgo 38:83
$85 was a {so .0 c8538 .m 2:9“.

 

 

 

 

 

Anacghweozuoesa

.8 8» .8 8n .8 8n .8 8. .

...L...L::..ELZ.........ILII On
a... .id§ "
M m . \II. \\\ i
N u \\ .8 .
wssL \\\A§:§E. I...

m .\\\ 28:..ng 3M

.. \ Seaman—soari- .8...

«oonm 00— 55. l

22

I..-

 

Field ma:

Fir
bit the Er
1997 anc
7.5 minu
(Figure l
Haven, 1
SUDDIer.
Mr'iln'rste
'Sland, l
Reoent

parant I

SECTION II: GEOLOGIC UNITS AND LANDFORMS

Methods

Field mapping.

Field mapping of the Lakeshore geology, supported by funding provided
by the Educational Mapping Program (EDMAP), was done over the summers of
1997 and 1998 and employed the most recent US. Geological Survey (U.S.G.S.)
7.5 minute topographic quadrangles available (1982—1983). The quadrangles
(Figure 9), covering parts of Leelanau and Benzie Counties, are as follows: Glen
Haven, Glen Arbor, Good Harbor Bay, Empire, Burdickville, and Beulah.
Supplemental adjacent quadrangles covering parts of Grand Traverse and
Manistee Counties were also used (Figure 9). They include: North Manitou
Island, South Manitou Island, Maple City, Platte River, Lake Ann, and Frankfort.
Recent aerial photographs and Leelanau County soil survey maps (1973) and

parent material descriptions were also used to assist mapping.

23

-- 45‘N

 

 

 

 

 

 

 

 

 

 

 

 

 

“'9!

Figure 9. U.S.G.S. 7.5 minute Quadrangles associated with the

Lakeshore.

24

 

gravel pit
notes. 8
Proportic
and com

Itnestor

SW.

Surficial deposits were studied in exposures along the Lakeshore, in
gravel pits when available, and in shallow (0.25 m to 1 m), hand-dug burrow
holes. Special attention was paid to sediment texture and the relative
proportions (e.g. “silty-clay"; or, “fine sand with little silt, trace gravel”; and so on),
and composition and clast lithology was also noted (granitoid, dolomite,

limestone breccia, etc).

Well-lgg data.

Deep wells are scarce in the Lakeshore area and their logs provide few
details regarding the geologic characteristics of the unconsolidated deposits
beneath the surface. Most deep wells are primarily exploratory in nature for oil
and gas and in most cases the well logs reflect a general lack of interest by the
driller in describing the glacial overburden. Three wells (here termed as ‘b’ for
bedrock wells) in the Lakeshore area were drilled as far as the bedrock and are
useful for determining the overall thickness of the unconsolidated sediment

overlying bedrock. The locations of these wells (Figure 11) are :
M1: SW‘/4,SW‘/4, NW‘A,Sec.35,T.28N.,R.12W.

m: SE%,SE%,NE%,Sec.5,T.29N.,R.12W.
m: SW‘/4,SW%, NVV‘A,Sec.34,T.28N.,R.14W.

25

wells dntt
hotness

(Figure 1
38

38

38

but Ile ou
some ca:
0w3Iburcj.
980Iogic

%

Fii

mosfiy on
“Sing a m
measllrec

factored ir

In addition, four more wells (here termed as “389’ for sand and gravel
wells drilled for water) stopped shy of the bedrock but provide at least minimum
thicknesses for the unconsolidated sediments. The locations of these wells

(Figure 11) are:

8&g#1: N%,SW‘/4,SW‘/4,Sec.4,T.29N.,R.12W.
88m: NW/a,SW‘/4,SW"/4,Sec.16,T.30N.,R.12W.
“9‘33 NE‘/4,NE‘/4,SE‘/4,Sec.33,T.30N.,R.12W.
3&9“: NE%,NE‘/4,NVV‘/4,Sec.30,T.29N.,R. 1 3W.

Other deep well records are available for Leelanau and Benzie counties
but lie outside the mapped area. Many of these are municipal water wells and, in
some cases, provide lithofacies descriptions and thicknesses of the glacial
overburden. These wells were useful to help describe the different surficial

geologic units in the Lakeshore area and to help construct cross-sections.

Measurement of sections.

Five Iakeshore sections were measured at the Pyramid Point bluff and
each section was positioned strategically approximately 350 m apart based
mostly on good exposures and consistent distances. Sections were measured
using a meter-tape from the top down to the beach and thickness of the
measured units was adjusted using a simple trigonometric nomogram that

factored in slope angle from the horizontal plane along with the average dip of

26

the beds
include i
compass
nietnodc

tetttree

traits:

the Laki

Iandion

the beds. Paleocurrent measurements were taken of current indicators that
include ripples, foresets, cross-beds, and aligned clasts using a Brunton
compass where feasible. Lithologic descriptions were made using the
methodology (Table 1) outlined in Eyles et al. (1983) which assigns a simple two-

to-three-Ietter alphabetical nomenclature to a unit.

Anabgis of geomprpholpgy

Geomorphic analysis consisted mostly of using 2° topographic maps of
the Lake Michigan basin to identify ridges around it and the Lakeshore
topographic maps to distinguish and describe the main landforms there. These
landforms included drumlins, ridges, lowlands, embayments, coastal and
perched dunes, as well as erosional landforms such as the bluffs and scarps.

In addition, geomorphology was also used to determine the elevations of
former glacial and post-glacial lake-level strandlines at the Lakeshore. A
strandline is formed generally during a high lake-level and may be represented
by a variety of landforms that are identified mostly from morphological attributes
(Larsen, 1985; Larsen, 1987; Taylor, 1990). These Iandfonns included: 1) raised
wave—cut scarps in areas where no fluvial activity occurred, 2) deltas, terraces,
benches, curvilinear beach and dune deposits, 3) large coastal swamps and
lacustrine plains, and 4) well-developed wave-cut scarps and benches (Leverett
and Taylor, 1915; Bretz, 1951; Evenson, 1973; Taylor, 1990; Colman et al.,
1994). Once these Iandforrns were identified and mapped, the general

elevations of their surfaces were read directly from the contours on the

27

topogtac
st'andEn

Dr
topog'a;

levels we

topographic maps. These elevations are therefore representative of lake-level
strandlines.

Due to the fact that strandline elevations were taken mostly from
topographic maps with a 5 m contour interval, exact elevations for former lake
levels were difficult to determine. Therefore, the paleolake strand data given in
this thesis are only estimates.

Sampling Strategy. Strandline data (elevations) were assembled, using the
criteria described above, from twenty-five sites (with over 70 data points; Table 3)
to cover about 130 km of Lake Michigan shoreline from Leland southwest to
Elberta, with North and South Manitou Islands included. These elevations, as
stated earlier, were read mostly from the five-meter contours on the topographic
maps used in this study and were then plotted on a scatterplot graph to show
their general spatial distributions. Therefore, the margin-of-error for any
strandline elevation using this approach is :I: 5 m. Lines were then drawn that
best-fit reasonable plot groups. Group criteria were based on finding the most
number of points that fell on a line drawn roughly horizontally. This line therefore
suggests an average strandline elevation for a paleolake level. This simple
group method was done using the general distribution of points that best
represented paleolake elevations. A point distribution of 6 points or more that fell
on or around the same line would suggest a well-defined strand, 5 to 3 points
would be considered a poorly defined strand, and less than 3 points on a line

would suggest a possible strand.

28

c“

~

....
i.

 

... h ......

r.-
Lesa

Ta':

(fro

Table 1. Glacial lithofacies codification (adapted from Eyles et al., 1983).

W

QIAMIQI D SAND S

Drnm matrix-supported, massive Sm massive

Dina ' " . stratified Sr rippled cross-stratified
Ding " ' , malty-graded Sh horizontaly laminated
Dca dost-supported, stratified St trough cross bedded
MUD F QBAXEI. G

Fm massive Gm massiw

Fl lunlnatad Gina massive. sand supported
Fld luninatad sily-clayldropstonas Gc dost-supported

Table 2. Diamict lithofacies code and symbols in addition to those in Table 1
(from Eyles et al., 1983).

 

 

 

 

 

 

 

 

 

  
   
   

 

  

 

 

 

 

“mucous m
“0'
On mm
°‘ -..... a A ....22... Mi
D-m m “ m2: ,‘
O-a unified -
0-. new W
mum-mum i : [g M
D—Irl; mm .
0—10): m W a...
.... .
an: i: mum-mu-
2‘2 -~Ilhdtundcbvcbl
8! = W ;-_- -mm
8! : "MW —mqu
CH : WW
8"! : ”I. CC”:
.0 = III-ll
Id : “mm
menu I
Fl bum
Fin : m
H victim

 

 

 

 

29

Map as-

hand-cc

 

Map assembly

Final paper copies of the completed maps were transferred to mylar and
hand-colored with all pertinent metadata and unit descriptions added. Three
cross-sections (A-A’, B-B’, C-C’) across representative areas were also
constructed and included on the map. Finally, the completed map was digitized
into an Arc/lnfo—ArcVIewT" (Version 3.1) format and was then overlain with the
most recent Digital Elevation Models (DEM) available from the U.S.G.S..

A color copy of the digitized surficial geologic map of the Lakeshore is
shown as Figure 12. Other images in this thesis are also presented in color. In
addition, 7.5” quadrangle-sized copies of the map are available (1 :25,000 scale)
as an open-file report (EDMAP#1434-HQ-97-AG-01730) at the Department of
Environmental Quality (DEQ), Michigan Geological Survey, Lansing, Michigan.

30

Befisd

enxse
lshnd.
Miter
oftSO
afihng
1957;!
include

EHSWOI

White ti
QrOUp ;

genera

Data

Bedrock eo

No bedrock crops out in the Lakeshore region, however rock units are
exposed to the east and northeast around Petosky, Charlevoix and Mackinac
Island. The bedrock geology of the Lakeshore, shown in Figure 10, is based on
Milstein’s (1987) bedrock geology map of Southem Michigan published at a scale
of'1:500,000. The map shows that the area is underlain mostly by northeast
striking Mississippian to Devonian sedimentary rocks (Melhom, 1954; Martin,
1957; Milstein, 1987) that dip toward the southeast at about 5°. The bedrock
includes, from oldest to youngest, the Detroit-Traverse Group and the Antrim-
Ellsworth-Coldwater (shales) Group (Milstein, 1987).

The Detroit-Traverse Group is mostly Devonian in age and consists of
white to black dolomite and limestone. Pebbles, cobbles, and boulders from this
group are common within the glacial deposits around the Lakeshore and are
generally prominent along most of the shoreline, especially areas with high bluffs.

The Antrim-Ellsworth-Coldwater Group is Devonian to Mississippian in
age, overlies the Traverse Group, and consists mostly of light-brown to black
shale with nodules of limestone and pyrite cubes. The Antrim shale is thin-
bedded, fissile, and dark-brown-to-black while the Ellsworth and Goldwater
shales consist of mostly soft, semi-indurated green shale with bands of
calcareous material (Melhom, 1954; Martin, 1957). In general, rocks from this

group are scarce within glacial deposits at the Lakeshore.

31

 

 

 

 

  

BEDROCK UNITS
i Goldwater Shale
D Ellsworth Shale Group

. Antrim Shale

Detroit-
Traverse Group

 

 

Figure 10. Map of bedrock units at the Lakeshore and vicinity (from
Milstein et al., 1987).

32

 

 

 

 

 

m...

N WW0 (>150 m)

M2 (103
(>95 m)
(>122 m)
RM)

67
Q,“ K

K\ (M

 

 

 

 

 

Figure 11. Map showing elevation of bedrock surface and well locations with
spot thicknesses of glacial drift (in meters a.s.l.) at the Lakeshore and vicinity
(from the Hydrologic Atlas of Michigan, Geologic Survey Division, Michigan
Department of Natural Resources, 1981).

33

Drift thickngs. The elevation of the bedrock at the Lakeshore is generally not
well known and therefore only general estimates on drift thickness are possible.

Figure 11 shows the bedrock surface elevation for the Lakeshore and
vicinity as redrawn from the Hydrologic Atlas of Michigan (1981). Drift thickness,
calculated by subtracting the elevations of the land surface taken off the U.S.G.S.
7.5' quadrangle maps from elevations of the bedrock surface reported in the
atlas, ranges from 50 to 322 m and thin towards the east and northeast. Drift
thickness ranges from 60 to 180 m at the Manitou Islands (Martin, 1957). It is
thickest near the mainland shoreline where it is about 230 m (well b#3) and thins
towards the east near Lake Leelanau where it ranges from 75 to 200 m (wells
b#1,2; s&g#1,2,3).

Surficial geology

Fourteen surficial geologic units were identified and mapped at the
Lakeshore (Figure 12) and are described (oldest to youngest) as follows (all
names given in this description are informal):
Leland formation. The “Leland formation” is a 5 to 120 m thick continuous layer
directly overlying the bedrock. It consists predominantly of massive, planar-to-
cross bedded, moderately sorted, pebbly-to—cobbly, medium-to—coarse sand.
The sand facies is mostly composed of rounded-to-subrounded quartz whereas
the pebbles and cobbles are composed chiefly of subangular-to-subrounded
carbonates and some crystallines. Cross-stratification within the sand is

common and includes cross-sets and truncated beds. Also common are

 

 

mil ...

 

.68.. 602.3 new Eon=m>>v 5922.2 £3550 cicom can 35.00.. 629.8330 8:5E m.» .333
new 6:33.225 .QEEm Sam Lona... c000 ...oE< :20 .cc>mI 5.0 9: *0 30.000 _m_o_.t:w .NF 2:9".

35

 

 

 

swap)“;

QM?)

   

Lillie

Gian Lakg _‘

Figure 12

36

        
 

Elev- -E‘fli‘i

 
   

\. W Monica“
a I

  
     

Good Harbor Bay

  
     
 

I

r

I

x
—<...i
r

I
I

     

 

 

 

A

Sleeping Bear blulll
Gib

 
   

       

arm: of“ m?“ ,. QM: sleepingaaaraay

elevation (m)

Suaartear Min.
ng

 

elevation (m)

Empire blulil
Club
300 1‘

 

elevation (m)

m fisqFlSIlfigFI‘hI-or IiiliyI IInd.

Surficial Geology of the Glen Haven,
Glen Arbor, Good Harbor Bay, ,.
Empire, Burdickville, and Beulah
7.5 Minute Quadrangles,
Leelanau and Benzie
Counties, Michigan
by
Todd E. Wallbom
and
Grahame J. Larson
Michigan State University
1998

PREPARED iN COOPERATION WITH
THE U S. GEOLOGICAL SURVEY
(EDMAP #1434-HQ—97-AG-0‘l730) AND THE
MICHIGAN GEOLOGICAL SURVEY

MAP REVIEWED BY KEVIN A. KINCARE, SENIOR GEOLOGIST,
MICHIGAN GEOLOGICAL SURVEY

Description of Map Units

0 ERNIDEIITA DEPOBT- ,
IIryII titled llIl Ind no r ' ' A

08 T-
I ”J um 5:35?" IL?!“ with main. Lee-11y IncludII ngvel.

    

H U MPLEXS
oI meeum a-mm ,

stifle! ifagurh‘agell In AME, .tl “mule: Ilum lend. CreII-Itrelllleellon :.nmmnh

fife “Inifilfiti’fisni. mixed with em Ind “115.301". IIIndlng w-ler.

I
I
I
l
I
I
I

EP Ofiljrmfll clung" Id cl I!“ II e-teandlum IInd.
on?!“ we! fol I IrnUI "ha" I end Im- I.
EIEII| fiWJIInI-Ie- medium Ilnd.
TR 53.153351"Lmnylnciuau Ich pebble Imbnclllen

fiedzlhI-lc-ceen II 'flhlub d'b ram dElk": oIrI “Yb: iTlllJiiil'p‘fi'u'nflI°"'"d“d

i I grey
g fi... ”I: .Iiiliiiiihihiéigfilifliiii...““Iii'i’iii-“WI9

- EédAIIie'imIfiT-m DB‘P narirTe; Lydkmhgelme. as; 3%.." chI-Ilrellflclllen common. DIIIIIII term

IIrgI Iprcn Trill II pm
R’MID INT,“ DEF 8| T- .
I W .rrl'i’m d flay :t..r...rrr.1.::tz'.it.'r.°'§sr: ml rat swam: : warm .
*‘ BI'rlImlIinLhIdEra'l WM. clgjlch dlImlct summit-um um; run-all _
- Whfimm Ind, pebblel. Ind cotheI croIleIllflcellon common. Thicknen >10 m.

14 m lhick.

  
 

Morphology Contour interval 20 Feet

\ Drumlins /\/561- 660
‘ 661- 760

    

. unes
/A Erosional Channel 76-1860
Scam Aye 861 940

 

-1 120
0 4 Kilometers
1 O 1 MIIII
=22:

Dlgll Ind Ind meted by John Linker
MIp IIynuI Ind dulgn by Todd WIlIbIm DIvld amt. Ind Chrlllopher Hoard

M‘j’

I19

i

 

 

 

remune:
Individua'
nfiequer
t>5cntfi
Itol4c:
40anir

Ti
long unit
Ewan
In add‘m
WEII-SOF

V
Stony, r
iaYErs,

interim

We is
Duck L
14) loc
Al'bor i

tomc

 

“0th

diamid

 

imbricated pebbles and cobbles as well as coarsening-upward sequences.
Individual bed thickness ranges from 0.1 to 1.5 m. Other characteristics include:
infrequent, gray-to-red, thin clay and silt clasts 1 to 4 cm in diameter, frequent, 1
to 5 cm thick and 8 to 22 cm long, gray-to-light tan silt and sand lenses; frequent,
1 to 14 cm thick and 4 to 60 cm long gravel lenses; and infrequent boulders 15 to
40 cm in diameter.

The Leland formation is also characterized by 1 to 3 cm thick and 1 to 2 m
long units of massive-to-Iaminated silt and fine sand facies as well as rippled
sand and 2 to 5 cm thick and 0.6 to 2 m long, gray-blue, wavy, laminated muds.
In addition, the formation includes 0.05 to 1.0 m thick and 0.7 to 3 m long units of
well-sorted sand, pebbles, and cobbles.

Well logs (b#2; s&g#3) show that the formation also includes massive,
stony, reddish diamict layers, massive-to-Iaminated, reddish-brown clay and silt
layers, and stratified, thinly-bedded sand and gravel layers that includes several
interbedded ‘yellow” sand stringers.

There are two good exposures of Leland formation in the Lakeshore area.
One is at the Leelanau County gravel pit (Figure 13) located about 3 km south of
Duck Lake (NE%,Sec.32,T.30N,R.12W.). The other is a 2.2 km long blUff (Figure
14) located at the northwest end of Pyramid Point about 6 km northeast of Glen
Arbor (NE%,Sec.36,T.30N,R.14W.). At the pit, Duck Lake diamict overlies about
10 m of the sand of the Leland formation (Figures 13, 15). In some places,
northward dipping 0.5 to 2 cm thick and 2 to 6 cm long masses of Duck Lake

diamict are thrust downward into the sand. The long axes of rock clasts within

37

. -"l'- 2' »' I-
. ‘f-Ducflske diamict
NH " ‘

 

Figure 13. Stratified sand and gravel of the Leland formation overlain by
Duck Lake diamict at the Leelanau County gravel pit.

 

Figure 14. Laminated clay and silt of the Leland formation with some
small-scale normal faulting at Pyramid Point. Note deformed beds near
pencil.

38

SYUCI
and 2
are 0
POOH
fine!

Qfair
(55:

and

the unit have a general north-south orientation. At Pyramid Point, 5 to 12 m of
the Leland formation is exposed and directly underlies the Duck Lake diamict
(Figure 14). Here, it consists mostly of a blue—to-gray, laminated clay and silt
layer which grades upwards to a more massive, blocky, and chaotic clayey-silt
and fine sand layer and then into a pebbly, clayey-silt layer. The laminated clay
and silt layer has frequent current ripples often in the form of ripple trains and is
locally intercalated with 0.2 to 5 cm thick and 6 to 45 cm long ribbons of sand.
The clayey-silt and fine sand layer includes frequent soft-sediment deformation
structures such as contorted bedding (Figure 14) and frequent, 0.4 to 2 cm thick
and 2 to 6 cm long, diamitic lenses that are generally planar. Also frequent here
are 0.3 to 5 cm thick and 5 to 18 cm long rock clast horizons.

Duck Lake diamict. The “Duck Lake diamict” is a massive, red-to-reddish brown,
poorly sorted, matrix-supported diamict. The diamict matrix consists chiefly of
fine, rounded-to-subrounded sand composed mainly of quartz and some fine-
grained sediments. The proportions of fine-grained sediments include mostly silt
(55-75% of the total amount of fine-grained sediments within the diamict matrix)
and some clay (ZS-45%). Clasts within the diamict are common, subrounded,
0.5 to 30 cm in diameter, and consist mostly of carbonates, some crystallines.
and pink-to-red sandstones. Other characteristies of the diamict include:
frequent, 0.3 to 6 cm thick and 2 to 38 cm long, gray-to-Iight-tan, fissile, silt
stringers; frequent, 0.1 to 12 cm thick and 5 to 100 cm long, light-tan, fine-to-
medium, sand stringers; and infrequent, 3 to 19 cm thick and 10 to 40 cm long
pebble and cobble lenses.

39

 

Locally, the diamict is more red than brown, sandier, and includes a
greater frequency of rock clasts but in some areas it is mostly light-gray-to-Iight—
blue. In a few places, it also includes poorly developed, discontinuous, rock-clast
horizons.

The Duck Lake diamict extent is about 96 km2 and is exposed mostly in
the eastern uplands where it underlies an undulating, drumlinized surface. Well
logs (b#2; s&g#1,2,3,4) and exposures of the unit show that its thickness
generally ranges from 2 to 35 m.

There are two good exposures of Duck Lake diamict in the Lakeshore
area. One is at the Leelanau County gravel pit whereas the other is at the
northwest end of Pyramid Point. In the pit, about 1 to 5 m of the unit is exposed
and directly overties the Leland formation (Figure 15). The contact between it
and the Leland formation is sharp and unconfon'nable. Also, in some places
weakly developed rock clast horizons are present near the contact. Above the
contact, 1 to 2.5 m of massive, brown-red, silty-clay diamict is overiain by a more
red, sandier, and more stratified diamict. The stratified diamict is 0.25 to 2 m
thick and occurs above a thrust-induced contact with the core diamict. Locally
above this contact and concentrated within the stratified diamict lies a zone of 0.5
to 2.3 m long thrust faults that dip northward at 3 to 7° and are spaced 0.5 to 1.2
m apart.

At Pyramid Point, a 2 to 11 cm thick and 0.3 to 1.6 m long bed ofred
clayey-silt diamict overlies the Leland formation (Figure 16). Wrthin this diamict

are infrequent, 1 to 3 cm thick and 2 to 5 cm long, blue-gray-to-red-brown,

40

diamict

arbor

oefom

diamict clasts as well as frequent pebbles and cobbles that consist of mostly
carbonates and crystallines. The diamict shows increasing degrees of
deformation approaching the contact with the overlying sands and gravels of the
Pyramid Point deposit.

flramid Point deposit. The “Pyramid Point deposit” consists of stratified, planar-
to-crossbedded, poor-to-moderately sorted, pebbly-to-cobbly, fine-to—medium,
rippled sand and includes light-blue—to-red-to-reddish-brown, massive-to-
laminated silt and clay layers. The sand consists mostly of rounded-to-
subrounded quartz whereas the pebbles and cobbles are chiefly subangular-to—
subrounded carbonates and some crystallines. Cross-bedded pebble and gravel
layers are common and include cross-sets, frequent troughs, truncated beds,
ripples, and coarsening-upward sequences. Thicknesses for the sand and sand
and gravel beds range from 0.2 to 3 m. Thicknesses for the silt and clay layers
range from 0.01 to 2.5 m. Other characteristies of the deposit are: infrequent, 1
to 3 cm wide and 3 to 8 cm long, blue-gray-to-red, thin silt and clay clasts and
lenses; infrequent, 0.8 to 6 cm thick and 4 to 26 cm long, light-tan, fine, sand
lenses; and infrequent boulders 0.2 to 1.3 m in diameter.

In some areas, the deposit contains more silt, clay, and fine sand as well
as wavy laminated muds. In other places, the deposit is more pebbly and cobbly
and includes 0.4 to 1.0 m thick and 0.1 to 4 m long beds of pebbles and cobbles
with a coarse sand matrix. Locally, the deposit contains aligned pebbles and
cobbles within a coarse sand matrix. In addition, some of these aligned, matrix-

supported pebbles and cobbles are contained within diapiric-shaped horizons

41

 

Figure 15. Detail of massive Duck Lake diamict superimposed on Leland

formation deposits, Leelanau County gravel pit. Note quarter on contact
(arrow).

\\
\
major shear plane

 

Figure 16. View upsection of a major shear plane (dashed line) in
deformed Duck Lake diamict shown slightly offset by post depositional
slumping at Pyramid Point. Shovel for scale is about 1 m in length.

42

 

Ilia «Ill. FIT. .Fnr EL. «L.

...: .

FD

\
h.
I.
r
K\
I.
t:
t.

l-
Flg

9e?

- 'f'ixzaéz‘w-rxe-
. ‘ wusfi ,

 

Figure 17. View of the bluff at Pyramid Point from Lake Michigan.
Direction of view is to the south.

 

Figure 18. Closer view of glaciolacustrine sediments at Pyramid Point. The
general direction of flow was from left to right. Shovel is about 1 m in length.

43

 

__.._.~

_ _.—:_-=_._.—v._——.. _ .-_

that arq
born"

from}

12. 17:
tel kn
tends

Lakesh

Point (F
directly

Unconf'

”billing
mafon

sedlme

_(
—————i—

rod(an

%

CIQSS b

consists
Dinkm I

quarm.

 

that are about 1 m wide upwards and taper to a point downwards. These
horizons are infrequent, occur within mostly deformed areas of the unit, range
from 0.6 to 2 m long, and generally coarsen upwards.

The Pyramid Point deposit is associated with the high ridges (Figures 5,
12, 17) that occur within the Lakeshore. The dimensions of these ridges are 0.3
to 1 km wide, 50-120 m high, and 3-12 km long and the orientation generally
trends northwest to southeast. A well log (s&g#4) and frequent exposures at the
Lakeshore show that its thickness ranges from 45 to 120 m.

The best exposure of the Pyramid Point deposit is the bluff at Pyramid
Point (Figures 17, 18). There, about 56 to 120 m of the deposit is exposed and
directly overlies Duck Lake diamict (Figure 19). Its contact with the diamict is
unconformable and in general dips to the southeast at about 20 to 35".

Primary structural features at Pyramid Point include trough cross-beds,
rippling, erosional contacts and scoured bases, small-scale (<2 m)
intrafonnational faulted blocks, conjugate faulting, and jointing. Mean direction of
sediment transport associated with ripples, cross beds (Figure 18), and aligned
rock and diamict clasts is from north to south.

Pearl Lake (legit. The “Pearl Lake deposit” consists of thick-bedded, planar-to-
cross bedded, moderate-to-well sorted, pebbly-to-cobbly, clast-supported gravel
with medium-to—coarse sand beds. The gravel is pebble and cobble-rich and
consists of mostly subangular-to-rounded carbonates, crystallines. and infrequent
pink-to-red sandstones. The sand consists mostly of rounded-to-subrounded

quartz. Cross-stratification is common as are cross-bedding, truncated beds,

mace
d‘a-rac

I"
v UL":

layers
thick 2
OOtltai
bOUldl

1.8m

imbricabd pebbles and cobbles, and fining-upward sequences. Other
characteristics include: frequent, 0.4 to 2 m thick and 1.3 to 2.5 m long, planar
bedding with discontinuous rock clast horizons; infrequent, weakly developed
troughs and channel deposits; frequent, weakly-developed clast imbrication;
infrequent, 0.2 to 4 cm thick and 3 to 25 cm long silt and fine sand beds; and
frequent, subangular-to-subrounded boulders 15 to 45 cm in diameter.

In some areas, the deposit contains more well-sorted fine-to—medium sand
layers 2 to 15 cm thick and 0.5 to 2 m long and whitish fine sand lenses 1 to 3 cm
thick and 3 to 8 cm long. In other places, the deposit is more cobbly and
contains frequent boulders whereas other areas has fewer cobbles and fewer
boulders with an overall increase in coarse sand layers that range from 0.15 to
1.8 m thick.

The Pearl Lake deposit extent is about 130 km2 and is associated with a
high flat plain that slopes to the southeast at about 6 mlkm, or 0.6%. Where the
deposit is partially collapsed, elongate northwest-southeast trending lakes mostly
fill the lows. The highest elevation of the deposit occurs near Bow Lake at 295 m
and the lowest elevation occurs to the south around Pearl Lake at about 265 m.
A well log (b#3) and frequent exposures at the Lakeshore show that its thickness
ranges from 60 to 105 m.

The type section for the Pearl Lake deposit is located at the Kasson Sand
and Gravel Mine, (NW‘/4,Sec.16,T.28N,R.13W; Figures 20, 21), a large active
gravel pit located about 2 km south of Bow Lake. At the pit, about 17 m of the

45

.mcosatomcu «6.03 5— mu 2:9“. 2 Beam .w-< used 28min. «a 80:03 oiamzngw .3 Sage

46

 

. .12.. I... . 1.
. vvzh a. u J:\</\:.~

.J-a<.§2-]

 

 

 

g §§§§8 isms

 

 

2 2:9“.

 

47

deposit is exposed and shows a general fining-upwards as well as imbricated
rock clasts (Figure 21) with long axes that dip towards the north-northwest.

Sand and gravel-undifferentiated. This deposit consists of weakly-stratified,
planar-to-crossbedded, poor-to-moderately sorted, pebbly-to-cobbly, matrix-
supported, medium-to—coarse sand and gravel. The sand consists mostly of
rounded-to-subrounded quartz whereas the gravel consists mostly of subangular-
to-subrounded carbonates and some crystalline and pink-to-red sandstones.
The cross-stratification and cross-bedding often includes foresets. Also common
are truncated beds, imbricated pebbles and cobbles, and fining-upward
sequences. Individual bed thickness ranges from 1.0 to 35 cm. Other
characteristics are: infrequent, laminated, clay layers 0.1 to 2 cm thick;
infrequent, 0.4 to 3 cm thick and 2 to 7 cm long, gray-to-red clay and silt lenses;
infrequent, 0.2 to 1 cm thick and 3 to 10 cm long, gray-to—light tan silt and sand
lenses; frequent, 1 to 4 cm thick, massive-to—laminated, silt beds; frequent,
massive, fine-to-medium sand; frequent, 1 to 7 cm thick and 4 to 30 cm long,
gravel lenses; and infrequent boulders 15 to 40 cm in diameter.

In some areas, the undifferentiated sand and gravel deposit contains more
fine sand and includes 1 to 3 cm thick and 0.5 to 2 m long, rippled sand and
wavy, laminated muds. In other places, the deposit is more pebbly and cobbly
and includes beds of well-sorted sand, pebbles, and cobbles 1 to 15 cm thick.
Frequent exposures at the Lakeshore show that it is mostly underlain by the
Duck Lake diamict, the Pyramid Point and the Pearl Lake deposits, and is 5 to 29
m thick.

48

 

Figure 20. Thick (~2 m) sets of stratified, coarse-grained Pearl Lake
glaciofluvial sediments at the Kasson Sand and Gravel Mine, Leelanau
County. The view is to the north.

 

Figure 21. Closer view of Pearl Lake glaciofluvial sediments at the
Kasson Sand and Gravel Mine. Well-developed clast imbrication
indicates that the general current flow was southward.

49

The type section for the undifferentiated sand and gravel deposit is at the
NPS gravel pit located just off Indian Hill Road about 5.5 km south of Empire
(NE%,Sec.7,T.27N,R.14W.). There, about 1.5 m of the deposit is exposed and
consists mostly of stratified, coarse sand with gravel, pebbles, and cobbles. Also
present are laminated silt and fine sand beds 0.3 to 4 cm thick, as well as clast-
supported pebbles and cobbles 5 to 30 cm thick. The long axes of imbricabd
rock clasts within the deposit dip to the northwest.
Honor depgit. The "Honor deposit“ consists of thin-to-thick bedded, planar-to-
cross-stratified, poor-to-moderately sorted, pebbly-to-cobbly, matrix-supported,
fine-to-medium sand. The sand consists chiefly of rounded-to—subrounded
quartz whereas the pebbles and cobbles consists of mostly carbonates, some
crystallines, and some shales and mudstones. Cross-bedded pebble and gravel
layers are common and include cross-sets, truncated beds, and fining-upward
sequences. In some locations, moderately-dipping (520°) foresets are common.
Other characteristics include: infrequent, 1 to 4 cm thick and 3 to 8 cm long,
whitish-tan, well—sorted, fine sand lenses; frequent, 0.3 to 5 cm thick and 0.4 to 2
m long, tan, laminated, silt beds; frequent, 0.3 to 5 cm thick and 2 to 12 cm long,
tan, silt lenses; and frequent, 6 to 25 cm thick and 0.4 to 3 m long, rock clast
horizons.

The Honor deposit in the Lakeshore area is associated with flat surfaces
with an areal extent of about 1 to 2.2 km2 and is found near to or within lowlands
and large stream channels. Elevations of these surfaces vary but most fall within

a range of 210 to 235 m. Deposit thickness generally ranges from 10 to 40 m.

50

 

Figure 22. The gravel pit of the lower delta northwest of Honor, Benzie
County.

‘ macaw '_,;,v*»""~

 

Figure 23. Detail of laminated sediments in the Honor delta complex.
Pencil is about 10 cm in length.

51

The best exposure of the Honor deposit is located in the Benzie County
Road Commission gravel pit (NE%,Sec.8,T.26N.,R.14W.; Figures 22, 23) located
near the town of Honor in Benzie County. There, about 8 m of the deposit is
exposed and consists of mostly stratified sand and gravel (Figure 22). Most of
the bedding is planar or has a low-to-moderate (5-20°) southwesterly dip. In
some areas, the deposit consists of more finely laminated silt and fine sand
(Figure 23) with frequent 1 to 3 cm thick and 2 to 7 cm long, white to-tan, fine
sand lenses.

Stream terrace deposit. This deposit consists chiefly of thin-to-thick bedded,
poorly-sorted, pebbly-to-cobbly, matrix-supported gravel. The gravel is pebble
and cobble-rich and consists mostly of subangular-to-subrounded carbonates
and some crystallines whereas the matrix consists chiefly of medium-to-coarse
sand that is mostly subangular—to-subrounded quartz. Cross-bedded gravel
layers are common and sometimes include troughs, foresets, and imbricated
pebbles and cobbles. Other characteristics are: infrequent, 0.2 to 3 cm thick and
0.5 to12 cm long silt and fine sand lenses; frequent, 4 to 25 cm thick and 2 to 35
cm long, massive, fine-to-medium sand; frequent, 2 to 13 cm thick and 5 to 40
cm long, gravel lenses; and infrequent boulders 15 to 40 cm in diameter.

The stream terrace deposit is associated mostly with a pair of terraces in
an abandoned channel located east of Empire. A photograph (Figure 24) taken
from Norcronk Road shows one of the well-developed terraces that has a surface
elevation of 225 m (a.m.s.l.), or an elevation of about 5 m above the channel

floor. Deposit thickness generally ranges from 3 to 7 m.

52

The only exposure of the deposit occurs at the NPS gravel pit (inactive)
located near M-22 just south of Empire in Benzie County about 300 m east of the
intersection of Aral Hills and Aral Road (NW‘/z,Sec.6,T.27N.,R.14W.). At the pit,
about 2 m of the deposit is exposed. The long axis of imbricated rock clasts
within the deposit dip to the north-northwest.

Glen Lake depgit. The “Glen Lake deposit” consists chiefly of stratified,
moderate-to-well sorted, fine sand with beds of red-to-red brown, laminated silt
and clay. The sand consists mostly of rounded-to-subrounded quartz and the
thickness of individual sand beds range from 0.4 to 27 cm. The silt and clay
beds range from 0.1 to 3 cm thick and 0.3 to 4 m long. Other characteristics
include: frequent, 0.2 to 6 cm thick and 0.5 to 1 m long, rock clast horizons;
infrequent, 0.5 to 1.2 m thick and 2 to 8 m long, massive-to-weakly stratified,
blue-gray, fossiliferous, marl beds; and frequent, 0.3 to 1.7 m thick and 1.2 to 12
m long, tabular, rippled, fine-to-medium sands.

In some areas, the deposit includes fossiliferous sands and marl with
frequent mollusc shell fragments. In other areas, the deposit has more fine sand
than clasts, whereas locally it is more gravel and pebble-rich. A well log (s&g#2)
and frequent exposures show that the deposit ranges from 0.04 to 8 m thick.

The Glen Lake deposit is extensive and is usually associated with low
relief areas at the Lakeshore. The best exposure of the deposit is locamd at the
easternmost end of Pyramid Point bluff (NW‘/4,Sec.32,T.30N.,R.13W.; Figures
17, 19) near Hidden Lake. Here, about 1.2 m of the deposit is exposed and

shows mostly 1.2 m thick and 6 m long, blue, marl beds overlying stratified, red-

53

brown, clay beds that range from 15 to 50 cm thick. Overlying the marl are
massive-to-paralIer-laminated, mollusc-bearing, cross-bedded sands about 4 m
thick.

Sl_e_eping Bear degsit. The ‘Sleeping Bear deposit” consists of mostly light-tan-
to-orange, massive-to-laminated, well-sorted, fine sand. Cross-stratification is
common as are truncated beds. Other characteristics are: frequent, planar-
bedding; infrequent, 0.5 to 3 cm thick and 0.3 to 2 m long, clast horizons;
infrequent, 0.2 to 7 cm thick, dark-brown-to—black, continuous, organic-rich
layers, infrequent, 0.1 to 3 cm thick and 2 to 6 cm long, organic-rich lenses, and
infrequent, 3 to 15 cm thick zones of disturbance and mixing by plants and
animals. Overall thickness of the unit ranges from 2 to 40 m.

The best exposure of the deposit is the Sleeping Bear Dune, which is part
of the Sleeping Bear complex (Figure 25) located near Sleeping Bear Point.
Here, about 15 to 32 m of the deposit is exposed and consists of mostly massive,
fine sand floored by pebble and cobble-rich sand. In other areas, the deposit is
superimposed on continuous, organic-rich layers 0.3 to 5 cm thick as well as
rootlets, tree, plant, and animal remains.

Beach/dune complex. This deposit consists of tan-to-reddish brown, massive-to-
laminated, moderate-to-well sorted, fine-to-medium, sand deposits with gravel
and pebble layers. The sand consists chiefly of subrounded-to-subangular,
quartz sand whereas the gravel and pebbles are mostly carbonates and
crystallines. Primary structures range from horizontal parallel laminations to low-

angle and high-angle parallel laminations. Other characteristics include:

 

Figure 24. A ~5-meter high terrace in an abandoned channel near Empire.
Direction of flow was from left to right. View is towards the southwest.

 

Figure 25. A large blowout surrounded by stabilized dunes perched atop
Sleeping Bear bluffs. Note possible Algonquin-age terrace (arrow)
developed near the southern end of South Manitou Island.

55

Muent, 0.4 to 6 cm thick and 0.2 to 1 m long gravel and pebble horizons;

frequent, coarse, sand beds 5 to 20 cm thick; frequent, 0.2 to 3 cm thick and 4 to

25 cm long, reddish-brown-to-black laminae of heavy minerals; and infrequent,
0.4 to 3 cm thick and 0.6 to 5 m long, dark-brown-to-black, organic-rich layers

' mixed with fine sand.

Beach/dune deposits are associated with distinctive, arcuate, ridge-like
landforms (Figure 12) 1.3 to 20 m high and 0.2 to 5 km long and are generally
located within enclosed semi-protected embayments. The deposit oftentimes is
superimposed on Glen Lake deposits but may also be found superimposed on
the Pyramid Point deposit and on the undifferentiated sand and gravel deposit.
Deposit thickness ranges from 1 to 20 m.

The best exposure of the deposit is near Glen Haven just north of M109
(S%,Sec.20.,T.29N.,R.14W.; about 0.3 km west of the DH. Day Campground).
There, about 3 m of mostly tan-to-light-orange, massive, medium sand is
exposed and includes a continuous, brown-black, planar, organic layer about 0.5
to 6 cm thick. The uppermost 0.3 m is heavily bioturbated and has a 10 cm thick
dark-brown soil layer that is partially covered with dune grasses and ferns.

Fan d_eposit. This deposit is composed of mostly massive-to-weakly-stratified,
planar-to—crossbedded, poorly-sorted, pebbly-to-cobbly, matrix-supported gravel.
The gravel consists chiefly of subrounded to-subangular carbonates, crystallines,
and red sandstones 0.5 to 40 cm in diameter. The matrix consists mostly of
medium-to-coarse, rounded-to-subrounded, quartz sand. Foresets are common

as are truncated beds and imbricated pebbles and cobbles. Other characteristics

are: infrequent, gray-red, silt laminae 0.15 to 2 m long and lenses 2 to 8 cm thick
and 5 to 22 cm long; frequent, 5 to 15 cm thick and 0.2 to 7 m long, clast
horizons; and infrequent boulders 15 to 40 cm in diameter.

In some areas, usually at or near the fan toe, the deposit contains more
silt and fine sand whereas in other places it includes 0.4 to 3 cm thick and 1 to 2
m long, wavy-to-laminated, mud horizons. In other places, usually at or near the
fan head, it is more pebbly and cobbly and includes beds of well-sorted sand 4 to
20 cm thick. Frequent exposures show that most of the fan deposits are 3 to 12
m thick.

Fans at the Lakeshore are common (Figure 12) but most are less than 1
km2 in area. The best exposure of a fan deposit is located near the southwestem
area of Little Traverse Lake (W‘A,Sec.14,T.29N.,R.13W.) and consists of about
2 m of fine-to-medium sand interfingering with 0.4 to 3 cm thick and 2 to 5 cm
long, silt laminations and massive, red, clay horizons located near the fan toe or
margin that is mostly superimposed on beach/dune and Glen Lake deposits.
Locally, the upper part of the fan deposit is more pebbly and cobbly and includes
1 to 4 cm thick and 3 to 8 cm long, gravel lenses. Foresets dipping 40-70° are
also common on the north and south ends of the fan.

Modern delta and terrace demsit. This deposit consists of complexes of thin
bedded, planar-to-cross-stratified, moderate-to-well sorted, gravel-to-pebbly,
matrix-supported, fine-to-medium sand. The sand consists chiefly of rounded-to-
subrounded quartz whereas the pebbles consist of mostly carbonates, some

crystallines, and some shales. Cross-bedded pebble and gravel layers are

57

common and include cross-sets, truncated beds, and fining-upward and fining-
away sequences with topsets, foresets, and bottomsets. Other characteristics
include: frequent, 0.2 to 3 cm thick and 1 to 3 cm long, clay laminae; frequent,
0.1 to 0.4 cm thick and 0.7 to 3 m long clay and silt lenses; frequent, stratified,
laminated-to—thinly bedded silt and fine sand layers; infrequent, 1 to 3 cm thick
and 4 to 15 cm long, clast horizons; and frequent, 2 to 8 cm thick and 0.2 to 6 m
long, organic-rich layers. Locally, the unit is massive and chaotic. Unit thickness
ranges from 1 to 3 m.

Modern delta and terrace complexes are located near bodies of water
around the Lakeshore area. The best example of one of fliese complexes is
located where the Platte River empties into Platte Lake near Honor (Figure 12).
Here, the river, assisted by its north branch, has formed a delta nearly 1.2 km2 in
area and mostly covered by swamp and marsh deposits.

Marsh and swamp deposit. This deposit consists mostly of dark-gray-to-md-
brown, well-sorted, laminated, fine-to-medium silt mixed with fine-to-medium
sand. The silt and sand consists of mostly rounded-to-subrounded quartz.
Individual bed thickness ranges from 0.5 to 3 cm. Other characteristics include:
frequent, laminated, red-to-red-brown, clay beds 0.4 to 7 cm thick; frequent, dark-
brown-to-black, thin-to-thick, peat deposits along with decaying plant and animal
matter mixed with silt and sand; and infrequent, 0.6 to 3 cm thick and 2 to 18 cm

long gravel and pebble lenses.

58

In some areas, the deposit contains more coarse sand as well as 2 to 6
cm thick and 0.4 to 5 m long, wavy-to-laminated muds. Locally, it is massive and
chaotic. Deposit thickness generally ranges from 0.2 to 3 m.

Exposures of the deposit are common around the Lakeshore (Figure 12)
and are generally associated with inland bodies of water located within protected
embayments and lowlands. The best exposures occur in the vicinity of Otter
Creek, located in Benzie County (S‘/z,Sec.12,T.27N.,R.15W.) where Otter Creek
flows northward from Otter Lake. Here, about 1 m of the deposit is exposed and
consists of mostly peat and fine sand interfingering with tan-to-red-brown clay
and silt. Locally, it contains rock clast horizons 4 to 17 cm thick and 0.3 to 2 m
long. Some localized bodies of standing water are also common.

Modem beach deposit. This deposit consists mostly of tan-to-light brown,
massive-to-laminated, moderate-to-well sorted, medium-to—coarse sand with
gravel and pebble layers. The sand consists of subrounded-to-subangular,
quartz sand whereas the gravel and pebbles consist of mostly subrounded-to-
subangular carbonates and crystallines. Primary structural features include
horizontal parallel lamination to low-angle and high-angle parallel lamination.
Other characteristics of the deposit are: frequent, 0.2 to 3 cm thick and 4 to 25
cm long, mddish-brown-to—black laminations of heavy minerals; infrequent, 0.4 to
3 cm thick and 0.6 to 5 m long, dark-brown-to—black, organic-rich layers mixed
with silt and fine sand; frequent, 0.4 to 6 cm thick and 0.2 to 1 m long gravel and

pebble horizons; infiequent, subrounded to subangular cobbles 4 to 12 cm in

59

diameter, and infrequent, subrounded to subangular boulders up to 2 m in
diameter.

The modern beach deposit is associated with proximal sediments along
the Lakeshore margin (Figure 26). It is superimposed mostly on older
beach/dune complexes and Glen Lake deposits or on eroded surfaces. Locally,
it is mixed with slumped material from bluffs. Deposit thickness ranges from 0.2
to 2 m.

The best exposure of the deposit is near the outlet of North Bar Lake,
located about 4 km northeast of Empire. Here, about 0.6 to 1.5 m of it is
exposed and shows mostly well-sorted, thick—bedded sand with a discontinuous
0.5 to 4 cm thick and 0.3 to 1.2 m long, gravel horizon occurring near the base of
the exposure. Also common are 0.2 to 1 cm thick and 5 to 14 cm long, black-to-

red, planar-to-wavy, fine laminations of heavy minerals.

Geomorphom
Drumlins. An elevated plain (Figure 5), 235 to 290 m in elevation and covering

about 96 km’, is located around Lime Lake, Little Traverse Lake, Duck Lake, and
Lake Leelanau. The plain includes numerous, well-developed drumlins and
drumlinoidal hills and has a gently undulating topography. Most of the drumlins
are oriented roughly north-south but locally trend northeast-southwest. They are
commonly elongated, low, and relatively mounded with smooth surfaces and
range from 6 to 42 m high, 18 to 250 m wide, and 0.2 to 1 km long. Length-to-
width-to-height ratios range from 0.120.422 to 0.5:6:17 (Figure 12, or see the

60

Good Harbor Bay U.S.G.S. 7.5” quadrangle). The drumlins vary in shape from
oval to linear with the largest drumlins and drumlinoids exhibiting fluted slopes to
the south and blunt noses to the north. They are generally separated from one
another by swales and depressions 10 to 500 m wide, some of which are filbd
with standing water, swamps, and bogs.

Ridges and kames. High flat ridges with long axes that are in large part
transverse to the margin of Lake Michigan form the most prominent landforms at
the Lakeshore. The ridges include (all ridge names listed here are unofficial)
Sleeping Bear, Empire, Point Betsie (or “Point Betsie Moraine”, or 'PBM’),
Alligator Hill, Miller Hill, and Pyramid Point (Figure 17), many of which intersect
the Lake Michigan shoreline to form steep bluffs (Figures 17, 26).

The Sleeping Bear ridge (Figure 5) is truncated at the Lake Michigan
shoreline and forms a bluff nearly 100 m high (Figure 26) that towers over Lake
Michigan. Along the shore, the ridge is generally capped by both active and
partially vegetated, stabilized dunes. The Sleeping Bear ridge is oriented roughly
northwest-southeast, is about 3 km wide and 8 km long, and has a flattened
surface that slopes gently to the southeast.

Directly south of the Sleeping Bear ridge is Empire ridge (Figure 5). It
consists of two to three rounded hills and also forms a bluff (Figure 26) that
stands about 65 m above the surface of Lake Michigan. Like the Sleeping Bear
ridge, Empire ridge is locally capped by dunes. About 14 km south of the Empire
ridge is the PBM (Figure 5) which is about 85 m high and has a relatively

61

hummocky surface. This ridge is the widest and longest and is 2 to 3 km wide
and about 12 km long.

To the east of Sleeping Bear ridge is Alligator Hill (Figure 5) which stands
about 50 m above Glen Lake (Figure 26). Like the PBM, it is oriented northwest-
southeast but it is smaller and only 0.3 to 2 km wide and about 4 km long. Its
surface consists of small, sharp peaks and crests that resembles an alligator's
back, henceforth its name.

About 4 km east of Alligator Hill is a large ridge complex named
collectively Miller Hill (Figure 5). It is 80 to 129 m high and separates the Glen
Lake lowland from the Bass/School Lakes lowland. Unlike the adjacent ridges,
Miller Hill is oriented roughly north-south and consists of mainly three
interconnected ridges 1 to 2 km wide and 2 to 8 km long and two proximal
isolated ridges 0.6 to 1 km wide and 1.5 to 3 km long that are separated by
lowlands, sags, and channels. The surface of Miller Hill is mostly hummocky with
wide, flat sections near its southern end as well as localized, incised channels
about 5 to 23 m deep and oriented roughly east-west (Figure 5).

About 8 km northeast of Glen Lake and directly north of Miller Hill is
Pyramid Point (Figure 17) which forms a bluff about 2.2 km long and stands
about 80 m above the surface of Lake Michigan. Its surface is mostly hummocky
and is locally trenched by several shallow channels 2 to 6 rn deep and oriented
roughly northwest-to-southeast. The ridge itself is also oriented roughly
northwest-to-southeast and is about 2 km wide towards the northwest and tapers
to a point toward the southeast (Figure 5).

62

.3855: 9: 3832 m_ is; 9:. .QEEm Lao: mcEsooo £33 9: co 32> .8 2:9".

 

63

Also within the Lakeshore are several smaller ridges, many of which are
less than 1 km wide and less than 2 km long. These form either sharp ridge
crests or gently undulating hills oriented roughly northwest-southeast.

Kames and kamic hills. A north-south trending line of kames and kamic hills
or mounds occur between Little Traverse Lake and Lake Leelanau. The most
prominent of these is Sugarloaf Mountain, which has an elevation of about 323 m
and is about 74 m of relief. Directly south of it is a series of progressively
smaller, unnamed kames 30 to 55 m high while to the north of it in Good Harbor
Bay is a submerged kame with a relief of about 22 m. This kame may have been
partially exposed when lake levels were lower.

Lowlands, embaments, and channels. The Lakeshore also includes several ‘
wide, shallow, bay-shaped lowlands proximal to Lake Michigan (Figures 5, 12,
26). They are generally flat with elevations that range from 178 to 204 m a.s.l
(T able 3) and are widest near the shoreline and nanowest from it. They are
often partially occupied by lakes and swamps.

The Glen Lake lowland (Figure 5) is a semi-circular-to—rectangular—shaped
depression about 6 km wide and 4 km long and is partially surrounded by the
ridges of Sleeping Bear, Alligator Hill, and Miller Hill (Figures 5, 12). The lowland
is partially filled by Glen Lake (Figure 5).

About 4 km northeast of Glen Lake is the Bass/School Lakes lowland.
This is an irregularly-shaped depression about 1 km wide and 4 km long. It is

bordered by Miller Hill to the west and a high drumlinized ridge to the east

(Figure 5). It is partially filled by Bass Lake and School Lake located north and
south from each other, respectively. About 3 km east of Bass Lake is the Little
Traverse/Lime Lakes lowland (Figure 5). This is a tongue-shaped depression
that is 2 to 5 km wide and about 7 km long that lies between a high drumlinized
ridge to the west and a high drumlinized plain to the east (Figure 5). The Little
Traverse/Lime Lakes lowland is partially filled by Little Traverse Lake and Lime
Lake that are located north and south from each other, respectively.

About 3 km south of the Glen Lake lowland is the Empire lowland (Figures
5, 26), which is an arcuate depression nearly 5 km wide that separates Sleeping
Bear and Empire ridges. About 4 km south of the Empire lowland is the Plate
Lakes lowland. It is a large arcuate depression about 6 km wide that separates
Empire ridge to the north from PBM to the south. The lowland is partially
occupied by Little Platte Lake and Platte Lake, located north and south from each
other, respectively, and is divided by a low-relief linear ridge that separates the
two lakes (Figure 5).

To the south of the Platte Lakes lowland is the Crystal Lake lowland. This
is a rectangular depression 3 km. wide and 7 km long and is partially filled by
Crystal Lake (Figure 5). It is bordered to the north by the PBM and to the south
by the hills and ridges that surround Benzonia. It is also separated from Lake
Michigan to the west by dunes, spits, and bars.

Empire channel. Located between the Glen Lakes and Plate Lakes lowlands
is a singular, north-south trending channel (Figures 12, 24, or see the Empire
U.S.G.S. 7.5” quadrangle) with a floor (Figure 24) that ranges in elevation from

65

219 to 228 m, and consists of three large prominent loops or meanders. The
meanders have a wavelength of 2 to 4 km and amplitude of about 3.5 km. The
channel floor is 15—20 m deep. It is relatively flat, featureless, boulder-free, and
is otherwise unremarkable except for a single, linear, elliptically-shaped sag
roughly 12 m deep, 120 m wide, and 1 km long located around the center of the
channel (SVIP/s,Sec.28ISE‘/4,Sec.29lNE‘/4,Sec.32,T.28N.,R.14W.). The channel
also includes several smaller elliptical and oval-shaped depressions 6 to 9 m
deep, 35t0200 mwide, and 250t0600m long Iocatedfurtl'lertothesouth.
Bluffs and scarps. Nearly 70 km of Lakeshore bluffs (Figures 5, 12, 17, 26) are
exposed along the Lake Michigan shoreline. Dunes cap many of the bluffs.
Most of the bluffs (Figures 17, 26) rise 50 to 120 m- above the lake with slope
gradients of 36 to 57° and are in varying stages of stability. Erosion and mass-
wasting have modified the bluff faces along the Lakeshore with periodic slumps
and landslides, especially at Sleeping Bear Point, Pyramid Point (Figure 17), and
occasionally at the west-facing bluffs of North and South Manitou Islands. In
general, those bluffs along the Lakeshore margin that are the most stable and
less prone to mass-wasting are either protected to a degree from the elements
by facing away from direct northeasterly storm wind and wave activity or are
relatively stabilized due to their compositional nature.

m Elevations for well-developed scarps were noted from various
locations (Table 3) around the Lakeshore area and its vicinity from Lake

Leelanau south to the Upper/Lower Herring Lakes (Figure 5).

The scarp with the highest elevation lies at 225 m and occurs along the
northeast-facing slope of Pyramid Point ridge and on west, north, and northeast-
facing slopes of North Manitou Island. The next lower scarp elevation occurs at
213 m and is found only along some of the slopes at North Manitou Island below
the 225 level. Below this elevation is a series of closely-spaced scarps that
range from 200 to 203 m in elevation and are found on several westward-facing
ridges and slopes near Duck Lake, Pyramid Point, Port Oneida, and North
Manitou Island. Below this is another series of scarps that are mostly around
196 to 199 m in elevation and along both the westward-facing slopes of North
and South Manitou Islands and the westward-facing ridges near Frankfort and
Elberta.

Dunes and beaches. Dunes at the Lakeshore take the form of either perched
dunes on bluffs and ridges (Figure 25) or lower-lying coastal beach/dune
complexes usually located in lowlands and depressions. The perched dunes
include Michigan’s most famous dune, the Sleeping Bear Dune, a 32 m thick,
rapidly eroding sand dune that is located at the edge of Sleeping Bear bluff
(Figure 26). Sleeping Bear Dune and the other similar perched dunes along the
Lakeshore margin are mostly parabolic in shape (Waterman, 1926; Snyder,
1985) and are best-developed on the steep bluffs 80 to. 143 m high that face the
strong northeasterly winds blowing off Lake Michigan. The perched dunes are
locally both active and inactive, and in places are stabilized by dune grasses,
thistles, ferns, and other vegetation. Locally, large blowouts occur (Figure 25),

some over 100 m in diameter and 16 m deep. In some cases, “lag gravel” of

67

pebbles are left behind or exposed at the base of some of the deeper blowouts
(Figure 25) while in others, wind-polished boulders, some more than 1 m in
diameter, lie within them.

Beach/dunes at the Lakeshore are mostly arcuate ridges of sand (Figure
12), 1.3 to 20 m high and 0.2 to 5 km long, that mostly parallel the shoreline and
are generally located within protected embayments. The elevations for these
ridges range from 181 m to 189 m (Table 3). In general, beach/dune ridges
conform to the present shoreline configuration. However, with increasing
distance from the shoreline margin, some ridges trend either roughly normal or

oblique to it.

68

 

 

.Em«§£§9.op=8pa§6§8a§§ ”802

 

 

 

 

 

 

 

 

a. 3. 8. e... ease: 58: mu
2: 8. a... case: 333 2s.
a: 8. steam am
a" 8. taxes“. -
... ... _ ._ s... 8. .5m $835 a
.3 226 a an... E "Na 8.. a: 38m em
= a. 8 b a 8 ea sou Lesa as: a.
z 2 (r a a8 SN 2" ea 3" B. see: 2
a: a: a: :86 .80 t
a I as 8a 3. 8. 2.3m 2
a" SN 8. case: see 2
. man can «.2 9.3 939m 3
35.... .3 .3 So a. an" a: a: 8. a2 93 5.0 2
E 8. 8. .32 8.0 N.
K... 3 w 2: a: n2 2.. .e. asses. .w 5
D \1 3a 2a 8a 8" a2 a: a. 32.52 .z 2
a a. .... 8a a: 835 So e
h
a a K. «A Ra ”8 SN Eon. 355 a
. :1.“ ad.) 2: 8. 9.3 =25 N
was. fix. 8. .320 8.2.0. e
1‘6 «an 3a 8a. 9.3 85¢ a
// . e in... 8...... 8a .2 a: 9a.. 8.: s
, / a: a: 8. a: a... 392» as: a
//x no... 358.. 93 N
/// .8 8. 93 5.5 s
/ ugfiAAAAAAAAAAgA
26.5.3 . E .. szF<>mlfi c¢<om 29.300.— emtm

 

.2258. 23 >95... .2 some: gas. 202
.3368 .5 3.503 new 2059.3 9: 5 «eggs-«... Boom .m 2an

 

69

 

SECTION III: GLACIAL AND POST-GLACIAL HISTORY

Deglaciation History

The Leland formation is the oldest surficial geologic unit at the Lakeshore
and is interpreted to represent a complex sequence of glaciofluvial and
glaciolacustrine sediments deposited at, or near, an ice margin. It consists
mostly of three different facies: 1) a reddish-brown stony diamict facies, 2) a
stratified sand facies, and 3) a blue-gray silt and clay facies with intercalated
sand and gravel.

The origin of the reddish stony diamict facies (as shown in well logs b#2;
s&g#3) is unclear but may represent debris flows and/or rainout till deposited in a
subaqueous setting. Evidence to support this interpretation includes massive,
structureless silt and clay that is described in well logs as highly heterogeneous
and includes several interbedded 'yellovrf' sand layers. These layers could
represent finely-textured stringers in the diamict produced by mobility of
sediment-laden fluids along marginal shear planes (Evenson et al., 1977;
Boulton, 1979; Huddart, 1983).

The stratified sand facies (Figure 13) most likely represents glacial
outwash of subaerial and subaqueous origin. Evidence in support of a subaerial
origin includes laterally-continuous, coarsening-upward sequences of stratified,
moderately-sorted, fine-to—medium sand to mixed sand and gravel (Gustavson,

1974; Boothroyd and Ashley, 1975; Blewett, 1990). Evidence in support of a

70

subaqueous origin includes parallel-bedding, rippling, cross-bedding, truncated
beds (Osborne, 1950; Shaw and Archer, 1978; Clifton and Dingler, 1984; Rovey
II and Borucki, 1995), and rockclast imbrication (Eyles and Eyles, 1983). The
imbrication of clasts indicates that the dominant flow direction was generally
towards the south.

The blue-gray clay and silt facies (Figure 14) is most likely of low energy
origin with probably nearly all of it deposited distal to the ice margin. Evidence to
support this interpretation includes distinctly laminated (varves?) bottomsets, silty
ripple trains and sandy horizons that show distinct signs of bottom current activity
(Rust and Romanelli, 1973; Ashley, 1975; Karrow et al., 1975; Rovey II and
Bomcki, 1995), isolated clasts that are suggestive of dropstones (Eyles and
Eyles, 1983), frequent silt and clay lenses (Eyles and Eyles, 1983), and primary
sedimentary structures that include rippling (Shaw and Archer, 1978; Clifton and
Dingler, 1984). The intercalated sand and gravel beds are interpreted either as a
series of subaqueous debris flow, subaqueous fan, or rain-out deposits formed
either in local proglacial lakes or in a single lake that was likely impounded by ice
subject to sudden episodes of large volumes of water released periodically
(Eyles and Eyles, 1983). In addition, the presence of well-sorted fine-grained
material (Figure 14) suggests that deposition was subaqueous (Dreimanis and
Goldthwait, 1973; Ashley, 1975; Huddart, 1983; Rovey II and Borucki, 1995).

Although the age of the Leland formation is uncertain, it clearly predates
the other mapped units at the Lakeshore. Evidence in support of this is the fact

that it directly underlies the Duck Lake diamict (Figure 13) which has been

71

interprebd to represent an ice advance (Monaghan et al., 1986; Monaghan and
Hansel, 1990). Therefore, the sand most likely mostly formed during an earlier
ice margin retreat.

The Duck Lake diamict is represented by two facies at the Lakeshore
exposed at the Leelanau County gravel pit (Figures 15,16) and at Pyramid Point
(Figure 16) and most likely represents subglacial till. Evidence in support of this
interpretation includes the fact that the diamict is generally massive, relatively
structureless, poorly-sorted, and contains frequent flattened, striated, and
elongated clasts and clast pavements (Boulton, 1968; 1979). The
unconfonnable contact between the diamict and the underlying Leland formation
(Figure 15), as well as downthrusted diamict wedges into the Leland formation,
also supports this interpretation (Boulton, 1979; Acomb et al., 1982; Dreimanis,
1989). At the pit, the red-brown till thrusted over the more massive till that forms
the core of the Duck Lake diamict may also represent a separate basal till formed
during a subsequent ice readvance (Boulton, 1968; Lawson, 1979; Dreimanis,
1989). The drumlinized topography developed on the diamict surface would
likewise suggest it was deposited as basal till when ice flowed in a southerly
direction (Leverett and Taylor, 1915; Melhom, 1954).

At Pyramid Point (Figure 16), the Duck Lake diamict probably represents
deformation till. Evidence in support of this interpretation includes the fact that
the diamict shows evidence of compressional folds, major shear planes (Figure
16), matrix-supported, rock clast lenses, inclusion of clasts with a preferred N-S

orientation, and overconsolidation (Boulton, 1968; 1979; Dreimanis, 1989). The

72

overconsolidation was likely produced by the heavy overburden of the ice while
the folds, shear planes, rock clast lenses, and oriented clasts were likely
produwd by the stresses applied by the moving ice or compaction due to its
weight or both (Boulton, 1968; 1979; Lawson, 1979; Dreimanis, 1989).

The age of the Duck Lake diamict is uncertain, but it is probably Port
Huron or younger. Evidence that it is no older than Port Huron is based on the
fact that it can be traced no further south than the Manistee (inner Port Huron)
moraine as defined by Melhom (1954). However, since the Greatlakean ice
advance (Figure 2) may also have extended as far south as the Manistee ,
moraine (Evenson, 1973; Hansel et al., 1985; Taylor, 1990), it is possible that all
or part of the Duck Lake diamict was deposited by this advance. On the other
hand, exposures and one well log (8891“) show that the Duck Lake diamict is
locally overlain by the sand that comprises the Pyramid Point deposit, hence it
must be younger than the Pyramid Point deposit.

The Pyramid Point deposit (Figures 17-19) includes ice-proximal, medial,
and distal facies and probably represents a series of localized subaqueous fan
deposits.

The ice-proximal facies probably represents high-energy subaqueous
gravity deposits formed at the ice margin near vents and/or tunnel-mouths.
Evidence in support of this interpretation includes stratified, moderately-sorted
gravel intercalated with fine sand (Ashley, 1975; Eyles and Eyles, 1983; Rovey II
and Borucki, 1995) while primary sedimentary structures that include cross-

bedding, truncated surfaces, and troughs suggest upper flow regimes probably

73

due to high discharge events occurring at the ice margin (Morgan, 1970; Rust
and Romanelli, 1973; Boothroyd and Ashley, 1975; Wright, 1977; Blewett, 1990).
Cross-bedding also suggests that the general flow direction was to the south.

The medial facies probably represents moderate-to-low energy
subaqueous gravity flow deposits accumulating near the ice margin. Evidence in
support of this interpretation are stratified, laterally continuous, coarsening-
upward sequences of sand (Figure 18) that are locally intercalated with clay, silt,
and gravel (Ashley, 1975; Karrow et al., 1975; Shaw and Archer, 1978; Eyles and
Eyles, 1983). These sequences include what appear to be dropstones as well as
primary sedimentary structures such as cross-bedding (Figure 18), which
indicates that the flow direction was generally to the south, parallel bedding
(Figure 19), frequent troughs and rippling, truncated beds, and rock clast
imbrication (Rust and Romanelli, 1973; Boothroyd and Ashley, 1975). The distal
facies was probably deposited within static water distal to the ice margin.
Evidence in support of this interpretation includes the fact that it consists of
massive-to-laminated silt and clay layers (Rust and Romanelli, 1973) that include
what appear to be dropstones (Eyles and Eyles, 1983).

Pyramid Point deposits were most likely limited to restricted, or semi-
enclosed, deep basins probably related to water-filled longitudinal crevasses
developed along the active ice margin. Evidence in support of this interpretation
includes layer-cake stratigraphy (Flint, 1928; Osborne, 1950; Klajnert, 1966;
Huddart, 1983) and morphological evidence that includes long linear ridges with

truncated flanks that roughly parallel the known southeasterly ice flow direction

74

(Leverett and Taylor, 1915; Calver, 1947; Klajnert, 1966; Huddart, 1983). The
above evidence supports the theory that the ridges are representative of
sediment containment by walls of ice during ridge construction (Flint, 1928;
Huddart, 1983). The age of the Pyramid Point deposit is uncertain but its
stratigraphic position above the Duck Lake diamict suggest it must have formed
during either the Port Huron or Greatlakean stades.

The Pearl Lake deposit (Figures 20, 21) is of glaciofluvial origin. Evidence
in support of this interpretation includes laterally continuous, fining-upward
sequences of planar-bedded, well-sorted gravel and coarse sand that suggest an
ice-proximal depositional environment (Boothroyd and Ashley, 1975; Blewett,
1990) and primary sedimentary structures (Figure 19) that include troughs,
channels, truncated beds, and rock clast imbrication (Gustavson, 1974;
Boothroyd and Ashley, 1975). The clast imbrication also suggests a southerly
flow direction (Eyles and Eyles, 1983). Morphological evidence also indicates
that the outwash of Pearl Lake deposit forms a sandur with a surface
characterized by an elevated, extensive, mostly well-drained plain that is partially
collapsed and filled by kettle lakes (Boothroyd and Ashley, 1975; Blewett, 1990).

A well-defined ice-contact margin occurs near the Glen Lake, Empire, and
Platte Lakes lowlands (Figures 5, 12) and represents the proximal margin of the
deposit. Near Glen Lake, the margin rises 66 m above the surrounding
landscape and slopes northward at about 27° while near Empire and Honor, it
slopes westward at about 22°.

75

The thickness of the Pearl Lake deposit suggests that large volumes of
sediment—charged meltwater exited the ice front and quickly built up the outwash.
The absence of fines in the upper part of the outwash (Figure 21) indicates that
streams were highly efficient in winnowing out suspended sediment and the
general fining-upwards characteristic suggests a steady decrease in flow velocity
accompanied by an increase in the amount of suspended sediment (Wright,
1977; Blewett and Winters, 1995).

The age of the Pearl Lake deposit is uncertain but since it locally buries
sediments associated with the Duck Lake diamict and Pyramid Point deposit, it is
probably related either to the Port Huron or Greatlakean ice sheet and was most
likely formed when the ice margin stood near the southern periphery of Glen
Lake and near Empire and Honor.

The undifferentiated sand and gravel deposit is interpreted to be of
glaciofluvial and/or glaciolacustrine origin and deposited locally against the
proximal edge of the Pearl Lake deposit. Evidence in support of a glaciofluvial
origin is weakly-stratified, moderately-sorted, coarse sand and gravel (Boothroyd
and Ashley, 1975; Blewett, 1990) and primary sedimentary structures that
include planar-bedding, cross-bedding, with occasional foresets, truncated beds,
and rock-clast imbrication (Gustavson, 1974; Boothroyd and Ashley, 1975).
Rock-clast imbrication indicates that the main direction of flow was to the
southeast. Evidence in support of a glaciolacustrine origin includes the
interbedding of laminated clay and silt layers (Ashley, 1975; Shaw and Archer,

1 978), clay and silt lenses (Eyles and Eyles, 1983; Rovey II and Borucki, 1995),

76

ripples, and wavy, laminated muds suggestive of active bottom currents that
reworked and shaped the lake sediments (Morgan, 1970; Rovey II and Borucki,
1995). .

The age of the undifferentiated sand and gravel deposit is uncertain but
because it locally buries sediments associated with the Duck Lake diamict,
Pyramid Point deposit, and Pearl Lake deposit, it is probably related either to the
Port Huron or Greatlakean stades and deposited during the retreat of the ice
margin from the edge of the Pearl Lake deposit.

The Honor deposit (Figures 22, 23) is interpreted as a delta complex.
Evidence in support of this interpretation includes planar-to-cross-beds,
moderately-dipping foresets (Gilbert, 1890; Morgan, 1970; Wright, 1977), and
truncated surfaces. Foresets suggest that the delta prograded out into a lake
and truncated surfaces suggest that the lake level fluctuated. The morphological
evidence in support of a delta includes a relatively planar surface, sharp slopes
along its margin (Gilbert, 1890; Wright, 1977), its position at the former outlet of
the Platte River (Figure 12), and the fact that it is located along the margin of a
paleolake (Calver, 1947; Taylor, 1990) that existed within the Plate Lakes
lowland (Figure 4).

The age of the Honor deposit is uncertain, but because it locally rests
against Port Huron or Greatlakean Pyramid Point and Pearl Lake deposits, it
must be post-Port Huron in age.

Sediments associated with the stream terrace deposit are interpreted as

fluvial in origin. Evidence in support of this interpretation includes such primary

77

structures as crossbedding, troughs, foresets, and imbricated pebbles and
cobbles (Rust and Romanelli, 1973; Boothroyd and Ashley, 1975; Lawson,
1982). The imbricated pebbles and cobbles suggest that the general flow
direction was towards the south. In addition, the fact that the terraces are
localized, restricted to a broad channel (Figure 24), and lack a kettled surface are
good indicators that they are stream terraces.

The age of the stream and terrace deposit is uncertain, but it must be
post-Port Huron because it rests on undifferentiated sand and gravel previously
shown to be either Port Huron or Greatlakean in age. In addition, since it is
restricted to the floor of a meltwater channel (Figure 24) formed during
deglaciation, initial deposition and construction of the terraces must have begun
shortly after the ice margin retreated from the head of the undifferentiated sand
and gravel deposit to allow glacial melt waters to carve out the channel.
Deposition and build-up of the terraces likely continued until flow through the
channel ceased.

Sediments associated with the Glen Lake deposit are interpreted as
lacustrine/glaciolacustrine in origin and were probably deposited proximal and
distal to the ice margin. Evidence in support of a lacustrine origin includes the
presence of fossiliferous, stratified, well-sorted, fine sand (Ashley, 1975; Shaw
and Archer, 1978; Eyles and Eyles, 1983; Clifton and Dingler, 1984; Rovey II and
Borucki, 1995), laminated silt and clay [(possibly varves?) (Ashley, 1975; Shaw
and Archer, 1978)], and beds of bluish, fossiliferous marl (Shaw and Archer,

1978). Evidence in support of a glaciolacustrine origin proximal to the ice margin

78

includes the presence of isolated clasts that are most likely dropstones (Eyles
and Eyles, 1983), while evidence in support of a glaciolacustrine origin distal to
the ice margin may include tabular, rippled, sand beds (Clifton and Dingler, 1984)
as well as the laminated silt and clay previously described. Morphological
evidence in support of a lacustrine/glaciolacustrine origin includes the fact that
the deposit forms a poorly drained, broad, flat plain throughout the lowlands and
Is locally associated with elevated beach and delta deposits. The extent of the
Glen Lake deposit would suggest that the lake into which it was deposited
extended well into the Platte, Empire, and Glen Lakes lowlands (Figures 5, 12).

The age of the Glen Lake deposit is uncertain, but because it locally rests
on the undifferentiated sand and gravel deposit previously shown to be
associated either with Port Huron or Greatlakean ice, initial deposition of the
Glen Lake deposit must have begun shortly after the ice margin withdrew north
and westward from the head of the Pearl Lake deposit to allow glacial lake
waters to flood the lowlands. Deposition continued as long as the lowlands
remained flooded and ceased when lake levels fell or the land surface rose
above lake level due to isostatic rebound. The Glen Lake deposit is time-
transgressive and is at least as old as the Honor delta complex.

Sediments associated with the Sleeping Bear deposit (Figure 25) are of
eolian origin. Evidence in support of this interpretation are well-draining,
massive-to-laminated, well-sorted, fine sand (Bagnold, 1936) and primary
structures that include cross-stratification and truncated beds (McKee, 1979).

The morphological evidence in support of an eolian origin includes the fact that

79

the Sleeping Bear deposit is associated with dunes that are perched on high
bluffs along the shore well above any known lake level. The source of the sand
that produces the dunes is mostly from the exposed bluffs (Waterman, 1926;
Dow, 1938; Snyder, 1985). Dow (1938) and Snyder (1985) have also suggested
that nearby modern beaches could also provide a secondary sand source.

The age of the Sleeping Bear deposit is uncertain, but because it rests on
Pyramid Point deposits shown previously to be Port Huron or Greatlakean in age,
deposition probably began soon after lake waters invaded the lowlands following
ice margin retreat (Dow, 1938; Snyder, 1985). Snyder (1985) states that the age
of the Sleeping Bear deposit is about 3,000 yrs B.P. based on 1‘0 dating of
paleosols on which the deposit is superimposed. However, he mentions that the
majority of eolian activity and dune buildup generally began about 1,000 yrs B.P
(1985).

Sediments associated with beach/dune complexes are of littoral and
eolian origin and were deposited proximal to the lake margin. Evidence in
support of this interpretation are well-draining, massive-to—laminated, well-sorted,
medium sand locally mixed with gravel and pebbles (Shepard and LaFond, 1940;
Komar, 1971; DuBois, 1978), frequent occurrences of dark, fine, heavy minerals
which suggest nearshore deposition (DuBois, 1978), and occasional buried,
organic-rich layers that suggest partial lake level stability followed by dune
migration as lake levels shifted (Kelley and Farrand, 1967). The morphological
evidence in support of a littoral origin includes the fact that the deposits form

arcuate-shaped, well-draining ridges that conform roughly to the present

80

shoreline configuration and are mostly located within enclosed, semi-protected
lowlands (Calver, 1947).

The age of the beach/dune complex is uncertain, but because it rests on
Glen Lake deposits shown previously to be younger than the undifferentiated
sand and gravel deposit, deposition must have begun shortly following lake
regression out of the lowlands.

Sediments associated with fan deposits are interpreted to be mostly
alluvial in origin. Evidence in support of this interpretation includes well-drained,
weakly stratified gravel interbedded with wavy, laminated mud and well-sorted
sand, and the fact that the coarsest sediments (pebbles and cobbles) usually
occurs near the head of the deposit and grades into finer sediments downslope
of the deposit (Boothroyd and Ashley, 1975). The morphological evidence
includes low, triangular-shaped deposits at the mouth of valleys which issues
from upland areas (Ryder, 1971; Boothroyd and Ashley, 1975), and multiple
stream incisions along the surfaces of the deposit (Boothroyd and Ashley, 1975).

The age of the fan deposit is unclear, but construction probably began not
long after deglaciation and continues today. The fact that many fan sediments
overlie undifferentiated sand and gravel and Glen Lake deposits as well as on
some beach/dune deposits supports this conclusion.

Sediments associated with modern delta and terrace deposits are
interpreted to represent alluvial complexes. Evidence in support of this
interpretation include planar-to-cross stratified clay, silt, and fine sand deposits

with steeply-dipping foresets which include some gravel accumulations in the

81

foreset beds (Gilbert, 1890; Wright, 1977). The morphological evidence in
support of this interpretation includes planar surfaces that extend like arrowheads
into inland lakes from stream outlets (Wright, 1977). In some cases, small spits
extend laterally along the delta front.

The age of the modern delta deposit is uncertain, but deposition and build-
up probably began shortly after the inland lakes at the Lakeshore stabilized to
assume their present outline. The fact that they mostly overlie Glen Lake
deposits, extend into the lakes, and are partially covered with marsh and swamp
deposits on stabilized surfaces supports this interpretation.

Sediments associated with the marsh and swamp deposit are interpreted
to be lacustrine in origin. Evidence in support of this interpretation includes
poorly-drained, well—sorted, stratified accumulations of mud, fine sand, and peat
(Swift, 1975). The morphological evidence includes the fact that many of these
deposits occupy low areas associated with Glen Lake deposits as well as the fact
that these deposits often occur around the margins of some inland lakes (Swift,
1975; DuBois, 1978).

The age of the marsh and swamp deposit is uncertain but because they
overlie Glen Lake deposits and partially cover modern delta sediments they
probably formed soon after lake regression out of the lowlands and inland lake
levels stabilized.

Sediments associated with the modern beach deposit are of littoral origin
and formed mostly by wave action and partly by eolian processes. Evidence in

support of this interpretation include well-draining, massive-to-laminated, well-

82

sorted sand with some gravel and pebble accumulations (Shepard and LaFond,
1940; Komar, 1971; DuBois, 1978) and the fact that these deposits rim most of
the lakeshore with the slip face sloping lakeward (Komar, 1971).

The modern beach deposit is the youngest unit at the Lakeshore and it is
likely that initial deposition began in the late Holocene and is continuing today.
The fact that the deposit overlies Glen Lake deposits, older beach and dune
deposits, marsh and swamp deposits, or eroded surfaces, supports this

conclusion.

Extent and tills 0f the Port Huron and Greatlakean ice advances

Ice advances of the LML during the Port Huron Substage occurred at least
twice between 13,300 and 12,200 years B.P. (Farrand and Eschman, 1974;
Fullerton, 1980; Hansel et al., 1985; Blewett and Winters, 1995). The first
advance formed the outer Port Huron moraine in Michigan (Figures 1-3) beMen
13,300 and 13,000 yrs B.P. (Farrand and Eschman, 1974; Hansel et al., 1985;
Blewett, 1990) while the second advance formed the Manistee morainefinner
Port Huron moraine (Figures 1-3) between 13,000 and 12,200 yrs B.P. (Evenson,
1973; Farrand and Eschman, 1974; Eschman, 1985; Hansel et al., 1985; Taylor,
1990; Blewett and Winters, 1995).

According to Leverett and Taylor (1915), Bergquist (1954), Melhom (1954,
pg. 32), and Monaghan et al. (1990, pg. 47) the Port Huron advance is marked
by a red-brown clay drift. Bergquist (1954) distinguished two tills near the

Lakeshore and described the lowest as a highly-compacted brown till deposited

83

during the Cary ice advance. Melhom also stated that he found multiple tills
underlying the drumlins around Leelanau County near the Lakeshore and
suggested that the oldest till was deposited during the Cary and Port Huron ice
advances. However, he states that most of the till underlying the drumlin
surfaces is composed of a red, clay-rich till that he interprets as Valders. Later,
Monaghan et al. (1990) noted several tills occurring within the drumlins around
Leelanau County and found that the clay mineralogy of the lowermost till is
similar to Ozaukee, Haven, and Valders tills from eastern Wlsconsin deposited
during the Port Huron ice advance (Figure 3).

As previously discussed, Leland formation likely formed during the retreat
of the ice margin from the outer Port Huron moraine and may include till or
diamict related to the ice advance that formed the outer Port Huron moraine
(Figures 1-3). On the other hand, the Duck Lake diamict was probably deposited
by Port Huron ice that advanced to the position of the inner Port Huron moraine
(Figures 1-3). The diamict has similar color and sedimentological characteristics
to other Port Huron tills observed by Melhom along the Lake Michigan basin up
to the position of the inner Port Huron moraine (Figure 4).

The extent of the second Port Huron ice advance is also marked at the
Lakeshore by several large mega crevasse-fill, kame, and outwash deposits
(Figure 12) that signify an active ice-margin which produced enormous volumes
of glacially-derived sediment via complex subglacial conduits, vents, as well as

englacial and supraglacial drainages.

The assignment of a Port Huron age to the Duck Lake diamict modifies
Melhom’s hypothesis (1954) that the till coring the drumlins (Duck Lake diamict),
as well as the pink, sandy-silty till (T aylor, 1990) underlying the crevasse-fill
(Pyramid Point deposit) and outwash (Pearl Lake deposit) deposits at the
Lakeshore, was produced by Valders (Greatlakean) ice (Figure 2). The Duck
Lake diamict is probably stratigraphically correlatable to the Port Huron till of
Leverett and Taylor (1915) and perhaps even to the Haven and Valders Member
of the Kewaunee Formation of eastern Wlsconsin (Evenson, 1973; Hansel et al.,
1985; Peterson, 1986; Monaghan et al., 1990; Rovey II and Borucki, 1995).
However, this linkage is tenuous and further work is necessary before any
stratigraphic relationships can be properly determined.

The advance of Greatlakean ice (Valders of Melhom) to the position of the
Manistee moraine (Figures 2-4) occurred beMen about 11,800 and 11,200 yrs
B.P. (Thwaites and Bertrand, 1957; Evenson, 1973; Schneider, 1990; Schneider
and Hansel, 1990). According to Melhom (1954), Evenson (1973), and Taylor
(1990), it is marked by a thin deposit of distinctive, red, clay-rich till whereas, by
comparison, the color and texture for the Port Huron ice advance is marked by a
brown, sandy till (Bretz, 1951; Bergquist, 1954; Melhom, 1954; Hough, 1955).
Melhom's limit of the Valders ice advance (1954, pg. 159, Plate 11) in northwest
lower Michigan was based on his view of the extent of red-clay till. His limit
would have Valders ice extending over the drumlins cored by Duck Lake diamict
to the outer flanks of the Manistee moraine (inner Port Huron). This view was

supported by Taylor (1990) who describes what he believes is a “Greatlakean’

85

red sandy till occurring along the flanks of the PBM and individual ridges in the
Benzie County area. In addition, Monaghan et al. (1990) analyzed the clay
mineralogy of till in Leelanau County and found it to be similar to that of Two
Rivers Till in eastern Wlsconsin which is of Greatlakean age (Evenson, 1973).

Despite Melhom (1954) and Taylor’s (1990) arguments, the extent of their
“Valders/Greatlakean” till at the Lakeshore is unsupported since there is no clear
field evidence to support the opinion that Greatlakean ice ever extended along
the PBM or over ridges or high plains in the Benzie County area. In essence,
there is no evidence at the Lakeshore for a stratigraphic marker that shows
Valders/Greatlakean-age, red, clay-rich till superimposed on Port Huron-age,
brown, sandy till (Bretz, 1951, Bergquist, 1954, Melhom, 1954, Hough, 1955,
Evenson, 1973, Taylor, 1990). Furthermore, the surficial geologic map of the
Lakeshore (Figure 12) shows no characteristic landforms like Greatlakean-
related end, recessional, or ground moraines overlying Port Huron-related
landforms (Eschman, 1985; Hansel et al., 1985; Taylor, 1990; Rovey II and
Borucki, 1995). These facts suggestthat Greatlakean ice either never advanwd
out of the Lake Michigan basin into the Lakeshore region to the position of the
Manistee moraine, or it only extended as thin lobes into the lowlands for no more
than 10 km and never covered the high ridges and plains.

The view stated above is in agreement with other studies in the Lake
Michigan basin that suggest the Valders/Greatlakean ice was thin, occupied
mainly lowlands, and was constrained by preexisting topography (Evenson,

1973; Farrand and Eschman, 1974; Hansel et al., 1985; Schneider, 1990; Larson

86

 

et al., 1994). What's more, others have suggested that this advance is better
represented in northern lower Michigan by erosion than by accumulations of till
(Evenson, 1973; Eschman, 1985; Taylor, 1990). Finally, many workers have
recognized that the red-clay lithology is not unique to the Greatlakean ice

advance and may instead occur in deposits related to the Port Huron ice

advance (Evenson, 1973; Farrand and Eschman, 1974; Eschman, 1985;
Schneider and Hansel, 1990). Such observations probably also apply to the
Lakeshore in that the high landforms there likely prevented the lateral spread of
Greatlakean ice beyond the lowlands (Figure 27). If any Greatlakean till/diamict
was deposited in the lowlands, it is now probably buried by subsequent lacustrine

sediments associated with the Glen Lake deposit.

87

 

 

 

 

   

EXPLANATION

fi Pearl Lake
‘ outwash

3:. Pyramid Point
'24!" deposit

    
 

Duck Lake
diamict
direction of ice
movement

 

O 2 4“
E:

l O l 4-

 

 

F59 ure 27. Hypothetical extent of Greatlakean ice conforming mainly
to lowland areas along the Lakeshore (~11,800 yrs B.P.).

88

 

Origin of the ridges at Sleeping Bear Dunes National Lakeshore

As previously discussed, sediments associated with the Pyramid Point
deposit (Figures 17-19) likely developed within longitudinal fissures along an
active ice-margin. These fissures deepened and widened with time to
accommodate high volumes of sediment that entered via subglacial and englacial
tunnels located along the ice margin. Once the ice margin retreated fully into the
Lake Michigan basin, construction of the Pyramid Point deposit ceased, partial
collapse redistributed some of the sediments, and water levels fell below the tops
of the corresponding ridges.

The overall heterogeneity inherent in the complex depositional
environment associated with the crevasses is probably best illustrated in the bluff
stratigraphy at Pyramid Point as shown in Figures 19 and 28. Here, at best
three different depositional facies are present: an ice-proximal facies, a middle
facies, and a distal facies. The gravels (Figure 28) represent the ice-proximal
facies, where coarse detritus is ejected from the nearby ice margin and deposited
almost immediately. This facies has numerous planar-convex boundaries
(Figure 19) and isolated flow structures and was likely derived from vent or
tunnel-mouth discharge in sediment-charged plumes, rainout from overturning
and melting of icebergs, and superglacial meltwater depositing supraglacial
sediment along file ice margin. Coarse sediment could also have been
transported away from the ice margin into the crevasse by debris flows that were

energized by meltwater pulses and deposited as massive gravel beds and lags.

89

The upper surface of some of the gravels appear to have been
subsequently shaped and eroded by fluvial processes producing troughs and
channels and are overlain by fluvial deposits with ripples and cross-sets (Figure
19) associated with the middle facies of the Pyramid Point deposit. This middle
facies includes stratified sand bodies (Figure 28) that were deposited as either
debris flows or sediment gravity flows. Sorting and remobilization of the sand
probably also occurred subaqueously as sediment flows which were either
fanned out in broad aprons forming thin sheets, or as individual massive bodies
that moved as a unit and filled in holes, erosional contacts, and depressions
caused by scour.

The fine-grained sediments (Figure 28) of the distal facies probably settled
over the middle-facies as the ice margin backed away and produced thick
accumulations of clay and silt rythmites. The distal facies includes red-brown
clay and silt deposits (Figure 28) with possibly dropstones that were probably
deposited during periods of slack water, or when sediment outflow was
temporarily restricted.

The entire sequence probably occurred within a trough-type subaqueous
system walled by ice. As the ice margin regressed, oscillatory sediment pulses
were produced forming intraforrnational truncated surfaces that grade away from
the ice margin. These surfaces were probably caused by channels that eroded
into the deposit during powerful spring and summer flows. Finally, the continued
regression of the ice-margin retreated from the crevasse fills and other

associative deposits such as kames and caused the water-filled fissures to drain.

90

Eon. 259E a see oases: see es .o 5.35% use 85580 .8 use".

91

 

 

 

   

  
 

AEOLIAN/BEACH DUNE
SAND. Locally includes peat,
soils. buried soils.

SAND, medium—coarse, planar
bedded with occasional troughs,
ripples. Gravel-rich in upper unit.

CLAY, SILT, and fine
SAND, massive to laminated.
Locally includes dropstones.

 

GRAVEL, massive, with medium—
coarse sand in lower unit.

 

Figure 28

92

  
   

       
 
     
  

Z
Z

  

  
  

    
    

6 1‘
=3.

l!

,g!
as
§

0

 
 
 
  
  

   
  

5»
s\\'::
l\\
is
I

is ‘

 
 

     
  

     

   

’f:

\

  
 
         

’//Ill
l

s

t'—

 

 

       

I _
3.

  

gt:

 

SAND, massive to trough cross-bedded.
Locally includes clay, silt ribbons,
gravel lenses, and boulders.

SAND, massive to trough cross-bedded.
Locally includes gravel lenses.

- CLAY, SILT, and fine SAND,

massive to laminated. Deformed in
upper unit.

 

 

 

A r conce tual model of a suba ueous crevasse-fill de it

A model for the origin of the subaqueous crevasse-fills at the Lakeshore is
given in Figure 29 and shows characteristics typical of such deposits. Tunnel-
mouths (“TM”) debouch sand and gravel directly into a large, open, water-filled
crevasse (Figure 29: # 3) to form planar-bedded, graded sand and gravel beds
that directly overlie basal till with current flow direction normal to the ice margin.
Supraglacial and englacial runoff (# 6) cany loose particles into the crevasses,
while bottom currents redistribute sand and gravel away from the ice, forming
ripples, truncated ripples, and cross-sets, and leaving behind pebbles, cobbles,
and boulders. Thick (meter-scale) massive-to-laminated deposits of fine material
occur and are locally punctuated by dropstones and rain-out (# 16) from icebergs
floating overhead. As the ice margin retreats, the ridge is completely abandoned
and the supply of sediment is terminated. Truncated sides, partly offset by listric
normal faulting and partial collapse (# 7) as the deposit expands out of the
confinement of the crevasse in the ice, mark the extent of the filling inside of the
crevasse.

This model is based on personal field observations and interpretations
made about ridges at the Lakeshore as well as on written accounts and
interpretations noted by researchers describing similar ridges in Canada
(Osborne, 1950; Duckworth, 1978), Europe (Klajnert, 1966; Huddart, 1983), and
the United States (Flint, 1928; Colton and Jensen, 1953; Blewett, 1990).

93

However, most of these ridges are not quite as large and extensive as those at
the Lakeshore.

The lack of normal faulting along the margins of the bluff exposure at
Pyramid Point is puzzling (Figures 17, 28). However, several simple theories
may help to explain this. They include the possibility that: 1) the east and west
ends of the bluff are poorly exposed due to local slumping (likely along normal
faults) and talus cover which may conceal any structures; 2) evidence of faulting
may have been removed by subsequent erosion by Greatlakean ice; and 3)
erosion associated with post-glacial lake-level fluctuations may have undercut

and eroded me flanks of the filling and removed any evidence of faulting.

05533
05 an 35:: $358 mooE co. EoEco.__>co .mcoEmoaou .228 can _ouoE .mafioocoo < .3 05mm

95

.... 25-5!
u o. ”3.890 3.9.9323. .— n. "out; 8. n v. ".... 3o... 385.32..» u n. ”8.8..
03.3 a N. ”3:58 ..l. . . ”9.88.. 53:. .055. u e. ”8. v0.55 u c 2:88.33
u m ”2.35? 330.323 n h “.055. 38.9.» a o ".... 383...: u w 683.68
7.3.983 u v ”moan—gob aqu u n .8. 55.3 «.2... Bo... u N woo. 03.0.. n .
......SE n 3. 35:3
5.29:. n U... ”58E .05.... u 3.... ”.0535 n :U ...... 38>2o 33:92.3 u mom
.8... 3338.82 a a. :5...

no. «£2.92» a .o .358 2.. wet—... eon—3.. 35:... 33.80 3833.6

 

. ... .dit.” \l\\
....u..............H ....... % %¢

»
.............. ..... 0‘ I\

   

I
a
O
‘
T
............

om use:

 

    

96

Discussion

Excellent Iakeshore exposures of the Manistee moraine at the Lakeshore
show distinct bedforrns and structures that do not fit Leverett’s (1915) classical
definition of a moraine. Instead, the ridges at the Lakeshore seem to be
indicative of infilling of large (mega) crevasses by subaqueous fan deposits,
proximal to an active ice margin, as originally proposed by Calver in 1947.

The origin of the massive-to-laminated, pink-to—red, lacustrine, clay beds in
the bluffs and ridges at the Lakeshore has long been controversial. Desor (1851)
was the first to discuss the clay beds and observed that they are widespread and
overlie older deposits. Leverett (1915) too noted the clay beds and stated
correctly that they likely represent a water-laid glacial deposit formed either within
the Wisconsin stage of glaciation or during a pre-Wrsconsin interglacial. He also
noted the high elevations of the clay which are higher than any known lake level
and suggested that it was a result of subaqueous deposition between lobes of ice
(pg. 315) but added that more data were needed before its origin could be
determined with any accuracy. However, despite the fact that the ridges
contained lacustrine sediments, he concluded that they are part of the Manistee
moraine and were likely overridden by later ice.

Leverett (1915) also described the Manistee moraine as it appears within
the Lakeshore as “a series of loops” and suggested that the ice that formed the
moraine “...had made a readvance and had adjusted its border to these
topographic features” (pg. 308). He found the inter-ridge depressions, partially

occupied by Glen, Platte, and Crystal Lakes (Figure 5), blanketed with till to

97

some degree and speculated that the till was deposited by a later ice advance
following formation of the Manistee moraine (pg. 309). Clearly, the timing of the
deposition of the Lakeshore ridges troubled Leverett, and he hinted that they may
have formed prior to the ice advance (Figure 2) that formed the Manistee
moraine (Figure 4).

Leverett and Taylor’s use of the term “moraine” when describing the
Manistee moraine follows accepted views at the time, when any kind of
hummocky topography inferred a moraine deposit. Martin’s (1955) map adhemd
to this definition, as did Farrand and Bell’s (1982) map.

A more recent definition by Flint (1971, pg. 199) expanded on this
classical approach by stating that moraines are composed chiefly of till, and are
the result of direct deposition by active glacial ice. In addition, during Leverett’s
time, there was also greater emphasis on the importance of an advancing ice
margin in producing modern glacial landscapes.

Not all early researchers accepted Leverett’s general observations about
the origin of the ridges at the Lakeshore. Calver (1947), working on developing
the glacial and post-glacial history of the Plate and Crystal Lakes lowlands
(Figure 5) in Benzie County, expressed doubt that all the ridges there (PBM and
other associated linear ridges) were moraines. Instead, he suggested that
“. . .these ridges may have formed as crevasse fillings in a stagnant block of ice
that once occupied the Platte Valley.” (pg. 13). However, Calver failed to provide
any solid sedimentological evidence for this interpretation, nor did he suggest

when the ridges formed. Nonetheless, he was particularly intrigued by the

98

apparent linearity of the ridges and correctly observed that they were generally
parallel to ice flow (northwest-southeast).

Melhom (1954) later reported that “between Benzonia and Empire, the
moraine (Manistee) is predominantly sand and there is very little till” (pg. 88).
Taylor (1990) described thick accumulations of stratified sand and gravel
occurring beneath the surface of the PBM and suggested that it represents an
outwash deposit of Port Huron age.

To summarize, this thesis has demonstrated that the clay beds in the bluff
exposures first noted by Desor ( 1851) were deposited in a lacustrine
environment as observed by Leverett (1915) but not directly by advancing glacial
ice, and that the ridges themselves are not composed of till. They were formed
at a stagnant Port Huron ice margin. Therefore, they are not moraine deposits
as originally suggested by Leverett (1915). Finally, they are not outwash
deposits as suggested by Taylor (1990) since they are isolated, truncated ridges
of lacustrine material oriented normal to the ice margin.

The major implication therefore is that the western part of the Manistee
moraine (Figures 4, 5), as it occurs in the Lakeshore area as mapped (Figure
12), has been misinterpreted. Instead, the Manistee moraine at the Lakeshore
was not produced by advancing Port Huron ice about 13,000 yrs B.P. but was

formed by active, unstable, retreating ice probably during either the Port Huron or

Two Creeks retreat sometime between 13,000 and 12,200 yrs B.P.

99

Histou of crevasse-fills
Crevasse fill deposits have been described in the literature since 1928

(Flint, 1928) and are referred to under a variety of names that include kames,
kame ridges, disintegration ridges, ice-crack ridges, ice disintegration ridge
networks, or ice-walled lake deposits (Flint, 1928; Calver, 1947; Colton and
Jensen, 1953; Klajnert, 1966; Huddart, 1983; Blewett and Winters, 1995; Ham,
1998). Generally, the basic model is the same in most cases, where in-filling of
supraglacial, englacial, or subglacial cracks, fissures, or crevasses with mostly
ice-contact sediments is delivered via runoff, debris-bands, plumes, and/or
drainage networks over, within, or beneath a stagnant glacier.

There have been few detailed observations made regarding mega
crevasse fillings in the us. and Canada. Application of the term “mega-crevasse
fill” for large linear ridges of the type observed at the Lakeshore is nonexistent
and instead the emphasis historically has been on describing smaller landforms.

Flint (1928) was the first to mention and describe the chief
sedimentological characteristics of crevasse fillings as they occur in the
Connecticut area. Flint observed that these features are linear, flat-topped
ridges, with the largest less than 11 min height and 1.6 km in length, and most
less than 3 to 5 m in height and about 400 m in length (pg. 414). Bedding is
mostly horizontal to subhorizontal with transverse sections exhibiting lateral
slump. According to Flint, these crevasse fill deposits formed within water-filled

fractures or crevasses at or near the edge of a stagnant ice mass (pp. 414-415).

100

Ice necessarily must be stagnant, Flint states (1928), otherwise movement would
have disrupted these deposits.

Slightly larger crevasse fill deposits were mapped in northeastern
Montana by Colton and Jensen (1953) who described essentially straight ridges
of till 6 m in height, 30 to 60 m in width, and up to 3.2 km in length, with some
ridges intersecting at angles ranging from 30 to 90 degrees. However, they
found little evidence for subaqueous deposition, and, as a result, concluded that
ice wastage was slow and produwd only minimal amounts of meltwater.

Duckworth (1978) described coarse-textured, buried crevasse fillings with
heights up to 20 m and with lengths up to 450 m of Port Huron age in the vicinity
of Stouffville, Ontario. He describes deposition as either occurring at a tunnel
mouth into standing water within a wasting, retreating ice front or the filling of
linear crevasses in an ice front projecting into standing water. Both scenarios
involve subaqueous deposition near a stagnating ice margin. The lack of
extensive clay beds, however, seems to support the likelihood that these
deposits formed by debouching of detritus via a tunnel mouth of some sort, or
may even be eskers.

Klajnert (1966) described 1 to 1.5 km wide, 40 to 80 m thick, and 4 to 8 km
long flat-topped ridges which comprise the Domaniewice Hills occurring in
Central Poland in the vicinity of Lowicz. These deposits were previously thought
to be an end moraine produced by the waning phase of the Wartian ice sheet
and were reinterpreted by Klajnert to represent crevasse fillings . These ridges

are roughly the size and scale of the ridges at the Lakeshore as well as similar in

101

general sedimentology and structure. However, Klajnert ( 1966) interprets their
formation as having occurred into “dry' crevasses by glaciofluvial deposition.
The extent of ridges (crevasse-fills?) in the Lake Michigan basin. As stabd
eariier, for the Lake Michigan basin, the emphasis has historically been on
describing deposits and landforms associative with advancing ice margins and
not stagnating ones (Leverett and Taylor, 1915; Thwaites and Bertrand, 1957;
Hough, 1958). Recently, several workers have noted a marked absence of till in
landforms associated with Late Wlsconsin ice advances in the basin (Blewett,
1990; Clark and Rudloff, 1990; Blewett and Vlflnters, 1995; Rovey II and Borucki,
1995; Ham, 1998). Blewett (1990) and Blewett and Mnters (1990; 1995) contest
that the inner and outer Port Huron moraines near Kalkaska are actually
collapsed heads of glaciofluvial outwash for the nearby Mancelona and outer
Port Huron outwash plains formed by melting Port Huron ice and are therefore
not classic end moraines. Blewett’s sedimentological work on the ridges reports
very little till but mostly water-sorted, stratified sand and gravel which is
interpreted as representative of major heads of outwash. Blewett and Winters
(1995) also described several small, linear crevasse-fills that radiate away from
the main inner Port Huron ridge like spurs, which they associated with the
outwash and which therefore act as climatic markers for the transition to warmer
temperatures with final deglaciation.

In the southern region of the basin along the western shore of Lake
Michigan in northern Illinois, coastal bluffs showing stratified deposits which had

been earlier interpreted as till sheets, were reinterpreted as subaqueous units by

102

Clark and Rudloff (1990). In addition, Rovey and Borucki (1995) reported on
similarly stratified sediments in the St. Francis bluffs located in southeastem
Wisconsin (Figure 2) and suggested that they represent subaqueous fans. Ham
(1998, pg. A-113), in an abstract presented at the Geological Society of America
national convention, described broad, hummocky moraines partially underlain by
collapsed lacustrine sediment occurring in north-central Wisconsin. He
interpreted these landforms as ice-disintegration landforms developed at the
margin of the Laurentide ice-sheet about 18,000 to 15,000 yrs. B.P. and
suggested that stagnation of the ice sheet persisted for several thousand years
after the margin reached its maximum extent.

High transverse ridges are not unique to the Lakeshore area but are
present elsewhere along the Lake Michigan basin. Analysis of topographic and
2° (4°X6°) surficial geologic maps of the Lake Michigan basin suggest that
morphologically-similar ridges are widespread and seem to occur predominantly
along the eastern margin of the basin. Specifically, the greatest frequency is
along the modern Iakeshore in western Michigan, especially north of Muskegon,
whereas a few smaller subdued ridges occur in the northeast part of the basin
around Charievoix and Emmet Counties.

In the area between Frankfort/Elberta and Manistee are several ridges
associated with the Manistee moraine that trend roughly east-west to northwest-
southeast. There, the largest ridge, located just south of Elberta, is about 1 km
wide and nearly 8 km long. To the south are several east-west trending ridges in

the Big Sable/Ludington area that are associated with the Lake Border moraine

103

and are 0.8 to 2 km wide and up to 12 km long. A single, long, broad, northeast-
southwest-trending ridge, associated with the Whitehall moraine, occurs near
Muskegon and is about 1 km wide and 7 km long. Just west of the Straits of
Mackinac and around the Indian River lowland (Figure 2) are several ridges of
various dimensions but with low relief. These low ridges mostly trend north-south
or northwest-southeast and are associated with, and are oblique to, the inner
Port Huron morainal belt (Blewett, 1990; Blewett and Winters, 1995).

Less distinct ridges occur along the western margin of the Lake Michigan
basin and are generally more difficult to separate from adjacent morainal
landforrn assemblages. In the Kewaunee area of eastern Wlsconsin along the
southern part of the Door Peninsula (about 100 km west-southwest of the
Lakeshore-Figure 2) are several indistinct ridges that trend northeast-southwest
and are therefore mostly normal to the western margin of the LML. In addition,
several low ridges 30 to 60 m high also occur further to the south along the
shoreline between Manitowoc and Sheboygan.

At this time, no workers have yet provided an explanation for the
predominance of larger, more definable transverse ridges along the eastern
margin of the Lake Michigan basin. However, several simple theories are
possible. These include the following: 1) the LML during the Port Huron
substage was thickest along its eastern margin and therefore formed the thickest
accumulations of material, 2) the LML margin was “dirtier” along its eastern
periphery since it had dredged up lacustrine sediments out of the basin and so

provided the large volumes of material needed to build the high ridges during its

104

melting, and 3) there were a higher occurrence of tensional fractures along the
eastern ice margin than along its western margin as the LML splayed out over
the Michigan mainland. In addition, these fractures were likely formed because
the shear strength of the ice front was overcome due to possibly higher shear
stresses (due to obstacles? greater material loads?) over on the Michigan side of
the LML. This shear component probably helped to produce some of the cracks,
joints, and fracture planes necessary to entrap sediment along the ice margin
(Boulton, 1979). Foreland and/or subglacial material was, as a consequence,
entrained or intruded into the fracture planes within it (Boulton, 1979; Lawson,
1979; Huddart, 1983). During the subsequent retreat of the ice margin, melting
occurred preferentially along the fiacture planes within it. As a result, these
fractures widened and eventually were filled in with material (all subaqueously?)
to form the ridges. One or all of these reasons may provide the answer. Even
so, they are but a few out of multiple explanations. However, a proper treatment

of this issue lies outside the scope of this thesis.

105

Lake Level History

A series of lake-levels at the Lakeshore (Table 3, Figure 30) are
represented by morphological landforms as well as lacustrine sediments. The
landforms include well-developed wave-cut scarps (Eyles and Eyles, 1983;
Larsen, 1985), lacustrine plains (Bretz, 1951; Hansel et al., 1985; Taylor, 1990),
erosional tenaces (Leverett and Taylor, 1915; Larsen, 1985; Taylor, 1990),
stream terraces (Bretz, 1963; Evenson, 1973; Larsen, 1985; Taylor, 1990), and
delta surfaces (Calver, 1947; Taylor, 1990). Sedimentological evidence includes
lake sediments (Evenson, 1973; Shaw and Archer, 1978; Eyles and Eyles, 1983;
Hansel et al., 1985), beach/dune sand (DuBois, 1978; Clifton and Dingler, 1984;
Colman et al., 1994), stream terrace deposits (Calver, 1947; Bretz, 1951; Fraser
et al., 1990; Rovey II and Borucki, 1995), and deltaic sediments (Gilbert, 1890;
Morgan, 1970; Wright, 1977).

The highest strandline in the study area occurs between 222 to 225 m
elevation (Figure 30) and is represented by wave-cut scarps traceable for nearly
2 km on the north and east-facing slopes at Alligator Hill (Figures 5, 26) just
northwest of Glen Lake, for about 1 km on the westward and northeastward
slopes of North Manitou Island, for nearly 4 km around Lake Manitou at North
Manitou Island, and for about 2 km on the eastward facing slopes of South
Manitou Island. It is also represented by stream terraces near the west-facing
slope of Miller Hill ridge (Figure 5) at about 222 m elevation, along the eastern

rim of the Glen Lake lowland at about 222 to 224 m elevation and on the north

106

facing slope of the Empire ridge in the Empire lowland at about 223 to 225 m
elevation. In addition, two well-developed terraces, roughly graded to each other
(a photograph of one is shown in Figure 24), also occur in the Empire channel
(Table 3IFigure 30: Site 16) at about 225 m elevation and another well-developed
terrace occurs near the Honor delta complex in the Platte River channel (Site 19)
at about 220 m elevation. Lastly, a broad terrace occurs on the north-facing
slope of the PBM (Site 18) also at about 225 m elevation.

This strandline is illustrated in Figure 30 by a roughly horizontal line that
dips slightly southward and is drawn through points representing the above
elevations taken around the Lakeshore. The line extends almost 100 km along
the Lakeshore area, is nearly 48 m higher than the present level of Lake
Michigan (177 m elevation), and its continuity and extent suggests the level of a
single lake.

A lower strandline also occurred between 213 to 220 m elevation (Figure
30) and is mainly represented by poorly-developed wave—cut scarps along some
of the westward-facing slopes of North Manitou Island (Site 10), Honor delta (Site
18), broad planar surfaces at about 213 to 218 m elevation, as well as by an
extensive, planar surface (erosional? delta?) located along the southern side of
the PBM just east of Beulah (Site 20) at about 215 to 220 m. This strandline is
poorly developed at the Lakeshore and its localized nature mostly in the Platte
Lake and Crystal Lake lowlands may indicate that it developed within an

interconnected lake that had once existed there.

107

E .o. l 8. 8......
a 8. l 8. 5:...
”33...:

a o. a l 8. v6.2....
8 om. l 8. :2...
”5:53“:

a com. I 8. vas...
a com I 8. :3...
"68350—0

.23 2 ..e 8 8.8.0
8 mm.

683...:

a om.
"auvdou:

a 2m

5053

8 man

”...... 8833...
88 a Lens...

8 cm.

Jog—d0

38 .8. <2
”sumo—m

83 2 888..

 

 

.3565? 33.9% 288.2 >_o.m.onoE Eng..."
mos: 35.... new 356:8.» eoao_o>oc.=o; £3.» mos: Eow ..amE co». m 0.2.... 5 3.»: 3% >023» o.
8.9 min-.. :0 BgEaz .39. ..m .0 {so new .88: 6:6... .85 S comco>m E9. mco_.m>o_o 05.035

2.... E59.» om_< .ococmoxm. on. .m 32.90.». 2:39.» _o>o_ 9.3 .o .0... conflmfiucgo... .8 05m...

 

 

 

 

 

 

 

 

 

I58 .8..- oem Emoz
an as a a a 2 2 2 2 4. :2 = 2 2.23.. s .

r, Pkkr fr w. p w: r, bwrw’». i

8... a .....I.o..I.|o.|o.|l..ool|T .8.
lololllral ollrllollul
|nl o
8.. I? olfildlllobllbllllubl .8.
O O .

o e e

8.. _ a d: s s s 11:11.: a...
. a a o o a o o
a... o a:
. llllllllllll b I h
a
an? lo'fll'odlllolurfillfilo. .8.
emu. O .99
o o
33 ...:
3
...... .18
_ _ _
2.9.3 2.9 Said..-

gait—n.—

ecu-5.“.—

108

An even lower strandline occurs between 198 to 201 m elevation (Figure
30) and is represented by well-developed wave-cut scarps along some of the
westward-facing slopes and ridges for 0.5 to 1 km near Frankfort and Elberta, as
well as several high, flat, lacustrine plains located mostly within lowlands
associated with Little Traverse Lake, Lime Lake, Glen Lake, Platte Lake, Little
Platte Lake, and Crystal Lake, or near channels, and along the margins of the
Manitou Islands. This strandline extends almost 100 km along the Lakeshore
(Figure 30), is about 25 m higher than the present level of Lake Michigan, and its
relative continuity suggests the level of a single lake or several lakes. However,
the water-plane for this lake-level is not continuous along the entire Lakeshore
and is either poorly developed or not present for nearly 50 km between Glen
Arbor (Site 12) and Honor (Site 18). Between these two sites, scarp elevations
deviated from the 198 to 201 m water-plane by about :I: 5 m. One possible
explanation for this gap may be that the area between Site 12 and Site 18
includes two broad lowlands (Empire and Platte Lakes lowland-Figure 5) that
extend several krn inland. The scouring energy from wave-action would have
been diminished over this distance and therefore would have been ineffective in
producing a defined scarp.

A still lower strand occurs at 189 to 192 m elevation (Figure 30) and is
represented by wave-cut scarps along parts of the southern slope of South
Manitou Island (Figure 25), well-developed ridges of beach/dune deposits
developed upon lake-bottom deposits in the lowlands around Glen Lake, Little

Traverse Lake, Platte Lake, Little Platte Lake, and Crystal Lake, and storm wave

109

lag gravel along some of the scarps around Empire (Site 16) and Beulah (Site
20). It is also represented by lacustrine sediments (Glen Lake deposit) that occur
in the lowlands along the Lakeshore, particularly in the Empire area (Site 16)
where the deposit extends to the base of a steep, north-south trending slope just
east of Empire, around the present margins of Little Glen and Glen Lakes, and
along the northern boundary of the Plate Lake lowland at 191 m. In addition, it is
also represented by an arcuate beach/dune ridge (beech/dune complex) around
Little Traverse and Lime Lakes, is just over 2 km in length, overlies Glen Lake
deposits, and separates the two lakes from one another. This strandline extends
almost 100 km along the Lakeshore area (Figure 30), is nearly 15 m higher man
the present level of Lake Michigan, and provides evidence for a lake that once
occupied the Lake Michigan basin.

The next lower strand occurs at 184 to 186 m elevation (Figure 30) and is
represented by an extensive lacustrine plain (Glen Lake deposit) on the mainland
and on the Manitou Islands (Sites 10, 11), well-developed ridges of beach-dune
sediments on the mainland and on the islands, and fossil-rich sands at the
easternmost end of the Pyramid Point bluff. It is also represented by lacustrine
deposits (Glen Lake deposit) in the lowlands along the Lakeshore, and is
especially prominent where the lake sediments skirt along the northern margin of
Alligator Hill (Site 15), along the northeastern margin of Pyramid Point near Shell
Lake (Site 7), and along the flanks of ridges that occur around Onkeonwe Beach
(Site 21). It is likewise represented by lacustrine benches along the eastern side

of South Manitou Island (Site 11 ) that extend for nearly 3 km, and near the

110

eastern margin of the Platte Lakes lowland around Honor (Site 18). Also near
this elevation are two well-developed, extensive ridges of sand (beach/dune
complex? spits?) about 1 to 1.5 km long that extend across isolated basins
located on North Manitou Island (Site 10). Associated with these deposits at
about 184 to 186 m are numerous curvilinear ridges of sand and gravel
(beach/dune complex) which vary from 0.25 to just over 1 km in length that are
superimposed on lacustrine sediments (Glen Lake deposit). Finally, fossiliferous
sands occur at Pyramid Point near Hidden Lake (Site 7) at about 185 to 186 m.

This strandline extends almost 120 km along the Lakeshore area (Figure
30), is about 9 m higher than the present level of Lake Michigan, and the
continuity of this strandline provides evidence for the minimum elevation of a lake
that once occupied the Lake Michigan basin.

The lowest strand occurs at 179 to 181 m elevation (Figure 30) and is
represented by beach/dune deposits and modern beach deposits that occur in
the lowlands partially occupied by Glen Lake near Glen Arbor (Site 12), Little
Traverse Lake (Site 3), and up to Empire near Otter Creek (Site 17). The
strandline represented by beach/dune ridges is less well-developed south of Site
17 around Platte Lake, Little Platte Lake (Sites 18, 19), Crystal Lake (Sites 20,
21), and Lower Herring Lake (Site 24). The beach/dune ridges at the 179 to 181
m elevation form mostly curvilinear shapes that vary both in width and extent.
However, they are usually less than 100 m wide and 0.5 km long and become
more fragmented as they approach the modern shoreline, probably due to

reworking by storm waves. On the other hand, the modern beach deposit

111

extends along large areas of the shoreline at the Lakeshore. Oftentimes, the
beach width is thinnest along exposed headlands such as along Pyramid Point,
Sleeping Bear Point, and Empire bluff where the beach is about 1 to 6 m wide,
and is widest in semi-sheltered bays along the shore such as along Good Harbor
Bay, Sleeping Bear Bay, and Platte Bay where the beach is about 60 to 150 m
wide.

This strandline extends along the Lakeshore for 60 to 70 km (Figure 30)

 

and is almost 4 m higher than present. Its relative continuity provides evidence
for the level of a lake that also occupied the Lake Michigan basin. However, as
pointed out earlier, this strandline is not ubiquitous along the Lakeshore and is
generally best represented in protected bays such as within Good Harbor Bay
and Sleeping Bear Bay. One possible explanation for the gradual disappearance
of the 179 to 181 m strandline south of Brooks Lake (Site 14) is likely due to
reworking of nearshore deposits by storm waves along unprotected regions of

the shoreline.

Discussion

Table 4 summarizes the elevations and interpretations of five paleolake
strandlines recorded at the Lakeshore. The 222 to 225 m strand is believed to
be a pre-Algonquin lake and probably represents a late Calumet stage of glacial
Lake Chicago that probably formed at about 11,800 yrs B.P. (Hansel et al., 1985;
Schneider and Hansel, 1990) as a consequence of the northward retreat of the

Greatlakean ice margin. The continuity of this strand and the fact that it is the

112

highest supports the argument that it formed in a large, open, proglacial lake and
suggests that during the retreat of the Greatlakean ice margin, the Calumet stage

of glacial Lake Chicago extended into the northern Lake Michigan basin.

Table 4. Lakeshore paleostrandline elevations and interpretations.

 

Late Pleistocene glacial lakes:

1. 222 - 225 m: a series of levels representing the main Calumet stage
of glacial Lake Chicago (to a pre-Lake Algonquin stage?)

2. 198 — 201 m: a second Calumet stage of glacial Lake Chicago or a
pre-Lake Algonquin stage.

3. 189 - 192 m: Lake Algonquin (Main?)

Eafly-mlddle Holocene Estglacial lakes:
4. 184 - 186 m: Lake Nipissing.

 

5. 179 - 181 m: Lake Algoma.

 

 

The presence of a Calumet stage of glacial Lake Chicago at the
Lakeshore significantly conflicts with interpretations made by earlier workers. For
example, as shown in Figure 31, this Calumet strand or level at 222 to 225 m is
nearly 32 m higher than the level suggested by Evenson (1973) and about 10 m
higher than the level suggested by Taylor (1990). Evenson based his Calumet

113

 

level on a flat water-plane from the Indiana-Michigan border to Muskegon,
Michigan fixed at 189 m elevation. Taylor (1990) argued that the 210 to 215 m
elevation of the Honor delta as well as planar surfaces located near Beulah
(located only about 4 km to the southwest) were representative of the Calumet
level of glacial Lake Chicago.

Also in the area around Beulah, Taylor attributes a well-developed terrace
(NE%,Sec.14,T.26N.,R.15W.) with an elevation of about 235 m on top of the
PBM to the Glenwood ll stage of glacial Lake Chicago. He also identifies a
terrace located over on the northeast side of Crystal Lake along the south flank
of the PBM (NE% ,Sec.15,T.26N.,R.15W.) and concluded that it too represents
the Glenwood ll stage of glacial Lake Chicago. Taylor (1990) correlates the
highest planar surface (about 230 to 235 m elevation) of the delta at Honor
(Figure 22) to the Glenwood ll stage because of elevational similarities with the
aforementioned terraces around Crystal Lake (pg. 108).

Taylor's (1990) Glenwood II and Calumet levels at the Lakeshore were
probably produced by an interconnected terminoglacial lake that existed during
the retreat of the Greatlakean ice margin. This lake, as Calver had first
recognized (1947), once occupied the Plate Lake-Crystal Lake lowland, was
about 40 m deep, likely fluctuated, and eventually equilibrated at 235 m in
elevation (highest?) to form a strandline. Evidence in support of this
interpretation includes the fact that the delta at Honor (Figures 22, 23) and
terrace landforms around Crystal Lake are well preserved, show no evidence of

erosion, and are localized to the lowlands (Calver, 1947). Their preservation and

114

lack of eroded surfaces at the 235 in elevation suggest that these landforms
were built mostly during the breakup of Greatlakean ice or nearly 1,000 years
later than the age suggested by Taylor. Furthermore, since the landforms are
localized to Plate and Crystal Lake (Figure 30, Sites 18, 20 respectively), they
cannot represent the Glenwood II or Calumet stage of glacial Lake Chicago.

In addition to producing a lake in the Plate Lakes/Crystal Lakes lowlands,
the Greatlakean ice-margin eventually retreated out of the lowlands along the
Lakeshore and produced a series of several cascading lakes. These lakes
formed a chain, as illustrated in Figure 34, from Lake Leelanau southwest to
Crystal Lake over a distance of 30 km, and drained from northeast to southwest
along the ice margin. Specifically, meltwater drained westward along the ice-
margin from Lake Leelanau along a wide channel into the ice-blocked Little
Traverse/Lime Lake lowland, then along another channel that narrowed into a
southerly-bending drainageway out between two ridges and eventually flowed
into the Glen Lake lowland (Figure 34), which was also blocked by ice. Water
then drained out of the lake occupying the Glen Lake lowland through a narrow
gorge cut into the Sleeping Bear ridge and meandered along the margin of the
ice filling the Empire lowland. This drainage eventually flowed into the large

terminoglacial lake occupying the Platte Lake lowland (Figure 34).

115

 

.68: 823 a can a. as. $8:

a .e {so a see 8669.. a. 88: 8:963 8.22s _

 

 

 

 

 

.58. 8...: 928.3.

t8: 823 8 can a. new .23: ..a a {so .a
Emu 36.3... 2 38: 3.55.0 c.5205 c5989...
5...: 92.3%.. o... .o comtmanO .3 2:9".

A§.e8:fl...m=o.<8§a§
8.. 8a 8... 8.. .8 .8 8. .

PhD—bbbI_bhbh-P.Pbb-DIPLPP_DDDP—bbbbtpberlPF .

 

—
clacealma

(U) "OWMIEI

lljll

 

 

8a
a
\\ $2.85....l .8
85823.1. .3

8m

\ :\:\.....0 W85 8. u3...... t r -

 

. mean. 32.33. $

9.76.92 2059.3 c... .o coetmano .8 2:9“.

9.... 885.... mac: 3.335
.3 8v 3 o

Db—Db-F—rlrrblphbbb

 

8
F!

8
H

§
(m) uopakelg

 

GDWWAWH
3
jlllLllllllLlll

 

xx 68:82... .
.2. . sen-.0 .025 com Moe I I t 3a
xx .02.... 8. :3... I
s .28: .8383. mw 8"

 

.88: .2...» Ba .85:
comco>w .0 San o. 885 :o_.m>c_o c.5905
.oEamo 2059.3 9... .o comtmano .5959”.

c5: 88:8... mac? 3:53

80 02. to 8m 89 .3 8n 8n 0
—>.-b-IEDP—PPLPPDED-thh-Dbhb-bIii-I’D.-

aasfi... ......

 

 

(ii) none/tore
§

 

116

The centerpiece for the ice-marginal drainage at the Lakeshore is the
Empire channel (Figure 34) that once connected Glen Lake to Platte Lake. This
channel is shown in Figure 34 (stippled). It is a well-defined, meandering,
erosional landforrn that is entrenched mostly in collapsed outwash
(undifl'erentiated sand and gravel deposit). Also within this channel lies a pair of
stream tenaces whose concordant elevations at about 225 m elevation (Table 3:
Site 16) suggest that water flowing through the channel was at least 5 m deep.
Imbricated clasts in the terrace deposit indicate that the main direction of flow
was from north (Glen Lake) to south (Platte Lake).

Drainage from the Platte Lake lowland terminoglacial lake flomd through
a narrow channel cut into the PBM and into the terminoglacial lake occupying the
Crystal Lake lowland (Figure 34). Morphological evidence for this lake level
(highest?) include Taylor’s “Glenwood Il-age' (1990) strandlines and terraces at
about 235 m elevation (Table 3: Site 20) as well as two large, planar surfaces,
also interpreted by Taylor (1990) as Glenwood II in age, located just east of
Beulah.

During this period of ice-marginal drainage at the Lakeshore, the main
body of the Honor delta was being developed. The Honor delta (Honor sand and
gravel deposit) was built primarily by the Platte River as it discharged sediment-
concentrated water from sources to the east into the terminoglacial lake
occupying the Plate Lake lowland. The delta is slightly unusual in that sections
of it consist of planar bedded to shallowly-dipping (5 20°) sand and gravel facies

(Figure 20) with laminated-to-wavy clay beds (Figure 23). Brodzikowski and van

117

Loon (1991) point out that these textural changes may occur because barriers
surrounding the terminoglacial lake (such as “ice masses and irregular
topography’; pg. 111) can cause disturbances affecting the zonal buildup of the
deltas, such as the supply of water and debris, and ultimately the water level.
Consequently, factors like these would make it impossible to distinguish between
topsets, foresets and bottomsets in any deltaic sequence.

Continued ice retreat out of the lowlands along with decreases in water
input via ice-marginal drainage channels into the Platte Lake-Crystal Lake
lowlands subsequently caused a drop in lake level and resulted in a cutoff
between Platte Lake and Crystal Lake. Lastly, complete ice retreat out of the
lowlands caused waters from the terminal lakes to escape and equilibrate with
the 222 to 225 m Calumet level of glacial Lake Chicago.

The 198 to 201 m strand that occurs at the Lakeshore is believed to
represent a pre-Algonquin lake level or possibly a low Calumet stage of glacial
Lake Chicago (Table 4) formed as the Greatlakean ice margin retreated
northward sometime between 11,800 yrs. B.P. and 11,200 yrs. B.P. (Hansel et
al., 1985; Schneider and Hansel, 1990; Larson et al., 1994). The drop in level
from about 225 m elevation to about 201 m elevation is perhaps due to the
deglaciation of the Indian River lowland (Hansel et al., 1985). This interpretation
is shared by Larsen (1987) and Taylor (1990) who suggest that the Calumet
levels in the northern part of the Lake Michigan basin were highly unstable and
short-lived and probably experienced other sharp drops just prior to the main

Algonquin level.

118

 

 

 

     
   
   
  

EXPLANATION

.3“

as.

Undilf. sand. gravel
Pearl Lake outwash
Pyramid Point deposit

Duck Lake diamid

 

 

K I 4 m
E:
o n 4 .-

\,,I
X t ‘ \ direction of water flow
Honor delta

‘ complex ~ direction of ice

movement

 

= etsie channel

 

 

 

Figure 34. Mature stage of Greatlakean ice retreat/stagnation with formation
of a series of large cascading lakes partially or completely ice blocked
(11,800 yrs-10,700 yrs B.P.). Main channels or drainageways are shown by
a stippled pattern. Arrows show probable direction of drainage.

119

The 189 to 192 m strand at the Lakeshore is believed to represent Main
Lake Algonquin (T able 4) which formed sometime between 11,500 yrs. B.P. and
11,200 yrs. B.P. (Larsen, 1987). Main Lake Algonquin developed as a
consequence of retreat of the Greatlakean ice margin north of the Straits of
Mackinac (Larsen, 1987; Schneider and Hansel, 1990; Taylor, 1990; Larson et
al., 1994). The 189 to 192 m strand at the Lakeshore (Figure 32) corresponds
well with the Main Lake Algonquin strand suggested by Larsen (1987) and Taylor
(1990), as well as that predicted by Clark et al.’s thin-ice model (1994).

The 184 to 186 m strand at the Lakeshore is believed to represent Lake
Nipissing (Table 4) which formed sometime after 9,000 yrs B.P. (Larsen, 1985).
Lake Nipissing formed when the outlet at North Bay, Ontario was progressively
isostatically raised to an elevation above that of the southern outlet at Port Huron
(Larsen, 1985), which resulted in a rise of lake level elevation close to that‘of
earlier Lake Algonquin between 9,000 yrs and 5,000 yrs B.P. (Larsen, 1985).
This event likely marked the establishment of the Nipissing Great Lakes whose
level, at first, was controlled by three outlets; Chicago, Port Huron, and North Bay
(Hansel et al., 1985; Larsen, 1985). By comparison, the 184 to 186 m strand at
the Lakeshore (Figure 33) is within one or two meters of the Nipissing strand
reported by Larsen (1987), about three meters higher than the Nipissing strand
reported by Taylor (1990), and one to three meters of the Nipissing level
predicted by Clark et al.’s thin-ice model (1994).

The 179 to 181 m strand at the Lakeshore is believed to represent Lake

Algoma (Table 4) which formed between 4,700 to 2,500 yrs B.P. (Larsen, 1985)

120

due to incision of the Port Huron outlet and abandonment of the outlets at
Chicago and North Bay (Hansel et al., 1985). The 179 to 181 m strand is within
0.5 meter of the Algoma strand for the northern Lake Michigan basin reported by
Larsen (1987). Following Lake Algoma, lake level at the Lakeshore steadily
equilibrated due to incision resulting in only about a 2 m drop in the last several
thousand years to form present day Lake Michigan (177 m elevation).

Correct placement of glacial and post-glacial lake levels at the Lakeshore
has been difficult and inconclusive mostly due to the lack of datable material in
the lacustrine deposits. Also, inaccuracies on lake level elevations are
unavoidable using mostly 5 m contours drawn on 7.5” topographic maps. Such
errors could be greatly reduced with the use of modern survey equipment such
as a GPS receiver or a Total Station. Nonetheless, morphologic and scattered
sedimentologic evidence does suggest that a series of lakes did at one time exist
at the Lakeshore, that the oldest level there was at about 222 to 225 m, and likely

represents the Calumet stage of glacial Lake Chicago.

121

CONCLUSIONS

Over the last several decades, there have been numerous studies of the
glacial and post-glacial history of the Sleeping Bear Dunes National Lakeshore
region. Three fundamental questions, however, remain unresolved: 1) how far
did Greatlakean ice advance into the area, 2) what is the origin of the ridges at
the Lakeshore, and 3) what glacial lake stage followed the retreat of the ice
margin.

Detailed surficial mapping of the Lakeshore (Figure 12) shows the region
to be characterized by a westward sloping drumlinized till plain partially buried by
a complex of glaciolacustrine and glaciofluvial deposits, as well as by littoral and
eolian sediments. Analysis of the map suggests the following:

1) The Duck Lake diamict represents the Port Huron ice advance and is
probably a subglacial till. In addition, it is clear that deposition of Duck Lake
diamict predates deposition of both the Pyramid Point and Pearl Lake
deposits. This would imply that the Pyramid Point and Pearl Lake deposits,
as well as most of the Duck Lake diamict, must have been formed prior to the
Greatlakean advance and probably represent deposits associated with Port
Huron-age ice.

2) There is no evidence for the Greatlakean ice advance at the Lakeshore.

3) Greatlakean ice probably invaded into the lowlands and was likely

constrained by high Port Huron-age topography.

122

4)

5)

The ridges are most likely subaqueous crevasse fillings (Pyramid Point
deposit) formed during the retreat/stagnation of the Port Huron ice sheet
sometime during the Two Creekan Substage.

Paleolake-levels are represented by a series of strands: the strand at 222 to
225 m elevation probably represents the Calumet stage of glacial Lake
Chicago; the strand at 198 to 201 m elevation probably represents a low
Calumet stage of glacial Lake Chicago; the strand at 189 to 192 m elevation
probably represents the Main Lake Algonquin stage; the strand at 184 to 186
m elevation probably represents Lake Nipissing; and the strand at 179 to 180

m elevation probably represents Lake Algoma.

123

LIST OF REFERENCES

Acomb, L. J., D. M. Mickelson, and E. B. Evenson, 1982, Till stratigraphy
and late glacial events in the Michigan Lobe of eastern Wisconsin:
Geological Society of America Bulletin, v. 93, p. 289-296.

Alden, WC, 1904, The Delavan lobe of the Lake Michigan glacier of the
Wisconsin stage of glaciation and associated phenomena, US. Geological
Survey Professional Paper 34, 106 p.

1918, The Quaternary geology of southeastern Wisconsin, US.
Geological Survey Professional Paper 106, p. 356.

Ashley, GM, 1975, Rhythmic sedimentation in glacial Lake Hitchcock,
Massachusetts, Connecticut, in Jopling, A. V., and McDonald, B. C., eds.,
Glaciofluvial and glaciolacustrine sedimentation, Society of Economic
Paleontologists and Mineralogists, p. 304—320.

Bagnold, R.A., 1936, The movement of desert sand: Proceedings of the
Royal Society of London, v. 163, p. 250-264.

Bergquist, S.G., 1954, Red drift of the Manistee Moraine in Michigan:
Geological Society of America Abstracts with Programs, Dec. Meeting,
Toronto, p. 1394-1 395.

Blewett, W.L., 1990, The Glacial Geomorphology of the Port Huron
Complex in Northwestern Southern Michigan [Ph.D. thesis]: Michigan State
University, 191 p.

Blewett, W. L., and H. A. Winters, 1995, The Importance of Glaciofluvial
Features within Michigan's Port Huron Moraine: Annals of the Association of
American Geographers, v. 85, no. 2, p. 306-319.

Boothroyd, J.C., and GM. Ashley, 1975, Processes, bar morphology, and
sedimentary structure on braided outwash fans, northeastern Gulf of Alaska,
in Jopling, A. V., and MacDonald, B. C., eds., Glaciofluvial and

124

glaciolacustrine sedimentation, Society of Economic Paleontologists and
Mineralogists Special Paper 23, p. 193-222.

Boulton, GS, 1968, Flow tills and related deposits on some
Vestspitzbergen glaciers: Journal of Glaciology, v. 7, p. 391-412.

' 1979, Processes of glacier erosion on different substrata: Journal of
Glaciology, v. 23, p. 15-38.

Bretz, J.H, 1951, The stages of Lake Chicago: Their causes and
correlations: American Journal of Science, v. 249, p. 401-429.

1959, The double Calumet stage of Lake Chicago: Journal of
Geology, v. 67, p. 675-684.

1963, Correlation of glacial lake stages in the Huron-Erie and
Michigan Basins: Journal of Geology, v. 72, p. 618-627.

Brodzikowski, K., and AJ. VanLoon, 1991, Glacigenic Sediments:
Developments in Sedimentology, 49: Amsterdam, Elsevier Science
Publications, 674 p.

Broecker, W.S., and W.R. Farrand, 1963, Radiocarbon age of the Two
Creeks Forest Bed, Wlsconsin: Geological Society of America Bulletin, v. 74,
p. 795-802.

Calver, J. L., 1947, The Glacial and Post-Glacial History of the Platte and
Crystal Lakes Depressions, Benzie County, Michigan, Occasional Papers on
the Geology of Michigan for 1946, Part II: Geological Survey Division
Publication 45, Geological Series 38, Michigan Dept. of Conservation,
Lansing, 70 p.

Chamberlin, TC, 1877, Geology of eastern Wisconsin, Geology of
Wisconsin, survey of 1873-1877, Wisconsin Geological Survey, p. 91-405.

Clark, J.A, M. Hendriks, T.J. Trmmermans, C. Struck, and K.J. Hilverda,
1994, Glacial isostatic deformation of the Great Lakes region: Geological
Society of America Bulletin, v. 106, p. 19-31.

125

Clark, J.A., H.S. Pranger ll, J.K. Walsh, and JA. Primus, 1990, A
numerical model of glacial isostasy in the Lake Michigan basin, in Schneider,
A. F., and Fraser, G. S., eds., Late Quaternary History of the Lake Michigan
Basin, Geological Society of America Special Paper 251, p. 111-123.

Clark, P.U., and GA. Rudloff, 1990, Sedimentology and stratigraphy of
late \Nrsconsinan deposits in Lake Michigan bluffs, northern Illinois, in
Schneider, A. F., and Fraser, G. S., eds, Late Quaternary History of the
Lake Michigan Basin, Geological Society of America Special Paper 251, p.
2941.

Clifton, HE, and JR. Dingler, 1984, Wave-fonned structures and
paleoenvironmental reconstruction: Marine Geology, v.60, p. 165-198.

Colton, RB, and F.S. Jensen, 1953, Crevasse Fillings on the glaciated
plains of Northeastern Montana: Geological Society of America Bulletin
Abstracts with Programs, May Meeting, Butte, v. 64, no. 12, p. 1542-1543.

Desor, E., 1851, On the Superficial Deposits of the District, in Foster, J.
W., and Whitney, J. D., eds, Report on the Geology of the Lake Superior
Land District, Part II: Washington 0.0., p. 240.

Dow, K.W., 1938, The Origin of Perched Dunes on the Manistee Moraine,
Michigan: Papers of the Michigan Academy of Science, Arts and Letters, v.
23, p. 427-440.

Dreimanis, A., 1989, Tills: their genetic terminology and classification, in
Goldthwait, R. P., and Matsch, C. L., eds., Genetic classification of glacigenic
deposits: Rotterdam, AA. Balkema, p. 17-83.

Dreimanis, A., and RP. Goldthwait, 1973, Wisconsin glaciation in the
Huron, Erie and Ontario lobes: Geological Society of America Memoir, v.
136, p. 71-106.

Drexler, C.W., 1975, Geological report on Sleeping Bear Dunes National
Lakeshore: University of Michigan Biological Station, Douglas Lake, Pellston,
and Ann Arbor, in Natural History Surveys of Pictured Rocks National
Lakeshore and Sleeping Bear Dunes National Lakeshore, p. 41-49.

126

DuBois, RN, 1978, Beach topography and beach cusps: Geological
Society of America Bulletin, v. 89, p. 1 133-1139.

Duckworth, PB, 1978, Buried Crevasse Fillings of Port Huron age,
Stouffville, Ontario: American Association of Petroleum Geologists Abstracts,
v. 62/3, p. 510-511.

Eschman, D.F., 1985, Summary of the Quaternary History of Michigan,
Ohio and Indiana: Journal of Geological Education, v. 33, p. 161-167.

Evenson, E. B., 1973, Late Pleistocene Shorelines and Stratigraphic
Relations in the Lake Michigan Basin: Geological Society of America Bulletin,
v. 84, p. 2281-2298.

Evenson, E.B., A. Dreimanis, and J.W. Newsome, 1977, Subaquatic flow
till: a new interpretation for the genesis of some laminated till deposits:
Boreas, v. 6, p. 115-133.

Evenson, E. B., W. R. Farrand, and D. F. Eschman, 1974, Late
Pleistocene Shorelines and Stratigraphic Relations to the Lake Michigan
Basin: Reply: Geological Society of America Bulletin, v. 85, p. 661-664.

Eyles, CH, and N. Eyles, 1983, Sedimentation in a large lake: A
reinterpretation of the late Pleistocene stratigraphy at Scarborough Bluffs,
Ontario, Canada: Geology, v. 11, p. 146-152.

Eyles, N., C. H. Eyles, and A. D. Miall, 1983, Lithofacies types and vertical
profile models; an alternative approach to the description and environmental
interpretation of glacial diamict and diamictite sequences: Sedimentology, v.
30, p. 393-410.

Farrand, W.R, and D.L. Bell, 1982, Quaternary geology of southern
Michigan; scale 12500000, Michigan Dept. of Natural Resources.

Farrand, W.R., and DP. Eschman, 1974, Glaciation of the southem
peninsula of Michigan; A review: Michigan Academician, v. 7, p. 35-62.

Flint, R. F ., 1928, Eskers and Crevasse Fillings: American Journal of
Science, v. 5, no. 15, p. 410-416.

127

1971, Glacial and Quaternary geology: New York, Wiley, 892 p.

Fullerton, D.S., 1980, Preliminary Correlation of Post-Erie lnterstadial
Events (16,000-10,000 Radiocarbon years Before Present), Central and
Eastern Great Lakes Region, and Hudson, Champlain, and St. Lawrence
Lowlands, United States and Canada, US. Geological Survey Professional
Paper 1089, 52 p.

Geologic Survey Division, 1981, Bedrock Surface Elevation for Southern
Michigan; scale 12500000, in Hydrologic Atlas of Michigan, prepared by the
Dept. of Geology, College of Arts and Sciences, Kalamazoo, Michigan,
Michigan Dept. of Natural Resources: Underground Injection Control
Program, US. Environmental Protection Agency, Plate 13.

Gilbert, GK, 1890, Lake Bonneville: US. Geological Survey Monograph
1, p. 438.

Goldthwait, J.W., 1906, Correlation of the raised beaches on the west side
of Lake Michigan: Journal of Geology, v. 14, p. 411-424.

1908, A reconstruction of water planes of the extinct glacial lakes in
the Lake Michigan basin: Journal of Geology, v. 16, p. 459-476.

Gustavson, TC, 1974, Sedimentation on Gravel Outwash Fans,
Malaspina Glacier Foreland, Alaska: Journal of Sedimentary Petrology, v. 44,
p. 374-389.

Ham, NR, 1998, Conceptual model for the development of ice-walled
lakes along the southern margin of the Laurentide Ice Sheet: Implications for
deglacial dynamics, Abstracts with Programs, Toronto 1998 Annual Meeting,
Geological Society of America, p. A-113.

Hansel, A. K., D. M. Mickelson, A. F. Schneider, and C. E. Larsen, 1985,
Late IMsconsinan and Holocene history of the Lake Michigan Basin: in
Karrow, PF. and Calkin, P.E., eds, Quaternary Evolution of the Great Lakes,
Geological Association of Canada Special Paper 30, p. 39-53.

Hough, J.L., 1958, Geology of the Great Lakes: Urbana, lll, University of
Illinois Press, 313 p.

128

1963, The prehistoric Great Lakes of North America: American
Scientist, v. 51, p. 84-109.

Huddart, D., 1983, Flow tills and ice-walled lacustrine sediments, the
Petteril Valley, Cumbria, England, in Evenson, E. B., Schluchter, C., and
Rabassa, J., eds., Tills and Related Deposits: Genesis, Petrology,
Application, Stratigraphy: Rotterdam, Netherlands, AA. Balkema, p. 81-94.

Karrow, P.F., T.W. Anderson, A.H. Clarke, L.D. Delorrne, and MR.
Sreenivasa, 1975, Stratigraphy, Paleontology, and Age of Lake Algonquin
Sediments in Southwestem Ontario, Canada: Quaternary Research, v. 5, p.
49-87.

Kelley, R.W., and W.R. Farrand, 1967, The glacial lakes around Michigan:
Michigan Geological Survey Bulletin, v. 4, p. 10-17.

Kempton, JP, and D.L.. Gross, 1971, Rate of advance of the
Woodfordian (Late Wisconsinan) glacial margin in Illinois: stratigraphic and
radiocarbon evidence: Geological Society of America Bulletin, v. 82, p. 3245-
3250.

Klajnert, Z., 1966, Origin of Domaniewicz Hills and remarks on the mode
of waning of the Middle Polish ice sheet: Acta Geographica Lodziensia
(Poland), v. 23, p. 1-134.

Komar, PD, 1971, The mechanics of sand transport on beaches: Journal
of Geophysical Research, v. 76, p. 713-721.

Larsen, C. E., 1985, Lake Level, Uplift, and Outlet Incision, the Nipissing
and Algoma Great Lakes, in Karrow, P. E., and Calkin, P. E., eds.,
Quaternary Evolution of the Great Lakes, Geological Association of Canada
Special Paper 30, p. 63-77.

1987, Geological History of Glacial Lake Algonquin and the Upper
Great Lakes: US. Geological Survey Bulletin 1801, 36 p.

Larson, G. J., T. V. Lowell, and N. E. Ostrom, 1994, Evidence for the Two
Creeks interstade in the Lake Huron basin: Canadian Journal of Earth
Sciences, v. 31, p. 793-797.

129

Lawson, DE, 1979, A sedimentological analysis of the western terminus
region of the Matanuska Glacier, Alaska: US. Army Cold Regions Research
and Engineering Laboratory, CRREL Report 79-9, 122 p..

1982, Mobilization, Movement and Deposition of Active Subaerial
Sediment Flows, Matanuska Glacier, Alaska: Journal of Geology, v. 90, p.
279-300.

Leverett, F., and F .B. Taylor, 1914, Geology of the Southem Peninsula of
Michigan; scale 1:500000, US Geological Survey.

1915, The Pleistocene of Indiana and Michigan and the History of
the Great Lakes: U. S. Geological Survey Monograph 53, p. 308-315; 458.

Martin, H., 1957, Geologic map of Michigan: Geological Survey Division,
Michigan Dept. of Conservation, scale 1:500000.

McKee, ED, 1979, Sedimentary structures in dunes, in McKee, E. 0.,
ed., A study of global sand seas, US. Geological Survey Professional Paper
1052, p. 83-113.

Melhom, W.N., 1954, Valders Glaciation of the Southern Peninsula of
Michigan [Ph.D. thesis]: University of Michigan, 170 p.

Mickelson, D. M., and E. B. Evenson, 1975, Pre-Twocreekan age of the
type Valders till, Wisconsin: Geology, p. 587-590.

Milstein, R.L., 1987, Bedrock Map of Southern Michigan; scale 1:500000,
Michigan Dept. of Natural Resources.

Monaghan, G. W., and A. K. Hansel, 1990, Evidence for the intra-
Glenwood (Mackinaw) low-water phase of glacial Lake Chicago: Canadian
Journal of Earth Sciences, v. 27, p. 1236-1241.

Monaghan, G.W., G.J. Larson, and GD. Gephart, 1986, Late Wisconsinan
drift stratigraphy of the Lake Michigan lobe in southwestern Michigan:
Geological Society of America Bulletin, v. 97, p. 329-334.

130

Morgan, JP, 1970, Deltaic Sedimentation: Modern and Ancient Society
of Economic Paleontologists and Mineralogists Special Publication 15.

Osborne, F.P., 1950, Marine Crevasse Fillings in the Lotbiniere Region,
Quebec: American Journal of Science, v. 248, p. 874—890.

Rovey ll, C.W., and MK. Borucki, 1995, Subglacial to proglacial sediment
transition in a shallow ice-contact lake: Boreas, v. 24, p. 117-127.

Rust, BR, and R. Romanelli, 1973, Late Quaternary subaqueous
outwash deposits near Ottawa, Canada, in Jopling, A, and McDonald, 8.,
eds., Society of Economic Paleontologists and Mineralogists, Special
Publication 23, p. 177-192.

Ryder, J.M., 1971, The stratigraphy and morphology of paraglacial alluvial
fans in south-central British Columbia: Canadian Journal of Earth Sciences,
v. 8, p. 279-298.

Schneider, AF, 1990, Radiocarbon confirmation of the Greatlakean age
of the type Two Rivers till of eastern Wisconsin, in Schneider, A. F., and
Fraser, G. 8., ads, Late Quaternary History of the Lake Michigan Basin,
Geological Society of America Special Paper 251, p. 51-55.

Schneider, AF, and A. K. Hansel, 1990, Evidence for post-Two Creeks
age of the type Calumet shoreline of glacial Lake Chicago, in Schneider, A.
F., and Fraser, G. S., eds., Late Quaternary History of the Lake Michigan
Basin, Geological Society of America Special Paper 251, p. 1-8.

Schoolcraft, HR, 1821, Narrative Journal of Travels through the
Northwestern Regions of the United States to the Sources of the Mississippi
River in 1820, 401 p.

Shaw, J., and J.J.J. Archer, 1978, Winter turbidity current deposits in Late
Pleistocene glaciolacustrine varves, Okanagan Valley, British Columbia,
Canada: Boreas, v. 7, p. 123-130.

Shepard, F.P., and EC. LaFond, 1940, Sand movements near the beach
in relation to tides and waves: American Journal of Science, v. 238, p. 272-
285.

131

Snyder, F .S., 1985, A Spatial and Temporal Analysis of Sleeping Bear
Dunes Complex, Michigan (A contribution to the geomorphology of perched
dunes) [Ph.D. thesis]: University of Pittsburgh, 201 p.

Soil Conservation Service, Michigan, in cooperation with Michigan
Agriculture Experiment Station, 1973, Soil Survey of Leelanau County; scale
12190080, US. Dept. of Agriculture, Washington DC.

Swift, D.J.P., 1975, Barrier island genesis: evidence from the central
Atlantic shelf, eastern U.S.A: Sedimentary Geology, v. 14, p. 1-43.

Taylor, L. D., 1990, Evidence for high glacial-lake levels in the
northeastern Lake Michigan basin and their relation to the Glenwood and
Calumet phases of glacial Lake Chicago, in Schneider, A. F., and Fraser, G.
S., eds., Late Quaternary History of the Lake Michigan Basin, Geological
Society of America Special Paper 251, p. 91-109.

Thwaites, F.T., 1943, Pleistocene of part of northeastem Wisconsin:
Geological Society of America Bulletin, v. 54, p. 87-144.

Thwaites, PI, and K. Bertrand, 1957, Pleistocene Geology of the Door
Peninsula, Wisconsin: Geological Society of America Bulletin, v. 68, p. 831-
880.

Wallbom, TE, and G.J. Larson, 1998, Surficial Geology of the Glen
Haven, Glen Arbor, Good Harbor Bay, Empire, Burdickville, and Beulah 7.5
Minute Quadrangles, Leelanau and Benzie Counties, Michigan; scale
1225000, Michigan State University: Michigan Geological Survey, Dept. of
Environmental Quality, US. Geological Survey.

Waterman, W.G., 1926, Ecology of Glen Lake and Sleeping Bear Region:
Papers of the Michigan Academy of Science, Arts and Letters, v. 6, p. 351-
376.

Wright, L.D., 1977, Sediment transport and Deposition at River Mouths: A
Synthesis: Geological Society of America Bulletin, v. 88, p. 857-868.

132

HICHIGQN STRTE

UNIV. LIBRRRIES
II I | IIHIIIllllllllllllllllllllHUI
0586354

3129302