Dune-pocketed wetlands along the eastern Lake Michigan shoreline

[From the U.S. Government Printing Office, www.gpo.gov]

DUNE-POCKETED
WETLANDS

along the
Eastern Lake Michigan Shoreline

December 1987

vital resources, inc.

DUNE-POCKETED WETLANDS

ALONG THE

EASTERN LAKE MICHIGAN SHORELINE

Prepared for

Michigan Department of Natural Resources
Geological Survey Division, and
Land and Water Management Division

DPO* 88-GA8266
OQ
'Cr
by

James A. Lively and F. Glenn Goff

December 1987

Vital Resources Incorporated
1518 River Terrace Drive
East Lansing, Michigan 48823
Phone: 517/337-1211

TABLE OF CONTENTS Page

INTRODUCT.ION . . . . . . . . . . . . . . . . . . . . . I

METHODS . . . . . . . . . . . . . . . . . . . . . . . 6

II.A Initial Map Analysis . . . . . . . . . . . . . . 7

II.B Aerial Photo Interpretation . . . . . . . . . . 10

II.C Ownership and Disturbance . . . . . . . . . . . 11

II.D Field Inspections . . . . . . . . . . . . . . . 13

II.E Extended and Refined Map Analysis . . . . . . . 15

II.F Classification of Shoreline Drainage System 16

II.G Screening, Ordination, and Classification of
Dune-Pocketed Wetlands Drainage Systems . . . . 17

II.H Preparation of Reference Maps . . . . . . . . . 17

EASTERN LAKE MICHIGAN SHORELINE LANDSCAPE FORMATION
PROCESSES . . . . . . . . . . . . . . . . . . . . . . 19

III.A Glacial Processes . . . . . . . . . . . . . . . 19

III.B Surface Drainage Processes . . . . . . . . . . 26

III.C Shoreline Processes . . . . . . . . . . . . . . 30

III.C.1 Dune Building . . . . . . . ... . . . 30

III.C.2 Littoral Drift . . . . . . . . . . . . 33

III.D Vegetational Processes . . . . . . . . . . . . 34

III.D.1 Species Migration and Colonization . . 34

III.D.2 Succession . . . . . . . . . . . . . . 36

III.D.3 Mineral Enrichment . . . . . . . . . . 38

III D.4 Seasonal Fluctuation of Water Level 40

IV. CLASSIFICATION OF EASTERN LAKE MICHIGAN DRAINAGE
SYSTEMS . . . . . . . . . . . . . . . . . . . . . . . 42

IV.A Interdunal Wetlands . . . . . . . . . . . . . . 47

IV.B Bay Mouth Bar Lakes . . . . . . . . . . . . . . 48

IV-C Dune-Pocketed Wetlands . . . . . . . . . . . . . 49

IV.D Somewhat Impeded Shoreline Streams . . . . . . . 52

IV.E Intermittent Stream Channel . . . . . . . . . . 52

IV.F Unimpeded Shoreline Streams . . . . . . . . . . 54

IV.G Drowned River Mouth Lakes . . . . . . . . . . . 55

IV.H Unimpeded Inland Streams . . . . . . . . . . . . 58

V. DESCRIPTION OF DUNE-POCKETED WETLANDS WATERSHEDS . . . 59

V.A Warren Dunes System . . . . . . . . . . . . . . . 59

V.B Van Buren Dunes System . . . . . . . . . . . . . 63

V.C Saugatuck Dunes System . . . . . . . . . . . . . 66

V.D Michilinda Dunes System . . . . . . . . . . . . . 69

V.E Fruitvale Dunes System . . . . . . . . . . . . . 71

V.G Little Sable Point South Dunes System . . . . . . 74

V.H Big Sable Point/Hamlin Lake/Nordhouse Dunes
System . . . . . . . . . . . . . . . . . . . . . 77

V.I County Line Road Watersheds . . . . . . . . . . . 81

V.J Disturbance Analysis . . . . . . . . . . . . . . 84

VI. ORDINATION AND CLASSIFICATION OF DUNE-POCKETED
WETLANDS WATERSHEDS . . . . . . . . . . . . . . . . . 86

VII. DISCUSSION . . . . . . . . . . . . . . . . . . . . . . 94

VIII. SUMMARY . . . . . . . . . . . . . . . . . . . . . . . 100

IX. LITERATURE CITED . . . . . . . . . . . . . . . . . . . 101

I. INTRODUCTION

The sand dunes of eastern Lake Michigan are recent and dyna-

mic geologic features. They developed upon the framework of the

glacially-formed landscape, and concurrently with the youthful

drainage systems of Michigan, to produce a unique and attractive

shoreline. In order to understand the present ecology of the

dunes, and their relationships with other ecologic al communities

within the shoreline zone, particularly the wetlands associated

with the dunes, it is necessary to examine the processes involved

in the formation of these interlocking systems and subsystems.

Many previous studies of the dunes have dealt with specific

issues or areas rather than with the entire shoreline. Some have

focused on dune ecology and vegetation processes (e.g. Wells and

Thompson, 1975 et seq). Others have looked at the process of Lake

Michigan dune development (Cressey 1928, Scott and Dow 1937, Tague

1@946, Olson 1958, etc.). Some more recent studies have concen-

trated on mapping and inventorying the dunes (Powers 1958, Kelley

1962, MDNR 1976, MDNR 1978, Buckler 1979, MNF1 1986). Management

of the dunes as separate ecosystems was legally addressed within

Michigan in 1976 by enacting the Sand Dune Management and Protec-

tion Act (Michigan PA 222). Resulting from this Act was the

establishment of a set of Designated Sand Dune Areas occurring

within two miles of the shoreline. This procedure provided an

ecologically arbitrary though legally definitive landward boundary

beyond the limits of the high dunes for many shoreline dune areas.

In 1978, the barrier sand dunes were outlined within these areas.

The landward boundary of the Barrier Dunes was defined as "the

most landward base of the first sand dune formation from the Great

Lake Shoreline which displays the greatest relative relief within

a designated sand dune area". Based upon Buckler's 1979 study, a

set of barrier dunes was inventoried and classified as legally

defined above.

Many of the studies of the dunes have been limited to the

area defined as Barrier Dunes. Recent ecological studies in the

Nordhouse Dunes system (Wells and Thompson 1984, Hazlett 1987,

Goff et al 1986, 1987) have included land further inland as part

of the dune system, mainly because the area was under consider-

ation for wilderness protection. In Goff et al's tnvironmental

Study (1986) and Supplement (1987) a "landward dune-pocketed

wetlands" zone was identified as an integral part of the dune

system. The interrelationship between the dunes and the landward

wetlands became a central issue in the Supervisor of Wells' de-

cision to limit hydrocarbon development in the Nordhouse Area

(Guyer, 1987): "the barrier dune formation and associated depen-

dent lands is an example of an ecosystem type that is unique in

its process of development and in the structural and functional

relationships among land units .... the confined natural hydrologic

systems identified within the Nordhouse Dunes Area are of critical

importance to the maintenance of a valuable wetland and associated

surface water environment."

The Nordhouse Dunes issue focused attention on the importance

of ecologically defining the landward limit of the dune system.

The 1987 Supplement to the Nordhouse Environmental Study suggested

that "a more complete survey of dune associated wetlands, consid-

ering the effects of barrier dune formation on upgradient drainage

patterns, is needed before the degree and significance of ecolog-

ical uniqueness of the Nordhouse dune-pocketed wetlands can be

deter mined." In the course of the studies on the Nordhouse Dunes

System it became apparent that the outer boundary of the abutting

dune-pocketed watershed defines a logical and defensible dune

ecosystem zone. Figure I is a reproduction of the surface

watershed divides recognized in that study.

An advantage of using the watershed boundary for defining the

dune system is that it includes all assoc-iated wetlands. These

are ecologically important subsystems, rich in species and high in

productivity. Numerous plants, birds, fish, amphibians, insects

and mammals thrive in these wet areas. Wetlands also act as water

purifiers by allowing silt, sediment and other impurities to

settle out and by chemically transforming what they receive from

surrounding uplands. In the case of "pocketed wetlands", surface

water contamination anyplace in the watershed may concentrate in
the catchment along the base of the dunesi.

iUnder still air conditions, heavier air contaminants (e.g.
hydrogen sulfide gas) would also settle into the catchments of
these pocketed watersheds, concentrating air borne toxic materials
at the base of the dunes.

d..

SIP
0

Ir;

IVA,

. ..... ....

1!,.Lgure 1. Surface watersheds of the Nordhouse Dunes
Drainage patterns from the eastern watershed
66 vide to the base of the dunes can be seen in uT.154
%'Uadu of lowland vegetation (from Goff et al, 1987)
BD Barrier Dunes (Lake Michigan)
LL Lost Lakes
PC = Porter Creek
DPW = Dune-Pocketed Wetlands
(n) northern unit
(s) southern unit
HL Hamlin Lake

Although both sand dunes and wetlands are protected by re-

spective laws, there has been little systematic examination of the

interaction or developmental relationships between them. Ecologi-

cal and hydrologic relationships between the dunes and wetland

areas lying behind the dunes have hardly been considered. Where

these two important resources, dunes and wetlands, intermingle,

they should be studied and managed as a system, rather than as

isolated features. Wetlands and sand dunes are two of the states

most valued resources, and establishing a link between them gives

these areas.added significance. The present study contributes to

that understanding by examining the formation processes and

physiographic relationship of shoreline drainage systems and by

providing an inventory and a series of maps of landward dune-

pocketed wetlands.

Because of the presence of dunes which have blocked their

flow to the big lake, many of the watersheds along the shoreline

have not developed the deeply cut stream channels typical of

unimpeded streams with watersheds of a similar size. Several

of these undeveloped watersheds remain pocketed behind the dunes.

Other deeply cut drainage systems have also been pocketed, appar-

ently more recently. A hypothesis of this study is that the

history of shoreline drainage systems can be reconstructed by

correlating the pattern and degree of their development with post

glacial lake shoreline chronology, with particular reference to

dune formation. Additionally, we will examine evidence that

suggests that some presently unimpeded stream drainage systems

were formerly blocked or "pocketed" behind dunes that have been

removed by rising lake levels. Specific objectives of the study

include:

1. To describe the processes that have resulted in the

formation of the drainage systems of the eastern Lake

Michigan basin, especially those that are associated with

shoreline dunes.

2. To generally classify, describe, and inventory by type

the drainage systems of the eastern Lake Michigan shoreline.

3. To more specifically map, describe and inventory

all dune-pocketed wetlands of the eastern Lake Michigan

shoreline.

II. METHODS

The study area stretches along the eastern Lake Michigan

shoreline from the Michigan/Indiana border north to the Straits of

Mackinac. The counties included are Berrien, Van Buren, Allegan,

Ottawa, Muskegon, Oceana, Mason, Manistee, Benzie, Leelanau,

Antrim, Charlevoix, and Emmet. Most of the effort was concen-

trated on the area between the Indiana/Michigan state line and

Frankfort in Benzie County. North of this, crustal rebound be-

comes an increasingly dominant factor and the dune-drainage system

linkage is more obscure.

Much of the study was conducted in the laboratory examining

maps and various sources of information in order to identify and

classify wetland areas within a zone along the shore, especially

those behind sand dunes. A representative series of dune-assoc-

iated wetland drainage systems was inspected in the field to

describe the vegetation and collect other relevant information.

Field studies were for the purpose of understanding and defining

the types of shoreline drainage systems and related wetlands,

rather than for comprehensive ecological inventory. Field work

was followed by further map studies, literature review, glacial

and recent geomorphological process reconstructions, and data

analysis. The resulting classification was used to inventory the

shoreline drainage systems based mainly on drainage patterns and

topography.

II.A Initial Map Analysis

Our initial task was to identify all nearshore wetlands and

catchments. A set of USGS topographic maps covering the shoreline

was used to identify lakes, ponds and wetland areas. The first

criterion applied was the presence of marsh or aquatic symbols on

the topographic map within two miles of the shoreline. Some low-

lying areas not mapped as wetlands but shown as unforested, and

judged to be possibly seasonally wet, were also included. The

topographic maps range in age from 1970 to 1983, and are divided

into two scales at Muskegon (1:24,000 south and 1:25,000 north).

They were sufficient for identifying many of the wetland catchment

areas and lakes, but were considered not detailed enough to use as

the sole source of information.

The National Wetlands Inventory, conducted by the USDOI --

Fish and Wildlife Service was consulted, but this survey has

completed only portions of two counties within our study area so

it was not available as a consistent information base. These data

were used where available. The USDA -- Soil Conservation Service

provided detailed soil surveys of nine of the thirteen counties

included in the study area. For the surveyed counties, the loca-

tion of wet and mucky soils was added to the topographic base map.

Hydrologic information was noted. In addition, land cover/use

maps available from the MDNR -- Division of Land and Water Manage-

ment, were consulted to help identify wetlands. In most instances

these proved to be too general to be useful for the purposes of

this study. Drain commissioners in several of the counties were

interviewed by phone and drain maps were obtained for several

southern shoreline areas. A set of nearshore wetlands was identi-

fied using these sources of information.

In order to determine which of the nearshore wetlands were

associated with dunes, we overlaid the designated sand dune and

barrier dune maps available from the MDNR (1976, 1978) onto the

USGS base maps with shoreline wetland catchments and lakes desig-

nated. Some of these wet areas occurred in conjunction with

dunes, but were not included in the designated or barrier dune

areas. To augment the shoreline dune designation, we used the

state-wide dune distribution map (Kelley, 1962). Both the

designated sand dune and barrier dune map series (MDNR 1976, 1978)

were compared with Kelley's map. In addition, the USGS quadrangle

maps were used to identify all areas of high relief along the

shoreline that might be of dune origin. In sandy areas with high

4ills, the Michigan glacial geology map by Farrand (1982) was

consulted to make a general distinction between dune sand and

materials of other origin. This information was compared with the

soil survey reports. It was possible, by studying all of these

sources, to determine practical limits for all shoreline dune

areas. We were liberal in identifying sand dunes, considering

most nearshore sandy areas with high topographic relief as dunes.

We then determined which of the wetland catchment areas initially

identified abut against sand dunes on their lakeward side. Those

that do, we considered dune-associated wetlands.

The next step of the survey was to delineate the watershed

boundaries of all identified dune-associated wetland areas. The

watershed divides provide a more definitive basis for defining the

natural system limits than does simply distance from the shore-

line. The watershed areas were delineated on the USGS topographic

base maps by studying the contour lines.

II.B Aerial Photo Interpretation

Within the watershed boundary we defined the upland, lowland,

and catchment components of the system. These distinctions were

made from aerial photographs using stereoscopic infrared imagery

at the scale of 1:24,000 available from the MDNR. Each watershed

was examined, and, based on the structure of the vegetation and

relief patterns of the land, these three land classes were differ-

entiated. Catchment basins include open water areas, emergent

aquatic beds, marshes, shrub carrs, and other non-forested areas.

In many of the watersheds a distinct change in vegetation struc-

ture could be detected between the catchment and the surrounding

lowland. Lowlands are usually forested areas on relatively flat

topography associated with the catchments. The boundaries of

these were usually indicated quite accurately on the topographic

maps, but they were confirmed by photo interpretation and by field

checking in some instances. Uplands are the portion of the water-

shed from the boundary of the flat lowlands to the watershed

divide.

Initially, the areas of the watersheds and of their upland,

lowland, and catchment components were estimated by copying the

map of each onto paper of consistent weight and then cutting out

the areas. Each piece was weighed on a precise balance and the

areas were calculated using the weight of a known area of the same

paper. Later measurements made use of a dot grid or square grid.

II.C Ownership and Disturbance

The ownership patterns of the sites was also included for

management reference. Using the most recent plat maps available

at the State of Michigan Library a determination was made between

private, state or federal ownership, or a combination. This data

was expressed as a percentage of the watershed area. A planimeter

projecting machine was used to bring the maps to scale, and then

the procedure described above (cutting out the areas and weighing

them or counting a dot grid) was used to calculate acreage by

ownership class.

Information for the estimation of the degree of human devel-

opment within each area was taken from the USGS topographic maps.

Information from field inspection and plat maps was also included.

The scale and criteria used are shown in Table 1.

Table 1. Degree of development scale. Numerical
values were assigned as follows to each dune-pocketed
wetlands watershed to rate disturbance, The
lowland/catchment was combined and received a
disturbance value of L 0 to 10. The upland was
assigned a value of U 0 to 10. Because disturbance
in the lowland/catchment area is considered more
significant, it was given twice the weight in
calculating a weighted mean disturbance for the
watershed as a whole. The watershed disturbance index,
W = [(L x 2) + U1 / 3, was calculated using the scaling
criteria listed below.

Lowland/Catchment Upland 1 Watershed!

'Major Dev. 10 10 W [(L x
!Moderate Dev. 7 7 2) + Ul!
'Minor Dev. 3 3
\ 3
No Apparent Dev. 0 0

Major Development -- Impacts which greatly affect
drainage (e.g. ditches, dams, or extensive
roadways), water quality (e.g. extensive
agriculture, city storm drains), or have great
potential for other adverse impacts (sand mining,
hydrocarbon drilling, sewage treatment plant)

Moderate Development -- Impacts which have less
@hysiographic or water quality effects, but
increase traffic and detract from aesthetics
(e.g. county or state roads, railroads, small
towns, campgrounds, etc.)

Minor Development -- Small-scale impacts which
have a minor effect on the ecological quality of
the area, but may increase traffic or detract
from aesthetics (e.g. individual or few houses,
trail roads, powerlines, etc.)

II.D Field Inspections

Many of the map and aerial photo procedures were undertaken

concurrently with the field inspections. As soon as the domain of

dune-associated wetlands was identified, an initial set of candi-

date areas for field examination was selected. Field sites were

chosen to represent both the dune-pocketed wetlands and other

shoreline wetland types that we wished to differentiate from dune-

pocketed wetlands. Inspections began in the areas close to the

Nordhouse Dunes. These were reconnaissance level examinations to

assist in the development of the classification and inventory.

Detailed studies of the sites were not possible within the time

constraints of this project. We limited reconnaissance site

examinations to a central portion of the shoreline'from Bar Lake

Swamp on the north to the Fruitvale Dunes-area on the south be-

cause this zone appeared to be representative and appeared to

contain the greatest diversity of shoreline wetland drainage

system types. At each site we made systematic observations at one

of three levels of intensity.

Level I: General Scouting -- The purpose of this level of

observation was to gain general impressions of the wetland water-

shed areas by driving perimeter roads or hiking through each area

scouted. Notes on drainage, topography, soils, cover types and

development were made and referenced to the USGS topographic maps.

Significant features were documented by narrative field notes and

photography.

Level II: Specific Site Description -- The purpose of this

level of observation was to locate a predetermined candidate site

of interest (e.g. a catchment basin, drainage divide, upgradient

wetland, etc.) and describe its characteristics. Photos were

taken in most instances. A narrative description was written to

summarize the area's significance. Vegetation cover type was

noted and dominant species were recorded. Initially, based on the

hypothesis that certain variables would prove to be important in

distinguishing and understanding these wetlands, each was ranked

subjectively on three scales using vegetation community composi-

tion and other indicators: (A) a mineral enrichment scale [0

ombrotrophic to 10 = m inerotrophic], (B) a closure/succession

scale [0 = open water to 10 = lowland forest], and (C) a surface

water variability scale [0 = constant seasonal water level to 10

highly variable water level between early spring and late summer).

The basis of these scales is described more fully in the "Wetland

Vegetational Processes" section of this report (Section IV).

Level III -- Vegetation/-Species Survey -- This level of

observation was made for certain sites that seemed to typify the

wetland drainage type classes. It consisted of a Level II obser-

vation plus a species search within the general location of the

identified site. Rather than keying unknown species on site, they

were collected, labeled and later pressed and identified. Vege-

tation summaries of several of the representative sites examined

in the field are presented in Appendix 1.

II.E Extended and Refined Map-Analysis

We found it necessary to map all watershed divides within the

shoreline zone in order to adequately delineate and understand the

limits of the dune-pocketed wetlands watersheds. From the Michi-

gan/Indiana state line northward to Portage Lake in Manistee

County, every drainage area and watershed along the shoreline was

delineated, assigned a number, and named. Because watersheds and

drainage basins are naturally hierarchical and vary in scale, our

ability to recognize them was dependent on the degree of detail

presented on the topographic maps. In this sense,'the entire

study is scale dependent, but the concepts developed can be ap-

plied at a more specific level as more refined topographic infor-

mation becomes available. For the purposes of a st'atewide inven-

tory, the USGS quadrangle sheets provide an adequate level of

detail. In areas having relatively high relief the boundary

designation was straightforward. In flatter areas, there was

often a degree of uncertainty as to where the divide line should

be placed due to the width of the contour interval (generally 10

feet or 3 meters). Delineation of the boundary in these areas was

more difficult and often somewhat arbitrary, especially where

artificial drainage has altered the landscape. Many of the mapped

watershed boundaries extended inland well beyond the two mile

limit initially established. We mapped them back from the shore-

line a sufficient distance to assure that every point on the

landscape had been included in one of the watersheds or drainage

areas. From the topographic maps and other map sources, each

section of the shoreline was ranked on the basis of the extent of

dunes (0 = no evidence of dunes; s = slight dune development; m

moderate dune development; h = highly developed dunes].

II.F Classification of Shoreline Drainage Systems

All drainage systems associated with shoreline dunes were

grouped into natural classes on the basis,of the process of their

origin. This was a procedure of reflective dialogue -- making

educated speculations about the process of formation of each

drainage basin, particularly its relationship to dunes, and then

determining its membership within one of the classe s that we had

hypothesized. During this stage of our work, we repeatedly

re-examined the topographic maps with watersheds designated,

consulted the field data, and reviewed literature sources. If one

of the drainage systems did not seem to fit the classification, we

adjusted our concept of the natural classes and worked back.

through the entire set of mapped watersheds. We went through

several iterations of this process, finally arriving at a neces-

sary and sufficient natural classification of all shoreline dune

associated drainage systems.

IIA Screenina, Ordination, and Classification of Dune-Pocketed

Wetlands Drainage Systems

The criteria for each shoreline drainage system class, based

on the process of origin of its members, permitted us to screen

out all but the dune-pocketed wetlands drainage basin class.

Eighteen dune-pocketed wetlands watershed systems were recog-

nized. Each example of this class was characterized in terms of

size, extent of upland, lowland, and catchment, topographic re-

lief, ownership, and disturbance.

A simple watershed area/topographic relief ordination was

prepared and patterns of distribution of the dune-p ocketed water-

sheds were interpreted. Finally a classification of dune-pocketed

watersheds was developed by delineating natural clusters within

the ordination having similar drainage pattern features. A gen-

eral hypothesis regarding the development process separating the

major types is presented. This is based on literature review and

reconstruction of likely historical geological sequences employing

topographic data and hydrologic theory.

IIA Preparation of Reference Maps

Using the USGS topographic sheets as a base, a set of drain-

age basin maps was prepared. From the Indiana/Michigan State Line

northward to Frankfort (Benzie County) these covered all streams

draining to Lake Michigan as well as shoreline direct drainage

areas. All dune-pocketed wetlands watersheds were included. In

the area north of Frankfort, where more recent glaciation and

crustal rebound has altered drainage patterns, we listed only the

first order stream watersheds. Characterizations of the systems

containing dune-pocketed wetlands is given in Section V, and the

listing of all watersheds and drainage areas is given in Appendix

II. T he maps of dune-pocketed wetlands drainage systems are

presented as Appendix II in a form suitable for digitizing and

entry in the Michigan Resource Information System.

III. EASTERN LAKE MICHIGAN SHORELINE LANDSCAPE FORMATION PROCESSES

Four principle agents have altered the surface geologic

landscape of Michigan: 1) glacial ice; 2) running water; 3) waves

and shore currents; and 4) wind. Within the last 15,000 years,

each of those forces has acted on the present shoreline area of

eastern Lake Michigan to produce an interesting and unique coastal

landscape. This report presents a classification of the dune-

associated drainage systems of that shoreline. It also identi-

fies, analyses, and inventories a new landscape type -- the dune-

pocketed wetlands watershed type. In order to understand these

systems, it is essential to consider the processes that formed

them. A large body of literature has developed since the late

1800's on the history of Michigan's landforms and the processes of

their development. The following is a brief description of how

the four primary agents of landscape development have effected the

eastern Lake Michigan shoreline, with specific regard to develop-

iment of dune-pocketed wetlands drainage systems. We also describe

some of the vegetational processes that are important to the

formation of plant communities within dune-associated wetlands.

III.A Glacial Processes

The work of continental glaciation upon North America began

approximately a million years ago during the Pleistocene epoch.

The ice age was brought about by a radical change in world cli-

mate, the cause of which is still open to speculation. During

this time there were several glacial advances and retreats over

the Great Lakes region. At the peak of each advance, the ice was

as much as two to three miles thick. The tremendous weight of

this ice mass caused the earth 's crust to be temporarily pressed

downward. As the ice melted back, the crust slowly rebounded.

As a glacier advanced, it cut deep basins in the areas where

the bedrock was weakest, and deposited the eroded material on top

of the stronger areas as it melted back. This process of repeated

erosion and deposition scooped out the Great Lakes basins and

formed the undulating upland landscape of Michigan. The present

character of the Lake Michigan shoreline began to be formed with

the last major glacial advance. This glacier separated the Lake

Michigan basin from the other Great Lakes and forced drainage to

the south through the Chicago River. The northwestern section of

-the lower peninsula was considerably altered by this advance. The

ice cut many deep north-south trending lakes and bays (Torch Lake,

Elk Lake, Traverse Bay) and also left many kettle lakes throughout

the area. The effects of glaciation on drainage are very pro-

nounced in this region. The area was also more effected by crust-

al rebound. Therefore one would not expect to find as many dune-

pocketed drainage systems along the northern shoreline as south of

Manistee. This last major advance of glacial ice began to retreat

from the Lake Michigan basin approximately 14,000 years ago.

With the retreat of each glacial advance, meltwater filled

the basin between the ice and the material the glacier had depos-

ited. Each glacial period had pro-glacial Great Lakes whose

shorelines, depths, and drainage outlets were very different from

those of the present. The historical sequence of glaciation and

the subsequent lakes has been interpreted and re-interpreted by

various authors since the late 1800's. Until very recently these

interp retations were based on a rigid earth crust concept which

postulated a "hinge-line" effecting the pro-glacial lake sequence.

A modification of the theory of crustal rebound published recently

by Larsen (1987), based on the earth's crust responding as a

visco-elastic crust to ice loading, invalidated the hinge-line

notion and prompted a significant adjustment in chr onology of the

Lake Michigan Basin. Larsen's interpretation is adequate to

explain salient aspects of drainage development. Accordingly, the

following synopsis is provided from his recent paper:

"...Separate proglacial lakes draining southward
through the Chicago and Port Huron outlets occupied the
Michigan and Huron basins, respectively. Lake Chicago
stood at its Calumet level of 189 m (620') while Early
Lake Algonquin filled the southern Huron basin, possi-
bly at a 184.5 m (605') level (Figure 2a -- reproduc-
tion of Larsen's figure 15]. Ice retreat into a pro-
gressively deepening isostatic depression caused
deglaciation of successively lower northern outlets,
which in turn caused the lakes to fall..."

"Deglaciation of the Lake Michigan basin was
accompanied by northward expansion of Calumet-level
Lake Chicago until drainage channels to the Huron basin
were uncovered north of Traverse City, Mich. Drainage
through the Indian River lowland possibly linked the
two lake basins briefly, joining the Calumet level of
Lake Chicago with the Main Algonquin of Huron [Figure
2B -- reproduction of Larsen's figure 16)."

21 -

zi -------

Mal.

Like AO-quiti 1j,

LaM

Fliw,

0 '00 chicg. oull.,
ch"go outlet i I
a too 200 MOMETERS
W66 '__W
Ntij& TU.=auffiW_Teve1 a i Cfticitgo aiO iFe Main n4vin
Figure 15. Calumal-leval LaIre Chicago and Early Lake AJgonquin, Overflow was to the south at Chicago and Poll Huton Overflow was to the south of Chicago and eastwiud to Early Lake
atthe maximum advance of Two Rivers ice about 11.Myrfil.P. Early Lake Erie drained to the Lake Ontario basin. were possibly linked by drainage through the Indian River lowland sou
Lake Ontario basin.

Figure 2. Early Pro-glacial lakes of the Lake Michigan basin.
A. Early Lake Chicago
B. Later Lake Chicago
Notice that Lake Algonquin did not occupy the Lake Michigan
during this time according to the chronology presented by L
figure is reproduced from Larsen, 1987 -- figure numbers 15
Larsen's numbers).

"Deglaciation of the Mackinac Straits at about
11,200 yr B.P. joined the basins at depth. The Main
Algonquin of Michigan dates from this period and repre-
sents a single water plane draining to the north
through the Fossmill system and probably controlled by
the Mink Lake sill at 328 m. This isostatically de-
pressed outlet channel caused the level of Lake Michi-
gan to fall to a low of about 61 m (200') in the south-
ern basin... (Figure 3A - reproduction of Larsen's
figure 17]."

"The lakes underwent their final fall when ice
receded from North Bay. The level of Georgian Bay
dropped to as low as 62m .... The Lake Michigan level, in
turn, fell to a plateau near 160m at the Mackinac
Straits and subsequently eroded the Mackinac River
channel to a controlling threshold now at about 140m.
This spillway governed a lowered level of the lake,
creating a separate lake in the south at an altitude of
about 56 m (180') that overflowed northward through a
channel near Muskegon (Hough, 1958, p241). The slow
rise from the early Chippewa low level in the southern
basin was controlled by uplift of the Mackinac
Straits..."

"The low-level phases of Lake Michigan... (Figure
4, bottom left, from Larsen's table 4) were progres-
sively ended by uplift of the controlling spillway at
North Bay .... A transgression, led by the expanding lake
in the Huron basin and caused by the steadily rising
North Bay outlet, rejoined the Michigan and Huron
basins at the Mackinac Straits after 8,150 yr B.P. The
episode marked the end of separate low-level phases in
the lakes and initiated the pre-Nipissing trans-
gression."

"Once drainage to the south was established, the
outlet configurations and probable paleoclimatic influ-
ences on water volume became major variables in the
system. Lake level fluctuated about a mean altitude
adjusted to the cross sections of the southern outlets.
Before 4,500 yr B.P., the record of lake-level change
is poorly known; however, distinct high levels occurred
at 4,500, 4,000, and 3,200 yr B.P. (Larsen 1985a,b).
These highs, referred to as the Nipissing I, Nipissing
II, and Algoma levels are identified by terraces that
rise exponentially with distance from the southern
shores of Lake Michigan to the North Bay region
(Larsen, 1985b). They reflect probable high-amplitude
fluctuations related to runoff variations in the drain-
age basins. The permanent outlet channel linking North
Bay with the Mattawa River maintained a level adjusted
to its rising sill. Isostatic uplift finally raised
the North Bay sill above the southern outlets between

23 -

Lk. S.P.d. b-in 7 A= I

T in
Channeft
Mink We
41%@
fleta awo,
RZI Ftive,

@Io

t, S1 Rwa,
_j

Niso.'a ftiv.,

N
0 100 KES

2W) @Iownft -J
Figure 17. Main Algonquin level of Lake Michigan confluerd wiIh the Ofillia (1) level of the Lake Huton basin (AA,20G to Figure 18. Chippewa and Stanley low levels. Marquette Michi
I i,000 yr B.P.). Overflow was eastward to the Ottawa valley at the Mink Lake sill. Lake Eris anti Lake Ontario basins ice filled the Lake Superio rbasin and supplied overflow chant
contained low lakes controlled by rising northeastern outlets (fig. 5, case 3), which show rising water levels along their through the Whiiefish-Au Train channels (9.900 yr B.P.). Huro
western shores.
The upper Great Lakes drained eastward to the Ottawa into I
valley across a controlling SRI at North Bay final was Lake
cleglacialed as early as 10,300 yi B.P. Chippewa4evel Lake their

Figure 3. Main Algonquin of Michigan and Lake Chippewa pro-glacial la
Lake Michigan basin.
A. Main Algonquin of Michigan (MAM)
B. Lake Chippewa
This figure is reproduced from Larsen, 1987 -- figure numbe
18 are Larsen's numbers.

4,500 and 4,000 yr B.P. The Chicago outlet was aban-
doned after 4,000 yr B.P., and the modern drainage
system at Port Huron came into being. Since then, lake
levels have continued to fluctuate adjusted to a single
outlet channel."

The lowest stage, termed Lake Chippewa, has been recognized

for several decades. It occurred just after the the North Bay

outlet opened but before isostatic rebound had raised the level of

the northern outlets. This extremely low lake stage (some 400'

below.present lake level) had profound effects upon drainage in

the Lake Michigan basin. These effects will be described more

fully in the next section on drainage processes.

Crustal rebound is an important factor in the development of

shoreline landscapes. The uplift of each lake stages' shoreline

features (beaches and dunes) becomes progressively (exponentially)

more pronounced to the north. This effect dominates drainage

patterns north of approximately Portage Lake in Benzie County, but

the entire shoreline has been uplifted. Drainage patterns between

approximately Ludington (Mason County) and Portage Lake (Benzie

County) show some noticeable effects of isostatic rebound and this

area way be considered a transitional zone between the minimally

effected areas to the south and the greatly effected northern

areas.

III.B Surface Drainage Processes

The erosive action of surface water runoff on the Michigan

landscape has only been operating since the last glacier retreated

from the land, about 14,000 years ago. That is a short time in

terms of most geologic processes, and the Great Lakes drainage

system is poorly developed, especially in comparison to older

systems such as the Mississippi River. The ultimate result of

surface water drainage over a very long time is a reduction of the

upland to a featureless plain at base level. Base level for the

Great Lakes drainage system is sea level, but for the landscapes

in western lower Michigan, base level is 580', the present level

of Lake Michigan.

The erosive energy of a surface drainage system is approxi-

mated by stream discharge (cross-sectional area per unit time).

Very generally, a stream's discharge is determined by the size of

its surface watershed and the gradient within the watershed.

Watershed area approximates total water volume collected, and

gradient approximates velocity due to gravity. Other factors such

as loss of volume to groundwater and evapotranspiration are also

important.

Drainage systems develop by a process called stream grada-

tion. Generally, the erosive energy in the stream works to smooth

out the gradient to base level, and as it does it erodes upgrad-

ient. This erosion may "capture" a nearby stream or lake, enlarg-

ing the watershed. Further enlargement and steepening of the

gradient as the watershed grows increases erosive energy, contin-

uing the cycle. Ultimately, the result of the expansion process

would be only a few very large rivers draining to base level.

Within the Lake Michigan basin, drainage systems are young and

their development has been complicated by changes in base level.

As the glacier first uncovered the landscape, most well-

defined stream watersheds were probably small, and the surface

gradient was not much steeper than it is today. Early drainage

into Lake Chicago was likely quite slow in development. However,

the opening of the North Bay drainage outlet about 4,000 years

later must have had a dramatic effect on stream gradation as base

lake levels dropped nearly 400' to the Lake Chippewa stage. This

effect was probably most pronounced along deeper nearshore areas

where the position of the shoreline did not shift very far as lake

levels dropped. Figure 4A, simplistically depicts the "Chippewa

drawdown" in relation to Larsen's chronology, and shows the area

occupied by Lake Chippewa at its minimum (Figure 4B -- from Hough,

1958). This period probably accounts for much of the present

upland drainage development. As the gradient steepened, the

larger streams cut deep gorges and their drainage systems rapidly

extended inland, except where sand dunes blocked drainage. Many

of the streams that had been small, low-energy nearshore creeks

draining into Lake Chicago became significant drainage systems

with deeply eroded channels. There were probably many rapids and

PRE-CHIPPEWA DUNKS POST-CHIPPEWA DUNES

CHIPPEWA DRAWDOWN
r
W W
C,

Lake Chicago chivP4.4 L" ft.s.

p
Aw

L
C

Pot-

Co 10
SOUTHERN
LAKE
a PEINA
to-J.T L- Ph."

F-u t""

Figure 4. The "Chippewa Drawdown" (Figure 4A) from the Calumet level of L
through the Main Algonquin of Michigan, to a minimum level during Lake Chi
affected drainage patterns along the Lake Michigan shoreline. From the vi
relationship to drainage development, dunes can be broadly distinguished E
or post-Chippewa. Figure 4B shows the approximate shoreline location at n
QKEE

level in Lake Chippewa (based on modern lake bottom contours), and the sho
exposed at that time (Map from Hough, 1958).

waterfalls during this period as streams flowed over unconsol-

idated glacial material and possibly bedrock in some areas.

Including the falling, low, and rising stages, the Chippewa

drawdown probably lasted at least 5,000 years -- a relatively long

duration among the post-glacial lakes. During this period of

base-level lowering, upland drainage was accelerated, and deep

river channels were cut. Many inland lakes and wetlands were

probably drained by expanding stream watershed boundaries. Even

after the base lake level began to rise, upgradient erosion prob-

ably continued at nearly the same rate for some time because of

the gradient already established. When the base lake level reach-

ed 605' (Lake Nipissing), the deep river gorges were flooded well

inland with lake water. Slowly they were filled with sediment

eroded from the upland. Lake Nipissing levels were reached about

6,000 years ago and probably represented a slowdown in the devel-

opment of upland drainage. Even though the total gradient of the

Lake Michigan basin watershed was similar to that established

earlier during Lake Chicago drainage, the pattern of drainage was

different. Upland channels had been cut during the Chippewa

drawdown. When Lake Nipissing dropped to the present level of

580'. there was probably much less effect on the upland drainage

patterns. The major rivers cut down through their bed deposits

and are now meandering within the width of these alluvium filled

channels.

III. C Shoreline Processes

While the erosive power of stream gradation constantly works

to level the land, dune building and littoral drift have acted as

depositional processes to move materials landward along the mar-

gins of the coast. Active forces that drive these processes

include direct wind energy, wave action, and lake currents. All

three of these forces simultaneously erode and deposit materials.

The character of the resulting shoreline depends upon the balance

between erosion and deposition.

The force of the direct winds that strike the eastern shore-

line is undiminished as they blow across the expanse of Lake

Michigan. In addition to their direct abrasive and erosive ac-

tion, these winds carry sand inland from Where it is exposed along

the shoreline, depositing it as they lose energy. They also

generate waves that both erode the shoreline and carry material

shoreward. The steady force of the wind produces currents within

the lake that move large amounts of material along the shoreline.

III.C.1 Dune Building -- The high sand dunes along the

eastern and southern shores of Lake Michigan are not associated

with the present stable lake level. A relatively constant lake

level is not conducive to sand dune formation because sand that is

piled on the beach is still within the reach of waves, either

during periods of high water or during storms. The waves wash

away the dunes before they can become large and stabilized.

Buckler (1979) emphasized that, "The environmental (climatic and

geomorphic) conditions under which the dunes formed no longer

exist; once destroyed, they are not likely ever to regain their

present significant size and extent."

The important event that allows for the development of the

high dunes is the lowering of lake level. When the lake level is

lowered, waves cannot reach the forming dunes and erode them.

Also, more sand is exposed to the wind when water level is drop-

ped, providing the material for even higher dunes. Vegetation

soon begins to grow on the dunes, stabilizing them from further

wind erosion.

Most of the high dunes along the present shoreline were

formed either formed during the drawdown*of Lake Chicago

(eventually to Lake Chippewa level), or as Lake Nipissing was

being lowered to the present lake level. Ignoring minor

variations in lake level that have produced a complicated series

of beach terraces and subsequent corresponding dunes, we have

labeled these major dune building periods as "pre-Chippewa" and

"post-Chippewa."

Pre-Chippewa dune development during the Chippewa drawdown

period may have been very extensive. However, because the level

of Lake Chicago and Lake Nipissing was approximately the same,

erosional washing by the rising waters of Lake Nipissing leveled

any dunes below about 605'. What remains of pre-Chippewa

dunes is probably only a small fraction of those originally

formed. Large beds of lake bottom sand apparently occurred

beneath Lake Nipissing. Some of the same sand that formed pre-

Chippewa dunes during the drawdown to Lake Chippewa in locations

presently offshore was probably redeposited later as part of the

Nipissing and Algoma dunes. In most cases, the post-Chippewa

dunes formed over top of, or at least adjacent to the remaining,

most landward (oldest) pre-Chippewa dunes. These two sets of

dunes essentially merged together into one large, dense dune

barrier. However, due to the greater uplift of the land to the

north, the pre-Chippewa dunes have been raised more, and the post-

Chippewa dunes have formed on the old lake bottom somewhat

lakeward and below the pre-Chippewa dunes.

Generally, the southern dune systems'are more compact sand

barriers, whereas the northern systems are more spread out both

vertically and horizontally, and therefore less effective as

barriers to drainage.

The Sleeping Bear Dunes of Leelanau County are a good example

of a northern dune. Although the Great Sleeping Bear Dunes form

an impressive feature that rises high above the shoreline, uplift

of the underlying glacial deposits is responsible for much of the

topographic relief. High dunes formed on glacial deposits are

called perched dunes.

III.C.2 Littoral Drift -- Another important shoreline pro-

cess of the Great Lakes is that of longshore currents and beach

drift. During periods of beach erosion (storms or high water),

sand is moved laterally along the beach by wave action or wind.

When it reaches an area of decreased wave energy, such as a bay

mouth or protruding shore feature, the lateral movement is de-

creased and sand deposition begins. This process can accumulate

large amounts of sand in some areas, forming hooked spits, bays,

or even closing off a lake. During the present period of rather

stable lake levels and general wave erosion, longshore currents

account for the majority of sand deposition. There are two forms

of littoral drift: beach drift and longshore drift. Beach drift

is caused by waves striking the beach at an oblique angle, lifting

sand off the beach and depositing it a few feet up the beach.

Longshore drift is a more dynamic process,of currents that follow

the shoreline. The prevailing winds cause the waters of Lake

Michigan to flow in counterclockwise currents. Thus, along the

eastern shore the water current is from south to north. These

currents are generally quite strong, and they carry large amounts

of sand with them. However, in areas where the current is weak-

ened, such as behind a jutting point or across the mouth of a bay,

the sand tends to be deposited. Reverse eddy currents are set up

in areas behind these deposits which eventually establish opposing

north to south deposits.

III.D VeAetational Processes

III.D.1 Species MiAration and Colonization -- Within the

frame of reference of geologic time, an important process effect-

ing vegetation is species migration. The glaciers destroyed all

vegetation in the area covered by the ice. The once circumboreal

arctotertiary geoflora had earlier been divided into continental

components (North American, European, Asiatic), but glaciation

further segmented it into subcontinental isolates (Eastern North

American, Western North American). Many of the species popula-

tions in these areas are similar but have undergone a degree of

genetic evolution and now occur as species or subspecies pairs

[e.g. eastern white pine (Pinus strobus) and western white pine

(Finus monticola)]. During the peak of the glacial epoch, plant

populations survived in "refugia" south o-f the ice, particularly

the Appalachian mountains. Some boreal species persisted along

the ice front (as evidenced, for example, in the buried "Two

Creeks" forest bed and in preserved trees beneath Lake Michigan).

Between advances of the glacier these populations evidently

migrated northward repeatedly, colonizing the exposed landscape.

Migration establishes the geographical range of species and thus

regulates the populations available to form a plant community.

Species that are available within the dune associated wet-

lands zone in northern Michigan are mostly similar to those found

in wetland communities throughout the lower peninsula. Many

species (e.g. Sassafras albidum, Nyssa sylvatica) extend further

northward along the Lake Michigan shoreline than they do further

inland within the state at the same latitude. This has been

presumed to be due to the ameliorating effect of Lake Michigan on

climate within the shoreline zone, and that no doubt is an impor-

tant factor. However, patterns of colonization and migration are

also important. These are related directly to prehistoric habi-

tats that were available as source areas for species populations.

it is noteworthy that during the Lake Chippewa low water

period extensive wetland and dune areas probably occurred in the

area presently covered by Lake Michigan. These exposed lands may

have been important in the development of the flora of the dunes

and eastern Lake Michigan shoreline areas. During'this stage,

water levels dropped by approximately 400'. One need only to con-

sider the extensive areas laid bare (see Figure 4B) in comparison

to recent dune building on exposed areas, to gain some appeciation

for vegetation colonization that might have existed during this

time. For example, the Great Sleeping Bear Dunes area has been

uplifted approximately 40 feet, equivalent to a drop in water

level only about one-tenth as great as that experienced during the

Chippewa drawdown. Yet extensive dune building has occurred.

Large areas of beach, dune, and shoreline habitat have developed.

Approximately three million acres of this type of habitat was

probably opened during the Chippewa period. Many of the species

presently found in the shoreline zone probably colonized this vast

area, persisting along the narrowing shoreline as the lake rose,

and then re-expanding their range as the lake fell from Nipissing

level (605') to its present level (580').

III.D.2 Succession -- Another process affecting vegetation

composition in the dune associated wetlands is plant succession.

Succession was first described in relation to dune communities.

On the dunes a complete series of vegetation communities is laid

out spatially from the earliest colonizing species and species

combinations along the shoreline and the foredunes, to stable

climax forests farther inland. Succession has also effected the

dune associated wetlands from their beginning until the present

time. Much of the information that has developed during the past

several decades regarding hydrosere succession has recently been

summarized by Reuter (1986). The model developed by Reuter de-

picts a successional sequence similar to that occurring in many of

the dune associated wetland areas along the Lake Michigan shore-

line, particularly minerotrophic sites with a relatively high,

stable watertable. We have employed this model as the basis for a

successional development scale (Figure 5, upper portion). This

scale ranges from 0, representing emergent aquatic vegetation, to

10, representing lowland forest. We term values on this scale

"wetland succession values".

As depicted in Figure 5, at each stage along the successional

spectrum disturbance factors also come into play. The effect of

catastrophic disturbance on wetland succession has been well

documented by a number of authors. Flooding and fire are the two

0.0 1 0 2f 0 -4.0 4.0 5.0 6.0 7.0 8.0 910 10.0
WETLANDS SUddESSIONAL DEVELOPMENT

Bedford at al.
severe Flooding
moderate F

Primary Sedge Meadow Shrub Carr Lowland
Hydrosere Habitat Habitat Forest
Succession Habitat
(Read* & Cattails) (Sedge Invasion) (Shrub invasion) (Tree invasion)
raised stable Novitaki -T Whitis
water water Lj:::t e
regime regime Curtis Moderate Fir
Linde Vogl
Severe Fire Aspen, Nettl*&
L lowered Vogl & Ragweed
water
regime

Figure 5. Wetlands succession model (lower portion)
from Reuter (1986). This model shows typical plant
community succession and disturbance of high water
table, mineral rich areas. Authors names associated
with reports of disturbance phenomena are indicated
on the lower arrows. Wetland succession value (WSV)
scale is shown in the upper portion of the figure.

37 -

natural factors having the greatest impact in retarding the suc-

cessional process or setting a community back to an earlier stage

of succession. Encroaching sand is also important in some areas

close to the dunes. There is evidence that both flooding (e.g. by

beaver works) and fire (Indian burning) were much more widespread

and significant in presettlement time. Plant community composi-

tion represents an equilibrium between succession and catastrophic

disturbance.

III.D.3 Mineral Enrichment -- Yet another factor affecting

succession and the character of the vegetational community, is the

degree of mineral enrichment (minerotrophy or ombrotrophy).

Apparently most wetland communities in northern Michigan, includ-

ing the dune-associated wetlands, began as minerotrophic commun-

ities. Wetlands usually occupy topographically lower positions on

the landscape and therefore receive enrichment from the surround-

ing upland areas, either by surface runoff or through groundwater

movement. For wetlands connected to groundwater the mineral

regime of the wetland is related to the natural fertility of the

surrounding landscape. Depending on the mineral characteristics

of the surrounding watershed and the position of the wetland basin

within the watershed, many wetlands, especially in sandy, natural-

ly infertile areas located near the head of their watershed where

little groundwater is possible, become increasingly acidic and

lacking in the nutrients needed for organic matter decomposition.

Eventually, they accumulate organic matter and develop a seal,

'A
- 38

process, we assigned a mineral enrichment value (0-10) to each of

the sites examined (See Methods Section). Application of the

succession model presented earlier (Figure 5) must take into

account the mineral characteristics of the site. Mineral

enrichment might be considered as an additional dimension of this

model.

III.D.4 Seasonal Fluctuation of Water Level -- The most

important single factor affecting many of the dune-pocketed wet-

lands catchment areas appears to be fluctuations in water level

between early spring and late summer. Seasonally fluctuating

water levels may be considered as a form of catastrophic distur-

bance. In this sense they could be thought of as a retrogression-

al factor in the model presented above. However, 'whether an

environmental effect is considered a disturbance or simply as a

component of the normal environment depends upon its frequency and

predictability. In the catchments of landward dune-pocketed

wetlands periodic flooding and drying is apparently the rule

rather than the exception. Many of these wetland areas undergo

extreme variation in water levels between spring and late summer.

Because of their relatively large watershed area and comparatively

small catchment area, it is not uncommon for catchment basins to

be flooded to a depth of three to four feet in early spring. This

kills most trees and other woody vegetation except those few

species that are able to tolerate flooding. In some basins a

small open water catchment pond or lake may persist throughout the

summer due to connection to the ground water. Otherwise, the

40 -

isolating them from groundwater. This process, termed ombrotro-

phication, is a positive feedback process. The developing organic

seal further isolates the wetland from mineral input from ground-

water which further reduces the rate of organic decomposition,

increasing peat accumulation, etc. In contrast to the classic

concept of bog vegetation formation by a floating mat in an open

bog lake, most of the peatland areas associated with the shoreline

zone have developed by paludification -- the outward growth of

peat resulting from ombrotrophic conditions within the wetland.

For this reason we prefer the term "peatland" to the term "bog" to

characterize these communities.

Once wetland communities begin to shift over from marsh to

peatland vegetation they seldom revert to marshes. Rare instances

can be found of reversal of ombrotrophication where highways have

been built, agricultural fertilization of the surrounding land-

scape has occurred, or where fire has resulted in a sudden and

massive influx of cations into the wetland, but it appears to be a

one-way process under most circumstances. Species typical of

minerotrophic, marsh-like wetlands include cattail (Typha lati-

folia), Scirpus atrovirens, etc., many willows (Salix spp., etc.),

whereas those more typical of ombrotrophic conditions include

Carex oligosperma, Scirpus cyperinus, leatherleaf (Chamaedaphne

calyculata), chokeberry (Nemopanthus murconata), etc. With some

practice it is possible to estimate the degree of ground water

influence (minerotrophication) of a wetland community by examining

the species composition of the vegetation community. Using this

water level in these flooded areas gradually drops as the season

progresses and by late summer these sandy areas may become very

dry. The extreme drying experienced by the plant community fur-

ther limits the complement of species persisting on the site.

We devised and applied a water constancy index to each site

(see Methods Section), but found that it was meaningful only in

the ca tchment areas. Even there scaling of water fluctuation is

complicated by zonation of the vegetation due to overall water

level as well as the extent of variation. The most extremely

fluctuating portions of the landward dune-pocketed wetlands catch-

ment areas remain denuded of vegetation throughout the year.

Surrounding these there is usually a zone dominated by amphibious

perennial plants (e.g. Polygonum amphibium, Proserpinaca palus-

tris) and weedy annuals (e.g. Conyza canadensis). Continuing

outward toward the margins of the catchment basin one typically

finds a sedge meadow zone, surrounded by shrubs (particularly

Salix petiolaris). Sedge meadow, shrub carr, and open lowland

forest form extensive wetland areas upgradient from the wetland

catchments in some of the dune-pocketed wetlands watersheds.

IV. CLASSIFICATION OF EASTERN LAKE MICHIGAN DRAINAGE SYSTEMS

The formation of post-glacial sand dunes of both pre-Chippewa

and post-Chippewa age had a pronounced effect on the development

of drainage throughout the eastern part of the Lake Michigan

watershed. The glacial process established the framework for

drainage patterns and also altered those patterns by fluctuation

of base lake levels and rebound of the earth's crust. Dunes

impeded drainage in many areas. The interaction of these pro-

cesses formed the present shoreline of eastern Lake Michigan.

This section will describe and classify the shoreline drainage

systems in terms of their process of formation.

The purpose of this classification scheme is to place dune-

pocketed wetlands in proper order with pr-eviously recognized

shoreline drainage systems. We have defined the dune-pocketed

wetlands drainage system type as an ancient stream watershed that

previously drained to Lake Michigan but has since been blocked by

the formation of sand dunes. Therefore, the drainage system

classification is generally based on the degree of sand dune

impedance of all Lake Michigan drainage areas.

The Lake Michigan watershed is subdivided into a set of

stream watersheds. Each stream that enters at the Lake Michigan

shoreline is considered a first-order stream and it drains an

entire upgradient watershed. Within that watershed, there may be

subordinate streams that drain into the first-order stream. Also,

42 -

there may be areas that do not drain by surface flow to any stream

in the watershed, such as glacial kettle lakes. If areas are

within a first-order watershed divide they are considered part of

the first-order stream watershed, and generally drain to that

stream through groundwater flow. Figure 6 is a map that shows the

larger first-order stream watersheds within Michigan's lower

peninsula.

In classifying the shoreline, the area north of Manistee has

shoreline drainage patterns that are much less well-defined than

the southern region. The northern area was affected greatly by

the last glaciation, as well as by uplift of the land. In the

cross-hatched area in Figure 6, north of Frankfort (Crystal Lake)

the drainage systems are so poorly defined, and the effect of

dunes on drainage is so slight, that we d-id not attempt to map all

of the watershed divides in that area. Repeated examinations

found no dune-pocketed wetlands in the northern region, with one

possible exception. The Billiau Hill area in Wilderness State

Park, Emmet County (Bliss Quadrangle) was originally suspected as

a possible dune-pocketed wetland, and has some similar

characteristics, but the process of formation (considering uplift

of the land) is decidedly different from the southern set, and it

is not included in our analysis. The boundaries of all drainage

systems (watersheds and drainage areas) along the shoreline from

the Indiana border to the Betsie River (Benzie County) were

mapped. All are classified and cited in Appendix II. Listings of

all dune-related types are given below.

43 -

Figure 6. Watersheds
and drainage areas of
western lower Michi-
gan. There are five
major watersheds that
extend to the central
divide of the state,
fifteen intermediate
watersheds south of
Manistee, and numerous
small watersheds and
drainage areas in the
shoreline zone (shaded
black). The northern
area (cross-hatched)
has been greatly af-
fected by recent glac-
iation and isostatic
rebound. Drainage
patterns in this area
are quite different
from those farther
south.

The area shaded black in Figure 6 is the shoreline drain

zone. Within this zone, there are numerous small first-order

streams. Some of the creeks only flow a few hundred yards inland

and may be intermittent, but they have developed definite stream

channels. There are also shoreline drainage areas that do not

drain to a first-order stream. These are shoreline drainage types

where surface water enters Lake Michigan either by overland sheet

flow or by groundwater seepage. Generally, direct surface sheet

flow occurs only nearshore, and depends on soil type, gradient,

and amount of water. All other surface water falling within the

shoreline drain zone that does not drain to a stream is isolated

from Lake Michigan. Ultimately, this water will evaporate or flow

to Lake Michigan as groundwater.

The shoreline drain zone generally consists of sand dune

systems and glacial landscapes. Within the set of isolated sur-

face water drainage areas, each water or wetland feature is class-

itied as either dune-formed or glacially-formed. The basis for

distinction is that glacial catchments have been isolated since

the retreat of the last glacier. The dune-formed areas have only

been isolated since development of the dunes (mostly Main

Algonquin of Michigan, Nipissing, Algoma, or Recent), and were

probably draining to the lake basin prior to that.

Non-dunal isolated surface drainage may occur because of

glacially formed depressions, a very gentle gradient, or an undul-

ating glacial terrain (moraines and till fields). These areas do

not have adequate surface water erosive energy to form channels

and drain the land. Glacial lakes and wetlands may occur proxi-

mate to sand dunes, and in some cases may not be immediately

discernable from those that are dune-formed. Generally, if there

is a ridge of non-dune material between the lake or wetland and

Lake Michigan, it is a glacial lake. In cases where that distinc-

tion is difficult to determine from topographic maps, the landward

watershed boundary can usually still be delineated. If the water-

shed is small relative to the size of the lake, it can generally

be assumed to be of glacial origin. Many glacial lakes were

formed from large pieces of ice that were left buried in the till

as the glacier receded, or in basins goug-ed out by the erosion of

a glacier.

Within the set of dune-formed drainage areas, we recognize

three general wetland formation types. Because of the highly

permeable soils, much of the surface water that falls on dunes is

absorbed and flows to Lake Michigan as groundwater. The dune-

formed surface water catchment types are described briefly in the

following paragraphs.

IV.A Interdunal Wetlands

The interdunal wetland has been recognized by the Michigan

Natural Features Inventory (MNFI) as a distinct community type.

MNFI describes these wetlands as "typically found in long troughs

between dune ridges, in wind-excavated depressions at the base of

blowouts, in hollows of dune fields wedged between inland lakes

and the Great Lakes, and in old river channels that once flowed

parallel to the lakeshore behind a foredune". Many interdunal

wetlands have fluctuating watertables seasonally and yearly,

related to changes in base lake level. In some instances they may

be ephemeral features. They occur as ponded areas only because of

the basins formed by sand dunes, and most of these areas probably

drained by sheet flow prior to dune formation.

In conducting our inventory, we initially identified several

areas as small dune-pocketed wetlands which we are now calling

transitional interdunal wetlands. These are wetlands that occur

at the landward base of the dunes, but have very small drainage

areas and essentially no evidence of drainage development. It was

not within the scope of this study to inventory interdunal wet-

lands, as MNFI is responsible for that task.

47 -

IV.B Bay-Mouth Bar Lakes

These lakes were originally bays of Lake Michigan, formed

either by glacial carving or sand deposition of hooked spits.

Where these bays occurred, wave energy was decreased at their

mouth, allowing for the deposition of sand bars by littoral drift.

As these bay mouth sand bars accumulated more sand, they began to

effectively close off the mouth of the bay from Lake Michigan,

forming a nearshore bay mouth bar lake. Generally, the sand bars

that began to close the bay were overridden by the formation of

sand dunes. In some cases, the level of the bay mouth bar lake

has remained at the 605' elevation it was formed at, or even been

uplifted higher. In others, when lake level dropped to 580', the

bay mouth bar lake also dropped to that level.

All first order streams that flowed into the original bay are

now re-ordered. If the lake has maintained a stream outlet, the

Qriginal streams become second order. If the bay mouth bar

lake is entirely closed off from Lake Michigan, the streams drain-

ing into the new lake are essentially isolated from Lake Michigan

in terms of surface drainage. In either case, the original

streams will continue to develop as if they were first order

streams, unless affected by uplift. The formation process of this

dune-associated wetland type was further described by Scott and

Dow (1932) in a descriptive case study of the Herring Lakes in

Benzie County.

A list of all bay mouth bar lakes formed along the eastern

shoreline of Lake Michigan is given in Table 2.

IV.C Dune-Pocketed Wetlands

The dune-pocketed wetland type developed as a first order

stream, but has since been blocked by sand dunes. Many of these

naturally dammed streams have since been artificially drained.

These areas normally have some dendritic drainage in the upland,

and collect into a catchment right at the base of the dunes. The

watershed boundary for the dune-pocketed wetland nearly always

defines the landward limit of the shore.line drain zone. The

possible formation processes of this type will be further de-

scribed in the following section. Table-3 lists the Dune-Pocketed

Wetland Watersheds of the eastern Lake Michigan shoreline.

Although the dune-pocketed wetland is the only first order

stream now blocked by sand dunes, it can be assumed that many of

the streams that presently drain may have been impeded by dune

development. The factors that determine a stream's degree of

impedance are its erosive energy (discharge), and the size of the

dune barrier. During the two major dune-building periods, base

lake level was dropping, so stream energy was increasing. Below

is the classification of the first-order streams south of Frank-

fort in terms of their degree of dune-impedance.

Table 2. Bay mouth bar lakes along the eastern shoreline
of Lake Michigan (listed south to north).

Number Name County Quadrangle Sheet

IM025 Grand Mere Lakes Drainage Area BERRIEN BRIDGEMAN/STEVENSVILLE
IM118 Silver Creek (Lake) Watershed OCEANA LITTLE SABLE POINT/MEARS
IM122 Bass Lake Watershed OCEANA+ PENTWATER
IM136 Big Sable (Hamlin Lake) Watershed MASON HAMLIN LAKE
LM154A Bar Lake Watershed MANISTEE PARKDALE
LM154B South Bar Lake Swamp Watershed MANISTEE PARKDALE
LM154C North Bar Lake Outlet Watershed MANISTEE PARKDALE
IM158 Portage Lake Watershed MANISTEE PARKDALE
IM801 Arcadia Lake MANISTEE ONEKEMA,
LM802 Upper and Lower Herring Lake BENZIE ELBERTA
IM804 Crystal Lake BENZIE. FRANKFURT/BEULAH
IM807 North and South Bar Lakes LEELANAU EMPIRE

Table A2-2. Dune-pocketed wetlands watersheds of the eastern
Lake Michigan shoreline (listed south to north).

Number Name/Code County --Quadrangle Sheet

IM019 Warren Dune-Pocketed
Wetlands Watershed (WD) BERRIEN BRIDGEMAN

LM022 Thornton Valley Dune-Pocketed
Wetlands Watershed *1 (TVI) BERRIEN BRIDGEMAN

LM023 Thornton Valley Dune-Pocketed
Wetlands Watershed *2 (TV2) B ERR IEN BRIDGEMAN

LM039 Mud Lake Dune-Pocketed
Wetlands Watershed (ML) BERRIEN+ COLOMA/COVERT

1,M040 Fish Cemetery Dune-Pocketed
Wetlands Watershed (FC) VAN BUREN COVERT

IM043 Dyckman Swamp Pocketed
Wetlands Watershed (DS) VAN BUREN COVERT

IM074 Saugatuck Dune-Pocketed
Wetlands Watershed (SD) ALLEGAN SAUGATUCK

IM076 Gilligan Lake Dune-Pocketed
Wetlands Watershed (GL) ALLEGAN SAUGATUCK

LM078B Kelly Lake Dune-Pocketed
Wetlands Watershed (KL) OTTAWA HOLLAND WEST+

IM095 Muskrat Lake Dune-Pocketed
Wetlands Watershed (MS) MUSKEGON DALTON/MICHILINDA

IAM103 Fruitvale South Dune-Pocketed
Wetlands Watershed (FS) MUSKEGON FLOWER CREEK

IM104 Fruitvale North Dune-Pocketed
Wetlands Watershed (FN) MUSKEGON FLOWER CREEK

IM113 Long Lake Dune-Pocketed
Wetlands Watershed (LL) OCEANA TOWN CORNERS

IM114 Skinny Lake Dune-Pocketed
Wetlands Watershed (SL) OCEANA TOWN CORNERS

IM117 Bigsbie Lake Dune-Pocketed
Wetlands Watershed (BL) OCEANA BIGSBEE LAKE+

LM138 Nordhouse Lake Dune-Pocketed
Wetlands Watershed (NL) MASON HAMLIN LAKE

LM139 Spartina Marsh Dune-Pocketed
Wetlands Watershed (SM) MASON HAMLIN LAKE

LM146 County Line Road Dune-Pocketed
Wetlands Watershed (CL) MASON+ MANISTEE
51 -

IV.D Somewhat Impeded Shoreline Streams

These streams occur within the shoreline drain zone and do

drain to Lake Michigan but their flow is somewhat impeded (Table

4). Most of these streams appear to have developed as dune-pock-

eted wetlands, and still have a dune-base catchment, but have

since developed drainage. Some of them (Sloan Pond, Pierson

Drain) run along the base of the dunes until they empty into a

first order stream. Others (Little Flower Creek, Section 33

Creek) drain behind the dunes until they reach a low spot and then

drain directly to the lake. Two of them (New Buffalo Lake

Watershed, Little Pigeon Creek) appear to have developed as drown-

ed river mouth lakes, but do not extend inland beyo nd the shore-

line drain zone.

IV.E Intermittent Stream Channel

This class of first order streams may also be considered as a

type of somewhat impeded shoreline stream (See Table 4, lower

part). Most its members occur in very sandy soils and are limited

by the amount of water that they drain. The only area south of

Frankfort in which such streams occur is in the vicinity of the

Fruitvale Dunes. In that area there are several relatively well

developed intermittent stream channels. These may have carried

more water during ancient times or their channels may simply

Table 4. Upper Portion -- Somewhat impeded streams of the
eastern Lake Michigan shoreline (listed from south
to north). Lower portion -- intermittent stream
channels of the eastern Lake Michigan shoreline zone.

Number Name County Quadrangle Sheet

IM004 New Buffalo Lake Watershed BERRIEN NEW BUFFALO WEST
LM041 Brandywine Creek Watershed VAN BUREN COVERT
LM075 Halfway Creek Watershed ALLEGAN+ SAUGATUCK+
LM080B Ten Hagen Creek (Sloan Pond) W. OTTAWA HOLLAND WEST+
IM083 Little Pigeon Creek Watershed OTTAWA PORT SHELDON
IM098B Pierson Drain Watershed MUSKEGON FLOWER CREEK
LM105 Little Flower Creek Watershed MUSKEGON FLOWER CREEK
LM147 Section 33 Creek Watershed MANISTEE MANISTEE

LM100 Lehman Road Intermitent
Dendritic Drainage Area MUSKEGON FLOWER CREEK

53 -

reflect a highly erodible and yet very porous substrate. They lie

in a sandy area between the heavy tills of the "Claybanks"area.

and the shoreline. In this area there is a complete gradation

from deeply cut channels presently occupied by lakes (drowned

river mouths), through dune-pocketed wetlands apparently not

having sufficient energy to cut through and limit dune develop-

ment, to these intermittent stream channels which do not have

enough water even to make it to the lake. North of Frankfort,

intermittent streams are very common flowing off the high glacial

bluffs and then being absorbed into the wide, sandy beach that has

been uplifted.

IV.F Unimpeded Shoreline Streams

These are first-order shoreline drain zone watersheds with no

apparent impedance to stream flow from dunes. These streams drain

through both dune and non-dune areas. Many may have been blocked

by ancient pre-Chippewa dunes which formed during the Chippewa

drawdown. When Lake Nipissing rose back to the original Algonquin

level, waves would have eroded much of the pre-Chippewa dunes,

allowing some blocked streams to begin draining again. Careful

study of the drainage patterns of each of these streams suggests

which were dune-pocketed in ancient times and which were not, but

such analysis is beyond the scope of this report.

IV.G Drowned River Mouth Lakes

These are lakes that are generally associated with the larger

inland-draining rivers in the eastern Lake Michigan basin. They

are related to dunes. Spits of bars formed across the mouths of

these rivers to partially block their flow to the lake. Dunes

have enhanced the process. Table 5 lists all drowned river mouth

drainage systems of the eastern Lake Michigan shoreline. Evans

(1937) early characterized this partially unformed drainage type

well:

"Closely connected in their origin with

the history of the terraces and dunes is a

series of lakes along the shore from Saugatuck

to Frankfort. They are of all sizes up to

four or five square miles in area and sep-

arated from Lake Michigan only by the sand

dunes. Their surfaces are generally only few

inches above that of the big lake, and the

natural drainage is through shifting, sand-

choked channels .... In many places the natural

outlets were straightened and channels were

cut (by man) that are easily maintained at

depths sufficient for navigation; in this way

perfect landlocked harbors were formed ....

The physiographic history of these harbor

lakes is interesting and unique and is, of

course, intimately linked with that of the

Table 5. Drowned river mouths of the eastern Lake
Michigan shoreline (listed from south to north).

Number Name County Quadrangle Sheet

LM005 Galien River Watershed BERRIEN NEW BUFFALO EAST
LM028 St. Joseph River Watershed BERRIEN BENTON HARBOR+
LM072 Kalamazoo River Watershed ALLEGAN SAUGATUCK+
IM078A Lake Macatawa Watershed ALLEGAN SAUGATUCK+
LM080A Pigeon River Watershed OTTAWA PORT SHELDON+
LM085 Grand River Watershed OTTAWA GRAND HAVEN
LM089 Mona Lake (Black Creek) Watershed MUSKEGON MUSKEGON WEST
IAM091 Muskegon River Watershed MUSKEGON MUSKEGON WEST+
IM096 Duck Lake Watershed MUSKEGON DALTON/MICHILINDA
LM098 White Lake Watershed MUSKEGON MICHILINDA+
IM107 Flower Creek Watershed MUSKEGON FLOWER CREEK
LM115 Stony Lake Watershed OCEANA TOWN CORNERS+
IM120 Pentwater Lake Watershed OCEANA PENTWATER
IM132 Pere Marquette River Watershed MASON LUDINGTON
JLM134 Lincoln River Watershed MASON LUDINGTON
LM152 Manistee River Watershed MANISTEE MANISTEE/PARKDALE
IM803 Betsie River BENZIE ELBERTA/FRANKFURT

56 -

diversified history of the Great Lakes them-

selves ....

The development of the coastal lakes

after the retreat of the ice may be treated as

follows: As the waters of Lake Chicago and

Lake Algonquin fell, the streams cut down.

When the waters rose again during Nipissing

time, the streams were ponded up at the mouth

and did some sidecutting. This ponding was

aided somewhat by the deposition of dune sand.

The fall of the Lake Nipissing level caused

the streams again to cut down rapidly; but

near the mouth they continued clogged with

sand, so that sidecutting also continued...'.

During the Chippewa drawdown, base level for these rivers

dropped by approximately 400'. It is apparent from the size of

their watersheds and the width and depth of their channels that

they were not impeded by the pre-Chippewa dunes. The sidecutting

that occurred as the river meandered within its alluvium-filled

channel produced the equivalent of a bay, and allowed sand to

accumulate across its mouth by longshore drift. When Nipissing

levels fell, dunes may have formed on those sand bars. However,

because of the large amount of energy associated with the large

watersheds, it is doubtful that these sand barriers could ever

completely separate the lakes occurring at their mouths from Lake

Michigan.

57 -

I-V.H Unimpeded Inland Streams

These inland streams are not sufficiently ponded at their

mouth to be considered drowned river mouth lakes. Generally their

watersheds are not as large as the drowned river mouth lakes, and

they occur in areas where the gradient to Lake Chippewa was not as

steep. These streams are essentially the opposite of the dune-

pocketed wetland type -- not at all dune-impeded, and draining

well inland. Some of these streams may have been somewhat dune-

impeded during the Chippewa drawdown by dunes that have since been

removed by erosion. However, for present purposes, these streams

are considered non-dune-associated.

V. DESCRIPTION OF DUNE-POCKETED WETLANDS WATERSHEDS

Application of the classification and screening criteria to

all drainage systems of the Lake Michigan shoreline resulted in

recognition of eighteen dune-pocketed wetlands watersheds. They

are listed in Table 3 above. These watersheds have blocked drain-

age that suggests they would have been normal direct drain stream

watersheds had they not been blocked by the dunes. They are

characterized by wetlands that lie at the base of sand dunes.

They differ from one another in several ways. The following

paragraphs provide descriptions and data summaries for each of the

dune systems having dune-pocketed wetlands watersheds (topographic

maps of each site are included in Appendix II). Figure 7 shows

the location of each. Vegetational characteristics of several of

these (as well as other case study examples) are described in

Appendix I.

V.A Warren Dunes System

The general character of the inland in this system is old

lake bottom, and rather flat. There is extensive agricultural

development here. The Warren Dunes are substantial barriers.

This system dates back to the earliest dune forming period

(Glenwood). The weakly developed drainage channels provide evi-

dence of pre-Chippewa dune pocketing. The South Branch of Paint-

erville Drain has well developed, deeply cut drainage channels,

59 -

---------- ---------

@-4 ------ --
+

CL
Sid
NL
@A.
!+ +

-----------

BL I--------

+
j.
FS
OW
MS 4

----------

KL
GL
SD

j-
DS
FC
ML

TV2

TVI
WD

Figure 7. Location of dune-pocketed wetlands watersheds along the
eastern Lake Michigan shoreline. Names corresponding to these
codes are: WD-Warren Dunes, TV1-Thornton Valley *1, TV2- Thornton
Valley *2, ML-Mud Lake, FC-Fish Cemetery, DS-Dyckman Swamp, SD-
Saugutuck Dunes, GL-Gilligan Lake, KL-Kelley Lake, MS-Muskrat
Lake, FS-Fruitvale South, FN-Fruitvale North, LL-Long Lak e SL-
Skinny Lake, BL-Bigsbie Lake, NL-Nordhouse Lake, SM-Spartina
Marsh, CL-County Line Road.

60 -

Chippewa). The North Branch of Painterville Drain is much less

deeply incised, indicating recent artificial drainage. Drainage

development is apparently pre-Chippewa behind the entire dune

system north to Grand Mere Lakes. However, artificial agricul-

tural drainage and the construction of 1-94, Red Arrow Highway,

and the town of Bridgeman, have altered original drainage patterns

considerably. Most of the wetlands have been drained.

We recognized three dune-pocketed wetlands watersheds in this

system, but due to the extent of artificial drainage the divides

between them are somewhat arbitrary. The creek draining west of

Bridgeman is probably artificial and drains what appears to have

formerly been a dune-pocketed wetland. However, since it shows no

current evidence of pocketing, this watershed was n'ot included as

dune-pocketed. There is a distinct landward boundary behind all

of these watersheds, most likely an ancient beach ridge. Inland

areas behind this drain south to the Galien River or north to the

St. Joseph River. Table 6 gives the development, area, and owner-

ship characteristics of each of the dune-pocketed wetlands water-

sheds in this system.

61 -

Table 6. Dune-pocketed wetlands watersheds within the Warren Dunes
System -- development, area, and ownership. Development index
values range from 0 (undeveloped) to 10 (highly developed) -- see
Table I in Section II for an explanation of scale values (U
Upland; L = Lowland/Catchment; W = Watershed weighted mean).
Ownership percentages are given for Private (P), Federal (F), and
State (S).

Dune-pocketed Degree of Area Ownership
wetlands watershed Development (acres)! percent
name and code U L W P F S

Warren Dunes (WD) 1 10 10 10 875 1 79 0 21

Thornton Valley 1 (TVI)! 10 10 10 1070 !100 0 0

Thornton Valley 2 (TV2)! 10 10 10 445 !100 0 0

V-B Van Buren Dunes System

The southern boundary of the Van Buren Dunes System is loc-

ated south of Lake Michigan Beach in northern Berrien County. The

system extends northward to just north of Van Buren park. From

Lake.Michigan Beach to the county line, the shoreline landscape is

predominantly glacially formed. There are numerous glacial lakes

throughout the area. The largest three, Hibbard, Harris, and

Jacob's Chapel Lakes, occur adjacent to the dunes. However, by

mapping the watershed boundaries and carefully studying the topo-

graphy it is evident that these are not dune formed. Jacob's

Chapel Lake may be artificially deepened due to the construction

of the ramps to 1-94. Rogers Creek, which drains the area around

these lakes, has well developed drainage -channels suggesting that

at one time this creek was cutting to a lower lake level. The

inland topography north of the county line is much flatter, appar-

ently an old lake bed. Upland drainage is much less well devel-

oped in the Rogers Creek area, and collects to a broad marshy flat

that extends behind the dunes. Mud Lake is the only well defined

dune-pocketed catchment. The Mud Lake dune-pocketed wetlands

watershed is artificially drained to the south into Rogers Creek

and to the north by Brandywine Creek. Nonetheless, it maintains a

very marshy character. The watershed to the north, with the Fish

Cemetery marked, is developmentally similar to the Mud Lake area,

but it is more directly drained by Brandywine Creek. Prior to

artificial channelization of Brandywine Creek, the Mud Lake wet-

land was apparently completely dune-pocketed and it remains essen-

tially so to the present time. Brandywine Creek appears to run

under 1-94 and Blue Star Highway by a culvert, through an old

sandpit excavation and out to Lake Michigan just north of Pali-

sades Park. Dyckman Swamp is a broad, marshy area, artificially

drained both to the north and to the south. The shoreline drain

zone behind this entire dune system extends nearly four miles

inland. It is a very flat area, with almost no drainage develop-

ment, suggesting that it has been pocketed since the formation of

the earliest dunes. Table 7 shows the development, area, and

ownership of the dune-pocketed wetlands watersheds in the Van

Buren Dunes System.

Table 7. Dune-pocketed wetlands watersheds within the Van Buren
Dunes System -- development, area, and ownership. Development
index values range from 0 (undeveloped) to 10 (highly developed)
see Table I in Section II for an explanation of scale values (U
Upland; L = Lowland/Catchment; W = Watershed weighted mean).
Ownership percentages are given for Private (P), Federal (F), and
State (S).

Dune-pocketed Degree of I Area Ownership
wetlands watershed Development (acres)! percent
name and code U L W P F S

Mud Lake (ML) 10 7 8 3422 1100 0 0

Fish Cemetery (FC) 7 10 9 1884 !100 0 0

Dyckman Swamp (DS) 10 7 8 1451 1 78 0 12

V.C_ Saugatuck Dunes System

This system extends from just west of Kalamazoo Lake north to

Lake Macatawa. Both of these lakes are drowned river mouth type.

The shoreline between them is dominated by sand dunes and drains

to the Kalamazoo River through Goshorn Creek. North of this

drainage system is the Saugatuck Dunes pocketed wetlands water-

shed, directly behind Saugatuck Dunes State Park. This watershed

is relatively small and does not have obvious upland stream chan-

nels, but the gradient slopes gradually to the dunes and there is

a dune-base catchment (elevation 620'). Apparently dunes formed

prior to the development of a stream channel, and the gradient has

remained low. Surface drainage is probably connected to ground

water at the catchment.

Halfway Creek drains the area to the north, collecting into a
*dune-base catchment at about 600' elevation. The dunes are weaker

here, and apparently drainage has broken through to the big lake.

However, upgradient stream channels contain numerous ponded areas,

implying a high watertable and poorly developed surface drainage.

Gilligan Lake dune-pocketed wetlands watershed is north of

Halfway Creek. The upland is glacial material with rather high

topographic relief. Drainage channels have been incised, indicat-

ing that this area was probably previously drained by a first-

order stream. The apparently shallow catchment is at elevation

610'. Another smaller catchment at the same elevation occurs just

to the north of the first, behind a jutting dune. Judging from

the degree of drainage development and catchment elevation, this

area may have drained to Lake Chicago, but has been blocked by the

first dunes formed by the Chippewa drawdown.

The Kelley Lake watershed to the north has similar

characteristics, but has been drained to Lake Macatawa. This

watershed continues to show evidence of dune pocketing, despite

the artificial drainage so we have included it as a landward dune-

pocketed wetlands watershed. Table 8 shows the disturbance, area,

and ownership characteristics of the dune-pocketed wetlands

watersheds within the Van Buren Dunes System.

Table B. Dune-pocketed wetlands watersheds within the Saugatuck
Dunes System -- development, area, and ownership. Development
index values range from 0 (undeveloped) to 10 (highly developed)
see Table 1 in Section TI for an explanation of scale values (U
Upland; L = Lowland/Catchment; W = Watershed weighted mean).
Ownership percentages are given for Private (P), Federal (F), and
State (S).

Dune-pocketed Degree of Area Ownership
wetlands watershed Development (acres)! percent
name and code U L W P F S

Saugatuck Dunes (SD) 3 3 3 545 1 86 0 14

Gilligan Lake (GL) 1 7 7 7 1569 1100 0 0

Kelley Lake (KL) 7 7 7 1684 !100 0 0

V.D Michilinda Dunes System

The Michilinda Dunes begin at Muskrat Lake and end just north

of White Lake. However, there are isolated dune features south-

ward along the shore all the way to Muskegon dunes. The landscape

is an ancient pro-glacial lake bottom. It is relatively flat,

with sandy soils. In this area of shoreline there are numerous

ancient stream channels that were apparently incised during the

Chippewa drawdown. However, many of the smaller channels are dry

due to their porous soils and small watershed area.

Muskrat Lake dune-pocketed wetlands watershed was formed by

the same process as Muskegon Lake to the south and Duck and White

Lakes to the north. These systems are exemplary dr'owned river

mouth lakes, formed in very erodible substrate. The watershed

sizes of the streams are directly related to the size of the

catchments they formed. All are apparently groundwater connected.

Ip contrast to the larger lakes watershed, the drainage system

collecting to Muskrat Lake was not large enough to maintain the

erosive energy needed to maintain an outlet. The channels between

Duck Lake and White Lake have even smaller watersheds than Muskrat

Lake and therefore did not erode a sufficient channel to form a

groundwater connected water body. Table 9 summarizes characteris-

tics of the Michilinda Dunes System.

Table 9. Dune-pocketed wetlands watersheds within the Michilinda
Dunes System -- development, area, and ownership. Development
index values range from 0 (undeveloped) to 10 (highly developed) -
see Table 1 in Section II for an explanation of scale values (U
Upland; L = Lowland/Catchment; W = Watershed weighted mean).
Ownership percentages are given for Private (P), Federal (F), and
State (S).

Dune-pocketed Degree of Area Ownership
wetlands watershed Development (acres)! percent
name and code U L W P F S

Muskrat Lake (MS) 3 3 3 3739 1100 0 0

V.E Fruitvale Dunes System

This system begins approximately three miles north of the

White Lake outlet and ends just inside of the Oceana county line.

Although it is relatively close to the Michilinda area, it has

developed quite different drainage patterns. The Fruitvale Dunes

system was apparently formed over a glacial moraine and has much

less permeable soils. The "claybanks" are located east of the

Fruitvale System.

The Lehman Road intermittent stream channels formed upon clay

substrate but have such small watersheds that surface water goes

to ground before reaching the lake, except during large rain

storms and in the spring. Lehman Road Creek has a somewhat larger

watershed with glacial ponds at the head.' It is groundwater

connected. The Fruitvale South area is transitional between

interdunal and dune-pocketed wetlands. It has not developed an

upgradient stream drainage system. Throughout the area there are

small dune and shoreline ridge features inland, and between these

small wetland areas dominated by buttonbush (Cephalanthus occiden-

talis) occur over fine textured substrate materials.

The Fruitvale North dune-pocketed wetlands watershed has

developed drainage channels that collect to several pockets at the

base of the dunes. This is an exemplary dune-pocketed wetlands

watershed system (see Figure 10). The drainage pattern in this

area contrasts with that of Little Flower Creek to the north,

which has approximately the same gradient and watershed size but

71 -

has drained out to the big lake and therefore has developed much

more well defined and deeply cut river channels. Little Flower is

contrasted further by Flower Creek, which has a drowned river

mouth, evidencing that it drained to Lake Chippewa. Little Flower

Creek apparently has cut through to the lake more recently, pro-

bably with the erosion of formerly blocking pre-Chippewa dunes by

the rising waters of Lake Nipissing. Table 10 summarizes develop-

ment, size, and ownership characteristics of the dune-pocketed

wetlands watersheds of the Fruitvale Dunes System.

Table 10. Dune-pocketed wetlands watersheds within the Fruitvale
Dunes System -- development, area, and ownership. Development
index values range from 0 (undeveloped) to 10 (highly developed)
see Table 1 in Section II for an explanation of scale values (U
Upland; L = Lowland/Catchment; W = Watershed weighted mean).
Ownership percentages are given for Private (P), Federal (F), and
State (S).

Dune-pocketed Degree of 1 Area Ownership
wetlands watershed Development (acres)' percent
name and code U L W P F S

Fruitvale South (FS) 3 3 3 430 !100 0 0

Fruitvale North (FN) 5 2 3 265 1100 0 0

V.G Little Sable Point South Dunes System

This dune system begins where Whiskey Creek enters Lake

Michigan and extends north to the town of Little Point Sable.

The substrate here is similar to that of the Fruitvale system

to the south. The southern part of the Little Sable Point system

is within Claybanks Township. Whiskey Creek has a very well

developed stream drainage system but is now small and intermit-

tent. This stream appears to have cut its channels when it drain-

ed to Lake Chippewa. There is no evidence of blockage by pre-

Chippewa dunes. Immediately to the north is Long Lake, with a

larger watershed and deeper channel. The channel of Long Lake was

drowned by the rising waters of Lake Nipissing and then completely

pocketed by post-Chippewa (mostly Nipissing) dunes as the lake

fell to Algoma and then present levels. "Skinny Lake" between

Long and Stony Lakes, developed the same way but its channel has

been more filled with encroaching sand. On-site examination

indicated that filling by eroding sand is continuing at the west-

ern end of this wetland area. Stony Lake is a drowned river mouth

of a similar type but with a much larger watershed that has suff-

icient flow to keep the channel open.

Bigsbie Lake is exemplary of the post-Chippewa dunes pocketed

lake and wetlands type (see Figure 10). It is quite similar in

its formation process to Muskrat Lake in the Michilinda system.

74 -

It has deeply cut channels, apparently of pre-Chippewa age,

through very sandy substrate. Now the upper channels are essen-

tially dry, but Bigsbie Lake and a very small wetland at its upper

end remain where the channel is deep enough to connect to ground

water. The channel was closed off by formation of post-Chippewa

dunes.

Table 11 gives index values for disturbance, as well as area

and ownership characteristics, for watersheds in the Little Sable

Point South System.

Table 11. Dune-pocketed wetlands watersheds within the Little
Sable Point South Dunes System -- development, area, and owner-
ship. Development index values range from 0 (undeveloped) to 10
(highly developed) see Table 1 in Section II for an explanation
of scale values (U Upland; L = Lowland/Catchment; W = Watershed
weighted mean). Ownership percentages are given for Private (P),
Federal (F), and State (S).

Dune-pocketed Degree of Area Ownership
wetlands watershed Development (acres)! percent
name and code U L W P F S

Long Lake (LL) 5 4 5 1 1122 1100 0 0

Skinny Lake (SL) 2 3 3 864 !100 0 0

Bigsbie Lake (BL) 4 3 3 1000 !100 0 0

V.H Big Sable Point/Hamlin Lake/Nordhouse Dunes System

The formation of Nordhouse Dunes is due primarily to

longshore drift that began forming spits across the mouth of the

Big Sable River during Lake Chicago. The cause of those spits is

most likely shallow offshore water or a moraine which sent the

southerly currents lakeward and then deposited most of the sand on

the north side of the river mouth. When Lake Chicago dropped in

elevation first to Main Algonquin of Michigan and then to Lake

Chippewa, that sand formed an extensive dune field along the shore

to the north, extending approximately to the Manistee-Mason county

line. The first pre-Chippewa dune ridge pocketed Nordhouse Lake,

the Spartina Marsh area, Porter Creek, and Cooper Creek. That

ridge (improperly name "Nipissing trail ridge" on the Forest

Service trail maps in this area) probably extended well lakeward

during the Chippewa drawdown. The Big Sable River cut down during

the Chippewa period, as did many of the other streams that

presently enter Hamlin Lake. The bay-mouth bar that formed

offshore giving rise to Hamlin Lake has preserved several of these

downcut river mouths from alteration by dune formation and other

shoreline processes along the eastern shore of Hamlin Lake. Upper

Hamlin Lake is a drowned river mouth.

When levels rose back to Nipissing (605') most of the pre-

Chippewa dunes were eroded by waves and reworked as longshore

drift, forming a large baymouth bar in front of the Big Sable

River. The downcutting of the Big Sable and other streams that

formerly entered the big lake at this point is evidenced by the

deeply cut river mouth channels along the east side of Hamlin

Lake. Reverse eddy currents in Lake Nipissing apparently carried

sand along the shore from the north by longshore drift, and depos-

ited it in the south on the Big Sable point. That process eroded

the northern part of this once more extensive dune field and

eventually "unpocketed" some of the formerly blocked drainage

systems. Cooper Creek was probably opened in this way. It has

developed some drainage channels near the lake but remains weakly

channeled. Porter Creek was probably opened even more recently

and has a very poorly developed drainage system. Immediately

south of Porter Creek the shoreline is expanding. Therefore,

watersheds lying behind the dunes south of this point (i.e.

Spartina Marsh watershed and Nordhouse Lake watershed) remained

pocketed.

Lost Lakes appears to be a system of large interdunal lakes

that formed between the first Algonquin ridge and the succeeding

post-Chippewa dunes. These later dunes were formed to the west on

the developing bar platform, leaving a low area which is presently

occupied by the Lost Lakes, between these two sets of dunes. The

lakes formed on the inner margins of bay-mouth bar lakes should be

studied further. These constitute a unique formation type that

has not been specifically identified heretofore.

We have recently described the Big Sable Point/Hamlin

Lake/Nordhouse Dunes Environmental System in much more detail

78 -

(Goff et al, 1987). Table 12 shows the disturbance, area, and

ownership characteristics of the dune-pocketed wetlands watersheds

in the Big Sable Point/Hamlin Lake/Nordhouse Dunes System.

Table 12. Dune-pocketed wetlands watersheds within the Big Sable
Point/Hamlin Lake/Nordhouse Dunes System -- development, area, and
ownership. Development index values range from 0 (undeveloped) to
10 (highly developed) -- see Table I in Section II for an explan-
ation of scale values (U Upland; L = Lowland/Catchment; W =
Watershed weighted mean). Ownership percentages are given for
Private (P), Federal (F), and State (S).

Dune-pocketed Degree of Area Ownership
wetlands watershed Development (acres)! percent
name and code U L W P F S

Nordhouse Lake (NL) 2 1 1 938 0 100 0

Spartina Marsh (SM) 1 2 2 1181 0 100 0

V.I County Line Road Watersheds

This is a small dune area between Gurney Creek and Magoon

Creek. Apparently both watersheds drained to Lake Chippewa,

although the southern section most likely flowed into Gurney Creek

before entering the big lake, The character of the natural drain-

age systems in the area has been altered by artificial drainage.

There are two watersheds, weekly connected along the base of the

dunes. The first, County Line Road North, originates in springs

that are recharged from the glacial hills to the east. It flows

westward to the base of the dunes and then north along the dunes

for about a mile before it enters the big lake. From the stream

course it is obvious that the dunes have had an effect. However,

on-site inspection shows this to be a cedar swamp area with a

clearly flowing stream throughout its entire length. Other than

the fact that the cedar swamp is unusually wet (see Appendix 1),

there is no evidence of pocketing by the dunes.

The southern watershed unit apparently also receives water

from the east but is separated from the headwaters of the northern

watershed unit by a dividing ridge. At the base of the dunes it

is difficult to distinguish these two units. However, whereas the

stream draining the northern units turns to the north, the south-

ern unit apparently slopes slightly to the south from this point

so water moves southward, eventually pocketing against the dunes.

At the catchment of this southern unit there is evidence of dredg-

ing and other disturbance. Apparently a channel was dredged

across the low dunes several decades ago in an attempt to drain

this pocketed wetland area. A road was built across the south-

western tip of the wetland at the base of the dunes. Dead,woody

wetland vegetation throughout the southern watershed unit shows

the effect of draining the area. Then, probably quite recently,

the culvert collapsed (it is not now functioning) and the lower-

most portion of the area re-flooded. There is now evidence of

extreme seasonal fluctuation in water level (see Appendix I).

Development, size, and ownership characteristics of the County

Line South dune-pocketed wetlands watershed area are given in

Table 13.

Table 13. Dune-pocketed wetlands watersheds within the County Line
Road System -- development, area, and ownership. Development
index values range from 0 (undeveloped) to 10 (highly developed) -
see Table 1 in Section II for an explanation of scale values (U
Upland; L = Lowland/Catchment; W = Watershed weighted mean).
Ownership percentages are given for Private (P), Federal (F), and
State (S).

Dune-pocketed Degree of Area Ownership
wetlands watershed Development (acres)! percent
i
name and code U L W P F S

County Line South (CL) 3 10 8 1 315 1.100 0 0 11

V.J Disturbance Analysis

One of the motives for this study was to find out how

"unique" the dune-pocketed wetlands watersheds in the Nordhouse

Dunes System are in comparison to other examples of this land type

occurring elsewhere along the shoreline. We had stated that the

two dune-pocketed wetlands watersheds in the Nordhouse area were

the largest relatively undisturbed examples in public ownership.

We found during the course of this study that while there are

other exemplary areas, none are as undisturbed as Nordhouse.

Figure 8 shows a frequency plot of the disturbance index

values for the eighteen dune-pocketed wetlands watersheds included

in the study.

3. Cr

NUMBER

BL
5 SL CL
FN KL FC
FS GL TV1
Sm Ms DS TV2
NL SD LL ML WD

0-2 3-4- 5-6 7-8 9-1o

DISTURBANCE INDEX VALUE

Figure 8. Disturbance index values of eighteen
dune-pocketed wetlands watersheds along the eastern
Lake Michigan shoreline. Table 3 gives the names
corresponding to the abbreviations indicated inside
each bar of the graph. Watersheds having an index
value of 3-4 should be considered candidates for
special management.

VI. ORDINATION AND CLASSIFICATION

OF DUNE-POCKETED WETLANDS WATERSHEDS

In order to better understand the similarities and differ-

ences among the dune-pocketed wetlands watersheds, we prepared

several graphical presentations. The most meaningful of these

shows the distribution of the dune-pocketed wetlands watersheds

within a simple ordination (graph) of watershed area (logarithmic

scale) and topographic relief (Figure 9). These two factors are

related to the development of the drainage systems and to their

surface gradient and erosive energy. Presentation of the water-

sheds in this form provides a framework for interpretation of

relationships among them and for development of hypotheses regard-

ing their development.

The most obvious pattern in this ordination is the peaked

distribution. This can be accounted for by two effects, both

resulting from the relationship between topographic relief, water-

shed area, and the erosive energy available to cut through any

impedance at the mouth of the watershed (e.g. dunes). None of the

dune-pocketed watersheds included in the study are found in the

upper left sector of the ordination because the very steep grad-

ient implied by this combination of watershed size and topographic

relief would normally erode back-and enlarge the watershed. No

dune-pocketed watersheds are found in the upper right sector of

the ordination because the erosive energy of watersheds that large

OSL

OBL Type
150- A

OOL

*KL

Topographic
Relief 100- OFN Type 0 S
(FEET) - OCL B OFC
OSD ODS
OTV2 0 Tvi NL
OWD
50- OFS OSM

00 1OU0 10000
WATERSHM AREA (ACRES) log scale

Figure 9. Area/topographic relief ordination of dune-pocketed
wetlands watersheds along the eastern Lake Michigan shoreline.
Dashed lines enclose initial clusters. Solid lines enclose types
finally determined.

and with that much relief cut through the dunes and did not remain

pocketed. Had they been included, many interdunal wetland basins

would have fallen in the upper left portion of this ordination and

many direct drain streams and drowned river mouth lake watersheds

-along the shoreline would have fallen in the upper right portion.

Most of the dune-pocketed wetlands watersheds have between 50

and 100 feet of total relief. A few have more than 100 feet.

Comparison of the position of the watersheds within this simple

ordination, par-Licularly their topography, led us to initially

identify two clusters as possibly distinct types (Figure 9 --

dashed lines). The placement of watersheds that did not fit con-

veniently into one of these clusters required further reexamin-

ation of drainage characteristics.

The single low relief watershed at the lower left of the ord-

ination is a small area that occurs on the landward edge of the

dunes. Careful examination of its mapped topography and drainage

in relation to topographic characteristics of the s'urrounding

landscape, as well as field examination, convinced us that it was

an "borderline" dune-pocketed wetland, transitional to a interdun-

al area. To avoid overlap between this study and work on the

interdunal wetlands underway by the Michigan Natural Features Inv-

entory, we did not generally consider interdunal wetlands. Several

watersheds (not shown) were examined and analyzed because they

were thought to be small examples of the landward dune-pocketed

wetlands watershed type. Landward dunal features associated with

their formation were not obvious from published information

sources. However, they lack the internal lakeward drainage

patterns associated with dune-pocketed wetlands and in every

instance upon field examination we found them to be associated on

all sides with dune features. The process of their formation is

essentially interdunal. We have retained only one, as an example

of the transitional type.

Examination of the high relief watersheds (Type A) shows

localized relatively deeply incised channels suggesting a differ-

ent process of origin from the watersheds lower in the ordination.

The watershed of Muskrat Lake Watershed (MS) has a similar pattern

of relief to those higher in the graph even though its total

amount of topographic relief is not as great. Based on its pat-

tern of relief it was included in Type A., Gilligan Lake Watershed

(GL) and Kelley Lake Watershed (KL) occur within the Saugatuck

Dune System. South of them is the Halfway Creek somewhat impeded

stream drainage system which is topographically similar in general

pattern to GL and KL. None of these three watersheds are as

dramatically channeled as those of the other watersheds in Type A,

but based on their position in the ordination as well as the

presence of channels more deeply incised than those typical of

watersheds in Type B, these were included in the Type A cluster.

They are actually intermediate in their topographic pattern and

should be studied further. Figure 9 (solid lines) shows the two

types finally established (with Fruitvale South shown as an excep-

tion to either type).

We postulate that the Type B watersheds were all pocketed by

more ancient pre-Lake Chippewa (mostly Algonquin) dunes, and have

remained pocketed ever since, whereas the Type A watersheds were

unimpeded during the Chippewa low water stage. The enhanced

downcutting of channels during the Chippewa drawdown accounts for

the deeply cut ravines characteristic of Type B watersheds.

Figure 10 shows panoramic views of typical examples of these

two dune-pocketed wetlands watershed types, and Figure 11 shows

the corresponding topographic and drainage patterns. The upper

panorama shows much of the Bigsbie Lake watershed. The lake is a

permanent feature. There is little lowland. In the upper, shal-

low part of the lake, emergent aquatic vegetation is found, but

there is almost no shrub carr or sedge meadow development. The

well drained upland gives way abruptly to the dammed lake. This

pattern is typical of Type A watersheds. According to our hypo-

thesis the channel now occupied by Bigsbie Lake was deeply cut

during the Chippewa low water stage. With rising lake levels its

channel was flooded in a manner similar to that of drowned river

mouths. With the drop in water level from Lake Nipissing (605')

to present (580'), dunes eventually pocketed the lake. Dune sand

is slowly encroaching on the lower end of the lake and the upper

end is gradually filling in due to erosion of the adjacent hills

(which are veneered with older dune sand).

'i Af

41
qv/@@
ro,,t,

ij.

'j %@l

z )

'AN

Fv..

r-Y

1,@. -ir
7

iP@

J,4
ti

wAV a
MiX

-to.
1@ V,14;11

-SIN

Figure 10. Panoramic views of two
contrasting Dune-Pocketed Wetlands
watershed systems.

Bigsbee Lake Dune-Pocketed
Wetlands watershed was dammed
by Nipissing or more recent
dunes (Type A, upper photo-
graph). Its deep channel was
cut during the Chippewa
drawdown when it was an
unimpeded stream.

Fruitvale North Dune-Pocketed
Wetlands watershed (Type B,
bottom photograph) is a low
relief watershed pocketed in
ancient times by Algonquin or
older dunes prior to the
Chippewa drawdown.

eb
LEY,

@JYL

Figure 11. Topographic
maps of the Bigsbie Lake
Type A dune-pocketed
wetlands watershed area
(upper map), and the
Ek
Fruitvale North Type B
dune-pocketed wetlands
memn
watershed area (lower
map). Topographic
relief in much greater
in the Type A watershed.

-Sv

'.0
VA i

The watershed of the Fruitvale North dune-pocketed system

(lower panorama) has been pocketed by dunes since before the

Chippewa stage (pocketing dunes in this case are apparently of

early Algonquin age). Throughout the upgradient watershed (off

the picture to the upper right) the landscape is characterized by

low, poorly drained areas. Although it does slope gently toward

Lake Michigan and eventually forms a catchment at the base of the

dunes, drainage channels have developed only very weakly in this

watershed. Apparently during the time of maximum drainage channel

development for many of the other watersheds along the shoreline,

this area (and other areas of pre-Chippewa dune pocketing) were

protected from the downcutting process. Not only are they pock-

eted by the present dunes of the Lake Michigan shoreline, Type B

drainage systems have been geomorphologically preserved by dunes

formed very early in post-glacial time.

VII. DISCUSSION

This study was initiated in response to a land management

issue. While the study was underway the Nordhouse Dunes area in

northern Mason County was officially declared a wilderness, the

only area so designated in the lower peninsula, At the time of

this writing there are lawsuits in process that question the

authority of the government to limit hydrocarbon development

within that area. As we have conducted ecological studies of the

Nordhouse dunes and the surrounding landscape during the past two

years, we have become aware that this landscape was developmental-

ly a single system. The dunes block drainage, forming dune-pock-

eted wetlands stretching eastward clear to the watershed divide.

Moreover, the drainage systems in these dune-pocketed areas are

very poorly developed. In contrast to other shoreline areas,

blocking dunes stopped the normal process of stream downcutting,

effectively preserving the drainage system in the form that it had

soon after the glacier retreated. It was this realization that

led us to explore several questions, "How extensive are such dune-

pocketed wetland watersheds?" "How did they form, and how are

they related to other types of drainage systems?" "Where do they

occur and what are their present-day characteristics?"

These and related questions have led us into an exploration

of the surface drainage systems of the eastern Lake Michigan

shoreline. When we started the study we harbored some doubts

about the significance of this previously undescribed type of

94 -

watershed. Now we are completely convinced that these landward

dune-pocketed wetland watersheds form a set of dune-related sys-

tems that is of limited extent and deserving of special management

consideration. The distinction between the two major types of

dune-pocketed wetlands recognized in this study -- high relief

systems formed by more recent (Nipissing) dune development, and

the flatter, older systems where drainage has been blocked since

the early Algonquin recession to Lake Chippewa -- may provide

important clues for more general understanding of wetland drainage

systems in other areas.

One deficiency of this study needs to be pointed out. We

recognize that the study would have been greatly improved if we

had been able to deal more effectively with groundwater. Soil and

substrate characteristics vary regionally'as well as locally along

the shoreline. Patterns of variation in substrate and groundwater

processes may account for some of the distinctions observed be-

tween the drainage systems studied. But the time available and

the scope of study simply did not permit us to include much sub-

strate and groundwater information.

Our original hypothesis was that few if any of the drainage

systems that presently cut through the dunes were pocketed at an

earlier time. This reasoning was based on the assumption that the

pre-Chippewa dunes that are presently found along the shoreline

constitute virtually the complete extent of such dunes. If so,

and if a given stream along the shore is unimpeded now, it must

95 -

have always drained. Gradually we have come to realize that this

is not necessarily so. In some areas older dunes that previously

blocked stream drainage appear to have been eroded away on the

lakeward side by subsequent rising water levels. Comparison of

the degree of downcutting in watersheds that are pre-Chippewa

dune-pocketed (e.g. Nordhouse Lake, Spartina Marsh) and adjacent

watersheds of presently unimpeded direct drain streams (e.g.

Porter Creek), suggests that these presently unimpeded streams

emptying to Lake Michigan were formerly impeded, at least to some

degree.

Consideration of the conditions that would have been likely

during the time when the lake level was dropping from Main Algon-

quin of Michigan (605') to Chippewa (230') further suggests that

extensive dune fields may have formed in some presently offshore

areas. The initial drop to Main Algonquin of Michigan was abrupt,

due to opening of the Straits of Mackinaw. This probably resulted

i@n shoreline and dune development several hundred yards west of

the position of the present shoreline. Then the lake continued to

drop as the till was eroded in the Straits until another relative-

ly stable water level was achieved corresponding to the level of

bedrock in the Straits. Another shoreline and dune system can be

postulated corresponding to this pause in the receding lake level.

Finally, as the soft breccia bedrock was eroded away, the level

continued to drop in the northern part of the basin to a low of

approximately 230'. The southern portion of the basin did not

96 -

drop below 336' due to elevated land (Valders moraine), now be-

neath the lake, near Grand Haven. Extensive dune systems may have

formed over much of the land area laid bare by the receding levels

of Lake Chippewa. It is reasonable to suppose that the pre-Chip-

pewa dunes presently occurring along the lakeshore would have been

substantially wider and more extensive than at present during this

time, resulting in more extensive pocketing of drainage.

The channels of streams that cut through the pre-Chippewa

dunes were eroded much more deeply than those that remained im-

peded. During the Chippewa drawdown, the position of the receding

lakeshore was dependent on the bottom contours. In general, the

southern third of the lake (south of the Valders moraine) had a

more gentle gradient than the northern shore. This gradient needs

to be considered when estimating stream channel cutting. Where

the lake was shallow for a long distance out from the shore, the

gradient may not have increased when the waters receded even

though the total relief of the watershed would have.

As the water levels rose, all pre-Chippewa dunes below water

level were washed away by wave action, leaving only the oldest,

most landward portions of the Main Algonquin of Michigan and some

older dunes. It is probable that few of the formerly pre-Chippewa

dune-pocketed drainage systems remained impeded throughout this

period (essentially only the eleven that we have identified as'

Type R). However, because the erosional gradient had been reduc-

ed, there was little change in form of these drainage systems.

97 -

As the lake receded once more from 605' to the present 580'

level, Nipissing and Algoma dunes formed by reworking the sand of

the previously more extensive pre-Chippewa dunes. In many in-

stances, drainage channels that had breached the dunes during the

605' minimal dune stage remained open through the developing

Nipissing dunes. Post-Chippewa dunes did form at the mouth of

some drainage systems during this time, such as those on the south

side of Little Sable Point where the lower ends of the pre-Nipiss-

ing drainage channels were cut below lake level. Although we

treated these as dune-pocketed wetland watersheds, we recognized

that they were different from other drainage systems in this

class. They are really drowned river mouths that have subsequent-

ly been blocked by post-Chippewa dune development over the filled

channel. These are all very small watersheds for the amount of

relief that they display, so there was re'latively little runoff to

breach the developing post-Chippewa dunes.

Understanding is the key to effective management of our

resources. It is also vitally needed for realistic preservation,

which is actually a form of management. What to save and what to

develop are issues that land managers face constantly -- how to

control development that does occur in order to preserve our

natural heritage and yet use the resources of the earth in a

manner consistent with the total well-being of present and future

generations. In order to make responsible decisions regarding

these resources it is necessary to understand them and to have an

98 -

adequate inventory. Both dunes and wetlands are currently of

great concern. Both are important. The dunes are a geologic

resource formed under very different conditions than those of the

present day. They cannot be replaced. Some wetlands are geomor-

phically and developmentally related to the dunes and to the whole

shoreline process through geologic time. Seen in this context the

shorelines, dunes, and watersheds form a series of development-

ally, structurally, and functionally integrated systems along

eastern Lake Michigan.

VIII. SUMMARY

The Chippewa drawdown, a major lowering of Lake Michigan

between Lakes Chicago and Nipissing, was very significant in the

development of shoreline drainage systems. This event, in combin-

ation with littoral drift and dune formation, explains much of the

present character of the eastern Lake Michigan shoreline and

adjacent inland areas. We have described the development of

several different types of drainage systems occurring along the

shoreline and have inventoried the dune-pocketed wetlands.water-

shed type and two subtypes (pre-Chippewa and post-Chippewa). In

the area south of significant crustal rebound, running from the

Indiana/Michigan state line north to Frankfort in Benzie County,

these landlocked watersheds are found behind the dunes in several

areas. We recognized and described eighteen dune-pocketed wet-

lands watersheds. Field data were collected on the vegetation and

other characteristics of several representative sites of each

drainage system type. Watershed boundaries of all dune-pocketed

wetlands watersheds have been traced onto USGS topographic base

maps in a form suitable for digitizing and entry.into the Michigan

Resource Information System.

100

IX. LITERATURE CITED

Buckler, W.R. 1979. Dune type inventory and barrier dune
classification study of Michigan's Lake Michigan shore.
Michigan Department of Natural Resources, Geological Survey
Division. Report of Investigation 23. 31 pp.

Calvert, James L. 1946. The glacial and post-glacial history of
the Platte and Crystal Lake depressions, Benzie County,
Michigan. Occasional papers on the Geology of Michigan.
Publication 45. Geological Series 38. Published of State of
Michigan Dept. of Conservation. p. 83-153.

Cressey, G.B. 1928. The Indiana Sand Dunes and Shore Lines of
the Lake Michigan Basin: Geological Society of
Chicago. Bulletin No. 8, p. 13-73.

Evans, O.F. 1937. Origin of the coastal lakes of 'western
Michigan. Geological Review 27(l): 136-137.

Farrand, W.R. 1082. Quarternary Geology of Southern Michigan
(Map). Department of Geological Science, Univ. of Michigan.

Goff, F. Glenn, S.P. Voice, J.R. Wells, and J.A. Lively. 1986.
Environmental Study of the Hamlin Lake/Nordhou.se Dunes Area
Mason County, Michigan. Report to Michigan Department of
Natural Resources (contract 3193). Lansing, MI 391pp.

Goff, F. Glenn, R.O. Skoog, and James A. Lively. 1987.
Supplement to the January 1986 Environmental Study of the
Hamlin Lake/Nordhouse Dunes Area --- Mason County, Michigan.
Report to Michigan Department of Natural Resources (contract
3193b). Lansing, MI 65pp.

Guyer, Gordon E. 1987. Conclusive Findings and Determination of
the Supervisor of Wells in re: Hydrocarbon Development,
Nordhouse Dunes Area, Mason County, Michigan. Michigan
Department of Natural Resources. Lansing, MI. lopp.

Hazlett, Brian T. 1986. The Vegetation and flora of the
Nordhouse Dunes, Manistee National Forest, Mason County,
Michigan Part I. History and present Vegetation. Michigan
Botanist 25(2): 74-92.

Hough, J.L. 1958. Geology of the Great Lakes. Univ. of Illinois
Press Urbana, Illinois. 313pp.

Kelley, R.W. (Ed.) 1962. Sand dunes of Michigan: Michigan
Department of Conservation. Geological Survey Division,
2 sheets.

101

Larsen, Curtis E. 1985a. A stratigraphic study of beach features
on the southeastern shore of Lake Michigan: New evidence for
Holocene lake level fluctuations: Illinois State Geological
Survey. Environmental Geology Notes 112. 31pp.

Larsen, Curtis E. 1985b. Lake level, uplift, and outlet incision
the Nipissing and Algoma Great Lakes. In Karrow, P.F. and
Calkin, P.E. (Eds). Quarternary evolution of the Great
Lakes: Geological Association of Canada Special Paper 30.
p. 63-78.

Larsen, Curtis E. 1987. Geological History of Glacial Lake
Algonquin and the Upper Great Lakes. U.S. Geological Survey
Bulletin 1801. 36pp.

MDNR 1976. Sand Dune Designated Areas. Michigan Department of
Naural Resources. Geological Survey Division. Lansing, MI.
42pp.

MDNR 1978. Barrier Dune Formation Areas. Michigan Department
of Natural 'Resources. Geological Survey Division. Lansing,
MI. 90pp.

MNFI 1986. Natural Area Inventory of Designated Sand Dune Areas
in Michigan. Michigan Natural Features Inventory. Lansing,
MI. 46pp.

Olson, Jerry S. 1958. Lake Michigan dune development. Journal
of Geology 66: 254-263; 345-351; 473-483.

Powers, W.E. 1958. Geomorphology of the Lake Michigan Shoreline,
Final Report of Contract No. Nonr-1228(07), Project No.
NR387-015, Geograpy Branch, Earth Science Division, Office
of Naval Research, United States Department of the Navy,
103pp.

Reuter, D. Dayton. 1986. Sedge meadows of the upper midwest:
A Stewardship Summary. Natural Areas Journal 6 27-34.

Scott, I.D. and K.W. Dow 1932. Dunes of the Herring Lake
Embayment Michigan: Papers Michigan Acad. Science, Arts, and
Letters, Vol. 22, p. 473--450.

Tague, Glenn C. 1946. The post-glacial geology of the Grand
Marais Embayment, Berrien County, Michigan. Occasional
Papers on the Geology of Michigan. Publication 45.
Geological Series 38. Published by State of Michigan
Department Of Conservation. p. 1-82.

Wells, James R., Paul W. Thompson, and F.D. Sheldon. 1975.
Vegetation and geology of North Fox Island, Lake Michigan.
Michigan Botanist 14: 203-214.

- 102

Wells, James R., and Paul W. Thompson. 1980. Vegetation analysis
of the Martin-Marietta Aggregates site, Berrien County
Michigan. (unpubl.) Report to Mich. Dept. Nat. Res. 68 pp.

Wells, James R., and Paul W. Thompson. 1981a. Ecological and
floristic survey of the Rosy Mound Sand Dunes Tract.
(unpubl.) Rept. to Mich. Dept. Nat. Res. 37 pp.

Wells, James R., and Paul W. Thompson. 1981b. Ecological and
floristic survey of the Ferrysburg sand dunes tract.
(unpubl.) Rept. to Mich. Dept. Nat. Res. 33 pp.

Wells, James R., and Paul W. Thompson. 1981c. Ecological and
floristic survey of the Gilligan Lake sand dunes tract,
Consumer's Power property. (unpubl.) Rept. to Mich. Dept.
Nat. Res. 61 pp.

Wells, James R., and Paul W. Thompson. 1982a. Plant communities
of the sand dunes region of Berrien County, Michigan. Mich.
Bot. 21: 3-38.

Wells, James R., and Paul W. Thompson. 1982b. Ecological and
floristic survey of the Sturgeon Bay sand dunes tract.
(unpubl.) Rept. to Mich. Dept. Nat. Res. 56 pp.

Wells, James R., and Paul W. Thompson. 1983. Ecological and
floristic survey of P. J. Hoffmaster State Park, Ottawa
and Muskegon Counties, Michigan. (unpubl.) Rept. to Mich.
Dept. Nat. Res. 64 pp.

Wells, James R., and Paul W. Thompson. 1984. Ecological
observations on the Nordhouse Dunes, Mason County, Michigan.
Report to Michigan Department of Natural Resources. Contract
No. 9DPMA-84-A. 45pp.

102

APPENDIX 1.

Notes on the vegetation and ecology of
wetlands sites examined during the summer of
1987. Examples of several different types of
watersheds are included, with descriptive
information, maps, and photographs.

APPENDIX I
SITE-BY-SITE VEGETATION NOTES

The vegetation and site descriptions in this Appendix were
developed during the summer of 1987. The purpose of the site
examinations was to assist in the development of a shoreline
drainage system classification system. Three different levels of
field examination were employed.

Level I: General Scouting -- The purpose of this level of
observation was to gain general impressions of the wetland
watershed areas by driving perimeter roads or hiking through each
area scouted. Notes on drainage, topography, soils, cover types
and development were made and referenced to the USGS topographic
maps. Significant features were documented by narrative field
notes and photographically.

Level II: -- Specific Site Description -- The purpose of this
level of observation was to locate a predetermined candidate site
of interest (e.g. a catchment basin, drainage divide, upgradient
wetland, etc.) and describe its characteristics. Photos were
taken in most instances. A narrative description was written to
summarize the area's significance. Vegetation cover type was
noted and dominant species were recorded. Initially, based on the
hypothesis that certain variables would prove to be important in
distinguishing and understanding these wetlands, each was ranked
subjectively on three scales using vegetation community
composition and other indicators: (A) a mineral enrichment scale
[0 = ombrotrophic to 10 = minerotrophic], (B) a closure/succession
scale [0 = open water to 10 lowland forest], and (C) a surface
water variability scale [0 constant seasonal water level to 10
highly variable water level between early spring and late summer].
The basis of these scales is described more fully in the "Wetland
Vegetational Processes" section of the full report (Section IV).

Level III -- Vegetation/Species Survey -- This level of
observation was made for certain sites that seemed to typify the
wetland drainage type classes. It consisted of a Level II
observation plus a timed meander species search within the general
location of the identified site. Rather than keying unknown
specimens on site, they were collected, labelled and later pressed
and identified.

We were particularly interested in characterizing the dune-
pocketed wetlands watersheds. However, we recognized early that
it would not be possible within the time and cost of this study to
complete a systematic field examination of all examples of this
watershed type. Indeed, at that time we were not even sure which
of the wetland areas occurring along the shoreline area were
members of the logical set termed "dune-pocketed wetlands
watersheds". In order to make such a determination it was
necessary for us to study examples of several of the wetland
watersheds, observe and record their characteristics and then use
this data in conjunction with map studies and other information to

Al -- I

formulate a comprehensive classification. Several dune-pocketed
wetlands watersheds are included among the watersheds examined in
the field. Because we could not field examine all of the sites
along the entire shoreline within the time available, we
concentrated on a section of the shoreline from the Fruitvale
Dunes in northern Oceana County north to Bar Lake Swamp in
Manistee County. This shoreline section contains a range of
variation in shoreline drainage system types, as well as dune-
pocketed wetlands watershed types. The following site
descriptions are presented in order from south to north. The
names and code designations are the same as those used in Section
V of the main report. Watershed boundary maps are included.

A.1 Fruitvale South Dune-Pocketed Wetlands Watershed --
Photo 1 shows the vegetation structure in a catchment at the base
of the dunes in the Fruitvale South dune-pocketed wetlands
watershed system. This catchment (Map 1, location 1) is an
elongated area (several hundred meters long and only about 100
meters wide) with its long axis oriented north and south parallel
to the face of the dunes. The photograph was taken viewing
westward across the wetland toward a larger blowout ridge which
appears at the horizon of the photo. This dune is apparently
eroding and encroaching on the dune-pocketed wetland marsh. The
marsh is a sedge meadow-shrub carr combination. Soils are organic
at the surface and become progressively wetter toward the center
of the wetland where there is standing water covered with duckweed
(Lemna minor). Dominant shrubs are Cephalanthus occidentalis (up
to aproximately 5 inches in diameter), Ilex verticillata, Cornus
amomum, Spiraea tomentosa, Salix lucida,-and Salix discolor.
Herbaceous species include Iris virginica, Lycopus uniflorus,
Ludwigia palustris, Scirpus cyperinus, Carex crinita, Typha
latifolia, Dulichium arundinaceum, Leerzia oryzoides, Bidens sp.,
Osmunda cinnamomea, Osmunda regalis, Thelypteris palustris, and
Calamogrostis canadensis. Around the margins of the wetland,
Salix nigra occurs, with occasional Populus tremuloides and Acer
rubrum.

Photo 2 shows the ponding that occurs in the southern end of
the Fruitvale South dune-pocketed wetlands area. The pond (Map 1,
location 2) contains yellow water lily (Nuphar vuriegata). The
dominant shrub cover around the margin of the open water is button
bush (Cephalanthus occidentalis). Other species in this ecotone
zone are Polygonum amphibium, Polygonum hydropiperoides, and Carex
crinita. There are fish in the pond (suggesting that it is a
relatively permanent feature) even though it is only a couple of
feet deep. The wetland is blocked on the lower end by a dune
ridge.

Photo 3 shows the structure of vegetation in a small wetland
area (Map 1, location 3) to the east of a Iow ridge separating the'
catchment at the base of the high dunes from inland areas.
Throughout the area numerous wetland shrub carrs and marshes
similar to this example are found. These are separated from one
another by low ridges, possibly inland dune features. Apparently

Al -- 2

the lowland forest/shrub carr vegetation that previously occurred
in many of these areas was killed by flooding perhaps twenty-
thirty years ago. There are dead Thuja occidentalis and
Cephalanthus occidentalis in most of them. Low sprouts of
Cephalanthus are now becoming re-established and succession to
this and other shrub species is occurring. Other species include
Cicuta maculata, Bidens sp., Ludwigia palustris, Ilex
verticillata, Salix discolor, Solanum dulcamara, Sium suave,
Lycopus asper, Osmunda cinnamomea, Osmunda regalis, Thelypteris
palustris, Carex sp., Calamogrostis canadensis, and Iris
virginica. There is one small Thuja occidentalis and a few Nyssa
sylvatica trees along the edge of the wetland area; otherwise
lowland forest is limited. The upland forest comes right down to
the edge of the sedge meadow/shrub carr. The surrounding forest
is comprised primarily of hemlock, red maple, red oak, white pine,
black oak, black gum, and yellow birch.

Photo 4 shows a successionally more advanced shrub carr area
(Map 1, location 4) presently succeeding to lowland forest.
Dominant species are Salix discolor, Salix lucida, Pinus strobus,
Acer rubrum, and Betula papyrifera.

A.2 Horseshoe Lake wetland area -- The Horeshoe Lake
catchment area (Map 1, location 5) located relatively far inland
in the Fruitvale Dunes System was examined and found to be of
glacial origin -- part to the Fruitvale Creek watershed. This
wetland displays aquatic to upland zonation and also a gradient
from ombrotrophic peatland type vegetation in its upper (northern)
end (Photo 5a), to a much more minerotrophic aquatic and marsh
vegetation in its lower end (Photo 5b). The northern end of the
Horseshoe Lake wetland area is less influenced by mineral-rich
groundwater than the lower end. Consequently, the vegetation is
more ombrotrophic, with Chamaedaphne calyculata, Osmunda
cinnamomea, and Scirpus cyperinus dominant. Around the outer
edges, buttonbush (Cephalanthus occidentalis) occurs.
In the deepest part of the lower portion there is standing water
perhaps two feet deep, with Nuphar variegata and Potamogeton sp.
Around the edge of the open water there is Polygonum amphibium and
then a zone of Scirpus cyperinus dominated sedge meadow.
Cephalanthus occidentalis, Chamaedaphne calyculata, and Rosa
palustris form an outer zone of shrub cover. At the edge of the
forest a narrow zone of Nyssa sylvatica occurs, and this gives way
quickly LU up.Luxia species including Pinus strobus, Acer rubrum,
Quercus borealis, and Fagus grandifolia.

To the east of the Horseshoe Lake wetland another small
minerotrophic wetland area was examined. Species recorded in the
wetter part of this area include Typha latifolia, Scirpus
cyperinus, Osmunda cinnamomea, Matuccia strutheriopteris,
Impatiens pallida, Carex lacustris, Glyceria striata, Ilex
verticillata, Salix discolor, Cephalanthus occidentalis, Betula
papyrifera, Thuja occidentalis, and Tsuga canadensis. Around the
edges there is Acer rubrum, Punus strobus, Fraxinus americana,
Prunus serotina, and Betula lutea. This area is wetter at the

Al -- 3

surface than the leatherleaf portion of the Horseshoe Lake wetland
located immediately to the west across a separating ridge. After
careful examination of the Horseshoe Lake area including several
marshes, we concluded that it is a glacial landscape with no
direct relationship to the dunes to the west.

A.3 Fruitvale North Dune-Pocketed Wetlands Watershed -- This
watershed is located immediately to the north of the Fruitvale
south dune-pocketed wetlands area (in fact the two could be
considered as subunits of a single watershed). Fruitvale north is
an exemplary dune-pocketed wetlands watershed of the pre-Chippewa
type (Type B, see main report). Figure 10 of the main report
presents a panoramic view of the catchment of the watershed and
Map I (location 6) shows its location. Species in the primary
catchment area include: Bidens sp., Polygonum hydropiperoides,
Pilea pumila, Carex sp., Polygonum amphibium, Fraxinus
pensylvanica, Hypericum punctatum, Salix petiolaris, Cyperus
esula, Mentha arvensis, Cephalanthus occidentalis, Ludwigia
palustris, Hordeum jubatum, Salix nigra, Eragrostis spectabilis,
Eleocharis elliptica, Rorippa islandica, Cornus amomum, Eupatorium
perfoliatum, and Salix lucida.

Upgradient from this primary catchmemt area there is a series
of low shrub carr and marsh openings separated by low dunes or
beach ridges (Map 1, location 7). Some of these have open water
surrounded by buttonbush (Cephalanthus occidentalis) and other
shrubs (Photo 6a) while others are more marshy, with cattail
(Typha latifolia) and a variety of sedges, other herbaceous
species, and shrubs (Photo 6b). Many of-the species listed above
within the primary catchment area also occur in these upgradient
marshes. Additional species recorded include: Sassafras albidum,
Iris virginica, Cornus stolonifera, Ilex verticillata, Quercus
borealis, Juncus effusus, Salix discolor, Scirpus cyperinus, and
Acer rubrum.

A.4 Winston Road Wetland Area -- South of Long Lake we
examined a wetland area (Map 2, location 1) that is located close
to the shoreline but is apparently a headwater area for an inland
un impeded stream drainage system (Flower Creek). Casual
observation of the flora and vegetation revealed no differences
between this and dune-associated wetland areas. Photo 7 shows the
general appearance of the open shrub carr vegetation community in
this area.

A.5- Long Lake -- The Long Lake dune-pocketed catchment is
shown on Map 2 (location 2). Viewing eastward up Long Lake
(Photo 8) a large bed of yellow water lily (Nuphar variegata) is
observable at the eastern end of the open water zone. East of
that is cattail (Typha latifolia) and a variety of sedges, grasses
(e.g. Calamogrostis canadensis) and herbaceous species forming a
marsh zone. Around the edges of the basin there is a narrow strip
of lowland trees (e.g. Salix nigra) and shrubs (e.g. Cephalanthus
occidentalis). The upland forest cover is Pinus strobus, Quercus
borealis, and Quercus velutina. A noteworthy characteristic of

Al -- 4

this landlocked lake/wetland is the narrow lowland strip.
Vegetation goes abruptly from catchment (open water, open marsh)
to uplands.

A.6 Skinny Lake -- "Skinny" Lake is a wetland located in a
long east-west trough that is blocked by dunes on its western end
(Map 2, location 3). It is apparently similar in its genesis to
both Long Lake to the south and Bigsbie Lake to the north, but was
a shallower ravine and is more filled in. The catchment is narrow
and there is abrupt transition to upland, with almost no lowland
zone (Photo 9). Species recorded include: Lemma minor,
Calamogrostis canadensis, Betula papyri-fera, Typha latifolia,
Scirpus cyperinus, Sparganium eurycarpum, Scirpus validus, Nuphar
variegata, Polygonum amphibium, Carex stricta and Sassafras
albidum.

A.7 Bigsbie Lake Dune-Pocketed Watershed -- The Bigsbie Lake
watershed is a classical example of a Type A (post-Chippewa) dune-
pocketed area (see Figure 10 in main report).-- Map 3 (location 1)
shows the location, and Photo 10 shows the wetland area at the
head of the lake. The lake is between glacial hil.ls but is
blocked on its west end by dunes. Species present include Salix
petiolaris, Salix discolor, Thelypteris palustris, Cephalanthus
occidentalis, Nuphar variegata, Najas flexilis, Calamogrostis
canadensis, and Typha latifolia. There are a few Salix nigra
trees around the edge of the lake and wetland, but the open
catchment essentially gives way directly to upland.forest Jagus
grandifolia, Acer rubrum, Tsuga canadensis, Betula lutea, Quercus
borealis, etc.) with no lowland zone present.

A.8 Little Silver Lake -- The Little Silver Lake area (Map
4, location 1) was investigated as a possible dune-pocketed
wetland. There is a small minerotrophic pond. Although the
entire area is sandy, there is a thin layer of muck at the soil
surface. Blue herons frequent the area. Little Silver Lake is
beparated from Silver Lake by a low dune ridge. These were
apparently a single water body at one time. The shallow open
water of Little Silver Lake is surrounded by a marshy sedge meadow
zone and then shrub carr (Photo 11). Species recorded include:
Nuphar variegata, Bidens sp., Juncus effusus, Carex stricta,
Lythrum salicaria, Sparganium eurycarpum, Leerzia oryzoides, Rosa
palustris, Cephalanthus occidentalis, Salix discolor, Salix
lucida, Onoclea sensibilis, Osmunda cinnamomea, Prunus serotina,
Populus tremuloides, and Quercus velutina. Little Silver Lake is
best considered as an interdunal area -- part of the Silver Lake
baymouth bar complex.

A.9 Cedar Point Park Wetland -- Cedar Point Park wetland
area in Section 4 - T15N - RI8W (Map 5, location 1) was examined
as a possibly dune-associated wetland area. However, upon field
inspection and further map analysis this area appears to be a
glacial lake which just happens to occur in the shoreline zone.
A species list is given in Table 1. It is a small, relatively
species poor, bog lake (Photo 12) surrounded by leatherleaf

Al -- 5

Table 1. Species list from Cedar Point Park wetland area sampled
August 5, 1987 by F. Glenn Goff. The list includes species
occurring in a wet forest pocket (Photo 12) in the glacial (non-
dunal) landscape in Section 4 T15N - R18W. Sample employed the
timed meander-search procedure (Goff, Dawson and Rochow, 1982)
with a total search time of 30 minutes.

Acer rubrum
Aronia melanocarpa
Betula papyrifera
Brassenia schreberi
Carex canescens
Carex oliosperma
Cephalanthus occidentalis
Chamaedaphne calculata
Dulichium arundinaceum
Eleocharis elliptica
Eleocharis erythropoda
Gaylussacia baccata
Hypericum ellipticun
Iris virginica
Juncus canadensis
Lycopus uniflorus
Mattucia strutheriopteris
Nemopanthus mucronata
Nuphar variagata
Nyssa sylvatica
Pinus resinosa
Pinus strobus
Populus tremuloides
Pteridium aquilinum
Quercus ellipsoidalis
Ranunculus sp:
Scirpus cyperinus
Sphagnum sp.
Triadenum virginicum
Utricularia minor
Vaccinium angustifolium

Al 6

(Chamaedaphne calyculata), tamarack (Larix laricina), and white
pine (Pinus strobus). There is buttonbush (Cephalanthus
occidentalis) and black gum (Nyssa sylvatica) around the outer
edge of the mote.

A.10 Ridge Road Wetland -- The small wetland areas just
southeast of the Pentwater Dune System (Map 5, location 2) are of
limited extent and mostly successionally advanced to the stage of
lowland forest, although small patches of more open shrub and
occasional sedge-fern vegetation are also found. This area was
included in the study because it superficially resembles much of
the impeded drainage lowland forest in the Fruitvale area to the
south, the Nordhouse Dunes tract to the north, and elsewhere along
the Lake Michigan shoreline. It is included as an example of a
type of vegetation that forms in poorly drained glacial areas
where seasonal flooding is sufficient to retard forest development
but not sufficient to set succession back to the shrub carr or
sedge meadow stadge.

Photo 13 shows the general character of the vegetation in the
Ridge Road Wetland area. The photo shows Jim Lively standing in a
young lowland forest of Acer rubrum and Fraxinus pensylvanica.
Trees are mostly small and unhealthy -- the result of periodic
flooding. The canopy is densely closed in some areas and open in
others. In open patches there are usually direct signs of
seasonal flooding (e.g. leaves matted by flood waters, debris
stranded on lower portions of trees and shrubs, dried algae crust
on the soil surface). Soils are generally black and high in
organic matter to several inches in depth. In these open areas
the herbaceous flora is typical of northern wet hardwood forests,
including: Osmunda cinnamomea, Calamogrostis canadensis, Lycopus
uniflorus, Viola conspersa, Carex intumescens, Osmunda regalis,
Onoclea sensibilis, Thelypteris palustris, Glyceria striata, Carex
crinita, Carex canescens, Scutellaria lateriflora, and Coptis
trifolia. Shrub species occuring more abundantly in the wetter
openings include: Ilex verticillata, Aronia melanocarpa, Alnus
rugosa, Spiraea alba, and Nemopanthus mucronata. Old hemlock
(Tsuga canadensis) and yellow birch (Betula lutea) trees occurring
on slightly higher topography throughout the area suggest that
despite its second-growth appearance, the vegetation mosaic of
this area represents a stable plant community at equilibrium with
its environment -- i.e. with seasonal flooding setting back
succession periodically in the lower, wetter areas to mainatin the
community in a late shrub carr -- early lowland forest
successional stage. This is not a dune-pocketed wetlands area.

A.11 Spartina Marsh -- The vegetation and ecology of the
Nordhouse Dunes area has been extensively described elsewhere
(Hazlett, 1986, Wells and Thompson 1984, Goff, et al 1986, 1987).
This area has recently been officially designated as a wilderness
area, the only wilderness in the Lower Peninsula. It was on the
basis of our earlier review and investigations of the landward
dunes associated with the Nordhouse Dunes that we recognized the
existence of the category of wetlands termed "landward barrier

Al -- 7

dune-pocketed wetlands", the principal focus of this study and
inventory. However, while we recognize the existence of two
completely pocketed wetland watersheds within the Nordhouse Dunes
subsystem (Goff, Skoog and Lively, 1987) (Map 6, locations I and
2), because of the extensive work already done in the Nordhouse
System and owing to the more general objectives of this study, we
examined the flora of only the catchment portion of one of these
(Map 6, location 2) -- the Spartina Marsh dune-pocketed watershed
catchment area near the campground of the Lake Michigan Recreation
Area. Photo 14 shows the structure of vegetation in this
catchment area. Photo 14A shows Jim Lively standing in the grassy
sedge meadow near the center of the catchment (indicating a nearly
dead clump of Salix petiolaris). Clumps of dead or near dead
Salix petiolaris appear on the right side of the photo and are
found throughout the area. In photograph 14B Jim is indicating
the seasonal high water mark that is obvious on trees throughout
this area. in the early spring, water accumulates in this
catchment to a depth of about one meter. Then, apparently by
seepage into the dune sand, the water level drops slowly but
steadily until, by mid- to late-summer the area is completely dry.
This area, typical of such dune-pocketed wetland sites, acts as a
natural runoff catchment basin. It is surrounded by open xeric
forest with jack pine (Pinus banksiana), Hill's oak (Quercus
ellipsoidalis), trembling aspen (Populus tremuloides), and other
species typical of northern Michigan dry forests. Table 2
provides a species list for the Spartina Marsh catchment area.
.Only twenty-nine species were found.

A.12 Cooper Creek Road -- This is a-small leatherleaf
(Chamaedaphne calyculata) peatland area (Map 6, location 3).
Because of its location, apparently on the inland margin of the
high dunes, we examined it as a potential dune-pocketed wetland.
The entire wetland (peatland) area is only about 2 acres in
extent. It is surrounded on all sides by dunes although the dunes
are lower on the eastern side. We called it "semi-interdunal" in
the field and now consider it an interdunal type. A list of the
species found in the area is included in Table 3. The mineral
enrichment value was estimated at 1.0; the wetland succession
value at 5.5; and the water constancy value at 8.0. Photo 15
shows the general appearance of the vegetation in the area. It is
apparently a small paludified peatland area. White pine (Pinus
strobus) grows throughout.

A.13 Manistee-Mason County Line Watersheds -- Two
contrasting basins that occur behind the barrier dunes in the
vicinity of the Mason/Manistee County Line (Map 7, locations I and
2) depict very well the contrast between groundwater recharged
("spring-fed") cedar swamp vegetation and vegetation of the
landlocked, seasonally flooded dune-pocketed wetlands catchment.
The northernmost of these contrasting watersheds (Map 7, location
1) has an outlet to Lake Michigan (Map 7, location 4). The outlet
stream flows westward from the surrounding glacial hills to the
front of the dunes and then turns to the north, finally entering
Lake Michigan. Although the stream course has been altered by the

Al -- 8

Table 2. Species list from the Spartina Marsh area in the Big
Sable -- Nordhouse Dunes System. The list includes species
occurring in the seasonally inundated catchment area at the base
of the barrier dunes. By F. Glenn Goff, July 30, 1987. This
species list was composed using the timed meander-search procedure
(Goff, Dawson and Rochow, 1982) with a total search time of 30
minutes.

Andropogon gerardi
Aster lateriflorus
Calamogrostis canadensis
Carex buxbaumii
Carex cryptolepis
Carex lanuginosa
Eleocharis acicularis
Eleocharis elliptica
Fraxinus americana
Fraxinus pensylvanica
Hypericum kalmianum
Hypericum punctatum
Juncus acuminatus
Juncus canadensis
Lycopus uniflorus
Mentha arvensis
Oenothera perennis
Panicum implicatum
Panicum virgatum
Pilea pumila
Populus tremuloides
Salix petiolaris
Sassafras albidum
Scirpus cyperinus
Scutellaria lateriflorus
Spartina pectinata
Spiraea alba
Thelypteris palustris
Viola lanceolata

Al 9

Table 3. Species list from Cooper Creek Road peatland -- a small
leatherleaf peatland pocket. By F. Glenn Goff, August 5, 1987.
This species list was composed using the timed meander-search
procedure (Goff, Dawson and Rochow, 1982) with a total search time
of 30 minutes.

Acer rubrum
Aronia melanocarpa
Betula papyrifera
Carex sp.
Chamaedaphne calyculata
Eriophorum spissum
Gaylussacia baccata
Glyceria canadensis
Ilex verticillata
Juncus nodosus
Larix laricina
Mathicia strutheriopteris
Menyanthes trifoliata
Nemopauthus mucronata
Pinus resinosa
Pinus strobus
Quercus borealis
Sphagnum sp.
Triadenum virginanum
Vaccinium angustifolium
Vaccinium macrocarpon
Vaccinum myrtelloides

Al 10

dunes, the vegetation is cedar swamp throughout and there is
little evidence of seasonal flooding, although there is apparently
some impedance of flow due to the dunes.

Photo 16 (a & b) show the vegetation conditions that are
typical of this area. Photo 16a shows Jim Lively standing in one
of the headwater spring areas. The vegetation here is
predominantly tamarack (Larix laricina) woods with cattail (Typha
latifolia) dominated sedge-meadow areas in the more open areas.
Conditions lower in the watershed are depicted by Photo 16b. This
is a typical northern Michigan wet cedar swamp community. The
moisture supply appears to be predominantly from groundwater and
is apparently relatively constant through the year. Although the
area is wetter during the spring months, it does not become
seasonally flooded as does the dune-pocketed wetlands catchment
areas. If flooding does occur it is apparently temporary. Soils
are wet, mucky, and in places there is shallow standing water at
the surface with floating duckweed (Lemna minor). There are many
down cedar trees and profuse white cedar regeneration suggesting a
low deer population. The stream flows westward to the base of the
dunes, then it meanders northward for over a half-mile to its
confluence with Lake Michigan. Table 4 provides a listing of the
species found in the cedar swamp area.

To the south there is a low divide separating the cedar swamp
area from a dune-pocketed catchment area (Map 7, location 2) that
apparently drains to the south. These two watersh'ed units merge
together at the base of the dunes (Map 7, location 3), making it
difficult to tell exactly where the divide occurs. Proceeding
southward along the landward base of the dunes, cedar swamp
gradually gives way to shrub carr vegetation dominated by Salix
petiolaris and then to an open area that was obviously flooded
during the spring and earlier in the summer. Photo l7a@ shows Jim
Lively standing in the formerly flooded area of the catchment, and
Photo l7b is a closeup of the ground surface showing snail shells
and dragonfly larvae -- evidence of the earlier flooded condition.
Vegetation in the open area consists primarily of patches of
Polygonum amphibium and Proserpinaca palustris. There are also
small patches of sedge meadow within the shrub carr. A species
list made in the shrub carr, sedge, open meadow, and catchment
area is provided as Table 5. This area is apparently flooded in
the spring to a depth of a meter or more. Almost all of the water
that it receives obviously comes in the form of surface runoff.
Due to the lack of groundwater influence, the area dries by mid-
to late-summer. This annual cycle of flooding and drying, which
appears to be typical of the post-Chippewa barrier dune-pocketed
wetlands, provides suitable habitat for relatively few species,
many of which are amphibious plants (e.g. Polygonum amphibium,
Polygonum hydropiperoides, Proserpinaca palustris) and facultative
wetland species (e.g. Rorippa islandica, Lysimachia terrestris,
Lobelia cardinalis). The almost ubiquitous occurrence of Salix
petiolaris in the dune pocketed wetland catchment areas is also
noteworthy. Farther to the south in the state this species
becomes less frequent and buttonbush (Cephalanthus occidentalis)

Al -- 11

Table 4. Species list from the spring fed cedar swamp watershed
in the Mason-Manistee County Line Dune System. By F. Glenn Goff,
July 30, 1987. This species list was composed using the timed
meander-search procedure (Goff, Dawson and Rochow, 1982) with a
total search time of 30 minutes.

Acer rubrum
Alnus rugosa
Betula lutea
Bidens cernua
Caltha palustris
Cardamine pensylvanica
Carex bebbii
Carex disperma
Carex hystericina
Cinna latifolia
Coptis trifolia
Cornus stolonifera
Epilobium ciliatus
Eupatorium maculatum
Fraxinus nigra
Gaultheria hispidula
Geum rivale
Glyceria striata
Ilex verticillata
Impatiens pallida
Juncus brachycephalus
Juncus canadensis
Larix laricina
Lemna minor
Lobelia cardinalis
Lobelia siphilitica
Ludwigia palustris
Lycopus asper
Maianthemum canadense
Mentha arvensis
Onoclea sensibilis
Osmunda regalis
Pilea pumila
Ranunculus abortivus
Rhamnus alnifolia
Ribes americanum
Rubus pubescens
Scutellaria lateriflora
Solanum dulcamara
Solidago rugosa
Thelypteris palustris
Thuia occidentalis
Trientalis borealis
Vaccinium angustifolium
Viola incognita
Viola pallens

Al 12

Table 5. Species list for a dune-pocketed wetland area along the
southwestern edge of Mason/Manistee County Line Dune System. The
species list was made by F. Glenn Goff on July 30, 1987. It was
piled using the timed meander-search procedure of Goff, Dawson
and Rochow (1982) and represents a 30 minute search period.
com

Search includes margins as well as central portion of the
catchment area.

Agrostis alba
Alnus rugosa
Asclepias incarnata
Betula papyrifera
Carex bebbii
Carex lacustris
Carex stricta
Chamaedaphne calyculata
Cornus stolonifera
Epilobium ciliatum
Eupatorium maculatum
Eupatorium perfoliatum
Glyceria canadensis
Glyceria striata
Ilex verticillata
Iris virginica
Larix laricina
Lobelia cardinalis
Lycopus asper
Lysimachis terrestris
Osmunda regalis
Poa compressa
Polygonum amphibium
Polygonum hydropiperoides
Proserpinaca palustris
Quercus borealis
Rorippa islandica
Rosa palustris
Rubus pubescens
Salix discolor
Salix lucida
Salix nigra
Salix petiolaris
Scirpus cyperinus
Solidago altissima
Thelypteris palustris
Thuja occidentalis
Vitis riparia

Al 13

becomes more common, but Salix petiolaris was found in nearly
every dune pocketed catchment that we examined.

A.14 Bar Lake Swamp -- Bar Lake Swamp illustrates the
contrast in vegetation types between a groundwater maintained
("spring fed") swamp (Map 8, location 1) and a highly seasonal,
principally surface water recharged dune-associated wetland (Map
8, location 2). Bar Lake Swamp was formed when postglacial uplift
of the land drained a former bay-mouth bar lake. Along the
eastern side of the swamp glacial hills of reddish unassorted till
rise abruptly above the swamp. On the west side Bar Lake Swamp
abuts the barrier dunes.

A fresh-water stream runs from south to north along the
eastern side of the swamp. This stream is apparently maintained
by groundwater recharge from the hills to the east. The general
appearance of the cedar swamp community along this stream is shown
in Photo 18. This is a typical cedar (Thuja occidentalis) lowland
forest. Soils are muck. Numerous small openings occur in the
forest cover and in these patches a rich understory flora of
boreal species is found.:- A species list made on July 29, 1987 in
the cedar swamp area is provided as Table 6.

Along the western edge of Bar Lake Swamp, areas of sedge
meadow and shrub carr occur at the base of high dune ridges.
Like the pre-Chippewa dune-pocketed wetlands, these wetland sedge
meadow -- shrub carr areas are apparently seasonally flooded to a
depth of a meter or more judging from water marks and debris
stranded on trees and shrubs by the receding water. Although
this was formed as a bay-mouth bar lake system, along its western
side, at the base of the dunes, it is similar to the pre-Chippewa
dune-pocketed wetlands areas farther south. Like the dune-
pocketed wetlands, dammed by highly porous dune sands that impede
water movement only temporarily, this area de-waters during the
summer and may become very dry in the late summer months of low
rainfall years. The open herbaceous and shrub cover typical of
such wetland areas is the result of periodic disturbance flooding
and drought and is in marked contrast to the more successionally
advanced lowland forest cover found on the east side of the swamp
where moisture levels are more constant. The vegetation here is
apparently a stable community type, in the sense that it is at
equilibrium with the edaphic and disturbance conditions typical of
the landscape it occupies -- i.e., it is predictably maintained at
a relatively early stage of succession by frequent natural
extremes in water level and site moisture conditions.

Photo 19 shows the structure of the plant community. Jim
Lively is standing at the boundary between the open, sedge
dominated area in the foreground and the shrub (principally Salix
petiolaris) dominated area in the background. A few cattail
(Typha latifolia) fruiting stalks are shown on the left. Cattails
are rare in the area, apparently because of the very dry
conditions that may prevail in late summer. The species list for
the dune-pocketed wetland areas (Table 7) contrasts sharply with
the species found in the spring-fed swamp area to the east.

Al -- 14

Table 6. Species list from the spring fed cedar swamp portion
(east side) of Bar Lake Swamp, by F. Glenn Goff, July 29, 1987.
This species list was composed using the timed meander-search
procedure (Goff, Dawson and Rochow, 1982) with a total search time
of 30 minutes.

Abies balsamea Pilea pumila
Acer rubrum Polygonatum biflorum
Acer saccharum Prunella vulgaris
Adiantum pedatum Prunus serotina
Alnus rugosa Pyrola secunda
Amelanchier canadensis Quercus borealis
Anthoxanthum odoratum Ranunculus recurvatus
Aralia racemosa Ribes hirtellum
Arisaema triphyllum Rubus pubescens
Aster lateriflorus Rubus strigosus
Aster macrophyllus Sambucus canadensis
Athyrium filix-femina Solitlago patula
Berberis thunbergii Solidago rugosa
Betula lutea Thelypteris palustris
Betula papyrifera Thuja occidentalis
Carex arctata Tilia americana
Carex disperma Trientalis borealis
Carex leptonervia Tsuga canadensis
Carex retrorsa Ulmus americana
Cerasteum vulgatum Veronica americana
Cinna latifolia Viola incognita
Circaea alpina
Circaea quadrisulcata
Clintonia borealis
Coptis trifolia
Cornus alternifolia
Cryptotaenia canadensis
Cystopteris bulbifera
Rryopteris disjuncta
Dryopteris spinulosa
Epilobium coloratum
Epipactus helleborine
Fagus grandifolia
Fragaria virginiana
Fraxinus americana
Fraxinus nigra
Galium triflorum
Geum canadense
Glyceria striata
Impatiens pallida
Lonicera dioica
Lycopus asper
Maianthemum canandese
Mitchella repens
Mitella nuda
Onoclea sensibilis
Osmunda cinnamomea
Osmunda regalis

Al 15

Table 7. Species list for a dune-pocketed wetland area along the
western edge of Bar Lake Swamp, by F. Glenn Goff, July 29, 1987.
The list was compiled using the timed meander-search procedure of
Goff, Dawson and Rochow (1982) and represents a 30 minute search
period.

Acer rubrum Rumex obtusifolius
Alnus rugosa Salix discolor
Asclepias incarnata Salix glaucophylloides
Aster lateriflorus Salix nigra
Boehmeria cylindrica Salix petiolaris
Calamogrostis canadensis Scirpus fluviatilis
Carex alata Scirpus pendulus
Carex aquatilis Scutellaria lateriflora
Carex interior Smilicina stellata
Carex lacustris Taraxacum officinale
Carex rosea Thelypertis palustris
Carex sp Triadenum virginianum
Carex stricta Typha latifolia
Cicuta bulbifera Ulmus americana
Cornus amomum Vitis riparia
Cornus stolonifera
Dulichium arundinaceum
Eleocharis smallii
Eupatorium maculatum
Eupatorium maculatum
Fraxinus americana
Fraxinus pensylvanica
Galium concinnum
Galium tinctorium
Glyceria striata
Hypericum ellipticum
Ilex verticillata
Iris virginica
Juncus canadensis
Juncus effusus
Leerzia oryzoides
Lobelia cardinalis
Ludwigia palustris
Lycopus asper
Lycopus uniflorus
Myrica gale
Onoclea sensibilis
Osmunda regalis
Pilea pumila
Polygonum amphibium
Polygonum hydropiperoides
Populus tremuloides
Potentilla palustris
Proserpinaca palustris
Rhamnus alnifolia
Rhus radicans
Rosa palustris
Rubus pubescens

Al 16

z"r
Y,)
C3
0.,
E,!-L ==f
FIECK

/SY 5 r,
Y,
Meinert
PT
BAA Z// T.3
'9 7

F '@@L Z03
LE

-7

Y.9. Y,
Zoe 49

@A,

407

--F L

"V 9'
Z, I@h

T
MAP 1

Al 17

STONY
8m

v'JU

Township'

@-qo.5 r

JC
19 lyl.5 If

T lq
0.5 -.0 411" 7)0

r 7ZA

2015

V I

(C
A
c

6
MAP 2

vv@jl
L

C) C,

k3m rl
7

Al 18

32' 30" 538 539 540 86-,36;'
1420000IFEET -- 43* 37' 30'
7 0
15 u

tso. 5

24 780000
EET

4829

-9 1 s

_@5
Z

4028

(4j

4027

tj L,

4826

All

35'

(1 14825

MAP 3

Al 19

4836
IK-7-7 5Z J,

o
f u
ID x

30
Jel 4035
0

40'
Alt,
0

/,YC5

. . . . . . . P 4034

1B 9

0 q)
9 K

ff
Uttle Point Sa Die -C ci
ff-o
4633

J,
Iv

s
- @nn,6
7@V
4

'IS
101 -

44 -
-d.
--A

MAP 4

A 1 20

19b. 5

!P E

@e r
ul
-'@22
C--

I T
I N

U1,
pa*

C- I"

............

WOO

19sr
(3t, 9.LSZ
5 r
or

CE 64D
'-oor ?
Y IV!

(07 5
Igs r G-- -
195 5 T

2p's
rr @( A\,
'T) zf

6'9

i7
MAP 5

A 1 21

LDDADUT
00

-SrA/AJ

LOOKOU T 10,

i I c Z06

REA 0 A

0
1,

?a/. S
Xf'@
S I") zot 5
zoo. S 11,

0
_nw rY
-4,0 S 7
ZOO 5A

ZVO.5 T

A N I S JIN A T
3-

.0 f.,

w /S
1,
77-
r

/Yl I
a19eq %
Vmt r
Lll@l 2.
3 5@1
00

ROAD

'N
=F T BDY
It. ' . - '. I - -In- MAP 6

Al 22

28 0

199.-f
3

o2 -

0 119-5 1-
V/I L

@-Oo
T21 N
T20N STE cd
N

61. f

SASS'
)p r C(A@f

20 75:

Cl.

ct.
9 fq

@U9
2104 r 2-0)
7.-5 T
T, ct

202
?0/
NN %@_

Z.
x
0 zo

C) 0 02.5
o
?02 3-5'
7 R04r

a0e
MAP 7
CC
1)7 0
-0

Al 23

V.,

Q13\

7,
00"

7Y. 5 r

U/I WGRS
'kVkL4

(7
-:2

1 79. s r @i tfOED EL
75,:r
ITT
A .1
W, 14
01

7'
14N.

KI
RENTO! ROAL

DRIVE- 00
S Jo.
1,9 7

1301 \N Ir
r--, MAP
p
I P@110-

Al 24

j,
Z"

Photo 1. View to the west across a dune-
pocketed wetland area immediately landward of
a barrier dune in the Fruitvale Dunes system.
The dune appears as higher topography at the
horizon of the photo. Trees in the foreground
are on a low ridge along the east side of-the
shrub carr/marsh area appearing in the
midground of the photo.

Al 25

Trmw
If

Photo 2. View from the barrier dune to the
northeast across the lower end of a dune-
pocketed wetland area in the Fruitvale Dunes
System. This area is at the southern end of
the elongated landward dune-pocketed wetland
shown in the previous photograph.

Al -- 26

. . . . . . . . . .

Photo 3. View to the southeast across a dune-
pocketed wetland area in the Fruitvale Dune
System. This shrub carr/marsh area is
separated from the wetlands at the foot of the
barrier dune by a low ridge, possibly of dune
or shoreline origin (see foreground of Photo
1). Otherwise it is very similar to and
essentially continuous with the dune-pocketed
wetland shown in the previous two
photographs.

Al -- 27

%%SOL

Photo 4. View to the west from a dune ridge
looking down on a small dune-pocketed wetland
in the Fruitvale Dune System. This wetland
area is undergoing succession from shrub carr
vegetation dominated by Salix discolor and
Salix lucida to tree cover -- Pinus strubus,
Fraxinus pensylvanica, Acer rubrum, and Betula
papyrifera.

Al -- 28

Photo 5. The Horseshoe Lake wetland area in
the Fruitvale Dune System. A -- View across
the ombrotrophic northern end of the wetland;
B -- View to the north across the lower, more
minerotrophic part of the wetland.

Al -- 29

4M_

IN@Ir IN

r-7

AL
dim

Photo 6. Wetland areas located upgradient
from the primary catchment in the Fruitvale
North dune-pocketed wetlands watershed.

Al 30

Photo 7. Winston Road wetland. This area is
located close to the shoreline but drains
inland.

Al -- 31

Photo 8. View of Long Lake basin looking
eastward from dune blocking lower end.

Al -- 32

-A6

Photo 9. Skinny Lake dune-pocketed wetland
(Type A).
lk
I

Al -- 33

Photo 10. WetlaDd area at the northeast end
of Bigsbie Lake.

A] -- 34

Photo I I TA L t Ltt SIA ver L a lie a tic.] surroun (I i ng
wet I ands.

In.
4;1 1..

4Ld'.

v4f
-@"W@w

;6M W,

L:o

Photo 12. Bog lake and surrounding
leatherleaf and wetland area at Cedar Point
Park Road.

Al -- 36

F7 P,

'N.-T
t7

Photo 13. Ridge Road wetland area (not dune-
pocketed).

Al 37

Z@l

Photo 14. Spartina marsh catchment area in
the Nordbouse Dunes System.

Al 38

Photo 15. Cooper Cr eek interdunal peat area.

Al -- 39

IV -A"

Photo 16. Manistee/MasOD county line wetland
area.

Al -- 40

@-7

77- 40

Photo 17. Catchment area of Manistee/Mason
County line dune-pocketed wetland watershed.

Al -- 41

Photo 18. Cedar swamp with patchy openings
along the east side of Bar Lake Swamp.

Al -- 42

Photo 19. Willow shrub carr (background) and
sedge (foreground) vegetation in locally
pocketed areas along the base of the dunes in
Bar Lake Swamp.

Al -- 43

APPENDIX II.

A. Listing of watersheds and drainage areas
along the eastern Lake Michigan shoreline by
type.

B. Listing and topographic map copies of
dune-pocketed wetlands watersheds along the
eastern Lake Michigan shoreline (original
watershed maps are provided in sheet form,
suitable for digitizing).

Table A2-1. Listing of watersheds and drainage areas associated with the eastern Lake Michigan
shoreline. Series numbered LMOOl through LM167 (from the Indiana/Michigan state line
north to Portage Lake in Benzie County) includes all watersheds and drainage
areas in the following types:

UIS - Unimpeded Inland Strom Watershed
USS - Unimpeded Shoreline Strom Watershed
DHM - Drowned River Mouth Watershed (Inland Impeded Stream Watershed)
SIS - Somewhat Dune-Impeded Shoreline Stream Watershed
Ise - Intermittent Strom Channel Watershed
DPW - Dune-Pocketed Wetlands Watershed
BNB - Bay Muuth Bar Lake Watershed

DOD - Dunal Direct Drain Areas [includes inter-dunal drainage areas (not mapped)]
ODD - Other Direct Drain Areas [includes glacial kettle areas, etc.)

Dire t drain areas (DDD and ODD) are not included in the IMBOI through LMB44 series (Worth of
Portage Lake).

Name Type Code County Quadrangle Sheet

LMOOl Michiana Drainage Area DOD BERRIEN NEW BUFFALO WEST
LM002 White Ditch Watershed USS BERRIEN NEW BUFFALO WEST
L4003 Grand Beach Drainage Area DOD BERRIEN NEW BUFFALO WEST
LM004 New Buffalo Lake Watershed SIS BERRIEN NEW BUFFALO WEST
LM005 Galien River Watershed OHM BERRIEN NEW BUFFALO EAST
LM006 Lakeview Drainage Area ODD BERRIEN NEW BUFFALO EAST
LM007 Lakeside Creek Watershed USS BERRIEN NEW BUFFALO EAST
LM008 North Lakeside Creek Watershed USS BERRIEN NEW BUFFALO EAST
LM009 Turner Shores Drainage Area ODD BERRIEN NEW BUFFALO EAST
L4010 North Turner Shores Creek Watershed USS BERRIEN NEW BUFFALO EAST
LMOll Youngren Road Watershed ODD BERRIEN NEW BUFFAL0 EAST/BRIDGEMAN
LM012 Harbert Shores Drainage Area DOD BERRIEN NEW BUFFALO EAST/BRIDGEMAN
LM013 Hazelhurst Creek Watershed USS BERRIEN BRIDGEKAN
LM014 Bethany Beach Drainage Area DOD BERRIEN BRIDGEKAN
LM015 Chickming Creek Watershed USS BERRIEN BRIDGEMAN
LM016 Tower Hill Shorelands Drainage Area DDD BERRIEN BRIDGEMAN
LM017 Painterville Drain Watershed USS BERRIEN BRIDGEMAN
LM018 Warren Dune State Park Drainage Area DOD BERRIEN BRIDGEMAN
LMO19 Warren Dune-Pocketed Wetlands Watershed DPW BERRIEN BRIDGEMAN
LM020 Bridgman Creek Watershed USS BERRIEN BRIDGEMAN
LM021A Warren Dunes North Drainage Area DOD BERRIEN BRIDGEMAN/STEVENSVILLE
LM021B Warren Dunes North Interdunal Wetland Drainage Area DOD BERRIEN BRIDGEMAN
LM022 Thornton Valley Dune-Pocketed Wetlands Watershed #1 DPW/ART BERRIEN BRIDGEMAN
LM023 Thornton Valley Dune-Pocketed Wetlands Watershed #2 DPW/ART BERRIEN BRIDGEMAN
LM024 Grand Mere Bar Drainage Area DDD BERRIEN BRIDGEMAN
LMM025 Grand Mere Lakes Drainage Area BMD BERRIEN BRIDGEKAN/STEVENSVILLE
LM026 Section 16 Drainage Area ODD BERRIEN STEVENSVILLE
LM027 Shoreham Drainage Area ODD BERRIEN STEVENSVILLE/BENTON HARBOR
LM028 St. Joseph River Watershed OHM BERRIEN BENTON HARBOR/BENTON HEIGHTS
LM029 Higman Park Drainage Area ODD BERRIEN BENTON HARBOR/BENTON HEIGHTS
LM030 Bundy Creek Watershed USS BERRIEN BENTON HEIGHTS/COLOMA
LM031 Lake Michigan Beach Creek Watershed USS BERRIEN BENTON HEIGHTS/COLOMA
LM032A South Blue Star Dunes Drainage Area DOD BERRIEN/VAN BUREN COLONA/COVERT
LM033 Harris Lake Basin ODD/GLB BERRIEN COLOMA
LM034 Hibbard Lake Basin ODO/GLD BERRIEN COLOMA
LM035 Jacobs Chapel Lake Basin ODD/GLB BERRIEN COLOMA
LM036 Berrien-Van Buren Lake Basin ODD/GLB BERRIEN/VAN BUREN COLOMA
LM037 Rogers Creek Watershed USS BERRIEN/VAN BUREN COLOMA/COVERT
LM038 Linden Hills Drainage Area DOD VAN BUREN COVERT
LM039 Mud Lake Dune-Pocketed Wetlands Watershed DPW/ART BERRIEN/VAN BUREN COLOMA/COVERT
LM040 Fish Cemetery Dune-Poketed Wetlands Watershed DPW/ART VAN BUREN COVERT
LM041 Brandywine Creek Watershed SIS VAN BUREN COVERT
LM042 Van Buren State Park Drainage Area DOD VAN BUREN COVERT
LM043 Dyckmon Swamp Pocketed Wetlands Watershed DPW/ART VAN BUREN COVERT
LM044 Ruggles Creek Watershed USS VAN BUREN COVERT
LM045 Blue Star Creek Watershed USS VAN BUREN COVERT
LM046 Deerlick Creek Watershed USS VAN BUREN COVERT/SOUTH-HAVEN
LM047 11th Avenue Drainage Area ODD VAN BUREN SOUTH HAVEN
LM048 Armory Creek Watershed UIS VAN BUREN SOUTH HAVEN
LM049 South Haven Township Drainage Area ODD VAN BUREN SOUTH HAVEN
L4050 Black River Watershed UIS VAN BUREN SOUTH HAVEN +
L4051 North Shore Drainage Area ODD VAN BUREN/ALLEGAN SOUTH HAVEN
LM052 North Shore Drive Creek Watershed USS ALLEGAN SOUTH HAVEN
LM053 Mt. Pleasant Drainage Area ODD ALLEGAN SOUTH HAVEN/LACOT/GLENN
LM054 Creekwood Watershed USS ALLEGAN LACOTA/GLENN
LM055 Glenn Creek Watershed USS ALLEGAN GLENN
LM056 71st Street Drainage Area ODD ALLEGAN GLENN
LM057 Section 19 Creek Watershed USS ALLEGAN GLENN
LM058 118th Street Drainage Area ODD ALLEGAN GLENN
LM059 Orchard Creek Watershed USS ALLEGAN GLENN
LM060 Section 18 Drainage Area ODD ALLEGAN GLENN
LM061 Plummerville Creek Watershed USS ALLEGAN GLENN
LM062 County Park Creek Watershed USS ALLEGAN GLENN
LM063 West Side County Park Drainage Area ODD ALLEGAN GLENN
LM064 Pier Cove Creek Watershed USS ALLEGAN GLENN
LM065 Townline Shores Drainage Area ODD ALLEGAN GLENN
LM066 Ramps Creek Watershed USS ALLEGAN GLENN

A2 -- 1

LM067 Johnson Marher Creek Watershed USS ALLEGAN GLENN
LM068 Two Ponds Creek Watershed USS ALLEGAN GLENN
LM069 Bench Herb 633 Drainage Area ODD ALLEGAN GLENN
LM070 Trailer Park Creek Watershed USS ALLEGAN GLENN
LM071A Douglas Beach Drainage Area ODD ALLEGAN GLENN/SAUGATLICK
LM071B Mt. Baldhead Dunes Drainage Area DUD ALLEGAN SAUGATUCK
LM072 Kalamazoo River Watershed DRM ALLEGAN SAUGATUCK +
LM073 Pelican Peak Drainage Area DOD ALLEGAN SAUGATUCK
LM074 Saugatuck Dune-Pocketed Wetlands Watershed DPW ALLEGAN SAUGATUCK
LM075 Halfway Creek Watershed SIS ALLEGAN/OTTAWA SAUGATUCK/HOLLAND WEST
LM076 Gilligan Lake Dune-Pocketed Wetlands Watershed DPW ALLEGAN SAUGATUCK
LM077 Castle Park Dune Drainage Area DDD OTTAWA SAUGATUCK/HOLLAND WEST
LM078A Lake Mucatawe Watershed DRM ALLEGAN SAUGATUCK/HOLLAND WEST
LM078B Kelly Lake Dune-Pocketed Wetlands Watershed DPW/An OTTAWA HOLLAND WEST +
LM079A Holland State Park Drainage Area DUD OTTAWA HOLLAND WEST
LM079B West Ottawa Drainage Area ODD OTTAWA HOLLAND WEST/PORT SHELDON
LM080 Cooling Water Drain (CPCO) Area ODD/ART OTTAWA PORT SHELDON
LM080A Pigeon River Watershed DRM OTTAWA PORT SHELDON +
LM408B Ten Hagen Creek (Sloan Pond) Watershed SIS OTTAWA HOLLAND WEST/PORT SHELDON
LM4081 Campbell - Mt. Baldy Dunes Drainage Area DOD OTTAWA PORT SHELDON
LM082 Hiawatha Low Dunes Drainage Area DDD OTTAWA PORT SHELDON
LM083 Little Pigeon Creek Watershed SIS OTTAWA PORT SHELDON
LM084A Pine Ridge Drainage Area ODD OTTAWA PORT SHELDON/GRAND HAVEN
LMO84B Grand River Watershed DRM OTTAWA GRAND HAVEN
LM085 Hoffmaster State Park Dunes Drainage Area DDO OTTAWA/MUSKEGON GRAND HAVEN/MUSKEGON WEST
LM086 Little B lack Creek Watershed USS/ART OTTAWA/MUSKEGON GRAND HAVEN/MUSKEGON WEST+
LM088 Norton Dunes Drainage Area DDO MUSKEGON MUSKEGON WEST
LM089 Mons Lake (Black Creek) Watershed DRM MUSKEGON MUSKEGON WEST
LM090 Laketon Township Dunes Drainage Area DDD MUSKEGON MUSKEGON WEST
LM091 Muskegon River Watershed OHM MUSKEGON MUSKEGON WEST/DALTON
LM092A Muskegon State Park Dunes Drainage Area ODD MUSKEGON MUSKEGON WEST/DALTON
L4092B Pioneer Park Drainage Area ODD MUSKEGON DALTON
LM093 Pioneer Park Creek Watershed USS MUSKEGON DALTON
LM094A Laketon Dunes Drainage Area ODD MUSKEGON DALTON/MICHILINDA
LM094B Duck Lake Dunes Drainage Area DDD MUSKEGON MICHILINDA
LM095 Muskrat Lake Dune-Pocketed Wetlands Watershed DPW MUSKEGON DALTON/MICHTLINDA
LM096 Duck Lake Watershed OHM MUSKEGON DALTON/MICHILINDA
LM097A Michilinda Drainage Area ODD MUSKEGON MICHILINDA
LM097B Michilinda Dunes Drainage Area DDD MUSKEGON MICHILINDA
LM098 White Lake Watershed URN MUSKEGON MICHILINDA/FLOWER CREEK
LM098B Pierson Drain Watershed SIS MUSKEGON FLOWER CREEK
LM099 Old Channel Trail Drainage Area DDD MUSKEGON FLOWER CREEK
LM1OO Lehman Road Intermitent Dendritic Drainage Area ISC MUSKEGON FLOWER CREEK
LMlOl Fruitvale Creek Watershed USS MUSKEGON FLOWER CREEK
LM102 Fruitvale Dunes Drainage Area DUD MUSKEGON FLOWER CREEK
LM103 Fruitvle South Dune-Pocketed Wetlands Watershed UPW MUSKEGON FLOWER CREEK
LM104 Fruitvale North Dune-Pocketed Wetlands Watershed DPW MUSKEGON FLOWER CREEK
LM105 Little Flower Creek Watershed SIS MUSKEGON FLOWER CREEK
LM106 Flower Creek Dunes Drainage Area DOD MUSKEGON FLOWER CREEK
LM107 Flower Creek Watershed URN MUSKEGON FLOWER CREEK
LM108 Claybanks Township Shoreline Drainage Area ODD MUSKEGON/OCEANA FLOWER CREEK/TOWN CORNERS
LM109 Township Park Creek Watershed USS OCEANA TOWN CORNERS
LM110 South Whiskey Creek Drainage Area ODD OCEANA TOWN CORNERS
LM111 Whiskey Creek Watershed USS OCEANA TOWN CORNERS
LM112 Benona Township South Dunes Drainage Area DDO OCEANA TOWN CORNERS/BIGSBER LAKE
LM113 Long Lake Dune-Pocketed Wetlands Watershed DPW OCEANA TOWN CORNERS
LM114 Skinny Lake Dune-Pocketed Wetlands Watershed DPW OCEANA TOWN CORNERS
LM115 Stony Lake Watershed DRM OCEANA TOWN CORNERS/BIGSBEE LAKE
LM116A Cobnoosa Shores Dunes Drainage Area DDD OCEANA BIGSHEE LAKE
LM116B Rest-of-Comoosa Drainage Area ODD OCEANA BIGSBEE LAKE
LM116C South Little Sable Drainage Area ODD OCEANA BIGSBEE LAKE/LITTLE SABLE PT
LM116D Little Point Sable Dunes Drainage Area DDD OCEANA LITTLE SABLE POINT
LM117 Bigabie Lake Dune-Pocketed Wetlands Watershed DPW OCEANA BIGSBEE LAKE/LITTLE SABLE PT
LM118 Silver Creek (Lake) Watershed BMB OCEANA LITTLE SABLE POINT/MEARS
LM119A Silver Lake State Park Drainage Area DUD OCEANA LITTLE SABLE POINT/MEARS
LM119B Richmonds inlet Shoreline Area ODD OCEANA MEARS
LM119C Behind-the-Bluffs Drainage Area ODD OCEANA MEARS
LM119D Ridge Road Drainage Area ODD OCEANA MEARS/PENTWATER,
LM120 Pentwater Dunes Drainage Area DDD OCEANA PENTWATER
LM121 Pentwater Lake Watershed URN OCEANA PENTWATER
LM121 Pentwater North Dunes Drainage Area DDO OCEANA/MASON PENTWATER
LM122 Bass Lake Watershed BMB/USS OCEANA/MASON PENTWATER
LM123 Summit Township Dunes Drainage Area DDD MASON PENTWATER
LM124 Section 23 Creek Watershed USS MASON PENTWATER
LM125 Section 23 Shoreline Drainage Area ODD MASON PENTWATER
LM126 Meisenheimer Road Creek Watershed USS MASON PENTWATER
LM127 South Reservoir Lakeshore Drainage Area ODD MASON PENTWATER/LUDINGTON
LM128 Section 15 Creek Watershed USS MASON PENTWATER/LUDINGTDN
LM129 Radio Tower Drainage Area ODD MASON LUDINGTON
LM130 Ludington Hydroelectric Plant Outlet ART MASON LUDINGTON
LM131 White Pine Village Dunes Drainage Area DDD MASON LUDINGTON
LM132 Pere Marquette River Watershed OHM MASON LUDINGTON
LM133A Ludington Shoreline Drainage Area ODD MASON LUDINGTON
LM133B North Epworth Dunes Drainage Area DDD MASON LUDINGTON
LM134 Lincoln River Watershed DRM MASON LUDINGTON
LM135 Ludington Dunes State Park Drainage Area DOD MASON HAMLIN LAKE
LM136 Big Sable (Hamlin Lake) Watershed HMB/DRM MASON HAMLIN LAKE
LM137A Big Sable Point Dunes Drainage Area DDD MASON HAMLIN LAKE

A2 -- 2

LM137B Nordhouse Dunes Drainage Area ODD MASON HAMLIN LAKE
LM138 Nordhouse Lake Dune-Pocketed Wetlands Watershed DPW MASON HAMLIN LAKE
LM139 Spartina Marsh Dune-Pocketed Wellands Watershed DPW MASON HAMLIN LAKE
LM140A Porter Creek Watershed Uss MASON LUDINGTON/MANISTRE
LM140B Cooper Creek Road Interdunal Peatland Area DDD/INT MASON MANISTEE
LM141 Northern Outlier Drainage Area DOD MASON LUDINGTON/MANISTEE
LM142 Cooper Creek Watershed Uss MASON LUDINGTON/MANISTEE
LM143A Old Freesoil Road Drainage Area DOD MASON MANISTEE
LM1438 Old Freesoil Road Flats Drainage Area ODD MASON MANISTEE
LM144 Gurney Creek Watershed Uss MASON MANISTEE
LM145 County Line Road South Low Dune Drainage Area UIS MASON/MANISTEE MANISTEE
L4146 County Line Road Dune-Pocketed Wetlands Watershed DPW KASON/MANISTEE MANISTEE
LM147 Section 33 Creek Watershed SLS MASINTEE MANISTEE
LM148 Magoon Creek Watershed Uss MANISTEE MANISTEE
LM4149 Red Apple Road Drainage Area ODD MANISTEE MANISTEE
LM4150 Golf Coarse Creek Watershed Uss MANISTEE MANISTEE
LM151 Section 10 Shoreline Drainage Area ODD MANISTEE MANISTEE
LM152 Manistee River Watershed OHM MANISTEE MANISTEE/PARKDALE
LM153 Orchard Beach State Park Drainage Area ODD MANISTEE PARKDALE
LM154A Bar Lake Watershed BMB MANISTEE PARKDALE
LM154D South Bar Lake Swamp Watershed BMB MANISTEE PARKDALE
LM154C North Bar Lake Outlet Watershed BMB MANISTEE PARKDALE
LM4155 Bar Lake Village Drainage Area ODD MANISTEE PARKDALE
LM156 Bar Lake Shoreline Dunes Drainage Area DOD MANISTEE PARKDALE
LM157 Williamsport Dunes Drainage Area DOD MANISTEE PARKDALE
LM158 Portage Lake Watershed BMB MANISTEE PARKDALE
LM8OI Arcadia Lake BKB MANISTEE ONEKEMA
LM802 Upper and Lower Herring Lake BMB BENZIE ELBERTA
LM14803 Betaie River OHM BENZIE RLBERTA/FRANKFURT
LM804 Crystal Lake BMB BENZIE FRANKFURT/BEULAH
LM4805 Platte River UTS BENZIR FRANKFURT/BEULAH
LM4806 Otter Creek Uss BENZIE EMPIRE
LM807 North and South Bar Lakes BME LEELANAU EMPIRE
LM808 Crystal Run Uss LEELANAU EMPIRE
LM809 Good Harbor Creek Uss LEELANAU GOOD HARBOR BAY
LMBIO Kovorick Creek USS LEELANAU GILLS PIER
LMBII Gills Creek Uss LEELANAU GILLS PIER
LM812 Roaring Brook Uss LEELANAU GILLS PIER
LM4813 Onomonee Dry Gulch Uss LEELANAU GILLS PIER
LM814 Onominee Dry Gulch Uss LEELANAU GILLS PIER
LM815 Christmas Cove Creek Uss LEELAPAU NORTHPORT NW
LM816 Mud Lake Creek Uss LEELANAU NORTHPORT
LM817 Creswel Creek Uss ANTRIM CENTRAL LAKE
LM818 Golden Beach Creek Uss ANTRIM CENTRAL LAKE
LM819 Section 35 Creek Uss ANTRIM CENTRAL LAKE
LM820 Guyer Creek Uss ANTRIM ATWOOD
LM821 West Guyer Creek Uss ANTRIM ATWOOD
LM822 Antrim Creek Uss ANTRIM ATWOOD
LM823 County Line Creek Uss CHARLEVOTX/ANTRIM ATWOOD
LM824 Whiskey Creek Uss CHARLEVOIX ATWOOD
LM825 Inwood Creek UIS CHARLEVOIX CHARLEVOIX
LM826 McGreen Creek UIS CHARLEVOIX CHARLEVOIX
LM827 Pine River (Lake Charlevoix) Uss CHARLEVOIX CHARLEVOIX
LM828 Range Line Creek Uss CHARLEVOIX IRONTON
LM829 Susau Creek UIS CHARLEVOIX TKNTON
LM830 5 Mile Creek Uss EMMET FOREST BEACH
LM831 Forest Beach 1, 2, and 3 Isc EMMET FOREST BEACH
LM832 Went Traverse Township Creek Uss EMMET FOREST BEACH
LM833 Friendship Creek Uss EMMET FOREST BEACH
LM834 Ryan's Corner Dry Gulch IsC EMMET GOOD HART
LM835 Horseshoe Bend Uss EMMET GOOD HART
LM836 Devil's Elbow Creek IsC EMMET GOOD HART
LM837 Good Hart Dry Gulches 1-6 IsC EMMET GOOD HART
LM838 Beckons Corners Creek Uss EMMET GOODHART
LM839 Cross Villae Campground Creek Uss EMMET CROSS VILLAGE
LM840 Wycamp Creek (Wycamp Lake) UIS EMMET CROSS VILLAGE
LM841 Billisu, Hill Dune-Pocketed Wetland Area DPW EMMET BLISS
LM842 Big Sucker Creek Uss EMMET BLISS
LM843 Little Sucker Creek Uss EMMET BLISS
LM844 Big Stone Crack Uss EMMET BLISS

Table A2-2. Dune-pocketed wetlands watersheds of the eastern
Lake Michigan shoreline (listed south to north).

Number NamejCode County Quadrangle Sheet

LM019 Warren Dune-Pocketed
Wetlands Watershed (WD) BERRIEN BRIDGEMAN

IJ4022 Thornton Valley Dune-Pocketed
Wetlands Watershed #1 (TVI) BERRIEN BRIDGEMAN

IM023 Thornton Valley Dune-Pocketed
Wetlands Watershed *2 (TV2) BERRIEN BRIDGEMAN

LMO 39 Mud Lake Dune-Pocketed
Wetlands Watershed (ML) BERRIEN+ COLONA/COVERT

LM040 Fish Cemetery Dune-Pocketed
Wetlands Watershed (FC) VAN BUREN COVERT

IM043 Dyckman Swamp Pocketed
Wetlands Watershed (DS) VAN BUREN COVERT

LM074 Sangutuck Dune-Pocketed
Wetlands Watershed (SD) ALLEGAN SAUGATUCK

IM076 Gilligan Lake Dune-Pocketed
Wetlands Watershed (GL) ALLEGAN SAUGATUCK

IM078B Kelly Lake Dune-Pocketed
Wetlands Watershed (KL) OTTAWA HOLLAND WEST+

114095 Muskrat Lake Dune-Pocketed
Wetlands Watershed (MS) MUSKEGON- DALTON/MICHTLINDA

IM103 Fruitvale South Dune-Pocketed
Wetlands Watershed (FS) MUSKEGON FLOWER CREEK

IM104 Fruitvale North Dune-Pocketed-
Wetlands Watershed (FN) MUSKEGON- FLOWER CREEK

IM113 Long Lake Dune-Pocketed
Wetlands Watershed (LL) OCEANA TOWN CORNERS

IM114 Skinny Lake Dune-Pocketed
Wetlands Watershed (SL) OCEANA TOWN CORNERS

IM11 7Bigsbie Lake Dune-Pocketed
Wetlands Watershed (BL) OCEANA BIGSBER LAKE+

IM138 Nordhouse Lake Dune-Pocketed
Wetlands Watershed (NL) MASON HAMLIN LAKE

LM139 Spartina Marsh Dune-Pocketed
Wetlands Watershed (SM) MASON HAMLIN LAKE

LM146 County Line Road Dune-Pocketed
Wetlands Watershed (CL) MASON+ MANISTEE

IM841 Billiau. Hill Dune-Pocketed
Wetland Area (BH) 24CT BLISS

A2 -- 4

. . .. . ........

n
28

,6T
19

(D
200
T21 N IATW@_Co
r TZIM
,
T20N WWI

County Line Road Dune-Pocketed
Wetlands Watershed (CL)
.9a.r

SAS;.

cc:

0
'197

20P4 7

<
In

4b
LLI
6@__
'AIN" '3,4
0
J-)

0
% 0 1?0 2. 5
02 % < -
% 7 20 T .1,.
\11@,..
\A

A2 5

_STAIAJ

Spartina Marsh Dune-Pocketed
Wetlands Watershed (SM) 211 19
J L

S
REA

-cO

A N",
Ajk,_44
7

20.@ 5 7
ZOO. 5,g_

Z00.5 T

N
I S/*T,I;-E T,
F

1! 4,11

V7. C
i e411J,

NURAN(BAfmoc 61 &j, _4fe@ . ROAD
T_

Nordhouse Lake Dune-Pocketed BDITY.
Wetlands Watershed (NL)
I @0
0/

FOU I DR I DOE
Frg BDY
L

L K E
1)0

A2 6

3 4830
9
2
00. 5

2 "d 780000

4829

E N sr

Qf,

4827

Bigsbie Lake Dune-Pocketed
Wetlands Watershed (BL)
c

V
.3

4025

A2 7

7
\j Oj@ )v
I - ROADI. 10/0
f&

f
J4

N J

4 ff
Stpn 1.

LAK
4m /go q,77-S N
STONY i-s r
Skinny Lake Dune-Pocketed
Wetlands Watershed (SL)

/.11 T
Township'-('

1yo.5 r [email protected] 7
a r

lyl.5

790.5 1 9 0
el 14q

JV
@T-
P

-A
Long Lake Dune-Pocketed
Wetlands Watershed (LL)

-OIL
I-A

% r\
f
<

_C:

A2 8

N. Y OC
RD
@ 6 T fe [us

183. 5'
. . . .------ 9,190

I @Ilj FK D Z,

tl F1

v"I
Meinert
Ri
74
197
Fruitvale North Dune-Pocketedllt,<'
rshed (FN)
Wetlands Wate
>

tit/
cketed
Fruitvale South Dune-Pocketed
Wetlands Watershed (FS)

19 5,

06.S'
9
VY
--77-

A2 9

v6w
D"Cl lakr@@

Duck Lclke

W /77

.........

ED
Muskrat Lake Dune-Pockete
@d
Wetlands Watershed (MS)

71
IN

0
10111',

P, 7

0
211
THN
N n

A2 10

-I, ight- :r,
--0/-'5Uat In
g
6
ill
Liah t,. 4Y
or St
5.1111A L
L
lacatawa':r.'-*'.
Ix I
1K (1)
ST
T .5N IL 0 TOGA
N

.5
T 4 N ..,go

Lo)

U) .
k -2 625 . . . @'@TH

'00 Kelly Lake Dune-Poc
keted 0
116 Wetlands Watershed (KL)

J*@
A

635.-. 146THI
64
-6/9

--f 620 0
01
10
L k@',@
14
KELL Y
Castle ar
A Kelly
110m@

Raldy'
>Io"

is'?3 144TH
'635
\C
01 0
640

c?)
Green Mountain
0
Beach

6
@15 143L
I kUHAIIr
(j Mligan,
Vak

Gill 650
_j J
i9an Lake Dune-Pocketed'@'
Wetlands Watershed (GL)
0 k'

Township Park
6

144 TH
623-1
:635

91
1@40
V\
ntairl
Sreen Mou
Bead",
31
'16
PRAW-
Q, -62@
62f
0
aks

650

Gilligan Lake Dune-Pocketed, 0
Wetlands Watershed (GL)
JV
Township Park4 6

(7,
4
I : W

10
10
mop
f@A'IQI
2
6/4

_J
6c,
Ir lu,;

f
7.

Gibs4 7/0
r 1/v
,A 1 7-@.
Saugatuck Dune-Pocketed
Wetlands Watershed (SD) C@)
G
GAT KRIJN 'Re f
_UC

6
27
ffA
SO
Q 50

GibsDp.
AVE NUE
648 676
I STATE

A2 12

;A,

Dyckman swamp Pocketed
Wetlands Watershed (DS) ;25
34
'5

.P.
yck7 1
UBZI

fog
8

jt@)
C)

625
17 1j 612 T
.4,,. 61,
T7-77 T 2 S 24rH I 7v"
L 6/0-

0
10

4
fb .,677

-- ----------- ----- --------------
)----------
_00 -------

28, 630
-A
I.dp
600

626
6
684
9 L ,-AV

�And R -

A2 13

t 3 672;

'@Sard i
C

0
0

643 E. j,
347H
16'
44

Q>
Q-
"J
Nil,

co, 10
G/
.667
38TH . 2 .65
1 613
6 @711 1.644'
0
Lake 2 M-1
Q
Idpi
4_6

Fish Cemetery Dune-Pocketed
Wetlands Watershed (FC) I

28 679
9,_ 6.
ICU
11-77)

Comars
68
667
6,-6
644

A2 14

j'7

38TH 21
613
Mud Lake 20 629 cem 1644
dpi

kj'
Mud Lake Dune-P ket d
oc e
Wetlands Watershed (ML)@,(

?
616 40TH - A VE

it
f
C

0 r

C
7) 630;' (C:, 636
29-
. . . . . . . . . .
f

'41

it t.
It c
(@N\ si It
it
Fish Conit rs
1 - 634 667
16, 3 1
644 6@6 In

Q 680'

32
33
49 (7@79

I OD
K .

671

r
J, 0
J1,

o cp/

-------- T6S
".T@ 6bO_
6OU

e C
Rosemary a

Thornton Valley Dune-Po
cketed,1
Z101
5"
Wetlands Watershed *2 (TV2)
V
C n

or)
'RD
61 _7
BA
L V4 ga @on

4
'@d
0"
Thornton Valley Dune-Pock ted
Wetlands Watershed *1 (TV

0'
LEMONA_
39@ 6S

61
h Ch,.
(7_;
WA-2
0/
667
e
Landin tI

---Cqfr

1. tt
a

iBM
622 . .........
-I CA...
J,

MA EE 676
9' -J Y -w

U Goi Ourse
@A`
C
Q,
h
ridgma

Ce
r__2
1119

A2 16

3
fl
!BI
62
j...
Weko Beach j
Recreational Are I--

V, PUAY f
QUA Bridg

W
arren Dune-Pocketed
Wetlands Watershed (WD) 2

-0

650

L vvawen
n
0

-SNOW;i
5
0
@K,t-
Nt

55
80)
4 6

6) V:
650
6; \j
84
7 =11-:7 (.
'T

650- 0.

T C,
BRWNTO@N.
621 654. ROA D 659
674

A2 17

NOAA COASTAL SERVICES CTR LIBRARY I
3 6668 14110357 4 1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I