r... a... k n...
3 I ;.. :

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.2! :1.

 

 

"I?“ LIBRARY
1 000‘ Michigan State
University

 

 

 

This is to certify that the
thesis entitled

CHARACTERIZATION AND PALEOENVIRONMENTAL
SIGNIFICANCE OF THE NORTH CENTRAL WISCONSIN
LOESS SHEET

presented by

KRISTINE EMMA STANLEY

has been accepted towards fulfillment
of the requirements for the

MS. degree in GEOGRAPHY

 

 

Mogul—L—

Major Profeébr’s Signature Y
i L— 0 3 200 6

Date

MSU is an Affirmative Action/Equal Opportunity Employer

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

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:IProj/Acc&Pres/ClRC/DaleDueIIndd

CHARACTERIZATION AND PALEOENVIRONMENTAL SIGNIFICANCE OF THE
NORTH CENTRAL WISCONSIN LOESS SHEET

By

Kristine Emma Stanley

A THESIS
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

MASTERS OF SCIENCE
Geography

2008

ABSTRACT

THE CHARACTERIZATION AND PALEOENVIRONMENTAL SIGNIFICANCE OF
THE NORTH CENTRAL WISCONSIN LOESS SHEET

By
Kristine Emma Stanley
In this thesis, I investigate the relatively thin loess sheet in north central
Wisconsin. Locss thickness on uplands exceeds 70 cm in northern and central Clark
County, thinning to the southeast and east, to < 35 cm in western Marathon County. Silt
loam-textured on its eastern margins, the loess sheet coarsens towards the west and
northwest, such that, in western Clark County, the loess mantle is comprised of very fine
and fine sands. It mantles at least three different substrates; from west to east they are (l)
fine-grained, Cambrian sandstone residuum, (2) sandy Bakerville till, and (3) silty Edgar
till. Thickness, particle size data, and silt mineralogy point to two distinct sources for this
loess: (l) the Late-Wisconsin terminal moraine to the northwest (including the numerous
ice-walled lake plains on and behind the moraine) and (2) fine-grained, Cambrian-aged
sandstones and siltstones, which crop out at its western edge. Because of its
environmentally sensitive source areas, the timing of loess deposition may be able to
constrain regional paleoenvironmental changes. It is likely that large quantities of freshly
exposed sediment would have been made available for deflation by the thawing of a
permafrost-dominated landscape, c.g., the destabilization of slopes in the areas dominated
by dissected Cambrian sandstones and the draining of numerous ice-walled lake plains.
Three OSL ages for central Wisconsin loess indicate that loess entrainment and
deposition occurred between approximately 15.2 and 12 ka; thus, regional permafrost

degradation likely occurred during this time span.

Copyright by
Kristine Emma Stanley
2008

ACKNOWLEDGEMENTS

It seems appropriate to start by acknowledging my undergraduate mentor, Peter
Jacobs. His charismatic teaching style and enthusiasm for soils were instrumental in my
decision to pursue graduate studies in physical geography. I so greatly appreciate the
time he spent chatting with me about academia and life in general. Additionally, he
provided helpful comments on this project. I am certain that I would not be here without
his guidance and encouragement.

I am greatly indebted to my advisor, Randy Schaetzl, who provided invaluable
advice, feedback, and encouragement along the way. Randy also supplied much of the
equipment necessary to complete this project. He has taught me so much about soils and
physical geography, as well research and academia. I am so thankful for his quick and
helpful feedback, his constantly open office door, and for always keeping my best interest
in mind. I feel very lucky to be Randy’s student, and I am eternally grateful.

I would also like to acknowledge the faculty here at Michigan State University
who provided support in so many ways. Antoinette WinklerPrins has been a continual
source of encouragement during my time at MSU. Her support in both my academic
pursuits and personal life has been vital to my success. Alan Arbogast provided many
insights into academia and also introduced me to the varied topography of Kansas.
Grahame Larson, in the best class I took at MSU, introduced me to the nuances of glacial
sediments and the applications of this knowledge. I also want to specially thank Dave
Lusch, Catherine Yansa, Ashton Shortridge, Kyle Evered, Richard Groop, Bruce Pigozzi,

and Kirk Goldsberry for mentoring me and assisting in different ways. I also need to

thank Claudia Brown, Judy Reginek, and Sharon Ruggles, who have always been so
helpful and friendly to me.

I would also like to recognize a number of people from various affiliations who
assisted in this project. Remke VanDam shared both knowledge and equipment with me
with respect to this project, and Steve Forrnan ran my OSL samples. I also need to
especially thank Ed Hanson, at Hope College, for generously letting me use his X-ray
diffraction machine, as well as his student, Amanda Brisbin for teaching me how to use
the machine. Thanks to all of the people who participated in Quiders. I appreciate the
time you spared to support and foster a sense of community and interest in Quaternary
research, as well as the constructive comments you provided towards my project. Kent
Syverson assisted with the collection of some samples in Chippewa and Taylor Counties,
and provided a draft of the Pleistocene Geology of Chippewa County, Wisconsin early
on. Additionally, I need to thank people at the Wisconsin Geological and Natural History
Survey including Susan Hunt, Peter Schoephoester, Bruce Brown, and especially, John
Attig.

Many friends and colleagues I have encountered need to be recognized as well. I
would like to especially acknowledge Leslie F usina, Meleia Egger, Steve Aldridge, Bilal
Butt, Audrey Joslin, Crosby Savage, Asger Nielsen, and Jen Holmstadt, among many
others, for their support. I can never thank you enough for helping me though tougher
times. In addition to this support, Brad Blumer has been my go-to person for ArcMap
questions and frustrations. Trevor Hobbs, as a fellow student of Randy, has been helpful
in too many ways to mention. I am very much indebted to Mike Bigsby, who assisted

with portions of laboratory work, and most importantly assisted with the final submission

phase of this thesis. Beth Weisenborn was a pleasure to work with and helped me more
than she probably knows. I appreciate her support, understanding, and advice. I also
need to specially thank Jocl Gruley for continually broadening my geographical
perceptions and for being my source of calm during these last several months. Finally, I
need to thank the landowners for permission to access their properties and my family for
their love and support.

This research was supported by the National Science Foundation under grant
number BC S-0422108 made to R. Schaetzl, and by fellowships from the Geography

Department at Michigan State University.

vi

TABLE OF CONTENTS

List of Tables ....................................................................................... x
List of Figures .................................................................................... xii
1. Introduction ..................................................................................... 1
1.1 Research Objectives ................................................................ 4

2. Literature Review .............................................................................. 5
2.1 Background .......................................................................... 5

2.1.1 Loess Production ......................................................... 6

2.1.2 Loess Transportation and Deposition ................................. 6

2.1.3 Preservation ............................................................... 8

2.1.4 Spatial Characteristics of Loess Deposits ............................. 9

2.1.5 Source Areas and Provenance .......................................... 10

2.2 Loess Chronology in North America ............................................. 13

2.3 Source Areas and Provenance in North America ................................ 15

2.4 Loess in Wisconsin ................................................................. 16

2.4.1 Loess in the Driftless Area ............................................. 17

2.4.2 Loess in North Central Wisconsin ..................................... 19

2.5 Summary .............................................................................. 21

3. Study Area ..................................................................................... 24
3.1 Bedrock Geology .................................................................... 27

3.1.1 Mt. Simon Formation ................................................... 29

3.1.1.1 Mt. Simon Lithostratigraphy ................................ 31

3.1.1.2 Mt. Simon Texture ........................................... 36

3.1.1.3 Mt. Simon Mineralogy ....................................... 36

3.1.2 Eau Claire Formation .................................................... 38

3.1.2.1 Eau Claire Lithostratigraphy ................................ 38

3.1.2.2 Eau Claire Texture ........................................... 39

3.1.2.3 Eau Claire Mineralogy ...................................... 40

3.1.3 Wonewoc Formation ................................................... 40

3.1.3.1 Wonewoc Lithostratigraphy ................................. 41

3.1.3.2 Wonewoc Texture ............................................ 42

3.1.3.3 Wonewoc Mineralogy ....................................... 43

3.1.4 Summary of Bedrock Geology ........................................ 43

3.2 Glacial History and Stratigraphy ................................................. 45

3.2.1 Pre-Late Wisconsin Glaciation ........................................ 46

3.2.1.1 Marathon formation .......................................... 47

3.2.1.1.] Edgar Plain Extent ................................. 47

3.2.1.1.2 Edgar Plain Topography ........................... 48

3.2.1.1.3 Edgar Till Characteristics ........................ 49

3.2.1.2 Lincoln Formation ............................................. 53

3.2.1.2.] Bakerville Plain Extent ........................... 53

vii

3.2.1.2.2 Bakerville Drift Topography ..................... 53

3.2.1.2.3 Bakerville Till Characteristics .................. 54

3.2.1.2.4 Merrill Plain Extent ............................... 55

3.2.1.2.5 Merrill Plain Topography ........................ 55

3.2.1.2.6 Merrill Till Characteristics ...................... 56

3.2.2 Late Wisconsin Terminal Moraine .................................... 57

3.2.2.1 Ice-walled Lake Plains ....................................... 57

3.3 Paleoenvironments .................................................................. 60
3.3.1 Periglacial Vegetation ................................................... 60

3.3.2 Periglacial Geomorphology ............................................. 61

3.3.3 Evidence of Past Permafrost in Wisconsin ........................... 62

3.3.4 Extent and Chronology of Late Wisconsin Permafrost ............. 65

3.3.5 Paleoenvironmental Summary of North Central Wisconsin ........ 66

3.4 Contemporary Climate and Vegetation .......................................... 67
3.5 Soils .................................................................................... 70
3.5.1 Soils with loess mantles ................................................. 73

3.5.2 Soils with sandy mantles ................................................ 75

3.5.3 Other Soils ................................................................ 78

3.6 Summary of Study Area ............................................................ 78
4. Methods .......................................................................................... 80
4.1 Field Methods ........................................................................ 80
4.2 Laboratory Methods ................................................................ 85
5. Results and Discussion ......................................................................... 89

5.1 Particle Size Comparisons of Pre-Late Wisconsin Till Samples to

Literature ............................................................................. 90

5.1.1 Cambrian Sandstone Residuum and Outcrops ....................... 90

5.1.2 Pre-Late Wisconsin Tills ................................................ 94

5.2 Characteristics of Loess in North Central Wisconsin ........................ 100
5.2.1 Soil Profile Data and Interpretations ................................ 100

5.2.2 Loess Sample Data and Interpretations .............................. 105

5.3 Spatial Characteristics of Loess in North Central Wisconsin ................ l 12
5.3.1 Thickness Data .......................................................... 112

5.3.2 Particle Size Data for the Loess Mantle .............................. 116

5.3.2.1 Spatial Trends in Sand Fractions .......................... 116

5.3.2.2 Spatial Trends in Silt Fractions ............................ 1 19

5.3.3 Silt (Feldspar) Mineralogy Data ....................................... 123

5.3.4 Summary of Spatial Trends ............................................ 125

5.4 Potential Source Sediments ....................................................... 127
5.4.1 Northwestern Source ................................................... 127

5.4.2 Westem/Southwestem Source ........................................ 133

5.4.3 Summary ................................................................. 138

5.5 Age and Paleoenvironmental Significance ..................................... 139
5.6 Summary and Implications ....................................................... 146

viii

Appendix A — Soil Profile Description Sheets .............................................. .151
Appendix B — Particle Size Data for Loess/Mantle Samples .............................. 155

References ........................................................................................ 160

ix

Table 2.1:

Table 3.1:

Table 3.2:

Table 3.3:

Table 3.4:

Table 3.5:

Table 3.6:

Table 3.7:

Table 3.8:

Table 4.1:

Table 5.1:

Table 5.2:

Table 5.3:

Table 5.4:

LIST OF TABLES

Loess chronology and stratigraphy in the Upper Mississippi
Valley .................................................................................. 14

Stratigraphic Nomenclature of Wisconsin Bedrock (after Ostrom
1967) ................................................................................ 29

Summary of Cambrian sandstone formation characteristics (after
Asthana 1969; Morrison 1968; Distefano 1973; Ostrom 1970). ............ 37

Pre Late Wisconsin glaciations in north central Wisconsin (after
Clayton et a1. 2006). ................................................................ 46

Moist Munsell colors for pre-Late Wisconsin and Late Wisconsin tills
in north central Wisconsin. ...................................................... 51

Average Sand:Si1t:C1ay ratios of pre-Late Wisconsin and Late
Wisconsin tills in north central Wisconsin. ................................... 52

Characteristics of loess mantled upland soils in the study area (after
Bartelme and Strelow 1977; Fiala 1989; Simonson and Lorenz 2002) ...... 71

Characteristics of sand mantled upland soils (after Simonson and
Lorenz 2002). ..................................................................... 72

Characteristics of upland soils with no recognized mantle (after
Simonson and Lorenz 2002). ................................................... 72

X-Ray diffraction peaks, and their intensity factors, as determined by
Grimley (1996). ................................................................... 87

Particle size data for Cambrian sandstone outcrop and sandstone
residuum samples. ................................................................ 93

Particle size data of pro-Late Wisconsin till samples in north central
Wisconsin .......................................................................... 97

Average Sand:Si1t:C1ay ratio comparisons of pro-Late Wisconsin tills in
north central Wisconsin between cited literature and this study .............. 98

Clay and clayfree particle size data for the three soil pits sampled in
Clark County, WI .................................................................. 101

Table 5.5:

Table 5.6:

Table 5.7:

Table 5.8:

Summary of mantle (loess) particle size data, grouped according to
substrate type. .................................................................... 107

Fine earth and clayfree particle size data for ice-walled lake plain
samples in north central Wisconsin. .......................................... 130

Silt mineralogy of source and loess samples in north central
Wisconsin. ........................................................................ 133

Optically stimulated luminescence (OSL) ages of loess samples from
north central Wisconsin. ....................................................... 142

xi

LIST OF FIGURES

*Figures in this thesis are presented in color.

Figure 2.1:

Figure 2.2:

Figure 3.1:

Figure 3.2:

Figure 3.3:

Figure 3.4:

Figure 3.5:

Figure 3.6:

Loess and sand deposits of Wisconsin (after Hole 1968) ..................... 17

Outwash and loess deposits mapped in Chippewa County, Wisconsin
(after Jake] and Dahl 1989); Syverson 2007). Cahow (1976) discussed

the thick loess (outlined in the yellow). He surmised the thinner loess to
the southeast (outlined in pink), was derived from the large valley train

of the Chippewa Valley/Lake Wissota .......................................... 22

Spatial distribution of loess soils in north central Wisconsin (after
Bartelme and Strelow 1977; Fiala 1989; Simonson and Lorenz 2002).
Dark red represents soil series mapped where loess thickness is >100

cm. Red represents soil series mapped where loess thickness is 51-100 _
cm. Pink represents soil series mapped where loess thickness is 25-50

cm .................................................................................... 24

DEM of Wisconsin showing the study area outlined by a black rectangle
in Clark, Marathon, and Wood Counties. The extent of the last glacial
maximum (the Late Wisconsin Glaciation) is shown by a dark, hashed
line. The extent/ approximate extent of the Driftless Area (an area
lacking evidence of being glaciated at any point during the Quaternary)

is represented by a dashed line/dashed line with question marks where

the extent is difficult to delineate (afier Clayton et a1. 2006) ............... 26

Bedrock of Wisconsin (after Mudrey et a1. 2007). The study area lies
within a transitional zone in which Precambrian crystalline bedrock
outcrops decrease and Cambrian sandstone outcrops increase towards

the southwest ...................................................................... 27

Spatial distribution of bedrock formations in north Central Wisconsin
(after Mudrey 1987; Brown and Patterson 1988) ............................. 30

Stratigraphic column of the Mt. Simon formation at Mt. Simon, WI
(after Asthana 1969) .............................................................. 31

Photographs of shale partings and laminae found within Cambrian

sandstone outcrops in Clark County, WI. Top: 88]; Bottom: SS3.
Photographs by K. Stanley ....................................................... 32

xii

Figure 3.7:

Figure 3.8:

Figure 3.9:

Figure 3.10:

Figure 3.11:

Figure 3.12:

Figure 3.13:

Figure 3.14:

Figure 3.15:

Figure 3.16:

Figure 3.17:

Figure 3.18:

Photograph of the sandstone exposure at 882. This exposure is likely

part of the upper Mt. Simon Formation based on the alternating beds of
yellow-gray, coarse to fine grained sandstones. Photograph by K.

Stanley .............................................................................. 33

Iron stained beds at the top of the Mt. Simon Formation (Shawtown
Formation) at exposure site SSS. Photograph by K. Stanley ............... 34

Detailed stratigraphic column of the Eau Claire Formation from Eau
Claire County, T.27N., R. 10W, Sec. 26, SW 'A, center; scale: 1” = 4’
(after Morrison 1968) ............................................................. 35

Stratigraphic column showing the generalized characteristics of the five
subunits within the Eau Claire formation according to Morrison (1968)...39

Detailed stratigraphic column of the Wonewoc formation (after Ostrom
1970) ................................................................................ 42

Locations of pre—late Wisconsin drift members (after Clayton et al.

2006). Edgar drift is the oldest drift found in north central Wisconsin

and is found mostly in Marathon and Wood Counties. Bakerville

overlies Edgar drift, but exhibits no glacial topography. Merrill also
overlies Edgar drift and drift shares similar characteristics as Bakerville
Drift, but does exhibit some glacial topography ............................... 45

Photograph of the low-relief Edgar till plain. Photograph by K. Stanley..48

Photograph of the low-relief Bakerville till plain. Photograph by K.
Stanley ............................................................................... 54

Photograph of an ice-walled lake plain in Taylor County, Wisconsin.
Photograph by K. Stanley ......................................................... 58

Photograph illustrating hummocky topography of the Late Wisconsin
moraine in Taylor County, Wisconsin. Photograph by K. Stanley ......... 58

Distribution of selected ice-wedge polygons and polygonal ground in
Wisconsin (after Black 1965, Johnson 1986, Clayton 1986, Clayton and
Attig 1989, Mickelson and Syverson 1997, and Clayton et al. 2001).

The minimum extent of continuous permafrost is illustrated in shades of
gray emanating from the glacial margin during the periods shown (Black
1965; Clayton and Attig 1989, Clayton et al. 2001) .......................... 63

Image of ice-wedge cast in friable Cambrian sandstone (Johnson 1986;
Figure 21) ........................................................................... 64

xiii

Figure 3.19:

Figure 3.20:

Figure 3.21:

Figure 3.22:

Figure 3.23:

Figure 3.24:

Figure 4.1:

Figure 4.2:

Figure 5.1:

Figure 5.2:

Pre-settlement vegetation of north central Wisconsin (after Finley 1976).67

Low-relief, agricultural landscape typical of the study area. Photograph
by R. Schaetzl ..................................................................... 68

Contemporary forest cover in Clark County Wisconsin. Forest cover
(dark green) is most abundant in the bedrock uplands and valleys of
western and southwestern Clark County ....................................... 68

Mean monthly temperatures and precipitation (1971-2000) for
Neillsville Wisconsin (NCDC 2002) ............................................ 69

Upland soils of the study area with a silt loam (loess) mantle (after
Bartelme and Strelow 1977; F iala 1989; Simonson and Lorenz 2002). ....74

Other upland soils of the study area (after Simonson and Lorenz 2002).
Soils formed in glacial till, with no recognized mantle are represented by
blue and violet tones. Soils with a sand mantle are shown in warm

tones ................................................................................. 76

Pre-Late Wisconsin drift boundaries (after Clayton et a1 2006) with a
DEM in the background. Locations of OSL samples are represented by
yellow triangles ..................................................................... 82

Logarithmic particle size distribution for all 79 loess samples in north
central Wisconsin .................................................................. 86

Location of sandstone sample sites in north central Wisconsin overlaid
on a map of Cambrian sandstone formations (after Mudrey 1987; Brown
and Patterson 1988). Most of the Cambrian sandstone that underlies
soils within the study area is of the Mt. Simon Formation. The
Wonewoc and finer grained Eau Claire Formations lie to the southwest
of the study area. The following samples: 881, SS4, SSS, were
identified as Mt. Simon Formation samples based on bedrock
distribution maps of Mudrey et al. (1987) and Brown (1988). The 832
samples were later classified as a sample from the upper unit of the Mt.
Simon formation based on particle size data, color, and bedding
characteristics described by (Morrison 1968) and Asthana (1969) ......... 91

Photograph of a shale layer in sandstone sample SSlC (bluish green

from approximately 29.5 to 30 in). Dark brown layer also included in
bulk sample particle size data. Photograph by K. Stanley ................... 92

xiv

Figure 5.3:

Figure 5.4:

Figure 5.5:

Figure 5.6:

Figure 5.7:

Figure 5.8:

Figure 5.9:

Locations of substrate (pre-Late Wisconsin till) samples, soil pedons,

and the boundaries of pre-Late Wisconsin till sheets and the Late
Wisconsin terminal moraine (LGM) in north central Wisconsin (after
Clayton et a1. 2006) ............................................................... 92

Photograph of soil profile at Clark 1, with cumulative particle size depth
plot overlain at approximately the same scale in cm. This soil

presumably formed in 42 cm of loess, over very fine sand-textured loess
(30cm thick), over Bakerville till, over sandstone residuum. The last

two samples (3Btb2 at 120 cm and 3Cg at 1450m) were collected by
auger. The lithologic discontinuity (LD) between the sandy loess and

the till is at 70 cm (3Bt/E2). A thick, red (SYR 5/3) paleosol is

interpreted to extend from 84 to 145 cm in the till. Photograph by K.
Stanley ............................................................................. 102

Photograph of soil profile at Clark 2, with cumulative particle size depth
plot overlain at approximately the same scale in cm. This soil formed in
loess, over Bakerville till, over an outwash-like diamicton, over

crystalline bedrock residuum. The lithologic discontinuity (LD)

between loess and till is at 50 cm. Photograph by K. Stanley ............. 103

Photograph of the soil profile at Clark 3, with cumulative particle size
depth plot overlain at approximately the same scale in cm. This soil
formed in 91 cm of loess overlying Merrill till; the lithologic

discontinuity (LD) between the two is shown by the dashed line. The

2C horizon (at 114 cm) was sampled by auger. Loess below the E

horizon (29cm) steadily increased in sand (dominantly very fine sand)
with depth. Photograph by K. Stanley ........................................ 104

Locations of loess samples, soil pedons, and boundaries of pre-Late
Wisconsin till members, and the Late Wisconsin terminal moraine
(LGM) in north central Wisconsin (after Clayton et al. 2006) .............. 106

Particle size distribution of the less than 2 mm fraction for mantle (loess)
samples grouped according to underlying substrate in north central
Wisconsin. A. Loess samples overlying Edgar till. B. Loess samples
overlying Merrill till. C. Loess samples overlying Bakerville till. D.

Loess samples overlying sandstone residuum ................................. 108

Kriged map of predicted loess thickness based on 79 sample locations in

north central Wisconsin. Dark shades represent thickest areas of loess
(>75 cm), and light/white shades represent loess thickness <25 cm ....... l 13

XV

Figure 5.10:

Figure 5.11:

Figure 5.12:

Figure 5.13:

Figure 5.14:

Figure 5.15:

Figure 5.16:

Figure 5.17:

Kriged maps of clayfree particle size distributions in north central
Wisconsin. Darkest shades represent highest concentrations, and

light/white shades represent the lowest concentration of the following
particle size fractions: A. 175-250 urn; B. 125-175 um; C. 75-125 um;

D. 50-75 pm. B. 35-50 pm; F. 25-35 pm; G. 12-25 pm; H. 2-12 pm ...... 117

Hillshade of a DEM overlain by the kriged loess thickness surface (gray
tones; this study) overlain by soils formed in loess (red, pink and fiichsia
colors; Bartelme and Strelow 1977, Fiala 1989, Simonson and Lorenz
2002). Note that the kriged loess thickness surface closely corresponds

to the extent of loess mapped Clark, Marathon, and Wood Counties,
Wisconsin ........................................................................... 121

Spatial distribution of plagioclase contents in loess of north central
Wisconsin. Plagioclase percentage determined using equation 5.1
(Grimley 1996) ..................................................................... 123

Spatial distribution of K-feldspar contents in loess of north central
Wisconsin. K-feldspar percentage determined using equation 5.1
(Grimley 1996) ....................................................................... 125

Location of ice walled lake plains (blue) in Chippewa and Taylor

Counties, Wisconsin (Attig 1993, Syverson 2007) with respect to kriged
loess thickness. Locations of loess sample points represented by black

dots, and the locations of soil pedon descriptions are represented by

white triangles ...................................................................... 128

Sediments from ice walled lake plain samples (red) peak between 20-45

um as do the majority of the silt loam loess samples (black). Loam loess
samples are shown in blue. Sandy loam loess samples are shown in

pink. Loamy sand loess samples are shown in yellow. Top: Particle size
(logarithmic um) distribution of ice-walled lake plain and loess samples
according to texture class. Bottom: Locations of ice walled lake plain

and loess (mantle) samples in north central Wisconsin ....................... 131

X-ray diffraction results of the 20-45 pm fraction for ice walled lake

plain samples and sandstone and residuum samples in north Central
Wisconsin. Ice walled lake plain samples (dark lines) contain quartz

(26.5 2(9) and plagioclase (approximately 27.8 26)), and very little k-
feldspar (27.5 26)). Sandstone and sandstone residuum samples contain

no plagioclase, and higher amounts of k-feldspar ............................. 132

Distribution of soils formed in loess with respect to the Black River in
north central Wisconsin (after Bartelme and Strelow 1977; F iala 1989;
Jakel and Dahl 1989; Mitchell 1996; Simonson and Lorenz 2002;

Boelter 2005) ...................................................................... 134

xvi

Figure 5.18:

Figure 5.19:

Figure 5.20:

Figure 5.21:

Distribution of soils (yellow) mapped in north central Wisconsin where
underlying sandstone interbedded with shale bedrock is at shallow

depths (< 152 cm; after Bartelme and Strelow 1977; Thomas 1977; Fiala
1989; Jakel and Dahl 1989; Langton and Simonson 1998; Simonson and
Lorenz 2002). In the southwest corner of the map, sandstone is near the
surface on ridgetops. To the north and west bedrock is near the surface

in all landscape positions with the exception of alluvium-filled river

valleys. In the center of the study area, near-surface bedrock is only on
backslope positions ................................................................ 137

Kriged loess thickness map showing the locations of OSL samples and
estimated ages results .............................................................. 140

Chronology of permafrost and loess deposition in north central

Wisconsin and adjoining regions. Dark bars represent timing of event;
light bars represent conservative timing estimates for the same event. 1.
Permafrost extends at least 90 km beyond the Late Wisconsin Moraine,
based on ice-wedges found in the Driftless Area of Wisconsin. (Clayton

et al. 2001 ). 2. Accelerated erosion in the Driftless Area (Mason 1995;
Mason and Knox 1997). 3. Period of loess deposition based on OSL

age estimates from loess in north central Wisconsin (this study). 4.
Spruce-Tundra paleoenvironment surrounds Wood Lake, Taylor County,
approximately 30 km north of the study area (Heide 1984) ................. 145

Summary of particle size data of the north central Wisconsin loess sheet;
darker tones represent higher concentrations. Left: This map shows the
kriged particle size distribution of 75-125 um contents. Thawing
permafrost likely exposed large quantities of sediment from friable, fine-
grained Cambrian sandstones, rich in quartz and K-feldspar, to deflate
and be deposited downwind approximately 15.2 ka. The lower contents
in the north are likely due to dilution of finer grained loess derived from
the northwest source. Right: This map shows the kriged particle size
distribution of the 12-25 pm contents, which suggest a finer-grained
source sediment to the northwest (likely ice-walled lake plains in and
behind the moraine). Thawing permafrost underlying and ice within the
Late Wisconsin moraine thawed, allowing ice-walled lake plains on and
behind the moraine to drain, exposing the lake beds (composed of off-
shore silts and clays) to wind erosion approximately 13 ka. The lower
contents of 12-25 pm particles in the south central region of the study
area are likely due to the dilution of coarser sediment from the
western/southwestern source ................................................... 147

xvii

1. Introduction

Loess deposits occur in many parts of the world, and are often associated with
glacial and periglacial environments (Smith 1942; Frye et a1. 1962; Smalley 1966; Ruhe
and Olson 1980; Pye 1984; Bettis et a1. 2003). Not only are these deposits important as
productive soils for agriculture and forests, but they also serve as important proxies of
paleoenvironmental change because they commonly record periods of landscape
instability (loess accumulation) and stability (soil development) (Willman and Frye 1970;
Follmer 1996; Bettis et al. 2003).

In North America, thick loess deposits are typically associated with, and thought
to be derived from, valley trains and wide river valleys (i.e. the Mississippi, Missouri,
and Illinois Rivers), based on knowledge of eolian and glacial systems, and observations
within modern glacial environments (Smith 1942; Snowden and Priddy 1968; Smalley
1972; Lea 1990; Muhs et a1. 2004). In this model, rivers draining glacial ice would have
been rich with sediment. During times of low flow, fine sands and silts are deflated from
valley trains and broad river valleys and subsequently deposited downwind, and thus,
record ice sheet dynamics and dominant wind directions (Smith 1942; Ruhe 1954; Frye et
al. 1962; Fehrenbacher et al. 1965; Smalley 1966; Olson and Ruhe 1979; Ruhe 1.983;
Johnson and Follmer 1989).

Loess deposits, regardless of origin, require three conditions in order to form: (1)
a sufficient and sustained source of unconsolidated sediment, (with dominantly silt and
very fine sand particles), (2) sustained winds of sufficient velocity for their transport, and
(3) adequate sites for sediment deposition and accumulation (Pye 1995). Once these
conditions are met, loess deposits can develop, often with specific spatial patterns of

1

thickness, particle size distribution, and mineralogy, in relation to the sediment source
area (Smith 1942; F ehrcnbacher et a1. 1965; Olson and Ruhe 1979; Johnson and Follmer
1989; Muhs and Bettis 2000; Mason 2001). Thus, mapping these characteristics can
provide necessary clues (i.e. paleowind directions) as to the source area for a particular
loess sheet. Knowing where the source area exists is important because it may provide
insights into the paleoenvironmental conditions under which loess deposition occurred.
In addition, if the age(s) of loess deposits can be determined, as has been done repeatedly
by OSL dating (Duller 1996; Forman and Pierson 2002), the timing of the eolian activity
and possibly the inferred paleoclimate chronologies of its formation can be constrained
and interpreted.

Although thick loess deposits have been the focus of much detailed research to
date, relatively thin loess deposits blanketing much of the Midwest have been, until
recently (Schaetzl and Loope 2008; Schaetzl in press), largely ignored. For example, the
relatively thick loess of the Driftless Area in southwestern Wisconsin has been the focus
of extensive research, while the relatively thin loess that covers the majority of the state
has received little, if any, attention. Often, these thinner and smaller loess sheets do not
have an obvious source because they do not lie adjacent to broad meltwater river valleys.
Not only do these loess sheets present an interesting question as to where the loess was
derived, but they can also provide age constraints to local paleoenvironmental conditions,
if a source area can be identified.

Constraining the ages of local and regional paleoenvironment events are
important in places like Wisconsin, where radiocarbon-datable materials are largely

absent between approximately 25 ka and 12 ka (Clayton and Moran 1982; Attig et a1.

1998; Colgan 1999; Clayton et al. 2001; Syverson and Colgan 2004). Currently, the ages
of glacial events in north central Wisconsin are based on correlations located hundreds of
kilometers away (Clayton et al. 2001; Attig 2008). These correlations are predicated on
the assumption that the entire glacial margin advanced in one synchronous event;
however, more recent research suggests that glacial lobes advanced and retreated
asynchronously (Attig 1993; Ham and Attig 1997; Attig et al. 1998).

Based on the pollen record and numerous fossil permafrost features such as ice-
wedge casts found in the regions adjacent to, and within, Wisconsin, it is apparent that
beyond the glacial margin, much of the state experienced a periglacial climate at the time
of the last glacial maximum (Black 1965; Birks 1976; Heide 1984; Baker et al. 1986;
Johnson 1986; Clayton and Attig 1987; Clayton et al. 2001). Degradation of this
permafrost may have supplied copious amounts of newly exposed and devegetated
sediment to erosional and deflationary processes (sensu Mason 1995; Mason and Knox
1997). If a loess sheet can be tied to such a source area, it could, theoretically, constrain
the timing of permafrost degradation, or at the very least, a period of rapid climate and
landscape change.

With this background in mind, the focus of this research is the central Wisconsin
loess sheet, which is located in Clark, Marathon, Wood, and Taylor Counties. In his
1942 dissertation, Hole argued that this deposit was eolian in origin rather than weathered
till or of aqueous origin. However, the focus of his research was not the loess deposit
itself, but rather the till that underlies much of the loess. Many other studies also
recognize the deposit as loess (Hole 1942; Hole et al. 1976; Bartelme and Strelow 1977;

Attig and Muldoon 1989; Fiala 1989; Sutherland 1989; Clayton 1991: Attig 1993;

Simonson and Lorenz 2002; Syverson 2007), but to date, it has not been the focus of
detailed research. This particular loess sheet has a complex thickness pattern and
presents an excellent opportunity for a case study on thin loess deposits in relation to
sources areas and regional paleoenvironmental chronologies, and thus, is the focus of this

research.

1.1 Research Objectives

This study has three main research objectives:
( 1) To document, characterize, and map the north central Wisconsin loess sheet.
(2) To determine the likely source area(s) for the north central Wisconsin loess sheet.
(3) To constrain the timing of eolian activity and loess deposition in north central
Wisconsin.
(4) To discuss and evaluate likely paleoenvironmental scenarios under which loess

deposition occurred.

2. Literature Review

Examples of well studied, thick loess deposits occur in China, central Asia, the
Ukraine, Siberia, central Europe, Argentina, as well as the Great Plains and Midwest of
North America (Pye 1984; Bettis et al. 2003). Loess deposits have been used to infer
changes in paleoenvironmental conditions during the Quaternary Period. in many areas of
the world. In order to use loess in the interpretation of paleoenvironmental change,
background information is needed as to its formation and typical characteristics. In this
chapter, I will discuss how eolian transport and deposition processes form the
distinguishing spatial characteristics of loess deposits, as well as discuss the implications
of loess preservation. I will then discuss how the spatial characteristics of loess deposits
can be used to infer loess provenance. Finally, I will conclude with a brief presentation

of loess in North America in general, followed by a discussion of loess in Wisconsin.

2.1 Background

Loess is widely recognized as a deposit of windblown, dominantly silt-sized (20-
60 um) quartz and feldspar particles (Smalley 1966, 1975), often associated with
glaciofluvial processes (Smith 1942; Frye et al. 1962; Ruhe and Olson 1980; Bettis et al.
2003). Loess deposits share certain spatial characteristics with respect to thickness and
particle size distributions related to eolian transport and deposition mechanisms. In this
section, I will provide background information regarding loess dispersal theories. Next,
I will describe the typical spatial trends in loess thickness and particle size characteristics,
and how provenance can be determined, using these spatial properties. I will also briefly

discuss implications of loess preservation on the landscape.

2.1.1 Loess Production

Until recently, the Midwestern conceptual model of loess dispersion, linking loess
deposits to broad river valleys which carried meltwater and sediment from the Laurentide
Ice Sheet (LIS), was prevalent in loess research (Smith 1942; Snowden and Priddy 1968;
Olson and Ruhe 1979; F ollmer 1996; Mason 2001; Bettis et al. 2003). In this model,
glacial ice grinds rocks and sand—sized quartz grains into dominantly silt-sized grains,
which are transported, sorted, and deposited as outwash in broad river valleys draining
the ice sheet (Smalley 1966, 1975; Ruhe 1983). In times of low flow, sediment from the
broad river channels and floodplains can be deflated by prevailing winds, which are later

deposited downwind (Frye et al. 1962; Johnson and Follmer 1989).

2.1.2 Loess Transportation and Deposition

Eolian transport can occur by (1) suspension, (2) saltation (bouncing of particles
along a surface), or (3) creep, depending on the size of the particle and wind
characteristics. It is important to remember for this discussion that eolian sediment is not
only transported in the form of single grains, but also by aggregates of silt and clay
(Mason et a1. 2003). With this in mind, saltation usually takes place with fine and
medium sand (125-500 um) particles (Tsoar and Pye 1987). Particles <50 um can be
transported by suspension in the atmosphere as long as a minimum velocity and/or
turbulence is obtained (Tsoar and Pye 1987). Grains between these particle ranges (50-

125 um) can be transported by short-term suspension or modified saltation, depending on

the wind velocity and turbulence (Tsoar and Pye 1987). The break between these forms
of transportation is likely within the 70-100 um range (Franzmeier 1970).

Deposition of particles traveling in suspension occurs dominantly when (l) the
wind velocity or turbulence decreases, (2) particles are washed out of the atmosphere by
rain, and/or (3) particles in transit are stopped due to surface roughness, e. g. vegetation
(Tsoar and Pye 1987). Although loess deposits develop due to any combination of the
reasons above, particles in transit closer to the ground are more apt to becoming trapped
on rough surfaces.

Mason et al. (1999) expanded the surface roughness theory described by Tsoar
and Pye (1987) to include changes in landscape relief. In this study, Mason et al. (1999)
provided a supplemental approach for loess deposits that are separated from the source
area by a low-relief surface of transport (little, if any net loess accumulation). In this
model, sands from the source area saltatc along this intermediate surface, and in the
process, stir up silt and clay sized particles that may have deposited on the surface of
transport. These deflated, fine particles can become re-entrained and carried further
downwind. This process continues until the sand grains fall into a steep sided valley or
get trapped against a steep topographic break, e.g., an escarpment, preceding an area of
increased relief. Protected from saltating sand grains, the sediment in short-term
suspension is deposited downwind from the topographic barrier. As the transportation
surface stabilizes, a thin loess deposit develops on the transportation surface itself, on the
windward side of the topographic barrier. In this process, an abrupt boundary between

the area ofthin and thick loess deposits often coincides with a steep-sided stream valley

too small to be a glaciofluvial source (Mason et al. 1999) or on the escarpment/upland

itself (Schaetzl and Loope 2008).

2.1.3 Preservation

Loess preservation is greatly affected by landscape position and can significantly
influence interpretations of loess deposits. Knowledge of landscape positions that are
susceptible to post depositional additions or losses is important, especially when choosing
sample locations/study sites, as well as interpreting the results of analysis.

Loess is initially deposited uniformly across an area, blanketing all parts of it,
regardless of position along a catena (Schaetzl and Anderson 2005). Due to post
deposition slope processes, however, loess can be redistributed along the eatena. For this
reason, loess in dissected landscapes has a tendency to be best preserved and thickest on
upland drainage divides and interfluves (Ruhe 1954; Frye et al. 1962). Loess on
sideslopes tends to be thinner, better sorted, and coarser than summit positions, due to
erosion of fine grained sediments (Milne 1936; Ruhe 1954; Sommer and Schlichting
1997; Schaetzl and Anderson 2005). Erosion not only decreases loess thickness and
changes particle size characteristics, but also slows soil development. Conversely, foot
and toeslope positions are depositional foci of fine-grained sediment from upslope areas
(Schaetzl and Anderson 2005). Colluvial additions of loess, also known as secondary
loess, can be incorporated into loess deposits through pedogenic or biogenie processes,
leading to overestimation of initial loess thicknesses (Obruchev 1945; Smalley 1972;
Pecsi 1990; Pye 1995). Alternatively, accretion of colluvium can potentially result in a

sequence of paleosols within the colluvium (Schaetzl and Anderson 2005). These

additional sequences of loess derived deposits and paleosols may be misinterpreted as

changes in paleoelimatic conditions, when, in fact, they reflect only local slope processes.

2.1.4 Spatial Characteristics of Loess Deposits

The thickness of loess deposits predictably and systematically thins with
increasing distance from the source area (Smith 1942; Fehrenbacher et al. 1965; Olson
and Ruhe 1979; Johnson and Follmer 1989; Muhs and Bettis 2000; Mason 2001; Bettis et
al. 2003). In cross section, loess deposits occur as a wedge-shape body, with the thickest
accumulation in areas proximal to the source (Fehrenbacher et al. 1965; Snowden and
Priddy 1968; Frazee et al. 1970; Tsoar and Pye 1987; Johnson and F ollmer 1989; Pye
1995). Thickness trends of the wedge-shaped deposits vary, based partially on the type
of vegetation present at the time of deposition (Tsoar and Pye 1987). For example, when
a large change in surface roughness is present i.e. the edge of a forest, loess is easily
trapped by the tall vegetation (and washed to the surface) creating a steep gradient loess
wedge (Tsoar and Pye 1987; Pye 1995). Where the height of the vegetation is lower (i.e.
grassland or tundra), suspended particles are less apt to be ensnared, resulting in a more
gradually sloping loess wedge (Tsoar and Pye 1987). In addition to paleovegetative
impact on the wedge shape, topography can also influence loess thickness trends. A
topographic barrier, i.e. an upland or narrow valley, can effectively halt saltating sands,
allowing particles in suspension to be deposited with little re-entrainment (Mason et al.
1999; Schaetzl and Loope 2008). Additionally, Putnam et al. (1988) and Leigh (1994)
showed a relationship between width and orientation of the source valley and loess
thickness downwind, which is ultimately connected to the amount of sediment available

for deflation and the paleowind direction.

Particles fall out of suspension according to Stokes Law (Pye 1995). This means
larger particles (i.e. particles moving in modified saltation or short term suspension) fall
out of suspension more quickly than smaller particles. Because of this, loess deposits
typically exhibit systematic fining of particle size with increasing distance from the
source (Smith 1942; Ruhe 1954; Frazee et al. 1970; Olson and Ruhe 1979; Muhs and
Bettis 2000). Coarse and very coarse silt contents decrease at predictable rates with
increased distance from the source. Conversely, the amount of fine silt and clay increases
systematically with increasing distance downwind (Ruhe 1954; Frazee et al. 1970; Olson
and Ruhe 1979).

The particle size distribution of loess sheets vary based on the characteristics of
the source and regional topography. For example, loess sheets adjacent to and downwind
from broad river valleys or valley trains exhibit a classic wedge-shape, where particles
traveling closer to the ground are trapped by well-vegetated, proximal areas and finer
particles traveling higher in the atmosphere are deposited further downwind (Pye 1995).
In situations that involve source areas with less-discrete boundaries (i.e. desert margins),
loess sheets are often separated from the source area by dunes or sandsheets (Pye 1995).
The latter scenario represents a continuum downwind of the source area, transitioning
from dunes to coversands to sandy loess to (silty) loess, and possibly to clayey loess
before becoming so thin that it is difficult to recognize the eolian inputs as a discrete

parent material (Pye 1995; Crouvi et al. 2008).

2.1.5 Source Areas and Provenance

Determining the source areas and provenance of loess deposits is a persistent goal

in loess research (Smith 1942; Fehrenbacher et al. 1965; Snowden and Priddy 1968;

10

Rutledge et al. 1975; Olson and Ruhe 1979; West et al. 1980; Johnson and Follmer 1989;
Aleinikoff et al. 1999; Grimley 2000; Mason 2001; Schaetzl and Loope 2008; Schaetzl
and Hook 2008). Most commonly, spatial variations in thickness and/or particle size
distribution (PSD) are used to determine source areas (Smith 1942; Fehrenbacher et al.
1965; Putman et a1. 1988; Johnson and Follmer 1989; Mason 2001; Schaetzl and Hook
2008); whereas chemical composition, mineralogy, and magnetic susceptibility have been
used to determine provenance (ultimate source) of loess (Ruhe and Olson 1980; Grimley
et al. 1998; Grimley 2000; Aleinikoff et al. 1999; Muhs et al. 1999; Muhs and Bettis
2000; Schaetzl and Loope 2008).

The wedge-shaped spatial characteristics of loess described in the previous
section allow for identification of prevailing paleowind direction at the time loess
deposition. Decreasing loess thickness and PSD trends indicate a downwind spatial
arrangement. Once a prevailing paleowind direction is identified, potential source areas
can be discerned. In North America, most loess deposits are adjacent to and downwind
from valley trains and broad river valleys, which have often been determined to be loess
sources (Smalley 1966, 1975; Ruhe 1983). For example, Fehrenbacher et a1. (1965) used
thickness trends in loess of southern Illinois and Indiana to determine that the dominant
paleowinds during loess deposition were westerly. Using this information they identified
the Wabash River (Illinois) and Ohio River (Indiana) as primary sources for loess in
southern Illinois and southern Indiana, and the White River as a secondary loess source.
Similarly, Muhs and Bettis (2000) used trends in particle size data, in addition to
thickness data, to determine that westerly winds were dominant during loess deposition in

western Iowa.

11

Some loess deposits in North America do not lie adjacent to large river valleys or
valley trains, making it more difficult to readily identify a source area. In these cases,
other loess properties are used to determine the source area, and potentially provenance,
of sediment. Ruhe and Olson (1980) were able to identify source areas and provenance
for loess in Indiana using clay mineralogy. Additionally, feldspar silt mineralogy has
been used to determine the source area (Mississippi River Valley) for a specific loess
deposit in Illinois (Leigh 1994). Grimley (1996, 2000) and Grimley et al. (1998)
expanded on Leigh’s research and found the provenance of the same loess deposit to be
mostly of Superior lobe sediment, using silt mineralogy and magnetic susceptibility.

Characteristics of loess deposits have been shown to be useful in
paleoenvironmental reconstruction, in addition to determining provenance (Aleinikoff et
a1. 1999; Muhs et al. 1999; Muhs and Bettis 2000; Wang et a1. 2006). For example,
relative wind speed, in addition to direction, can be inferred based on changes either in
the modal grain size of the coarse silt fraction or ratios between coarse and medium silt
with depth (Wang et al. 2006). Thickness and PSD analysis suggested a change from
glaciogenic to nonglaeiogenic sources over time for a lithostratigraphic sequence of loess
in Nebraska (Mason et al. 2007). Several properties (i.e. sedimentology, mineralogy,
geochemistry, and lead isotopes in feldspars) were used to determine loess sources for
deposits in Colorado and Nebraska (Aleinikoff et a1. 1998; Aleinikoff et al. 1999; Muhs
et a1. 1999; Mason 2001). Results from these data suggested two possible sources of
loess (1) the Platte River and (2) the White River Group, a tertiary siltstone that crops out
to the northwest of the loess sheet. Researchers suggested that, during warmer periods,

the Platte River, which drained melting alpine glaciers in the Rocky Mountains, was the

12

dominant source; however, during colder periods, the White River Group was the
dominant source of loess (Aleinikoff et al. 1999). They argued that vegetation likely
decreased in the area where the White River Group crops out, subjecting it to wind

erosion; its sediments then became entrained and carried downwind.

2.2 Loess Chronology in North America

Loess units are often named based on the age of the deposit, which can be
obtained using various methods i.e. radiocarbon dating (RC) of organic material that
underlies the deposit (maximum limiting age) or overlying the deposit/soil organic matter
from the soil formed in the loess deposit (minimum limiting age), optically stimulated
luminescence dating (OSL), and thermoluminescence dating (TL). The following
section presents a brief discussion of the various, major loess lithologic units and
chronology found in North America.

The oldest widespread loess unit known in North America is Loveland loess
(Table 2.1), which is thought to have been deposited during the Illinoian period (Willman
and Frye 1970; Forrnan and Pierson 2002). Based on 10Be and TL ages, Loveland loess
in the middle Mississippi Valley was first deposited approximately 185 ka and ended
approximately 130 ka (Markewich et a1. 1998). The Sangamon Geosol subsequently
formed in the Loveland loess during a time of slow or nondeposition between 130 and 60
ka (Leigh and Knox 1993, 1994; Markewieh et al. 1998). Usually overlying the
Loveland Formation and Sangamon Geosol is Roxana loess, which is a middle-
Wisconsin aged deposit (Table 2.1; Johnson and Follmer 1989; Leigh and Knox 1993;
Grimley et al. 2003). Ages for the Roxana formation vary, based on location along the

Mississippi River Valley and by the use of different dating methods. Extrapolated RC

13

ages indicate that loess deposition occurred between approximately 601 and 32 ka in the
upper Mississippi Valley (Leigh and Knox 1993). However, Markewieh et al. (1998)
suggested that in the middle Mississippi Valley, the Roxana Formation is as old as 65 to
55 ka, using 10Be and TL dating techniques. These older dates may be more reliable,
because the dates reported by Leigh and Knox (1993) are at and beyond the limit of
radiocarbon dating. Correlative units are the Pisgah Formation in Iowa (Fonnan et al.
1992) and the Gilman Canyon Formation in areas west of the Missouri River (Mason et
al. 2007). It should be noted that these correlative units have younger ages

(approximately 49 ka) (Leigh and Knox 1993; F ollmer 1996; Mat and Johnson 1996).

Table 2.1: Loess chronology and stratigraphy in the Upper Mississippi Valley.

 

   

   

 

 

 

 

Age* Glacial Lithostratigraphy/Soil
(cal. vr) Episode Strati_ra h-v
4““ Modem Soil
Late

.g Peoria Loess

'5'

E

3
~25ka
~30ka 7 F armdale Geosol

Middle/
Early _ . j...

~65ka
430'“! Sangamon Geosol

=

a

'5

.E

E Loveland Loess
~185ka

 

 

*Age denotes timing of loess deposition. and does not relate to timing of soil development.

 

l . . . . . .
Upper RC age lrmrt (55 yr BP) converted to calendar years by extrapolation the calibration chart provrded
at: http. ” " ' lden mlumbia edu/cgi-bin/radcarbcfig (Fairbanks et al. 2005).

14

Peoria loess, which was deposited during the last glacial maximum, is the
youngest and thickest loess unit east of the Missouri River (Bettis et al. 2003). Although
Peoria Loess is found throughout the Midwest and Great Plains, deposition was neither
synchronous nor consistent. In the upper and middle Mississippi Valley, deposition of
Peoria Loess began approximately 30 ka and ended approximately 12.9 ka (Leigh and
Knox 1993; Markewieh et al. 1998; Grimley 2000; Forman and Pierson 2002; Bettis et a1.
2003). However, accumulation west of the Missouri River began between approximately
26.5 and 23.9 ka (Arbogast and Johnson 1998; Bettis et a1. 2003). Additionally, thick
deposits of Peoria Loess reveal that loess accumulation was variable overtime. For
example, PSD and chemical properties of the Peoria Formation in western Iowa indicate
that at least three episodes of deposition occurred, two of which were from different

source areas (Muhs and Bettis 2000).

2.3 Source Areas and Provenance in North America

Most thick loess deposits in North America are adjacent to and derived from river
valleys and valley trains (F ehrenbaeher ct al. 1965; Smalley 1966; Ruhe and Olson 1980;
Ruhe I983; Grimley 2000; Bettis et al. 2003). In the central lowlands region,
glaciofluvial sediment is often the dominant source of loess; thus, loess sheets derived
from these sources record, at a minimum, changes in ice sheet dynamics (Johnson and
Follmer 1989; Bettis et al. 2003). When nonglaeial source sediments (i.e. periglacially or
nonglaeially eroded bedrock or unconsolidated sediments) are cited as a constituent to a
loess deposit, the sediment is still often thought to have been first washed into a drainage
basin before it is deflated (Ruhe and Olson 1980; Johnson and Follmer 1989; Grimley

2000). Loess in northeastern Colorado and Nebraska, however, was derived mostly from

15

the Tertiary-aged White River group, which is a silt-rich bedrock unit that crops out north
and northwest of the loess deposit (Aleinikoff et al. 1998; Aleinikoff et al. 1999; Muhs et
al. 1999; Mason 2001). Aleinikoff et al. (1999) also suggest that the contributions of
glaciogenic and non-glaciogenic loess in northeastern Colorado occurred during different
seasons. They proposed that during warmer periods of the last glacial, the loess source
was dominantly from melting alpine glaciers in the Rocky Mountains. During colder
periods when vegetation cover was likely reduced, the loess source was dominantly the
White River Group. Loess of the Palouse region of Washington State was not directly
derived from a glaciogenic source, but instead from fine-grained lacustrine sediment
following the draining of proglacial Lake Missoula (McDonald and Busacca 1992).
Similarly, in the Great Lakes region, Schaetzl and Loope (Schaetzl and Loope 2008)
suggested thin loess in the upper peninsula of Michigan deflated from recently drained

glacial lake beds.

2.4 Loess in Wisconsin

Loess covers much of Wisconsin (Figure 2.1; Hole 1968). The thickest loess is
found in the Driftless Area of Wisconsin (DA), where it is up to ~ 5 m thick on uplands
east of the Mississippi River Valley. Elsewhere in Wisconsin, loess cover is generally
less than 1.5 m thick, and in some areas (i.e. eastern Wisconsin) it is undetectable, if not
absent altogether (Figure 2.1; Hole 1968). The following section presents a brief

overview of loess characteristics and research in Wisconsin.

40 Miles

(I’—I-"12;‘—l

0 20 40 Kilometers

 

 

 

 

 

 

 

Loess ,
1:] > 0.2 m I. ' . ,
1:] 0.2 - 0.6 m ; . ”LT 7
-0.6-1.2m ' I
-1.2-2.4m "TERI
- 2.4 - 4.9 m , r ‘5.
D Study Area 3.11.2
Sand Deposlts m-.-)
- Active Dunes Present __ I.
5 Sandy Areas Subject to Blowing

D Study Area

Figure 2.1: Loess and sand deposits of Wisconsin (alter Hole 1968).

2.4.1 Loess in the Driftless Area
In general only four or five loess deposits can be found on upland summits of
stream divides in the DA (Leigh and Knox 1994). Jacobs and Knox (1994) identified six

units at an interfluve within the DA, but this is uncommon, and a strictly eolian origin is

17

not certain for all six. These loess deposits differ from their type sections in Illinois and
Iowa, as they are typically finer grained and thinner (Leigh and Knox 1994). The
thickest total accumulations of loess are adjacent to the Mississippi and Wisconsin River
Valleys, where they are typically 10-12 m thick (Figure 2.1; Hole 1968; Leigh and Knox
1994). Leigh and Knox (1994) also noted that loess thickness decreases toward the east,
to less than one meter thickness over a distance of 40 km.

Intense erosion associated with periglacial conditions has completely erased pre-
late Wisconsin loess deposits from the stratigraphic record (Jacobs et al. 1997; Mason
and Knox 1997). Jacobs et a1. (1997) argued that the incomplete record of
paleoenvironmental change prior to the late Pleistocene is likely due partially to the
erosion of older deposits, but mainly to extensive weathering, rendering the deposits
unrecognizable, and thus, classified as residuum. Additionally, as Jacobs and Knox
(1994) pointed out, older loess deposited on landscape positions other than the flattest,
most stable uplands, were likely reworked by solifluction, resulting in secondary loess
deposits (loess-derived colluvium). This does not imply that preserved loess deposits in
the DA have been spared modification, as even relatively flat, stable uplands probably
experienced alteration by cryoturbation processes (Leigh and Knox 1994; Mason 1995;
Mason and Knox 1997).

The oldest, named loess deposit in the DA is the Wyalusing Formation (Leigh and
Knox 1994). The Wyalusing Loess was likely deposited during, if not prior to, the
Illinoian glaciation (Leigh and Knox 1994). Overlying the Wyalusing Formation is the
Loveland Formation, which consists of Loveland loess and the Sangamon Geosol (Leigh

and Knox 1994). In the DA of Wisconsin, the Loveland Loess is typically < 2 m thick

18

(Leigh and Knox 1994). Ages of the Loveland Formation have not been determined in
the DA. However, it can be identified based on pedologic, sedimentologic, and
stratigraphic similarities to type sections elsewhere in the Midwest (Leigh and Knox
1994). The Roxana Formation overlies the Loveland Formation, and in the DA of
Wisconsin, the Roxana loess is no more than 1.5 m thick (Leigh and Knox 1994).

Peoria loess is the thickest (approximately 7 m) and most well preserved of the
four loess deposits commonly found in the DA of Wisconsin (Leigh and Knox 1994); and
was deposited between approximately 25 and 11 ka (Leigh and Knox 1993; Markewieh
et al. 1998; Bettis et al. 2003). The base of the Peoria Loess is a massive, calcareous, and
light brownish-gray (2.5YR 6/2) to yellowish brown (lOYR 5/4) silt loam, indicating that
it has not been weathered (Leigh and Knox 1994). Peoria loess was likely deposited
initially during a time of landscape instability as the contact between Peoria loess is
interbeded with secondary (colluvial) Roxana loess (J aeobs and Knox 1994; Jacobs et al.

1997; Bettis et al. 2003).

2.4.2 Loess in North Central Wisconsin

Loess, albeit thin, has long been recognized as a surficial deposit that is generally
less than 1.5 m thick across much of north central Wisconsin (Hole 1942, 1968; Cahow
1976; Hole et a1. 1976; Bartelme and Strelow 1977; Attig and Muldoon 1989; Fiala 1989;
Sutherland 1989; Clayton 1991; Attig 1993; Simonson and Lorenz 2002; Syverson 2007).
Even though loess is widely accepted as a soil parent material in this part of Wisconsin,
few studies have characterized those deposits, and no studies have focused on the north
central Wisconsin loess sheet itself. The following is a summary of research that has

included some aspect of the north central Wisconsin loess sheet.

19

In his dissertation, which sought to correlate the ages of drift in north central
Wisconsin, Francis Hole (1942) described a surficial deposit between 60 and 90 cm thick,
overlying drift as well as sandstone and crystalline bedrock residuum. Hole believed the
surficial deposit to be late Pleistocene loess, due to its 1) tendency to blanket all types of
topography, 2) overlying different substratum (glacial drift, sandstone residuum,
crystalline residuum), 3) having a mechanical composition different than the underlying
substratum (often with sharp contact between the two) and yet similar to that of the
Peoria Loess in Iowa, and 4) being relatively unweathered. Hole noted that the silt
mineralogy of the loess was similar to that of the underlying drift; however, the same
varieties of minerals were found on Rib Hill, and therefore could not likely be ascribed
only to admixing with the underlying drift. He further surmised that the source of the
north central Wisconsin loess was the local drift sheets as well as the exposed Cambrian
sandstones to the south. In this proposed scenario, eolian erosion on the local drift sheets
would have happened synchronously with loess deposition — the process of which is
difficult to explain.

Sutherland briefly discussed north central Wisconsin loess in her 1989 Master’s
thesis. While the focus of her research was the glacial tills underlying loess deposits in
the area, she noted that areas of thickest loess accumulation were along the Late
Wisconsin terminal moraine in Taylor County, and along the Black River in Clark
County. Based on the general thickness trends, she hypothesized that the Black River
was the source of the thick loess in Clark County, and the outwash plains associated with

the Late Wisconsin terminal moraine were the sources of thick loess in Taylor County.

20

In Cahow’s dissertation (1976), he focused primarily on the glacial
geomorphology of the southwest Chippewa lobe moraine, but also discussed the thick
loess deposits in western Chippewa County (Figure 2.2). Cahow suggested the outwash
of the Copper Falls Member as the source of the thick loess directly east of the outwash
plains. He also described thinner (60-120 cm thick) loess in southeastern Chippewa
County (outside his thesis area). Cahow classified this loess as anomalous because there
is no outwash plain/valley trains associated with the adjacent north-south part of the
Yellow River. Instead, he argued that the extensive valley train of the Chippewa River

(in the vicinity of Lake Wissota) is the source of the loess.

2.5 Summary

Mapping the spatial characteristics of loess deposits is an important step to
understanding the paleoenvironmental conditions that occurred during loess deposition.
Loess deposits have characteristic spatial trends with respect to thickness and particle size
data, sometimes enabling the determination of source areas. The thickest and coarsest
areas of loess deposits are usually proximal to source areas, whereas distal parts of the
loess deposit are often thinner and have finer (fine silt and clay) particle sizes. After
loess provenance is identified, paleoenvironmental conditions during loess deposition,

e. g., temperature, wind direction, etc. can often be inferred. Additionally, obtaining ages
of loess deposits can be used to constrain the chronology of the paleoenvironmental

conditions under which loess deposition occurred.

21

.30.: 0580.: 52.323 33 JJ.
omem D

SETH

              

cones: 2.3. 3&8. 5235 o2 - B D
a. on? I o2 A I
con—5: 2.3. 335 £33.30 Ea. $25.25 300..
292.22 N o
.JIIJ

438%? 3.3.3.3?) «539:0

05:0 Eat 5:5.» ewes 2: 50¢
giant me? .335 E «SEE—53 7.35:?
2: 2 3.5. 5:55 2: BEE—am

o: <50—.9». 2.: E 35:53 £92 x35
e5 comm—.36 32: V .3230 team
sombiw ”3mm: Ema Ea 3an 5:3
Emsoomrg SEED «325:5 E 332:
$5096 $02 95 £3330 "N.~ 2:»:—

uu .

 

22

Although thick loess deposits have been studied in detail where they occur across
North America, relatively thin loess sheets are often ignored. Thick loess deposits in the
Central Lowlands, US, are adjacent to and mostly derived from broad river valleys or
valley trains, which record ice sheet dynamics, not climatic fluctuations as in the case of
the Great Plains loess sheet and that of the Palouse region (McDonald and Busacca 1992;
Aleinikoff et al. 1999; Bettis et al. 2003). Thick loess in the Driftless Area of Wisconsin,
like thick loess elsewhere in the Central Lowland region, has been thoroughly studied;
however, the relatively thin loess that covers much of the rest of Wisconsin has not
received much attention. One relatively thin loess sheet, in north central Wisconsin, has
a more complex spatial distribution. Understanding not only the provenance, but also the
paleoenvironmental conditions under which this loess was deposited may provide
important clues into regional paleoenvironmental variations that the thick loess of the

Driftless Area cannot.

23

3. Study Area
The focus of this study is the north central Wisconsin loess2 sheet, which is

located primarily in Clark, Marathon, and Wood Counties (Figure 3.1). Loess in this area

4}
1
l

y!

       

Figure 3.1: Spatial distribution ofloess soils in Loess thickness (cm) 0 2 5 5' Kilometers
north central Wisconsin (after Bartelme and - > 100 ‘
Strelow 1977; Fiala 1989; Simonson and

Lorenz 2002). Dark red represents soil series E 51 ' 100
mapped where loess thickness is >100 cm. Red [:1 25 _ 50
represents soil series mapped where loess

thickness is 51—100 cm. Pink represents soil A '5'“
series mapped where loess thickness is 25-50 D Study Area

cm' I I County line

 

2 Although this is the first study to focus on the characterization of the north central Wisconsin silt mantle,
numerous studies have previously recognized the silt learn mantle as loess (Hole 1942, I943: Cahow 1976;
Hole et al. 1976; Bartelme and Strelow 1977: Attig and Muldoon 1989; Fiala 1989; Sutherland 1989;
Clayton 1991; Attig 1993; Simonson and Lorenz 2002). Because this sediment has been widely accepted
as loess. and for the sake of writing clarity. I will refer to the silt loam mantle in north central Wisconsin as
loess for the remainder of this study, rather than as a silt mantle.

24

has been widely recognized, especially where it overlies glacial drift (Weidman 1907;
Hole 1942; Hole et al. 1976; Attig and Muldoon 1989; Clayton 1991), but has been also
recognized overlying bedrock residuum (Hole 1942; Hole et al. 1976). The loess sheet
appears to have an abrupt western boundary in central Clark County, based on soil series
that list loess as a parent material (Figure 3.1; Bartelme and Strelow 1977; F iala 1989;
Simonson and Lorenz 2002). To the west of this border, Cambrian sandstones
interbedded with shale crop out, as the topography transitions into the dissected bedrock
uplands of the Driftless Area (Figures 3.1, 3.2). This chapter provides background
information as to the Cambrian lithostratigraphy, glacial lithostratigraphy,
paleoenvironments, modern environment, and soils relevant to the north central

Wisconsin loess sheet.

25

 
    

LLLL Late Wisconsin moraine (LGM)

D Study Area

Uncoln Q.

Taylor
Chippewa
Marathon
Eau Claire
Portage

Wood
Jo ckson

Adams
Juneau

   

   

0 20 40 Miles

0 20 40 Kilometers

Figure 3.2: DEM of Wisconsin showing the study area outlined by a black rectangle in Clark,
Marathon, and Wood Counties. The extent of the last glacial maximum (the Late Wisconsin
Glaciation) is shown by a dark, hashed line. The extent/ approximate extent of the Driftless Area (an
area lacking evidence of being glaciated at any point during the Quaternary) is represented by a

dashed line/dashed line with question marks where the extent is difficult to delineate (afier Clayton et
al. 2006).

26

3.1 Bedrock Geology
The study area lies in a transitional zone between Precambrian crystalline bedrock
of the Laurentian Shield in the north and northeast, and the overlying, Paleozoic

Cambrian sandstones in the west and southwest (Figure 3.3; Mudrey et al. 2007). In

.3 I

I

   
 
 
 

0 20 40 80 Miles

    

80 Kilometers

  

Wisconsin Bedrock
Devonian
Silurian
[:1 Ordovician
:1 Cambrian
- Precambrian

Water
D Study Area

Figure 3.3: Bedrock of Wisconsin (after Mudrey et al. 2007). The study area lies within a transitional
zone in which Precambrian crystalline bedrock outcrops decrease and Cambrian sandstone outcrops

increase towards the southwest.

27

some locations within this transitional zone, rivers have incised through the Cambrian
bedrock, exposing the underlying Precambrian bedrock (Figure 3.3). Collectively, the
Cambrian sandstones of north central Wisconsin are weakly lithified, readily weathering
into sandy or sandy loam rcgolith and soils (Simonson and Lorenz 2002; Weidman

1907). These sandstones are subdivided into three formations. Using the nomenclature
Ostrom (1967) provided, they are from oldest to youngest, the Mt. Simon, Eau Claire,
and Wonewoc Formations (Table 3.1). The following section provides a brief
description of the lithostratigraphy, texture, and mineralogy of the Mt. Simon, Eau Claire,

and Wonewoc Formations.

28

 

8.08 EoEomem 2358805

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

E m8 2: E @5253 con: .
cm _ ”ocoumccem eoEflm be: use 5562: .3300 .m :2 3
ozoamecfi 3:525 M. 552585
E on new .mzoooazaae .eoESM on: bo> ecu oEm EEO :mm W
. . . o S
ecouwncem eoEEw E5682 BESEO m u. no
E cm BoswzoE. cease: 83253 P m w.
958.52 ”ocoumccew coEEw 25 can E3632 w. m.
28.35% 89: E u u
E oEcoozflw EB .oEummmEfl .eoEEw 5:602 v— .m oo 25 3 M cEconscwE
co 2852mm soteEE 62:me 82on .bzm :anH v_ m A ,M. m
0.5523 02.5233 .eoEEw EEBE ocom 1
395—2.:
.52: 322:: ESE-:5 33:52 Eta—Eon— QEZU uufim 829m Eoummm
.xo...5<
A52 5050 Baxooecom FEW—58.3.3 EEEozoEoZ oimmcwzebw "fin 93.;

 

29

3.1.1 Mt. Simon Formation

The Mt. Simon F omiation is the uppemiost bedrock unit in the custom parts of the
study area along the C ambrian/‘Preeambrian transition. and unconfomrnbly overlies

Precambrian bedrock (Figure 3.4: Table 3.1).

F Int ‘1‘! {u

‘-
r

L Chippewa Co.

Clark C0.

 

 

 

‘ -.r . .Trempealeau Co. ‘ 3: Jackson Co,
‘.‘~ .1 r - - .

 

Bedrock Formation « 7 '
[:I Wonewoc Dandy Area /5/. . If},
B Eau Claire [:1 County borders 0 5 10 Miles ;' “ " ' ~ ‘-
-Mt.Simon I—T—‘r—J . g.
- Precambrian 0 5 10 Kilometers ', , .2 ‘ ' a
Figure 3.4: Spatial distribution of bedrock formations in north Central Wisconsin (after Mudrey 1987;
Brown and Patterson 1988).

30

3.1.1.1 Mt. Simon Lithostratigraphy

The Mt. Simon formation can be
subdivided into three lithologically distinct
units (Figure 3.5; Asthana 1969). The
lowest unit unconforrnably overlies
Precambrian bedrock. Thick conglomerate
beds with overlying pebbly sandstones are
usually present at the base of this unit
(Ostrom 1966, 1970; Asthana 1969). The
lower unit is the most poorly sorted of the
three units within the Mt. Simon Formation,
with grain sizes ranging from 8 to <0.625
mm; shale partings (Figure 3.6) are more
common at the base and top of this unit
(Asthana 1969). The sandstones of the
middle unit are also classified as poorly
sorted, with grain sizes ranging from 1.4 to
<0.625 mm; however, beds consist of 0.71
to 0.25 mm (coarse and medium) grains
(Asthana I969). The middle unit is
massively bedded, with shale partings more

common at the top (Figures 3.5, 3.6). The

 

 

,. .. --.-.-'.
-0 .J.‘ .

Ejfe 1’
0 . -

Upper unit: Sandstone,
yellow, coarse to fine, some
very fine, some shale and silt,
cross-laminated, massive
bedded, fossiliferous, worm

' . borings.

 

' _° Middle unit: Sandstone,

. 7.4
i

—‘--_4

3:1

'- base poorly to very poorly

” - yellow brown to yellow gray,
' coarse to medium, some fine
_ f shale partings at or near top,
‘ " poorly sorted, cross-
' laminated, massive bedded.

 

1

 

 

31

Lower unit: Sandstone,
conglomerate, yellow brown,
very coarse to fine, very little

. "3 fine, thick conglomerate beds

31.9%.“!?9

at the base, pebbly above,
some shale and silt near the

sorted, cross-laminated,
massive bedded.

Figure 3.5: Stratigraphic column of the
Mt. Simon formation at Mt. Simon, WI
(after Asthana 1969).

 

Figure 3.6: Photographs of shale partings and laminae found within Cambrian sandstone outcrops in Clark
County, WI. Top: 851; Bottom: SSS. Photographs by K. Stanley.

32

upper unit is often characterized as a transitional unit where medium and coarse grained
beds (typical of Mt. Simon lithologies) alternate with finer-grained beds (typical of Eau
Claire lithologies) (Ostrom 1966, 1970; Morrison 1968; Asthana 1969; Distefano 1973).
In general, this upper unit consists of yellow-gray (Figure 3.7), coarse to fine sandstones
in which silt and clay are abundant (Asthana 1969). Sorting within the beds of the upper
unit are better than the
other two units within
the Mt. Simon
Formation, and the
contact between the beds
is sharp (Morrison 1968;
Asthana 1969).

The transitional
nature of the upper unit
makes it difficult to
place a discrete
boundary between the
Mt. Simon and overlying

Eau Claire Formations.

 

This difficulty,

Figure 3.7: Photograph of the sandstone exposure at $82. This
exposure is likely part of the upper Mt. Simon Formation based on the
alternating beds of yellow-gray, coarse to fine grained sandstones.
Photograph by K. Stanley.

combined with the
presence of strongly

iron-stained beds at the

33

top of the unit, provoked some researchers to designate this transition zone the Shawtown

Formation, colloquially known as the ‘rusty foot’ of the Eau Claire Formation (Figure

3.8, 3.9; Morrison 1968; Raasch and Unfer 1964).

 

Figure 3.8: Iron stained beds at the top of the Mt. Simon Formation (Shawtown Formation) at exposure
site SSS. Photograph by K. Stanley.

34

 

l
I

I

r|:;ilii";i|'riiti
. . i ii
flil‘llill"

     

l-

lill
I

 

. I'II'"
.;Il=

 

..
I 4|

u
.—..o

 

 

Sandstone: tan, mostly fine grained, some clay,
very glauconitic, thin-bedded.

Sandstone: tan to red-tan; fine grained, sub-
rounded to rounded, angular; medium sorting;
very glauconitic, thick bedded-massive, ripple
marks, few thin clay interbeds.

Sandstone-clay: similar to unit #8.

Sandstone: like unit #7; delicate laminations in
the glauconite; abundant brachiopods and
Crepicephalus.

Sandstone—clay: red-tan, fine grained, thin-
bedded or irregular bedded; often very
glauconitic, non-resistant, especially near top.
Sandstone: tan, fine to very fine grained,
angular, glauconitic, well-indurated, massive
bedding.

Sandstone: dark buff, fine grained, well-sorted,
subrounded, thick-bedded; Cedaria near bottom
in thinner beds.

Sandstone: similar to unit #4, but thin-bedded,
many clay partings; very few fossils; non-
resistant.

Sandstone: buff; fine grained, well-sorted,
glauconitic, thick-bedded; few clay interbeds.
Sandstone-Shale: interbedded; sandstone is
gray-buff, fine grained, glauconitic, thin-bedded;
clay partings, bottom marks; generally “dirty”
appearance in upper 5’; definite bedding.
Sandstone: medium grained, well-rounded, well-
sorted, iron-stained.

Sandstone-Clay: sandstone is tan to purplish tan,
fine grained, well-sorted, angular to subangular,
slightly glauconitic; interbedded, irregular
bedding; sands become slightly thicker bedded
going up the section.

Sandstone: type similar to unit X; “Rusty Foot.”
Sandstone: light-tan to buff; fine to very fine
grained, medium sorting; subangular, slightly
glauconitic, numerous greenish-gray clay
partings; bottom marks (trace fossils?) and
Oboloid brachiopods abundant.

Sandstone: light tan to red-brown; medium
grained, well-rounded, well-sorted, iron
concretions in bands; fairly massive, but
interbedded with thin beds of sandstone type Y.

1.8m

1.2m
0.5m

2.6 m

0.5 m

1.6m

0.9 m

2.5 m

0.7 m

2.7 m

0.2 m

5.5 m
0.3 m

1.8m

2.1m

Figure 3.9: Detailed stratigraphic column ofthe Eau Claire Formation from Eau Claire County. T.27N.,
R.10W, Sec. 26. SW ‘A. center: scale: 1” = 4’ (after Morrison 1968).

35

3.1.1.2 Mt. Simon Texture

The Mt. Simon formation tends to be poorly sorted, with grain sizes ranging from

8 to <0.0625 mm (Asthanal969); adjective beds are, however, commonly composed of
medium and coarse grained sand (Distefano 1973). Distefano (1973) reported a sorting
coefficient of 0.85 (moderately sorted) and a mean grain size of 0.277 mm (medium
sand) (Table 3.2). Distefano suggested that the grain size and sorting results may be
biased due to the number of samples taken from the upper portion of the Mt. Simon
formation, which is finer and better sorted. Shale partings and laminae, siltstones, and
granules are a minor constituent throughout the most of the formation (Asthana 1969;

Ostrom 1970).

3.1.1.3 Mt. Simon Mineralogy

The light mineral component of the Mt. Simon Formation is dominantly quartz
(Asthana 1969; Distefano 1973). Although a minor component, feldspar contents
commonly average 3% (Ostrom 1970); it is most abundant in fine-grained beds (Table
3.2; Distefano 1973). The high feldspar contents that Asthana (1969) reported (as high as
40%) may reflect a sampling bias in the upper, finer grained unit. Heavy minerals make
up less than 1% of the bulk mineralogy and include magnetite, ilmenite, luecoxene,
zircon, tourmaline, and garnet (Asthana 1969; Ostrom 1970). Out of these minerals,
garnet was the most common, and its concentration appears to decrease with depth
(Asthana 1969; Ostrom 1970). Asthana (1969) also found zircon to be concentrated in

the fine (<0.0625 mm) fraction.

36

 

.83 eoEEw
ecu E 3:38

wagon
mmoeo 0E8 AEE

cones—:8
05 Socwzoefi 958

8:3 EEBEV

 

 

 

 

 

 

$9. 9 a: "fem 380m 0:53 3632 :5 5:03:28 5E2 SE Fwd 2.8 0858 use 8:622 :oEm s2
83 95 .erEE
.metma .meEoo
Efiw mm 958 22m
:03 o. :25 833%: 52.258 Seawaefi 3:3 26 but
{OER 288282 0603 E5. ween E coEEoU EE de 2.8 25 .63 use oEm EEO new
erEE e5... 85% 59:82
metea 8 e55.“ 293 EEBE 2 on: boi 98m 858 25 8:622 o_=>mo_a0
can we :wE we Eozmccfl on. Co a2 EE 83:82
33 Eutazm ax. _ ._ :03 3.08 63322 05 E 5888 802 Fed -EE 3 _.o 9:3. 25 ecu 82:52 :25: 8325?
2.2.2.53 35:35.5 9W5:
Ema—:0.— ouaeo>< ”Etom com ch—oom 225 ES Em on: 5an 5:52 on: 55% 5.55.5 5.35.8..—

 

 

 

 

 

 

 

 

.853 :8wa ”m3. c.5035 ”woo. 52:32 602 «SEW/x. 5:3 mozmtouofleco .5szqu 89328 :EBEaU Co Emacs—am "Nd 2:5.

37

3.1.2 Eau Claire Formation

The Eau Claire Formation, which overlies the Mt. Simon Formation, occurs to the

west of the study area (Figure 3.4; Table 3.1).

3.1.2.1 Eau Claire Lithostratigraphy

The Eau Claire Formation can be separated into 4-5 units based on particle sizes,
sorting coefficients, and mineralogies (Figs. 3.8, 3.9). The lowest unit, which overlies the
Mt. Simon Formation, averages approximately 10.5 m in thickness, and is composed of
fine and very fine sandstone, thinly interbedded (5-9 cm thick on average) with lenses of
shale and siltstone (Distefano 1973). The interbedded shales and siltstones ofien exhibit
deformation due to lithification (Morrison 1968; Ostrom 1970; Distefano 1973). In some
locations, this unit is subdivided into two units, in which the basal portion (approximately
4.6 m) contains more clay and the individual beds are indistinct, whereas the upper 6.1 m
of this lowest unit contains distinct thin beds and more glauconite (Morrison 1968).

Overlying the lowest thin-bedded unit is a massive, glauconite-rich unit, which is
approximately 7.6 m thick (Figure 3.10). This lower, massively bedded unit contains
minor amounts of thinly interbedded sandstones and shales; however each massive
sandstone bed is separated by sandstone beds that contain much clay-rich shale (Morrison
1968). The lower unit is overlain by an upper thinly-bedded, glauconitic sandstone unit
that is approximately 4.6 m thick, which is sometimes overlain by 6.1 m of the upper
massively-bedded, glauconitic sandstone unit (Morrison 1968; Distefano 1973).
Regardless if the unit is missing or not, contact between the Eau Claire Formation and the

overlying Galesville Member of the Wonewoc Formation is sharp, and has resulted in

38

arguments supporting both the presence of an unconformity (Ostrom 1966; Morrison

1968) and the absence of one (Distefano 1973).

Upper Massive Beds

Upper Thin Beds

Lower Massive Beds

Lower Thin Beds

Shaly Beds

 

Sandstone: massive to submassive.
glauconitic: exposed only at Whitehall and
strum. About 6 m

Sandstone: thin. distinct bedding: very
glauconitic: usually missing from sections.
About 4.5 m

Sandstone: thick to massive bedding; often

very glauconitic; few very clay-rich,

irregularly bedded units separating the more
characteristic massive units. About 7.5 m

Sandstone: mixed thin and thick bedded; high

clay content (less than A); more glauconite

than below; thin beds more regular and

distinct; mica common. About 6 m

Sandstone-Shale: very thin bedded; very high
clay content; individual beds often indistinct
and seldom over 3” thick. About 4.5 m

Figure 3.10: Stratigraphic column showing the generalized characteristics ofthe five subunits within the

Eau Claire Formation according to Morrison (1968).

3.1.2.2 Eau Claire Texture

The Eau Claire formation has finer textures than either of the formations

confining it. Clay that is present in this formation is found either coating the fine sand

grains, as thin shale partings, or as shale beds that can be up to 3.7 m thick (Ostrom

1970). The mean particle size reported by Distefano (1973) is 0.073 mm: very fine sand

(Table 3.2). Distefano also found the Eau Claire formation to be moderately well to well-

sorted; however, he excluded shale samples from both the sorting and texture data.

39

3.1.2.3 Eau Claire Mineralogy

Distefano (1973) classified the Eau Claire Formation as a fine and very fine,
feldspathic sandstone. Although most of the grains within it are quartz (usually detrital
quartz, Morrison 1968), feldspar contents can exceed 40% and is most highly
concentrated in the <0. 125mm (fine sand) fraction (Distefano 1973; Odom 1975).
Morrison (1968) also found most of the feldspar present was a monoclinic form of
orthoelase. Less than one percent of the sandstone is comprised of heavy minerals,
including garnet (dominantly), tourmaline, zircon, ilmenite, magnetite, and leucoxene
(Morrison 1968; Ostrom 1970).

Glauconite, a clay mineral with a 10A mica structure, commonly found in the
form of green pellets in sandstones, is more abundant in the Eau Claire Formation than
either of the underlying Mt. Simon or overlying Galesville Formations (Distefano 1973).
Greenish glauconite pellets are found in the very fine sand and silt facies, but never in the

coarser sands or shales (Morrison 1968; Ostrom 1970; Distefano 1973).

3.1.3 Wonewoc Formation

The Wonewoc Formation is found overlying the Eau Claire formation west of the
study area and is subdivided into the Galesville (lower) and Ironton (upper) members

(Figure 3.4; Table 3.1).

40

3.1.3.1 Wonewoc Lithostratigraphy

The Galesville and Ironton members are differentiated based on grain size
characteristics (Figure 3.1 l). C onglomeratc can sometimes be found at the base ofthe
Galesville member. overlying the upper. thin bedded units of the Eau Claire Formation
(Morrison 1968). Otherwise. the Galesville Member consists ofelean medium and fine
grained sandstone beds that are usually massively bedded and cross—bedded (Emrieh
1966; Morrison 1968; Ostrom 1970). Each bed tends to be well sorted with few coarse or
very fine grains (Ostrom 1970). In general, the overlying Ironton Member is coarser
grained than the Galesville Member and is massively bedded and cross bedded (Emiich
1966; Ostrom 1970). Like the upper unit of the Mt. Simon formation, the Ironton
Member appears to be a transitional unit between the cleaner, massively bedded
Galesville Member and the overlying argillaceous and thin bedded sandstone of the Lone
Rock Formation (Ostrom 1970). Coarse grained beds of the lronton are found
interbedded with poorly sorted silty to clay-rich strata (Ermrich 1966; Ostrom 1970). At
the type section in Trempealeau County, Wisconsin (approximately 50 km southwest of
the study area), the Ironton Member fines and becomes fossiliferous towards the top,

where it becomes thin bedded and glauconitic (Figure 3.1 l; Ermrich 1966).

3.1.3.2 Wonewoc Texture

In general, the Galesville Member ofthe Wonewoc Formation is finer grained
than the overlying Ironton Member. The beds within the Wonewoc Formation have
median grain-size diameters that range from 0.1 14 (very fine sand) to 0.471 mm (medium

sand) (Table 3.2; Emrich 1966). According to Distefano (1973), the Galesville Member

41

East -* inc and very fine grained, trace of medium and

 
   

LOW :7 . coarse; few brachiopod shell fragments.
. Wonewoc Fm. edium and fine grained, trace of coarse;
Medium and coarse grained, little - medium bedded.
fine; few brachiopod shell 3‘ Fine and very fine grained, trace of medium; thin
fragments; thick- bedded bedded.
Medium grained, trace of coarse _, inc and medium grained, trace of coarse; thick-
and fine. ' . '.-.-j.‘_- bedded.
: " Medium grained, little coarse and fine; thin to
medium-bedded.
Coarse to medium-grained, trace fine; thick-
bedded with some thin beds.

‘ Ironton Member
5.- .. Galesville Member
Fine to very fine gi-ainettL~ 1.3? g,’ Fine to medium grained, little coarse; thick bedded

4'. -' $.-

trace medium; thin bedded. '7’ ._ ° —- with few thin beds. . .
Fine to medium grained, e oarse to medium grained, little fine; thick-bedded.
“me marge: thick—h ed d ed. —>/..</ Medium grained, little fine and coarse, thick-bedded.
:7 ,_ Fine to very fine grained, trace medium; thick-bedded.
10 F 2’7 edium to fine grained; thick bedded.
93’ a. Medium grained, little fine and coarse; thick-bedded.
- ' ' “- Fine grained, little medium and coarse; thick-bedded.

, , , \Medium grained, little fine and coarse; thick-bedded.
Fine grained, ““16 VCFY— , .2. \Fine grained, little very fine; thick-bedded.

fine; thick-bedded. 7
,,°_{'\.;l-,-. Medium and fine grained, little very coarse and very
&/ fine; thick-bedded.

,gK'f-jzé‘gfi Medium and fine grained, little coarse and fine; thick-bedded.
K-o; Very fine and fine grained; thin-bedded.

;~»/”‘Very fine and fine grained; Oboloid brachiopod shells
fés .gfi locally abundant; medium and thin-bedded; weathered
Wonewoc Fm surface pitted due to burrows and clay pebbles.
3‘“ 0““ “WE-”5' Very fine and fine grained; thin-bedded.

Figure 3.11: Detailed stratigraphic column of the Wonewoc formation (after Ostrom 1970).

has a mean grain size of 0.207 mm (fine sand) and has many fewer very fine sand beds
than the Mt. Simon Formation. Both members of the Wonewoc Formation are classified

as well sorted according to Emrich (1966). Distefano (1973) reported the Galesville

42

Member as moderately well sorted, but postulated that this result was biased due to the
number of samples taken in the more poorly sorted base where it grades into the upper

Eau Claire Formation.

3.1.3.3 Wonewoc Mineralogy

Feldspar content of both members of the Wonewoc Formation is low, averaging
1.1% (Table 3.2; Emrich 1966). Whereas very fine sandstone beds are rare, when found,
they and other silt-rich beds contain the highest (<5%) feldspar contents (Emrich 1966;
Odom 1975; KunleDare 2005). In these silt-rich beds, microcline, plagioclase, and
orthoclase feldspar have been identified. Heavy minerals account for less than 1% of the
sediments in the Galesville and Ironton Members, of which zircon and tourmaline are
consistently the most abundant (Wilgus 1933; Emrich 1966). Other heavy minerals
present in these members include garnet, ilmenite, and 1eucozene(Wilgus 1933; Emrich
1966; Ostrom 1970). Pyrite is only found in subsurface samples (Emrich 1966). Illite
was the dominant clay mineral in all samples Emrich (1966) studied, with only minor
amounts of kaolinite. Chlorite was present in the overlying Lone Rock Formation and

underlying Eau Claire Formation, but not in the Wonowoc Formation (Emrich 1966).

3.1.4 Summary of Bedrock Geology

Parts of North central Wisconsin are underlain by both Cambrian sandstones (to
the south and west) and Precambrian bedrock (to the north and east); the study area
occurs at a transitional zone between the two. The sandstones that make up the Cambrian
formations are weakly cemented and very friable. The Cambrian sandstones immediately

west of the study area include the Mt. Simon, Eau Claire, and Wonewoc Formations have

43

dominant grain sizes that range from medium to very fine sands depending on the
formation and bed examined. Some of these beds, especially within the Eau Claire
Formation, can be comprised of relatively thick (up to 3.7 m) siltstones and shalestones
(Figure 3.9, 3.10, 3.11). Shale partings and laminae can be found throughout the
formations, regardless of the individual bed grain size, sorting, or thickness (Table 3.2).
With the friable and fine grained nature of these sandstones interbedded with shale,
combined with their location west of the north central Wisconsin loess sheet it is not
surprising that Hole (1942) speculated that these Cambrian formations are the source of

the north central Wisconsin loess sheet.

44

3.2 Glacial History and Stratigraphy
North and east of the Driftless Area (Figure 3.2), the Cambrian sandstones
interbedded with shale, west of and within the study area, are overlain by an increasing

thickness of glacial deposits of varying ages (Figure 3.12). These deposits include sandy

Rusk Co.

Prrce Ca. _

Lincoln C5

I
-, .- Taylor Co.
‘ tang/ode Co.

I Chippewa Co .

  

Marathon (0 '
" .1 '4 "l 7 DStudyAroa
" J—l- Last Glacial Maxlmum
Llncoln Fomlatlon
- Merrill Member
[3 Bakerville Member
Marathon Formation
Wood (0 ‘ m Edgar Member

0 10 20Miles

   

Clark Co.

,I
.2

 

Jackson C o

_ . r_T—-fi
.. - ~-‘ 0 10 20 Kilometers
Figure 3.12: Locations of pre-late Wisconsin drifi members (afier Clayton et al. 2006). Edgar drift is the
oldest drift found in north central Wisconsin and is found mostly in Marathon and Wood Counties.
Bakerville overlies Edgar drift, but exhibits no glacial topography. Merrill also overlies Edgar drift and
drift shares similar characteristics as Bakerville Drift. but does exhibit some glacial topography.

or silty tills, outwash sand and gravels, and fine-grained lake sediments. Generally, the
glacial deposits in north central Wisconsin are grouped into two categories: Late
Wisconsin (formerly latc—Woodfordian) and pre-Late Wisconsin (sometimes referred to
as Border) drift (Weidman 1907; Hole 1943; Stewart 1973). In north central Wisconsin,

the terminal moraine of the Late Wisconsin glaciation winds west-to-east, mostly through

45

Chippewa, Taylor, Lincoln, and Langlade Counties (Figure 3.12). This moraine also
covers a small portion of northwestern Clark County, which lies within the study area
(Figure 3.12). Late Wisconsin glacial sediments are found on and behind the terminal
moraine, whereas pre-Late Wisconsin glacial sediments are found far beyond the moraine
(Figure 3.12). This section provides characteristics of the three pre-Late Wisconsin till
plains that occur at or near the surface within the study area, as well as characteristics of

the Late Wisconsin terminal moraine in Chippewa, Clark, and Marathon Counties.

3.2.1 Pre-Late Wisconsin Glaciation

Ages of Late Wisconsin Glacial events are not well constrained in Wisconsin
(Clayton et al. 2001), and the ages of pre-Late Wisconsin glaciations are even less
certain. Attempts to correlate pre-Late Wisconsin drift members to glacial events of
known ages are often based on speculation and relative-dating assumptions, rather than
quantifiable, absolute (numeric) age data (Weidman 1913; Hole 1943; Stewart 1973;
Stewart and Mickelson 1976); in north central Wisconsin, relative ages have been
determined for these deposits based on stratigraphy, topography, and clay mineralogy
(Weidman 1907; Stewart 1973; Stewart and Mickelson 1976). The following is a brief
discussion of the pre-Late Wisconsin drift deposits of the Marathon and Lincoln

Formations within the study area (Figure3.12; Table 3.3).

Table 3.3: Pre Late Wisconsin glaciations in north central Wisconsin (afier Clayton et a1. 2006).

 

 

Formation Members Phase Yrs BP
Lincoln Merrill l-lamberg >40,000
Bakerville Nasonville Illinoian? (>130,000?)
Marathon Edgar Milan (earlier than Illinoian)
Medford Stetsonville Possibly 460,000 or 780,000

 

46

3.2.1.1 Marathon formation

The Marathon Formation consists of three members, which are the oldest
Pleistocene deposits known to occur in the study area. From oldest to youngest these are
the Wausau Member, the Medford Member, and the Edgar Member (Attig and Muldoon
1989). Of these, only the Edgar Member occurs as an identifiable surface unit within the
study area and represents the youngest deposit of the Marathon Formation (Table 3.3;

Attig and Muldoon 1989). Characteristics of the Edgar Member are described below.

3.2.1.1.1 Edgar Plain Extent

Edgar drift exists at the surface or is overlain by less than a meter of loess in the
western third of Marathon County, northwestern Wood County, and in small portions of
northeastern Clark and southwestern Taylor Counties (Figure 3.12). The till sheet thins
and becomes patchy because of long-term erosion in central Marathon County, where the
regolith is delineated as ‘Undifferentiated Marathon Member’ (Attig and Muldoon 1989).
In areas north and west of Clark County, Edgar till underlies younger till units (Clayton
1991). In Clark County, the southern extent of the Edgar member is just south of the
Marshfield Moraine, within which 30-50 m of Edgar till is overlain by sand and gravel
followed by several meters of the younger Bakerville till (Clayton 1991). In Wood
County, Edgar till has been identified in hillslope material and on relatively stable stream
divides; thus, the Milan glaciation likely extended past the moraine and into central

Wood County (Clayton 1991).

47

3.2.1.1.2 Edgar Plain Topography

Edgar till is typically 6-12 m thick, allowing the dendritic drainage pattern of the
underlying bedrock to be expressed, albeit in subdued form, at the surface (Figure 3.12;
Attig and Muldoon 1989). The absence of prominent glacial geomorphology, i.e., the
lack of poorly integrated drainage patterns and hummocky topography, on the low—relief

Edgar till plain is commonly recognized as evidence of long-term erosion (Figure 3.13;

Weidman 1907; Attig and Muldoon 1989; Clayton 1991; Attig 1993), possibly

 

Figure 3.13: Photograph of the low-relief Edgar till plain. Photograph by K. Stanley.

accelerated by periglacial conditions during the Late Wisconsin (Attig and Muldoon
1989). Attig (1993) and Clayton (1991) attributed the lack of surface boulders in the
Edgar drift to long-term weathering and erosion. On the other hand, Stewart (1973)

suggested that the thinness and lack of terminal moraines associated with the drift could

48

be explained by the quick advance and retreat of a relatively clean ice mass, rather than
the effects of long-term erosion.

For many years, a moraine associated with Edgar till could not be found, mostly
due to the misinterpretations of the Marshfield Moraine. For example, Weidman (1907)
and Mode (1976) hypothesized that the moraine was comprised entirely of younger
Bakerville till (what Weidman called ‘Second Drift’) because this till is found at the
surface of the moraine. Alternatively, Hole (1942, 1943) and Hole et a1. (1976) surmised
that the Marshfield Moraine was a bedrock ridge overlain by thin till. However, through
careful examination of well logs, Clayton (1991) concluded that the Marshfield Moraine
was a palimpsest feature, mostly comprised of Edgar till overlain by a thin layer of
Bakerville till. Clayton (1991) also found evidence for three distinct episodes of till
deposition by the Milan glaciation within the Marshfield Moraine, where Edgar till is the
thickest and best preserved. Here, Edgar till is separated by thin layers of lake sediment,

outwash, or organic material (Clayton 1991).

3.2.1.1.3 Edgar Till Characteristics

In general, the Edgar till deposited by the Milan glaciation is distinct from other
till members in the region due to its color, high silt content, and calcareous and
fossiliferous lithologies. The most commonly reported moist Munsell color for Edgar
drift is 7.5YR 4/4 (brown), but it can range between lOYR 5/1 (gray) when unoxidized to
SYR 4/4 (dark reddish brown) (Table 3.4). Edgar till is usually loam in texture based on
the weighted mean of sandzsiltzclay ratios, which are 40:41 :19 (Table 3.5). The median
particle size diameter of Edgar till in Marathon County is 0.039 mm (medium sand)

(Attig and Muldoon 1989). Mode (1976) specifically noted high calcite contents in the

49

coarse sand fractions, while Hole (1943) actually quantified the calcite content,
estimating that the drift as a whole is 20% calcite by weight. Ninety percent of the 2-4
mm fraction in the Edgar member is calcite, in the form of both primary and secondary
carbonates (Hole 1942, 1943).

Hole described the primary carbonate as compact, water-worn, homogenous fine-
grained granules, while be likened the secondary carbonate to conglomerate fragments, in
which carbonate cemented sand, silt, and clay particles were less compact towards the
exterior. The high limestone/calcite content of the drift remains enigmatic, as local, up-
ice sources (Superior and Lake Michigan basins) are currently nonexistent (Attig 1993;
Mode 1976). The flow direction of the Milan Glaciation was determined to be toward the
southwest based on the long axis orientation of pebbles found within the Edgar till (Mode
1976). Clayton (1991) hypothesized that the source of this drift is in Manitoba, based on

the southwesterly flow direction and the carbonate Silurian fossils found in the drift.

50

as: 632 a 3m: 9 5:3 3:53» on 33:68 £6 :a 58 Ammo: 59.6% .3 cotomoe £28 .3222 _

 

v 9 Qm 8 m m>m 2 Mi:
“owcmm

 

<2 :5. ”:5 .555 .5553. 3.. E55 .555 5.5 o. .555 :5. .5555
55.555 5.. 5.... 3.5 55 .555 5:53.
555:5 5.55 .5585 5 Em 5: B. 5.5
Homeow—
Ee Eat .555 535 52.5.5 55 .555 5553.
5:. 53 .555 555.5 55.3 .555 555.... 5.5 55 55.3 .555 5552 58 .32 5.555
5:. ”:6 : .555
:5. 2.5 55 .555 5552 58
3.5 55 .555 .5552 58 -3... 52 .555 5552 53 5.5-3.. E55 .555 53. 252
3.... 55.: .555
:5. 53 .555 .5553. Sz 32 SS .5255.
2.5-3. 5.. E5 .555 .5553. :5. E56 .555
55 .555 .558. 5.5 5:. get 5.8m 32 :5. ES: .555 5.55:3 5.5 32 53.
2:. ~53 .555 55.53. :5. 52 .555 5553. 3.2. Est .555 5.5 5 .555 $5.
5:. EEC .555 55.5 5:. ES: .555 2.5. ES: .555 555:5 55 552:: E... 55.
S35 ”:5:
3:. m>m§ 230.5 haw 56?: 9 .580 63638::
5:. «>3 .555 5553. .555 5 a... o. 5:55.. 5:13. mid .555 55550 23. 5.5.555
2.x..— .SnneU 5.292 o___>._3_am haw—em— 3:383—

 

.EmcoUEB 35:8 65: E m:: E2823 Boi— cca E3833 east—rouafl £28 :85: E02 3am 2:5.

51

.E... ow :5. E: N 5252. @2228 mm 3.5% 33 Em £93 £5 E 3.
.81 mo :5. E: N 503.2. 833.8 mm cacao: £3 :6 .353 £5 E...

 

 

 

:55. 5.55 2.2.8 €83.55 3 . . :2 :55. 5.55 5. 5:8 .55: 5......8 55... 3.353
a u 5 2-92-8552 .8 u 5 2-253.933 5.3. .5550
2... u 5 3 . 5.. a. u 5 o. 52.5 5.... 25.55
5 n5 253m 5. u 5 9.2.5 5.15.5532 .52. 5.5...
:5 u 5 ... H . :3 E...” .5525
:5 u 5 253m .3 u 5 25mm 3.5. n =5 .85. ”.3.
2. u 5 252... :w u 5 m. 59% .3. u 5 5.9.55 5.5. .5852 E... 5.2

35.8 .5 u 5 5.2. . m :25: a u 5 532....

:25: 2.. u 5 53.05.. 35.8 Gm n 5 :53...
:N n 5 3.5.2.8 E u 5 5.5. 5.53.9. .55 u 5 5.25.53... :55. 5.5.55.5

2....— ..2550 Eta—2 2.3—3:5 amen.—

 

.:_m:oom_>> 35:8 5.5: E 2:. 5208.3 83 :5. E2853 2.5-830 8:8 hflunfimfifiwm ommb>< .m.n Saab

52

3.2.1.2 Lincoln Formation

The Lincoln Formation has two members, Bakerville and Merrill. It is younger
than the Marathon Formation. The Bakerville Member is associated with the Nasonville
Phase and overlies Edgar till (Attig and Muldoon 1989) (Table 3.3). Bakerville till has
been tentatively correlated with that of the River Falls Member in northwestern
Wisconsin, which is thought to be Illinoian in age (Baker et a1. 1983; Johnson 2000;
Syverson and Colgan 2004). The Merrill Member was deposited during the Hamburg

Phase of the Lincoln Formation (Attig and Muldoon 1989). A minimum limiting age of
45.3 ka3 for the Merrill till was based on two l4C ages obtained from organic rich silt and

clays (in a bog) overlying till of the Merrill Member (Stewart and Mickelson 1976).

3.2.1.2.1 Bakerville Plain Extent

Bakerville drift is at the surface (or buried under approximately 1 m of loess or
colluvium) across much of central and eastern Clark County (Figure 3.12). Bakerville
drift overlies Edgar till on the Marshfield moraine, where it extends from Clark into
Wood and Marathon Counties (Figure 3.12). Clayton et al. (2006) mapped Bakerville till
in both northeastern Eau Claire and southwestern Chippewa Counties; however, Merrill

till has been recently remapped in this same area by Syverson (2007).

3.2.1.2.2 Bakerville Drift Topography
Bakerville till is usually less than 3 m thick in Marathon and Wood Counties, but

has been reported to be up to 25 m thick in Marathon County and 15 m thick in Wood

 

3 . . . . .
All age estimates are reported in calendar years. Ages that were originally reported in RC years BP have

been converted to calendar years using the following web site:
http://radiocarbon.ldco.columbia.edu/rescarch/radcarbcal.htm (Fairbanks ct al. 2005).

53

County (Attig and Muldoon; Clayton 1991). Due to the thin nature of the Bakerville till
sheet, the dendritic pattern of the underlying bedrock is apparent on the drifi surface
(Figure 3.12). Clayton (1991) attributed the absence of glacial geomorphology of the
Bakerville landscape, like that of the Edgar Member, to the great age of this till sheet, as

inferred by the considerable amount of erosion which occurred post deposition (Figure

3.14; Clayton 1991).

 

Figure 3.14: Photograph of the low-relief Bakerville till plain. Photograph by K. Stanley

3.2.1.2.3 Bakerville Till Characteristics

Bakerville drift of the Lincoln Formation is associated with the Nasonville Phase,
during which ice is assumed to have advanced out of the Superior basin, based on the

composition and orientation of clasts within the till (Attig and Muldoon 1989).

54

Bakerville till is typically reddish brown (SYR 4/4) (Table 3.4), non calcareous, and
gravelly (Attig and Muldoon 1989; Clayton 1991). The weighted mean of sandzsiltzclay
ratios reported for Bakerville till is 56:28:16, making it typically sandy loam in texture
(Table 3.5). The average clay content of Bakerville till samples reported by Sutherland
(1989) is slightly higher than average clay content of that found by Attig and Muldoon
(1989), which Sutherland attributed to pedogenic illuvial clay accumulation, or to the
incorporation of Edgar till into Bakerville sediments during glaciation. Bakerville till is
usually redder and contains an average of 20% more sand (at the expense of silt and clay)

than the Edgar till. In addition, Edgar till is calcareous; Bakerville till is not.

3.2.1.2.4 Merrill Plain Extent

Merrill till is at or near the surface from southeastern Chippewa and northeastern
Eau Claire Counties, through northern Clark, southeastern Taylor, northern Marathon,
southern Lincoln, and western Langlade Counties (Figure 3.12). It thins and becomes
patchy as it approaches its southern extent, where it blends with the subdued topography
of the underlying Edgar Member in Marathon County (Attig and Muldoon 1989; Attig
1993). In Chippewa County, Merrill drifi overlies bedrock (Syverson 2007). The
northern and northwestern extent of the Merrill Member is unknown as it is overlain by

drift of the Late Wisconsin Glaciation (Attig 1993; Syverson 2007).

3.2.1.2.5 Merrill Plain Topography

The Merrill drift plain exhibits less glacial topography than drift plains of the Late
Wisconsin Glaciation, which include prominent, but discontinuous moraines, ice-walled

lake plains, and lineated features (Figure 3.12; Attig ct al. 1985; Attig 1993; Attig and

55

Clayton 1993; Syverson 2007). Merrill till still exhibits small scale glacial topography,
such as hummocky areas, less-well integrated drainage, and more wetlands and undrained
depressions than that of Bakerville or Edgar drifi (Attig and Muldoon 1989). The
Bakerville Member does not exhibit this glacial topography, which suggests that
Bakerville drift has experienced more erosion, and thus, is older (Attig and Muldoon
1989; Syverson 2007). A stratigraphic section has not been found that includes both
members, however, which would conclusively provide relative ages of these two

members.

3.2.1.2.6 Merrill Till Characteristics

Till of the Merrill Member shares many of the same characteristics as that of the
Bakerville Member, which also allows it to be readily distinguished from the Edgar
Member. Merrill till is non calcareous, gravelly, and was probably derived from the
Superior Basin (Attig and Muldoon 1989; Attig 1993; Syverson 2007). Till of the Merrill
member is typically 5YR 4/3 (reddish brown), but can range between 7.5YR 4/6 (brown)
and 5YR 4/4 (reddish brown) (Table 3.4). Merrill till is also sandy loam in texture with a
weighted mean sand:silt:clay ratio of 55:31 :14 (Table 3.5). The mean silt content of the
Merrill Member (31 %) is slightly higher than that of the Bakerville Member (28 %) at
the expense of both sand and clay content. Considering the thin nature of the Merrill till
in northern Marathon County, it is possible that the Hamberg glaciation incorporated
some of the underlying Edgar Member, which could account for the siltier contents

reported (Table 3.5; Attig and Muldoon 1989).

56

3.2.2 Late Wisconsin Terminal Moraine

In Taylor and Lincoln Counties, the ice marginal ridge of the Late Wisconsin
terminal moraine is discontinuous, being absent in some areas. This characteristic makes
it difficult to map the exact extent of the LIS at the time of the last glacial maximum
(LGM) (Attig 1993). Where the ridge is present, numerous outwash deposits are found
along and beyond its distal side (Attig et al. 1985; Attig 1993). The Late Wisconsin
terminal moraine is comprised of hummocky collapse topography, including ice-walled
lake plains, kettles, disintegration ridges, and associated outwash fans (Attig 1993; Ham
and Attig 1997). During the Late Wisconsin glaciation, the LIS was likely at its
maximum extent from approximately 21.3 to 18.1 ka (Clayton and Moran 1982; Attig et
al. 1985). However, these ages are poorly constrained due to the ubiquitous lack of
radiocarbon-datable material in northern Wisconsin between 31 and 15 ka (Clayton and

Moran 1982; Attig et al. 1998; Clayton et al. 2001; Syverson and Colgan 2004).

3.2.2.1 Ice-walled Lake Plains

Ice—walled lake plains, which are stagnant-ice features closely associated with
permafrost, are some of the most distinctive features of the Late Wisconsin terminal
moraine in Chippewa and Taylor Counties (Cahow 1976; Attig 1993; Syverson 2007).
Lakes form on the stagnant glacial ice-drifi landscape, and collect supraglacial sands and
silts. Well preserved ice-walled lake plains are only found where the bottom of the lake
contacted solid ground (Attig 1993). Well-preserved ice-walled lake plains are typically
flat or low relief plains that are elevated above the surrounding hummocky terrain

(Figures 3.15; 3.16). The centers of these plains are comprised of laminated offshore silts

57

 

Figure 3.15: Photograph of an ice—walled lake plain in Taylor County. Wisconsin. Photograph by K.
Stanley.

 

Figure 3.16: Photograph illustrating hummocky topography ofthe Late Wisconsin moraine in Taylor
County. Wisconsin. Photograph by K. Stanley.

58

and clays which grades to coarser beach sediment along the rim (Attig 1993; Syverson
2007; Clayton et al. 2008). The laminated offshore sediment is often at least as thick as
the difference in elevation between the plain and the surrounding inter-hummock
depressions (Clayton et al. 2008).

Ice-walled lake plains are not only an indicator of stagnant ice, they are also an
indicator of permafrost. Attig (1993) outlined a number of arguments as to the
connection between ice-walled lake plains and permafrost. First, permafrost would have
restricted melting of the stagnant ice, forming a stable environment for the lake to form.
Second, permafrost would have limited drainage of supraglacial lakes, causing the lakes
to melt down through the ice (like modern thaw lakes in permafrost). Once the
permafrost degraded, the ice surrounding the lakes melted, creating a high plain above
surrounding hummocky, collapse topography (Attig 1993; Attig and Clayton 1993;
Clayton et a1. 2008). Although permafrost existed beneath the stagnant ice of the Late
Wisconsin glacier, the extent to which these conditions affected north central Wisconsin

landscapes beyond the ice margin is not well understood.

59

3.3 Paleoenvironments

Permafrost is defined as ground that has been continuously frozen (remaining
below 0°C) for a minimum of 2 years (W ashbum 1980, 21); its presence can inflict
changes to the landscape both while frozen and during degradation. Periglacial
conditions are currently limited to arctic and high alpine locations, but during the
Quaternary, extended into the midlatitudes during glacial events (W ashbum 1980, 2;
Péwe’ 1983; French 1996, 5). Areas of continuous permafrost are closely correlated to
specific ecosystems (tundra and boreal forests), as well as the presence of active
periglacial processes, i.e. actively forming ice-wedge polygons. The following section

provides background and evidence of permafrost in north central Wisconsin.

3.3.1 Periglacial Vegetation

Tundra and boreal forest communities are closely associated with periglacial
environments. The zone between these two ccotones is referred to as the treeline, which
is typically 50-100 km wide if not in an alpine environment (Price 1972; French 1996,
20). The border between discontinuous and continuous permafrost can be delineated at
the treeline (French 1996, 60). On the colder side of the treeline, tundra vegetation
usually overlies continuous permafrost and consists of sedges, grasses, mosses, herbs and
lichens (Price 1972; Carter ct al. 1987). Boreal forest currently exists south of the
treeline in North America. Western boreal tree species include Black spruce (Picea
mariana), Lodgepole pine (Pinus contorta), and Tamarack (Larix laricina), whereas
eastern boreal forests are comprised of White spruce (Picea glauca), Balsam fir (A bies

balsamea) and Jack pine (Pinus banksiana) (Price 1972). Where eastern and western tree

60

species are found together, Black spruce is found in more poorly drained sites and White

spruce on better drained sites (Price 1972).

3.3.2 Periglacial Geomorphology

Ice-wedge polygons are particularly indicative of permafrost, as actively forming
wedges spatially coincide with the area of continuous permafrost (Péwe’ 1966, 1983,
1984; Black 1976; French 1996, 90; Shur and J orgenson 2007). In profile, large wedges
of ice roughly approximate a V-shape; when viewed from above, they express a
polygonal pattern on the landscape. In general, ice-wedges form when thermal
contraction cracks (no more than a few mm per winter) open in permafrost when mean
annual temperatures are below freezing; however, rapid and sufficient drops in
temperature are more important to the formation of thermal cracks (Lachenbruch 1962;
Péwé 1966; Washbum 1980, 279; Hamilton et a1. 1983). After the thermal cracks open,
water freezes in the cracks, not only adding a layer of ice but also widening the crack
(Péwé 1966; Washbum 1980, 111). Old ice-wedges grow at a rate of roughly one mm
per year (Black 1976; Washbum 1980, 111); however, Mackay and Burn (2002)
determined that young ice-wedges grow at much faster rates, averaging 10-30 mm per
year. Regardless, it would take several hundred to approximately 1000 years for an ice-
wedge one m wide to form by the repeated process of cracking and freezing of moisture
within the crack.

As the mean annual temperature warms above 0°C, permafrost begins to degrade
from the top down, causing any ice-wedges to melt (Péwe’ 1966). The resulting void left
by the ice-wedge is commonly filled with slumped host material or eolian sediment,

forming an ice-wedge cast (Péwé 1966; Black 1976; Washbum 1980, 1 14; Clayton et al.

61

2001). The infllled material of these ice-wedge casts are often crudely stratified (Black
1976)

In addition to frost processes associated with periglacial conditions, the effects of
permafrost thawing can also significantly impact landscapes. Permafrost is usually
overlain by a seasonally thawed (active) layer, which is usually <1 m thick (Price 1972).
Where this occurs, the zone of permafrost inhibits water percolation, which destabilizes
slopes leading to accelerated runoff, solifluction, and mass wasting (Carter et al. 1987;

Mason 1995; French 1996, 139; Mason and Knox 1997).

3.3.3 Evidence of Past Permafrost in Wisconsin

Paleovegetation studies in Wisconsin and adjacent areas support the chronology
discussed below. The northern boreal treeline was likely in Illinois and Iowa at the time
of the LGM, because pollen accumulation rates during this time are comparable to the
modern arctic treeline (Baker et a1. 1986). In east-central Minnesota, taxa consistent with
tundra vegetation date to 24.4-17.5 ka (Birks 1976). Boreal forest, as inferred from
spruce (Picea) pollen influx diagrams, dominated much of the area from approximately
16.3-15.1 ka until 12.9-11.5 ka, when pines (Pinus) began to outcompete spruce (Picea)
(Wright et al. 1963; Wright 1968; Birks 1976; Webb 1981; Maher 1982). Heide (1984)
similarly found that a spruce parkland (Picea-herb) surrounded Wood Lake
(approximately 30 km north of the study area, on the Late Wisconsin moraine) in Taylor
County, Wisconsin from 15.1 until 1 1.5 ka. Afier Pinus entered north central Wisconsin
approximately 12.9 ka, it quickly replaced Picea, forcing the southern limit of the boreal

forest into northern Wisconsin and Minnesota (Wright 1968).

62

Relict permafrost features associated with the climate of the Late Wisconsin

glaciation have been recognized across the state of Wisconsin (Figure 3.17; Black 1965;

 

 

.S/‘\‘ or ' ‘
«w/f’ '< 3 o 20 4o 80 Miles
1 ‘ Z; \m. N l I I I l I I I J
\ | l l l l I l l |
.' I I ‘ l O 25 50 100 Kilometers

 

 

  
      

, . /

Ice-wedge polygons J """""" -71- .. é—l

1 Black 1965

v Clayton 1986 " . ,. .____i__fi ,__i_ \
7 Johnson 1986 a

v Clayton and Attig 1989 \

Polygonal ground Minimum extent of continuous permafrost
- Mickelson and Syverson 1997 [:1 20 km (”129 ka)
.r... Late WI moralnelLGM [:1 70 km (18.1-15.1 ka)
D Study Am - 90 km (21.3-18.1 ka)
Figure 3.17: Distribution of selected ice-wedge polygons and polygonal ground in Wisconsin (after Black
1965, Johnson 1986, Clayton 1986, Clayton and Attig 1989, Mickelson and Syverson 1997, and Clayton et

al. 2001). The minimum extent of continuous permafrost is illustrated in shades of gray emanating from
the glacial margin during the periods shown (Black 1965; Clayton and Attig 1989, Clayton et al. 2001).

63

Johnson 1986; Clayton and Attig 1987; Clayton et a1. 2001) and surrounding areas

(Johnson 1990; Lusch et al. in press). Relict features associated with a permafrost

environment include ice-wedge casts, patterned ground, and talus streams, as well as

erosional features such as gullies following the draining of Glacial Lake Wisconsin

(Clayton and Attig 1987; Clayton et al. 2001). As previously discussed, in north central

Wisconsin, the strongly hummocky terrain and ice-walled lake plains in the Late

 

Figure 3.18: Image of an ice-wedge cast in
friable Cambrian sandstone (Johnson 1986,
Figure 21).

Wisconsin moraine have also been attributed
to the presence of permafrost (Attig 1993;
Attig et al. 1998; Clayton et al. 2001;
Syverson 2007). Most of the ice-wedge casts
in Wisconsin are set within fluvial or
lacustrine sands; the remainders are found in
sandy till or in Cambrian sand and sandstone
(Johnson 1986; Clayton et al. 2001). Eolian
sands and some sand-blasted pebbles with
random orientations can be found within well
documented ice-wedge cavities in Wisconsin
(Figures 3.17 3.18); thus, Clayton et al.
(2001) argued that intense eolian activity was

synchronous with ice-wedge degradation.

64

3.3.4 Extent and Chronology of Late Wisconsin Permafrost

Previous studies have speculated as to the extent of continuous permafrost in the
region (Pe'wé 1962, 1983; Wayne and Guthrie 1993; Clayton et al. 2001); however, the
extent of permafrost beyond the ice margin is supported in very few cases, because the
relict features often cannot be accurately dated. Based on the distribution of relict ice-
wedge features, French (1996) suggested that the zone of continuous permafrost extended
beyond the ice margin by 80-250 km. Similarly, Clayton et a1. (2001) concluded that
permafrost was present at least 90 km from the LIS margin during the LGM in Wisconsin
(21 .3-15 .1 ka) based on the distance between the Late Wisconsin terminal moraine and
ice-wedges identified by Black (1965) in the Driftless Area. At approximately 18.1 ka,
permafrost probably existed at least 70 km from the LIS margin, based on ice-wedge
polygons formed in the proglacial Lake Wisconsin basin after it drained (Clayton and
Attig 1989; Clayton et al. 2001); however, around 12.9 ka, the permafrost likely extended
less than a few tens of kilometers past the ice margin (which would have been near the
border between the upper peninsula of Michigan and Wisconsin) (Figure 3.17; Clayton et
al. 2001).

Relative ages of relict permafrost features are based on glacial events (Mickelson
et al. 1983; Attig et al. 1985; Clayton et al. 2001; Clayton et al. 2006); thus, these relict
features are assumed to have begun forming at approximately 30 ka, and ceased between
approximately 16.3 and 15.1 ka in southern Wisconsin and 11.5 ka in northern Wisconsin
(Clayton et al. 2001). This chronology is consistent with that of active solifluction in the
Driftless Area of Wisconsin and Minnesota, which occurred from approximately 22.3 ka

until approximately 15.1-13.8 ka, as determined by radiocarbon dates found within

65

colluvial sediments of the Driftless Area (Leigh and Knox 1994; Mason 1995; Mason and

Knox 1997).

3.3.5 Paleoenvironmental Summary of North Central Wisconsin

Unfortunately, due to the low-relief terrain and lack of bogs and deep lakes within
north central Wisconsin, few pollen sinks exist within the study area which could provide
specific vegetation records of deglaciation following the LGM. The study area lies
within 75 km of the Late Wisconsin terminal moraine (Figure 3.13); thus, it would have
been underlain by continuous permafrost at the time of the LGM. Both pollen data and
relict permafrost features from surrounding regions suggest that permafrost existed in
north central Wisconsin from approximately 30 to 15.1 ka. When permafrost began to
degrade, some landscapes, especially in the region of southwestern Clark County
underlain by friable sandstones interbedded with shale, probably experienced high rates
of instability and mass wasting, similar to that found in the Driftless Area of Wisconsin
and Minnesota (Leigh and Knox 1994; Mason 1995; Mason and Knox 1997). The slope
instability would have likely exposed large quantities of sediment, which would. have

been subject to further erosion, including deflation by prevailing paleowinds.

66

3.4 Contemporary Vegetation and Climate

Pre-settlement vegetation in the study area consisted of mixed hardwood forest,

which covered much of the area underlain by various glacial drifi members. Areas

underlain by Cambrian sandstones interbedded with shale supported pine and oak forest

(Figure 3.19). Currently, uplands and sloping surfaces covered by loess, glacial drift, or a

Taylor Co.

Marathon Co.

 

D Study Area [3 County line 0 5 10 Kilometers
Pn-oottlamont Vegetation I I ,

D Oak

:3 Pine

[:1 Maple-Oak

- Northern hardwoods including Hemlock
- Northern hardwoods minus Hemlock
E] Other (Aspen-White Birch. swamp. prairie. and brush)

 

Figure 3.19: Pre-settlement vegetation of north central Wisconsin (alter Finley 1976).

67

combination of the two are usually in crops or pasture. across most of the study area

(Figure 3.20). Very steeply sloping surfaces and low (wet) areas are typically forested.

 

Figure 3.20: Low-relief. agricultural landscape typical ofthe study area. Photograph by R. Schaetzl.

In western Clark County, where sandy soils derived from the Cambrian sandstones are
present, most of the land is
unsuitable for farming and thus
supports a mixed hardwood forest

(Figure 3.21).

 

Figure 3.21: Contemporary forest cover in
Clark County Wisconsin. Forest cover (dark
green) is most abundant in the bedrock
uplands and valleys of western and
southwestern Clark County.

 

l_'_g‘__l
0 5 10 Kilometers

North central Wisconsin has a cool, humid continental climate. The mean annual
temperature (1971-2000) at Neillsville, the county seat of Clark County is 42.2°F, with a
mean daily maximum of 8 1 °F in July and a mean daily minimum of 01°F in January
(Figure 3.19; National Climatic Data Center 2002). The mean annual precipitation is 823
mm (National Climatic Data Center 2002); 80% of which occurs in between April and
October (Figure 3.19). In a typical year, 25.4 mm of snow covers the ground for 66 days,

and the average seasonal snowfall is 1031.2 mm (Simonson and Lorenz 2002).

 

 

 

 

 

 

90 E140
80 5120
k. 70
°.. .5100
r. 60 *5
= u
E 50 3'.- 80
a 40 g 60
5 30 a 40
r- 20 “5 20
10 g
0 g 0
.9 "" P" >\ C'- '_' DD 9- “ > O <
~£§§§33§$525
OMean Temperature — Mean daily max/min lMonthly mean precipitation (mm)

Figure 3.22: Mean monthly temperatures and precipitation (1971-2000) for Neillsville Wisconsin (NCDC
2002)

69

3.5 Soils

Jenny’s (1941) state factor model provides a conceptual framework to define a
soil system in terms of controls on pedogenesis and soil distribution. According to this
model, soil formation is the product of five environmental factors: climate, organisms,
relief, parent material, and time. By looking at how these factors change across a
landscape, the state factor model can be used to explain the morphological differences
among the soils on that landscape. Within the study area, the most important of these
factors (in terms of soil morphology with respect to this research) are parent materials,
both in the surface and subsurface, and relief. Soils, and the sediments they formed in,
are most likely to be protected from erosion on relatively flat upland positions on the
landscape. For this reason, upland soils will be the focus of this section. For the sake of
discussion, the upland soils found within the study area will be discussed in this section
based on three categories: those with silt loam (loess) mantles, those with sand mantles,

and other soils (Tables 3.6, 3.7, 3.8).

70

Aflmmv @0596 .3509 323088 AER/SC BEBE :03 328358 ADZC wofific :03 ”$20 omaEEQ...

 

woo? wan
Eu - 60 Geo _ :5 >58. wEbLoncs 85808 :38 memo—moo“ £35320 .5 8m 8 _
mo. 2 G _ Em ace SSS—Em 33 .0 $004 . c 0 use mono—Soot.“ £ng Dam oS¢< Jake—DE 5.3
- :3 .380. 28 87.8% 8:658 958w mono—303 can $393320
Eu mm o :8 2: :82 Em 53 850 Lo 8004 v.3 2:an £52.85 $288.6 Q32 E6905 EEO Loocoam
:5 >88. 8582: 98 392333 can £18385
E0 05.3 :6 mm :52 Em >28 85.6 math—Lows: can mocfionmE £52.85 .EEEsm 03 Emmi coo? owgcmm
nan EE>==~ 3.6 .8 $84 9520 . .
- :3 .982 mach—$95 mono—8.23 can £35320 :oEEmZ
Eu m2 2 :5 E Emo— Em can 8233.5 .03 8 3004 8:229: 2580 £82305 .EEfism 02:2 05355 25 E30 3.34
- 53668 .350— ES 8:2:qu mono—$33 can £25320
:5 E m Eu cm Ego; Em 822:? .93 5 $004 ES 2:: 304 £52.55 .EEEzm 0mm 2:3. woo? tav—
- c.5382 >582 can 3:2:qu mecca—QB van £83320 .
Eu _m m Eu we Eco. Em 823:5 E6 .5 $004 was 2:: 264 502:9? .mzcESm 032 2:9“sz EEO YEI
223—3.; A: 3 9.5.8... man—U
”yawn—<1 .3qu 8a....sm 2.3.3.2: 2.9::— Eceueaa Sign 2.88:3 .09 252.835. 8:550 85m

 

.Amoom ~551— wcm :omcoszm ”amo— EQE KR: arc—ohm can oEEtwm $th 88 N95. 2: E E8 can—Q: coca—«E £02 mo mocmcfioego 66 95:.

71

5mm. 288:: >508: 823885 .53.): .888: :83 3888.88 5520 89.289,

 

 

 

 

 

8.85225 :8. £82585
Eu 52.3 8.81. E. .283 >884 8882). 88:0 £82.85 £288..m 932 22.885 28:82...
8.8. 8828:. 2.2.5520
Eu mo 72 >::am E. 882w >884 8:888 .8880 :8. 8.8580... 0mm 2.8< 288E...
885—2.;
.888: 238,—. 28.88:. 28.8.— 8835?— :e...8._ 2.32.55 DD mun—U 8.8m
. coatsm . . . .1 28.833... .
2...»...
.958 888‘. :8. :8:08_m 8:3 28.8 82:38. 0: :85 E5 .82.... .8 85.88820 in 2...;
5an .888: >500: 8.3888 .533: .888: :83 338288 5520 03880..
8.8— 885225 :5.
>85 8.828.. >85 £82.85 £2885 28.82%—
:8 Ed. 80 mm 88“. :8. 82228 >884 888:8: £285 802 Dam 2.5 82:82).
8:23.25 :8. 28::23:
88.28. >88. £8285 £885 0 2.5
80 SA 80 Vm gum :5 8.8.2:. >::wm 3.88:8; £825 882 >92 2.8.5.0 8.8.2.4
255 .23 828988.
8.8. 8:85.85 w:_>_8::.. 8:222... :8. 2858.8:
>83. .8 8.828: >o>2o £8285 £2885 D 2.5
88 .2: Eu mm 82“. :8 882:... >884 388.8: .885 802 .52 2.8.8.5 858:3
88238.5 :8.
8.828: >88. £82.85 £2885 28.82am
Eu mm-m 80 ac :cnm :8 8:32... >::mm 3.88:8: £285 882 Cam 2:: 2288;
£82....
8:05.85 8.88.... .8 285.8822
8:5 E. w:.>_8::.. m82.85 :8. 2.885 882.85 D 2335.0
80 om-_m :8 2. >882 :8 8.8.2... >::..m :o 8:888 8.883. :8. £._88=m 32 oE< 8.22.3
wwwfluuohfi‘ GA 8 33.8% m «.8 a8 :9... 8.8 :a :25...— onaomuga 520 8.8
u==.w..< .395 3......m _ . . . m .=. A . . .09 28.888... . m

 

.Amoom 8882 :8. :o£:o8_m 8:... 28m .82.... 828.8 .85 .8 85.88895 "Em 25.?

72

3.5.1 Soils with loess mantles

Within the study area, most of the soils with silt loam-textured loess mantles
overlying loamy glacial till are mapped in central Clark County, extending eastward into
Marathon and Wood Counties (Figure 3.23). Loess mantles here generally range
between 50 and 110 cm thick (Table 3.6; Hole et al. 1976; Bartelme and Strelow 1977;
Fiala 1989; Simonson and Lorenz 2002). The combination of the low-relief terrain and
the presence of glossic and argillic horizons, upon which water may seasonally perch,
combine to make most of the upland soils in the study area moderately well drained or
somewhat poorly drained.

Specific soil series are delineated based on a number of factors. In the study area,
the Hiles silt loam series (Oxyaquic Glossudalfs) is mapped where loess (typically 48 cm
thick) overlies loamy sandstone interbedded with shale residuum on low hills and
pediments (Figure 3.23; Table 3.6). The Kert series (Aquic Glossudalfs), which is
mapped in Wood County, is similar with respect to landform, landscape position, and
parent materials to the Hiles series; however, the Kert series is somewhat poorly drained
and typically has a slightly thicker loess mantle (Table 3.6). Santiago soils are also
mapped in Wood County and are found primarily on ridgetops on and near the
Marshfield Moraine. Santiago soils have a loess mantle that is approximately 58 cm
thick overlying dense sandy loam till, and are well drained. Where loess mantles loamy
glacial till on ground moraines, the Loyal silt loam (Oxyaquic Glossudalfs), Spencer silt
loam (Oxyaquic Glossudalfs), and Withee silt loam (Aquic Glossudalfs) series are
mapped (Figure 3.23; Table 3.6). Of these soils, both Loyal and Spencer soils are

moderately well drained and located on summit and shoulder slope positions; however,

73

Lat/’09

.la.‘
.,

.me 3!...

If.
N
our
.6_
d
, w
W.

x

t,-

. for Clark Co. ‘.'

. 3'

0

Clark County

 

- Spencer silt loam (> 100 cm)

i: Withee silt loam (~60 cm)

[:3 Loyal silt loam (~50 cm)

[:] Hiles silt loam (~50cm)

Marathon County

[:Z] Withee silt loam (~60 cm)

[3 Loyal silt loam (~50 cm)

Wood County

[3 Withee silt loam (~60 cm)

 

[:1 Santiago silt loam (~60 cm)

0 1
Ll_i

[:1 Kert silt loam (~55cm)

Figure 3.23: Upland soils of the study area with a silt loam (loess) mantle (after Bartelme and Strelow

1977: Fiala 1989; Simonson and Lorenz 2002).

74

they are separated taxonomically. based on loess mantle thickness (approximately 50 and
100 cm respectively). In the study area. the Withee silt loam series is by far the most
abundantly mapped series with a loess mantle (Figure 3.23). Withee soils have loess
mantles similar in thickness to Loyal soils (approx 60 em). but are somewhat poorly
drained, either due to their location on lower (foot and toeslope) or flat landscape
positions. Spencer silt loams, which have loess thicknesses approximately one meter
thick, are located in the northwestern part of the study area (southwest of the Late
Wisconsin terminal moraine) and in the south central part of the study area (Figure 3.23:
table 3.6). East of where the Spencer series is mapped, loess thicknesses are generally

between 50 and 60 cm, depending on landscape position.

3.5.2 Soils with sandy mantles

Upland soils with sandy mantles are mapped in western and southwestern Clark
County, where the landscape is dominated by dissected bedrock uplands (Figure 3.24).
Sand mantles typically range in thickness between 50 and 90 em and have loam, fine
sandy loam, or sand textures (Table 3.7). In addition to having a sand mantle, these soils
are bisequal; they exhibit spodic development in the sand mantle and argillic horizons
that usually coincide with the lithologic discontinuity between the sand mantle and the
underlying till or bedrock residuum (Appendix 1).

Eauclaire soils have sand mantles that are loamy sand in texture and that overly
sandy loam glacial till on dissected bedrock uplands (Figure 3.24; Table 3.7). Also,
Eauclaire loamy sands (Alfie Oxyaquic Haplorthods) have the thickest (approximately 86
cm) sand mantles. Other sand-mantled upland soils in the study area are found on

pediments overlying sandstone and interbedded shale residuum, and can be grouped into

75

Marathon Co.

I!

 

J—l- lGM I: Study Area 0

  

Soils with sand mantle

[:] Eauclaire loamy sand
Ludington sand

[:] Fairchild sand

[:] Merrillan sand

E] Humbird fine sandy loam
Soils with no mantle

- Flambeau loam

- Fallcreek loam

 

Figure 3.24: Other upland soils ofthe study area (after Simonson and Lorenz 2002). Soils formed in
glacial till. with no recognized mantle are represented by blue and violet tones. Soils with a sand mantle
are shown in warm tones.

76

two drainage sequences based on the texture of the sand mantle (Figure 3.24; Table 3.7).
F airehild and Ludington soils have sand-textured mantles and are found dominantly
mapped in the far southwestern comer of the study area (Figure 3.24). Fairehild sands
(Ultic Epiaquods) are somewhat poorly drained; thus, are usually mapped on lower
landscape positions than the moderately well drained Ludington sands (Oxyaquic Ultic
Haplorthods) (Figure 3.24). The other drainage sequence includes the Humbird
(Oxyaquic Ultic Haplorthods) and Merrillan (U ltic Epiaqods) series, which have fine
sandy loam-textured mantles. The moderately well drained Humbird series mapped
higher on the landscape than is the somewhat poorly drained Merrillan series (Figure
3.24; Table 3.7). Humbird and Merrillan fine sandy loams are mapped in the
southwestern corner of the study area, but are mostly found north and east of where
Ludington and Fairehild sands dominate. Additionally, Fairehild and Ludington sands
have thick mantles (are approximately 70-80 cm), whereas the Humbird and Merrillan

fine sandy loams have mantles that are approximately 50 cm thick (Table 3.7).

77

3.5.3 Other Soils

In central and west central Clark County, soils that lack a recognized mantle are
mapped on ground moraines in the area generally between those with sand mantles in the
southwest, and those with loess mantles to the north and east (Figure 3.24; Table 3.8;
Simonson and Lorenz 2002). Examples of such soils include the Fallereek (Aquic
Glossudalfs) and Flambeau (Oxyaquic Glossudalfs) series, which are formed entirely in
loam glacial till. However, it is interesting that soils having a fine sandy loam mantle
(i.e. the Humbird series) are mapped immediately east and west of locations where
Flambeau and Fallereek soils are mapped (Figure 3.24).

Fallereek and Flambeau soils are delineated based on the landscape position,
drainage class, and surface texture. F lambeau loams are mapped on summit and shoulder
positions, and as a result, are better drained than Fallereek sandy loam, which are mapped
on footslopes and toeslopes (Figure 3.24; Table 3.8). Both probably have some seasonal
perching of water in the lower glossic and upper argillic horizons, which can exceed one

meter in thickness (Table 3.8; Appendix A).

3.6 Summary of Study Area

The study area, set within north central Wisconsin, is a complex and diverse
landscape. In western and southwestern Clark County, the terrain is dominated by
dissected bedrock uplands. Here, soils with sandy mantles overly fine grained Cambrian
sandstones interbedded with shale. Only in well protected landscape positions within
these bedrock controlled uplands can eroded remnants of glacial deposits be found

(Simonson and Lorenz 2002). East of this region, the terrain becomes more subdued, as

78

the thicknesses of glacial deposits increase. Multiple glacial episodes are evident for this
area, based on characteristics such as color, texture, lithology, and terrain of the till sheets
left by these glaciers. The most recent glacial advance not only influenced the
topography in the far northwestern portion of Clark County with its terminal moraine, but
it also influenced the rest of the study site through periglacial conditions. Continuous
permafrost likely extended beyond the study area by at least 20 km, and probably further
than 90 km. When the glaciers receded from the LGM and the climate ameliorated, the
landscape underlain by degrading permafrost probably experienced various degrees of
instability and mass wasting. Under these conditions, it is likely that large amounts of
newly exposed sediment became available for deflation. At the same time as the
permafrost began to degrade, a period of intense eolian activity occurred, as evident from

ice-wedge casts infilled with wind-blown sediment.

79

4. Methods
4.1 Field Methods

Prior to going in the field, target locations for soil sampling were identified and
recorded in a geographic information system (GIS). Specifically, using a DEM,
orthoimagery (1992-2000), and SSURGO (digital soils) data in a GIS, target locations
were selected (1) where loess is mapped, (2) on broad uplands, where loess is thickest
and best preserved (3) within woodlots, where anthropogenic disturbances (plowing) are
minimal, and (4) approximately five kilometers apart, so as to achieve a generally
uniform geographic distribution. In the field, the general target locations were not always
viable for sample collection because the woodlot was not on the highest and flattest
position locally on the landscape, or the site was not wooded. In these cases, attempts
were made to locate a suitable woodlot for sample collection as near to the target location
as possible. Unfortunately, suitable woodlots near target locations could not always be
found, especially in central Clark, southwestern Marathon, and northwestern Wood
Counties, where the highest and flattest positions on the landscape are predominantly
cultivated. When suitable woodlots were found, additional criteria were used to narrow
locations for soil samples based on quality and accessibility. Poorly drained upland sites
were avoided, as chemical alterations often complicate interpretations. Locally, obvious
signs of bioturbation, i.e. tree uprooting mounds and pits (Schaetzl et al. 1990), were
avoided as well.

The coordinates of each soil sample were accurately recorded by using a Garrnin
GPS 76 unit, running in conjunction with real-time tracking software downloaded from
the mpzflwwwdnrstatemnus site, and ArcMap 9.2, a geographic information system

80

(GIS) software (ESRI software, Redlands, CA). In the field, additional information for
each site was recorded in the GIS, including: silt mantle thickness, a subjective measure
of loess sample quality (poor, ok, good, or great), characteristics of the contact between
the silt cap and the substratum i.e. presence of a stonelayer or paleosol, depth to substrate
(a transitional zone was sometimes interpreted between the loess and substrate), depth to
water table (if encountered), vegetation present, and the soil series mapped. For the
locations at which substrate samples were collected, I also noted the type of drift
(outwash or till), if present, texture class by hand-texturing, estimated volumetric content
of coarse fiagments, general color, and an estimate of the till member (Edgar, Bakerville,
or Merrill) at the site (Figure 4.1; Clayton et al. 2006). These data were recorded in field
notes as well as within the GIS attribute table. Because Bakerville and Merrill till sheets
become thin and patchy along their margins (Attig and Muldoon 1989, Attig 1993;
Syverson 2007), color and texture of the till samples were to identify outliers used in this
area.

During June and October of 2007, and May 2008, 67 loess samples and 19
substrate samples were collected using a standard three-inch bucket auger. At each
location, a soil sample was collected as deeply as possible from the silt mantle, but at
least 20 cm above the lithologic discontinuity or transition zone. Substrate samples were
collected if relatively unaltered material was obtained, which was ofien difficult due to an
increase in cobbles and the presence of a relatively thick (approx. 20 cm) Btb horizon
formed into the top of some of the drifi. At some locations, a layer of sand was found
between the silt cap and the drift; at three of these sites, samples of the sand layer were

also collected. Soil samples with obvious redox features were avoided, as a reducing

81

environment may affect the soil chemistry and clay mineralogy of the sample. Soil
samples were sequentially labeled based on the geographic location of each sample (1-
67) and nominally labeled based on the sample type (‘cap’ for loess samples, ‘resid’ for
samples identified as bedrock residuum. and ‘sub’ for non-differentiated substrate

(mostly till samples).

 

 

J.

[:3

 
   

OSL sample

B Study Area :1 Bakerville Till‘ P—‘T—H
1: County border Edgar Till" ° 4 8 "m

4—L Late WI moraine (LGM)"
*Clayton et al. 2006

 

Figure 4.1: Pre-Late Wisconsin drifi boundaries (afler Clayton et al 2006) with a DEM in the background.
Locations of OSL samples are represented by yellow triangles.

82

Target locations for ice-walled lake plain samples were selected based on the
locations of ice-walled lake plains delineated by Attig (1993) and Syverson (2007). In
the field, sample locations were selected from the interior of the ice-walled lake plain
where possible. Since most of the sites were cultivated, it was necessary to acquire
landowner permission at these locations. In total, ten ice-walled lake plain samples were
collected in Taylor and Chippewa Counties during May 2008. I collected six soil
samples from cultivated fields from a depth greater than 50 cm. Three soil samples
collected by R. Schaetzl and K. Syverson were also used in analysis. These three
samples were collected from a road ditchs from a depth of approximately one meter.

In order to determine the age of the sediments, three soil samples were collected
from locations in Clark County where the silt mantle is > 50 cm thick for OSL analysis
(Figure 4.1). At the Clark 1 and 2 sites, soil exposures were found along a roadside and
in a quarry respectively. These exposures were cleaned to expose fresh, vertical soil
profiles. The third location for an OSL sample (Clark 3) was chosen based on the
thickness of loess and amenability of the landowner at soil sample location 1. A pit was
dug at this last location. At all three locations, soil profiles were described according to
NRCS guidelines (Staff 1996; Schoeneberger et al. 2002). OSL samples were taken
within a 20 cm-long, two-inch PVC pipe, which was coated (after sampling) with at least
two layers of duct tape in order to minimize light contamination. Samples were collected
below the Bt horizon (in BC or preferably C horizon), to minimize contamination due to
pedoturbation, which could lead to underestimated ages, yet at least 20 cm above the
lithologic discontinuity, to minimize contamination from the substrate, which would

result in overestimated ages. Approximately 500 g of loess immediately surrounding the

83

PVC pipe was carefully collected for dose rate analysis. Collecting the dose rate samples
from the material immediately surrounding the PVC pipe served two purposes: 1) the
dose rate sample is more likely to be more similar to the OSL sample and 2) removal of
sediment surrounding the pipe aided in pipe removal from the exposed face. After the
OSL and dose rate samples were collected, approximately 500g of soil was collected
from each genetic horizon described for particle size analysis (Appendix A). The
location of each sample was recorded in a GIS using a GPS unit, as was the depth of the
sample, general quality of the site, thickness of silt mantle, and quality of the loess.
Eleven samples were also collected from five different outcrops of sandstone bedrock
in southwestern Clark County; one sample was collected in Eau Claire County.
Approximately 5 cm of the weakly cemented bedrock was shaved off using a shovel or
knife in order to expose fresh material, prior to sample collection. Photographs were
taken of the fresh exposures showing the relative location and characteristics of facies
within the outcrop. Approximately 500 g of sandstone were preferentially collected from
facies containing shale layers and finer textured sandstone using a knife and soil pan as to
carefully collect within the specific facies. As with all other samples, the locations were
recorded in a GIS using a GPS unit. Final identification of sandstone samples within the
Cambrian stratigraphic column for this discussion was made based on a georectified
digital image of maps (Mudrey 1987; Brown and Patterson 1988) showing the
distribution of bedrock formations in north central Wisconsin, with sample locations

overlaid in a GIS.

84

4.2 Laboratory Methods

After oven-drying (at 30°C), the samples were ground with a mortar and wooden
pestle by myself or M. Bigsby at the Michigan State Geomorphology Laboratory.
Substrate samples were ground more gently, and coarse fragments (larger than 4mm)
were often removed by hand prior to sieving, to avoid artificially altering the <2mm
fraction due to the highly weathered nature of these samples. Alternatively, sandstone
samples were thoroughly ground until nearly single grained. Afier each sample was
ground, it was passed through a 2 mm sieve and sent through a sample splitter four times
in order to homogenize the sample.

I prepared soil samples for particle size analysis by adding approximately 1 g of
soil to a water-based solution with (NaPO3)13-Na20 as the dispersant in a 25 ml vial. The

vials were agitated for a minimum of 2 hours at 120 rpm. Particle size analysis was
completed using a Mastersizer 2000B laser particle size analyzer (Malvem Instruments
Ltd., Worcestershire, UK), which like other laser diffractometers, consistently
underestimates the clay sized (<2 um) fraction (Konert and Vandenberghe 1997;
Beuselinck et al. 1998; Mason et al. 2003; Sperazza et al. 2004; Arriaga and Lowery
2006). To correct for this underestimation, the clay/silt boundary used to interpretation
standard texture classes was set to 6 um, which most closely mimics particle size data
using the sieve-pipette method, based on prior experiments conducted in the MSU

geomorphology lab (Schaetzl 2007).

85

 

Figure 4.2: Logarithmic particle size distribution for all 79 loess samples in north central Wisconsin.

After examining particle size data for the 79 loess samples and finding that most
of the silt peaks overlap within the 20-45 pm range (Figure 4.2), this range was chosen
for silt mineralogy analysis on a subset of samples. Twenty loess samples were chosen
based on the qualitative descriptions recorded in fieldnotes and geographic location to
ensure a representative distribution across the study area. Eight sandstone and sandstone
residuum samples were chosen based on silt contents relatively high silt contents. All
nine ice-walled lake plains were analyzed for silt mineralogy.

Once the subsets were determined, the 20-45 um fraction was separated and

retained in three steps. First, clay was dispersed by agitating a mixture of approximately
10 g of soil and a water-based solution with (NaPO3)13-Na20 as the dispersant in a 120
ml vial for 2 hours at 120 rpm. From this dispersed sample, the > 45 um fraction was

86

removed by wet sieving using distilled water; the <45 um fraction was retained. Third,
the <20 um fraction was removed by redispersion, settling, and decantation no less than
three times in a 1000 ml cylinder until the supernatant was clear or light gray in color.
Depth of decantation was determined using Stokes law. The 20-45 pm sediment, which
had settled to the bottom of the cylinder, was transferred to a 100 ml beaker and oven
dried at 30°C. In order to produce the sharpest peaks for X-ray diffraction analysis
(especially for feldspar), approximately 2 g of the 20-45 pm silt particles were
micronized (dry) for 3.5 minutes using a Fritsch Analysette 3 Spartan Pulverisette 0 mini
mill (Fritch GmbH, Idar — Oberstein, Germany). The sample powders were then packed
into a sample container, carefully tapping the powder into the container using the edge of
a glass slide in order to optimize random orientation of silt particles. X-ray diffraction
(XRD) was performed using a MiniFlex+ X-ray Diffractometer (Rigaku Corporation,
The Woodlands, TX) between the angles of 25° 20 and 31 .5° 20, using a step size of .02°
20 (Grimley 1996). Mineral peaks and background values were determined using the
computer software program, JADE. The mineral peaks identified by JADE were
manually verified by comparing the results to the peaks listed in Table 4.1, identified by

Grimley (1996).

Table 4.1: X-Ray diffraction peaks, and their intensgy factors, as determined by Grimley (1996).

 

 

Mineral d-spaeing (A) 2 9 Peak intensity factor *
Quartz 3.34 26.5 1.0
K-feldspar 3.25 27.5 4.0
Plagioclase 3.20 27.8 3.5

 

*Peak intensity factor determined after 3.5 minutes of mieronization utilizing standard mixtures by Grimley
(1996).

87

Relative feldspar contents were calculated by the following conversion based on
F/Q peak heights (Grimley 1996; Equation 4.1). In this equation, Pm is the relative
percentage of the mineral (quartz, K-feldspar, or plagioclase), Hm is the peak height
(counts per second) for the mineral, I is the intensity factor for the mineral, and 2H is the

sum of all quartz, K-felspar, and plagioclase peak heights for the sample.

Hm*I
2H

 

Pm = ( ) * 100 Equation 4.1

The peak intensity factors (Table 4.1), were determined by Grimley (1996)
through analyzing the X-ray diffraction patterns of micronized standard samples. The
peak intensity factors are used in equation 4.1 to normalize peak heights of quartz,
plagioclase, and K-feldspar. This converts the P/Q and K/Q ratios into a proxy for

content.

88

5. Results and Discussion

The silty mantle present in north central Wisconsin has been widely accepted as
loess (Hole 1942, 1968; Cahow 1976; Hole et al. 1976; Bartelme and Strelow 1977; Attig
and Muldoon 1989; Fiala 1989; Sutherland 1989; Clayton 1991; Attig 1993; Simonson
and Lorenz 2002; Syverson 2007), and therefore, I will refer to it as loess in this thesis.
Although loess is mentioned in many studies situated in north central Wisconsin, it has
not yet been the focus of research. The focus of this chapter, therefore, will be to
characterize this loess deposit. I will begin by presenting and discussing particle size
data for samples of Cambrian sandstone and the pre-late Wisconsin tills that underlie the
loess in Clark, Marathon, and Wood Counties. The remaining sections of this chapter
focus on the loess itself. First, I will introduce results and discuss the soil profile and
loess sample data. Next, I will present and discuss the spatial characteristics of the north
central Wisconsin loess sheet. This section is important because spatial characteristics
not only define this type of eolian sediment, but also suggest its source area(s). Using the
identified source areas, I will present particle size and mineralogy data of potential source
sediments. Finally, by combining OSL ages of the loess sheet with the presumed regions
of provenance, the paleoenvironmental significance of the north central loess sheet will

be evaluated and discussed.

89

5.1 Particle Size Comparisons of Sandstone and Pre-Late Wisconsin
Till Samples to Literature

Sandstone and substrate samples collected in the field were characterized in order
to compare them to data reported in the literature. The following is a summary and

discussion of particle size data for all sandstone, residuum4, and till samples collected

during the summer of 2007.

5.1.1 Cambrian Sandstone Residuum and Outcrops

Cambrian sandstones interbedded with shale are located to the west and southwest
of the study area (Figure 5.1; Mudrey 1987; Brown and Patterson 1988). These
formations are generally fine-grained and friable, readily weathering to loam and sandy
loam soils, i.e. Humbird fine sandy loam (W eidman 1907; Hole et al. 1976; Simonson

and Lorenz 2002).

 

4 . .

Substrate samples (collected beneath loess samples With a hand augar) that were determined to be
sandstone residuum will be referred to as ‘residuum’ samples throughout this chapter. Sandstone samples
collected from sandstone outcrops will be referred to as ‘sandstone’ samples.

90

p Tat/or Co.

7 EC1R
‘0

l Chippewa Co.

Clark Co.

 

 

 

 

Trempealeauffi -. Jackson C o
Bedrock Formation "/FN
:1 Wonewoc 0 Sandstone Samples 6‘ l 5,11?
Eau Claire D Study Area 0 s 10 Miles jT__r:1‘L‘14,g,x/

- l—i—H—' ‘9‘ .T ‘t
- Mt. Simon [:1 County borders 0 5 1° Kimms t , g
- Precambrian t/

Figure 5.1: Location of sandstone sample sites in north central Wisconsin overlaid on a map of Cambrian
sandstone formations (alter Mudrey 1987; Brown and Patterson 1988). Most of the Cambrian sandstone
that underlies soils within the study area is of the Mt. Simon Formation. The Wonewoc and finer grained
Eau Claire Formations lie to the southwest of the study area. The following samples: SSl, SS4, SS5, were
identified as Mt. Simon Formation samples based on bedrock distribution maps of Mudrey et al. (1987) and
Brown (1988). The SS2 samples were later classified as a sample from the upper unit ofthe Mt. Simon
formation based on particle size data, color, and bedding characteristics described by (Morrison 1968) and
Asthana (1969).

91

In seven of the eight Mt. Simon sandstone samples, clay contents were 5 6%, and
silt contents generally ranged between 1.0 and 8.3% (Table 5.1). The sample (SSlC) that
did not fall within these ranges included a 0.5 cm thick shale layer (and the dark brown
layer above it), which likely accounts for most, if not all of the abnormally high clay and
silt contents (Figure 5.2; Table 5.1). In all eight Mt. Simon samples, the 125-250 um
(fine sand) contents varied between 20 and 52%, whereas the 50-125 um (very fine sand)
contents varied more widely (8-41% Table 5.1). According to NRCS texture
nomenclature, the Mt. Simon beds sampled ranged from fine sand to coarse sand; the

mean and median of which was sand.

 

Figure 5.2: Photograph of a shale layer in sandstone sample SSlC (bluish green from approximately 29.5
to 30 in). Dark brown layer also included in bulk sample particle size data. Photograph by K. Stanley.

92

dogwoo— zowo 8 x25 :5 2 new 2 52.33 203 3358 moon 22m .8358 08:33. 3 Swan “8.25 e 53» 360:8 @2958 22mm
Adm oazwr: 3958 28323 2mm 2: Set :83 .5?— oEfi Bag 30232 a Co 29:28 38 20:55:

.Om-mm H.mU www-mm um—z “mmrfi awn MNTN H_mh~> mooomrooe ”my? Mcoo—roam Hm0 Moomémm “m2 6mm-m§ “mm ”mmTOm um; .81 E 32.08% 36 Botan—

e

.mUM Z 2: E cop—cow momma—o 22x2 :8 co women 2222.252 9

A83 0823 3 ofim 0.2th 35303 :88 “mg—2 a
E: om-~ azm a

 

 

 

 

 

 

 

 

 

E S m: 4.2 S 3“ M3 8.2 Now an; :52 Sam; we: $38

8.: 9: 93 as no on 3 on on 848 e827. 2538 we. 8%: mm

am 2: gm as S I M: as 3: Sam 68:5 comma 5:. .635 u w
2%

3 S. 3. an a... a... m... 8.8 «.9. a; 2.82.... 35» 3: 5.82

a." 3 am 3 a... E E in as. mama 2.82.: 58. can; 32 :82 a

S 3 a: 8.: _.o E 3 EN 2m 88 E8. memo; moo aim m

3 2 3 2 od 3. we 92 BM 2wa anaieas View 3: 03m m

to 08 2 we is od 3 8.2 NS :8 838: be, 388 3: mmmm a

3 od 3 3 ed 3 2 2m 3m 98 838mb? 38 5: 5%

N... 2 2 3 2. 2 ”.2 a: a: a: use as: we: 8:82

S. a... an 2 2 S. «.2 n.3, an” e8 .88 at: an: :82

3 to no is 3 2: 0:. 8.8 E :8 Ba 38 2.8 <mmm

3 ed 3 3 od 3 3 9; 2e 08 238: :98, :2 9% w

ed 3 2 3 ed 2 98 was 92 a8 Esme: 38 32 5mm “.8.

no 3. ed as 3 SN man as 4.: 92: 88388 to? 22 mmmm m.

S no 3. 2 3 3 3a Nam 3a 98 Ba 925 38 £3 m

2 3. <2 as 3 ed 3. 8.? 3m 38 581288: Nina 28 2mm

2 as am 2 is 2. n: 3% 38 a8 Edge 3% E: 2%

3 S 2 no 9o 2: 3w 8.8 mg 28 8a 923 38 5%

8:0 mm: mm... mm?» 88> emu am: am... em“; 5% 0:29 2.55; 9.3 8232 92:55 a...

Emma—>430

amt—E m2:

 

.moEEmm :5:ng 2.93:3 was @8230 2552.3 :wtnEmU .8 8% 33 203.8“— u_.m 2a.;

93

Most samples from the younger Eau Claire Formation (Figure 5.1) had higher
clay and silt contents than that of Mt. Simon samples. Clay contents were generally less
than 5% and silt contents varied between 3.8 and 11.7%, with the exception of sample
EClR, which was dominantly composed of shale laminae (Table 5.1). The 50-125 um
(very fine sand) contents ranged between 30 and 63%, and the range of 125-250 um (fine
sand) contents was between 18 and 37% for Eau Claire Formation samples (Table 5.1).
NRCS texture classes for these Eau Claire samples were generally either very fine sand
or loamy fine sand.

Samples from relatively pure shale beds from the Mt. Simon Formation had silt
loam or loam NRCS textures. Clay contents for the three relatively pure shale samples
varied between 16-25%; silt contents varied between 33 and 55% (Table 5.1). Total sand
contents accounted for approximately 25% or 50% of the particle size distribution in
shale samples, of which the 50-125 um fraction, was the largest component (Table 5.1).
NRCS texture classes for the three shale samples are silt loam or loam (Table 5.1).

In summary, samples from the Mt. Simon and Eau Claire Formations suggest that
these rocks are typically fine or very fine grained sandstones. On average, the 50-250 urn
(fine and very fine sand) content accounted for approximately two-thirds of volume of
the Mt. Simon and Eau Claire samples. Silt and clay contents were slightly higher for the
Eau Claire Formation samples than the Mt. Simon Formation samples, but constitute a
small percentage of both formations overall. Silt and clay contents increased
dramatically within shale laminae/stringers/beds, which were dominantly composed of

silt-sized grains.

94

5.1.2 Pre-Late Wisconsin Tills
Much of the bedrock within the study area is overlain by one of three different
tills (Edgar till, Bakerville till, and Merrill till), which exist as a surficial deposit or are

underlain by <1 .5 m of loess (Figure 5.3; Clayton et al. 2006). The Edgar till sheet,

     
 

.4‘
L‘L

    

:1 a.,.,.:(.r11 WK’COnl-E'J‘ 1 1";""I"—l

 
  

‘ 4L

    

* pre-Late WI rm samples CI Merrill rm; 0 2 4 Miles
OSL sample [:3 Bakerville Till‘ l-——L-r—H
D Study Area _' ° 4 8 km

[:3 County border -l—-|- Late WI moraine (LGM)*
*Clayton et al. 2006

 

Figure 5.3: Locations of substrate (pre-Late Wisconsin till) samples, soil pedons, and the boundaries of
pre-Late Wisconsin till sheets and the Late Wisconsin terminal moraine (LGM) in north central Wisconsin
(after Clayton et al. 2006).

95

which is probably the oldest (pre-Illinoian) of the three, is found near the surface in the
western third of Marathon County, but also occurs in a small portion of southwest Taylor
and eastern Clark Counties (Clayton et al. 2006). The Bakerville till sheet occurs
throughout much of Clark County, extending into Wood and Marathon Counties along
the Marshfield Moraine; its till can be found overlying Edgar till in some locations.
Bakerville till is estimated to be Illinoian in age (>130 ka) (Baker et al. 1983; Clayton et

a1. 2006). In southwestern Taylor and northern Clark and Marathon Counties, Merrill till

can be found at or near the surface and is at least 45 ka5 old (Stewart and Mickelson
1976; Clayton et al. 2006). With the exception of Sutherland’s (1989) work, literature
reporting particle size data for pre-Late Wisconsin tills use a sand/silt break at 63 um,
thus, the following section will compare particle size data using the same sand/silt break.
Mean particle size data from the three pre-Late Wisconsin tills show significant
differences among the three members. Consistent with the general particle size trends
reported in the literature, Edgar till samples have the highest silt contents and Bakerville
till samples have the highest sand contents (Tables 5.2, 5.3; Stewart 1973; Mode 1976;
Attig and Muldoon 1989; Sutherland 1989; Clayton 1991; Attig 1993; Syverson 2007).
The mean sand:silt:clay ratio for Edgar till samples in this study (24:56:20) is siltier than
the weighted mean sand:silt:clay ratios (39:43:18) reported by Attig and Muldoon (1989)
and Mode (1976); however, mean silt percentages are only slightly above the reported
range of values (Tables 5.2, 5.3; Mode 1976; Attig and Muldoon 1989; Clayton 1991;

Attig 1993). Similarly, Bakerville till samples in this study had higher silt contents

 

5 All ages are reported in calendar years; the dates originally reported in radiocarbon years BP were
converted to calendar years using http://radiocarbon.ldeo.columbia.edu/research/radcarbcal.htm (Fairbanks
et al. 2005).

96

(31%) than the weighted mean of Bakerville samples analyzed and reported in the

literature (27%), yet fell within the range of reported values (Tables 5.2, 5 .3; Mode 1976;

Attig and Muldoon 1989; Clayton 1991).

Table 5.2: Particle size data of pre-Late Wisconsin till samples in north central Wisconsin.
CLAYFREE

 

 

 

Sample ID MWPSa S:Si:Cb S:Si:C Texture Class VFSc FSc MSc CSC VCS
16 Sub 53.6 19:57:24 24:52:24 silt loam 18.3 7.1 4.5 1.7 0.1
37 Sub 96.4 29:52:19 33:48:19 loam 16.4 9.9 9.2 5.5 0.4
43 Sub 81.5 23:58:19 27:54:19 siltloam 14.4 7.1 7.1 4.7 0.4
i'u' 445ub 65.9 19:59:22 24:54:22 silt loam 16.9 4.8 4.8 3.8 0.3
g 246 Drifl 84.9 28:52:20 30:50:20 silt loam 11.4 12.1 10.5 3.7 0.1
249 Drifi 59.8 21:59:20 24:56:20 silt loam 11.9 9.4 7.0 1.5 0.0
250Drifi 77.0 24:55:21 28:51:21 silt loam 13.6 9.5 8.0 3.8 0.2
Mean 74.1 24:56:20 27:52:21 silt loam 14.7 8.6 7.3 3.5 0.2
Median 77.0 23:57:20 28:52:20 silt loam 14.4 9.4 7.1 3.8 0.2
3Sub 308.1 70:20:10 72:18:10 sandy loam 10.4 14.7 27.7 25.8 1.9
12 Sub 187.0 49:36:15 52:33:15 fine sandy loam 15.2 13.6 17.5 14.3 1.1
19 Sub 181.5 59:28:13 62:25:13 fine sandy loam 20.7 22.1 16.8 10.9 0.8
24 Sub 167.3 54:32:14 58:28:14 fine sandy loam 19.1 19.4 18.6 9.5 0.6
9 33 Sub 263.6 67:22:11 69:21:10 sandy loam 6.6 17.8 33.9 17.8 0.8
f3: 45 Sub 179.7 48:39:13 52:35:13 loam 14.9 13.3 17.9 12.7 0.9
g coarse sandy
1'3 178 Drifi 351.4 7421917 76:17:7 loam 9.2 11.0 29.3 30.2 2.3
180 Drift 128.5 43:41:16 46:38:16 loam 16.4 16.3 14.9 6.6 0.3
2C-
(Clark 2) 132.0 45:42:13 48:40:12 loam 14.8 18.6 15.8 5.6 0.2
Mean 211.0 57:31:12 60:28:12 sandy loam 14.2 16.3 21.4 14.8 1.0
fine sandy
Median 181.5 54:32:13* 57:28:13 loam 14.9 16.3 17.9 12.7 0.8
1Sub 123.8 35:52:13 41:46:13 loam 18.0 9.2 11.6 7.6 0.5
39 Sub 106.3 36:45:19 40:41:19 loam 17.5 13.1 13.4 5.1 0.1
3C-
__. (Clark 3) 153.3 51:38:11 55:33:12 fine sandy loam 18.0 17.8 19.3 7.0 0.2
g 55 Sub 87.7 25:63:12 31:57:12 silt loam 16.9 6.3 7.1 4.5 0.3
E 58 Sub 128.9 36:48:16 40:44:16 loam 14.4 10.6 13.9 8.1 0.5
61 Sub 181.9 48:40:12 52:36:12 fine sandy loam 14.4 12.2 20.0 12.2 0.7
Mean 130.3 38:48:14 43:43:14 loam 16.5 11.5 14.2 7.4 0.4
Median 126.4 36:47:13* 40:43:13* loam 17.2 11.4 13.7 7.3 0.4

 

a MWPS: mean weighted particle size by volume (um)

b Sand:Silt:Clay ratio based on a 63 um silt/sand break. All other columns based on a 50 um break.
c Particle size fractions in um. VFS: 50-125; FS: 125-250: MS: 250-500; CS: 500-1000; VCS: 1000-2000

97

 

 

 

 

 

 

 

 

 

 

 

 

 

 

25.8w Sufism 32522 5...: a... 52.. 5:82
2538- 2255 38m": :5: a... :55 :82
5.8. 5:55 2258 58. 2:35 22:55 5.8: 25.5... =85 8:53,
magnum-828-8 mafia-328-3 mas—“mménmm-fi 85.:
5 u 5 2-92-8598 8m 1 5 assume-amass 52 82:6 5:
§ .1. 5 2 528 :2 585 8d
a. .1525me 5152238 E 15 Emma? 22 8o:
:8 u 5 E n _ 3m 88 Scam
3.. u 5 253m 8-9. new 82 522
2. u 5 285. a u 5 2 ”on? :2 u 5 M: ”$55.. $2 8852 8a 9.2
283:. 253m 5323 read a... :35 :58:
Ease. 2525 :28": :5: a... 52.. :82 E:
E u 5 3:328 E u 5 5.28.538 am 1 5 3253.8 :8... 8.88m 52
38.8 Q n 5 we; _ m C255 3 u 5 am” 3.5m
cast a: n 5 : Mama. 958 6m 1 5 58$ $2 82855
ado-.m—
=_._._QE o=_>.5v_am— haw—um 99:95.3”— :5

 

Seam £5 new 2382: page 5953 5288;» 35:3 58: E 2.: 52853 owe-H-850 385950 2.8 afloazmficmm om80>< "mm 2::-

98

Merrill till samples in this study contain much more silt (48%) and much less sand
(38%) than the Merrill till samples analyzed and reported in the literature (31 and 55%
respectively) (Tables 5.2, 5.3; Stewart 1973; Mode 1976; Attig and Muldoon 1989; Attig
1993; Syverson 2007). The mean sand (3 8%) and silt (48%) contents of Merrill till
samples also fell outside the reported range of values for sand (49-60%) and silt (30-
38%) contents (Tables 5.2, 5.3; Stewart 1973; Mode 1976; Attig and Muldoon 1989;

Atti g 1993; Syverson 2007). At least two factors could account for the higher silt
contents of the Merrill till samples. First, the sandier Merrill Member becomes thin and
patchy towards its southwestern extent (Attig and Muldoon 1989); thus, some/all of the
samples classified as Merrill till in this study may, in fact, have been Edgar till. Second,
substrate samples were taken as deep as possible with a bucket auger; however, they were
usually within 30 cm of the lithologic discontinuity; thus, they are likely to have
experienced some amount of post-loess depositional mixing with the above, silt-rich
sediment (Foss and Rust 1962).

In sum, the pre-Late Wisconsin till samples matched the general texture
characteristics reported in the literature; Edgar till samples contained the most silt,
whereas Bakerville samples contained the most sand. However the specific weighted
mean values reported by the literature do not match the values found in this thesis. The
discrepancy is likely due largely to sampling error — I obtained the till samples as deep as
possible using a bucket auger; however, due the length of the auger and ubiquitous
cobbles, till samples were taken from the lower Bt horizon, instead of deep in the C
horizon, as was done by researchers more interested in characterizing the tills of the

region.

99

5.2 Characteristics of Loess in North Central Wisconsin

As discussed in previous chapters, loess has widely been accepted as a surficial
deposit in Clark, Marathon, Wood, and Taylor Counties (Hole 1942, 1968; Hole et al.
1976; Bartelme and Strelow 1977; Attig and Muldoon 1989; F iala 1989; Sutherland
1989; Clayton 1991; Attig 1993; Simonson and Lorenz 2002). The purpose of this
section is not so much to confirm the presence of loess and its underlying lithologic
discontinuity, which was done by Hole (1942), but rather to expand on Hole’s research
by presenting a more focused characterization of the loess itself. In this section I will (1)
present soil profile data and discuss interpretations of this data, and (2) present and

discuss loess sample particle size data.

5.2.1 Soil Profile Data and Interpretations

Data from the three soil pedon exposures (Clark 1, 2 and 3) indicate the presence
of loess and, below, at least one lithologic discontinuity at each site (Table 5 .4).
Locations of soil pedons are shown on figure 5.2, and complete pedon descriptions can be
found in appendix A. Soil at the Clark 1 site formed in what is interpreted as 40 cm of
silt loam loess, overlying what is interpreted as 30 cm of sandier (loam) loess (Figure 5.4;
Table 5.4). At approximately 70 cm depth, a notable increase in gravel content and a
slight increase in clay content presumably signifies the top of Bakerville till. The eroded
remnant of a buried argillic horizon (3Btb) is interpreted to begin at 84 cm and extend to
145 cm. Below this, 25 cm of unaltered Bakerville till overlies sandstone residuum. In
sum, this site contains a buried, but partially eroded paleosol, formed in Bakerville till,

overlain by loess, which gets increasingly siltier nearer the surface.

100

.Om-mm H.mU ”mm-mm H5.). ”mm-m. 5m“. mNTN gm“; .ooom-ooo. H.mU> ”cog-cow ”mu "com-omm .mE 5mm-m~. um... ”mm—5m .mu—> .E: E 5:265... 0N8 2255.. u
3.3 2:23 .3 38 205.8. BEE?» =52: ”ma—32 A.

5.. 52 5.5 ..

 

 

 

 

 

55 5.5 55. m5 N5 55 m5. 5.: 55. 855 5.8. 58.55585 .353 .5282). i... on
N... 55. 5.5. 5.5. m5 3 5... N5 55. $5.. 5.2. 58. 25.... ...:.Eo.2 ....-.5 .mm
.8. 55. 3.5 55. 55 55 3 m5 2. .555 5.8 58:8 5.555. 8.8. 3-8 88. 3
3. M5. 5.8 5.... 55 m5 5.. m5 5.: 555. 5.9.. 58:8 253. 88. 8.8 .58. m
.8. S. 5.8 3. 55 55 55 55 5.... £5. .55 58:8 :52. 88. 8-55 m m
3. 55. 555 3.. 55 55 .5 .5 5.5 FR 5.2 58:8 5.55.5 88. 55-: .2 c.
..... 5.2 5.55 S. .5 5.. 5.. 55 S. Rum-o. 3... 58.8 :5555 88. 2-5 .<
5-5 .o
2.558 + 55. 5..
.8 3 5.5. 5.5. 5.. 5.55 a: 55 5... 59.5 5.5mm 58185 5.53.5 88.553.57.828 8.52 um
55 55 55. 55. 5.. m5. 55. 55. 2. 5.55 5.85 58:58.. .335 88.589.58.88 52-52 8.5
55 2 m5. 2. N5 55 5.2 55. 5.... 5.8 5.5: 58. 2555.. ....o...>.8.£ 52.: um D
55. 5... 3.5 5.... .5 3. 5... 55. 5.... .58 5.8 58....8 5.52.5 ....o...t8...m 55.55 255 w
E. .8. 5.8 .5. 55 55 5.5 5.5 55. 32 we. 58:8 2.8“... 88. 8-5.. 55 z
.5. 2. 5.8 mm. 55 55 55 ..5 5.8 52.5 a: 58:8 5.8”: 88. 5..-: :m
55. 55. .55 5.: .5 5.. 55 N5 55. .55. E. 58:8 5.555. 88. 3.5. m.
5.... 3.. 5.8 m5. .5 2. 55 .5 EN 855 n: 58:8 553m 88. 5.5 5<
555ng occumficwm + 0N.— 5U?
:5 55585 85858 5: - 8. 52.6..
3 m5 5.: 5.2 5.5 2 55 5.5 3. Sam 5.? 58:8 5.525 ...8...te5m 8.8... mom
5.5 m5 555 .85 55 55 55 S. 5.5. 535 5.5.. 58:8 55535 ....o...ze._.m 8.55. 55m:
3 S 5.... N5 .5 E. .5. SN 5.... 53.5 .5... 581.558.. 5.525 ....2..te.mm 55.-.... .555 m
5... m... 555 5.55 m... 2 55 m5 .5. 8”: at. 58:8 :53.” ....o...>.8.£ 8-55 5.8: m
55 .5 5.: 55. N5 N5 .8. 55. 55. 5.8 5.2. 58. 25.5.. 5855588558... 55-; 5me I
E. 2.. 3.5 5.2 m5 2 m5 m5 5.5. 855 .55 58:8 5.52m 5.855.588.5588 8-5.. mm:
55. .5. SN 55. 55 .5 N5. 55 2. 855 555 58:8 25555 88. 5..-... m.
5.2 55. .55 55 55 55 m5 ..... 55m 52... 5.? 58:8 559:. 88. 5.-.. 5<
0.6 Q82 08... .85 08> .50 um: 05... 02> .55 5555:. 8.0 2585 9.55 5.852.855 .55 2
85.5.0 “:58. 52.... £85 2558

 

.59 .bssoU 4.5.0 5 @2958 $5. :8 8.5 2: .8 858 um; 20.58.325.83 9... .85 Sam 035,—.

101

 

Figure 5.4: Photograph of soil profile at Clark 1, with cumulative particle size depth plot overlain at
approximately the same scale in cm. This soil presumably formed in 42 cm of loess, over very fine sand-
textured loess (300m thick), over Bakerville till, over sandstone residuum. The last two samples (38th at
120 cm and 3Cg at 145cm) were collected by auger. The lithologic discontinuity (LD) between the sandy
loess and the till is at 70 cm (3Bt/EZ). A thick, red (5YR 5/3) paleosol is interpreted to extend from 84 to
145 cm in the till. Photograph by K. Stanley.

102

At the Clark 2 site, approximately 50 cm of silt loam loess directly overlies
approximately 80 cm of Bakerville till (Figure 5.5; Table 5.4). The lithologic
discontinuity between the loess and
till is readily apparent in the field,
and is also indicated by an abrupt
increase in cumulative sand content
(Figure 5.5; Table 5.4). The lower
60 cm of the Bakerville till is
relatively unaltered and overlies
what was interpreted in the field as a
remnant of a buried paleosol formed
in an outwash—like diamicton, at a
depth of 130 cm. If a paleosol exists
here, the clay content would likely
increase with depth, corresponding
to a Btb horizon. However, the
cumulative percentages of sand, silt,
and clay do not support the

interpretation of a paleosol, as the

 

clay content decreases with depth Figure 5.5: Photograph of soil profile at Clark 2, With

cumulative particle size depth plot overlain at

approximately the same scale in cm. This soil formed in

loess, over Bakerville till, over an outwash-like

diamicton, over crystalline bedrock residuum. The

diamicton is weathered crystalline lithologic discontinuity (LD) between loess and till is at
50 cm. Photograph by K. Stanley.

Underlying the outwash-like

bedrock residuum (likely grus). In

103

summary, four different parent materials exist at the Clark 2 site. Weathered crystalline
bedrock is overlain by an outwash-like diamicton, which is in turn overlain by Bakerville

till. Overlying the till is 50 cm of loess.

 

Figure 5.6: Photograph of the soil profile at Clark 3, with cumulative particle size depth plot overlain at
approximately the same scale in cm. This soil formed in 91 cm of loess overlying Merrill till; the lithologic
discontinuity (LD) between the two is shown by the dashed line. The 2C horizon (at 114 cm) was sampled
by auger. Loess below the E horizon (29cm) steadily increased in sand (dominantly very fine sand) with
depth. Photograph by K. Stanley.

Approximately 90 cm of silt loam loess overlies Merrill till at the Clark 3 site

(Figure 5.6; Table 5.4). Starting at the Bt/El horizon (at a depth of 36cm) the sand

104

content steadily increases with depth (Figure 5.6; Table 5.4). Of the sand in the loess, the
50-125 urn (very fine sand) fraction remains dominant, with depth (Table 5.4). Coarse
and very coarse sand contents are virtually non existent in the loess, but occur in the
underlying, fine sandy loam, Merrill till, in small amounts (Table 5.4). Fine (125-250
um) and medium (250-500 um) sand contents are low in the upper portions of the loess,
and increase slightly towards the contact (at 91 em) with upper Merrill till (Table 5 .4). In
sum, at this site Merrill till is overlain by 90 cm of loess, which becomes siltier towards
the surface.

Increases in very fine, fine, and medium sand with depth occur in the loess at all
three locations, especially at Clark 3; they can be explained as follows. Loess deposition
rates may have been slower at first, allowing pedogenesis to keep up with aggradation,
leading to mixing between the substratum and overlying loess. Second, sediments
deposited above the tills may have contained more very fine sand initially, but later may
have been more silt-rich. In this second scenario, the fining of loess towards the surface
could result from wind speed variations (stronger winds initially which decrease in
intensity over time), or from a change sediment supply (initially sediment deflated from a
sandier source, but changed over time to a more silt-rich source. Regardless, the increase
in sand contents with depth suggests changing paleoenvironmental conditions during

loess deposition.

5.2.2 Loess Sample Data and Interpretations

Loess samples were collected with the use of a bucket auger across Clark,
Marathon, and Wood Counties (see C haptcr 4). I collected 65 loess samples during

2007-2008. Fourteen supplemental loess samples, collected by R. Schaetzl during 2006

105

within the confines of the study area, were also included in this study. In total, data from

79 samples (Figure 5.7) were analyzed. The m

can sand:silt:clay ratio for all mantle6

 
   

Hwy-w

i- L.r ,. notice.

. 25
l . ' Clark 2
27_ '

    

“41342 iv}- ,‘26 V

;

 

 

Loess Samples

   

2 - Y ,
o Merrill Till'

OSL sample :1 Bakerville Till“
D Study Area Edgar Till’

[:3 County border J—L Late WI moraine
*Clayton et al. 2006

 
    

o 2 4 Miles
0 4 8 km
(LGM)*

 

Figure 5.7: Locations of loess samples, soil pedons, and boundaries of pre-Latc Wisconsin till members,
and the Late Wisconsin terminal moraine (LGM) in north central Wisconsin (after Clayton et al. 2006).

 

6 . .
Some of the loess samples collected in the study area are quite sandy (> 65% sand). and may not be
considered true loess. These samples are likely eolian. and instead. may be more akin to sand sheets or

coversands (sensu Pye 1995; Crouvi et al. 2008); thus, I

will refer to these samples as mantle samples.

106

.Om-mm ..mU ”mm-mm ..mS. flmm.m. H.mn. .m.-m H.m-..> Seem-coo. HmU> "ace-cow .mU Sew-cmm H.m.>. 5mm-mm. .mn. ”WE-ow Hm..> .E: a. 80.8.... 0E. 0.0.2.... 0
$5... 08...? .3 38 22:8. BEwBB SEE @355. ..

5.. ..m-N new ..

 

 

 

 

 

 

 

 

 

 

 

..... em. N.N ..... .... N.N e...- NN em. ENN
..N. ...... EN .8. N... ...N ..m 8.. ..8. at...”
E 8.. N... o... N... .... o... 2.. E. 9....
N8 ..8 o... ..... m... Na 8.... o...N .....N 8....
N... N... eoN .N. .... e. ..N n. ..m. 8.:
8.... .8. ..N e... 2. m... ..N 8.. 2. ..eNN
x... e... N.N m... .... N.N N... ..N ..m. Ear-N
o... ..... ..N S. N... .2 a. m... m... ..uoN
3.. m... ..N n: .... ..N ..... ..N E. FMN
3.. 3. noN v: N... ..N ..N ..N e... FMN
08.. U8.). 08... 05..» 08> omu um: um. am“; .3

mm¢m><au

Eu... :8 mfimeug 32m 5:55. tea-.0
E5... 2.. mtamneN 333 535. e529
58. 2 ”on . m 2...... .2 u 5. 8.8.). .5858...
Eeo. .658 2.... o. 3.0% 3..wm. 8. H 5 :32 .26 88-.
E8. .2. £60.... .mfl. .m. n a. .5602
En.
Sec. :8 m: 69w. $6.. G. n c. :82 A... 5..
Eu... .2. m . .35..” 3.3 .mm H 5 52.55. 3.22885
:80. 2.. m. 59mm wwdn Aw». n 5 585. .95 800-.
58. .8 o. ”3.... E... e: u 5.. 8.8.2 .88..
E8. :8 m. £93 mam Aw. n c. :85. .92. m8...
wan—U 9.5.3..- Uufium awn—3.). 8.2.21.9
«Fry—«.m— mZ—m .25 .8...—

 

dmb 25.2.3 0. $6.08.. 39.9% 5.... on; 0.”..th 93.... 2.5:: ..o b.2555 "Wm «Ea-F

107

samples is 26:59:15 and the median is 20:65:15, both of which are silt loam in texture;
however, some outliers were notably more sandy (sandy loam and loamy sand) (Table
5.5).

The texture of the 33 loess samples overlying either Edgar or Merrill till is silt
loam (Figure 5.8 A, B). The mean and median sand:silt:clay ratios were similar to each

other, showing little deviation from the central trend of the particle size distributions.

 
 
   

 
 
   

I I
A. Loess C ay B. Loess Cay
overlying overlying
Edgar till Merrill till

(n=18) (n= 15)

  
 
 
 
   

  
 
    

 

 

 

 

Sand Silt Sand Silt

Clay Clay

C. Loess D. Loess

overlying overlying
Bakerville till sandstone
(n = 38) residuum

 

(n= 10)

  

 

 

 

 

Sand Silt Sand Silt

Figure 5.8: Particle size distribution of the less than 2 mm fraction for mantle (loess) samples grouped
according to underlying substrate in north central Wisconsin. A. Loess samples overlying Edgar till. B.
Loess samples overlying Merrill till. C. Loess samples overlying Bakerville till. D. Loess samples
overlying sandstone residuum.

108

Loess overlying Merrill till tends to be slightly finer than the loess overlying Edgar till,
based on the mean and median mean weighted particle sizes (MWPS) determined by
laser diffraction analysis (Table 5.5). Several factors could account for the finer textured
loess overlying Merrill and Edgar tills. First, both Merrill, and especially, Edgar till have
higher silt contents than Bakerville till samples; thus, the siltier textured loess could be
due, in part, to inmixing of finer-grained substrate into the overlying loess. Since the
mean MWPS for Edgar till samples is 74.1 um and that of Merrill till samples is
considerably higher (130.3 um), in-mixing of underlying till does not sufficiently explain
the finer MWPS of loess overlying Merrill till. Second, the finer textured loess overlying
Merrill till may be due to it being further from the presumed source area (see section 5.3
below) compared to loess overlying Edgar till (Smith 1942; Ruhe 1954; Frazee et a1.
1970; Olson and Ruhe 1979; Muhs and Bettis 2000). This explanation assumes one
general source area. Alternatively, multiple source areas, which differ in particle size
characteristics, could account for the finer texture of loess overlying Merrill till.

Mean sand:silt:clay ratios for loess overlying Bakerville till is 25:60:15, and the
median ratio is slightly more silty (20:65:15); both are silt loam in texture (Table 5.5).
The particle size distribution for locss samples overlying Bakerville till mostly fell within
the silt loam texture class; however, loam and sandy loam textured outliers are noticeable
(Figure 5.8 C). The sandier mantle outliers skew the mean MWPS by more than 20 pm
from the median. These outliers could be samples taken from disturbed sites, where
another mechanism, i.e., bioturbation, has allowed underlying till to be mixed into
overlying loess. However, this seems an unlikely explanation as (1) soil pit

investigations revealed clear to abrupt contacts at the lithologic discontinuity, (2)

109

sampling protocol specifically sought out sites devoid of obvious disturbance, and (3) the
average loess thickness overlying Bakerville till is >55 cm, which is greater than most
bioturbative processes can operate. Alternatively, and depending on the spatial
distribution of particle size data, the sandy outliers may be indicative of the transition
from silty loess to sandy loess/coversands with increasing proximity to a source area
(Catt 1977; Pye 1995; Crouvi ct a1. 2008).

The mantle samples overlying sandstone residuum had a much larger range of
particle size distributions, compared to that of loess overlying any of the three till
members (5.5D). The mean sand:silt:clay ratio for loess overlying sandstone residuum
was 56:34:10 (fine sandy loam), whereas the median ratio was 51:39:10 (loam). The
mean MWPS was slightly higher than the median MWPS for loess samples overlying
residuum, both of which were much higher than the mean and median MWPS of loess
overlying till (Table 5.5). Additionally, of the sand contents present in the various
samples, only the very fine sand fractions remained consistently high. In the loess
samples overlying any one of the three tills, the 125-2000 um fractions were relatively
low; however, in the loess samples overlying residuum, only the 500-2000 um fractions
were relatively low.

In sum, 66 out of the 78 loess samples had a silt loam texture. Mantle samples

overlying sandstone residuum were the sandiest and had particle size distributions that
varied the most (ranging from loamy sand to silt loam). Loess samples overlying
Bakerville till were dominantly silt loam in texture, but did contain a number of sandier

(loam and sandy loam) outliers. All loess samples overlying Edgar and Merrill till were

110

silt loam in texture and trendcd toward MWPS values of approximately 51 and 45 pm

respectively.

111

5.3 Spatial Characteristics of Loess in North Central Wisconsin

Thickness, particle size data, and silt mineralogy for loess samples were analyzed
spatially in ArcMap 9.2 (ESRI software, Redlands, CA). Prediction (interpolation) maps
were created using ordinary kriging, with a smoothing factor of 0.5 within the
geostatistieal analyst extension. Spatial characteristics of loess deposits not only aid in
the identification of a sediment as eolian, but also can point to provenance (Schaetzl and
Hook 2008). Loess deposits, in cross-section, take on wedge-shape form, with the
thicker areas adjacent to the source, and thinning downwind (Smith 1942; Olson and
Ruhe 1979; Muhs and Bettis 2000). Similarly, loess deposits are coarsest near the source
and fine, texturally, downwind (Smith 1942; Ruhe 1954, Frazee et al. 1970, Olson and
Ruhe 1979; Muhs and Bettis 2000).

Mineralogy data have been also used to determine provenance and source regions
of loess in numerous studies in the Midwest (Leigh 1994; Grimley 1996, 2000; Grimley
et a1. 1998; Schaetzl and Loope 2008). These studies ofien reveal how paleowind patterns
and source areas change over time. If two distinct sources for loess in north central
Wisconsin exist, spatial trends in texture, thickness and mineralogy may reflect these two

sources.

5.3.1 Thickness Data

In general, the north central Wisconsin loess sheet exhibits a wedge-shaped
thickness trend consistent with that of loess elsewhere (Smith 1942; Olson and Ruhe
1979; Muhs and Bettis 2000). Within the study area, sampled loess thicknesses ranged

between approximately 35 and 100 cm. This range is generally consistent with other

112

reports of loess thickness in north central Wisconsin; however, pedons with slightly

thicker loess (100-150 cm) have been reported in north central Wisconsin (Hole 1942;
Cahow 1976; Hole et a1. 1976; Bartelme and Strelow 1977; Attig and Muldoon 1989;
Fiala 1989; Sutherland 1989; Clayton 1991; Attig 1993; Simonson and Lorenz 2002).

Loess is thickest (> 75 cm thick) in central and northern Clark County (Figure 5.9).

 

Marathon C'.--

Wood Co.
“/2,

 

 

 

Loess Thickness (cm) . ‘ . \
[:1 < 24.6 - 50.0 - 54.2 - Loess sample (537i 1;.I;:,»m}
|:] 24.6 — 38.3 - 54.2 - 61.7 J—‘- Late W1 moraine (LGM) ’ 7 i Z l
:71.
1:] 38.3 - 45.8 - 61.7 - 75.4 D Study area 0 2 4 Mflesv,

45.8 - 50.0 - > 75.4 [:3 County border

       
 

Figure 5.9: Kriged map of predicted loess thickness based on 79 sample locations in north central
Wisconsin. Dark shades represent thickest areas of loess (>75 cm), and light/white shades represent loess

thickness <25 cm.

113

Loess thickness patterns show a primary trend of decreasing, gradually, from the
northwest to the southeast, across the study area. In northern Clark and northwestern
Marathon Counties, loess thickness decreases from the thickest areas (> 75 cm thick) in
the northwest, to approximately 40 cm thick towards the southeast, over a distance of
approximately 30 km. This trend may reflect dominant paleowinds from the northwest,
consistent with reported wind directions from other areas in the Midwest during the last
glaciation (Ruhe 1954; Muhs and Bettis 2000). Northwest and proximal to the thickest
areas of the loess sheet in northern Clark County, is the Late Wisconsin terminal moraine,
which makes it a possible source of loess.

In central Clark County, where loess is >75 cm, a secondary, and more
complicated, decrease in thickness occurs from west to east (Figure 5.9). Loess thickness
decreases by at least 30 cm over a distance of approximately 40 km, from where it is
thickest, in central Clark County, into western Marathon County, where it is only about
40 cm thick. Based on this trend, it is very likely that more directly westerly winds also
occurred during a time of intense eolian activity, which is consistent with findings from
elsewhere in the upper Mississippi River Valley (Leigh and Knox 1994; Mason et a1.
1994; Muhs and Bettis 2000). These spatial trends suggest a secondary, western source
for the loess. In central Clark County, loess also decreases, albeit much more sharply,
towards the west. Here, the loess sheet decreases in thickness from > 75 cm thick to
approximately 30 cm thick over a distance of approximately 10 km. This trend could be
due to storm events or periodic changes in wind direction from westerly to easterly
winds, which has been documented in other loess sheets along the Upper Mississippi and

Missouri River Valleys (Fehrenbacher et al. 1965, Muhs and Bettis 2000). This

114

explanation requires a source area with relatively discrete eastern and western extent
(valley train or wide river valley/ floodplain). Since no such source is evident in the study
area, variable wind directions do not sufficiently explain this pattern. Alternatively, the
sharp decrease in thickness to the west could be a transitional zone between the source
area, which supplied eolian sediment to the west, and the depositional area to the east.
The area exhibiting the sharp decrease in loess thickness to the west approximately
corresponds to the eastern extent of near-surface Cambrian sandstones, which suggests
Hole (1942) correctly identified the area dominated by Cambrian sandstones as a western
loess source.

In the southeastern comer of the study area, a relatively thick area of loess is
shown on the kriged map (Figure 5.9); however, this trend is based on a single
measurement. It is difficult to say whether this measurement is an outlier, and thus,
should be excluded from the interpolation, without the benefit of additional, nearby data
points. Unfortunately in this area, as well as other areas (i.e. east central Clark and
southwestern Marathon Counties), additional samples were unable to be collected in
accordance with protocols, due to the low relief terrain and ubiquitous farming practices
on uplands.

Based on the loess thickness trends, it appears the loess sheet has two main source
areas. Loess is thickest in the northwestern part of the study area, tapering to the
southeast. This trend points to a source associated with the Late Wisconsin terminal
moraine, or points nearby. A secondary area of relatively thick loess is located in central

Clark County, decreasing both to the east and southeast. This trend suggests that a

115

secondary, western source may have contributed eolian sediment to the north central

Wisconsin loess sheet.

5.3.2 Particle Size Data for the Loess Mantle

5.3.2.1 Spatial Trends in Sand Fractions

The geography of particle size trends in the loess sheet shows a gradual, but
dominant, fining southeastward pattern (Figure 5.10). The 125-175 um (fine-fine sand)
and 175-250 urn (coarse fine sand) contents of the loess sheet are most highly
concentrated in the southwestern part of the study area (Figure 5.10 AB) and coincide
with the approximate eastern extent of locations where fine grained Cambrian sandstones
(interbedded with shale) occur at or near the surface (Figure 5.1). Concentrations of
these particle size fractions gradually decrease toward the north and central parts of the
study area, where concentrations of the 125-175 and 175-250 um fractions are the lowest.

A secondary area of higher concentrations of these two size fractions occurs in the
eastern half of the study area, which is underlain by Edgar till. Of the sand present in
Edgar till samples, very fine and fine sand are the most abundant fractions; thus, the
secondary area of higher concentrations of the 125-250 urn fractions is probably, at least
partially, the result of long-term pedoturbation (i.e. bioturbation and/or cryoturbation), in
which coarser substrate material was mixed into the finer (and thinner) loess mantle.
These patterns suggest that little, if any, of the 125-250 um particles were transported
great distances by eolian processes. It is much more likely the relatively high
concentration of these fractions are due to winnowing processes of soils formed in the
fine grained sandstone residuum in the southwestern study area or by in-mixing from the

substrate below, in the eastern region.

116

.E...

meow .N. 5.. 273
.u .51 £22 .m .51
03.2. .< H28.85..
own 0.2.5.. 9.552....
2.. .0 2035.588
.825. 2.. .5852
85.... 2535....
E... acormhcoucg
52w... .5859.
83% $8.55
5.258.? .558
..to: E 22.2.55...
N.N.m 0.3.1:...

02.330 .0 mums.
.555. "....m 2:5

 

 

 

 

 

 

117

.E... «.-N .= ”c...
m~-N. .U E... mm-m~ ..—
N..... 3.3 .m 8:28....

36 22...... m...>>o..o.
o... .o 8.8.2828
.83... 2.. E383.

move—E 2.2.52.3.
.2... .mco..m..=oo:8
.wosw... E383.
3.52m Sufism.
5.58.? 3.2.8
..to: E 22.2....3...
..N; 20...»... no???
t... 3...: pewtv.
5:3 c—.m 2:»..—

i”

..‘v7
A a

a_-¢‘-|‘

7.3

‘.
..

 

two.

 

 

 

 

 

 

 

118

The 75-125 um (coarse very fine sand) fraction in the loess exhibits the same
basic trend as does the 125-175 and 175-250 um fractions; the highest concentrations are
located in the southwestern corner of the study area, decreasing in concentration to the
northeast (Figure 5.10 C). Unlike the previous particle size fractions discussed, however,
the 75-125 um fraction exhibits a parabolic protrusion of moderately high concentrations,
eastwardly across the southern half of the study area. The 50-75 pm (very-very fine sand)
fraction exhibits the same protrusion towards the east; however, the areas of highest and
lowest concentration change dramatically (Figure 5.10 D). The southwestern comer of
the study area contains the lowest concentrations of all size fractions smaller than 75 um
(Figure 5.10 D-H). Taken collectively, these data suggest paleowinds from the west may
have winnowed the newly exposed sediment, leaving coarser fractions (> 125 pm) in the
southwestern comer of the study area (Figure 5.10 A, B) and depositing finer particle
sizes (< 125 um) downwind, to the east. In this scenario, the decrease in loess thickness
to the west discussed previously likely represents a transitional zone between erosional

and depositional areas, rather than due to variable wind directions.

5.3.2.2 Spatial Trends in Silt Fractions

Like the 50-75 pm fractions, the highest concentrations of 35-50 pm fractions in
the loess are in the south central parts of the study area, but there also occurs an area of
high concentration in the northwest, near the Late Wisconsin moraine (Figure 5.10 D, E).
From here, the 35-50 pm fraction decreases toward the southeast and northeast. The
highest concentration of 25 -35 um contents occur in the northwest and expand across the
area distal to the Late Wisconsin moraine. The concentrations of 25-35 pm particle sizes

decrease both to the southeast and northeast (Figure 5.10 F). The western edge of the

119

area most highly concentrated in particle sizes between 25 and 75 um extends from
northern through central Clark County and is nearly coincident with the western edge of
soils mapped with a loess mantle (Figures 5 .10 D-F, 5.11). The western edge of mapped
loess in central Clark County also corresponds to the Black River, which may have
trapped saltating sands moving from the west, allowing silt to remain deposited uplands
east of the river (sensu Mason et a1. 1999).

The concentrations of the 2-25 pm fractions are relatively high in the northern and
eastern half of the study area, with the highest concentrations located along the LGM
moraine in the north central parts of the study area (Figure 5.10 F-H). Equally high
concentrations of the 2-25 pm particle sizes are located in the southeastern comer of the
study area (Figure 5.10 G, H). Concentrations of these finer silt particle sizes decrease
gradually toward the southwest as well as toward the northwest; which may reflect source
regions not only to the southwest, but also to the northwest (in the direction of the LGM
moraine).

An area of relatively low concentrations of the 2-25 pm fraction protrudes
eastward, in a pattern similar to that of relatively high 50-125 um concentrations
described above (Figure 5.10 C, D, G, H) in the study area. This pattern is probably due
to the dilution of the 2-25 um particle sizes from a coarser, southwestern source.
Similarly, the northern region along the LGM moraine that contains the lowest
concentrations of >75 um particle size fraction also exhibits the highest concentration of
2-35 pm particle sizes. In this case, dilution via a northwestern source of finer grained
loess is likely responsible for these concentration patterns. The parabolic pattern of

relatively high 75-125 um concentration and low 2-25 um concentration in the southern

120

Figure 5.11: Hillshade ofa
DEM overlain by the kriged
loess thickness surface (gray
tones: this study) overlain by
soils formed in loess (red. pink
and fuchsia colors; Bartelme
and Strelow 1977, Fiala 1989,
Simonson and Lorenz 2002).
Note that the kriged loess
thickness surface closely
corresponds to the extent of
loess mapped Clark, Marathon.
and Wood Counties, Wisconsin.

| Marathon Co.

 

Wood Co.
- |'_——'—T—1

harass. o s ......

. Thickness (cm)
- Spencer Slit loam (> 100 cm)
1:] Withee silt loam (~60 cm) :1 < 24'6
[:1 Loyal silt loam (~50 cm) E 24.6 ‘ 38'3
[:1 Hiles silt loam (~50cm) 5:1 38-3 ' 45-8
Marathon County 45'8 ' 50
C] Withee silt loam (~60 cm) - 5° ‘ 54-2
:1 Loyal silt loam (~50 cm) - 54-2 ‘ 61-7
Wood County - 61.7 - 75.4
E] Wlthee silt loam (~60 cm) - > 75'4
[:] Santiago silt loam (~60 cm) 4“- LGM
[:1 Kert silt loam (~55cm) D Study Area

121

half of the study area is probably due, at least in part, to the greater amounts of finer
grained loess from the northwest diluting the coarser westem-sourced loess.

An area with high concentrations of the 2-25 pm (fine and very fine silt) fractions
also appears in the southeastern comer of the study area. Unlike the thickness prediction
map, in which this area also exhibited thicker loess, the particle size trend is not based on
a single data point. This area also seems to have slightly lower 35-75 um contents. In
this case, the Marshfield moraine was likely enough of a topographic barrier that only
eolian sediment finer than 25 um were able to carried over it (sensu Pye 1995; Mason et
al. 1999).

In summary, the spatial trends of the different particle size fractions in the loess
mantle are consistent with loess deposits elsewhere in the Midwest, where coarse loess is
dominant nearest the source, and fine and very fine silt are most abundant downwind
(Smith 1942; Olson and Ruhe 1979; Muhs and Bettis 2000). The spatial trends of these
particle size fractions suggest two possible sources of eolian sediment for the north
central Wisconsin loess sheet. The relatively high concentrations of the 50-75 pm
fractions in the south central parts of the study area, combined with the gradual decrease
primarily towards the east (Figure 5.10 D), suggest that westerly paleowinds supplied
dominantly very fine sand sized sediment to the loess sheet, from this source region,
where fine-grained Cambrian sandstones are near to and crop out at the surface.
Similarly, the higher concentrations of 25-50 pm fractions to the northwest, decreasing
primarily to the southeast, suggest a finer source of loess in the direction of, or within, the

Late Wisconsin moraine.

122

5.3.3 Silt (Feldspar) Mineralogy Data

In general, the loess across most of the northern half of the study area contains
relatively high amounts of plagioclase feldspar, compared to the southern half of the
study area (Figure 5.12). Plagioclase contents are highest in the northwestern part of the
study area, and decrease to the southeast. This extension of relatively high plagioclase

towards the southeast, in addition to the higher contents of plagioclase in the north,

1 Marathon Co.
:0 WoodCo.

 

Plagioclase contents (%) 4 T

[:1 < 12.9 - 17.4 - 18.2 o Loess samples n ' "9 5:
[312.9449 -18.2-19.4 LGM ‘ " ‘
CI 14.9- 16.2 - 19.4-21.4:13tudyArea ° 2 “Wes i
-16.2-17.4->21.4 HAT

 

Figure 5.12: Spatial distribution ofplagioclase contents in loess of north central Wisconsin. Plagioclase
percentage determined using equation 5.1 (Grimley 1996).

123

support a model of northwestern paleowinds carrying sediment high in plagioclase,
downwind and to the southeast. This trend implies that the Late Wisconsin moraine to
the northwest, may have been a loess source area. High plagioclase contents also occur
in the far eastern regions of the study area (Figure 5.12). This interpolated area of
relatively high plagioclase contents, however, is based on one point, which may be an
anomaly, and thus maybe excluded from analysis. The lowest plagioclase contents occur
in the southwestern comer of the study area, near to where Cambrian sandstones occur at
or near the surface.

Like the spatial trends in plagioclase contents, the highest contents of K-feldspar
occur in the northwestern part of the study area and gradually decrease to the east, south
and southeast (Figure 5.13). This trend may reflect northwestern paleowinds carrying
sediment rich in K-feldspar from a source to the west and northwest. A secondary area of
relatively high K-feldspar contents occurs in the south central part of the study area, and
extends east and northeast. This pattern may be due to paleowinds carrying potassium-
rich feldspar from a source area to the west or southwest of the study area. The lowest
concentrations of K-feldspar occur in the northeastern comer of the study area, which
indicates that this area may be furthest from the K-feldspar source, or it is proportionately
diluted by quartz or plagioclase contents. A combination of these two explanations is

likely the cause of the relatively low concentration of K-feldspar in this area.

124

 

 

!

   

K-Feldspar contents (%)
I: < 7.4 - 10.6 - 11.7 o Loess samples

[37.4-9 -11.7-12.7 LGM
_ 9 -10 - >12.7 DStudy Area
- 10 -10.6

Figure 5.13: Spatial distribution of K-feldspar contents in loess of north central Wisconsin. K-feldspar
percentage determined using equation 5.1 (Grimley 1996).

5.3.4 Summary of Spatial Trends

The spatial characteristics of loess particle size, thickness, and feldspar
mineralogy data suggest paleowinds were from the northwest and west/southwest during
the time of loess deposition. These results do not support the anticyclonic paleowinds

(dominantly easterly winds) predicted by COHMAP members (1988), but instead, are

125

consistent with northwesterly and westerly paleowinds derived from loess distributions
elsewhere in the midcontinent (Smith 1942; Ruhe 1954; Fehrenbacher et a1. 1965; Frazee
et al. 1970; Putman et al. 1988; Mason et al. 1994; Muhs et al. 1999; Aleinikoff et al.
1999; Muhs and Bettis 2000; Mason 2001). Using the inferred paleowind directions, the
spatial patterns also indicate that the loess in north central Wisconsin was probably
derived from two source areas. The source to the northwest appears to have contributed
finer, siltier loess, richer in plagioclase feldspar. Conversely, the western, or
southwestern source appears to have contributed coarser (very fine sand-sized) sediment
with little plagioclase. K-feldspar seems to come mainly from both the northwest and the
west/southwest.

Loess is thickest in central Clark County, corresponding roughly with the eastern
extent of Cambrian sandstones, and also in northern Clark County, along the Late
Wisconsin terminal moraine, thinning to <40 cm to the east and southeast. The sharp
decrease in loess thickness towards the west in central Clark County is not likely due to
storm events (sensu Fehrenbacher et al. 1965; Muhs and Bettis 2000), as (l) a wide
valley source could not be identified within the area of thick loess and the and (2) particle

size trends do not fine westward.

126

5.4 Potential Source Sediments

Based on the spatial patterns of thickness, particle size data, and silt mineralogy
of the loess in north central Wisconsin discussed above, it is possible that two source
areas contributed sediment to the loess sheet in north central Wisconsin. The source area
to the northwest contributed silt-rich sediment with relatively high amount of plagioclase
(Figures 5.10, 5.12), whereas the source to the southwest/west was coarser (sandier), and
contained more K-feldspar (Figures 5 .10, 5.13). This section will focus on the
characteristics of the two potential source sediments (l) to the northwest and (2) to the
west/southwest, and compare them, in light of the characteristics of loess in north central

Wisconsin.

5.4.1 Northwestern Source

The entire Late-Wisconsin terminal moraine, located northwest of the loess sheet,
likely contributed much sediment downwind as the ice within the moraine began to
degrade causing great landscape instability. This section will focus on one particular
landform found on and behind the moraine, ice-walled lake plains, which could be
relatively large source areas fine-grained sediments (silts and clays) distilled from the
surrounding silty till.

Numerous ice-walled lake plains exist on and behind the Late Wisconsin moraine
in Chippewa and Taylor Counties (Figure 5.14; Cahow 1976; Attig 1993; Ham and Attig
1997; Syverson 2007). Ice-walled lake plains initially form as lakes in stagnant glacial
ice, collecting supraglacial sediment, and are best preserved where the bottom of the lake

contacts solid ground (Attig 1993). These features are typically the highest on the

127

landscape and are comprised of offshore silts and clays (often laminated) in the center,
which coarsen towards the rim (Attig 1993; Syverson 2007; Clayton et al. 2008). Once
these lakes drained, this fine grained sediment could have been easily entrained by the

wind, until they became vegetated.

Taylor Co.

Clarkl
A

r Marathon Co.

Clark 2
A

 

Wood Co.

        

z

 

Loess Thickne s cm) 1 - Ice walled lake plain sample
< 24-6 50 ' 54-2 o Loess sample
[:1 24.6 - 38.3 - 54.2 - 61.7 A 30" p" _
[:1 38.3 - 45.8 - 61.7 - 75.4 LGM °l 3 6, ”“5
45-8 - 50 - > 75-4 D Study Area
- Ice walled Lake Plaln n L,

Figure 5.14: Location ofiee walled lake plains (blue) in Chippewa and Taylor Counties, Wisconsin (Attig
1993. Syverson 2007) with respect to kriged loess thickness. Locations of loess sample points represented
by black dots. and the locations of soil pedon descriptions are represented by white triangles.

128

The mean sand:silt:clay ratio for the nine lake plains sampled in this study is
17:65:18 (silt loam), and the mean MWPS is 38.9 um (Table 5.6). Of the sand present in
the samples, the 50-125 um (very fine sand) fraction was the highest, which on a clayfree
basis accounted an average of 16.6% of the total clayfree sand content (~20%). Contents
of sand > 125 um usually accounted for <1%, and did not exceed 4% in any sample.
When the fine earth fraction is plotted on a logarithmic scale (in pm), the particle size
distributions of ice-walled lake plain samples and most of the loess samples are strikingly
similar. Sediment from most of the silt loam-textured loess samples and lake plain
samples peak within the 20-45 pm range (Figure 5.15). Even sandy loess samples (those
with loam, sandy loam, or loamy sand texture) have a secondary silt peak, albeit small in
some cases, within the 20-45 pm range.

Analysis of the feldspar mineralogy of the 20-45 pm fraction of ice-walled lake
plain samples reveals that both plagioclase and K-feldspar grains are present (Figure
5.16; Table 5.7). The plagioclase peak is < 14.1% of the quartz peak in all samples; the
average being 9.4%. A K-feldspar peak is present in 8 of the 9 samples, with K/Q peak
height ratios <4.5% (mean = 3.2%). The average plagioclase contents are 22.5%; no
individual samples exceeded 33%. K-feldspar contents, in general, ranged between 7.9
and 12.2%, the median of which is 9.6%. In comparison, plagioclase contents averaged
17.5% and did not exceed 25% in the loess samples analyzed, whereas the K-feldspar
contents averaged 10.5% and did not exceed 15%. As the kriged plagioclase contents
data suggests (Figure 5.12), the highest plagioclase contents were in the northwest,

nearest to the Late Wisconsin moraine; thus, it is likely much of the plagioclase in the

129

.om-mm H50 69mm ”$2 "3.2 Emu mNTN awn; mooom.ooo_ HmU> Mcoo—loom ”mo mcomémm ”m2 6mm-m~_ ”mn— mm~_-om ”m”; .81 E 20:23 85 22th 9
A83 0823 .3 on: 22th BEE?» =88 ”ma—>22 a

a: cam 25 a

 

 

 

 

<2 3. 3:” 2: 2. 3 2 2 am— 35 58. :7. ES”: 9mm :38:
am. 2: 33 2: 2. E .3 S as. as" .53. 5... meme: 3m :82
3.. S; 3“ we 3 3 am am 3: ER :32 as 2 ”Sam 3% a
I. a: 2% 2m 3 2 so no a: 3“: 88. ..a 8&3: 23 I 5% v__
3; NS cam o. _ N S 3 2 2 v.2 a“: :52 :5 Gas: 92 o 5% c:
_.2 _..: 3m v.2 3 3 go 3 a: 3E :52 .E M: as: m. _ m a sea V=
3: Q: «.2 :8 3 9o 3 3 mi £5 58. as E “Es: Sm m 5% U=
v.2 v.2 cam 52 3 3 mm 3 2: ER 58. as 2 ”Nana 22m 9 22a V=
a: 3: 3; S: 3 i. 2 3 a: :2: 58. ..s t “SE Em m 5% V=
E: N: EN 3. mo 3 E 3 :2 $2 :52 ..a 2 “8an mm < 5% V=
fl: 3: 3m <2 od 3 ed 3 S: :«2 58. am 8 ”was 3N seen
05 as: as... 05“; 08> omu om: am; am: im hump 95mm 99:52 aim—ham
mama—>30 «Egg m2:—

 

 

.=_m=3m_>> .9559 5.3: E 8.9:: Ea:- 33 3.73703 .8.— 32. on: 2322— conga—o E... 5.59 2:..— 66 05:.

130

Figure 5.15: Sediments
from ice walled lake plain
samples (red) peak between
20—45 pm as do the majority
of the silt loam loess
samples (black). Loam loess
samples are shown in blue.
Sandy loam loess samples
are shown in pink. Loamy
sand loess samples are
shown in yellow. Top:
Particle size (logarithmic
pm) distribution of ice-
walled lake plain and loess
samples according to texture
class. Bottom: Locations of
ice walled lake plain and
loess (mantle) samples in
north central Wisconsin.

— lee walled lake plain samples
— Silt loam loess samples
_. Loam loess samples

— Sandy loam loess samples
-— Loamy sand loess samples

  
    
   
  

    

a \
3 ‘5’; «

/‘\ ‘

l
/’L

, lk plain H ’Kg _
. _ [k plain F Ik plain [/1 \ .
Ik plain 8 IR plain G
- Ik plain E Taylor Co.

' ,‘ Ik plain D
Ik plain A /k plain C /
.:' , W

Eau Claire Co.,

‘ 3f . .

   

 

Samples Loess Thickness (cm) 0 8 km « ar’frv
- lce walled lake plain l: <24.6 - 50 - 54.2 ‘ ' " ”
Loess [:1 24.6 — 38.3 - 54.2 - 61.7 34,5;
. silt loam [:1 38.3 — 45.8 - 61.7 - 75.4 ‘
e loam - 45.8 - 50 - > 75.4 g y
' sandy loam LGM D Study Area Last Glacial Maximum
loamy sand [:1 County border (LGM) —

131

loess may have been derived from the ice-walled lake plains. Alternatively, K-feldspar
contents in some loess samples exceed that of ice-walled lake plain samples; therefore, a
secondary source must have contributed additional K-feldspar contents to the loess in
north central Wisconsin. In summary, the silt mineralogy data suggest ice-walled lake
plains and the Late Wisconsin terminal moraine would have contributed quartz and

plagioclase feldspar dominantly, and K-feldspar to a lesser extent.

 

 

 

 

 

 

 

 

 

 

I — Ice walled lake plain samples
.. Sandstone/sandstone residuum samples
Quartz l
K-feldspar
l Plagioclase
l , .2 fM
[ii 2 w .: :::
/7“ f - V
j. , ,2“ ,a-A-A
25 26 27 28 29 30 31

Figure 5.16: X-ray diffraction results of the 20-45 pm fraction for ice walled lake plain samples and
sandstone and residuum samples in north Central Wisconsin. Ice walled lake plain samples (dark lines)
contain quartz (26.5 2(9) and plagioclase (approximately 27.8 29), and very little k-feldspar (27.5 29).
Sandstone and sandstone residuum samples contain no plagioclase, and higher amounts of k-feldspar.

132

Table 5.7: Silt mineralogy of source and loess samples in north central Wisconsin.
Sample 11) Q 3.34/11 K 3.254 P 3.20A 109* Plot %Q % K§ % Pif

 

 

 

 

 

 

 

 

 

 

Delman 8146 316 677 3.9 8.3 69.2 10.7 20.1
Lake plain A 6738 282 523 4.2 7.8 69.5 11.6 18.9
.2 Lake Plain B 7868 258 673 3.3 8.6 69.9 9.2 20.9
L: Lake Plain D 7267 1022 0.0 14.1 67.0 0.0 33.0
3 Lake plain E 7519 280 883 3.7 11.7 64.1 9.5 26.3
g lake Plain F 8580 278 752 3.2 8.8 69.6 9.0 21.4
5.: Lake Plain G 7714 342 614 4.4 8.0 68.7 12.2 19.1
a Lake Plain H 7388 273 704 3.7 9.5 67.5 10.0 22.5
E 1k plain I 8648 237 679 2.7 7.9 72.2 7.9 19.8
Mean 7763 283 725 3.2 9.4 68.6 8.9 22.5
Median 7714 279 679 3.7 8.6 69.2 9.5 20.9
SSlC 1828 596 32.6 0.0 43.4 56.6 0.0
ss2A 4220 233 5.5 0.0 81.9 18.1 0.0
E ss3c 8133 637 7.8 0.0 76.1 23.9 0.0
g SS4A 3040 502 16.5 0.0 60.2 . 39.8 0.0
g 5848 2776 419 15.1 0.0 62.4 37.6 0.0
g 17 Sub 4408 976 22.1 0.0 53.0 47.0 0.0
g 50 Resid 5147 1372 26.7 0.0 48.4 51.6 0.0
E EClR 912 339 37.1 0.0 40.4 59.6 0.0
Mean 3808 634 20.4 0.0 58.2 41.8 0.0
Median 3630 549 19.3 0.0 56.6 43.4 0.0
1 Cap 8132 402 732 4.9 9.0 66.1 13.1 20.8
12 Cap 7804 304 616 3.9 7.9 69.8 10.9 19.3
14 Cap 7358 288 540 3.9 7.3 70.8 11.1 18.2
16 Cap 8804 372 844 4.2 9.6 66.5 11.2 22.3
20 Cap 9141 272 316 3.0 3.5 80.6 9.6 9.8
23 Cap 8925 142 551 1.6 6.2 78.1 5.0 16.9
24 Cap 8140 359 511 4.4 6.3 71.6 12.6 15.7
26 Cap 8268 353 517 4.3 6.3 72.0 12.3 15.7
27 Cap 7578 293 441 3.9 5.8 73.6 11.4 15.0
31 Cap 9444 330 468 3.5 5.0 76.1 10.6 13.2
g 33 Cap 8913 480 744 5.4 8.3 66.3 14.3 19.4
3 39 Cap 8597 362 585 4.2 6.8 71.1 12.0 16.9
41 Cap 7776 274 631 3.5 8.1 70.2 9.9 19.9
42 Cap 8485 230 555 2.7 6.5 74.8 8.1 17.1
45 Cap 9272 208 594 2.2 6.4 76.1 6.8 17.1
47 Cap 8206 301 598 3.7 7.3 71.3 10.5 18.2
55 Cap 7140 321 431 4.5 6.0 71.9 12.9 15.2
181 Cap 7099 280 767 3.9 10.8 65.1 10.3 24.6
244 Cap 8471 258 457 3.0 5.4 76.3 9.3 14.4
245 Cap 7305 229 599 3.1 8.2 70.8 8.9 20.3
Mean 8243 303 575 3.7 7.0 72.0 10.5 17.5
Median 8237 297 570 3.9 6.7 71.5 10.8 17.1
1Ratio of feldspar peak height to quartz expressed as a percentage. (Backgrounds determined and removed
by JADE).

§: (Mineral peak height*peak intensity)/Z all mineral peak heights)*100 (Grimley 1996).

133

5.4.2 Western / Southwestern Source

The Black River, with its headwaters in the Late Wisconsin terminal moraine,
runs along the westem extent of loess mapped in central Clark County, making it a
potential source area (Figure 5.17). During glacial retreat, it likely carried sediment—rich

meltwater from the LIS, and thus, may have contributed sediment to the loess sheet.

,,. ., ‘ V

l V: Z

t - V" ‘Q ‘ V . .
. “3mg? .3535;
. sq}... 9,1,“ ‘f

Clark Co.

 

Jackson Co. Wood Co.

Loess thickness cm
Figure 5.17: Distribution ofsoils formed in ( ) ° 5 1° km

  
 

loess with respect to the Black River in north - > 100
central Wisconsin (aficr Bartelme and Strelow [:1 51 - 100
1977: Fiala l989: Jakel and Dahl 1989: Mitchell

1996; Simonson and Lorcn7 2002; Boelter [:1 25 - 50
2005). .LL LGM

— Black River

D Study Area

{:3 County line

134

However, the Black River has a relatively narrow valley, making it an unlikely major
loess source. Additionally, in northern Clark County, relatively thick loess is found both
east and west of the river valley, which suggests the Black River was not a dominant
source of loess in this region.

The spatial trends of particle size data for the loess sheet also discount the Black
River as a major loess source. Particle size trends of the loess sheet (Figure 5.10) exhibit
a general coarsening westward pattern, which is especially apparent in central Clark
County. Here, the loess sheet becomes sandier west of the Black River. If the Black
River was the source of loess in north central Wisconsin, it would have contributed silty
sediments towards the east, and sandy sediments towards the west. The process by which
this could occur is difficult, if not impossible, to explain given our knowledge of eolian
systems. The particle size trends of the loess sheet are better explained using the surface
of transport model (sensu Mason et al. 1999). In this scenario, the sandy loess west of the
Black River is in a transitional zone between the sandy source area, dominated by
erosional processes, to the west/southwest and the area of thick, silty loess east of the
river. It is likely the transitional zone of sandy loess acted as a surface of transport (sensu
Mason et al. 1999), where particles traveling by saltation or short-term suspension stirred
up finer grained particles that settled, causing them to be re-entrained and carried further
downwind. Using this model, the Black River likely trapped the particles traveling by
saltation and short-term suspension, allowing finer particles traveling in suspension to
remain deposited east of its banks. This model accounts for both the sandy loess west of

the Black River and the silty loess immediately east of it.

135

West and southwest of the Black River (e. g. in the southwestern corner of the
study area), residual soils, such as Humbird fine sandy loam, were derived from the
friable fine-grained Cambrian sandstones interbedded with shale that underlie them, often
at a depth of < 150 cm (Figure 5.18; Weidman 1907; Hole 1976; Simonson and Lorenz
2002). Very fine sand (50-125 um) tends to be the most abundant particle size fraction
in nearly all samples of the Eau Claire Formation, ranging between approximately 26 and
63% (Table 5.1). On average, Mt. Simon samples contained lower contents of very fine
sand (22.1%) compared to that of Eau Claire samples (43.1%). Interpolated particle size
data show a protrusion of relatively high concentrations of this same fraction in central
and southern Clark County, which gradually decreases eastward (Figure 5.10 C, D), and
suggest that the friable Cambrian sandstones and their residuum are the western source of
the loess.

The feldspar mineralogy of the 20-45 pm fraction of the clay and silt-rich
sandstone samples illustrates that sandstone and sandstone residuum samples contain no
detectable plagioclase (Figure 5 .16; Table 5.7). Relative K-feldspar contents of
sandstone (and residuum) samples ranged between 18.1 and 59.6%, averaging 41.8%. K-
feldspar is concentrated more in samples rich in shale (i.e. 881C, 17 Sub, 50 Resid, and
EClR) regardless of formation (Tables 5.1, 5 .7), which in general, is consistent with
other reports (Morrison 1968; Asthana 1969; Ostrom 1970; Distefano 1973). Contents of
feldspar for the Mt. Simon Formation samples are higher than is reported in most of the
literature (Ostrom 1970; Distefano 1973), and are closer to, but still exceed, the values
reported by Asthana (1969). It is important to note that this discrepancy is most likely

due to the difference in particle sizes analyzed. This study sought only to determine

136

i'

i Marathon C o.

Wflfi —

Wood Co.

 

 

Loess thickness (cm) 0 5' 1'0 km
[:] < 24.6 - 50 - 54.2 o Loess sample
C] 24.6 - 38.3 - 54.2 - 61.7 1“; Sandstone sample

[:1 38.3 - 45.8 - 61.7 - 75.4 [:1 Study Area
- 45.8 - 50 - > 75.4 [:] Depth to Bedrock < 152 cm

  

Figure 5.18: Distribution of soils (yellow) mapped in north central Wisconsin where underlying sandstone
interbedded with shale bedrock is at shallow depths (< 152 cm; afler Bartelme and Strelow 1977: Thomas
1977: Fiala 1989; Jake] and Dahl 1989: Langton and Simonson 1998: Simonson and Lorenz 2002). In the
southwest comer of the map. sandstone is near the surface on ridgetops. To the north and west bedrock is
near the surface in all landscape positions with the exception of alluvium-fillcd river valleys. In the center
of the study area. near-surface bedrock is only on backslope positions.

mineralogy of the 20-45 pm fraction; whereas the mineralogy reported in the literature
was based on much broader fraction ranges (either the < 125 um fraction or a bulk

sample). In summary. feldspar mineralogy data for the 20-45 pm fractions indicate that

137

sandstones and their residuum would have mostly contributed K-feldspar and quartz, and

little if any plagioclase to the loess Sheet.

5.4.3 Summary

Loess in north central Wisconsin appears to have originated from two distinct
sediment source areas (1) the Late Wisconsin moraine to the northwest including
numerous ice-walled lake plains on and behind the moraine and (2) the fine-grained
Cambrian sandstones interbedded with Shale (and their residuum) to the west/southwest.
The ice-walled lake plains to the northwest probably account for much of the finer-
grained (2-50 pm), plagioclase-rich sediment in the loess. K-feldspar is a minor
component of the ice-walled lake plain silt mineralogy. A11 ice-walled lake plain samples
contain higher plagioclase contents compared to that of loess samples; however, some
loess samples contained higher contents of K-feldspar than any of the ice-walled lake
plain samples (Tale 5.7). Thus, the loess samples with higher amounts of K-feldspar
must have received additional inputs of K—feldspar from a different source. The
Cambrian sandstones to the west and southwest likely contributed dominantly coarser
grained sediment (SO-125 pm). They also likely contributed sediment to the loess Sheet
that is rich in quartz and K-feldspar, but largely devoid of plagioclase. The high K-
feldspar contents in the loess of the west-northwest comer of the study area are, therefore,
due to relatively high contributions from both western and northwestern sediment

sources.

138

5.5 Age and Paleoenvironmental Significance

Known ages of ice-wedge casts in Wisconsin, when combined with pollen data,
from regions adjacent to Wisconsin, suggest the presence of cold, permafrost-dominated
conditions across north central Wisconsin between approximately 30 and 15.1 ka (Black.
1965; Birks 1976; Heide 1984; Baker et al. 1986; Johnson 1986; Clayton and Attig 1987;
Clayton et al. 2001). To the southwest of the north central Wisconsin loess sheet, in the
Driftless Area, active solifluction associated with a permafrost-dominated landscape is
known to have occurred between 22.3 and 13.8 ka (Mason 1995; Mason and Knox 1997).
Although it is likely that continuous permafrost occurred beneath the Late Wisconsin
moraine, and extended south of the study area at the time of the last glacial maximum,
the timing of permafrost degradation, along with most glacial events in Wisconsin,
remain poorly constrained (Clayton et al. 2001). I intend to Show that OSL ages from
loess samples can be an important step in understanding the post-glacial chronology in
north central Wisconsin. Not only will such ages indicate the timing of regional eolian
activity, but also, because the identified sources were located downwind of areas possibly
controlled by permafrost during the LGM, the sediment would likely not have been
available for deflation until the permafrost began to thaw. For that reason, OSL ages
from this research can be used to potentially constrain the timing of regional permafrost
degradation.

OSL samples were acquired from three locations within the study area (Figure
5.19), as described in chapter 4. The detailed stratigraphy of each location is discussed in
section 5 .2 above, and illustrated in Figures 5.4-5.6. The Clark 1 sample was collected at

a depth of 30 cm. at a Site where the loess is approximately 70 cm thick. This site is

139

Clark 3

A
12,950 :945

Clark 1
A

 

Marathon C 0.

Clark 2

A
14,160 :1040
Wood Co.

 

 

 

Loess Thickness (cm)
E < 24.6 - 50.0 - 54.2 A Soil pit/OSL sample

[:1 24.6 - 38.3 - 54.2 - 61.7 4—L Late WI moraine (LGM)

1:] 38.3 — 45.8 - 61.7 - 75.4 D Study area 0 2
45.8 - 50.0 - > 75.4 {:3 County border 1 1 1
4

 

Figure 5.19: Kriged loess thickness map showing the locations of OSL samples and estimated ages results.

located in a roadeut on the shoulder of a south-facing, 4% slope on a dissected, bedrock
upland in western Clark County, where the loess is typically more sandy (loam or sandy
loam) than is typical for the central Wisconsin loess sheet (Figure 5.15). The Clark 2
sample was collected at a quarry, from a depth of 30 cm, where loess is approximately 50

cm thick. The Clark 2 site is the furthest (southern most) site with respect to the Late

140

Wisconsin moraine. The Clark 3 sample was collected from a depth of 30 cm, where
loess was approximately 90 cm thick. The Clark 3 Site is located in a woodlot on the
summit of a dissected upland. Of the three locations OSL samples were collected, this
Site is nearest to the Late Wisconsin moraine and in the area of thickest loess. OSL age

estimates obtained from these three locations suggest, conservatively, that the period of

loess deposition in the study area occurred between 15.2 and 12.0 ka7 (Figure 5.19; Table
5.8), which post-dates the last glacial maximum by at least LOGO-2,000 years, and likely
thousands of years more (Attig 2008).

The OSL age estimates exhibit an interesting spatial pattern that support two
periods of loess deposition from two distinct sources (one from the west and one from the
northwest). Although the error ranges associated with each mean OSL age overlap, the
sample collected farthest from the Late Wisconsin moraine (Clark 2; Fig. 5.19) returned
the oldest age, whereas the northern most sample (closest to the Late Wisconsin moraine;
Clark 3; Fig. 5 .19) returned the youngest age. The spatial pattern of the OSL age
estimates, in combination with particle Size data from the three soil pedons described in
section 5.3 and above, may account for the approximately 90 degree difference in
paleowind direction between the two source areas. A likely explanation for this spatial
arrangement is presented below.

Permafrost to the west and southwest of the loess sheet probably began to thaw
first, as it was farthest from the ice margin. Landscapes experiencing thawing permafrost

would have caused large amounts of loamy and fine sandy sediment, weathered from the

 

7 The range of ages (15.2 to 12.0 ka) is a conservative estimate of the period of loess deposition in north
central Wisconsin. This range was determined by adding the standard deviation (one sigma) of the oldest
OSL age estimate (14,160i1040) and subtracting the same from the youngest OSL age estimate
(12,950i945). The range of mean ages for loess deposition in north central Wisconsin is much smaller -
14.2 to 13.0 ka.

141

moocaom _S:oE:o:>:m
9:: 53m— do E25330 .owaoEU E m_o::: Co 43582:: 05 “a @2883 codenamed wczao 8:885:51— 2: 3 wombac< .mefi 25 8 Pa 285 :<_..
.e\a m H om do EquQ 2:29: 3:5 62553 SW :5 .345: 60:2: :5 3835 do 32.3.6.8 Eot 8:43:08 5 do 22 owe: b: 2:58 a 8:22: mow< a

62:50 65:5 .mocoztonei— :o_§>:o< a wide 43 e283.“ cmv— ecu .59 .3 s
.Amoom ..e do :63 26:58. 80: 038853: 82:? 29:2: 2: 3 3:558: 80: 253333 2: 5.3 bozsoomme
AS: cm H oxwv mfiwcfloaes 3:5,: 98 AS: cm H o3 Em: 2:: AE: om H 3 3 Em: :ooew 3 5:965 8:865 5:233: ~= ES .5 JG 2:. e

 

 

55 H omoag w _ .o H Zum No.0 H mmd _.o H Wm _.o H m.~ vmd H _o.wv Amwnommv—D 0v m vim—U

o3: H $2.3 w_.o H boa Nod H wNN _.o H _.w _.o H N.N mmd H 50. _ m AmnnomriD Om N eta—U

coo H cam.m_ :d H mm.m Nod H Omfl _.o H m6 _.0 H qm mvd H omww Amonomb=g Om A V:a—U
% h e a: an

Cb eewe i. m O of: \ 39:: DA \ev ac: V DAE V 3.29 ea b.3223; swam—MW— :. 0155

3a.. omen— O~v_ .C. D 33. «av—«253m

 

.EmaeomS’ .9559 5.8: Ea...— moEEam amoe— he wows Ai—mOv 3:33....5: 3.3—:5: ESE—O "Qm 933..

142

friable, fine-grained Cambrian sandstones, to thaw, Slump, flow, erode and otherwise
become available for deflation. During this time, paleowinds were dominantly from the
west/southwest during the time loamy and fine sandy sediment was available for
deflation. Eventually, when the permafrost was mostly or entirely thawed in this region,
vegetation likely stabilized the landscape, preventing entrainment of the sediment.

At the time permafrost was degrading in the region west and southwest of the
loess sheet, the Late Wisconsin terminal moraine was likely both ice-cored and underlain
by permafrost, which probably had an insulating effect on the frozen landscape. Because
of this insulation, it is likely the moraine would have remained ice-cored well after
thinner ice north of the moraine had retreated. Later (~13 to 12 ka), ice within and
permafrost underlying the Wisconsin terminal moraine may have finally begun to
degrade, destabilizing the landscape and allowing sediment to become available for
entrainment. Additionally, the thawing ice-cored moraine likely caused numerous ice-
walled lakes on and behind the moraine to drain, and exposed the fine-grained lacustrine
sediment (mostly Silt and clay) to deflation. During the time the Late Wisconsin terminal
moraine was thawing, paleowinds were dominantly from the northwest, and carried
sediment from the moraine to be deposited downwind.

Particle Size data from the three soil pedons described in section 5.2 above,
support this chronology (Figures 5.4-5 .6; Table 5.4). In all three pedons, sand contents
(most of which is fine and very fine sand) increase with depth within the loess (Table
5.4). For example, sand contents in loess at the Clark 1 pedon increase from 30% in the
E horizon (29 cm) to 48% in the Bt/E2 horizon (60 cm), which is approximately 10 cm

above the lithologic discontinuity and underlying Bakerville till below. The Clark 1 site

143

is located on the western margin of the loess Sheet, where it is near the area where
Cambrian sandstones crop out. Similarly, loess at the Clark 3 site, which is nearest to the
Late Wisconsin moraine, increases in sand contents with depth. Here, sand contents
increase from 13% in the B horizon (32 cm) tol9% in the Bt/E2 horizon (73 cm), which
is approximately 18 cm above the lithologic discontinuity with underlying Merrill till.
These data suggest that earlier eolian contributions at these Sites were sand-rich, whereas
later contributions were silt-rich and sand-poor.

Pollen data collected from Wood Lake, which is in the northeast corner of Taylor
County and approximately 30 km north of the study area, support this interpretation. The
pollen assemblage at this lake suggests that paleoenvironmental conditions between 15.1.
and 11.5 ka supported a Picea -herb (spruce with tundra) ecosystem (Figure 5.20; Heide
1984). When the climate ameliorated, the degrading permafrost in the area likely
produced an unstable landscape — one that was largely devoid of vegetation at the time.
Given the sparse vegetation at the time, more steeply sloping landscapes would have been
even more susceptible accelerated erosion. Large quantities of sediment were, therefore,
newly exposed via processes such as mass wasting and slopewash, sensu Mason (1995)
and Mason and Knox (1997), to processes of wind, water, and gravity (Figure 5.20).
During this period, strong winds could have entrained the newly exposed sediment and
deposited it downwind, where it accumulated on relatively stable, sparsely vegetated
uplands. OSL age estimates conservatively place the window of loess deposition in north
central Wisconsin between approximately 15.2 and 12.0 ka (Figure 5.20; Table 5 .8),

which closely coincides with the window of tundra-spruce environment in Taylor

144

Spruce-tundra environment; Late WI -
4 . .
termmal moraine, Taylor Co., WI
3 Permafrost degradation; [.3
north central Wisconsin
2 _-\ecc1c1‘;1tcd erosion; 1).\. \\‘1 -

 

1 Extensive permafrost; “isconsin

 

 

l l l 1 j

30 25 20 15 10
Chronology (cal. ka)

Figure 5.20: Chronology of permafrost and loess deposition in north central Wisconsin and adjoining regions.
Dark bars represent timing of event; light bars represent conservative timing estimates for the same event.

1. Permafrost extends at least 90 km beyond the Late Wisconsin Moraine, based on ice-wedges found in the
Driftless Area of Wisconsin. (Clayton et al. 2001). 2. Accelerated erosion in the Driftless Area (Mason 1995;
Mason and Knox 1997). 3. Period of loess deposition based on OSL age estimates from loess in north central
Wisconsin (this study). 4. Spruce-Tundra paleoenvironment surrounds Wood Lake, Taylor County,
approximately 30 km north of the study area (Heide 1984).

County (Figure 5.20; Heide 1984). In addition, pine forest, which generally signals an
ameliorated climate, did not enter the region until ~12.9 ka (Wright 1968), and did not
likely replace spruce until approximately 11.5 ka in north central Wisconsin (Heide
1984). Landscapes that had previously been largely devoid of vegetation and had
experienced accelerated erosion would have stabilized as pine became the dominant
vegetation type. Without the sediment source, loess deposition would have ended as
well.

Thus, based on the spatial pattern of OSL age estimates (Figure 5.19) and particle
size depth trends (Figure 5.4-5.6; Table 5.4), it is reasonable to suggest that (l) loess in

north central Wisconsin was initially mainly derived from fine and very fine grained,

145

quartz and K—feldspar rich sediment to the west/ southwest (weathered fine grained
Cambrian sandstones and residuum), on westerly winds, starting around 15 ka, and (2)
siltier loess, rich in quartz and plagioclase with minor amounts of K-feldspar, was later
(closer to 12 ka) derived from the deflation of drained, ice-walled lake plains on and
behind the late Wisconsin terminal moraine, on winds that derived mainly from the
(northwest). Although the window of eolian activity is likely to have begun with loess
derived from the west and ended with loess derived from the northwest, long periods of

overlap almost certainly occurred during the heart of this period of eolian activity.

5.6 Summary and Implications

Loess is thickest (>70 cm) in northern Clark County, along the Late Wisconsin
terminal moraine, and in central Clark County, approximately coincident with the eastern
extent of Cambrian sandstone uplands. Away from this region, loess thickness decreases
dominantly towards both the southeast and to the east to < 40 em. On its eastern margins,
the loess is silt loam in texture, becoming increasingly coarser toward the northwest, and
especially coarser toward the west, where it grades into a fine sandy loam mantle, also
eolian in origin.

The spatial characteristics of loess thickness, particle size data, and feldspar
mineralogy indicate loess within the study area was likely derived from two sources (1)
one to the northwest and (2) one to the west/southwest (Figure 5.21). To the northwest,
numerous ice-walled lake plains exist in and behind the Late Wisconsin terminal
moraine, and could have been the source of exposed sediment once ice and permafrost

began to melt, allowing the lake to drain. These features are dominantly comprised of

146

.858 E23353£E8$§ 05 Eot EoEEom
Emu—woo mo 8:36 2: 8 26 38:— Es «Ba 33% 2: mo 5&9. :3ch 5:8 2: E mBoEam E: 3-2 no 3:2:8 532 2C. .mx 2 boamfiioamm c285 was»
2 APE—o 98 95 20:53 mo comonEoUV mean 8.2 05 9:898 .Efiv 8 2:95.: 2: union use so $33 8:: 3:55-92 MERE—m 6953. 9:59: E3855

23 2.: 55:5 02 ES mix—SEE ”mofimctoa 9:392. .8552: 2: v5.33 was E 283 83 cu__m>»-ou_ 38:: .3355: 2: 8 ESE—8m 08:8 noEEwiocc
a “momma £023 £5280 E: 3-2 2: mo :omusnEmE vim 229.3 332 2: 2.55 92.: 25. ”EM;— ooSOm “main—to: 2: Eofi 35:29 $2: 853w 5.5 .6
5:36 9 use 38:. Ex 5.5: 2: E Ecuacou 532 of. .3 NM— buuméxoamm 35:38 vozmomov 3 van Sauce 2 came—8-x E; 5.5% E so: .8563an
:mtnEaU EESMAEC .oZaE E0: Eon—£8 Co 32:53. uwus @0898 box: 38%.:ch 9.36:... .3525". E1 mNTmn ho cocanmE “Em 22:3 vowtx

2: £59? 9:: fish “:91— .mcozmbzoucg EcwE E0858 mono. Bias zoocm $02 Emcouig 25:8 £5: 2.30 3% 35 32:3,? bficEsm “_N.m obi:

 

 

 

 

 

147

offshore silts and clays, with a median MWPS of 33.6 mm (coarse silt); thus, they may
have contributed mainly silt-sized, finer-grained sediment to loess deposited southeast of
the moraine. Second, fine-grained and friable Cambrian sandstones interbedded with
shale, and their residuum, which occur at or near the surface west and southwest of the
loess sheet, are likely the source of the coarser eolian sediment which has a median
MW PS of 130.5 um (fine sand) that dominates the western and central parts of the loess
sheet. The spatial patterns of feldspar mineralogy further suggest that both sources
dominantly contributed quartz. However, in terms of feldspar mineralogy, the Cambrian
sandstones are largely devoid of plagioclase, and thus, contributed mostly K-feldspar
eastward toward the loess sheet; whereas the ice-walled lake plains contributed both
plagioclase and to a lesser extent, K-feldspar, to the northwestern and northern parts of
the loess sheet.

In central Clark County, loess thickness also sharply declines towards the west.
Unlike similar trends reported by F ehrenbacher et al. (1965) and Muhs and Bettis (2000),
this decrease is not likely due to storm events because (1) a river valley source was not
identified within the area of thick loess and (2) particle size data do not indicate a fining
trend towards the west. Particle size data suggest, instead, the sharp decrease in thickness
towards the west is more likely a transitional zone, where the fine and very fine-grained
source sediment (wind-winnowed) grades into a still sandy but increasingly eolian (wind-
dcposited) sediment (sensu Pye 1995).

Because the sources of loess in north central Wisconsin are not associated with
major river valleys carrying glacial outwash, they provide a unique opportunity to

understand the regional paleoenvironmental dynamics, as opposed to fluctuations in ice

148

sheet dynamics. At the time of the last glacial maximum, the two loess source areas
discussed in this research (as with most/all of north central Wisconsin) were likely
underlain by continuous permafrost (Clayton et a1. 2001). Sediment from these source
areas would not have been available for deflation until ice retreated from the glacial
margin allowing regional permafrost as well as ice blocks within the moraine to degrade.
OSL age estimates indicate the period of loess deposition occurred between
approximately 15.2 and 12 ka; thus, the age estimates may constrain the timing of
sediment availability/permafrost degradation. At the least, they point to a period of rapid
environmental change and eolian activity in the study area. The spatial distribution of
these OSL age estimates, in conjunction with particle size depth plots, suggest the
landscape to the west/southwest thawed and became unstable first (closer to 15.2 ka),
leading to the deposition of sandier loess to the east and northeast. Ice blocks within, and
permafrost underlying, the Late Wisconsin terminal moraine likely persisted longer.
Once the permafrost underlying the moraine and the ice blocks within the moraine
degraded, ice-walled lake plains would have drained. The freshly exposed lake beds,
consisting of fine-grained off shore sediment was likely entrained and deposited
downwind to the southeast.

The importance, therefore, of this research is twofold. First, it substantiates an
earlier hypothesis that loess in north central Wisconsin was derived from the friable, fine-
grained Cambrian Sandstones to the west (Hole 1942) as well as from the Late Wisconsin
terminal moraine. Secondly, this research illustrates the significance of studying thin
loess sheets, which may provide valuable insights into the regional paleoenvironmental

conditions during loess deposition. These insights can be particularly meaningful in

149

combination with obtaining OSL age estimates in order to constrain the timing of the

paleoenvironmental changes.

150

APPENDIX A

Soil Profile Descriptions

151

Site Name: Clark 1

Site Description

Sample bag abbrev: Clark OSLl Date: 6/1/2007

Location: Roadcut in Clark Ctv

Slope at pit (%): fl

Slope element: Shoulder

Landform type: Bedrock upland, dissected SS landscape

Aspect: Mb

Drainage class: Q

Parent Material (stratigraphy if appropriate): Sandy loess over sandier (mixed?) zone. over till
over $5 residuum

Vegetation at site: hayfield

Erosion: Perhaps slightly

Water table (em): _N_/_A_

Other notes: Not good for loes_s thickness measurement

Profile description

 

 

 

 

 

 

 

 

 

 

 

 

 

Hor Depth Moist Structure Consistance Boundary Coarse Other
Color Frag.%
cm Gr 82 Shp H20 Feel Dist To
Ap 0-18 lOYR M F, Gran M Vfr C S 4 gr
3/3 Vf
co
B 18-40 lOYR M F. Sbk M Vfr C S 6 gr
5/3 Vf
co
2EB 40-51 lOYR M F Sbk M Fr C S 6 gr
5/4
co
ZBt/E 51-70 7.5YR M F, Sbk M Fr C S 10 gr 80% Bt,Skeletans weak
5/4 M patchy
7.5YR co
6/3
3Bt/Eb 70-84 5YR M F, Sbk M Fr C S 15 gr 90%Bcolor, skeletons
4/6 M lOYR 7/1 on ped faces,
7.5YR co thick and continuous
5/3
38th 84- 5YR S M, Sbk M Firm C S 10 gr Thick, red paleosol
120 5/3 C
co
38th 120- lOYR S M, Sbk M Firm gr Sampled by auger
145 C
co
3Cg 145- lOYR S M, Sbk M Firm gr Less clay
169 6/1 C
co
4Cr/3Cg 169- gr Till stringersinto SS
179 crevasses
co
4Cr 179+ gr SS residuum
co

 

 

 

 

 

 

 

 

 

 

 

 

 

152

Site Description
Site Name: Clark 2 Sample bag abbrev: Clark 2 Date: 6/5/2007

Location: gm

Slope at pit (%): 1_-2:/g

Slope element: Summit/Shoulder (hard to tell since in a quarry)
Landform type: Upland

Aspect: m

Drainage class: Well drained

Parent Material (stratigraphy if appropriate): Loess over till over an outwash-like diamicton over
gruS'LPaleosol in outwash-like diamicton

Vegetation at site: Grassland

Erosion: Perhaps slightly

Water table (cm): MA

 

Profile description

 

 

 

 

 

 

 

 

 

 

 

 

Hor Depth Moist Structure Consistance Boundary Coarse Other
Color Frag.
(cm) Gr 82 Sh H20 Feel Dist To %
Ap 0-18 lOYR M F& Gran M Vfr C S 0 gr
3/2 m
co
E 18-27 lOYR M Platy Thin M Vfr C S 0 gr
5/4 &
med co
Btl 27-42 7.5Y M Platy Thin M Fr C S 0 gr
R4/4 &
Med co
Bt2 42-50 7.5Y M Sbk F&m M Firm C S 2 gr
R4/4
co
28t3 50-70 5YR M Sbk F&m M Firm C S 5 gr Sand-filled krotrovina
4/4 in till
co
2C 70- 5YR M Sbk M& M Vfi C S ’10 gr
130 4/6 C
co
38th 130- 7.5 M Sbk C M VFr C S 10 gr
150 YR
3/4 co
3C 150- 2.5Y M Grain Mass M Fr A S 6 gr
186 R4/6
co
4Cr 186+ 2.5Y M Mass gr Yellowwxrind
R 5/6 apparent in top of the
co horizon

 

 

 

 

 

 

 

 

 

 

 

 

 

153

Site Name: Clark 3

Site Description
Sample bag abbrev: Clark 3

Location: Soil Sample Site 1; Woodlot
Slope at pit (%): 2
Slope element: Summit
Landform type: Dissected upland in northern Clark County
Aspect: E
Drainage class: WD
Parent Material (stratigraphy if appropriate): Loess over till
Vegetation at site: Sugar maple, basswood
Erosion: None-virgin soil
Water table (cm): N/A

Profile description

Date: 6/8/2007

 

 

 

 

 

 

 

 

 

 

 

'Hor Depth Moist Structure Consistance Boundary Coarse Other
Color Frag.
cm Gr 82 Sh H20 Feel Dist To %
Oi 0-3 C S gr Leaves, sticks
co
A1 3-13 10YR3/2 M F, Gr M Vfr C S 5 gr
M
co
A2 13-29 lOYR4/3 W F Gr M Vfr C S 5 gr
co
E 29-36 10YR5/4 M M Pl M Vfr C S 0 gr
+
Th 0 co
Bt/El 36-56 E210YR M Ff, Sbk M Fr C S 0 gr
6/3 in
Bt:7.5YR 0 co
4/6
Bt/E2 56-91 E210YR St M Sbk M Fr- C S 0 gr 70%B; vertical streaks of
6/2 to firm white interiors and red
Bt:5YR C 0 co halos bwon matrix 7.5YR
4/6 5/4
ZBt 91- 7.5YR St M Sbk M Fr C S 2 gr
114 4/6 to
C 4 co
3C 114+ 7.5YR W M Sbk M Fr 15 gr Sample byauger
5/6
4 co

 

 

 

 

 

 

 

 

 

 

 

 

154

 

APPENDIX B

Particle Size Data for Loess/Mantle Samples

155

 

 

Ne w: QM 8.: 2o 3 N: 2.: 3: 3m :5 58:5 :82: 8.8 2:525 56 mm
to m: 5.2 we 3 3 2m 3 N: v.8 5.2 58. 28mm. 3.2 5858: 80 mm

2:523

852
No S. n: E 3 no 3: 2: w: v.8 m: 58:5 :88” 2:: 2:525 8U:
do 3 mm 3 Z. 2. m: 5.: n: 3: 3: 58:5 :85 :8 2:583 :58
no 4.: w: 3: 3 3 :8 no we or 5.: 581858: 2558 no: 2:528: :82
_.o 3 no 3 E a: a: 2: S: mom :5 58_ :53: in: 58582 83:

2:538

852
N: 2 ma 3 2: 8.: m: .3 3 8.2 8.: 58_ :82: :8 5.558: :8:
No 3.. 3 3 N: 3. w: 2.: :2 v.8 3: 58:5 :88. :8 58: 58.2
m: 3 EH 2 2 3. o: of 0.2 3:. S: 58:5 :85 8.8m 88m 562
N: E 3 5.: m: 3. 2: 3: :2 SM 2: 58:5 28”: :8 2:583 803
3 2 w.~ o: 8.: on o: 2.2 we now 8.2 58:5 £822 :8 2:523 562
S E 3 8.: so 3. 2: we. as 8.8m E: 58:5 £552 3». 2:528 562
S E 5.2 3. 3 on 3 w: 8.: 3.: :2 58:5 2892 5.2: 2:528 :5:
m: 3 T: T.“ 3 no 88 N: 3: EN 5: 58_ :82: 22 2:52am :62
to E 8.: 3 3 ox :8 E S n8 5.2 58_ 285.. ~52 2:52am :88
3 :w E. Q 3 3w 3 S 3: 8.: E: 58. :88 3: 2:528. 8:
do 3 3: 3 NS 8.8 8.: :8 I w: .2: 58_ 26% f: 55:53: 88:

3:8 2:»: be»,

no 3 S: m: 88 E 2.. 2 SW 2: :2: 58. 92.5 22 5:853: 88:
9o _.: :2 of a: 2.2 E. Z 3 S 3 85:58. 3;: 3.2 58582 8cm
a: 8.: 8.8 2: 3 to E 3. S. f: E 58.555 $58 is 2:528 50:
to 02 SN _8 S. 2“ 3. :8 8.: a: w: 58.555 2838 $2 2:528 56».
_.o 3 2 2o 8.: on 2: w: w: to... we 58:5 £8”: 28 .582 50:
a: o: w: 5.: S 2. 2: w: 22 8.; 8.2 58:5 :58”: 8.8 :82 8o.
08> 08 um: ammo om“: 0mm: own?» 0:8 gas can 055» 520 558.: 0:3 59.32 2:... 22:58

mm~§><du «Fry—(..N HZ:— Satan—i

 

156

 

 

_.o M3 N5 :2 3 3‘ _6 ~12 :2 03. 3: :52 :5 2.3.2 :5 598m :5 _m
We 2: o9 Wm: v.2 5.2 ~.m Wm 2m 5.: 2m :52 >955 0:: 93.3. mam: 8.5282 :50 om
ed a: N3 3.: a: 02 we. 5 o.~ vé ~.m :58 .852 m; 35 35: 8.5282 50 ow
_.o 0.. 2.. M3 M3 ed «:0 02 0.2 gm tom :52 :5 2:92 5.? 2:58.32 90 wv
_.o o.~ 5m 2 :2 W: N: ~42 :2 wdm of :52 :5 2.8.2 2.5 2:533 :50 3
mo ~.m TM 3 v2 ow 3: mi of 2:: N: :52 :5 2.55am 2% 2:535 .50 3
_.o o.~ 5m .1 Z .35 No .22 0.2 35 3: :52 :5 £63: 23 25502.5: 90 m:
to 3. ca o._ f ~.m :6 ed 03 new of :52 :5 2.6.3 was 53m 50 3
od od :2 Do do m: S: _.m_ :— 62 92 :52 25 t 592 odw 528m .50 3
..o 2 N._ Go Yo we .2: 5: 5.2 0.; mi :52 :5 £592 35 5mg 96 mv
od od o.o od ed 5.: n: 5.2 w: 52 mi :52 :5 2.2.“: :cm 53m 90 G
No Om o.~ Yo :3 ed _.: 5.2 3: mam N: :52 :5 2 H:3: 0:: :552 .50 cc
_.o o._ :2 Do We ma ad v.3 .3. am 3: :52 :5 2:3; :5: :55: 90 am
No N.N N.N o._ N2 05 02 I: We. 2%. :2 :52 :5 2252: 0% 528m 96 mm
o.o to o._ 0.: w: .5 to :2 0.2 mam Wow :52 :5 2692 0.? 53m 9.0 R
we a: 0.: N2 2 we :2 5.2 .62 35 mi :52 :5 259$ 35 2=>§5m 50 cm
ed od o.o o: o.o 35 n: 52 I; v.2 :2 :52 :5 2593 52 225825 .50 mm
No \..~ 5.: 2o We a: «a of of m. _ m of :52 =5 M2592 5.3 2:53am 96 mm
No 0.: m: 3 Ev 3. M: 2: .2: m3 :2 :52 :5 2592 New 5.5282 96 _m
No 05 ~23 ms: do. :2: :5 :5 m: 2: 5.2 :52 25:5 00.2 5.5282 96 om
we :5 :5 Q: 2.: 5m 2: 2: 5:. ES .2: :52 :5 2 ”:90: N.Nw 2:535 90 am
_.o 3 _.~ We 0o :5 of 5.2 _.S no: NS :52 :5 2 ”$2: 9? 2:585m 9.0 mm
_.o N2 52 Yo wd Em x.: 0.2 v.3 mdm :2 :52 :5 3602: Wow 2.3.523: 5:0 mm
o: I 5.: I I 35 NE 3: 0.2 Vow of :52 :5 t 6.2: m2: 2:585m :50 cm
ed to m._ be od 5% 5.: 5.2 02 Now 5.: :52 =5 2.5:: mm: 2.3525 96 mm
no 05 Em ad ad Wm 5.: M3; _.m_ to: of :52 :5 2 U :3: ma: 2558.5 :5 cm
omu> emu om: own—U own—..— omm>u own?» 050 0:92 05... can; 820 23.8... U23 5232 2:... a. 2:.an
mmmm><qu al.—.13: m2...— 85.5.3.5

 

157

 

 

:.: :.: :.: 2 m2 :.: :.: :.: :2 :.:: :.: 58:5 2:52 :.:: 5:8 8:

3 2 :.~ :.~ :.: :.: :.: :.: :.: :.:: n2 58:5 285: :.:: 5:3 8m

3 :2 E :.: :._ :.: :2 :.2 :2 :.:: N2 58:5 :35: :.:m 58m 8:

S .2 E :.: :.: :.: :.: :2 n2 :.:: :.: 58:5 22:2 :.:: 2:585: 8m

:.: S :2 :2 :.~ :.: n2 :.: :.2 :.:: :2 58:5 2:58 :.: 2:395: _2

S 2 :2 :.: :.: :.: 5.2 m2 :2 :.:: :.2 58:5 2:52 :.:: 2:825 :2

S :.: _.: :.: :.: :.: m2 :2 :.: 3: S; 58:5 2:3: :.:: :58: :2

~.: :.: N.N :2 E :.: N2 :2 m2 :.:: S; 58:5 28:: :.:: 2:55: ::

:.: :.: :.~ :2 :.: :.: :.: :.2 N2 :.:: :.2 58:5 22:2: N.N: .5522 E

S :2 :.: :.: N.N :.: :2 :2 :.2 :.:: :2 58:5 28:5: :.:: :82 ::

S 3 :.: :.: :.: :.: :.2 :.: :.2 :.:: :2 58_ .5 255mm 2.: 2:585: :8 n:

:.: I :.: :.: n: :.: :.: n2 n2 :.:: :.: 58_ :5 :35: :.:: 2:88: :8 8

:.: :.: :2 :.: :.: :.: :.: :.2 _.2 :.:: :.:: 58. .5 25:2 2: 2:585: :8 m:

S 2 :.N :.: :.: :.: :.: :2 5.2 :.: we 58_ :5 : H$2 :.: 2:533 8.: 8

:.: :.: :.: :2 .2 :.: :.: :.2 :.2 :.:: 5.2 58. :5 28:: :.:: :82 :8 8

S :.N :.: .2 :.: :.: :.: 5.2 :2 :.:: :.:: 58. :5 2:5: :.:: 2:88: :5 m:

S 2 :: :.: :.: :.: 8.2 :.: 2.2 :.:: :2 58_ :5 :8”: 3: :52 :3 _:

N: :.: :.N N: :.: :.: :.: :.2 :.: :.:: :2 58_ :5 26:2: :.8 :58: 8,: ::

:.: I Z :2 :.: 3 :2 N2 5.2 S: :2 58. :5 2:2: :.:: :82 :5 :m

:.: :.: :.: :.: :.: :.: :.:: :.2 :.2 :.: m2 58_ :5 2 :2: :.:: _::22 :5 w:

3 :N :.: :.: :.: N: :2 N2 m2 3: :.2 58_ 5 28:H _m :.:: 2:58am 56 S

N: 2 :.~ 2 :.: :.: N: :.: N2 :.:: :.:: 58_ :5 25:2 :8 :52): :5 :m

S S :.: :.: :.: :.: :.: :.2 v2 :.:: n2 58_ .5 2822 E: :58: :8 mm

:.: :.: :.: :.: :.: :.: :.2 :.2 5.2 :.:: :.2 58_ 25 22:8: :.:: :52 :5 3.

N: N.N Z :.: :.: 3 :2 :.: N2 :.:: :.: 58. :5 3:52 :.:: 58m :8 mm

3 :.~ :.: :._ :.: :.: :.: :.2 n2 :.:: m2 58_ .5 2 “:52 :.:: 58: 5G N:
08> 0:0 um: 8:20 0:: 09:6 8:2?» :50 952 0:9. 52> :20 8385 92:": 5:23: 2:5 22:55:

2222550 5:523. 22: 2.5.8::

 

158

5363 “m2 52-3.— Hwho ”mu—-mm: “WE 6.2-: Hmn_>> mmbém Hmm>> uom.mm

.oooméoo. HmU> ”coo—dew “mu

:mU www-mm :mE ”3-9 :mu ”NTN 2m“; .E: E 80:88.: 8:: 20:22:: 0
2:: 2:29 B :2: 22:8 BEE?» :85 “ma—3.2 n

5: :2: 2:: a.

 

 

 

_é N.N in «A _.— _.m «.3 :.v— e.m_ 33 :5 58. 2:: m—"mencN :.:—82
N... :6 5m QN 2N ad ad w.~_ ad— 23 Q:— Eae. 2:: m- "amuen :82
Yo 2: m4: m2 Z 06 md. o.3 :2 $8 93 882:: 382mm 9: 2:8”: _mm
No Wm m.m o: :2 ..c Q: 3: NE QR :.f 882:: 262mm m.wm Ewcm cmm
_.o o; :6 ON :.m to :.:: 52 :3 ..wm 2: 882:: Sumouvm osm Swvm 2%
_.o _._ _.N Z a: 2m 9: 0.3 _.o_ wdm v.2 E82:W 2629 Wm: Ewcm cum
mU> omU om: own—U own—:.— umm>U omm>> :50 ”mm—2 :5”— _mr.—> mag—U :.::on Dumnm awn—>92 22,—. A: 2956
mmmh><do :FSZm Hz: 2.2.3.5

 

159

REFERENCES

160

Aleinikoff, J. N., D. R. Muhs, and C. M. Farming. 1998. Isotopic evidence for the sources
of late Wisconsin (Peoria) loess, Colorado and Nebraska: Implications for
paleoelimate. In Dust Aerosols, Loess Soils, and Global Change. Ed. A. Busacca.
College of Agriculture and Home Economics Miscellaneous Publication
MISC0190. Washington State University, Pullman. 124-127.

Aleinikoff, J. N., D. R. Muhs, R. R. Sauer, and C. M. Fanning. 1999. Late Quaternary
loess in northeastern Colorado: Part II—Pb isotopic evidence for the variability of
loess sources. Geological Society of America Bulletin 111:1876-1883.

Arbogast, A. F., and W. C. Johnson. 1998. Late-Quaternary Landscape Response to
Environmental Change in South-Central Kansas. Annals of the Association of
American Geographers 88 (l):126-l45.

Arriaga, F. J., and B. Lowery. 2006. A fast method for determining soil particle size
distribution using a laser instrument. Soil Science 171 (9):663-74.

Asthana, V. 1969. The Mt. Simon Formation (Dresbachian stage) of Wisconsin. [Ph.D.
thesis], University of Wisconsin, Madison.

Attig, J. W. 1993. Pleistocene geology of Taylor County, Wisconsin. Wisconsin
Geological and Natural History Survey.

 

. 2008. Personal communication.

Attig, J. W., and L. Clayton. 1993. Stratigraphy and origin of an area of hummocky
glacial topography, northern Wisconsin, USA. Quaternary International 18:61-
67.

Attig, J. W., L. Clayton, and D. M. Mickelson. 1985. Correlation of Late Wisconsin
Glacial Phases in the Western Great-Lakes Area. Geological Society of America
Bulletin 96 (12)21585-1593.

Attig, J. W., N. R. Ham, and D. M. Mickelson. 1998. Environments and processes along
the margin of the Laurentide Ice Sheet in north-central Wisconsin. Madison:
Wisconsin Geological and Natural History Survey.

Attig, J. W., and M. A. Muldoon. 1989. Pleistocene Geology of Marathon County,
Wisconsin: University of Wisconsin-Extension, Geological and Natural History
Survey.

Baker, R. G., R. S. Rhodes, D. P. Sehwert, A. C. Ashworth, T. J. Frest, G. R. Hallberg,
and J. A. Janssens. 1986. A fiill-glacial biota from southeastern Iowa, USA.
Journal of Quaternary Science 1291-107.

Baker, R. W., J. F. Diehl, T. W. Simpson, L. W. Zelazny, and S. Beske-Diehl. 1983. Pre-
Wisconsinan glacial stratigraphy, chronology, and paleomagnetics of west-central
Wisconsin. Bulletin of the Geological Society of America 94 (12): 1442-1449.

161

Bartelme, R. J ., and H. Strelow. 1977. Soil Survey of Wood County, Wisconsin: US
Department of Agriculture, Soil Conservation Service.

Bettis, E. A., D. R. Muhs, H. M. Roberts, and A. G. Wintle. 2003. Last Glacial loess in
the conterrninous USA. Quaternary Science Reviews 22 (18-19): 1907-1946.

Beuselinck, L., G. Govers, J. Poesen, G. Degraer, and L. Froyen. 1998. Grain-size
analysis by laser diffractometry: comparison with the sieve-pipette method.
CA TENA 32 (3-4):193-208.

Birks, H. J. B. 1976. Late-Wisconsinan Vegetational History at Wolf Creek, Central
Minnesota. Ecological Monographs 46 (4):395-429.

Black, R. F. 1965. Ice-wedge casts of Wisconsin. Wisconsin Academy of Science, Arts
and Letters 54:187-222.

. 1976. Periglacial Features Indicative of Permafrost - Ice and Soil Wedges.
Quaternary Research 6 (1)23-26.

 

Boelter, J. M. 2005. Soil survey of Taylor County, Wisconsin: US Department of
Agriculture, Soil Conservation Service.

Brown, B. A., and D. L. Patterson. 1988. Bedrock geology of Wisconsin west-central
sheet. Madison, Wisconsin: Geological and Natural History Survey.

Cahow, A. C. 1976. Glacial geomorphology of the southwest segment of the Chippewa
lobe moraine complex, Wisconsin. [Ph.D. thesis], Geography, Michigan State
University, East Lansing.

Carter, L. D., J. A. Heginbottom, and M. K. Woo. 1987. Arctic lowlands. In Geomorphic
Systems of North America ed. W. L. Graf, 583-628. Boulder, CO: Geological
Society of America

C att, J. A. 1977. Loess and coversands. In British Quaternary studies: recent advances,
ed. F. W. Shotton, 221—229.

Clayton, L. 1991. Pleistocene geology of Wood County, Wisconsin: University of
Wisconsin-Extension, Geological and Natural History Survey.

Clayton, L., and J. W. Attig. 1987. Lake-ice collapse trenches in Wisconsin, US A. Earth
Surface Processes and Landforms 12:167-172.

 

. 1989. Glacial Lake Wisconsin: Geological Society of America, Boulder, Colo.

Clayton, L., J. W. Attig, N. R. Ham, M. D. Johnson, C. E. Jennings, and K. M. Syverson.
2008. Ice-walled-lake plains: Implications for the origin of hummocky glacial
topography in middle North America. Geomorphology 97 (1-2):237-248.

162

Clayton, L., J. W. Attig, and D. M. Mickelson. 2001. Effects of late Pleistocene
permafrost on the landscape of Wisconsin, USA. Boreas 30 (3):l73 - 188-173 -
188.

Clayton, L., J. W. Attig, D. M. Mickelson, M. D. Johnson, and K. M. Syverson. 2006.
Glaciation of Wisconsin. Educational Series 36.

Clayton, L., and S. R. Moran. 1982. Chronology of Late Wisconsinan glaciation in
middle North America. Quaternary Science Reviews 1:55—82.

C OHMAP Members. 1988. Climatic Changes of the Last 18,000 Years: Observations
and Model Simultations. Science 241 (4869):]043.

Colgan, P. M. 1999. Reconstruction of the Green Bay Lobe, Wisconsin, United States,
from 26,000 to 13,000 radiocarbon years BP. Geological Society of America
Special Paper 337: 137-1 50.

Crouvi, 0., R. Amit, Y. Enzel, N. Porat, and A. Sandler. 2008. Sand dunes as a major
proximal dust source for late Pleistocene loess in the Negev Desert, Israel.
Quaternary Research 70 (2):275-282.

Distefano, M. 1973. The mineralogy and petrology of the Eau Claire Formation, west-
central Wisconsin. [M.A. thesis], Department of Geology, Northern Illinois
University, Dekalb.

Duller, G. A. T. 1996. Recent developments in luminescence dating of Quaternary
sediments. Progress in Physical Geography 20 (2)2127-145.

Emrich, G. H. 1966. Ironton and Galesville (Cambrian) Sandstones in Illinois and
Adjacent Areas. Illinois State Geological Survey Circular 403155.

Fairbanks, R. G., R. A. Mortlock, T.-C. Chiu, L. Cao, A. Kaplan, T. P. Guilderson, T. W.
Fairbanks, A. L. Bloom, P. M. Grootes, and M.-J. Nadeau. 2005. Radiocarbon
calibration curve spanning 0 to 50,000 years BP based on paired
230Th/234U/23 8U and 14C dates on pristine corals. Quaternary Science Reviews
24 (16-17):l781-1796.

Fehrenbacher, J. B., J. L. White, H. P. Ulrich, and R. T. Odell. 1965. Loess Distribution
in Southeastern Illinois and Southwestern Indiana. Soil Science Society of
America Journal 29 (5):566-572.

F iala, W. D. 1989. Soil survey of Marathon County, Wisconsin: US Department of
Agriculture, Soil Conservation Service.

F ollmcr, L. R. 1996. Loess studies in central United States: Evolution of concepts.
Engineering Geology 45 (l-4):287-304.

163

Forman, S. L., E. A. Bettis, T. J. Kemmis, and B. B. Miller. 1992. Chronologic evidence
for multiple periods of loess deposition during the Late Pleistocene in the
Missouri and Mississippi River Valley, United States: implications for the activity
of the Laurentide Ice Sheet. Palaeogeography, palaeoclimatologv, palaeoecology
93 (1-2):71-83.

Forman, S. L., and J. Pierson. 2002. Late Pleistocene luminescence chronology of loess
deposition in the Missouri and Mississippi river valleys, United States.
Palaeogeography Palaeoclimatology Palaeoecology 186 (1-2):25—46.

Foss, J. E., and R. H. Rust. 1962. Soil Development in Relation to Loessial Deposition in
Southeastern Minnesota. Soil Science Society of America Journal 26 (3):270-274.

Franzmeier, D. P. 1970. Particle Size Sorting of Proglaeial Eolian Materials. Soil Science
Society of America Journal 34 (6):920.

Frazee, C. J ., J. B. Fehrenbacher, and W. C. Krumbein. 1970. Loess Distribution from a
Source. Soil Science Society of America Journal 34 (2):296-301.

French, H. M. 1996. The Periglacial Environment: Harlow: Longrnan.

Frye, J. C., H. D. Glass, and H. B. Willman. 1962. Stratigraphy and mineralogy of the
Wisconsinan loesses of Illinois. Illinois State Geological Survey Circular 334:55.

Grimley, D. A. 1996. Stratigraphy, magnetic susceptibility, and mineralogy of loess -
paleosol sequences in southwestern Illinois and eastern Missouri [Ph.D. thesis],
University of Illinois, Urbana-Champaign.

. 2000. Glacial and nonglaeial sediment contributions to Wisconsin Episode loess
in the central United States. Geological Society of America Bulletin 112
(10): 1475-1495.

 

Grimley, D. A., L. R. Follmer, R. E. Hughes, and P. A. Solheid. 2003. Modern,
Sangamon and Yarmouth soil development in loess of unglaciated southwestern
Illinois. Quaternary Science Reviews 22 (2-4):225-244.

Grimley, D. A., L. R. Follmer, and E. D. McKay. 1998. Magnetic Susceptibility and
Mineral Zonations Controlled by Provenance in Loess along the Illinois and
Central Mississippi River Valleys. Quaternary Research 49 (1)224-36.

Ham, N. R., and J. W. Attig. 1997. Pleistocene geology of Lincoln County, Wisconsin.
Wisconsin Geological and Natural History Survey Information circular.

Hamilton, T. D., T. A. Ager, and S. W. Robinson. 1983. Late Holocene Ice-wedges near
Fairbanks, Alaska, U.S.A.: Environmental Setting and History of Growth. Arctic
and Alpine Research 15 (2)2157-168.

164

Heide, K. 1984. Holocene Pollen Stratigraphy from a Lake and Small Hollow in North-
Central Wisconsin, USA. Palynology 8:3-19.

Hole, F. D. 1942. Correlation of the glacial Border drift of north central Wisconsin.
[Ph.D. thesis], University of Wisconsin, Madison.

 

. 1943. Correlation of the glacial Border drift of north central Wisconsin.
American Journal of Science 241 (8):498-516.

 

. 1968. Aeolian silt and sand deposits of Wisconsin. Madison: Wisconsin
Geological and Natural History Survey.

Hole, F. D., M. T. Beatty, and G. B. Lee. 1976. Soils of Wisconsin. Madison, Wis.:
Published for the Geological and Natural History Survey, University of
Wisconsin-Extension [by the] University of Wisconsin Press.

Jacobs, P. M., and J. C. Knox. 1994. Provenance and pedology of a long-term Pleistocene
depositional sequence in Wisconsin's Driftless Area. CA T ENA 22 (1):49—68.

Jacobs, P. M., J. C. Knox, and J. A. Mason. 1997. Preservation and Recognition of
Middle and Early Pleistocene Loess in the Driftless Area, Wisconsin. Quaternary
Research 47 (2): 147-154.

Jain, M., L. Better-Jensen, and A. K. Singhvi. 2003. Dose evaluation using multiple-
aliquot quartz OSL: test of methods and a new protocol for improved accuracy
and precision. Radiation Measurements 37 (l):67—80.

Jakel, D. E., and R. A. Dahl. 1989. Soil survey of Chippewa County, Wisconsin: US
Department of Agriculture, Soil Conservation Service.

Jenny, H. 1941. Factors of soil formation; a system of quantitative pedology. lst ed. New
York, London,: McGraw-Hill.

Johnson, M. D. 1986. Pleistocene geology of Barron County, Wisconsin. Wisconsin
Geological and Natural History Survey Information Circular 55:42.

. 2000. Pleistocene Geology of Polk County, Wisconsin. Wisconsin Geological
and Natural History Survey Information circular 92.

 

Johnson, W. H. 1990. Ice-Wedge Casts and Relict Patterned-Ground in Central Illinois
and Their Environmental Significance. Quaternary Research 33 (l):51-72.

Johnson, W. H., and L. R. Follmer. 1989. Source and origin of Roxana Silt and Middle
Wisconsinan Midcontinent glacial activity. Quaternary Research 31 :3 19—33 1.

Konert, M., and J. E. F. Vandenberghe. 1997. Comparison of laser grain size analysis
with pipette and sieve analysis: a solution for the underestimation of the clay
fraction. Sedimentology 44 (3):523-535.

165

KunleDare, M. A. 2005. Petrographic investigation into the development of secondary
porosity in sandstones: A case study of the Cambrian Mount Simon and
Galesville Sandstones, Illinois Basin. [Ph.D. thesis], University of Missouri,
Rolla.

Lachenbruch, A. H. 1962. Mechanics of thermal contraction cracks and ice-wedge
polygons in permafrost. Geological Society of America Special Paper 70:69.

Langton, J. E., and D. T. Simonson. 1998. Soil survey of Jackson County, Wisconsin: US
Department of Agriculture, Soil Conservation Service.

Lca, P. D. 1990. Pleistocene periglacial eolian deposits in southwestern Alaska;
sedimentary facies and depositional processes. JOURNAL OF SEDIMENTARY

RESEARCH 60 (4)2582-591.

Leigh, D. S. 1994. Roxana Silt of the Upper Mississippi Valley; lithology, source, and
paleoenvironment. Bulletin of the Geological Society of America 106 (3):430-442.

Leigh, D. S., and J. C. Knox. 1993. AMS Radiocarbon Age of the Upper Mississippi
Valley Roxana Silt. Quaternary Research 39 (3):282-289.

 

. 1994. Loess of the Upper Mississippi Valley Driftless Area. Quaternary
Research 42 (1 )230-40.

Lusch, D. P., K. E. Stanley, R.J. Schaetzl, A. Kendall, R. L. Van Dam, A. Nielsen, B. E.
Blumer, T. C. Hobbs, J. L. F. Holmstadt, and C. L. May. in press.
Characterization and mapping of patterned ground in the Saginaw Lowlands,
Michigan: Possible evidence for Late-Wisconsin permafrost. Annals of the
Association of American Geographers.

Maat, P. B., and W. C. Johnson. 1996. Therrnolumineseence and new 14C age estimates
for late quaternary loesses in southwestern Nebraska. Geomorphology 17 (1-
3):l 15-128.

Mackay, J. R., and C. R. Burn. 2002. The first 20 years (1978-1979 to 1998-1999) of ice-
wedge growth at the Illisarvik experimental drained lake site, western Arctic
coast, Canada. Canadian Journal of Earth Sciences 39 (1)295.

Maher, L. ed. 1982. The palynology of Devil ’5 Lake, Sauk County, Wisconsin. Madison,
Wisconsin: Wisconsin geological and Natural History Survey.

Markewieh, H. W., D. A. Wysocki, M. J. Pavich, E. M. Rutledge, H. T. Millard, F. J.
Rich, P. B. Maat, M. Rubin, and J. P. MeGeehin. 1998. Paleopedology plus TL,
Be-lO, and C-14 dating as tools in stratigraphic and paleoelimatic investigations,
Mississippi River Valley, USA. Quaternary International 51-22143-167.

Mason, J. A. 1995. Effects of glacial-interglacial climate change on mass wasting,
southeastern Minnesota. [Ph.D. thesis], University of Wisconsin, Madison.

166

 

. 2001. Transport Direction of Peoria Loess in Nebraska and Implications for
Loess Sources on the Central Great Plains. Quaternary Research 56 (1):79-86.

Mason, J. A., P. M. Jacobs, R. S. B. Greene, and W. D. Nettleton. 2003. Sedimentary
aggregates in the Peoria Loess of Nebraska, USA. CA TENA 53 (4):377-397.

Mason, J. A., R. M. Joeckel, and E. A. Bettis. 2007. Middle to Late Pleistocene loess
record in eastern Nebraska, USA, and implications for the unique nature of
Oxygen Isotope Stage 2. Quaternary Science Reviews 26 (5-6):773-792.

Mason, J. A., and J. C. Knox. 1997. Age of colluvium indicates accelerated late
Wisconsinan hillslope erosion in the Upper Mississippi Valley. Geologi 25
(3):267-270.

Mason, J. A., E. A. Nater, and H. C. Hobbs. 1994. Transport Direction of Wisconsinan
Loess in Southeastern Minnesota. Quaternary Research 41 (1):44-51.

Mason, J. A., E. A. Nater, C. W. Zanner, and J. C. Bell. 1999. A new model of
topographic effects on the distribution of loess. Geomorphology 28 (3-4):223-236.

McDonald, E. V., and A. J. Busacca. 1992. Late Quaternary stratigraphy of loess in the
Channeled Scabland and Palouse regions of Washington state. Quaternary
Research 38 (2):]41-156.

Mickelson, D. M., L. Clayton, D. S. Fullerton, and H. W. Borns Jr. 1983. The late
Wisconsin glacial record of the Laurentide ice sheet in the United States. Late
Quaternary environments of the United States 1:3—37.

Milne, G. 1936. Normal erosion as a factor in soil profile development. Nature 1382548-
549.

Mitchell, M. J. 1996. Soil Survey of Lincoln County, Wisconsin: US Department of
Agriculture, Soil Conservation Service.

Mode, W. N. 1976. The glacial geology of a portion of north—central Wisconsin. [M.S.
thesis], University of Wisconsin, Madison.

Morrison, B. C. 1968. Stratigraphy of the Eau Claire formation of west-central
Wisconsin. [M.A. thesis], University of Wisconsin, Madison.

Mudrey, M. G. 1987. Bedrock geology of Wisconsin northwest sheet. Madison,
Wisconsin: Geological and Natural History Survey.

Mudrey, M. G., B. A. Brown, and J. K. Greenberg. 2007. Bedrock geologic map of
Wisconsin. Madison: University of Wisconsin-Extension, Geological and Natural
History Survey,.

167

Muhs, D. R., J. N. Aleinikoff, T. W. Stafford Jr, R. Kihl, J. Been, S. A. Mahan, and S.
Cowherd. 1999. Late Quaternary loess in northeastern Colorado: Part I—Age and
paleoelimatic significance. Geological Society of America Bulletin 11121861-
1875.

Muhs, D. R., and E. A. Bettis. 2000. Geochemical variations in Peoria Loess of western
Iowa indicate paleowinds of midcontinental North America during last glaciation.
Quaternary Research 53 (1)249-61.

Muhs, D. R., J. P. McGeehin, J. Beann, and E. Fisher. 2004. Holocene loess deposition
and soil formation as competing processes, Matanuska Valley, southern Alaska.
Quaternary Research 61 (3):265-276.

National Climatic Data Center. 2002. Monthly station normals of temperature,
precipitation, and heating and cooling degree days 1971-2000: National Oceanic
and Atmospheric Administration.

Obruchev, V. A. 1945. Loess types and their origin. American Journal of Science 243
(5):256-262.

Odom, l. E. 1975. F eldspar—grain size relations in Cambrian arenites, upper Mississippi
Valley. JOURNAL OF SEDIMENTARY RESEARCH 45 (3):636-650.

Olson, C. G., and R. V. Ruhe. 1979. Loess Dispersion Model, Southwest Indiana, U.S.A.
Acta Geologica Academiae Scientiarum Hungaricae, T omus 22 (l-4):205-227.

Ostrom, M. E. 1966. Cambrian stratigraphy of western Wisconsin. Madison,: University
of Wisconsin Geological and Natural History Survey.

. 1967. Paleozoic stratigraphic nomenclature for Wisconsin. [Madison,:
University of Wisconsin Geological and Natural History Survey.

 

 

. 1970. Lithologic cycles in lower Paleozoic rocks of western Wisconsin.
Madison,: Available from the Geological and Natural History Survey, University
of Wisconsin.

Pecsi, M. 1990. Loess is not just the accumulation of dust. Quaternary International 7
(8): 1-21.

Péwé, T. L. 1962. Ice-wedges in permafrost, Lower Yukon River area near Galena,
Alaska. Biuletyn Peryglacjalny 11:65-76.

 

. 1966. Paleoclimatic Significance of Fossil Ice-wedges. Biuletyn Peryglacjalny
15:65-73.

. 1984. Present and past permafrost in the temperate United States. Me 'langes
offerts a ' Andre' Journaux2255-66.

 

168

 

ed. 1983. The periglacial environment in North America during Wisconsin time.

Prescott, J. R., and J. T. Hutton. 1994. Cosmic ray contributions to dose rates for
luminescence and ESR dating: large depths and long-term time variations.
Radiation Measurements 23 (2-3):497-500.

Price, L. W. 1972. The Periglacial Environment, Permafrost, and Man.

Putman, B. R., 1. J. Jansen, and L. R. Follmer. 1988. Loessial Soils: Their relationship to
width of the source valley in Illinois. Soil Science 146 (4):241-247.

Pye, K. 1984. Loess. Progress in Physical Geography 8 (2)1176.

 

. 1995. The nature, origin and accumulation of loess. Quaternary Science Reviews
14 (7-8):653-667.

Raasch, G. O., and L. Unfer, Jr. 1964. Transgressive-regressive cycle in Croixan
sediments (Upper Cambrian). Kansas Geological Survey Bulletin 169:427—440.

Ruhe, R. V. 1954. Relations of the Properties of Wisconsin Loess to Topography in
Western Iowa. American Journal of Science 252 ( l 1):663-672.

. 1983. Depositional environment of late Wisconsin loess in the midcontinental
United States. Late-Quaternary Environments of the United States 1:130—137.

 

Ruhe, R. V., and C. G. Olson. 1980. Clay-Mineral Indicators of Glacial and Non-Glacial
Sources of Wisconsinan Loesses in Southern Indiana, USA. Geoderma 24
(4):283-297.

Rutledge, E. M., N. Holowaychuk, G. F. Hall, and L. P. Wilding. 1975. Loess in Ohio in
Relation to Several Possible Source Areas: 1. Physical and Chemical Properties.
Soil Science Society of A merica Journal 39 (6): 1 125-1 132.

Schaetzl, R. 2007. Personal communication.

Schaetzl, R. J. in press. The distribution of silty soils in the Grayling Fingers region of
Michigan: Evidence for loess deposition onto frozen ground. Geomorphology In
Press, Corrected Proof. [doiz10.1016/j.geomorph.2008.03.012]

Schaetzl, R. J ., and S. Anderson. 2005. Soils .' genesis and geomorphology. New York:
Cambridge University Press.

Schaetzl, R. J ., S. F. Burns, T. W. Small, and D. L. Johnson. 1990. Tree uprooting:
review of types and patterns of soil disturbance. Physical Geography 11 (3)2277—
291.

Schaetzl, R., and J. Hook. 2008. Evidence for Loess in Northwest Lower Michigan:
Characterizing the Silty Mantle on the Buckley Flats Outwash Plain. Physical
Geography 29 (2): 140- l 57.

169

Schaetzl, R. J., and W. L. Loope. 2008. Evidence for an eolian origin for the silt-enriched
soil mantles on the glaciated uplands of eastern Upper Michigan, USA.
Geomorphology 100 (3-4):285-295.

Schoeneberger, P. J., D. A. Wysocki, E. C. Benham, and W. D. Broderson. 2002. Field
book for describing and sampling soils, Version 2.0. Natural Resources
Conservation Service, National Soil Survey Center, Lincoln, NE.

Shur, Y. L., and M. T. Jorgenson. 2007. Patterns of permafrost formation and degradation
in relation to climate and ecosystems. Permafrost and Periglacial Processes 18
(1):7-19.

Simonson, D. T., and H. E. Lorenz. 2002. Soil survey of Clark County, Wisconsin: US
Department of Agriculture, Soil Conservation Service.

Smalley, I. J. 1966. The properties of glacial loess and the formation of loess deposits.
Journal of Sedimentary Research 36 (3):669-676.

 

. 1972. The interaction of great rivers and large deposits of primary loess.
Transactions, New York Academy of Sciences 34:534-542.

 

. 1975. Loess: Lithology and Genesis: Dowden Hutchinson and Ross.

Smith, G. D. 1942. Illinois Loess: Variations in Its Properties and Distribution, a
Pedologic Interpretation: University of Illinois, Agricultural Experiment Station.

Snowden, J. O., and R. R. Priddy. 1968. Geology of Mississippi loess. Mississippi
Geological Survey Bulletin 1 l 1:13—203.

Sommer, M., and E. Schlichting. 1997. Archetypes of Catenas in respect to matter - A
concept for structuring and grouping Catenas. Geoderma 76 (1-2):1-33.

Sperazza, M., J. N. Moore, and M. S. Hendrix. 2004. High-Resolution Particle Size
Analysis of Naturally Occurring Very F ine-Grained Sediment Through Laser
Diffractometry. Journal of Sedimentary Research 74 (5):736-743.

Staff, S. S. L. 1996. Soil survey laboratory methods manual. Soil Survey Investigations
Report 42.

Stewart, M. T. 1973. Pre-Woodfordian drifts of north-central Wisconsin. [M.S. thesis],
University of Wisconsin, Madison.

Stewart, M. T., and D. M. Mickelson. 1976. Clay Mineralogy and Relative Age of Tills
in North-Central Wisconsin. Journal of Sedimentary Petrology 46 (1):200-205.

Sutherland, A. W. 1989. Soil development and clay mineral composition of pre-late

Wisconsin till units in northeastern Clark and southeastern Taylor counties,
Wisconsin. [M.S. thesis], University of Wisconsin, Madison.

170

Syverson, K. M. 2007. Pleistocene Geology of Chippewa County, Wisconsin: University
of Wisconsin-Extension, Geological and Natural History Survey.

Syverson, K. M., and P. M. Colgan. 2004. The Quaternary of Wisconsin: a review of
stratigraphy and glaciation. Quaternary Glaciations: Extent and Chronology.

Thomas, D. D. 1977. Soil survey of Eau Claire County, Wisconsin: US Department of
Agriculture, Soil Conservation Service.

Tsoar, H., and K. Pye. 1987. Dust Transport and the Question of Desert Loess Formation.
Sedimentology 34 (1)2139-153.

Wang, H., J. A. Mason, and W. L. Balsam. 2006. The importance of both geological and
pedological processes in control of grain size and sedimentation rates in Peoria
Loess. Geoderma 136 (1-2):388-400.

Washbum, A. L. 1980. Geocryology. New York: Wiley.

Wayne, W., and R. S. Guthrie. 1993. Permafrost and periglacial winds around the
Wisconsinan ice margin in the northern plains of the USA. Permafrost,
Proceedings Sixth International Conference 1:694-699.

Webb, T., III. 1981. The Past 1 1,000 Years of Vegetational Change in Eastern North
America. BioScience 31 (7):501-506.

Weidman, S. 1907. The geology of north central Wisconsin. Madison, Wis.: The State.

 

. 1913. The Pleistocene Succession in Wisconsin. Science 37 (951):456-457.

West, L. T., E. M. Rutledge, and D. M. Barber. 1980. Sources and Properties of Loess
Deposits on Crowley's Ridge in Arkansas. Soil Science Society of America
Journal 44 (2):353-358.

Wilgus, W. L. 1933. Heavy minerals of the Dresbach Sandstone of western Wisconsin:
J our. Sedimentary Petrology 3 : 83-9 1 .

Willman, H. B., and J. C. Frye. 1970. Pleistocene Stratigraphy of Illinois: Illinois State
Geological Survey Bulletin.

Wright, H. E., Jr. 1968. The Roles of Pine and Spruce in the Forest History of Minnesota
and Adjacent Areas. Ecology 49 (5):937-955.

Wright, H. E., T. C. Winter, and H. L. Patten. 1963. 2 Pollen Diagrams from
Southeastern Minnesota - Problems in the Regional Late-Glacial and Postglacial
Vegetational History. Geological Society of America Bulletin 74 (11):]371-&.

171

   

mlmlsllllllll lllll ll llllllllllllllllllllllllllles
3 1293 03062 5093