COMPARISON OF SAND DUNE CHRONOLOGIES IN THE GREAT PLAINS
AND EASTERN LAKE MICHIGAN COASTAL ZONE
By
Daniel Michael Kowalski

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
Geography – Master of Science
2014

ABSTRACT
COMPARISON OF SAND DUNE CHRONOLOGIES IN THE GREAT PLAINS
AND EASTERN LAKE MICHIGAN COASTAL ZONE
By
Daniel Michael Kowalski

Extensive deposits of eolian sand occur throughout the Great Plains
region as well as along the eastern coastal zone of Lake Michigan. Numerous
studies have been conducted on dunes in the Great Plains and along the Lake
Michigan coast. Recent research suggests that dunes in both regions were active
contemporaneously during the Medieval Warm Period (MWP). This finding is
interesting because it suggests that broad regional climate patterns may have
influenced dunes in both systems.
Given the apparent synchroneity in dune systems within the Great Plains
and Great Lakes regions during the MWP, this research further compares the
chronology of sand dune evolution in both regions during the Holocene. To test
this relationship, published literature from both regions was reviewed and all
published radiocarbon and luminescence ages reported were logged, including
348 ages from the Great Plains and 246 ages from the Great Lakes region. Ages
were used to construct probability density distributions, inform Principal
Components Analysis (PCA), and construct time-slice maps to compare and
contrast dune evolution events over the past 7000 years. Based upon
interpretation of the results of this study, similar dune activation events have
likely been taking place in both regions from ~4400 years ago to the present.

ACKNOWLEDGEMENTS

Many have contributed over the past few years to the completion of this
thesis. First and foremost, I want to thank the MSU Department of Geography for
allowing me to earn the degree. Your various financial contributions (including a
teaching assistantship) made this whole thing possible, and for that I will be
forever grateful.
I also want to thank my advisor, Dr. Alan Arbogast. Al, your advice and
support throughout this whole process were invaluable. Our “off the record”
conversations about both school and life were appreciated more than you will
ever know. It’s a helluva hole, my friend.
A big thanks to my other two committee members, Dr. Randall Schaetzl
and Dr. Ashton Shortridge. Your feedback throughout this entire process was
incredibly helpful, and I couldn’t have done this without you. I also want to thank
Dr. Bruce Pigozzi for his help and encouragement during the writing process.
You made a statistician out of me, which I never thought was possible! I would
also like to thank Michael Michalek and Phillippe Wernette for their assistance
with ArcGIS and Adobe Illustrator, respectively. You both provided help when I
needed it most, and never complained. This never would have been completed
without you both.
Last, but certainly not least, I want to thank my best friend and partner,
Libbey. I’ll never be able to thank you enough for your never-ending support.

iii

TABLE OF CONTENTS

LIST OF TABLES................................................................................................. vi
LIST OF FIGURES ..............................................................................................vii
Chapter 1. Introduction ......................................................................................... 1
1.1: Statement of Problem ................................................................................ 6
Chapter 2. Literature Review ................................................................................ 8
2.1: Eolian Processes and Landforms .............................................................. 8
2.1.1: Eolian Processes .............................................................................. 8
2.1.2: Landforms Associated with Eolian Sand ......................................... 10
2.2: Dune Types.............................................................................................. 12
2.3: Dating Methods ........................................................................................ 15
2.3.1: Radiocarbon Dating ........................................................................ 15
2.3.2: Age Considerations ......................................................................... 17
2.3.3: Sources of Error within Radiocarbon Dating Method ...................... 18
2.3.4: Luminescence Dating...................................................................... 20
2.3.5: Sources of Error within Luminescence Dating Method.................... 24
2.4: Northern Great Plains Dune Fields .......................................................... 24
2.4.1: Manitoba ......................................................................................... 24
2.4.2: North Dakota ................................................................................... 26
2.5: Central Great Plains Dune Fields............................................................. 27
2.5.1: Nebraska......................................................................................... 27
2.5.2: Northeastern Colorado .................................................................... 30
2.5.3: Kansas ............................................................................................ 31
2.6: Southern Great Plains Dune Fields.......................................................... 36
2.6.1: Oklahoma........................................................................................ 36
2.6.2: Eastern New Mexico and Northwestern Texas ............................... 37
2.7: Eastern Lake Michigan Coastal Zone ...................................................... 39
2.8: Summary of Past Dune Chronology Research ........................................ 45
Chapter 3: Methods ............................................................................................ 46
3.1: Data Collection......................................................................................... 46
3.2: Probability Density Distributions (PDDs) .................................................. 47
3.3: Principal Components Analysis (PCA) ..................................................... 48
3.4: Time-Slice Maps ...................................................................................... 49
Chapter 4: Results and Discussion ..................................................................... 52
4.1: PDD Results ............................................................................................ 53
4.1.1: Analysis of Great Plains PDDs ........................................................ 53
4.1.2: Analysis of Eastern Lake Michigan Coastal Zone PDDs ................. 56

iv

4.1.3: Comparison of Great Plains and Eastern Lake Michigan
Coastal Zone PDDs .................................................................................. 58
4.1.4: Considerations Associated with Utilizing PDDs for Sand
Dune Chronological Research .................................................................. 61
4.2: PCA Results............................................................................................. 65
4.2.1: Radiocarbon PCA ........................................................................... 67
4.2.2: Luminescence PCA......................................................................... 70
4.3: Time-Slice Map Results ........................................................................... 73
4.3.1: Analysis of Time-Slice Maps ........................................................... 89
4.4: Discussion................................................................................................ 91
4.4.1: Advantages to Utilizing Time-Slice Maps in Dune
Chronology................................................................................................ 92
4.4.2: Potential Catalysts for Dune Activation and Stabilization
in the Great Plains and Eastern Lake Michigan Coastal Zone .................. 93
4.5: Conclusions ............................................................................................. 99
4.5.1: Contributions of this Research ...................................................... 101
4.5.2: Future Research ........................................................................... 102
APPENDIX ........................................................................................................ 103
BIBLIOGRAPHY ............................................................................................... 126

v

LIST OF TABLES

Table 4.1: Subregions and their abbreviations for PCA ...................................... 66
Table 4.2: PCA results from SYSTAT 13 for radiocarbon data ........................... 67
Table 4.3: Rotated loading matrix from SYSTAT 13 for radiocarbon data .......... 68
Table 4.4: PCA results from SYSTAT 13 for luminescence data ........................ 70
Table 4.5: Rotated loading matrix from SYSTAT 13 for luminescence data ....... 71
Table A.1: Radiocarbon and Luminescence data from the Great Plains........... 103
Table A.2: Radiocarbon and Luminescence data from the Eastern Lake
Michigan Coastal zone................................................................................. 116

vi

LIST OF FIGURES

Figure 1.1: Dune field locations in the Great Plains (modified from
Muhs et al.,1997b) ........................................................................................... 2
Figure 1.2: Dune locations along the eastern shore of Lake Michigan
(modified from Van Oort et al., 2001; Arbogast et al., 2004). ........................... 3
Figure 2.1: Illustration of the saltation and creep processes (modified
from Ritter, 1986) ............................................................................................. 9
Figure 2.2: Cross-section of a typical sand dune (modified from
Ritter, 1986) ................................................................................................... 11
Figure 2.3: Illustration of dune form in response to sand supply, vegetation,
and wind power. Constant wind direction is inferred (modified from Hack,
1941).............................................................................................................. 12
Figure 2.4: Basic dune forms. Arrows indicate wind direction (modified from
McKee, 1979)................................................................................................. 13
Figure 2.5: Illustration of the carbon cycle .......................................................... 16
Figure 2.6: Radiocarbon half-life curve (modified from Arnold and Libby,
1949).............................................................................................................. 17
Figure 2.7: IntCal04 tree-ring calibration curve. Red line indicates linear time,
with the blue line indicative of secular variation (modified from Reimer et
al., 2004) ........................................................................................................ 20
Figure 2.8: Illustration of the methods used to obtain a luminescence date
(modified from Aitken, 1998) .......................................................................... 22
Figure 2.9: Review of dating methods relating to environmental radiation and
process (modified from Stokes, 1999) ........................................................... 23
Figure 2.10: Dune fields in the Northern Great Plains......................................... 25
Figure 2.11: Dune fields in Nebraska and northeastern Colorado ...................... 28
Figure 2.12: Dune fields in central and southwestern Kansas ............................ 32
Figure 2.13: Dune fields in the Southern Great Plains ........................................ 38

vii

Figure 2.14: Major dune fields along the eastern Lake Michigan coastal
zone ............................................................................................................... 40
Figure 3.1: Hypothetical example of a probability density distribution (PDD) ...... 48
Figure 3.2: Example of regional maps utilized in construction of time-slice
maps, displaying the spatial distribution of all samples collected from
both regions ................................................................................................... 51
Figure 4.1: PDD showing peaks in radiocarbon dates in the Great Plains .......... 54
Figure 4.2: PDD showing peaks in luminescence dates in the Great Plains ....... 54
Figure 4.3: Stacked PDD showing peaks in radiocarbon and luminescence
dates in the Great Plains ................................................................................ 55
Figure 4.4: PDD showing peaks in radiocarbon dates obtained along the
eastern Lake Michigan coastal zone .............................................................. 56
Figure 4.5: PDD showing peaks in luminescence dates obtained along
the eastern Lake Michigan coastal zone ........................................................ 57
Figure 4.6: Stacked PDD illustration displaying peaks in radiocarbon and
luminescence dates along the eastern Lake Michigan coastal zone.............. 58
Figure 4.7: Stacked PDD illustration displaying peaks in radiocarbon
dates for dunes in the Great Plains and near the eastern Lake
Michigan coastal zone. Similar periods of stability are highlighted................. 59
Figure 4.8: Stacked PDD illustration displaying luminescence dates for
dunes in the Great Plains and near the eastern Lake Michigan coastal
zone. Similar periods of activity are highlighted ............................................. 60
Figure 4.9: PDD of Great Plains luminescence mean dates from 70001000 years ago, with the last 1000 years removed ........................................ 63
Figure 4.10: Great Plains luminescence PDDs arranged by subregion .............. 64
Figure 4.11: Scatterplot of scores for Factors 1 (Northern US) & 2
(Central/Southern Plains) of the radiocarbon data set ................................... 69
Figure 4.12: Scatterplot of scores for Factors 1 (Northern US) & 3 (Limited)
of the radiocarbon data set ............................................................................ 69
Figure 4.13: Scatterplot of scores for Factors 2 (Central/Southern Plains) & 3
(Limited) of the radiocarbon data set ............................................................. 69

viii

Figure 4.14: Scatterplot of scores for Factors 1 (Western Plains) & 2 (Eastern
Plains/Great Lakes) of the luminescence data set ......................................... 72
Figure 4.15: Time-slice maps for the Great Plains (GP) and eastern Lake
Michigan (LM) regions. a): map showing distribution of ages between
7000-6800 years ago; b): map showing distribution of ages between
6800-6600 years ago ..................................................................................... 74
Figure 4.16: Time-slice maps for the Great Plains (GP) and eastern Lake
Michigan (LM) regions. a): map showing distribution of ages between
6600-6400 years ago; b): map showing distribution of ages between
6400-6200 years ago ..................................................................................... 75
Figure 4.17: Time-slice maps for the Great Plains (GP) and eastern Lake
Michigan (LM) regions. a): map showing distribution of ages between
6200-6000 years ago; b): map showing distribution of ages between
6000-5800 years ago ..................................................................................... 76
Figure 4.18: Time-slice maps for the Great Plains (GP) and eastern Lake
Michigan (LM) regions. a): map showing distribution of ages between
5800-5600 years ago; b): map showing distribution of ages between
5600-5400 years ago ..................................................................................... 77
Figure 4.19: Time-slice maps for the Great Plains (GP) and eastern Lake
Michigan (LM) regions. a): map showing distribution of ages between
5400-5200 years ago; b): map showing distribution of ages between
5200-5000 years ago ..................................................................................... 78
Figure 4.20: Time-slice maps for the Great Plains (GP) and eastern Lake
Michigan (LM) regions. a): map showing distribution of ages between
5000-4800 years ago; b): map showing distribution of ages between
4800-4600 years ago ..................................................................................... 78
Figure 4.21: Time-slice maps for the Great Plains (GP) and eastern Lake
Michigan (LM) regions. a): map showing distribution of ages between
4600-4400 years ago; b): map showing distribution of ages between
4400-4200 years ago ..................................................................................... 79
Figure 4.22: Time-slice maps for the Great Plains (GP) and eastern Lake
Michigan (LM) regions. a): map showing distribution of ages between
4200-4000 years ago; b): map showing distribution of ages between
4000-3800 years ago ..................................................................................... 80

ix

Figure 4.23: Time-slice maps for the Great Plains (GP) and eastern Lake
Michigan (LM) regions. a): map showing distribution of ages between
3800-3600 years ago; b): map showing distribution of ages between
3600-3400 years ago ..................................................................................... 81
Figure 4.24: Time-slice maps for the Great Plains (GP) and eastern Lake
Michigan (LM) regions. a): map showing distribution of ages between
3400-3200 years ago; b): map showing distribution of ages between
3200-3000 years ago ..................................................................................... 82
Figure 4.25: Time-slice maps for the Great Plains (GP) and eastern Lake
Michigan (LM) regions. a): map showing distribution of ages between
3000-2800 years ago; b): map showing distribution of ages between
2800-2600 years ago ..................................................................................... 83

Figure 4.26: Time-slice maps for the Great Plains (GP) and eastern Lake
Michigan (LM) regions. a): map showing distribution of ages between
2600-2400 years ago; b): map showing distribution of ages between
2400-2200 years ago ..................................................................................... 84
Figure 4.27: Time-slice maps for the Great Plains (GP) and eastern Lake
Michigan (LM) regions. a): map showing distribution of ages between
2200-2000 years ago; b): map showing distribution of ages between
2000-1800 years ago ..................................................................................... 84
Figure 4.28: Time-slice maps for the Great Plains (GP) and eastern Lake
Michigan (LM) regions. a): map showing distribution of ages between
1800-1600 years ago; b): map showing distribution of ages between
1600-1400 years ago ..................................................................................... 85
Figure 4.29: Time-slice maps for the Great Plains (GP) and eastern Lake
Michigan (LM) regions. a): map showing distribution of ages between
1400-1200 years ago; b): map showing distribution of ages between
1200-1000 years ago ..................................................................................... 86
Figure 4.30: Time-slice maps for the Great Plains (GP) and eastern Lake
Michigan (LM) regions. a): map showing distribution of ages between
1000-800 years ago; b): map showing distribution of ages between
800-600 years ago ......................................................................................... 87
Figure 4.31: Time-slice maps for the Great Plains (GP) and eastern Lake
Michigan (LM) regions. a): map showing distribution of ages between
600-400 years ago; b): map showing distribution of ages between
400-200 years ago ......................................................................................... 88

x

Figure 4.32: Time-slice map for the GP and LM regions, displaying the
distribution of ages between 200 years ago and the present ......................... 89
Figure 4.33: Summary of time-slice maps, derived from data shown in
Figures 4.15-4.32. .......................................................................................... 90
Figure 4.34: Great Lakes radiocarbon and luminescence PDDs combined
with El Niño record and Lake Michigan lake level data (modified from
Moy et al., 2002; Baedke and Thompson, 2000; Arbogast et al.,
unpublished data). Yellow bars highlight similar intervals of high lake
level, El Niño occurrence, and dune stability ................................................. 97
Figure 4.35: Great Plains radiocarbon and luminescence PDD combined
with El Niño event and buried soil organic carbon data (modified from
Moy et al., 2002; Nordt et al., 2008; Arbogast et al., unpublished data)......... 99

xi

Chapter 1: Introduction

Sand dunes are common in the central part of the United States (e.g.
Thorp and Smith, 1952; Arbogast et al., 2009; Halfen and Johnson, 2013). They
occur in environments that range from the semiarid/sub-humid Great Plains to
the humid continental climate of the Great Lakes region. In the Great Plains
(Figure 1.1), large dune fields are present in Manitoba, North Dakota, South
Dakota, Nebraska, Colorado, Kansas, Oklahoma, New Mexico, and Texas (e.g.,
Holliday, 1989; Arbogast, 1996; Muhs et al., 1997a; Havholm and Running, 2005;
Lepper and Scott, 2005; Halfen et al., 2012). Dunes in the Great Lakes region
(Figure 1.2) range from interior dunes in Michigan (e.g., Arbogast et al., 1997) to
coastal dunes along the lakes (Peterson and Dersch, 1981; Snyder, 1986;
Arbogast and Loope, 1999; Blumer et al., 2012; Lovis et al., 2012). The best
developed of these dunes are those along the eastern coastal zone of Lake
Michigan. These dunes may collectively represent the largest body of freshwater
coastal dunes in the world (Peterson and Dersch, 1981).
Dunes throughout the Great Plains region have been studied extensively
since the late 19

th

century. Early studies (e.g., Hay, 1893; Haworth, 1897; Moore,

1920; Lugn, 1935; Smith, 1937, 1939, 1940; Melton, 1940; Simonett, 1960;
Ogden and Kay, 1965; Smith, 1965; David, 1968) were mostly qualitative in
nature and sought to determine sources of dune sand, as well as to construct
late Pleistocene and Holocene dune activity through relative dating methods,
utilizing stratigraphic research of dunes and associated soils.

1

Figure 1.1: Dune field locations in the Great Plains (modified from Muhs
et al., 1997b).

2

Figure 1.2: Dune locations along the eastern shore of Lake Michigan
(modified from Van Oort et al., 2001; Arbogast et al., 2004).

3

The first published radiocarbon dates in the Great Plains were from David
(1971) as part of a study from the Brandon Sand Hills in Manitoba. Similar
studies were conducted from the late 1970s to the present (e.g., Warren, 1976;
David, 1977, 1979; Gile, 1979; Ahlbrandt and Fryberger, 1980; Ahlbrandt et al.,
1983; Wright et al., 1985; Brady, 1989; Holliday, 1989; Swineheart, 1990; Ponte
et al., 1994; Loope et al., 1995; Mason et al., 1995; Muhs et al., 1995; Olson et
al., 1995; Arbogast, 1996; Arbogast and Johnson, 1998; Arbogast and Muhs,
2000; Forman et al., 2001; Holliday, 2001; Muhs and Holliday, 2001; Olson and
Porter, 2002; Feathers, 2003; Miao et al., 2005; Miao et al., 2007; Forman et al.,
2008; Hanson et al., 2009; Rich and Stokes, 2011; Werner et al., 2011; Halfen et
al., 2012; Halfen and Johnson, 2013). These studies collectively suggest that
dune activation and stabilization has occurred in the Great Plains several times
over the past 7000 years, and that mobilization of eolian sand is related to
enhanced drought conditions and reduced vegetative cover.
Geomorphic studies in the Great Lakes region also began in the late 19

th

century, with the most attention focusing on the coastal dunes along the eastern
coast of Lake Michigan. Early coastal dune studies were qualitative in nature; for
example, Cowles (1899) focused on dune vegetation and how differing plant
species can contribute to differences in dune evolution. Dow (1937) investigated
the formation of “perched” dunes on high bluffs north of Manistee, and possible
sources of eolian sand. Scott (1942) observed the differences in dune formation
north and south of the isostatic “hinge line” proposed by Goldthwait (1908). Other
qualitative studies (i.e., Tague, 1946; Olson, 1958a, 1958b, 1958c; Dorr and

4

Eschman, 1970; Buckler, 1979) contributed to the general understanding of
coastal dunes in the Great Lakes region.
Snyder (1985) reported the first radiocarbon dates from the region at
Sleeping Bear Dunes National Lakeshore. Since that time, an abundance of
geomorphic studies have contributed to the chronology of dune activation and
stabilization events in the Great Lakes region from the mid-1990s to the present
(Arbogast and Loope, 1999; Loope and Arbogast, 2000; Van Oort et al., 2001;
Arbogast et al., 2002; Hansen et al., 2002; Arbogast et al., 2004; Lepczyk and
Arbogast, 2005; Hansen et al., 2010; Blumer et al., 2012; Lovis et al., 2012).
These studies also suggested several episodes of dune activation and
stabilization along the eastern Lake Michigan coastal zone over the past 7000
years and have largely attributed eolian sand mobilization to periods of high lake
level.
To date, radiocarbon dating has been the most widely used method to
reconstruct dune chronology in the Great Plains and along the eastern coast of
Lake Michigan (e.g. Arbogast and Loope, 1999; Loope and Arbogast, 2000; Van
Oort et al., 2001; Madole, 1994; 1995; Arbogast, 1996; Stokes and Swineheart,
1997; Holliday, 2001; Olson and Porter, 2002; Goble, 2004), with most ages
derived from charcoal and wood fragments within buried paleosols. More
recently, luminescence dating techniques (i.e. optically-stimulated luminescence,
thermoluminescence, infrared-stimulated luminescence) have allowed
researchers to sample buried sands from various depths in the profile and
directly estimate the timing of past dune activation events (e.g. Hansen et al.,

5

2002; Hanson et al., 2009; Hansen et al. 2010; Rich and Stokes, 2011; Werner et
al., 2011; Blumer et al., 2012; Halfen et al., 2012; Lovis et al., 2012).

1.1: Statement of Problem
This research compares and contrasts dune chronologies in the Great
Plains and along the eastern shore of Lake Michigan for the past 7000 years. It is
motivated by a recent study conducted by Arbogast et al. (2011), which
suggested that dune activation occurred in both the Great Plains and along the
eastern coastal zone of Lake Michigan during the Medieval Warm Period (MWP)
about 1000 years ago. Their finding is interesting because the regions are in
different climate zones: the Great Plains in a semi-arid to arid climate zone, and
the Great Lakes region in a humid climate zone. Past research in both regions
has demonstrated very different drivers of dune activation in each respective
region. In the Great Lakes region, high lake levels as a result of heavy
precipitation and cold winters have generally correlated with dune building
activity (Loope and Arbogast, 2000; Van Oort et al., 2001; Arbogast et al., 2002;
Fisher and Loope, 2005), while drought conditions have corresponded with
periods of dune activation in the Great Plains (Holliday, 1989; Muhs et al., 1997a;
Wolfe et al., 2000; Werner et al., 2011). Given the simultaneous activity in both
regions during the MWP, it is possible that broad climate patterns may influence
dunes in both regions at other similar times during the Holocene.
Comparison of dune chronologies in both regions will be conducted
through collection of published radiocarbon and luminescence data for both

6

regions. Probability density distributions (PDDs) will then be constructed using
the data, with peaks in the OSL distributions used to infer periods of activation,
and peaks in radiocarbon distributions used to infer stabilization events or the
onset of activation events through the Late Holocene. Principal components
analysis (PCA) will then be utilized to determine if there are any multiregional
similarities from an exclusively numerical standpoint. Lastly, time-slice maps are
displayed to demonstrate the spatial patterns of sand dune activity in 200-year
intervals. This research will contribute (1) the first known collection of published
dates from both regions, (2) the first known utilization of PCA in dune chronology
research, (3) PDDs that display peaks in luminescence curves during stages of
activation and peaks in radiocarbon curves during periods of stabilization in both
regions, (4) time-slice maps of dune activity and stabilization for both regions in
200-year increments, and (5) a discussion of potential forcing variables relating
to activation and stability events in both regions.

7

Chapter 2: Literature Review

Abundant research has been conducted on sand dunes and dune
chronologies in the Great Plains and Great Lakes regions. The evaluation of this
literature is organized into four sections. The first section is a broad review of
eolian processes as they relate to the mobilization of wind-blown sand. The
second section is a discussion of the types of dunes associated with the study
regions. The third section is a survey of dating methods used in the
reconstruction of dune chronologies in both regions. The fourth section is a
review of previous studies and dates obtained from sand dunes in the Great
Plains and eastern coastal zone of Lake Michigan.

2.1: Eolian Processes and Landforms
2.1.1: Eolian Processes
Given the scope of this study it is important to have an understanding of
how flowing air can shape sandy terrain. Bagnold (1941) was the first to study
the physics of wind-blown sand through a series of wind tunnel experiments,
showing a relationship between wind velocity and movement of sand grains.
Sand grains are relocated in eolian environments mainly as contact load via
saltation and creep (Figure 2.1). Saltation describes grains that reflect across a
sand dune, and creep occurs when grains roll or slide across a sand dune
(Bagnold, 1941; Ritter, 1986; Bloom, 1991).

8

Figure 2.1: Illustration of the saltation and creep processes (modified
from Ritter, 1986).

The predominant variable for both saltation and creep is wind velocity. For
example, erosion by wind in dune fields occurs largely via deflation, which is the
effect of volatile air mobilizing loose sand. According to Ritter (1986), the
minimum velocity for desert sand mobilization is 16 km/hr, but specific values are
contingent upon density, shape, soil moisture, surface irregularities, mineralogy,
and sorting (Bagnold, 1941; Belly, 1964; Williams, 1964; Woodruff and
Siddoway, 1965; Gerety and Slingerland, 1983; Greeley et al., 1983).
The main factor dictating minimum velocity of wind is the diameter of
sand particles, with 0.84 mm the apparent upper limit for unassisted eolian
transport (Bagnold, 1941). Larger particles are kept in place by saltating grains
bouncing across the surface. Once a sand grain becomes airborne, wind
transport carries it on an irregular, short path within 3 cm of the surface (Bagnold,
1941). As the grains return/descend to the surface, their impact can entrain other
small particles on the landscape, propelling them into the air. Although sand
grains are initially moved by other smaller particles, the grains are ultimately
propelled forward because of wind moving above the surface; wind speed is
close to zero nearer to the surface, but increases dramatically as elevation from

9

the surface increases (Bagnold, 1941). Sand grains larger than 0.84 mm do not
enter the airstream above the surface, and move along the surface via creep,
which is the result of saltating grains propelling larger sand grains forward
without pushing them upward (Bagnold, 1941).

2.1.2: Landforms Associated with Eolian Sand
A variety of landforms can develop from the accumulation of eolian sand,
the largest being sand sheets and sand seas. Sand sheets are broad areas of
sand that exhibit little to no surface relief, while sand seas develop in massive
deserts (e.g. Sahara) where tremendous amounts of sand develop a multitude of
sand sheets and dune forms (McKee, 1979). Within the complex of eolian
landscapes, sand dunes have been the focus of the most geomorphic study
(Ritter, 1986).
Dune forms develop as a result of the interaction of three factors:
vegetation, prevailing winds, and sand supply (Olson, 1958a). When grains of
sand begin moving, they will continue to do so until wind speed drops below a
critical velocity. As this happens sand grains begin to fall out of suspension and
be dropped onto the surface. At microscale this usually happens when obstacles
are present, such as vegetation, which cause wind speed to decrease minimally
(Olson, 1958a). In the early stages of dune formation, a small mass of sand will
begin to accumulate.
As sand deposition continues, the sand mass starts to develop a profile
typically associated with sand dunes. These profiles consist of three separate

10

elements: an erosional surface or backslope, crest, and a depositional surface or
slip face (Figure 2.2). The typical backslope angle is ~8-15°, with the slip face
angle positioned between 30-35°, which is near the angle of repose (Livingstone
and Warren, 1996; Pye and Tsoar, 2009).

Figure 2.2: Cross-section of a typical sand dune (modified from Ritter,
1986).

Differences in wind direction and velocity, particle size, and vegetation, for
example, can result in very distinctive characteristics in many dunes (Ritter,
1986). In spite of the complexity of dune formations, several attempts have been
made in the literature to classify sand dunes based upon appearance (e.g.
Bagnold, 1941; Hack, 1941; McKee, 1966, 1979). Classification relies mostly on
any of a number of distinctive traits including shape, direction in which “arms” are
oriented, evolution, and wind direction. Figure 2.3 illustrates how certain types of
dunes might develop based upon sand supply, wind, and vegetation.

11

Figure 2.3: Illustration of dune form in response to sand
supply, vegetation, and wind power. Constant wind direction is inferred (modified from Hack, 1941).

2.2: Dune Types
According to McKee (1979), nine classes of dune forms are recognized
(Figure 2.4) that can, in turn, be divided into two groups: those that form in the
presence of vegetation, and those that form in extremely arid environments
where plants are absent. The two dune forms that develop in the presence of
vegetation are parabolic dunes and blowout dunes. Parabolic dunes are noted
for their crests bowing downwind, with elongated arms following the dune (Figure
2.4). The elongated arms are fixed by vegetation, while most of the sand in the
dune moves forward downwind (Ritter, 1986). Blowout dunes are common in
what was once the backslope of an existing parabolic dune (Figure 2.4), but can
also be the initial landform in development of parabolic dunes. Blowouts form as

12

a result of reduced vegetation, which allows wind to move sediments forward,
leaving a bowl-shaped depression.

Figure 2.4: Basic dune forms. Arrows indicate wind
direction (modified from McKee, 1979).

Compound parabolic dunes also occur in many places where several
dunes are superimposed, or often as part of a complex dune field where more
than one type of dune has formed (Breed and Grow, 1979). Parabolic and
blowout dunes both typically form when sand supply is high and in areas of
strong wind.
The remaining dune types in Figure 2.4 typically form without the
presence of dense vegetation. Barchan (or cresentic) dunes, barchanoid ridges,
and transverse dunes (Figure 2.4) are genetically similar and are the typical dune
forms in sandy terrain (McKee, 1979). Barchan dunes have a shape opposite to
that of parabolic dunes, with arms extended downwind, and typically form where
13

sand supply is limited but where strong winds are present. Several joined
barchan dunes can form barchanoid ridges, which are asymmetrical dunes that
form at right angles to the prevailing wind direction, with the barchan dunes
themselves forming a wave-like feature in the ridge (McKee, 1979). These ridges
form in areas where sand supply is not limited, and weak winds prevail (Bloom,
1991).
If consistently strong winds are present, or multi-directional winds
continue, the crest of a barchan dune may be reduced and the backslope
flattened, resulting in coppice (or dome) dunes (Figure 2.4). Coppice dunes are a
common landform in sand plains and desert areas, where blowing sand and
small shrubs are abundant. These dunes typically form around small bushes, and
are varying in size based upon sand supply.
Transverse and linear dunes (Figure 2.4) are both ridge-like features that
form in environments without vegetation and/or moisture. Transverse dunes
typically develop perpendicular to wind direction, whereas linear dunes normally
form parallel to the prevailing wind direction, with some seasonal shifts in
prevailing wind direction occurring (McKee, 1979). Reversing dunes (Figure 2.4)
typically take on any of the forms mentioned previously, and occur in regions
where wind direction reverses periodically. Reversing dunes often possess major
and minor slipfaces oriented in different directions (McKee, 1979).
The last dune form to discuss is the star dune (Figure 2.4). Star dunes are
predominantly found in sand seas and deserts, and develop as a result of
multidirectional winds, an abundance of sand, and no vegetation. These dunes

14

are pyramidal mounds of sand, with slipfaces on three or more arms that radiate
outward from the center of the mound (McKee, 1979).

2.3: Dating Methods
Researchers have attempted to construct dune chronologies since the late
19

th

century (e.g. Hay, 1893; Haworth, 1897; Cowles, 1899), with early studies

mostly qualitative in nature, utilizing stratigraphic interpretation and soil
development as relative proxies for chronology. More recently, researchers have
been able to quantitatively construct dune chronologies globally through the use
of two primary methods: radiocarbon and luminescence dating (Pye and Tsoar,
2009). The following is a discussion of both methods.

2.3.1: Radiocarbon Dating
Radiocarbon dating has been widely utilized for reconstruction of dune
chronologies. Developed by Arnold and Libby (1949), this method is based upon
the nature of the carbon cycle and the relationship among three carbon isotopes,
specifically

12

C,

13

C , and

14

C. The first pair are stable isotopes, whereas

14

C is

radioactive and thus unstable.
In order to clearly understand the radiocarbon dating process, a
14

discussion about the carbon cycle (Figure 2.5) is necessary. Carbon-14 ( C)

15

Figure 2.5: Illustration of the carbon cycle.

production is a result of a collision of high-energy neutrons produced by cosmic
14

rays and nitrogen ( N2) gas. The

14

C isotope is rapidly changed to

14

CO2 and

is absorbed by plants through photosynthesis and is incorporated into an
animal’s biomass through herbivory (Schaetzl and Anderson, 2005). While the
plant or animal is alive, it continues to absorb
death, at which time

14

C. Uptake of

14

C stops upon

14

C begins to decay logarithmically at the known half-life of

5730±400 years (Arnold and Libby, 1949) (Figure 2.6). Materials used to produce
a radiocarbon date must have been living at some period. These materials
include wood fragments, organic matter from decaying plants and animals,
shells, bone, ceramics, and bulk soil humates (Holliday, 1989).

16

Figure 2.6: Radiocarbon half-life curve
(modified from Arnold and Libby, 1949).

2.3.2: Age Considerations
Given the multitude of applications with radiocarbon dating, there are
many considerations to take into account with respect to dune-chronology
research. The first is the interpretation of the age obtained via sampling.
Laboratory ages are typically reported with a mean date plus or minus two
standard deviations. Therefore, a date provides a statistical statement – not an
exact period of time – and should only be viewed in context of probability.
Additional considerations are presented within the research design. For
instance, given the half-life of

14

C, radiocarbon dating can only be utilized for

samples ranging from ~50,000 years ago to the present (Kolstrup, 2007).
Another consideration to take into account is that radiocarbon dates on a buried

17

paleosol may only provide a maximum or minimum limiting date for a dunebuilding event, and do not provide information about sand movement over
shorter periods of time, most notably when paleosols are unable to develop (Rich
and Stokes, 2011). Thus, solely utilizing radiocarbon dating for chronologies of
buried soils in dunes may not show all periods of dune stabilization and activation
throughout the period sampled.

2.3.3: Sources of Error within the Radiocarbon Dating Method
Many potential sources of error are possible within the radiocarbon dating
method. One potential source of error exists with the type of material used to
produce a radiocarbon date. For example, shells often exchange carbon with
soils or water surrounding them, altering their radiocarbon age and making them
appear older (Bowman, 1990). The marine reservoir effect can also impact the
accuracy of radiocarbon dates. According to Stuiver and Braziunas (1993),
sample materials that obtain carbon from an alternative source (reservoir) than
atmospheric carbon could potentially yield what are termed “apparent ages.” For
example, the average difference between a terrestrial sample and marine sample
of the same age is about 400 radiocarbon years. The apparent age of the marine
sample is influenced by two possible causes: 1) a delay in exchange rates
between atmospheric CO2 and ocean bicarbonate, and 2) a dilution effect
caused by mixing of surface waters with upwelled deep waters that are very old
(Mangerud, 1972). For example, a shellfish that is presently alive in a marine
environment within a limestone catchment will likely yield a radiocarbon date that

18

is much older than the true age of the shellfish. This occurs because the
limestone, which is weathered and eventually dissolved into bicarbonate, has no
radioactive carbon. Through dilution, the radioactivity of the lake is depleted in
comparison to radiocarbon activity elsewhere. Therefore, marine dates often
require correction to account for this issue (Stuiver and Braziunas, 1993).
Fluctuations in atmospheric

14

C levels through time (secular variation)

caused by alterations in the earth’s magnetic field, as well as changes in sunspot
activity, pose a potential issue as far as accuracy (Kigoshi and Hasegawa, 1966;
Bowman, 1990) is concerned. Additionally, increased levels of atmospheric

14

C

post-1950 were detected after nuclear bomb testing and subsequent radiation
fallout (de Vries, 1958). According to Taylor (1987), bomb testing may have as
much as doubled the amount of

14

C in terrestrial carbon-bearing materials. To

correct for these variations through time, radiocarbon dates are often calibrated.
The most popular calibration method has been by dendrochronology (tree-ring
curves).
The first extensive tree ring correction curve was compiled by Suess
(1967), who examined

14

C levels in tree rings extracted from trees alive prior to

1950. Many calibration curves have been utilized in the years that followed as a
result of more robust data sets, duplicate sampling, and improved radiocarbon
dating methods. The current tree ring correction curve utilized by researchers is
that developed by Reimer et al. (2004), and is shown in Figure 2.7. Convention

19

dictates that calibrated radiocarbon dates are reported as “cal yrs BP” in the
literature to differentiate them from uncalibrated dates. The base year used for

Figure 2.7: IntCal04 tree-ring calibration curve. Red
line indicates linear time, with the blue line
indicative of secular variation (modified from
Reimer et al., 2004).

measurements of how old a sample is via

14

C methods is A.D. 1950, reported as

“yrs BP.” The year 1950 is used for no particular reason other than to recognize
the publication of the first radiocarbon dates calculated in December of 1949
(Taylor, 1987).

2.3.4: Luminescence Dating
In addition to radiocarbon dating, luminescence dating can also be used to
reconstruct dune chronologies. This method is broadly based on the premise that
the emission of luminescence in sand samples is the measure of time since the

20

last light exposure during transport (e.g. saltation, creep), deposition, and burial
of the host deposit (Aitken, 1998; Clarke et al., 1999). Developed by Huntley et
al. (1985), luminescence dating is best utilized on samples dating from ~200,000
years old to modern (Schaetzl and Anderson, 2005), but has been shown to
accurately date materials up to 780,000 years old (Watanuki et al., 2005; Wang
et al., 2006). A discussion of the geoscientific applications of luminescence
dating is presented here, but will not address the physics of the technique.
Details about the physics of the technique can be found in the literature (e.g.
Aitken, 1998; Berger, 1994; Forman, 1989; Wintle, 1993; Duller, 1996) and the
references found within those articles.
A luminescence date is obtained by measuring the amount of stored
radiation trapped within defects in a quartz grain, also called “crystal lattice”
defects (Keizars et al., 2008). Radiation within the crystal lattice comes from
alpha, beta, and gamma radiation emitted during the decay of
232

Th,

40

K, and

87

235

U,

238

U,

Rb, as well as their daughter products, both within the mineral

grain and in their surroundings (Lian, 2007), and is “trapped” within the crystal
lattice defects. When excited by heat or light, the radiation stored within the traps
is capable of escaping (Figure 2.8). A date is determined by measuring the
amount of electrons released from the crystal lattice through stimulation by heat
or light, depending upon the luminescence method used (Stokes, 1999). A
luminescence date, or age, is determined by the following equation:
AGE = DE / DR

21

with AGE estimating the date of burial, DE representative of the absorbed dose
or equivalent dose (total amount of radiation absorbed within the sediment traps,
using the conventional unit of “gray” (Gy)), and DR representing the dose rate, or
dose of radiation that the sand grain can store. The dose rate is determined in

Figure 2.8: Illustration of the methods
used to obtain a luminescence date
(modified from Aitken, 1998).

the lab from a separate sample collected at the same site as the luminescence
sample (Rhodes, 2011).
Three luminescence dating methods are commonly used (Figure 2.9):
specifically, thermoluminescence (TL), infrared stimulated luminescence (IRSL),
and optically stimulated luminescence (OSL), with OSL being the most widely
used method. Luminescence dating has been shown to produce consistent age
estimates on sand dunes on both the Great Plains and the eastern Lake
Michigan coastal zone (e.g. Wolfe et al., 2002; Rich and Stokes, 2011; Halfen et
22

al., 2012; Lovis et al., 2012). For example, Wolfe et al. (2002) collected samples
from the crest, lee slope, and stoss slope of an active dune in the Brandon Sand
Hills of Manitoba to determine the accuracy of IRSL dating with modern dune
activity. Samples conveyed dates of 1 ya from the lee slope, 8 ya from the crest,
and 38 ya from the stoss slope. The dates were found to be consistent with
expected high rates of sand deposition on the crest and lee slope of a dune, as
well as expected net erosion of the stoss slopes of dunes (Wolfe et al., 2002). In
addition, Wolfe et al. (2002) obtained IRSL samples from a dune roadcut and
radiocarbon samples from intervening paleosols in the dune. The IRSL ages
collected from eolian deposits in stratigraphic sections were found to be in the
correct chronological sequence, both in terms of stratigraphic position and to
radiocarbon ages obtained from organic matter in the buried soils, suggesting
that activation events occurred at ~5600, 4000-3100, and ~2000 years ago
(Wolfe et al., 2002). Convention dictates that luminescence dates are reported as
“ya” (years ago) in the literature to differentiate them from radiocarbon dates.

Figure 2.9: Review of dating methods relating to environmental radiation
and process (modified from Stokes, 1999).

23

2.3.5: Sources of Error within Luminescence Dating Method
There are several potential sources of error with luminescence dating that
may yield inaccurate ages. In particular, the OSL dating method assumes that
the sand grains were sufficiently exposed to sunlight before burial, effectively
“zeroing out” any radiation being held within the sand grain. If the sand grain was
only partially bleached when it was buried, it can potentially yield a date older
than burial (Schaetzl and Anderson, 2005). OSL can also yield a date younger
than burial if sand grains are partially bleached by post-depositional disturbance.
For example, Bateman et al. (2003) demonstrated that bioturbation could affect
the accuracy of OSL dating in pocket gopher burrows in the Great Plains due to
exposure and subsequent reburial as a result of excavation. Dose rates can also
differ based upon the sample and geographic location (H. Wang, personal
communication).

2.4: Northern Great Plains Dune Fields
2.4.1: Manitoba
Several dune fields occur in the northern Great Plains. The Brandon Sand
Hills (BSH) and Lauder Sand Hills (LSH) are located in southwestern Manitoba
2

(Figure 2.10). The BSH cover an area of ~1400 km and are derived from sandy
deposits of the Assiniboine delta of Glacial Lake Agassiz (Elson 1960). The Sand
Hills consist of southeast-trending, stabilized parabolic dunes and some
stabilized blowouts.

24

Early studies in the area were conducted by David (1968, 1971), who
sought to determine the chronology of dune activation. Twelve radiocarbon dates
obtained by David (1971) as well as an additional radiocarbon date from Lowdon
and Blake (1975) suggested a short period of dune stability prior until ~3700 yrs
BP, and ensuing activity around 2100, 1500, 900, and 400 yrs BP. Periods of
dune activity were assumed to be the result of regional drought events.
Additional research by David (1977, 1979) involved study of time-lapse aerial
photographs from the mid-20th century, with results showing more recent
stabilization trends throughout the region.

Figure 2.10: Dune fields in the Northern Great Plains.

Following the work by David (1968, 1971, 1977, 1979) and Lowdon and
Blake (1975), Wolfe et al. (2000) published 25 radiocarbon dates as part of a
study that assessed activation events in the region. Results suggested periods of
dune stability at ~2150, 1200, and 550 yrs BP, with dune activity occurring

25

between the periods of stability. Periods of activity were thought to correspond
with episodes of regional drought previously recorded in northern Great Plains
lakes (Fritz et al., 1991). The most recent study conducted on dunes in the BSH
was by Wolfe et al. (2002) that produced results suggesting activation events at
~5600, 4000-3100, and ~2000 years ago, using infrared stimulated luminescence
(IRSL) dates.
The LSH are located in southwestern Manitoba (Figure 2.10) and cover an
2

area of ~70 km . The Sand Hills consist of southeast-trending parabolic dunes
with long, northwest-oriented arms. The first study of the LSH was conducted by
Running et al. (2002), who examined the Holocene history of the area. Their
research, based on radiocarbon dates, suggested that parabolic dune migration
took place from 6700-5400 yrs BP, and again at least six times from 3250 yrs BP
to the present. Havholm and Running (2005) conducted the most recent research
in the LSH, which examines dune stratigraphy and sedimentology using four
radiocarbon dates and previously published (e.g. Forman et al., 1995; Mason et
al., 1997; Stokes and Swineheart, 1997; Wolfe et al., 2002) radiocarbon and
luminescence dates. Data suggest that dune mobilization occurred ~6100 years
ago, and was likely tied to broader-scale drought events in the Great Plains
(Havholm and Running, 2005).

2.4.2: North Dakota
A small number of dune fields also occur in North Dakota. The largest is
the Minot dune field, which is located in the north-central part of North Dakota

26

2

(Figure 2.10), and covers an area of ~1500 km . The dunes in the area consist of
southeast-trending stabilized parabolic dunes with limbs oriented to the
northwest (Muhs et al., 1997a). Early studies in the Minot area involved extensive
soil mapping (e.g. Knobel et al., 1926; DesLauriers, 1990) and sand mapping
(e.g. Clayton et al., 1980; Bluemle, 1982; 1985; and Lord, 1988).
The only quantitative study on the Minot dune field was conducted by
Muhs et al. (1997a). Ten radiocarbon dates were reported on paleosols, and
results indicated that there were at least two episodes of dune activity in the past
1200 years, with the earliest event represented by a radiocarbon age of 1260 yrs
BP. The authors attributed dune activation to a lack of vegetative cover and
drought (Muhs et al., 1997a).

2.5: Central Great Plains Dune Fields
2.5.1: Nebraska
Several dune fields occur within the central Great Plains, ranging from
Nebraska to central Colorado (Figure 2.11). The largest dune field in the region is
2

the Nebraska Sand Hills (NSH), which covers an area of about 57,000 km and
is the largest sand sea in the western hemisphere (Smith, 1965; Ahlbrandt and
Fryberger, 1979). The NSH are bordered by the Niobrara River to the north and
the North Platte and Platte Rivers to the south (Figure 2.11). Through early
geomorphic investigation, the NSH were thought to be no younger than late
Pleistocene in age, based upon the description of sediments below the dunes
and loess adjoining the area (Lugn, 1935; Smith, 1965). The source has been

27

debated through other studies (e.g. Reed and Dreeszen, 1965; Stanley and
Wayne, 1972), with the most commonly agreed upon origin for NSH sands
suggesting the upper Tertiary Ogallala Formation as their source (Lugn, 1960;
1962; Swineheart, 1990).
Dunes in the region consist of parabolic dunes, compound parabolic
dunes, barchan dunes, barchanoid dune ridges, and longitudinal dunes (Loope
and Swineheart, 2000; Mason et al., 2011). “Megabarchans” are also present in
the Sand Hills area, with some more than 100 meters high and several
kilometers long (Mason et al., 2011).

Figure 2.11: Dune fields in Nebraska and northeastern Colorado.

Previous studies in Nebraska focused on dune morphology and texture of
the dune sand (Lugn, 1935; Smith, 1965; 1968; Warren, 1968; Warren, 1976;
Ahlbrandt and Fryberger, 1980). Ahlbrandt and Fryberger (1983) were the first to
report radiocarbon dates from dunes in Nebraska. Their dates, ranging from

28

5150-860 yrs BP, suggested for the first time that the NSH were much younger
than the late Pleistocene timeframe previously thought (Lugn, 1935; Smith,
1965).
Muhs et al. (1995b, 1997b) reported 19 radiocarbon dates from the NSH.
The dates range from 4330-220 yrs BP and suggested that sand deposition
occurred at least twice in the last 4000 years. Stokes and Swineheart (1997)
reported 16 dates from the NSH, with eight radiocarbon dates ranging from 5300270 yrs BP and eight OSL dates ranging from 5730-210 ya. These dates
suggested that dune activation events took place at least twice during the midHolocene. Goble et al. (2004) collected six radiocarbon dates ranging from 4150modern yrs BP and 35 OSL dates ranging from 6180-150 ya, suggesting five
different dune activation events in the mid-to-late Holocene in the NSH. Mason
et al. (2004) collected seven radiocarbon dates ranging from 4150-modern yrs
BP and 10 OSL sample dates ranging from 3900-180 ya, and suggested the
youngest reactivation event likely took place between 1000-700 years ago during
an extended period of drought.
More recent studies in Nebraska have focused on the collection of OSL
dates from previously-studied dune fields, as well as unstudied fields. In a study
conducted in the western NSH and Wray dune field, Forman et al. (2005)
collected 32 OSL samples, with dates ranging from 1490-40 ya, and suggested
that six different dune activation events took place in western Nebraska in the
past 1500 years. Hanson et al. (2009b) collected 18 OSL dates as part of their
study of the Duncan dune field, the easternmost dune field in Nebraska. Dates

29

ranged from 5070-490 ya, suggesting that the Duncan dune field was active at
least twice in the past ~5000 years, between ~4400-3400 years ago and again
during the MWP. According to Hanson et al. (2009b), the dune field was likely
active as a result of regional megadrought in each case. The most recent study
on dunes in Nebraska was conducted by Mason et al. (2011). Their study
contributed 12 mid-to-late Holocene OSL dates, ranging from 7500-490 ya, with
their findings suggesting several periods of dune activation in the past 7000
years.

2.5.2: Northeastern Colorado
Several dune fields occur in northeastern Colorado, with the largest fields
being the Wray, Sterling, Fort Morgan, and Greeley dune fields with a combined
2

area of 4,700 km (Figure 2.11). Derived from South Platte River sediments,
dunes in the region are mostly parabolic in form with southeast-trending crests
and arms oriented to the northwest (Muhs, 1985, 2000). Early studies in
northeastern Colorado were conducted mostly for geologic mapping purposes
(Thorp and Smith, 1952; Hill and Tompkin, 1953; Colton, 1978; Scott, 1978;
Trimble and Machette, 1979; Bryant et al., 1981).
The first effort to determine the age of dunes in northeastern Colorado
was conducted by Muhs (1985) through the use of soil-stratigraphic methods.
Dates of 3000-1500 yrs BP were estimated based upon previous studies dating
paleosols from the southern High Plains (e.g. Gile, 1979). Forman and Maat
(1990) estimated that dunes most recently stabilized about 3000 years ago near

30

Hudson, Colorado, based upon the morphology of surface soils. Jorgensen
(1992) estimated that young, high relief dunes in the Fort Morgan Dune Field
formed about 1500 years ago based upon a regional comparison of soil
properties.
Madole (1994) reported the first radiocarbon dates from northeastern
Colorado. Eight radiocarbon dates were obtained and ranged from 1380-810 yrs
BP. These dates suggested periods of stability in the southern Platte River area.
Madole (1995) also reported 10 radiocarbon dates from paleosols in northeastern
Colorado that show episodic periods of stability from 5640-810 yrs BP. Forman et
al. (1995) obtained four radiocarbon dates from paleosols within dunes on the
south side of the South Platte River, reporting dates ranging from 5520-920 yrs
BP. Muhs et al. (1996) reported radiocarbon dates from the Fort Morgan and
Wray dune fields, reflecting episodic dune reactivation during the Late Holocene,
with the youngest date falling within 1500-1000 yrs BP. In the most recent study
of dunes in Colorado, Clarke and Rendell (2003) collected eight samples from
the Fort Morgan dune field, with reported infrared stimulated luminescence
(IRSL) dates ranging from 4850-370 ya. These dates suggested that dunes in
northeastern Colorado were also episodically active during the late Holocene
megadrought events in the region.

2.5.3: Kansas
A variety of dune fields exist throughout central and southwestern Kansas
(Figure 2.12). The largest dune fields in the region are the Great Bend Sand

31

Prairie, Hutchinson, Abilene, and Arkansas River dune fields, with a combined
2

area of ~5400 km . Derived from late Wisconsin deposits, reworked eolian
sands, and deflated alluvium from the Arkansas River (Simonett, 1960; Arbogast,
1996; Arbogast and Muhs, 2000; Halfen et al., 2012), dunes in Kansas are
mostly parabolic and crescentic in shape, with arms oriented in the direction of
northwesterly prevailing winds through time. Transverse dunes, sand sheets,
barchanoid ridges, and blowouts are also present throughout Kansas (Smith,
1940; Forman et al., 2008; Hanson et al., 2009a; Werner et al., 2011; Halfen et
al., 2012).

Figure 2.12: Dune fields in central and southwestern Kansas.

Early dune studies in Kansas were qualitative in nature and focused on
the Arkansas River valley. Studies conducted by Hay (1893) and Haworth (1897)
sought to determine the source of sand for dunes along the Arkansas River,

32

suggesting the local Ogallala Formation as a source (Hay 1893), as well as
nearby valley sands (Haworth, 1897).
The first study using paleowind direction to infer a source for dune sand
was conducted by Moore (1920). He suggested that the source of sand for the
Arkansas Valley dunes was the Arkansas River floodplain, based upon assumed
northwesterly paleowinds. Smith (1937) noted three separate soil sections in a
blowout near the Hutchinson dune field, and suggested that there were three
different periods of activation and stabilization in recent history. Strike and dip
measurements in a basal dune also led Smith (1937) to suggest that
northeasterly paleowinds formed the dune. Smith (1939, 1940) was the first to
discuss vegetation as a factor in dune morphology in Kansas. Given the
prevalence of parabolic dunes in the region, he suggested that some vegetation
had always been present during the course of dune formation.
Simonett (1960) sought to further determine the age and origin of dunes in
Kansas. Working along the Arkansas River near Syracuse, he obtained cores
across the area. The presence of Peoria loess in the cores showed a more
northerly source for the dune sand, as the loess section was thinner in the
southerly cores. In the same study, Simonett (1960) was also able to show that
more recent southerly winds had reworked dunes in the Syracuse area, based
upon the location and orientation of blowouts.
A study by Porter et al. (1994) in the Cimarron River valley indicated that
dune ages could be estimated based upon the morphology of soils. For example,
they found that older dunes possessed surface soils with A/Bt/2Bk horizonation

33

and younger dunes had a less developed A/C profile (Porter et al., 1994). In a
review of the accounts of early surveyors in the Arkansas River valley, Muhs and
Holliday (1995) cited evidence for active sand in the Great Bend Sand Prairie
(GBSP). Areas of active and inactive dunes extended west along the Arkansas
River into Colorado, suggesting that sand movement along the stream valley
varied extensively depending on location.
The first quantitative studies were conducted in the GBSP in the mid1990s. Arbogast (1996) reported 23 radiocarbon dates from the GBSP, ranging
from 6050 yrs BP-modern. The dates indicated that at least five different periods
of soil formation had occurred in the region at ~2300, 1400, 1100-900, 700-500,
and 300 years ago, and compared well with other late Holocene studies in the
Great Plains (e.g. Ahlbrandt et al., 1983; Muhs, 1985). A study by Arbogast and
Johnson (1998) indicated that dunes in the GBSP typically contain one to two
weakly developed soils, indicating dune activation periods in the late Holocene.
Additionally, the authors suggested that dunes could be easily reactivated if
vegetation is minimized, based on the poor development of surface soils
(Arbogast and Johnson, 1998). Arbogast and Muhs (2000) utilized mineralogy
and trace element concentrations to determine the source of dune sands in the
GBSP. This study indicated that dunes in the region are chemically similar to
sands in the Arkansas River valley, and also suggested that paleowinds were
northwesterly during early sand deposition in the GBSP (Arbogast and Muhs,
2000).

34

Following the GBSP studies, three radiocarbon dates were collected in the
Cimarron Bend area of southwestern Kansas by Olson and Porter (2002),
ranging from 5770-1450 yrs BP. Based upon their findings, the authors
suggested that two periods of dune activation and stability took place during the
mid-to-late Holocene. In the first study to use OSL dates in Kansas, Forman et al.
(2008) collected 23 OSL dates ranging from 6280 ya-present from dunes south
of the Arkansas River. This study suggested that the dates obtained were
reflective of dune activation during previously recorded continental-wide periods
of drought in the tree-ring record.
Hanson et al. (2009a) were the first to construct a chronology of the
Abilene dune field in eastern Kansas (Figure 2.12). They reported 15 late
Holocene OSL dates ranging from 1060-460 ya. The peak activity appears to
have occurred during the Medieval Warm Period (MWP). This interval correlates
with dune activation in other regions of the Great Plains (e.g., Muhs et al., 1996;
Mason et al., 2004; Forman et al., 2005; Hanson et al., 2009b) and suggested
that the MWP impacted areas further east than originally thought (Hanson et al.,
2009a). As part of a study in southwestern Kansas and northwestern Oklahoma
by Werner et al. (2011), eight OSL samples were collected south of the Cimarron
River near Liberal. Dates ranged from 6440-520 ya and suggested at least three
dune activation events occurred during the mid-to-late Holocene: ~6500-5700,
~3600-2600, and ~800-500 years ago (Werner et al., 2011).
In the most recent dune study in Kansas, Halfen et al. (2012) collected 60
OSL dates from the Hutchinson dune field in central Kansas (Figure 2.12). Dates

35

ranged from 2080-80 ya, and indicated that three major episodes of dune
activation took place over the past 2100 years: ~2100-1800, ~1000-900, and
after 600 years ago. According to Halfen et al. (2012), the period of activation at
~1000 years ago appears to be coincident with regional-scale dune activity in the
NSH (Miao et al., 2007), Duncan Dune Field (Hanson et al., 2009), and Abilene
Dune Field (Hanson et al., 2010) during the MWP.

2.6: Southern Great Plains Dune Fields
2.6.1: Oklahoma
Although a number of dune fields exist in northwestern Oklahoma (Figure
2.13), the dunes in the region remain some of the least-studied in the Great
Plains. Dune fields in the region are complex, consisting of mostly transverse
dunes and barchanoid ridges, with single and compound parabolic dunes also
present (Brady, 1989). Few quantitative studies have been conducted on dunes
in Oklahoma, with the first study conducted by Brady (1989) in Alfalfa and Major
Counties in northwest Oklahoma. He produced two radiocarbon dates from
paleosols within dunes, one at 6385 yrs BP and the other 1200 yrs BP. The next
quantitative study was conducted by Lepper and Scott (2005) in the Cimarron
River valley in southeast Major and northwest Kingfisher counties in northern
Oklahoma. Nine late Holocene OSL dates, ranging from 3330-770 ya, were
collected, as well as two late Holocene radiocarbon dates, ranging from 16301120 yrs BP.

36

The most recent dune study in Oklahoma was conducted by Werner et al.
(2011). Six late Holocene OSL dates were obtained, ranging from 810-460 ya.
According to Werner et al. (2011), these dates likely reflect one period of dune
activation during the late Holocene.

2.6.2: Eastern New Mexico and Northwestern Texas
Several dune fields are present in the Southern High Plains (SHP) of
eastern New Mexico and northwestern Texas (Figure 2.13). The largest dune
fields in the region are the Muleshoe, Mescalero, Lea-Yoakum, Andrews, and
2

Monahans dune fields, with a combined area of ~8750 km . Dunes in the region
2

are mostly stable except for an area of about 300 km in the Monahans dune
field that is still active. Dune forms in the region include parabolic dunes with
blowouts, barchan dunes, barchanoid ridges, and coppice dunes (Rich and
Stokes, 2011).
Early studies were conducted to study the geology and soils of the region
(e.g. Melton, 1940; Wendorf et al., 1955; Green, 1961; Frye and Leonard, 1964;
Reeves, 1976; Gile, 1981). The dunes overlie the Blackwater Draw Formation, a
late Quaternary eolian deposit thought to be derived from two sources: the
Pliocene-Miocene Ogallala Formation (Lugn, 1968), and sediments from the
Pecos River Valley (Holliday, 1989). The source of dune sands in the SHP has
also been previously attributed to two different sources, the first being the Pecos
River Valley (e.g. Huffington and Albritton, 1941; Green, 1961; Hawley et al.,
1976; Carlisle and Marrs, 1982). However, geochemical and mineralogical

37

analysis conducted by Muhs and Holliday (2001) indicates a source more like
that of the Blackwater Draw Formation.

Figure 2.13: Dune fields in the Southern Great
Plains.

The first detailed study of Holocene dune activation in this region was
conducted by Holliday (1989). Based upon geomorphological, paleontological,
and archaeological evidence, he suggested that widespread dune activation
occurred across the SHP between 5500 and 4500 years ago, which appeared to
correspond with activation events across the broader Great Plains region
(Holliday, 1989). As part of a study on the geochronology and stratigraphy of the

38

Southern High Plains sand, Holliday (2001) collected 19 radiocarbon dates from
the Andrews/Monahans, Lea-Yoakum, and Muleshoe dune fields, ranging from
6130 yrs BP-modern. According to Holliday (2001), the Muleshoe dunes were
active at least five times during the past 1400 years, and the Andrews/Monahans
dunes were active at least twice during the past 2300 years (Holliday, 2001).
Feathers (2003) contributed five mid-to-late Holocene OSL dates as part of a
study to determine the accuracy of OSL dating on a variety of sediments in the
SHP. Dates ranged from 6500-950 ya, indicating that several periods of
activation likely took place through the mid-to-late Holocene.
The most recent study in the region was conducted by Rich and Stokes
(2011). As part of a broad study of the Muleshoe, Lea-Yoakum, Mescalero, and
Monahans/Andrews dune fields (Figure 2.13), 19 OSL samples were collected,
with dates ranging from 5100-70 ya. The authors found that all of the dune fields
sampled were active at multiple times in the mid-to-late Holocene, as well as
during the “Dust Bowl” event in the 1930s.

2.7: Eastern Lake Michigan Coastal Zone
One of the largest systems of freshwater coastal dunes in the world
occurs along the eastern coast of Lake Michigan (Peterson and Dersch, 1981)
(Figure 2.14). Dunes in this part of the Great Lakes Region have been studied
since the late 19

th

century, with early studies being mostly qualitative in nature.

The first of these studies was conducted by Cowles (1899), who studied dune
vegetation and evolution of dune forms as related to plant species. The next

39

research was by Dow (1937), who studied the formation of “perched” dunes north
of Manistee. Subsequent qualitative dune research continued until the end of the
1970s (i.e. Scott, 1942; Tague, 1946; Olson, 1958a, 1958b, 1958c; Dorr and
Eschman, 1970; Buckler, 1979). These studies as a whole provided much of the
background for future qualitative and quantitative research.

Figure 2.14: Major dune fields along the eastern Lake Michigan coastal zone.

Snyder (1985) was the first to report radiocarbon dates from dunes along
the eastern coastal zone as part of a study at Sleeping Bear Dunes National
Lakeshore (Figure 2.14). Three radiocarbon dates were collected from paleosols

40

that were exposed as a result of bluff erosion. The stratigraphically lowest soil
within the dune produced a date of about 4600 yrs BP, with the middle and upper
soils producing dates of about 2800 and 700 yrs BP, respectively. Snyder (1985)
interpreted the older date as being from a soil that developed when the lakeshore
was far to the west during the Chippewa low phase, with the other two sites
representing periods of dune stabilization in the late Holocene.
Following the Snyder (1985) study, radiocarbon dating was utilized
extensively in the late 1990s to construct dune chronologies along the eastern
Lake Michigan coastal zone. The first of these studies was conducted by
Arbogast and Loope (1999) who presented five radiocarbon dates from the
Nordhouse dunes, Nugent and Jackson quarries, and the Rosy Mound Natural
Area. A sample date of about 4030 yrs BP from the Nordhouse dunes, dates of
3720 and 3730 yrs BP from the Nugent and Jackson quarries, and dates of 2920
and 2890 yrs BP from the Rosy Mound site suggested that dunes formed during
the Nipissing high lake phase at the Nordhouse site, and at the other two sites (to
the south) the dunes formed sometime later (Arbogast and Loope, 1999).
Following the Arbogast and Loope (1999) study, Loope and Arbogast
(2000) collected 75 radiocarbon samples from paleosol outcrops at 32 different
locations along Lake Michigan’s eastern shore, with dates ranging from 5330 yrs
BP to modern. The authors constructed a probability density distribution and
compared it with a late Holocene lake level curve created by Thompson and
Baedke (1999). The authors found that peaks in soil development coincided with
~150 year high lake-level periods over the past ~1500 years, suggesting cyclical

41

periods of stability (Loope and Arbogast, 2000). In a similar study, Van Oort et al.
(2001) collected 16 samples from paleosol outcrops at Van Buren State Park
along the southeastern coast of Lake Michigan. Dates ranged from 5160 yrs BP
to modern, indicating several periods of dune activation and stability during the
mid-to-late Holocene.
As part of a study on dunes along the central part of the eastern coastal
zone of Lake Michigan, Arbogast et al. (2002) collected 16 radiocarbon dates
from four dunes south of Holland, ranging from 4840-35 yrs BP. Based on the
dates, they found that dune activation likely began during the Nipissing high lake
stage, with a period of dune stability until about 4000 years ago. Dates
suggested that activation occurred around 4000 years ago, with later periods of
dune growth at about 3200, 2400, and 900 years ago (Arbogast et al., 2002). The
dune activation periods in the late Holocene appeared to coincide with high lake
level stages (Baedke and Thompson, 2000). As part of a study on backdune
activity near Holland, Hansen et al. (2002) collected four OSL samples from three
separate dunes, with dates ranging from 4990-3720 ya. The dates were all within
one standard deviation of each another, suggesting that deposition likely
occurred in a single activation event (Hansen et al., 2002). The authors attributed
this period of dune growth to sands supplied after the Nipissing high lake phase,
when sand supply was likely plentiful (Hansen et al., 2002).
In association with these studies, a relatively well-developed paleosol,
informally named the “Holland Paleosol,” was recognized in the upper part of the
stratigraphic sequence in dunes along the southeastern Lake Michigan coastal

42

zone. Arbogast et al. (2004) collected seven radiocarbon samples from dunes
from Montague south to the Indiana Dunes National Lakeshore, with dates
ranging from 3090-50 yrs BP. The soil varies in development from weakly
developed Spodosols with A-E-Bs-Bw-BC-C profiles, to Entisols with A-Bw-BC-C
profiles. As paleosols in dunes typically have weakly-developed A-C
horizonation, the development of the soils suggested a long period of dune
stability occurred in the region. This Holland Paleosol would probably qualify as a
formal pedostratigraphic unit if it were covered by an overlying formal lithographic
or allostratigraphic unit (Arbogast et al., 2004).
Following the Arbogast et al. (2004) study, Lepczyk and Arbogast (2005)
collected 12 radiocarbon samples from dunes in Petoskey State Park (Figure
2.14), with dates ranging from 4620 yrs BP-modern. The authors found that
several episodes of dune activation and stability have occurred over the past
5000 years.
Hansen et al. (2010) published a study conducted at P.J. Hoffmaster and
Warren Dunes State Parks that sought to further explain dune chronology in
southwestern Michigan. Twenty mid-to-late Holocene OSL dates were reported,
ranging from 4360-710 ya. Fourteen radiocarbon dates were also reported,
ranging from 2970-180 yrs BP. The authors broadly identified six stages of dune
development in southwestern Michigan: a series of dune activation and
stabilization stages after deglaciation until ~5700 years ago, activation during the
Nipissing phase from ~5700-3800 years ago, a period of stability from ~38003300 years ago, dune activation and stabilization as a result of the Algoma phase

43

~3300-1600 years ago, dune stabilization ~1600-500 years ago, and stages of
activation and stabilization in the last 500 years (Hansen et al., 2010).
Following the Hansen et al. (2010) study, Blumer et al. (2012) conducted a
study on perched dunes at the Arcadia dune field in northwest lower Michigan
using OSL dating to evaluate the chronology of perched dune growth. A total of
12 OSL dates were collected from three separate exposures, ranging from 4500320 ya. Through pedostratigraphic analysis and analysis of OSL dates, the
authors identified four distinct periods of dune activation: ~4500 ya (during the
Nipissing phase), ~3500 ya during the post-Nipissing phase, ~1700 ya, and
between 1000-500 ya (Blumer et al., 2012). Through comparison of their OSL
dates with previous dune chronologies that utilized uncalibrated radiocarbon
dates for their analysis (e.g. Snyder, 1985; Anderton and Loope, 1995), Blumer
et al. (2012) provided a model for comparisons between uncalibrated
radiocarbon and OSL chronologies in dune systems.
The most recent study conducted on dunes along the eastern shore of
Lake Michigan was by Lovis et al. (2012). As part of a geoarchaeological study
on dune activation, the authors collected 30 mid-to-late Holocene radiocarbon
dates ranging from 6550-150 yrs BP, and 28 mid-to-late Holocene OSL dates
ranging from 5150-540 ya, suggesting that several periods of dune stability and
activation occurred during the mid-to-late Holocene along the eastern Lake
Michigan coastline.

44

2.8: Summary of Past Dune Chronology Research
In summary, the study of dune chronology in both the Great Plains and
eastern Lake Michigan coastal zone is a cumulative product of more than a
century of early qualitative and more recent quantitative dune research in both
areas. Chronological studies in the early 1980s explored the variability of dune
activation and stabilization events in each region. A study by Arbogast et al.
(2011) drew attention to a similar period of dune activation in the Great Plains
and eastern Lake Michigan coastal zone during the Medieval Warm Period, but
was limited to data from three sites in the Great Plains and one Lake Michigan
site. This research provides an analysis of all radiocarbon and luminescence
dates in the literature for both regions, and utilizes probability density
distributions (PDDs) and Principal Components Analysis (PCA) along with
regional time-slice maps to compare and contrast dune chronologies in both
areas.

45

Chapter 3: Methods

In order to compare the dune chronologies of the Great Plains and eastern
Lake Michigan coastal zone over the past 7,000 years, chronological data from
published literature for both areas were collected. The methodologies used in this
research to compare and contrast dune chronologies are based on the
interpretation of peaks in probability density distributions (PDDs), interpretation of
dimensions using Principal Components Analysis (PCA), and the analysis of 200year time-slice maps. The use of PDDs has been well documented in previous
studies in both the Great Plains and eastern Lake Michigan coastal zone
(Forman et al., 2008; Hanson et al., 2009; Blumer et al., 2012; Lovis et al., 2012).
No previous record of the utilization of PCA for dune chronology research has
been found. Time-slice maps have been used more recently to display regional
activation and stability events during the late Holocene in 100-year “slices” of
time in the Great Plains (Halfen and Johnson, 2013). These techniques provide
different ways to compare and contrast dune chronologies graphically.

3.1: Data Collection
Data for this study were collected as part of the literature review process.
The collection procedure was based on three criteria: sites with a mean date
between 7000 ya-modern for luminescence dates, and an uncalibrated mean
date between 7000 yrs BP-modern for radiocarbon dates, the sample must have

46

been obtained from within a sand dune, and, if applicable, the sample was noted
as “reliable” by the author.
Data meeting all of the above requirements were entered into an Excel
spreadsheet, and were sorted by location, lab number, author, mean date,
standard deviation, and date type (radiocarbon or luminescence). Dates were
then separated by region (Great Plains or eastern Lake Michigan coastal zone)
into two separate spreadsheets (see Appendix A). Great Plains data were
divided further into three subregions for comparative analysis: Northern Great
Plains (Manitoba and North Dakota), Central Great Plains (northeastern
Colorado, Nebraska, and Kansas), and Southern Great Plains (Oklahoma,
eastern New Mexico, and northwestern Texas).

3.2: Probability Density Distributions (PDDs)
The most widely used method to present dune chronologies in the Great
Plains and eastern Lake Michigan coastal zone has been through the use of
probability density distributions (PDDs) of radiocarbon and luminescence ages
(e.g. Hanson et al., 2009a; Blumer et al., 2012; Lovis et al., 2012; Halfen and
Johnson, 2013). PDDs display the varying likelihood of probable ages through
time (on the x-axis) and normalize the function to unity; in the scope of this
research, unity is represented by a value of one (1) on the y-axis. However, the
y-axis does not infer actual probability similar to that of a normal curve. Rather,
the y-axis in a PDD shows that some dates are more likely than others, and
when taken together, have higher density distribution values and appear as

47

peaks in the PDD curve (Lovis et al., 2012). Alternatively, if a particular date does
not occur (for example, 500 years ago), the density value would be zero, and no
peak would be displayed on the plot (see Figure 3.1).

Figure 3.1: Hypothetical example of a probability density distribution (PDD).

For this research, PDDs were generated by incorporating mean dates and
their error (one standard deviation in this study, or 1σ) over the past 7,000 years
of dune activity, as published in the literature. Data were entered into CalPal
(v.2013) calibration software, and all radiocarbon dates were calibrated as part of
the PDD creation process. PDDs were also generated in CalPal, utilizing a
Gaussian kernel density function over a dataset with an unknown bandwidth (see
Figure 3.1). The generated PDDs were saved as .eps (Encapsulated PostScript)
files, and Adobe Illustrator was utilized to color and fill the PDDs. The PDD plots
were then stacked for comparative purposes when necessary.

3.3: Principal Components Analysis (PCA)
Aside from PDDs, the use of other statistical methods in dune chronology
research is not well documented. There are methods, however, that can be
utilized to better develop our understanding of dune evolution with respect to time
and location. Principal Components Analysis (PCA) is one such method that is
most often used for data summarization or data reduction. PCA is commonly

48

applied to sizable datasets that display evidence of collinearity, with the intent of
finding the least amount of independent variables that best represent the
variation of the total number of independent variables in the dataset. In the
context of this research, PCA is utilized to potentially uncover patterns or
interrelationships at a local scale, and allows for further analysis of data structure
(Jolliffe, 2002). As a result, geographical patterns can emerge from the dataset,
offering a useful tool in the comparison of interregional chronological studies of
dunes.
Radiocarbon and luminescence dates were entered into SYSTAT 13
statistical software, and a PCA was run for both datasets. Based upon generation
of eigenvalues greater than one, as well as factor loadings and percentage of
variance explained, factors were extracted and rotated using a VARIMAX
standard. This rotation allows for variance maximization of loadings on a
designated dimension, which can assist in the interpretation and explanation of
all dimensions (B. Pigozzi, personal communication).

3.4: Time-Slice Maps
Although the use of PDDs has been and will likely continue to be a
common method to construct sand dune chronologies, they do not show spatial
patterns. Although typically not an issue with chronologies at a local scale, the
lack of location data can be troublesome when comparing and contrasting sand
dune activity on a interregional scale.

49

As a result of this shortfall, time-slice maps have recently been utilized to
display spatial patterns of sand dune activity (e.g. Halfen and Johnson, 2013).
These maps display a specific study area, with points placed in locations of dune
activity for a particular period of time. As part of a review of Great Plains dune
chronologies, Halfen and Johnson (2013) used time-slice maps to display
location and timing of dune activity and stability through the late Pleistocene and
Holocene. However, maps of this sort have not yet been utilized for interregional
dune chronologies. In this study, regional maps were generated using ArcMap
10.1 and saved as .jpg (JPEG) files. Adobe Illustrator was then utilized to
combine the two regional maps into a single figure, with luminescence and
radiocarbon data plotted through the interpretation of locational data from
previous literature. To demonstrate how time-slice maps are constructed, a map
displaying the geographic location of all radiocarbon and luminescence samples
collected from both regions is shown in Figure 3.2. The map illustrates the spatial
distribution of the samples collected through decades of dune research in each
region, as well as displaying locations that have yet to be studied. Red squares
represent dune activity supported by luminescence dates, while green circles
represent dune stability supported by radiocarbon dates. Differing shapes were
utilized for interpretation of figures if color is not available.

50

Figure 3.2: Example of regional maps utilized in construction of
time-slice maps, displaying the spatial distribution of all samples
collected from both regions.

51

Chapter 4: Results and Discussion

This chapter is organized into five main sections. First, PDD results are
organized by region for comparison purposes, beginning with Great Plains
radiocarbon and luminescence plots. Radiocarbon and luminescence PDDs are
first shown separately, and then graphically stacked to compare regional
chronologies in a single illustration. Next, Great Plains and eastern Lake
Michigan coastal zone radiocarbon and luminescence plots are overlain into two
separate plots for comparison purposes. This comprehensive comparison of the
generated PDDs provides an understanding of chronological trends, and in
addition, can enlighten the discussion of issues associated with using PDDs to
assess the chronology of sand dune evolution.
The second part of the chapter is a discussion of Principal Components
Analysis (PCA) in this dune chronology research. PCA is most commonly applied
to large collinear datasets as part of a data reduction process. The typical
objective of such analysis is to find the least amount of independent variables
that best represent the variation of the entirety of independent variables in the
data set. The primary reason for utilizing PCA in this study is to investigate
possible underlying patterns of interrelationships within the data set, and thus
develop local or regional groupings based upon similar dune activity.
The third part of the chapter describes the development and interpretation
of comparative “time-slice” maps. These maps are designed to visually compare
the geographic distribution of luminescence and radiocarbon ages between the

52

two regions in 200-year “slices” of time over the past 7000 years. Similar to the
PDDs, the time-slice maps provide a technique to compare and contrast
activation and stabilization events in both regions through the mid-to-late
Holocene. The fourth part of the chapter is a discussion of the applications of the
methods used as part of this chronology research, and provides a deeper
analysis of potential drivers for dune activity and stability in both regions. The fifth
and final part of the chapter incorporates both the conclusions as well as
suggestions for future research.

4.1: PDD Results
4.1.1: Analysis of Great Plains PDDs
A total of 125 radiocarbon ages were acquired from samples taken within
Great Plains dune fields. Prior to constructing the PDD, dates were calibrated
using CalPal calibration software (v.2013, developed by Weininger et al., 2013)
to provide a similar time scale for comparison of radiocarbon and luminescence
dates. The PDD of radiocarbon dates from the Great Plains suggests that nine
main periods of dune stability and soil development have occurred during the
past 7000 years (Figure 4.1). Notable peaks in the probability of radiocarbon
dates occur ~6800-6400, ~6200-5600, ~4900-4600, ~4500-4100, ~3200-2900,
~2800-2500, ~2400-2000, ~1600-1300, and ~1100-500 years ago.

53

Figure 4.1: PDD showing peaks in radiocarbon
dates in the Great Plains.

A total of 223 luminescence dates were acquired from samples recovered
from Great Plains dune fields. The Great Plains luminescence PDD (Figure 4.2)
suggests that four potential activation events occurred over the past 7000 years,
most notably ~1000-600, ~500-300, ~200-100, and ~50 years ago.

Figure 4.2: PDD showing peaks in luminescence
dates in the Great Plains.

54

In order to more closely compare the probability of stabilization and
activation events in the Great Plains, radiocarbon and luminescence PDDs were
stacked in a single illustration (Figure 4.3). The plots suggest that both
stabilization and activation periods have occurred over the past 1500 years. A
spike in the radiocarbon peak ~400 years ago corresponds with a gradual
decrease in the luminescence PDD, suggesting ~1000 years of dune stability in
the region. Elevated luminescence peaks in the last ~300 years appear to
correspond with a small peak in the radiocarbon PDD, suggesting that many
dunes have recently been active.

Figure 4.3: Stacked PDD showing peaks in
radiocarbon and luminescence dates in the
Great Plains.

The abundance of luminescence dates younger than 1000 years is
noteworthy. At face value, a high number of younger dates suggests that sand

55

dunes in the Great Plains are very young. However, the prevalence of younger
luminescence dates may be explained in part by the Great Plains being in a
semiarid/sub-humid environment. Low precipitation inhibits vegetation growth,
which allows dunes to reactivate on a more frequent basis. As a consequence,
reworked sands, especially common in the upper parts of dunes, would often
return much younger luminescence dates (e.g., Wolfe et al., 2002).

4.1.2: Analysis of Eastern Lake Michigan Coastal Zone PDDs
A total of 166 radiocarbon dates were analyzed from sand dunes along
the eastern Lake Michigan coastal zone. The PDD generated from these data
suggests that five periods of dune stability and soil development occurred during
the past 7000 years (Figure 4.4). The highest probability of dune stabilization
occurs from ~4300-4000, ~3500-2900, ~2200-2000, ~1000-700, and ~500 years
ago to the present.

Figure 4.4: PDD showing peaks in radiocarbon
dates obtained along the eastern Lake Michigan
coastal zone.

56

Compared to the radiocarbon PDD in Figure 4.4, the luminescence PDD
data were less variable, and suggested that six dune activation events occurred
over the past 7000 years (Figure 4.5). Notable peaks occur from ~2900-2600,
~2300-1600, ~1000-700, and ~400-200 years ago. The subdued nature of the
peak from ~4600-3100 years ago is difficult to interpret.

Figure 4.5: PDD showing peaks in luminescence
dates obtained along the eastern Lake Michigan
coastal zone.

As previously demonstrated with the Great Plains radiocarbon and
luminescence PDDs (Figure 4.3), stacking the two plots provides an opportunity
to closely compare and contrast the probability of dune stabilization and
activation events. Stacking the eastern Lake Michigan coastal zone radiocarbon
and luminescence PDDs within a single illustration (Figure 4.6) supports the
hypothesis that five main periods of stabilization and activation have likely
occurred over the past 4000 years. A vertical variation in the radiocarbon peak
from ~3600-2900 years ago implies a short period of dune stability during that
time, followed by a period of dune activation from ~2900-1700 years ago. A
57

notable peak in the radiocarbon PDD from ~1900-1100 indicates that a period of
dune stability likely occurred at this time, subsequent to an interval of activation
between ~1000-800 years ago. The presence of notable radiocarbon and
luminescence peaks in the last 500 years suggests that similar dune stabilization
and activation events may have occurred at a local level.

Figure 4.6: Stacked PDD illustration displaying
peaks in radiocarbon and luminescence dates
along the eastern Lake Michigan coastal zone.

4.1.3: Comparison of Great Plains and Eastern Lake Michigan Coastal Zone
PDDs
The primary goal of this study is to compare and contrast dune
stabilization and activity in the Great Plains and eastern Lake Michigan coastal
zone over the past 7000 years. A useful tool for multiregional comparison of dune
evolution is the graphical stacking of radiocarbon and luminescence PDDs from

58

each respective region. Stacking radiocarbon PDDs from the Great Plains and
eastern Lake Michigan coastal zone into a single illustration provides for
comparison of dune stabilization events in both regions over the past 7000 years
(Figure 4.7). The stacked PDDs suggest that six similar periods of dune stability
have occurred over this interval of time in the two regions. The oldest
synchronous interval of stability appears to have transpired from ~5800-5600
years ago, with subsequent periods of stability occurring from ~4300-4100,
~3600-3400, ~3100-2900, ~1000-750, and ~500 years ago to the present. In
addition to displaying synchronous dune stabilization events, asynchronous
events for the past 7000 years are also displayed in Figure 4.7. For example,

Figure 4.7: Stacked PDD illustration displaying
peaks in radiocarbon dates for dunes in the
Great Plains and near the eastern Lake
Michigan coastal zone. Similar periods of
stability are highlighted.

59

notable peaks in the Great Plains from ~6700-6500, ~2800-2500, and ~24002100 suggest dunes were likely stable in the Great Plains during these periods of
time.
Similar to the stacked radiocarbon PDDs for both regions shown in Figure
4.7, stacked luminescence PDDs from the Great Plains and eastern Lake
Michigan coastal zone allow for comparison of dune activation events in both
regions (Figure 4.8). The illustration suggests that two main periods of similar
dune activation likely occurred in both locations in the past 1000 years. The
oldest appears to have taken place from ~1000-600 years ago, followed by a
period of activation from ~500-300 years ago.

Figure 4.8: Stacked PDD illustration displayIng luminescence dates for dunes in the
Great Plains and near the eastern Lake
Michigan coastal zone. Similar periods of
activation are highlighted.

60

Similar to the stability events shown in Figure 4.7, many asynchronous
activation events are also displayed in Figure 4.8. Two such events appear to
have occurred over the past 5000 years, the first from ~4800-1600 years ago
along the eastern Lake Michigan coastal zone, while relatively limited activity was
displayed on the Great Plains luminescence PDD during the same time interval.
Two notable peaks occur from ~400-100 years ago in the Great Plains, which
correspond with a decline in the Lake Michigan luminescence peak.
In summary, visual analysis of the radiocarbon PDDs (Figures. 4.1, 4.4,
and 4.7) suggests that several simultaneous periods of dune stability occurred in
both regions over the past 7000 years. The luminescence PDDs (Figures 4.2,
4.5, and 4.8) suggest that only two similar activation events likely occurred over
the past 7000 years in both regions, but provide limited information about
comparable dune activity prior to 1000 years ago in the Great Plains.

4.1.4: Considerations Associated with Utilizing PDDs for Sand Dune
Chronological Research
Although PDDs are commonly used to illustrate the chronology of sand
dune evolution (e.g. Hanson et al., 2009a; Blumer et al., 2012; Lovis et al., 2012;
Halfen and Johnson, 2013), several conceptual issues regarding this method at a
regional level have recently surfaced. According to Halfen and Johnson (2013),
this method may be problematic for a number of reasons. For example, PDDs
used for regional dune chronologies often lack an important geographical
component of where individual samples were collected. While typically not an
issue when presenting subregional data (e.g., the Minot dune field), locational

61

data are critical for interregional analysis, as when comparing distributions of
dune evolution among several locations (Halfen and Johnson, 2013).
Additionally, PDDs are particularly sensitive to clustering within datasets. An
abundance of radiocarbon dates for a certain time period (e.g., 1000-800 years
ago) may reflect a false high probability for that period, while disregarding other
potentially significant intervals of stability due to a smaller sample size for those
usually earlier time intervals. This section discusses the issues presented when
using PDDs for this research.
Although the absence of a geographical component undeniably impacted
the way PDDs displayed dune chronologies in Figures 4.1-4.8, the most
noteworthy issue with regard to PDDs as used in this research is potential
sample bias, particularly as regarding sample age. An excellent example of this
type of bias can be seen when analyzing the luminescence PDD from the Great
Plains (Figure 4.2). In this case, an abundance of dates younger than 1000 years
old (n=161, ~72% of data set) has significantly skewed the PDD, resulting in
fewer notable curves prior to 1000 years ago, even though several samples are
present. This is likely due to the fact that samples nearer to the surface are often
easier to collect, and have been potentially reworked through periods of
subsequent activation or bioturbation. In an effort to demonstrate how bias
affects the results of the PDD, a separate PDD of luminescence dates for the
Great Plains is presented in Figure 4.9 which removes dates younger than 1000
ya. Thus, the PDD values for older periods are proportionately increased.

62

Figure 4.9: PDD of Great Plains luminescence mean
dates from 7000-1000 years ago, with the last 1000
years removed.

The Great Plains luminescence PDD in Figure 4.9 is noticeably different
than the Great Plains luminescence PDD shown in Figure 4.2. Because of the
removal of dates younger than 1000 years old, substantial peaks are now visible
from ~6000-5100, ~4300-2700, ~2500-1800, and ~1500-700 years ago. The
overlap in the plot, i.e., its “tail,” appears younger than 1000 ya due to the
standard deviations of younger dates being factored into the PDD. To further
illustrate how sample bias can affect peaks in a PDD, Figure 4.10 shows four
Great Plains luminescence PDDs in a single illustration, with four subregions
displayed.

63

Figure 4.10: Great Plains luminescence PDDs
arranged by subregion.

The plots in Figure 4.10 display how PDDs are affected by sample size
and sample distribution for four subregions in the Great Plains. For example, the
PDD for the Northern Great Plains is influenced by the paucity of dates (n = 8).
The peaks show where the five dates OSL dates are present (three of eight
samples are modern dates and therefore not shown on the PDD). A younger
date with a smaller standard deviation results in a taller, thinner peak, as shown
from ~2300-1700 ya. Older ages typically have a larger standard deviation, with a
shorter, wider peak, as displayed from 6300-4900 ya. Additionally, the Central

64

Great Plains and Kansas luminescence PDDs (Figure 4.10) are skewed to the
late Holocene given the high number of ages younger than 1000 years. The
Southern Great Plains luminescence PDD data (Figure 4.10) show similar trends
in sample distribution to the Central Great Plains and Kansas, with peaks skewed
to the late Holocene as a result of a large quantity of younger dates. However,
with a smaller luminescence sample size (n=39) in the Southern Great Plains,
single older dates show as individual curves, whereas older dates in the Central
Great Plains and Kansas PDDs cluster and show only amalgamated curves due
to the increased sample size and clustering from 1000 years ago to the present.

4.2: PCA Results
PCA is a statistical method typically utilized for data reduction or
summarization. Although I could find no indication that PCA has been used in
dune chronology studies, the method does have application to this type of
research. This section demonstrates how the application of PCA to radiocarbon
and luminescence data offers the opportunity to evaluate large datasets using a
few factors, based upon factor values with an absolute value > 0.40. Based upon
the grouping of the locations into the factors, it may be possible to interpret
spatial relationships within those variables.
The objective of utilizing PCA in this study was to determine if
interregional similarity was present through time in the radiocarbon and
luminescence datasets. To do this, both radiocarbon and luminescence datasets
were separated into localized groupings in Excel. Depending upon available

65

data, in some cases specific dune fields were assigned their own column. When
locations with no data were encountered, they were removed from the dataset to
prevent correlation with other sites that had low probability of activity, as a zero in
the dataset could mean low probability in one particular location, and no data in
another.
In order to compare the subregions as accurately as possible, 100-year
intervals were used to interpolate probability, resulting in 71 time intervals for
each location. Data were divided by locality and PDDs were constructed using
CalPal (v.2013). Probabilities were interpolated at the 100-year interval for each
location, and values were noted at each of the 71 time intervals in the Excel
spreadsheet. Radiocarbon data were divided into nine subregions, while
luminescence data were separated into ten subregions (Figure 4.11).

Table 4.1: Subregions and their abbreviations used in the PCA.

66

4.2.1: Radiocarbon PCA
A total of nine subregions were used for the radiocarbon PCA. The
minimum eigenvalue was set to 0.01 with unlimited factors (Table 4.2). After
examination of eigenvalues and component loadings, and based upon
dimensions one and two having multiple high loadings (> |0.70|), extraction
of three dimensions may potentially draw “MANITOBA” and “OKSWKS” into the
third dimension. As shown in Table 4.3, extraction of three dimensions pulled

Table 4.2: PCA results from SYSTAT 13 for radiocarbon data. Highest
loadings are highlighted.

Latent Roots (Eigenvalues)
1
2
3
4
5
6
7
8
9
3.1461.745 1.1031.0080.8430.430 0.3840.1810.159

Component Loadings
1
2

3

4

5

6

7

8

9

SLM

0.837 -0.377 -0.091 0.145 0.111 0.091 0.042 0.199 -0.255

CLM

0.797 0.173 -0.164 0.098 0.244 -0.119 0.454 -0.084 0.104

NLM

0.797 -0.247 0.154 0.146 0.127 0.409 -0.202 -0.141 0.125

MINOT

0.777 -0.302 0.093 -0.209 -0.302 -0.287 -0.145 0.160 0.183

GBSP

0.713 0.506 0.114 -0.152 -0.272 -0.174 -0.132 -0.218 -0.174

TEXASNM

0.188 0.789 -0.173 -0.256 -0.325 0.326 0.087 0.157 0.041

NSHNECO

0.140 0.660 0.320 0.005 0.616 -0.111 -0.188 0.118 0.022

MANITOBA -0.073 0.094 0.812 0.460 -0.289 0.023 0.168 0.045 -0.003
OKSWKS

-0.046 -0.310 0.480 -0.782 0.148 0.095 0.160 -0.024 -0.042

“MANITOBA” and “OKSWKS” into the third dimension, indicating that substantial
similarity exists over time between the two subregions. Closer analysis shows
that “OKSWKS,” the microregion having the lowest of the high factor scores, is in
fact most unlike the other subregions with regard to the chronology of dune

67

Table 4.3: Rotated loading matrix from SYSTAT 13 for radiocarbon data.
Highest loadings are highlighted, and indicate three groupings are present
in the data set.
Rotated Loading Matrix (VARIMAX, Gamma = 1.000000)
1
2
3
SLM

0.910

-0.122

-0.093

NLM

0.838

0.032

0.130

MINOT

0.834

-0.037

0.081

CLM

0.702

0.369

-0.251

GBSP

0.527

0.706

-0.025

TEXASNM

-0.066

0.766

-0.313

NSHNECO

-0.060

0.718

0.197

MANITOBA

-0.079

0.212

0.789

OKSWKS

0.063

-0.219

0.526

stabilization through time. For example, relatively low probability of stabilization in
the microregion is not reflected in the data set until 1800 years ago. The same
could be said for the “MINOT” data, but higher probability with younger dates in
the dataset aligns well with the other locations conveying high loadings (e.g.,
SLM, NLM).
The importance of the high loadings in the first factor of the rotated matrix
(Table 4.3) grouping into three dimensions indicates that similarity in dune
stabilization occurred across several dune fields at various times throughout the
Holocene. Given these groupings, I named the first factor “Northern US” due to
the grouping of the Great Lakes regions and North Dakota, the second factor
“Central/Southern Plains” due to the similarity and grouping of those subregions,
and the third factor “Limited” due to the minimal amount of data from Manitoba,
Oklahoma, and southwestern Kansas. The factor scores were saved and
displayed on scatterplots (Figures 4.11-4.13) for comparison.

68

Factor 1 / Factor 2 Comparison

4.0
3.4

Factor Scores

2.8
2.2
1.6
1.0
0.4
-0.2
-0.8

Factor 1
Factor 2

-1.4
72
75

63
66
69

48
51

54
57
60

33
36

39
42
45

18
21

24
27
30

3

6

9
12
15

0

-2.0

Age

Figure 4.11: Scatterplot of scores for Factors 1 (Northern US) and 2
(Central/Southern Plains) of the radiocarbon data set.
Factor 1 / Factor 3 Comparison
4.0
3.4

Factor Scores

2.8
2.2
1.6
1.0
0.4
-0.2
-0.8

Factor 1
Factor 3

-1.4
75

60
63

66
69
72

45
48

51
54
57

30
33

36
39
42

24
27

9
12

15
18
21

0

3

6

-2.0

Age

Figure 4.12: Scatterplot of scores for Factors 1 (Northern US) and 3
(Limited) of the radiocarbon data set.
Factor 2 / Factor 3 Comparison

4.0
3.4

Factor Scores

2.8
2.2
1.6
1.0
0.4
-0.2
-0.8

Factor 2
Factor 3

-1.4
72
75

60

63
66
69

45
48

51
54
57

39
42

27

30
33
36

21
24

15
18

0

3

6

9
12

-2.0

Age

Figure 4.13: Scatterplot of scores for Factors 2 (Central/Southern
Plains) and 3 (Limited) of the radiocarbon data set.

69

Interpretation of the scatterplots of factor scores suggests that the most similar
period of stability between the subregions occurred during the early Holocene.
Most notably, however, the plots indicate that for the most part, the subregions
contrasted greatly with regard to intervals of stability through the Holocene.

4.2.2: Luminescence PCA
A total of 10 subregions were used for the luminescence PCA. The
minimum eigenvalue was set to 0.01, with unlimited factors (Table 4.4).

Table 4.4: PCA results from SYSTAT 13 for luminescence data. Highest
loadings are highlighted.
Latent Roots (Eigenvalues)
1
2
3
4
5
6

7

8

9

10

3.460 2.086 1.471 1.098 0.873 0.324 0.284 0.237 0.122 0.046
Component Loadings
1
2

3

4

5

6

7

8

9

10

NSHNECO 0.920 -0.145 0.046 0.051 0.033 -0.254 0.165 -0.123 -0.055 -0.128
OKSWKS

0.788 -0.329 -0.149 0.214 -0.392 -0.083 0.147 -0.010 -0.028 0.141

NLM

0.680 0.410 0.321 0.146 -0.294 0.158 -0.282 -0.230 0.042 -0.010

ABILENE

0.678 0.414 -0.437 -0.181 0.196 -0.030 -0.209 0.155 -0.197 0.017

HUTCH

0.600 -0.589 0.329 0.205 -0.066 0.153 -0.083 0.320 0.057 -0.051

DUNCAN

0.512 0.279 -0.759 0.038 0.147 0.064 0.044 0.016 0.233 -0.014

CLM

0.165 0.822 0.241 0.281 0.050 0.244 0.297 0.071 -0.067 -0.006

SLM

0.261 0.585 0.527 -0.456 -0.058 -0.256 0.011 0.141 0.121 0.036

MANITOBA -0.372 0.322 0.007 0.812 0.079 -0.270 -0.122 0.071 0.022 0.005
TEXASNM

0.464 -0.266 0.364 0.086 0.745 0.031 -0.008 -0.108 0.026 0.072

After examining the eigenvalues and loadings, and based upon the outcomes of
the radiocarbon data set, two dimensions were extracted and VARIMAX rotation
was utilized to determine if the single high loadings in dimensions 3-5 could
potentially load into the second dimension (Table 4.5).

70

Table 4.5: Rotated loading matrix from SYSTAT 13 for luminescence data.
Highest loadings are highlighted, and indicate two groupings are present in
the data set.
Rotated Loading Matrix (VARIMAX, Gamma = 1.000000)
1
2
HUTCH

0.834

-0.106

OKSWKS

0.827

0.214

NSHNECO

0.821

0.440

TEXASNM

0.531

0.069

CLM

-0.366

0.754

ABILENE

0.290

0.739

NLM

0.294

0.737

SLM

-0.146

0.624

DUNCAN

0.239

0.532

MANITOBA

-0.491

0.032

As shown in Table 4.5, all of the subregions had medium to high loadings
(> |0.40-0.99|) on the first two factors; notably, the first three subregions
(HUTCH, OKSWKS, NSHNECO). However, in the first factor of the
luminescence rotation, one negative loading appears in the first dimension for
“MANITOBA.” Upon closer analysis of the data set, “MANITOBA” has a high
probability of activity in the early-to-mid Holocene, and had a probability of zero
in the last 1500 years. This was almost the exact opposite of the other
subregions, and thus “MANITOBA” is similar in magnitude to the top four
loadings in the first dimension, but in an opposite manner, which is represented
by the negative loading.
Based upon the groupings, I named the first factor “Western Plains,” due
to its similarity in late Holocene dune activity to the westernmost dune fields in
the Great Plains. The second factor was named “Eastern Plains/Great Lakes”
due to the similarity in mid-to-late Holocene dune activity on the easternmost
fringes of the Great Plains and along the eastern coast of Lake Michigan. The
71

factor scores were saved and displayed on a scatterplot for interpretation (Figure
4.14).

Factor 1 / Factor 2 Comparison

5
4

Value

3
2
1
0
-1

Factor 1
Factor 2
72
75

66
69

60
63

54
57

48
51

42
45

36
39

30
33

24
27

18
21

6

9
12
15

3

0

-2

AGE

Figure 4.14: Scatterplot of scores for Factors 1
(Western Plains) and 2 (Eastern Plains/Great
Lakes) of the luminescence data set.

Interpretation of the scatterplots of factor scores in Figures 4.11-4.13
suggests that a period of similar stability occurred between the Western Plains
and Eastern Plains/Great Lakes during the late Holocene. Most notably,
however, the plots indicate that for the most part the subregions contrasted
greatly with regard to intervals of stability throughout most of the Holocene. The
high amount of variance in the late Holocene is most likely an example of how
the abundance of younger luminescence dates, as a result of reworked sands,
influences PCA when utilized in chronology studies. For example, both factors
peak late in the Holocene, and progressively flatten as time regresses into the

72

early Holocene (Figure 4.14). Factor 2 (Eastern Plains/Great Lakes), which
includes the eastern Lake Michigan coastal region and the easternmost Great
Plains sites (Duncan and Abilene Dune Fields), shows a period of prolonged
activity into the mid-Holocene which is also demonstrated by the luminescence
PDD from Lake Michigan (Figure 4.8).
In summary, utilizing PCA and a rotated loadings matrix (Tables 4.2-4.5)
for the radiocarbon and luminescence data sets broadly suggests that, to some
extent, subregions were both stable and active during similar intervals of time in
both regions. Further, plotting radiocarbon and luminescence factor scores
(Figures 4.11-4.14) suggests that dune stability and activity are not limited to
regional-scale events, but may be influenced by contrasting drivers at the
microregional level, as indicated in Figure 4.14.

4.3: Time-Slice Map Results
As previously discussed, many PDDs used in regional dune chronology
research often lack a spatial component. PCA allows for interpretation of factor
groupings on an interregional scale over time, but may still lack data for certain
areas. In an effort to display chronological data by location, Halfen and Johnson
(2013) developed 100-year time-slice maps to present radiocarbon and
luminescence dates for the Great Plains dunes for the past 1200 years.
However, time-slice maps have not yet been utilized to display geographic
patterns in dune evolution across multiple regions. This section displays timeslice maps for both the Great Plains and eastern Lake Michigan coastal zone,

73

beginning at 7000 years ago. As demonstrated in Figures 4.15-4.32, presenting
chronological data in a sequence of time-slice maps provides a tool to visualize
the spatial chronology of dune evolution within and between regions.
For this study, a 200-year interval was chosen for two reasons, the first
being similarity in how data are presented. Between a 100-year and 200-year
interval the maps looked almost identical, and as a result the 200-year interval
was chosen in an effort to be more conservative as far as probability is
concerned. The second reason chosen for using a 200-year interval was to
reduce the total number of maps from 70 to 35. In the maps, luminescence dates
are indicated by red squares, whereas radiocarbon dates are indicated by green
circles.

b)

a)

Figure 4.15: Time-slice maps for the Great Plains (GP) and eastern Lake
Michigan (LM) regions. a): map showing distribution of ages between
7000-6800 years ago; b): map showing distribution of ages between 68006600 years ago.

74

Due to a lack of data for the two intervals of time, the time-slice maps in
Figure 4.15 present a limited amount of information about stabilization and
activation events in both regions. Figure 4.15a displays no information about
dune evolution in either region, while Figure 4.15b suggests that at least some of
the dunes in the central and northern Great Plains were stable. However, neither
of the maps display any information for dunes in the Great Lakes region.

b)

a)

Figure 4.16: Time-slice maps for the GP and LM regions. a): distribution of
ages between 6600-6400 years ago; b): distribution of ages between 64006200 years ago.

The map in Figure 4.16a suggests that dunes in Manitoba and
northeastern Colorado were stable from 6600-6400 years ago, while at least
some dunes in southwestern Kansas and northern Texas were active. Figure
4.16b indicates that dunes in northeastern Colorado and central Kansas were
likely stable, while dunes in southwestern Kansas may have been active.

75

Radiocarbon data from northern Lake Michigan suggests that dunes were stable
in that part of the region from 6400-6200 years ago.

b)

a)

Figure 4.17: Time-slice maps for the GP and LM regions. a): distribution of
ages between 6200-6000 years ago; b): distribution of ages between 60005800 years ago.

Data displayed in Figure 4.17a, with one radiocarbon date from northern
Lake Michigan, suggest that a period of stability occurred from 6200-6000 years
ago. Figure 4.17b indicates that dune fields throughout much of the Great Plains
region were likely stable from 6000-5800 years ago, as well as in the southern
Lake Michigan region.
Figure 4.18a suggests that dunes were largely stable in the central and
southern Great Plains, as well as the southern Lake Michigan region.
Luminescence data from northern Texas indicates that dune activity may have
occurred from 5800-5600 years ago in this part of the Plains. In Figure 4.18b,
radiocarbon data from northeastern Colorado and the southern Lake Michigan

76

region indicate that both areas were likely stable from 5600-5400 years ago. A
luminescence date from the Brandon Dune Field in Manitoba suggests that dune
activity occurred there during the same time interval.

b)

a)

Figure 4.18: Time-slice maps for the GP and LM regions. a): distribution of
ages between 5800-5600 years ago; b): distribution of ages between 56005400 years ago.

The map in Figure 4.19a displays radiocarbon data from both regions,
which suggests a similar period of stabilization occurred in those areas from
5400-5200 years ago. In Figure 4.19b, luminescence data from the central and
southern Great Plains indicate that dunes may have been active there, whereas
radiocarbon data from Manitoba suggest an interval of stability occurred in the
northern part of the Plains at the same time. Radiocarbon data from southern
Lake Michigan indicates that a period of stability likely took place from 5200-5000
years ago, while local activation may have been occurring in the northern part of
the Lake Michigan region.

77

a)

b)

Figure 4.19: Time-slice maps for the GP and LM regions. a): distribution of
ages between 5400-5200 years ago; b): distribution of ages between 52005000 years ago.
Figure 4.20a suggests that dunes in the central Great Plains and central
Lake Michigan areas were largely active from 5000-4800 years ago, whereas

a)

b)

Figure 4.20: Time-slice maps for the GP and LM regions. a): distribution of
ages between 5000-4800 years ago; b): distribution of ages between 48004600 years ago.

78

radiocarbon data from northern Lake Michigan indicate that a period of local
stability occurred. Figure 4.20b suggests that dunes were largely stable in the
Great Plains region from 4800-4600 years ago, whereas dunes were likely active
in the northern part of the Lake Michigan region.

b)

a)

Figure 4.21: Time-slice maps for the GP and LM regions. a): distribution of
ages between 4600-4400 years ago; b): distribution of ages between 44004200 years ago.

The map in Figure 4.21a displays limited information about dune evolution
in the Great Plains. However, luminescence data from northern and southern
Lake Michigan suggest that dunes were active in those areas from 4600-4400
years ago. Radiocarbon data from central Lake Michigan indicate that a period of
stability likely occurred there during that same time. Figure 4.21b shows that
dunes were widely stable in the central Great Plains from 4400-4200 years ago,
with localized dune activity in the Duncan Dune Field and western Texas. Dunes
along the central and northern parts of the eastern Lake Michigan coastal zone

79

were likely active, with a radiocarbon date from northern Lake Michigan reflecting
a period of localized stability.
The map in Figure 4.22a suggests that the dunes in central and southern
Great Plains were largely stable from 4200-4000 years ago, with localized
activation in Manitoba, eastern Nebraska, and northern Texas. Activity likely
occurred in northern and southern Lake Michigan, with stability occurring in the
central part of the region.

b)

a)

Figure 4.22: Time-slice maps for the GP and LM regions. a): distribution of
ages between 4200-4000 years ago; b): distribution of ages between 40003800 years ago.

In Figure 4.22b, the maps indicate that the central and southern Great
Plains were largely active with localized dune stability in the Nebraska Sand Hills
and northeastern Colorado. Dune activity likely occurred in the central and
southern Lake Michigan from 4000-3800, whereas localized stability probably
took place in central Lake Michigan.

80

a)

b)

Figure 4.23: Time-slice maps for the GP and LM regions. a): distribution of
ages between 3800-3600 years ago; b): distribution of ages between 36003400 years ago.

The map in Figure 4.23a suggests that dunes in central Lake Michigan as
well as the central and southern Great Plains were active from 3800-3600 years
ago. Radiocarbon data from northern Texas may indicate localized dune stability
during that same time period. Figure 4.23b indicates that dunes were largely
active throughout most of the Great Plains region, with localized stability in
central Kansas and northern Texas. The northern and southern subregions of
Lake Michigan were likely stable from 3600-3400 years ago, with localized
activity likely occurring in the northern part of the Lake Michigan region.
Figure 4.24a suggests that localized stability and activation events
occurred in the Nebraska Sand Hills from 3400-3200 years ago. Localized
stability likely occurred in the northern and central subregions of Lake Michigan,
with localized dune activation occurring in the same subregions. Figure 4.24b

81

a)

b)

Figure 4.24: Time-slice maps for the GP and LM regions. a): distribution of
ages between 3400-3200 years ago; b): distribution of ages between 32003000 years ago.

is the only map that truly suggests that dunes in the Great Lakes and Great
Plains were mostly stable at the same time, with localized dune activation
probably occurring in Manitoba. Radiocarbon data from the Great Lakes region
largely indicates that dunes were stable, with localized dune activation in the
northern part of the region.
The map in Figure 4.25a shows that dunes were likely stable in the
northern and central Great Plains from 3000-2800 years ago, with localized
activity occurring in the Nebraska Sand Hills. Northern Lake Michigan was
probably stable during the same time interval, with dune activity occurring in the
central and southern parts of the region. Figure 4.25b suggests that dune stability
was occurring in northeastern Colorado whereas dunes were active in the
Nebraska Sand Hills. Localized stability and activation was likely occurring in the

82

northern part of the Great Lakes region from 2800-2600 years ago, with
simultaneous activity occurring in the central part of the region.

b)

a)

Figure 4.25: Time-slice maps for the GP and LM regions. a): distribution of
ages between 3000-2800 years ago; b): distribution of ages between 28002600 years ago.

In Figure 4.26a, the map shows that dunes across much of the Great
Plains were likely stable with some localized activation in the central part of the
region. In the Great Lakes, dunes appear to have been active in the northern and
southern subregions, with stability occurring in the central part of the region, as
well as localized stability in the northern part of the Great Lakes region. Figure
4.26b suggests that dunes were largely stable across the Great Plains from
2400-2200 years ago, with localized activation in the central Plains. Dunes were
likely active along most of the Lake Michigan coastal zone, with some localized
stability in the northern part of the region.

83

a)

b)

Figure 4.26: Time-slice maps for the GP and LM regions. a): distribution of
ages between 2600-2400 years ago; b): distribution of ages between 24002200 years ago.

The map in Figure 4.27a suggests that dunes were largely stable in the
northern and central Great Plains, with localized activity in the Nebraska Sand

a)

b)

Figure 4.27: Time-slice maps for the GP and LM regions. a): distribution of
ages between 2200-2000 years ago; b): distribution of ages between 20001800 years ago.
84

Hills and central Kansas. Dunes along the eastern Lake Michigan coastal zone
were largely stable, with localized activity occurring in the central and northern
parts of the region from 2200-2000 years ago. Figure 4.27b shows that dunes
were likely active in both regions, with localized stability in the northern part of
the Great Lakes region from 2000-1800 years ago.

b)

a)

Figure 4.28: Time-slice maps for the GP and LM regions. a): distribution of
ages between 1800-1600 years ago; b): distribution of ages between 16001400 years ago.

Limited data in Figure 4.28a suggest that dunes were stable from 18001600 years ago in the southern Great Plains, as well as the northern part of the
Great Lakes region. However, localized activity appears to have also occurred in
the northern part of the Great Lakes coastal zone. Figure 4.28b indicates that
dunes in both regions were largely stable from 1600-1400 years ago, with some
isolated dune activity. This time interval represents the core of the “Holland

85

Paleosol” interval – a time of regional, widespread dune stability in the Great
Lakes region (Arbogast et al., 2004).

b)

a)

Figure 4.29: Time-slice maps for the GP and LM regions. a): distribution of
ages between 1400-1200 years ago; b): distribution of ages between 12001000 years ago.

The map in Figure 4.29a suggests that dunes in both regions were largely
stable from 1400-1200 years ago. Some localized dune activity likely occurred in
the central and southern Great Plains, as well as in the northern part of the Great
Lakes region. Figure 4.29b shows that dunes were mostly stable throughout the
Great Plains, with localized activation in the central and southern subregions of
the Great Plains from 1200-1000 years ago. Dunes were likely stable throughout
most of the Great Lakes region, with localized activity in the northern part of the
region.

86

a)

b)

Figure 4.30: Time-slice maps for the GP and LM regions. a): distribution of
ages between 1000-800 years ago. b): distribution of ages between 800-600
years ago.

In Figure 4.30a, the map suggests that dunes were stable in the northern
and central Great Plains, with localized activity occurring throughout most of the
central and southern parts of the region. Dunes appear to have been stable
throughout the entirety of the Great Lakes region, with dune activity occurring at
a local level along the eastern Lake Michigan coastal zone from 1000-800 years
ago. Similar dune evolution is displayed in Figure 4.30b, with dune stability likely
occurring across the entirety of both regions. Similar dune activity is revealed in
the central Great Plains and central Great Lakes subregions from 800-600 years
ago. Additionally, the data in Figure 4.30 suggest that the MWP may not have
been an interval of widespread dune activation across both regions, but does
allude to the likelihood of similar dune activity in the eastern Great Plains and
northern Lake Michigan as previously suggested by Arbogast et al. (2011).

87

a)

b)

Figure 4.31: Time-slice maps for the GP and LM regions. a): distribution of
ages between 600-400 years ago; b): distribution of ages between 400-200
years ago.

The map in Figure 4.31a indicates that dunes were largely stable across
both regions, with an abundance of activity in the central Great Plains and
northern Lake Michigan from 600-400 years ago. Figure 4.31b suggests that
dunes were stable in the northern Great Plains and across most of the Great
Lakes region from 400-200 years ago. However, similar dune activation appears
to have occurred in the central Great Plains and in the northern Lake Michigan
region during that time interval.

88

Figure 4.32: Time-slice map for the GP and LM regions, displaying the
distribution of ages between 200 years ago and the present.

In Figure 4.32, dunes in the Great Plains appear to have been largely
active across the region from 200 years ago to the present, with some localized
stability. The Great Lakes region appears to have been stabilized with localized
activity in the northern part of the region from 200 years ago to the present.

4.3.1: Analysis of Time-Slice Maps
Through analysis of the time-slice maps, it appears that several activation
events occurred from 5200-4800 years ago, and from 4400-1800 years ago in
both the Great Plains and Great Lakes regions. The spatial variability in dune
activity has differed extensively throughout the past 7000 years in both regions,
and further suggests that dune activity was normally not constrained to a single
dune field, but that activity was spread across both regions. Similar to the

89

abundance of activation events shown in the time-slice maps for both regions,
the maps also display several stabilization events over the past 7000 years in
both regions. In order to easily compare the time-slice map results, Figure 4.33
displays dune evolution through time in both regions, based upon sample size for
each time interval. Individual rectangles represent a period of time in a particular

Figure 4.33: Summary of time-slice maps, derived from data shown in
Figures 4.15-4.32.

90

location, with color indicating if the subregion or region was mostly active or
mostly stable. If the same number of luminescence and radiocarbon dates
appeared during a particular time interval, then the rectangle was filled with a
wide crosshatch pattern, indicating a mixed signal. If a subregion or region had
no data for a particular time interval, the rectangle was left blank.
Similar to the time-slice maps (Figures 4.15-4.32), Figure 4.33 suggests
that whereas some similarity in dune evolution existed between the two regions,
dune activity and stability mostly contrasted among regions throughout the
Holocene. For example, there are only three intervals of time when both regions
were largely stable: from 6000-5800 years ago, 1600-1400 years ago, and 14001200 years ago. Whereas dune activity occurred subregionally in the Great
Plains and at similar times to dune activity in the Great Lakes, no similar interval
of activity is present across the entirety of both regions.
Figure 4.33 also provides information on the variety of mixed and missing
data from both regions. For example, although there are no samples from the
early Lake Michigan record, a consistent record of dune evolution exists from
6400 years ago to the present. Alternatively, the dune evolution record in the
northern and southern Great Plains is intermittent, with more consistency in
younger dates in the entirety of the region.

91

4.4: Discussion
4.4.1: Advantages to Utilizing Time-Slice Maps in Dune Chronology
Through utilization of time-slice maps and PDDs, dune chronologies in the
Great Plains and eastern Lake Michigan coastal zone have been constructed. In
the context of this research, the most notable advantage of using the time-slice
maps as opposed to PDDs is in the display of dates prior to 1000 years ago. For
example, the luminescence PDD from the Great Plains (Figure 4.2) suggests
little to no dune activity occurred prior to 1000 years ago. However, when
analyzing the time-slice maps (Figures 4.15-4.32), it is evident that significant
dune activation occurred in the Great Plains prior to 1000 years ago. By using
the time-slice maps for both areas in 200-year spans, older dates are equally
represented and not subject to the bias of an abundance of younger dates, as
occurs with the PDDs.
Additionally, the use of time-slice maps allows for the identification of gaps
in spatial data and temporal bias (Halfen and Johnson, 2013). For example, of
the 35 200-year time periods presented in Figures 4.15-4.32, eight periods of
time occur where data are missing for one or both regions: (1) from 7000-6800
years ago in both regions, (2,3) from 6800-6400 years ago (two periods) along
eastern Lake Michigan, (4) from 6200-6000 years ago in the Great Plains, (5)
from 5000-4800 years ago in the Great Plains, (6) from 4800-4600 years ago
along eastern Lake Michigan, (7) from 4600-4400 years ago in the Great Plains,
and (8) 3800-3600 years ago along eastern Lake Michigan. Because activation
or stabilization events must be occurring at any given time in either location, the

92

lack of luminescence or radiocarbon dates suggests that samples bracketed by
the previously mentioned time periods have not yet been collected as part of any
study in one or both respective locations.
In addition, throughout the course of geomorphic study in both regions,
certain dune fields have attracted more research interest than others (e.g.
Nebraska Sand Hills, Great Bend Sand Prairie, Sleeping Bear Dunes), and thus
an inadvertent research bias has developed. In much of the previous dune
chronology literature from both regions, an abundance of dates from a particular
area has led researchers to suggest a higher likelihood for activation/stabilization
events, when in fact this may not actually be the case. For this reason, sample
sizes were left off of the maps shown in Figures 4.15-4.32. Until equal amounts
of chronological data are collected for each dune field in both regions, an
abundance of dates for a particular area only lends weight to increased interest
in a particular area. In this regard, the time-slice maps are helpful in that they
provide geomorphologists the opportunity to visualize where data are missing for
particular time periods in different areas, and the opportunity for possible
development of a research design for obtaining dates from the areas where data
are absent on the maps.

4.4.2: Potential Catalysts for Dune Activation and Stabilization in the Great
Plains and Eastern Lake Michigan Coastal Zone
Interpretation of PDDs (Figures 4.1-4.8) and time-slice maps shown in
Figures 4.15-4.32 show that several periods of similar dune activation and
stabilization have taken place in the Great Plains and along the eastern Lake

93

Michigan coastal zone over the past 7000 years. This history suggests that
dunes in both regions may have responded to both similar and different forcing
variables. Previous dune chronology research in both regions has focused
primarily on specific dune fields (e.g. Ahlbrandt and Fryberger, 1980; Arbogast,
1996; Forman et al., 2001; Holliday, 2001; Arbogast et al., 2002; Hugenholtz and
Wolfe, 2005; Hanson et al., 2009b; Blumer et al., 2012), and has associated
sporadic dune activity in the Great Plains with periods of drought through the
Holocene (e.g. Mason et al., 2004; Cook et al., 2004; Miao et al., 2007). Prior
research conducted in eastern Lake Michigan dune fields has commonly
associated periods of dune activation with high lake levels (e.g. Loope and
Arbogast, 2000; Arbogast et al., 2002; Hansen et al., 2010). These conclusions
suggest that contrasting conditions trigger dune activation events in these
regions. However, a study by Arbogast et al. (2011) showed a similar period of
dune activation occurred in both regions during the Medieval Warm Period
(MWP), which was a period of warmer climate, lower lake levels, and associated
drought from ~1100-800 years ago in North America (Laird et al., 1996; GrissinoMayer, 1996; Schmeider et al., 2011).
Although the Arbogast et al. (2011) study suggests that broad climate
patterns may have concurrently affected dunes in both regions, it has been
shown with the time-slice maps in Figures 4.15-4.32 that the MWP may have
been a geographically isolated dune activation event, as dune fields on the
eastern fringes of the Great Plains (Duncan and Abilene) were concurrently
active with dunes on the eastern Lake Michigan coastal zone during that time.

94

Dune activity during the MWP may have been widespread in the Great Plains,
but it is possible that reworking of sands may have potentially altered the
activation record. Thus, the MWP appears to have been only one of several
intervals of similar dune activation across both regions during the past 7000
years.
Although the MWP is noteworthy because of the severity of drought and
its great extent in the midcontinent, additional work has also been done to
determine when other periods of Holocene drought occurred in the region.
Wetherald et al. (1999) argued that mid-continental warming can lead to
increased evaporation and decreased soil moisture. Forman et al. (2001) and
Booth et al. (2005) suggested that periods of drought might be associated with a
La Niña-influenced climate, and proposed that cool sea surface temperatures
(SST) in the eastern tropical Pacific Ocean, tropical Atlantic Ocean, and Gulf of
Mexico can significantly weaken cyclogenesis over central North America,
initiating both micro and macro scale drought (Booth et al., 2005; Forman et al.,
2008). Feng et al. (2008) proposed that SSTs in the North Atlantic Ocean might
also have an effect on the initiation of drought episodes in the central and
southern Great Plains. Other research has also suggested that drought in the
Great Plains and potentially North America as a whole may be linked to
variations in the Atlantic Multidecadal Oscillation (e.g. Knight et al., 2006), or to
the Pacific Decadal Oscillation (e.g. McCabe et al., 2004).
Significantly less research has been conducted within the Great Lakes
region pertaining to Holocene drought. Booth et al. (2006) examined one peat

95

core from the thumb in Lower Michigan and one from northern Minnesota. The
authors hypothesized that SST anomalies in the North Pacific, Tropical Pacific,
and North Atlantic likely had a combined impact on North American climate over
the past 1000 years (Booth et al., 2006). More recently, Arbogast et al. (personal
communication) found that dune activation and stabilization events in the Great
Lakes may correlate with the 7000-year El Niño record from Peru obtained by
Moy et al. (2002). In order to test this hypothesis and incorporate the lake level
data discussed in previous studies (e.g., Arbogast et al., 2002; Hansen et al.,
2010), Great Lakes OSL and radiocarbon PDDs from Figure 4.6 were placed
above a 4700-year lake level curve from Lake Michigan and the El Niño record
for the past 7000 years (Figure 4.34).

96

Figure 4.34: Great Lakes radiocarbon and luminescence PDDs
combined with El Niño record and Lake Michigan lake level data
(modified from Moy et al., 2002; Baedke and Thompson, 2000;
Arbogast et al., unpublished data). Yellow bars highlight
similar intervals of high lake level, El Niño occurrence, and dune
stability.

As shown in Figure 4.34, dune stabilization appears to have occurred
mostly when both lake levels and the number of El Niño events per century were
high (>5). Conversely, activation events appear to have transpired when both
lake levels and the number of El Niño events per century were low (<5). This
finding assumes that La Niña events were occurring when El Niño events were
not, and that the La Niña events may have caused increased storminess, leading

97

to landscape instability and associated dune activation (Arbogast et al., personal
communication).
Given the findings in the Great Lakes region, a similar figure (Figure 4.35)
was created to compare how stabilization events correlated with El Niño events
and the C4 plant carbon record from buried soils (Nordt et al., 2008) in the Great
Plains. The C4 plant carbon record was utilized because C4 plants thrive in a
semi-arid environment and respond positively to increases in temperature, thus
serving as a climate proxy (von Fischer et al., 2008). Based upon interpretation
of Figure 4.35, correlation between frequent El Niño events and periods of dune
stability in the Great Plains appears to be similar to that of the Great Lakes.
However, the C4 plant carbon record appears to correlate with both periods of
dune stability and activation in the Great Plains, which may be a result of C4
plants responding positively to a variety of climate conditions, including intervals
of time between drought events (Nordt et al., 2008).

98

Figure 4.35: Great Plains radiocarbon and luminescence PDD
combined with El Niño event and buried soil organic carbon
data (modified from Moy et al., 2002; Nordt et al., 2008; Arbogast
et al., unpublished data).

Nevertheless, an association between frequent El Niño events and dune
stabilization events in both the Great Plains and Great Lakes regions appears to
exist, and suggests that broad climate patterns likely affect dune systems in both
regions.

4.5: Conclusions
An abundance of research has been conducted on dune fields in both the
Great Plains and eastern Lake Michigan coastal zone for over a century. More
recently, quantitative methods and data collection have been utilized at a large
scale in an attempt to construct dune chronologies throughout the late
Pleistocene and Holocene through the use of radiocarbon and luminescence

99

dating. Substantial effort has been applied to determining the forcing variables
that cause dune activation and stabilization in both regions, as well as
determining the most effective way to display luminescence and radiocarbon
data obtained as part of recent dune studies in both regions.
As part of this research, three different methods of data interpretation
were utilized. Construction of PDDs, PCA, and time-slice maps displayed dune
chronologies for both regions in very different ways, both graphically and
quantitatively. For time-constrained and regional data where a geographic
component is implied, PDDs provided a suitable method for construction of dune
chronologies. For interregional dune chronology research over a broad period of
time (>1000 years), PCA and time-slice maps were the best methods to both
statistically and spatially display luminescence and radiocarbon data. For
example, given the abundance of younger luminescence dates in the Great
Plains, the PDDs did not have the capacity to pick up ~25% of the luminescence
data set, thereby falsely implying that dunes were not active in the Great Plains
region prior to 1000 years ago. Based upon the interpretation of PCA output
(Table 4.3, 4.5; Figures 4.11-4.14) and time-slice maps (Figures 4.15-4.32), dune
activation events have likely been taking place in both regions from ~4400 years
ago to the present. Given the apparent synchroneity of dune systems in both
regions, the potential capacity of climate factors such as El Niño frequency and
SST fluctuations to have major impacts on the activity of dune systems is a
strong possibility.

100

4.5.1: Contributions of this Research
There are four major contributions as a result of this research. The first
contribution is that this research provides a data set for both luminescence and
radiocarbon dates obtained from sand dunes in both the Great Plains and
eastern Lake Michigan coastal zone. Previous research has provided local and
regional data in one of the areas, but not for both within the same study. The
second contribution of this study is the construction of sand dune chronologies
for both the Great Plains and eastern Lake Michigan coastal zone. Incorporating
PDDs and time-slice maps into a particular study has only been done on a
regional level (e.g. Halfen and Johnson, 2013), but never in an interregional
study. The third contribution of this study is the use of PCA in the investigation of
dune chronology research. PCA introduces a data-reduction component that
allows for the identification of similarities and differences within and across
regions. As a result, corresponding locations are grouped, allowing for a
straightforward analysis of dune evolution. The fourth contribution is represented
in a discussion about how to best utilize both PDDs and time-slice maps on
different scales depending upon specific research design. Briefly discussed as
part of an application in the Great Plains by Halfen and Johnson (2013),
interregional analysis shows the importance of integrating the geographical
component into sand dune chronology research.

101

4.5.2: Future Research
The methods and results of this study provide ample opportunities for
future research. Through the use and subsequent analysis of time-slice maps,
gaps in spatial data and locational bias of existing chronologies have been
identified. Given these tools, future research should focus on filling those “spatial
gaps” to provide a more robust chronological data set, thus improving our
understanding of sand dune activity through the Holocene. Further, the
interregional analysis conducted as part of this study could be extended to
multiple other regions in North America or potentially the world in a variety of
dune systems.

102

APPENDIX

103

Table A.1: Radiocarbon and Luminescence data from the Great Plains.
Lum. Mean
Date

Error
(1σ)

RC Mean
Date

Site

Author(s)

Lab Number

Duncan dune field, NE

Hanson et al. 2009

UNL1468

490

Duncan dune field, NE

Hanson et al. 2009

UNL1338

560

50

Duncan dune field, NE

Hanson et al. 2009

UNL1342

590

50

Duncan dune field, NE

Hanson et al. 2009

UNL1346

670

60

Duncan dune field, NE

Hanson et al. 2009

UNL1347

670

60

50

Duncan dune field, NE

Hanson et al. 2009

UNL1348

690

60

Duncan dune field, NE

Hanson et al. 2009

UNL1349

690

70

Duncan dune field, NE

Hanson et al. 2009

UNL1343

700

60

Duncan dune field, NE

Hanson et al. 2009

UNL1344

720

70

Duncan dune field, NE

Hanson et al. 2009

UNL1345

830

90

Duncan dune field, NE

Hanson et al. 2009

UNL1631

3440

310

Duncan dune field, NE

Hanson et al. 2009

UNL1632

3640

280

Duncan dune field, NE

Hanson et al. 2009

UNL1634

3720

340

Duncan dune field, NE

Hanson et al. 2009

UNL1466

3990

350

Duncan dune field, NE

Hanson et al. 2009

UNL1467

4170

410

Duncan dune field, NE

Hanson et al. 2009

UNL1471

4360

360

Duncan dune field, NE

Hanson et al. 2009

UNL1469

4980

570

Duncan dune field, NE

Hanson et al. 2009

UNL1470

5070

430

Abilene dunes, KS

Hanson et al. 2009

OSL1

860

70

Abilene dunes, KS

Hanson et al. 2009

OSL2

610

40

Abilene dunes, KS

Hanson et al. 2009

OSL3

760

70

Abilene dunes, KS

Hanson et al. 2009

OSL4

780

70

Abilene dunes, KS

Hanson et al. 2009

OSL5

720

60

Abilene dunes, KS

Hanson et al. 2009

OSL6

710

80

Abilene dunes, KS

Hanson et al. 2009

OSL7

760

60

Abilene dunes, KS

Hanson et al. 2009

OSL8

790

100

Abilene dunes, KS

Hanson et al. 2009

OSL9

460

40

Abilene dunes, KS

Hanson et al. 2009

OSL10

640

70

Abilene dunes, KS

Hanson et al. 2009

OSL11

710

60

104

Error
(1σ)

Table A.1: (cont’d)
Site

Author(s)

Lab Number

Lum. Mean
Date

Error
(1σ)

RC Mean
Date

Abilene dunes, KS

Hanson et al. 2009

OSL12

750

80

Abilene dunes, KS

Hanson et al. 2009

OSL13

780

80

Abilene dunes, KS

Hanson et al. 2009

OSL14

820

80

Abilene dunes, KS

Hanson et al. 2009

OSL15

1060

120

Arkansas River, KS

Forman et al. 2008

UIC1407

180

15

Arkansas River, KS

Forman et al. 2008

UIC1449

70

7

Arkansas River, KS

Forman et al. 2008

UIC1448

80

10

Arkansas River, KS

Forman et al. 2008

UIC1446

430

30

Arkansas River, KS

Forman et al. 2008

UIC1445

340

30

Arkansas River, KS

Forman et al. 2008

UIC1411

1490

130

Arkansas River, KS

Forman et al. 2008

UIC2088

420

40

Arkansas River, KS

Forman et al. 2008

UIC2087

380

30

Arkansas River, KS

Forman et al. 2008

UIC1450

370

30

Arkansas River, KS

Forman et al. 2008

UIC2086

65

5

Arkansas River, KS

Forman et al. 2008

UIC2085

190

20

Arkansas River, KS

Forman et al. 2008

UIC1412

320

25

Arkansas River, KS

Forman et al. 2008

UIC1417

6280

670

Arkansas River, KS

Forman et al. 2008

UIC1408

220

20

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC830 (RLS)

430

40

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC856 (RLS)

670

90

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC857 (RLS)

450

50

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC858 (RLS)

100

10

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC859 (RLS)

690

70

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC783 (CLS)

540

40
15

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC784 (CLS)

75

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC785 (CLS)

520

40

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC786 (CLS)

70

10

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC816 (CLS)

520

40

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC817 (CLS)

1430

120

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC827 (CLS)

1250

100

105

Error
(1σ)

Table A.1: (cont’d)
Site

Author(s)

Lab Number

Lum. Mean
Date

Error
(1σ)

RC Mean
Date

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC829 (CLS)

480

40

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC919 (CLS)

40

10

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC918 (CLS)

460

40

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC920 (CLS)

60

10

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC1225 (CLS)

1480

160

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC1226 (CLS)

480

30

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC1227 (CLS)

450

40

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC968 (HVLS)

400

50

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC978 (HVLS)

80

10

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC977 (HVLS)

660

50

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC976 (HVLS)

420

40

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC967 (HVLS)

140

20

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC966 (HVLS)

240

40

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC1222 (HVLS)

1490

160

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC1223 (HVLS)

480

30

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC1224 (HVLS)

70

10

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC986 (BBP)

80

10

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC991 (BBP)

540

40

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC992 (BBP)

420

30

70

10

Nebraska Sand Hills/E Wray, CO

Forman et al. 2005

UIC993 (BBP)

Nebraska Sand Hills

Mason et al. 2004

00RJG1

180

10

Nebraska Sand Hills

Mason et al. 2004

00RJG3

810

60

Nebraska Sand Hills

Mason et al. 2004

00RJG4

860

60

Nebraska Sand Hills

Mason et al. 2004

00RJG6

3900

270

Nebraska Sand Hills

Mason et al. 2004

00RJG10

950

70

Nebraska Sand Hills

Mason et al. 2004

00RJG12

3400

250

Nebraska Sand Hills

Mason et al. 2004

00RJG17

930

70

Nebraska Sand Hills

Mason et al. 2004

00RJG15

2360

160

Nebraska Sand Hills

Mason et al. 2004

00RJG13

3360

230

106

Error
(1σ)

Table A.1: (cont’d)
Site

Author(s)

Lab Number

Lum. Mean
Date
910

Error
(1σ)

RC Mean
Date

Error
(1σ)

Nebraska Sand Hills

Mason et al. 2004

00RJG22

Nebraska Sand Hills

Mason et al. 2004

CURL-5322

0

0

Nebraska Sand Hills

Mason et al. 2004

Beta-4497

1590

70

Nebraska Sand Hills

Mason et al. 2004

CURL-5321

4150

40

Nebraska Sand Hills

Mason et al. 2004

CURL-5323

1380

35

Nebraska Sand Hills

Mason et al. 2004

Beta-50435, ETH-9088

980

55

Nebraska Sand Hills

Mason et al. 2004

Beta-50438, ETH-9089

2910

60

Nebraska Sand Hills

Mason et al. 2004

CURL-5324

2820

35

Minot dune field, ND

Muhs 1997

CAMS-23136

570

60

Minot dune field, ND

Muhs 1997

CAMS-23137

1030

60

Minot dune field, ND

Muhs 1997

CAMS-23138

0

0

Minot dune field, ND

Muhs 1997

CAMS-23139

0

0

Minot dune field, ND

Muhs 1997

CAMS-23140

570

50

Minot dune field, ND

Muhs 1997

CAMS-23141

330

60

Minot dune field, ND

Muhs 1997

CAMS-23142

1260

60

Minot dune field, ND

Muhs 1997

CAMS-23143

540

60

Minot dune field, ND

Muhs 1997

CAMS-23144

290

60

Minot dune field, ND

Muhs 1997

CAMS-24131

170

60

Great Bend Sand Prairie, KS

Arbogast 1996

7980

0

0

Great Bend Sand Prairie, KS

Arbogast 1996

7983

270

80

Great Bend Sand Prairie, KS

Arbogast 1996

7978

380

80

Great Bend Sand Prairie, KS

Arbogast 1996

7982

480

100

Great Bend Sand Prairie, KS

Arbogast 1996

7977

490

80

Great Bend Sand Prairie, KS

Arbogast 1996

7981

550

80

Great Bend Sand Prairie, KS

Arbogast 1996

8119

700

80

Great Bend Sand Prairie, KS

Arbogast 1996

8012

710

80

Great Bend Sand Prairie, KS

Arbogast 1996

9743

810

120

Great Bend Sand Prairie, KS

Arbogast 1996

8214

880

80

Great Bend Sand Prairie, KS

Arbogast 1996

9313

1030

80

107

70

Table A.1: (cont’d)
Site

Author(s)

Lab Number

Lum. Mean
Date

Error
(1σ)

RC Mean
Date

Error
(1σ)

Great Bend Sand Prairie, KS

Arbogast 1996

7777

1090

Great Bend Sand Prairie, KS

Arbogast 1996

8218

1500

120
80

Great Bend Sand Prairie, KS

Arbogast 1996

7979

1500

100

Great Bend Sand Prairie, KS

Arbogast 1996

7998

2310

100

Great Bend Sand Prairie, KS

Arbogast 1996

8216

2400

130

Great Bend Sand Prairie, KS

Arbogast 1996

8314

2730

180

Great Bend Sand Prairie, KS

Arbogast 1996

6745

2940

160

Great Bend Sand Prairie, KS

Arbogast 1996

8003

3220

80

Great Bend Sand Prairie, KS

Arbogast 1996

8215

3280

100

Great Bend Sand Prairie, KS

Arbogast 1996

8221

3820

100

Great Bend Sand Prairie, KS

Arbogast 1996

8011

5370

120

Fort Morgan dune field, CO

Muhs et al. 1997

1

1560

70

Fort Morgan dune field, CO

Muhs et al. 1997

2

2190

70

Fort Morgan dune field, CO

Muhs et al. 1997

3

2160

80

Fort Morgan dune field, CO

Muhs et al. 1997

4

2580

70

3600

70

410

Fort Morgan dune field, CO

Muhs et al. 1997

5

Fort Morgan/Wray/Sterling, CO

Clarke & Rendell 2003

NT02/09

595

100

Fort Morgan/Wray/Sterling, CO

Clarke & Rendell 2003

NT02/10

1065

125

Fort Morgan/Wray/Sterling, CO

Clarke & Rendell 2003

NT02/11

2370

210

Fort Morgan/Wray/Sterling, CO

Clarke & Rendell 2003

NT02/12

805

105

Fort Morgan/Wray/Sterling, CO

Clarke & Rendell 2003

NT02/14

535

115

Fort Morgan/Wray/Sterling, CO

Clarke & Rendell 2003

NT02/15

4850

325

Fort Morgan/Wray/Sterling, CO

Clarke & Rendell 2003

NT02/16

1060

95

Fort Morgan/Wray/Sterling, CO

Clarke & Rendell 2003

NT02/17

370

50

Keenesburg, CO

Forman et al. 1995

GX-15785

5520

Keenesburg, CO

Forman et al. 1995

AA-7017

5010

100

Keenesburg, CO

Forman et al. 1995

GX-15840

4760

305

Keenesburg, CO

Forman et al. 1995

GX-15841

920

260

Northeastern CO

Madole 1995

B70542

810

90

108

Table A.1: (cont’d)
Site

Author(s)

Lab Number

Lum. Mean
Date

Error
(1σ)

RC Mean
Date

Error
(1σ)

Northeastern CO

Madole 1995

B61143

860

90

Northeastern CO

Madole 1995

B52719

910

50

Northeastern CO

Madole 1995

B61144

940

110

Northeastern CO

Madole 1995

B62192

1000

100

Northeastern CO

Madole 1995

B59164

1150

70

Northeastern CO

Madole 1995

B70543

1370

80

Northeastern CO

Madole 1995

B52846

1380

90

Northeastern CO

Madole 1995

B53002

2860

60

Northeastern CO

Madole 1995

B72203

5640

90

Southern KS/Northern OK

Werner et al. 2011

1379

460

40

Southern KS/Northern OK

Werner et al. 2011

1381

500

70

Southern KS/Northern OK

Werner et al. 2011

1378

630

70

Southern KS/Northern OK

Werner et al. 2011

1372

670

130

Southern KS/Northern OK

Werner et al. 2011

1452

720

120

Southern KS/Northern OK

Werner et al. 2011

1453

810

90

Southern KS/Northern OK

Werner et al. 2011

1455

520

50

Southern KS/Northern OK

Werner et al. 2011

1456

630

50

Southern KS/Northern OK

Werner et al. 2011

1380

700

70

Southern KS/Northern OK

Werner et al. 2011

1373

2530

300

Southern KS/Northern OK

Werner et al. 2011

1374

6440

760

Southern KS/Northern OK

Werner et al. 2011

1463

3570

400

Southern KS/Northern OK

Werner et al. 2011

1460

3590

360

Southern KS/Northern OK

Werner et al. 2011

1461

3620

300

Northwestern TX/Eastern NM

Feathers 2003

UW569

2410

130

Northwestern TX/Eastern NM

Feathers 2003

UW570

5640

250

Northwestern TX/Eastern NM

Feathers 2003

UW582

4040

310

Northwestern TX/Eastern NM

Feathers 2003

UW583

6500

570

950

110
2340

40

Northwestern TX/Eastern NM

Feathers 2003

UW588

TX dunes (multiple locations)

Holliday 2001

A7432.1

109

Table A.1: (cont’d)
Site

Author(s)

Lab Number

Lum. Mean
Date

Error
(1σ)

RC Mean
Date

Error
(1σ)

TX dunes (multiple locations)

Holliday 2001

A7435.1

3215

348

TX dunes (multiple locations)

Holliday 2001

A7436

3475

100

TX dunes (multiple locations)

Holliday 2001

A7437

4720

320

TX dunes (multiple locations)

Holliday 2001

A6905

6130

165

TX dunes (multiple locations)

Holliday 2001

A6913

450

30

TX dunes (multiple locations)

Holliday 2001

A6912

755

35

TX dunes (multiple locations)

Holliday 2001

CAMS16006

850

60

TX dunes (multiple locations)

Holliday 2001

A7861

1480

160

TX dunes (multiple locations)

Holliday 2001

A7861.1

1480

60

TX dunes (multiple locations)

Holliday 2001

A7862.1

3890

60

TX dunes (multiple locations)

Holliday 2001

SI4585

4855

90

TX dunes (multiple locations)

Holliday 2001

AA7094

2500

60

TX dunes (multiple locations)

Holliday 2001

AA7095

3800

60

TX dunes (multiple locations)

Holliday 2001

A7445

720

193

TX dunes (multiple locations)

Holliday 2001

A7446

4120

208

TX dunes (multiple locations)

Holliday 2001

A7448.1

5110

388

TX dunes (multiple locations)

Holliday 2001

A7450.1

645

148

TX dunes (multiple locations)

Holliday 2001

A7872

0

0

Lauder Sand Hills, Manitoba

Havholm & Running 2005

2

5240

60

Lauder Sand Hills, Manitoba

Havholm & Running 2005

3

5250

50

Lauder Sand Hills, Manitoba

Havholm & Running 2005

4

5800

50

Lauder Sand Hills, Manitoba

Havholm & Running 2005

5

5790

50

Southwestern Manitoba

Running et al. 2002

111143

2500

40

Southwestern Manitoba

Running et al. 2002

165740

5780

50

Southwestern Manitoba

Running et al. 2002

165741

5760

50

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

WW1048

490

40

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

WW1049

0

0

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

WW1050

0

0

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

WW1044

0

0

110

Table A.1: (cont’d)
Site

Author(s)

Lab Number

Lum. Mean
Date

Error
(1σ)

RC Mean
Date

Error
(1σ)

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

WW1045

0

0

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

WW1046

140

40

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

WW1047

0

0

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

WW1051

0

0

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

WW1913

670

45

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

WW1914

1600

45

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

WW1915

2205

55

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

WW1941

920

150

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

WW1944

4180

75

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

WW1945

2180

55

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

WW1961

1430

60

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

WW1962

2150

60

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

QU1378

2690

170

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

QU1377

2950

160

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

QU1429

1090

90

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

QU1287

1310

330

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

QU1286

1370

100

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

QU155

1510

100

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

QU316

2780

170

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

QU315

4540

250

Brandon Sand Hills, Manitoba

Wolfe et al. 2000

QU314

4560

370

Brandon Sand Hills, Manitoba

Wolfe et al. 2002

O168

0

0

Brandon Sand Hills, Manitoba

Wolfe et al. 2002

O170

0

0

Brandon Sand Hills, Manitoba

Wolfe et al. 2002

O172

0

0

Brandon Sand Hills, Manitoba

Wolfe et al. 2002

O173

5600

270

Brandon Sand Hills, Manitoba

Wolfe et al. 2002

O174

4040

150

Brandon Sand Hills, Manitoba

Wolfe et al. 2002

O175

2050

120

Brandon Sand Hills, Manitoba

Wolfe et al. 2002

O176

3440

150

Brandon Sand Hills, Manitoba

Wolfe et al. 2002

O177

3040

150

111

Table A.1: (cont’d)
Site

Author(s)

Lab Number

Lum. Mean
Date

Error
(1σ)

RC Mean
Date

Hutchinson dune field, KS

Halfen et al. 2012

1874

270

30

Hutchinson dune field, KS

Halfen et al. 2012

1875

320

50

Hutchinson dune field, KS

Halfen et al. 2012

1876

330

40

Hutchinson dune field, KS

Halfen et al. 2012

1877

450

50

Hutchinson dune field, KS

Halfen et al. 2012

1878

320

30

Hutchinson dune field, KS

Halfen et al. 2012

1879

320

30

Hutchinson dune field, KS

Halfen et al. 2012

1880

1150

140

Hutchinson dune field, KS

Halfen et al. 2012

2091

120

10

Hutchinson dune field, KS

Halfen et al. 2012

2092

390

50

Hutchinson dune field, KS

Halfen et al. 2012

1881

290

20

Hutchinson dune field, KS

Halfen et al. 2012

1882

920

80

Hutchinson dune field, KS

Halfen et al. 2012

1883

300

30

Hutchinson dune field, KS

Halfen et al. 2012

2090

350

30

Hutchinson dune field, KS

Halfen et al. 2012

2553

180

10

Hutchinson dune field, KS

Halfen et al. 2012

2554

200

20

Hutchinson dune field, KS

Halfen et al. 2012

2555

170

20

Hutchinson dune field, KS

Halfen et al. 2012

2562

100

10

Hutchinson dune field, KS

Halfen et al. 2012

2563

520

50

Hutchinson dune field, KS

Halfen et al. 2012

2560

80

10

Hutchinson dune field, KS

Halfen et al. 2012

2561

140

30

Hutchinson dune field, KS

Halfen et al. 2012

2558

110

10

Hutchinson dune field, KS

Halfen et al. 2012

2559

220

20

Hutchinson dune field, KS

Halfen et al. 2012

2556

80

80

Hutchinson dune field, KS

Halfen et al. 2012

2557

160

20

Hutchinson dune field, KS

Halfen et al. 2012

2551

180

20

Hutchinson dune field, KS

Halfen et al. 2012

2552

190

20

Hutchinson dune field, KS

Halfen et al. 2012

2686

200

20

Hutchinson dune field, KS

Halfen et al. 2012

2687

240

20

Hutchinson dune field, KS

Halfen et al. 2012

2688

260

30

112

Error
(1σ)

Table A.1: (cont’d)
Site

Author(s)

Lab Number

Lum. Mean
Date

Error
(1σ)

RC Mean
Date

Hutchinson dune field, KS

Halfen et al. 2012

2689

200

20

Hutchinson dune field, KS

Halfen et al. 2012

2692

220

20

Hutchinson dune field, KS

Halfen et al. 2012

2693

240

30

Hutchinson dune field, KS

Halfen et al. 2012

2694

960

80

Hutchinson dune field, KS

Halfen et al. 2012

2695

960

80

Hutchinson dune field, KS

Halfen et al. 2012

2696

100

10

Hutchinson dune field, KS

Halfen et al. 2012

2697

600

50

Hutchinson dune field, KS

Halfen et al. 2012

2700

240

20

Hutchinson dune field, KS

Halfen et al. 2012

2701

920

90

Hutchinson dune field, KS

Halfen et al. 2012

2702

190

20

Hutchinson dune field, KS

Halfen et al. 2012

2703

420

50

Hutchinson dune field, KS

Halfen et al. 2012

2690

220

20

Hutchinson dune field, KS

Halfen et al. 2012

2691

200

20

Hutchinson dune field, KS

Halfen et al. 2012

2698

80

10

Hutchinson dune field, KS

Halfen et al. 2012

2699

190

20

Hutchinson dune field, KS

Halfen et al. 2012

2984

140

10

Hutchinson dune field, KS

Halfen et al. 2012

2985

140

20

Hutchinson dune field, KS

Halfen et al. 2012

2986

220

30

Hutchinson dune field, KS

Halfen et al. 2012

2971

2050

190

Hutchinson dune field, KS

Halfen et al. 2012

2972

2080

200

Hutchinson dune field, KS

Halfen et al. 2012

2974

1880

190

Hutchinson dune field, KS

Halfen et al. 2012

2975

2070

200

Hutchinson dune field, KS

Halfen et al. 2012

2976

210

20

Hutchinson dune field, KS

Halfen et al. 2012

2977

270

50
10

Hutchinson dune field, KS

Halfen et al. 2012

2981

80

Hutchinson dune field, KS

Halfen et al. 2012

2982

160

20

Hutchinson dune field, KS

Halfen et al. 2012

2983

90

10

Hutchinson dune field, KS

Halfen et al. 2012

2987

100

10

Hutchinson dune field, KS

Halfen et al. 2012

2988

200

20

113

Error
(1σ)

Table A.1: (cont’d)
Lum. Mean
Date

Error
(1σ)

RC Mean
Date

Error
(1σ)

Site

Author(s)

Lab Number

Hutchinson dune field, KS

Halfen et al. 2012

2969

110

10

Hutchinson dune field, KS

Halfen et al. 2012

2970

810

70

Southwestern KS

Olson & Porter 2002

92120

5870

60

Southwestern KS

Olson & Porter 2002

73448

3730

90

Southwestern KS

Olson & Porter 2002

75029

1600

80

Nebraska Sand Hills

Mason et al. 2011

1111

Nebraska Sand Hills

Mason et al. 2011

980

Nebraska Sand Hills

Mason et al. 2011

1122

710

90

Nebraska Sand Hills

Mason et al. 2011

1124

2800

200

Nebraska Sand Hills

Mason et al. 2011

1097

540

50

Nebraska Sand Hills

Mason et al. 2011

1099

3000

200

Nebraska Sand Hills

Mason et al. 2011

1103

700

60

Nebraska Sand Hills

Mason et al. 2011

1105

2500

200

Nebraska Sand Hills

Mason et al. 2011

1107

2100

200

Nebraska Sand Hills

Mason et al. 2011

1604

680

50

Nebraska Sand Hills

Mason et al. 2011

1605

740

60

Nebraska Sand Hills

Ahlbrandt & Fryberger 1983

1

3000

400

Nebraska Sand Hills

Ahlbrandt & Fryberger 1983

2

3560

70

Nebraska Sand Hills

Ahlbrandt & Fryberger 1983

3

3810

80

Nebraska Sand Hills

Ahlbrandt & Fryberger 1983

4

4900

500

Nebraska Sand Hills

Ahlbrandt & Fryberger 1983

5

5150

400

Nebraska Sand Hills

Ahlbrandt & Fryberger 1983

6

3600

400

Nebraska Sand Hills

Ahlbrandt & Fryberger 1983

7

3110

80

Nebraska Sand Hills

Ahlbrandt & Fryberger 1983

8

5040

80

Nebraska Sand Hills

Ahlbrandt & Fryberger 1983

9

860

55

Muleshoe dunes, NM/TX

Rich & Stokes 2011

98/2/1

Muleshoe dunes, NM/TX

Rich & Stokes 2011

Muleshoe dunes, NM/TX
Muleshoe dunes, NM/TX

1480

120

490

50

90

20

99/13/3

1300

200

Rich & Stokes 2011

99/13/4

890

90

Rich & Stokes 2011

99/13/5

80

10

114

Table A.1: (cont’d)
RC Mean
Date

Error
(1σ)

Author(s)

Lab Number

Muleshoe dunes, NM/TX

Rich & Stokes 2011

99/13/4

890

90

Muleshoe dunes, NM/TX

Rich & Stokes 2011

99/13/5

80

10

Muleshoe dunes, NM/TX

Rich & Stokes 2011

99/14/3

4000

700

Muleshoe dunes, NM/TX

Rich & Stokes 2011

99/15/2

830

70

Muleshoe dunes, NM/TX

Rich & Stokes 2011

99/15/3

70

10

Muleshoe dunes, NM/TX

Rich & Stokes 2011

99/16/1

1100

100

Muleshoe dunes, NM/TX

Rich & Stokes 2011

99/17/1

1200

100

Lea-Yoakum dunes, NM/TX

Rich & Stokes 2011

99/18/2

3600

400

Lea-Yoakum dunes, NM/TX

Rich & Stokes 2011

99/18/3

110

10

Mescalero dunes, NM/TX

Rich & Stokes 2011

99/19/2A

3900

400

Mescalero dunes, NM/TX

Rich & Stokes 2011

99/19/2B

3700

300

Mescalero dunes, NM/TX

Rich & Stokes 2011

99/19/3

2300

300

Mescalero dunes, NM/TX

Rich & Stokes 2011

99/19/4

70

20

Monahans/Andrews dunes, TX

Rich & Stokes 2011

99/21/2

2000

300

Monahans/Andrews dunes, TX

Rich & Stokes 2011

99/22/1

5100

500

Monahans/Andrews dunes, TX

Rich & Stokes 2011

99/22/2

Monahans/Andrews dunes, TX

Rich & Stokes 2011

803/2

Alfalfa County, OK

Brady 1989

GX-14706

1200

70

Alfalfa County, OK

Brady 1989

GX-14709

6385

285

SE Major Co/NW Kingfisher Co, OK

Lepper & Scott 2005

Beta-131206

1250

40

SE Major Co/NW Kingfisher Co, OK

Lepper & Scott 2005

Beta-131207

1730

40

SE Major Co/NW Kingfisher Co, OK

Lepper & Scott 2005

KL98-06

810

SE Major Co/NW Kingfisher Co, OK

Lepper & Scott 2005

KL98-06-1

800

50

SE Major Co/NW Kingfisher Co, OK

Lepper & Scott 2005

KL99-01

870

50

SE Major Co/NW Kingfisher Co, OK

Lepper & Scott 2005

KL99-01-1

880

50

SE Major Co/NW Kingfisher Co, OK

Lepper & Scott 2005

KL99-02C

830

50

SE Major Co/NW Kingfisher Co, OK

Lepper & Scott 2005

KL99-02B

870

50

115

Lum. Mean Date

Error
(1σ)

Site

70

10

4300

400

40

Table A.1: (cont’d)
RC Mean
Date

Author(s)

Lab Number

SE Major Co/NW Kingfisher Co, OK

Lepper & Scott 2005

KL99-02A

3330

SE Major Co/NW Kingfisher Co, OK

Lepper & Scott 2005

KL99-03B

770

40

SE Major Co/NW Kingfisher Co, OK

Lepper & Scott 2005

KL99-03A

830

50

116

Lum. Mean Date

Error
(1σ)

Site

180

Error
(1σ)

Table A.2: Radiocarbon and Luminescence data from the Eastern Lake Michigan Coastal zone.
Lum. Mean
Date

Error
(1σ)

RC Mean
Date

Error
(1σ)

Site

Author(s)

Lab Number

Northeastern Lake Michigan

Loope & Arbogast 2000

B83880

30

Northeastern Lake Michigan

Loope & Arbogast 2000

B87102

40

60

Northeastern Lake Michigan

Loope & Arbogast 2000

WW2058

60

40

Northeastern Lake Michigan

Loope & Arbogast 2000

B87195

60

60

Northeastern Lake Michigan

Loope & Arbogast 2000

B85491

70

60

Northeastern Lake Michigan

Loope & Arbogast 2000

WW905

80

60

Northeastern Lake Michigan

Loope & Arbogast 2000

B85490

80

60

Northeastern Lake Michigan

Loope & Arbogast 2000

B83235

110

60

Northeastern Lake Michigan

Loope & Arbogast 2000

B106926

120

30

Northeastern Lake Michigan

Loope & Arbogast 2000

B82537

120

70

Northeastern Lake Michigan

Loope & Arbogast 2000

WW1161

120

50

Northeastern Lake Michigan

Loope & Arbogast 2000

WW980

120

50

Northeastern Lake Michigan

Loope & Arbogast 2000

B87104

130

40

Northeastern Lake Michigan

Loope & Arbogast 2000

B83974

150

50

Southern Lake Michigan

Loope & Arbogast 2000

WW985

150

50

Southern Lake Michigan

Loope & Arbogast 2000

WW972

150

50

Southern Lake Michigan

Loope & Arbogast 2000

WW974

160

50

Northeastern Lake Michigan

Loope & Arbogast 2000

WW2057

170

50

Southern Lake Michigan

Loope & Arbogast 2000

WW909

190

60

60

Northeastern Lake Michigan

Loope & Arbogast 2000

B87103

190

60

Southern Lake Michigan

Loope & Arbogast 2000

WW975

200

60

Northeastern Lake Michigan

Loope & Arbogast 2000

B56525

220

50

Northeastern Lake Michigan

Loope & Arbogast 2000

B83877

240

70

Northeastern Lake Michigan

Loope & Arbogast 2000

WW971

240

60

Southern Lake Michigan

Loope & Arbogast 2000

WW982

260

60

Southern Lake Michigan

Loope & Arbogast 2000

WW977

270

50

Northeastern Lake Michigan

Loope & Arbogast 2000

WW1163

270

50

Southern Lake Michigan

Loope & Arbogast 2000

WW897

280

50

Southern Lake Michigan

Loope & Arbogast 2000

WW900

280

60

117

Table A.2: (cont’d)
Lum. Mean
Date

Error
(1σ)

RC Mean
Date

Error
(1σ)

Site

Author(s)

Lab Number

Northeastern Lake Michigan

Loope & Arbogast 2000

B56520

300

Northeastern Lake Michigan

Loope & Arbogast 2000

B83534

310

60

Southern Lake Michigan

Loope & Arbogast 2000

WW981

310

60

60

Northeastern Lake Michigan

Loope & Arbogast 2000

B87196

330

60

Southern Lake Michigan

Loope & Arbogast 2000

WW908

330

60

Southern Lake Michigan

Loope & Arbogast 2000

WW984

340

50

Southern Lake Michigan

Loope & Arbogast 2000

WW983

360

60

Northeastern Lake Michigan

Loope & Arbogast 2000

B56521

390

50

Southern Lake Michigan

Loope & Arbogast 2000

WW898

390

50

Northeastern Lake Michigan

Loope & Arbogast 2000

B85488

420

50

Southern Lake Michigan

Loope & Arbogast 2000

WW902

430

60

Northeastern Lake Michigan

Loope & Arbogast 2000

B83881

480

70

Northeastern Lake Michigan

Loope & Arbogast 2000

B56524

600

60

Southern Lake Michigan

Loope & Arbogast 2000

WW901

670

60

Northeastern Lake Michigan

Loope & Arbogast 2000

WW906

670

60

Northeastern Lake Michigan

Loope & Arbogast 2000

B56522

900

60

Northeastern Lake Michigan

Loope & Arbogast 2000

B83977

920

60

Northeastern Lake Michigan

Loope & Arbogast 2000

B83976

1040

80

Northeastern Lake Michigan

Loope & Arbogast 2000

B85487

1130

90

Southern Lake Michigan

Loope & Arbogast 2000

WW976

1230

60

Southern Lake Michigan

Loope & Arbogast 2000

WW899

1280

60

Northeastern Lake Michigan

Loope & Arbogast 2000

WW2059

1380

40

Northeastern Lake Michigan

Loope & Arbogast 2000

B83236

1480

70

Southern Lake Michigan

Loope & Arbogast 2000

WW907

1660

50

Northeastern Lake Michigan

Loope & Arbogast 2000

WW1065

1890

50

Northeastern Lake Michigan

Loope & Arbogast 2000

WW1066

2029

70

Northeastern Lake Michigan

Loope & Arbogast 2000

B83975

2280

70

Northeastern Lake Michigan

Loope & Arbogast 2000

B85489

2340

70

Northeastern Lake Michigan

Loope & Arbogast 2000

B56523

2460

100

118

Table A.2: (cont’d)
Lum. Mean
Date

Error
(1σ)

RC Mean
Date

Error
(1σ)

Site

Author(s)

Lab Number

Northeastern Lake Michigan

Loope & Arbogast 2000

B83879

2600

Northeastern Lake Michigan

Loope & Arbogast 2000

B85955

2690

50

Southern Lake Michigan

Loope & Arbogast 2000

WW903

2800

60

70

Northeastern Lake Michigan

Loope & Arbogast 2000

B83234

2840

60

Southern Lake Michigan

Loope & Arbogast 2000

TX2890

2890

60

Southern Lake Michigan

Loope & Arbogast 2000

CAMS36652

2920

60

Northeastern Lake Michigan

Loope & Arbogast 2000

B92229

2920

90

Northeastern Lake Michigan

Loope & Arbogast 2000

B83878

3070

80

Northeastern Lake Michigan

Loope & Arbogast 2000

WW2060

3250

50

Southern Lake Michigan

Loope & Arbogast 2000

CAMS39354

3720

50

Southern Lake Michigan

Loope & Arbogast 2000

CAMS39355

3730

50

Northeastern Lake Michigan

Loope & Arbogast 2000

B83233

3820

90

Southern Lake Michigan

Loope & Arbogast 2000

B106929

4030

50

Northeastern Lake Michigan

Loope & Arbogast 2000

WW1068

4250

50

Northeastern Lake Michigan

Loope & Arbogast 2000

WW1064

4380

60

Northeastern Lake Michigan

Loope & Arbogast 2000

WW1067

4500

50

Northeastern Lake Michigan

Loope & Arbogast 2000

B83876

5330

150

Arcadia Dunes

Blumer et al. 2012

UIC2128

410

40

Arcadia Dunes

Blumer et al. 2012

UIC2124

970

10

Arcadia Dunes

Blumer et al. 2012

UIC2121

1765

190

Arcadia Dunes

Blumer et al. 2012

UIC2119

3495

335

Arcadia Dunes

Blumer et al. 2012

UIC2118

3530

300

Arcadia Dunes

Blumer et al. 2012

UIC2122

320

50

Arcadia Dunes

Blumer et al. 2012

UIC2123

630

85

Arcadia Dunes

Blumer et al. 2012

UIC2129

710

80

Arcadia Dunes

Blumer et al. 2012

UIC2125

910

95

Arcadia Dunes

Blumer et al. 2012

UIC2127

4340

380

Arcadia Dunes

Blumer et al. 2012

UIC2120

4500

445

Arcadia Dunes

Blumer et al. 2012

UIC2126

4070

380

119

Table A.2: (cont’d)
Lum. Mean
Date

Error
(1σ)

RC Mean
Date

Error
(1σ)

Site

Author(s)

Lab Number

Holland

Arbogast et al. 2004

B179044

390

Holland

Arbogast et al. 2004

B175383

3090

40

Indiana Dunes Natl Lakeshore

Arbogast et al. 2004

B159504

50

50

Indiana Dunes Natl Lakeshore

Arbogast et al. 2004

B159508

160

40

Indiana Dunes Natl Lakeshore

Arbogast et al. 2004

B159509

240

40

Indiana Dunes Natl Lakeshore

Arbogast et al. 2004

B159506

2070

40

Montague

Arbogast et al. 2004

B172560

420

60

Holland

Hansen et al. 2002

LLAW1

3720

650

Holland

Hansen et al. 2002

LLAW2

4340

570

Holland

Hansen et al. 2002

LLAW3

4990

830

Holland

Hansen et al. 2002

LLAW4

3870

470

Holland

Arbogast et al. 2002

NSRL-10488

1050

65

Holland

Arbogast et al. 2002

NSRL-10489

2980

55

Holland

Arbogast et al. 2002

NSRL-10490

3560

55

Holland

Arbogast et al. 2002

NSRL-10491

3750

55

Holland

Arbogast et al. 2002

NSRL-10347

430

55

Holland

Arbogast et al. 2002

NSRL-10346

4090

55

Holland

Arbogast et al. 2002

NSRL-10345

4840

65

Holland

Arbogast et al. 2002

NSRL-10492

35

45

Holland

Arbogast et al. 2002

NSRL-10493

200

45

Holland

Arbogast et al. 2002

NSRL-10494

310

50

Holland

Arbogast et al. 2002

NSRL-10495

2390

65

Holland

Arbogast et al. 2002

NSRL-10496

3730

55

Holland

Arbogast et al. 2002

B-132389

130

50

Holland

Arbogast et al. 2002

B-132390

320

50

Holland

Arbogast et al. 2002

B-132391

390

40

Holland

Arbogast et al. 2002

B-132392

930

40

Rosy Mound Natural Area

Arbogast & Loope 1999

TX-8608

2890

60

Rosy Mound Natural Area

Arbogast & Loope 1999

NSRL-3518

2920

60

120

40

Table A.2: (cont’d)
Lum. Mean Date

Error
(1σ)

RC Mean
Date

Error
(1σ)

Site

Author(s)

Lab Number

Nugent Quarry

Arbogast & Loope 1999

NSRL-3676

3720

Jackson Quarry

Arbogast & Loope 1999

NSRL-3677

3730

50

Nordhouse Dunes Wilderness Area

Arbogast & Loope 1999

B-106928

4030

50

Northeastern Lake Michigan

Lovis et al. 2012

B-237014

860

40

Northeastern Lake Michigan

Lovis et al. 2012

B-237015

870

40

Northeastern Lake Michigan

Lovis et al. 2012

B-237016

1240

40

Southern Lake Michigan

Lovis et al. 2012

B-238129

820

40

Northeastern Lake Michigan

Lovis et al. 2012

B-40288

2630

70

Northeastern Lake Michigan

Lovis et al. 2012

B-40289

6550

80

Northeastern Lake Michigan

Lovis et al. 2012

B-40290

6430

70

Northeastern Lake Michigan

Lovis et al. 2012

B-40291

5440

70

Northeastern Lake Michigan

Lovis et al. 2012

B-40292

6450

80

Northeastern Lake Michigan

Lovis et al. 2012

B-40304

850

60

Northeastern Lake Michigan

Lovis et al. 2012

B-40305

5380

70

Northeastern Lake Michigan

Lovis et al. 2012

B-251817

190

40

Northeastern Lake Michigan

Lovis et al. 2012

DIC-651

2050

80

Northeastern Lake Michigan

Lovis et al. 2012

DIC-652

1830

120

Northeastern Lake Michigan

Lovis et al. 2012

DIC-653

1620

150

Northeastern Lake Michigan

Lovis et al. 2012

M-2311

870

120

Northeastern Lake Michigan

Lovis et al. 2012

M-2312

1290

130

Northeastern Lake Michigan

Lovis et al. 2012

M-2398

430

100

Northeastern Lake Michigan

Lovis et al. 2012

M-2401

1000

140

Northeastern Lake Michigan

Lovis et al. 2012

M-2405

670

100

Northeastern Lake Michigan

Lovis et al. 2012

M-2406

740

100

Northeastern Lake Michigan

Lovis et al. 2012

N-1268

905

115

Northeastern Lake Michigan

Lovis et al. 2012

M-2059

730

110

Northeastern Lake Michigan

Lovis et al. 2012

M-2060

240

100

50

Northeastern Lake Michigan

Lovis et al. 2012

M-2065

1320

120

Southern Lake Michigan

Lovis et al. 2012

NSRL-3969

3000

50

121

Table A.2: (cont’d)
Lum. Mean
Date

Error
(1σ)

RC Mean
Date

Error
(1σ)

Site

Author(s)

Lab Number

Southern Lake Michigan

Lovis et al. 2012

NSRL-3970

500

50

Southern Lake Michigan

Lovis et al. 2012

NSRL-3965

150

50

Southern Lake Michigan

Lovis et al. 2012

NSRL-3966

2080

40

Southern Lake Michigan

Lovis et al. 2012

NSRL-3967

2830

50

Northeastern Lake Michigan

Lovis et al. 2012

Shfd-06138

Northeastern Lake Michigan

Lovis et al. 2012

Shfd-06139

740

70

Northeastern Lake Michigan

Lovis et al. 2012

Shfd-06140

5150

390

Northeastern Lake Michigan

Lovis et al. 2012

Shfd-06141

1300

110

Northeastern Lake Michigan

Lovis et al. 2012

Shfd-06142

1950

180

Northeastern Lake Michigan

Lovis et al. 2012

Shfd-07001

2150

170

Northeastern Lake Michigan

Lovis et al. 2012

UIC-2134

1950

205

Northeastern Lake Michigan

Lovis et al. 2012

UIC-2135

575

70

Northeastern Lake Michigan

Lovis et al. 2012

UIC-2136

930

90

Northeastern Lake Michigan

Lovis et al. 2012

UIC-2137

1930

225

Northeastern Lake Michigan

Lovis et al. 2012

UIC-2138

3280

265

Northeastern Lake Michigan

Lovis et al. 2012

UIC-2139

3380

300

Northeastern Lake Michigan

Lovis et al. 2012

UIC-2143

3240

260

Northeastern Lake Michigan

Lovis et al. 2012

UIC-2144

2315

220

Northeastern Lake Michigan

Lovis et al. 2012

UIC-2145

2420

240

Northeastern Lake Michigan

Lovis et al. 2012

UIC-2146

3260

305

Northeastern Lake Michigan

Lovis et al. 2012

UIC-2147

3160

280

Northeastern Lake Michigan

Lovis et al. 2012

UIC-2148

2490

230

Northeastern Lake Michigan

Lovis et al. 2012

UIC-2149

2765

225

Northeastern Lake Michigan

Lovis et al. 2012

UIC-2150

540

65

920

90

Northeastern Lake Michigan

Lovis et al. 2012

UIC-2151

910

90

Southern Lake Michigan

Lovis et al. 2012

UIC-2178

920

80

Southern Lake Michigan

Lovis et al. 2012

UIC-2179

870

80

Southern Lake Michigan

Lovis et al. 2012

UIC-2207

2360

260

Southern Lake Michigan

Lovis et al. 2012

UIC-2208

2820

210

122

Table A.2: (cont’d)
Lum. Mean
Date

Error
(1σ)

RC Mean
Date

Error
(1σ)

Site

Author(s)

Lab Number

Southern Lake Michigan

Lovis et al. 2012

UIC-2209

2500

220

Southern Lake Michigan

Lovis et al. 2012

UIC-2227

2480

200

Southern Lake Michigan

Lovis et al. 2012

UIC-2228

1510

130

Hoffmaster State Park

Hansen et al. 2010

Shfd03078

2250

250

Hoffmaster State Park

Hansen et al. 2010

Shfd03079

2830

190

Hoffmaster State Park

Hansen et al. 2010

Shfd03080

2790

260

Hoffmaster State Park

Hansen et al. 2010

Shfd03082

4320

280

Hoffmaster State Park

Hansen et al. 2010

Shfd03083

2200

130

Hoffmaster State Park

Hansen et al. 2010

Shfd03084

1900

130

Hoffmaster State Park

Hansen et al. 2010

Shfd05052

1990

120

Hoffmaster State Park

Hansen et al. 2010

Shfd05053

3710

240

Hoffmaster State Park

Hansen et al. 2010

Shfd05054

2830

200

Hoffmaster State Park

Hansen et al. 2010

Shfd05055

3780

240

Hoffmaster State Park

Hansen et al. 2010

Shfd05056

2800

180

Hoffmaster State Park

Hansen et al. 2010

Shfd05057

2660

170

Hoffmaster State Park

Hansen et al. 2010

Shfd05105

710

80

Hoffmaster State Park

Hansen et al. 2010

Shfd05106

960

140

Hoffmaster State Park

Hansen et al. 2010

Shfd05107

880

120

Hoffmaster State Park

Hansen et al. 2010

Shfd05108

4360

290

Hoffmaster State Park

Hansen et al. 2010

Shfd05109

3290

230

Hoffmaster State Park

Hansen et al. 2010

CAMS-108778

330

35

Hoffmaster State Park

Hansen et al. 2010

CAMS-108779

925

35

Hoffmaster State Park

Hansen et al. 2010

CAMS-108780

915

35

Hoffmaster State Park

Hansen et al. 2010

CAMS-108892

180

40

Hoffmaster State Park

Hansen et al. 2010

CAMS-108895

1580

40

Hoffmaster State Park

Hansen et al. 2010

CAMS-108777

310

40

Hoffmaster State Park

Hansen et al. 2010

CAMS-108896

420

40

Warren Dunes State Park

Hansen et al. 2010

Shfd05058

3850

290

Warren Dunes State Park

Hansen et al. 2010

Shfd05059

2260

150

123

Table A.2: (cont’d)
Lum. Mean
Date

Error
(1σ)

RC Mean
Date

Error
(1σ)

Site

Author(s)

Lab Number

Warren Dunes State Park

Hansen et al. 2010

Shfd05060

Warren Dunes State Park

Hansen et al. 2010

CAMS-156730

290

60

Warren Dunes State Park

Hansen et al. 2010

CAMS-156731

2970

80

Warren Dunes State Park

Hansen et al. 2010

CAMS-108781

865

35

Warren Dunes State Park

Hansen et al. 2010

CAMS-108897

345

40

Warren Dunes State Park

Hansen et al. 2010

CAMS-116835

940

35

Warren Dunes State Park

Hansen et al. 2010

CAMS-116837

685

35

970

40

4030

270

Warren Dunes State Park

Hansen et al. 2010

CAMS-108893

Petoskey State Park

Lepczyk & Arbogast 2005

Beta132995

4620

195

Petoskey State Park

Lepczyk & Arbogast 2005

Beta132996

150

150

Petoskey State Park

Lepczyk & Arbogast 2005

Beta132997

128

123

Petoskey State Park

Lepczyk & Arbogast 2005

Beta132998

150

150

Petoskey State Park

Lepczyk & Arbogast 2005

Beta132999

380

90

Petoskey State Park

Lepczyk & Arbogast 2005

Beta133000

1740

115

Petoskey State Park

Lepczyk & Arbogast 2005

Beta133001

415

100

Petoskey State Park

Lepczyk & Arbogast 2005

PP11/31

2180

240

Petoskey State Park

Lepczyk & Arbogast 2005

Beta133003

2800

60

Petoskey State Park

Lepczyk & Arbogast 2005

PP14/31

1770

100

Petoskey State Park

Lepczyk & Arbogast 2005

PP16/31

1030

100

Petoskey State Park

Lepczyk & Arbogast 2005

Beta1333992

0

0

Van Buren State Park

Van Oort et al. 2001

INSTAAR1

16

30

Van Buren State Park

Van Oort et al. 2001

B2

4620

60

Van Buren State Park

Van Oort et al. 2001

B3

5160

60

Van Buren State Park

Van Oort et al. 2001

B4

0

0

Van Buren State Park

Van Oort et al. 2001

B5

200

50

Van Buren State Park

Van Oort et al. 2001

B6

2090

40

Van Buren State Park

Van Oort et al. 2001

B7

3220

40

Van Buren State Park

Van Oort et al. 2001

B8

3190

50

Van Buren State Park

Van Oort et al. 2001

B9

3800

40

124

Table A.2: (cont’d)
Site

Author(s)

Lab Number

Lum. Mean
Date

Error
(1σ)

RC Mean
Date

Error
(1σ)

Van Buren State Park

Van Oort et al. 2001

B10

4550

80

Van Buren State Park

Van Oort et al. 2001

INSTAAR11

155

30

Van Buren State Park

Van Oort et al. 2001

INSTAAR12

235

35

Van Buren State Park

Van Oort et al. 2001

INSTAAR13

3320

45

Van Buren State Park

Van Oort et al. 2001

INSTAAR14

4890

45

Van Buren State Park

Van Oort et al. 2001

INSTAAR15

300

60

Van Buren State Park

Van Oort et al. 2001

INSTAAR16

5000

40

Sleeping Bear Dunes Natl Lakeshore

Snyder 1985

S1

4559

225

Sleeping Bear Dunes Natl Lakeshore

Snyder 1985

S2

2781

160

Sleeping Bear Dunes Natl Lakeshore

Snyder 1985

S3

688

180

Northeastern Lake Michigan

Arbogast et al. 2012 (unpub)

ISGS163

4400

420

Northeastern Lake Michigan

Arbogast et al. 2012 (unpub)

ISGS160

4040

390

Northeastern Lake Michigan

Arbogast et al. 2012 (unpub)

ISGS158

5050

440

Northeastern Lake Michigan

Arbogast et al. 2012 (unpub)

ISGS157

4520

370

125

BIBLIOGRAPHY

126

BIBLIOGRAPHY

Ahlbrandt, T.S. & S.G. Fryberger. 1980. Geologic and paleoecologic studies of
the Nebraska Sand Hills. U.S. Geological Survey Paper 1120-A, 28 p.
Ahlbrandt, T.S., J.B. Swineheart, & D.G. Maroney. 1983. The dynamic
Holocene dune fields of the Great Plains and Rocky Mountain Basins, U.S.A.
in Brookfield, M.E. & T.S. Ahlbrandt, eds., Eolian Sediments and Processes,
New York, Elsevier, 379-406.
Aitken, M.J. 1998. An Introduction to Optical Dating: The Dating of Quaternary
Sediments by the Use of Photon-Stimulated Luminescence. Oxford University
Press, New York.
Anderton, B. & W.L. Loope. 1995. Buried soils in a perched dune field as
indicators of Late Holocene lake-level change in the Lake Superior Basin.
Quaternary Research, 44, 190-199.
Arbogast, A.F. 1995. Paleoenvironments and desertification on the Great Bend
Sand Prairie in Kansas. Ph.D. diss., Univ. of Kansas.
Arbogast, A.F. 1996. Stratigraphic evidence for late-Holocene aeolian sand
mobilization and soil formation in south-central Kansas, U.S.A. Journal of Arid
Environments, 34, 403-414.
Arbogast, A.F., P. Scull, R.J. Schaetzl, J. Harrison, T.P. Jameson, & S. Crozier.
1997. Concurrent stabilization of some interior dune fields in Michigan.
Physical Geography, 18 (1), 63-79.
Arbogast, A.F. & W.C. Johnson. 1998. Late-Quaternary landscape response to
environmental change in south-central Kansas. Annals of the Association of
American Geographers, 88 (1), 126-145.
Arbogast, A.F. & W. L. Loope. 1999. Maximum-limiting ages of Lake Michigan
coastal dunes: Their correlation with Holocene lake level history. Journal of
Great Lakes Research, 25 (2), 372-382.
Arbogast, A.F., M.D. Van Oort, E.C. Hansen, B.E. Bodenbender, J.P.
Buckingham, & M.J. Ingersoll. 1999. Reconstructing the geomorphic
chronology of massive lake-terrace dunes along Lake Michigan. Abstracts
with Programs, GSA 1999 Annual Meeting, 50.

127

Arbogast, A.F. & D.R. Muhs. 2000. Geochemical and mineralogical evidence
from eolian sediments for northwesterly mid-Holocene paleowinds, central
Kansas, USA. Quaternary International, 67, 107-118.
Arbogast, A.F., E.C. Hansen, & W.L. Loope. 2001. Geomorphic evolution of the
coastal dune landscape along the southeastern shore of Lake Michigan.
Abstracts with Programs, GSA Annual Meeting, A437.
Arbogast, A.F., E.C. Hansen, & M.D. Van Oort. 2002. Reconstructing the
geomorphic evolution of large coastal dunes along the southeastern shore of
Lake Michigan. Geomorphology, 46, 241-255.
Arbogast, A.F., R.J. Schaetzl, J.P. Hupy, & E.C. Hansen. 2004. The Holland
Paleosol: an informal pedostratigraphic unit in the coastal dunes of
southeastern Lake Michigan. Canadian Journal of Earth Science, 41, 13851400.
Arbogast, A.F., A.M. Shortridge, & M.E. Bigsby. 2009. Volumetric estimates of
coastal sand dunes in lower Michigan: Explaining the geography of dune
fields. Physical Geography, 30(6), 479-500.
Arbogast, A.F, G.W. Monaghan, & W.A. Lovis. 2011. Exploring the potential
linkage between coastal sand dune activation and drought episodes in the
northern Lake Michigan basin. Geological Society of America Abstracts with
Programs, 43 (5), 517.
Arbogast, A.F., G.W. Monaghan, W.A. Lovis, H. Wang, & D.M. Kowalski. 2013. A
5.5 ka Record of ENSO Cycling and Coastal Dune Activation in the Great
Lakes (USA). Unpublished data.
Arnold, J.R. & W.F. Libby. 1949. Age determinations by radiocarbon content:
Checks with samples of known age. Science, 110.
Baedke, S.J. & T.A. Thompson. 2000. A 4,700 year record of lake level and
isostasy for Lake Michigan. Journal of Great Lakes Research, 26(4), 416-426.
Bateman, M.D., C.D. Frederick, M.K. Jaiswal, & A.K. Singhvi. 2003.
Investigations into the potential effects of pedoturbation on luminescence
dating. Quaternary Science Reviews, 22, 1169-1176.
Belly, P.V. 1964. Sand movement by wind: U.S. Army Corps of Engineers
Coastal Engineering Research Center, Technical Memo 1, 85p.
Berger, G.W. 1994. Thermoluminescence dating of sediment older than ca. 100
ka. Quaternary Science Reviews (Quaternary Geochronology), 13, 445-456.

128

Bloom, A.L. 1991. Geomorphology: A systematic analysis of Late Cenozoic
landforms. Prentice Hall, Englewood Cliffs, New Jersey. 532p.
Blumer, B.E., A.F. Arbogast, & S.L. Forman. 2012. The OSL chronology of
eolian sand deposition in a perched dune field along the northwestern shore
of Lower Michigan. Quaternary Research, 77, 445-455.
Booth, R.K., S.T. Jackson, S.L. Forman, J.E. Kutzbach, E.A. Bettis III, J. Kreig, &
D.K. Wright. 2005. A severe centennial-scale drought in mid-continental North
America 4200 years ago and apparent global linkages. The Holocene, 15(3),
321-328.
Booth, R.K., M. Notaro, S.T. Jackson, & J.E. Kutzbach. 2006. Widespread
drought episodes in the western Great Lakes region during the past 2000
years: Geographic extent and potential mechanisms. Earth and Planetary
Science Letters, 242, 415-427.
Bowman, S. 1990. Radiocarbon Dating. University of California Press,
Berkeley and Los Angeles, California. 64p.
Brady, R.G. 1989. Geology of the Quaternary dune sands in eastern Major and
southern Alfalfa Counties, Oklahoma. Unpublished Ph.D. dissertation,
Oklahoma State University, 166p.
Breed, C.S. & T. Grow. 1979. Morphology and distribution of dunes in sand
seas observed by remote sensing. In McKee, E.D., ed., A Study of Global
Sand Seas, Chapter J, United States Geological Survey Professional Paper
1052, 253-305.
Bryant, B., L.W. McGrew, & R.A. Wobus. 1981. Geologic map of the Denver 1
x 2˚ quadrangle, north-central Colorado. U.S. Geological Survey Map I-1163.
Buckler, W.R. 1979. Dune Type Inventory and Barrier Dune Classification, A
Study of Michigan’s Lake Michigan Shore. Report of Investigation 23.
Michigan Geological Survey, Michigan Department of Natural Resources
Geological Survey Division, Lansing.
Carlisle, W.J. & R.W. Marrs. 1982. Eolian features of the Southern High Plains
and their relationships to wind flow patterns. Geological Society of America
Special Paper 192, 89-105.
Clarke, M.L., H.M. Rendell, & A.G. Wintle. 1999. Quality assurance in
luminescence dating. Geomorphology, 29, 173-185.

129

Clarke, M.L. & H.M. Rendell. 2003. Late Holocene dune accretion and episodes
of persistent drought in the Great Plains of Northeastern Colorado.
Quaternary Science Reviews, 22, 1051-1058.
Colton, R.B. 1978. Geologic map of the Boulder-Fort Collins-Greeley area,
Colorado. U.S. Geological Survey Map I-855-G.
Cordova, C.E., J.C. Porter, K. Lepper, R. Kalchgruber, & G.F. Scott. 2005.
Preliminary assessment of sand dune stability along a bioclimatic gradient,
north-central and northwestern Oklahoma. Great Plains Research, 15, 227249.
Cowles, H.C. 1899. The ecological relations of the vegetation on the sand
dunes of Lake Michigan, Part I. Geographical relations of the dune floras.
Botanical Gazette, 27, 95-117, 167-202, 281-308, 361-391.
David, P.P. 1968. Geomorphology, stratigraphy, chronology, and migration of
sand dunes in Manitoba and Saskatchewan (62G, 72K). Report of Activities,
Part A, May to October 1967: Geological Society of America, Paper 68-1,
155-157.
David, P.P. 1971. The Brookdale Road section and its significance in the
chronological studies of dune activities in the Brandon Sand Hills of Manitoba.
In Turnock, A.C. (Ed.), Geoscience Studies in Manitoba. Geological
Association of Canada, Special Paper 9, 293-299.
David, P.P. 1977. Sand dune occurrences of Canada: a theme and resource
inventory of eolian landforms in Canada. Indian and Northern Affairs, National
Parks Branch, Contract No. 74-230, 183.
David, P.P. 1979. Sand dunes in Canada. GEOS, Geological Survey of Canada
(Spring Issue), 12-14.
Dorr, J.A. & D.F. Eschman. 1970. Geology of Michigan. University of Michigan
Press, Ann Arbor.
Dow, K.W. 1937. The origin of perched dunes on the Manistee Moraine,
Michigan. Michigan Academy of Science Arts and Letters, 23, 427-440.
Duller, G.A.T. 1996. Recent developments in luminescence dating of
Quaternary sediments. Progress in Physical Geography, 20(2), 127-145.
Elson, J.A. 1960. Surficial geology, Brandon, west of Principal Meridian,
Manitoba: Geological Survey of Canada, Map 1067A. In: Halstead, E.C.
Groundwater Resources of the Brandon Map Area. Geological Survey of
Canada, Memoir 300.

130

Feathers, J.K. 2003. Single-grain OSL dating of sediments from the Southern
High Plains, USA. Quaternary Science Reviews, 22, 1035-1042.
Feng, S., R.J. Oglesby, C.M. Rowe, D.B. Loope, & Q. Hu. 2008. Atlantic and
Pacific SST influences on medieval drought in North America simulated by the
Community Atmospheric Model. Journal of Geophysical Research, 113, 1-14.
Forman, S.L. 1989. Applications and limitations of thermoluminescence to date
Quaternary sediments. Quaternary International, 1, 47-59.
Forman, S.L. & P. Maat. 1990. Stratigraphic evidence for late Quaternary dune
activity near Hudson on the Piedmont of northern Colorado. Geology, 18, 745748.
Forman, S.L., R. Oglesby, V. Markgraf, & T. Stafford. 1995. Paleoclimatic
significance of Late Quaternary eolian deposition on the Piedmont and High
Plains, Central United States. Global and Planetary Change, 11, 35-55.
Forman, S.L., R. Oglesby, & R.S. Webb. 2001. Temporal and spatial patterns of
Holocene dune activity on the Great Plains of North America: Megadroughts
and climate links. Global and Planetary Change, 29, 1-29.
Forman, S.L., L. Marín, J. Pierson, J. Gómez, G.H. Miller, & R.S. Webb. 2005.
Aeolian sand depositional records from western Nebraska: landscape
response to droughts in the past 1500 years. The Holocene, 15 (7), 973-981.
Forman, S.L., L. Marín, J. Gomez, & J. Pierson. 2008. Late Quaternary eolian
sand depositional record for southwestern Kansas: landscape sensitivity to
droughts. Palaeogeography, Palaeoclimatology, Palaeoecology, 265, 107120.
Fritz, S.C., S. Juggins, R.W. Battarbee, & D.R. Engstrom. 1991. Reconstruction
of past changes in salinity and climate using a diatom-based transfer function.
Nature, 352, 706-708.
Frye, J.C. & A.B. Leonard. 1964. Relation of Ogallala Formation to the southern
High Plains in Texas. University of Texas at Austin, Bureau of Economic
Geology, Austin, Texas, United States. 1-25.
Gerety, K.M. & R. Slingerland. 1983. Nature of the saltating population in wind
tunnel experiments with heterogeneous size density sands. In Eolian
Sediments and Landforms, edited by M.E. Brookfield and T.S. Ahlbrandt.
Developments in Sedimentology 38. Elsevier, Amsterdam.
Gile, L.H. 1979. Holocene soils in eolian sediments of Bailey County, Texas.
Soil Science Society of America Journal, 43, 994-1003.

131

Gile, L.H. 1981. Soils and stratigraphy of dunes along a segment of farm road
1731, Bailey County, Texas. ICASALS Publication No. 81-2, Lubbock, Texas.
1-78.
Goble, R.J., J.A. Mason, D.B. Loope, & J.B. Swineheart. 2004. Optical and
radiocarbon ages of stacked paleosols and dune sands in the Nebraska Sand
Hills, USA. Quaternary Science Reviews, 23, 1173-1182.
Goldthwait, J.W. 1908. A reconstruction of water planes of extinct glacial lakes in
the Lake Michigan Basin. Journal of Geology, 16, 459-476.
Greeley, R., S.H. Williams, & J.R. Marshall. 1983. Velocities of windblown
particles in saltation: Preliminary laboratory and field measurements. In Eolian
Sediments and Landforms, edited by M.E. Brookfield and T.S. Ahlbrandt.
Developments in Sedimentology 38. Elsevier, Amsterdam.
Green, F.E. 1961. The Monahans Dunes area. In: Wendorf, F. (ed.),
Paleoecology of the Llano Estacado. Museum of New Mexico Press, Santa
Fe, 22-47.
Grissino-Mayer, H.D. 1996. A 2129-year reconstruction of precipitation for
northwestern New Mexico, U.S.A. In: Dean, J.S., D.M. Meko, & T.W.
Swetnam (Eds.), Tree Rings, Environment, and Humanity. Arizona,
Radiocarbon, Tucson, 191-204.
Hack, J.T. 1941. Dunes of the western Navajo country. Geographical Review,
31, 240-263.
Halfen, A.F., W.C. Johnson, P.R. Hanson, T.L. Woodburn, A.R. Young, & G.A.
Ludvigson. 2012. Activation history of the Hutchinson dunes in east-central
Kansas, USA during the past 2200 years. Aeolian Research, 5, 9-20.
Halfen, A.F. & W.C. Johnson. 2013. A review of Great Plains dune field
chronologies. Aeolian Research, 10, 135-160.
Hansen, E.C., A.F. Arbogast, S. Packman, & B. Hansen. 2002. Post-Nipissing
origin of a backdune complex along the southeastern shore of Lake Michigan.
Physical Geography, 23, 233-244.
Hansen, E.C., T.G. Fisher, A.F. Arbogast, & M. Bateman. 2010. Geomorphic
history of low perched, transgressive dune complexes along the southeastern
shore of Lake Michigan. Aeolian Research, 1, 111-127.

132

Hanson, P.R., A.F. Arbogast, W.C. Johnson, R.M. Joeckel, & A.R. Young.
2009a. Megadroughts and late Holocene dune activation at the eastern
margin of the Great Plains, north-central Kansas, USA. Aeolian Research, 1
(3-4), 101-110.
Hanson, P.R., R.M. Joeckel, A.R. Young, & J. Horn. 2009b. Late Holocene
dune activity in the Eastern Platte River Valley, Nebraska. Geomorphology,
103, 555-561.
Havholm, K.G. & G.L. Running IV. 2005. Stratigraphy, sedimentology, and
environmental significance of late mid-Holocene dunes, Lauder Sand Hills,
Glacial Lake Hind Basin, southwestern Manitoba. Canadian Journal of Earth
Science, 42, 847-863.
Hawley, J.W., G.O. Bachman, & K. Manley. 1976. Quaternary stratigraphy in
the Basin and Range and Great Basin Provinces, New Mexico and western
Texas. In: Mahaney, W.C. (ed.), Quaternary Stratigraphy of North America.
Dowden, Hutchinson and Ross, Inc., Stroudsberg, Pennsylvania, 235-274.
Haworth, E. 1897. Physiography of western Kansas. Kansas Geological Survey
Bulletin, 2, 247-284.
Hay, R. 1893. The geology of the Great Plains. Kansas Academy of Science
Transactions, 13, 3-6.
Hill, D.R. & J.M. Tompkin. 1953. General and engineering geology of the Wray
area, Colorado and Nebraska. Geological Society of America Bulletin, 1001.
Washington, D.C.: U.S. Government Printing Office.
Holliday, V.T. 1989. Middle Holocene drought on the Southern High Plains.
Quaternary Research, 31, 74-82.
Holliday, V.T. 2001. Origin of late Quaternary dune fields on the Southern High
Plains of Texas and New Mexico. GSA Bulletin, 113 (1), 75-87.
Huffington, R.M. & C.C. Albritton. 1941. Quaternary sands on the Southern
High Plains of western Texas. American Journal of Science, 239, 325-338.
Hugenholtz, C.H. & S.A. Wolfe. 2005. Recent stabilization of active sand dunes
on the Canadian prairies and relation to recent climate variations.
Geomorphology, 68, 131-147.
Huntley, D.J., D.I. Godfrey-Smith, & M.L.W. Thewalt. 1985. Optical dating of
sediments. Nature, 313, 105-107.

133

Irwin, J.H. & R.B. Morton. 1969. Hydrogeologic information on the Glorieta
sandstone and the Ogallala formation in the Oklahoma Panhandle and
adjoining areas as related to underground waste disposal. U.S. Geological
Survey Circular, (630). United States Geological Survey.
Jolliffe, I.T. (2002). Principal Component Analysis, Springer Series in Statistics,
2nd ed., Springer, NY, 487 p.
Jorgensen, D.W. 1992. Use of soils to differentiate dune age and to document
spatial variation in eolian activity, northeast Colorado, U.S.A. Journal of Arid
Environments, 23, 19-34.
Keizars, K.Z., B.M. Forrest, & W.J. Rink. 2008. Natural residual
thermoluminescence as a method of analysis of sand transport along the
coast of the St. Joseph Peninsula, Florida. Journal of Coastal Research,
24(2), 500-507.
Kigoshi, K. & Hasegawa, H. 1966. Secular variation of atmospheric radiocarbon
concentration and its dependence on geomagnetism. Journal of Geophysical
Research, 71(4), 1065-1071.
Knight, J.R., C.K. Folland, & A.A. Scaife. 2006. Climate impacts of the Atlantic
multidecadal oscillation. Geophysical Research Letters, 33, LI7706.
Kolstrup, E. 2007. OSL dating in palaeoenvironmental reconstructions. A
discussion from a user’s perspective. Estonian Journal of Earth Sciences, 56
(3), 157-166.
Kudela, K. & P. Bobik. 2004. Long-term variations of geomagnetic rigidity
cutoffs. Solar Physics, 224, 423-431.
Laird, K.R., S.C. Fritz, K.A. Maasch, & B.F. Cumming. 1996. Greater drought
intensity and frequency before AD 1200 in the Northern Great Plains, USA.
Nature, 384, 552-554.
Lepczyk, X.C. & A.F. Arbogast. 2005. Geomorphic history of dunes at
Petoskey State Park, Petoskey, Michigan. Journal of Coastal Research, 21
(2), 231-241.
Lepper, K., & G.F. Scott. 2005. Late Holocene eolian activity in the Cimarron
River valley of west-central Oklahoma. Geomorphology, 70, 42-52.
Lian, O.B. 2007. Luminescence Dating. In: Encyclopedia of Quaternary Science.
Elsevier. 3576p.

134

Livingstone, I. 1986. Geomorphological significance of wind flow patterns over a
Namib linear dune. In Nickling, W.G., ed., Aeolian Geomorphology, Boston,
Allen & Unwin, Inc., 97-112.
Livingstone, I. & A. Warren. 1996. Aeolian Geomorphology: An Introduction.
Addison Wesley Longman, Harlow, England.
Loope, W.L. & A.F. Arbogast. 2000. Dominance of an ~150-year cycle of sandsupply change in Late Holocene dune-building along the eastern shore of
Lake Michigan. Quaternary Research, 54, 414-422.
Lowdon, J.A. & W. Blake, Jr. 1975. Geological Survey of Canada Radiocarbon
Dates XV, GSC Paper 75-7, 15.
Lugn, A.L. 1935. The Pleistocene geology of Nebraska. Nebraska Geological
Survey Bulletin, 10.
Lugn, A.L. 1960. The origin and sources of loess in the Great Plains in North
America: International Geology Congress, 21st, Copenhagen, 1960, Rept., pt.
21, 223-235.
Lugn, A.L. 1962. The origin and sources of loess in the central Great Plains and
adjoining areas of the central lowland: Nebraska Univ. Studies. New ser., no.
26, Pt. XI, 105pp.
Lugn, A.L. 1968. The origin of loesses and their relation to the Great Plains in
North America. In Schultz, C.B. and Frye, J.C. (eds.), Loess and related
eolian deposits of the world: Lincoln, University of Nebraska Press, 139-182.
Machenberg, M.D. 1984. Geology of Monahans Sandhills State Park, Texas.
Bureau of Economic Geology, Austin, Texas. 1-39.
Madole, R.F. 1994. Stratigraphic evidence of desertification in the west-central
Great Plains within the past 1000 yr. Geology, 22, 483-486.
Madole, R.F. 1995. Spatial and temporal patterns of Late Quaternary eolian
deposition, eastern Colorado, USA. Quaternary Science Reviews, 14, 155178.
Mangerud, J. 1972. Radiocarbon dating of marine shells, including a discussion
of apparent ages of recent shells from Norway. Boreas, 1, 143-172.
Mason, J.A., J.B. Swineheart, R.J. Goble, & D.B. Loope. 2004. Late-Holocene
dune activity linked to hydrological drought, Nebraska Sand Hills, USA. The
Holocene, 14 (2), 209-217.

135

Mason, J.A., J.B. Swinehart, P.R. Hanson, D.B. Loope, R.J. Goble, X. Miao, &
R.L. Schmeisser. 2011. Late Pleistocene dune activity in the central Great
Plains, USA. Quaternary Science Reviews, 30, 3858-3870.
Mason, J.P., J.B. Swineheart, & D.B. Loope. 1997. Holocene history of
lacustrine and marsh sediments in a dune-blocked drainage, southwestern
Nebraska Sand Hills, USA. Journal of Paleolimnology, 17, 67-83.
McCabe, G.J., M.A. Palecki, & J.L. Betancourt. 2004. Pacific and Atlantic Ocean
influences on multidecadal drought frequency in the United States.
Proceedings of the National Academy of Science, 101, 4136-4141.
McKee, E.D. 1966. Structures of dunes at White Sands National Monument,
New Mexico, and a comparison with structures of dunes from other selected
areas. Sedimentology, 7, 1-69.
McKee, E.D. 1979. Introduction to a study of global sand seas. In McKee, E.D.,
ed., A Study of Global Sand Seas, Chapter A, United States Geological
Survey Professional Paper 1052, 305-399.
Melton, F.A. 1940. A tentative classification of sand dunes: its application to
dune history in the Southern High Plains. Journal of Geology, 57, 13-145.
Miao, X., J.A. Mason, J.B. Swineheart, D.B. Loope, P.R. Hanson, R.J. Goble, &
X. Liu. 2007. A 10,000-year record of dune activity, dust storms, and drought
in the central Great Plains. Geology, 35, 119-122.
Moore, R.C. 1920. Geology of Kansas. Kansas Geological Survey Bulletin, 6.
Moy, C.M., G.O. Seltzer, D.T. Rodbell, & D.M. Anderson. 2002. Variability of El
Niño/Southern Oscillation activity at millennial timescales during the Holocene
epoch. Nature, 420, 162-165.
Muhs, D.R. 1985. Age and paleoclimatic significance of Holocene sand dunes
in northeastern Colorado. Annals of the Association of American
Geographers, 75 (4), 566-582.
Muhs, D.R. & V.T. Holliday. 1995a. Active dune sand on the Great Plains in the
th
19 century: Evidence from accounts of early explorers. Quaternary
Research, 43 (1), 198-208.
Muhs, D.R., T.W. Stafford, C.A. Bush, Jr., C.E. Hedge, J.B. Swinehart, S.D.
Cowherd, & S.A. Mahan. 1995b. Geochemistry of the Nebraska Sand Hills:
Evidence for mineralogical maturity in a recently active sand sea. Geological
Society of America, Abstracts with Programs, 27 (3), 75.

136

Muhs, D.R., T.W. Stafford, S.D. Cowherd, S.A. Mahan, R. Kihl, P.B. Maat, C.A.
Bush, & J. Nehring. 1996. Origin of the late Quaternary dune fields of
northeastern Colorado. Geomorphology, 17, 129-149.
Muhs, D.R., T.W. Stafford, Jr., J. Been, S.A. Mahan, J. Burdett, G. Skipp, & Z.
Muhs Rowland. 1997a. Holocene eolian activity in the Minot dune field,
North Dakota. Canadian Journal of Earth Science, 34, 1442-1459.
Muhs, D.R., T.W. Stafford, Jr., J.B. Swineheart, S.D. Cowherd, S.A. Mahan, C.A.
Bush, R.F. Madole, & P.B. Maat. 1997b. Late Holocene eolian activity in the
mineralogically mature Nebraska Sand Hills. Quaternary Research, 48, 162176.
Muhs, D.R. & V.T. Holliday. 2001. Origin of late Quaternary dune fields on the
Southern High Plains of Texas and New Mexico. Geological Society of
America Bulletin, 113, 75-87.
Nordt, L., J. Von Fischer, L. Tieszen, J. Tubbs. 2008. Coherent changes in
relative C4 plant productivity and climate during the late Quaternary in the
North American Great Plains. Quaternary Science Reviews, 27, 1600-1611.
Olson, J.S. 1958a. Lake Michigan dune development: Wind-velocity profiles.
Journal of Geology, 66, 254-263.
Olson, J.S. 1958b. Lake Michigan dune development: Plants as agents and
tools in geomorphology. Journal of Geology, 66, 345-351.
Olson, J.S. 1958c. Lake Michigan dune development: Lake level, beach, and
dune oscillations. Journal of Geology, 66, 473-483.
Peterson, J.M. & E. Dersch. 1981. A guide to sand dune and coastal ecosystem
functional relationships: Extension Bulletin, v. E-1529, MICHU-SG-81-501.
Porter, D.A., C.G. Olson, & M.D. Ransom. 1994. Formation and reactivation of
sand dunes within the Cimarron River valley, southwestern Kansas.
Geological Society of America, Abstracts with Programs, 26 (7), A-61.
Pye, K. & H. Tsoar. 2009. Aeolian Sand and Sand Dunes. New York, Springer,
458p.
Reed, E.C. & V.H. Dreeszen. 1965. Revision of the classification of the
Pleistocene deposits of Nebraska. Nebraska Geological Survey Bulletin, 23,
65pp.

137

Reeves, C.C. 1976. Quaternary stratigraphy and geologic history of Southern
High Plains, Texas, and New Mexico. In: Mahaney, W.C. (ed.), Quaternary
Stratigraphy of North America. Dowden, Hutchinson and Ross, Inc.,
Stroudsburg, Pennsylvania, 213-234.
Reimer, P.J., M.G.L. Baillie, E. Bard, A. Bayliss, J.W. Beck, C.J.H. Bertrand, P.G.
Blackwell, C.E. Buck, G.S. Burr, K.B. Cutler, P.E. Damon, R.L. Edwards, R.G.
Fairbanks, M. Friedrich, T.P. Guilderson, A.G. Hogg, K.A. Hughen, B. Kromer,
G. McCormac, S. Manning, C. Bronk Ramsey, R.W. Reimer, S. Remmele,
J.R. Southon, M. Stuiver, S. Talamo, F. W. Taylor, J. van der Plicht, & C.E.
Weyhenmeyer. 2004. IntCal04 terrestrial radiocarbon age calibration, 0-26 cal
kyr BP. Radiocarbon, 3, 1029-1058.
Rhodes, E.J. 2011. Optically Stimulated Luminescence Dating of Sediments
over the past 200,000 years. Annual Review of Earth and Planetary Sciences
(Jeanloz, R. & K.H. Freeman, eds.), 39, 461-488.
Rich, J., & S. Stokes. 2011. A 200,000-year record of late Quaternary Aeolian
sedimentation on the Southern High Plains and nearby Pecos River Valley,
USA. Aeolian Research, 2, 221-240.
Ritter, D.F. 1986. Process Geomorphology. W.C. Brown, Dubuque, Iowa. 579p.
Running, G.L., K.G. Havholm, M. Boyd, & D.J. Wiseman. 2002. Holocene
stratigraphy and geomorphology of Flintstone Hill, Lauder Sandhills, Glacial
Lake Hind Basin, southwestern Manitoba. Géographie physique et
Quaternaire, 56 (2-3), 291-303.
Schaetzl, R. & S. Anderson. 2005. Soils: Genesis and Geomorphology.
Cambridge University Press, New York. 817pp.
Schmeider, J., S.C. Fritz, S.C. Swineheart, A.L.C. Shinneman, A.P. Wolfe, G.
Miller, N. Daniels, K.C. Jacobs, & E.C. Gremm. 2011. A regional-scale climate
reconstruction of the last 4000 years from lakes in the Nebraska Sand Hills,
USA. Quaternary Science Reviews, 30, 1797-1812.
Scott, G.R. 1978. Map showing geology, structure, and oil and gas fields in the
Sterling 1 x 2˚ quadrangle, Colorado, Nebraska, and Kansas. U.S. Geological
Survey Map I-1092.
Scott, I.D. 1942. The dunes of Lake Michigan and correlated problems. 44
Annual Report, Michigan Academy of Science Arts and Letters, 53-61.

th

Simonett, D.S. 1960. Development and grading of dunes in western Kansas.
Annals of the Association of American Geographers, 50, 216-241.

138

Smith, H.T.U. 1937. Quaternary dune building in Kansas. Geological Society of
America Proceedings, 115.
Smith, H.T.U. 1939. Sand dune cycle in western Kansas. Geological Society of
America Bulletin, 50, 1934-1935.
Smith, H.T.U. 1940. Geologic studies in southwestern Kansas. Kansas
Geological Survey Bulletin, 34.
Smith, H.T.U. 1965. Dune morphology and chronology in central and western
Nebraska. Journal of Geology, 73, 557-578.
Smith, H.T.U. 1968. Nebraska dunes compared with those of North Africa and
other regions. In: Schultz, C.B. and J.C. Frye (eds.), Loess and related eolian
deposits of the world. Nebraska University Press, Lincoln, NE. 29-47.
Snyder, F.S. 1985. A spatial and temporal analysis of the Sleeping Bear Dunes
complex, Michigan. (A contribution to the geomorphology of perched dunes in
humid continental regions). Ph.D. diss., Univ. of Pittsburgh.
Stanley, K.O. & W.J. Wayne. 1972. Epeirogenic and climatic controls of early
Pleistocene fluvial sediment dispersal in Nebraska. Geological Society of
America Bulletin, 83 (12), 3675-3690.
Stokes, S. & J.B. Swineheart. 1997. Middle- and late-Holocene dune
reactivation in the Nebraska Sand Hills, USA. The Holocene, 7 (3), 263-272.
Stokes, S. 1999. Luminescence dating applications in geomorphological
research. Geomorphology, 29, 153-171.
Stuiver, M. & T.F. Braziunas. 1993. Modeling atmospheric 14C influences and
14C ages of marine samples to 10000 BC. Radiocarbon, 35(1), 137-191.
Stuiver, M. & G.W. Pearson. 1993. High-precision bidecadal calibration of the
radiocarbon time scale, AD 1950-500 BC and 2500-6000 BC. Radiocarbon,
35 (1), 1-25.
Suess, H.E. 1967. In Radioactive Dating and Methods of Low-level Counting,
143, I.A.E.A., Vienna, Austria.
Swineheart, J.B. 1990. Wind-blown deposits. In: Bleed, A. and C. Flowerday
(eds.), An Atlas of the Sandhills, Conservation and Survey Division, Institute
of Agriculture and Natural Resources, Resource Atlas No. 5a, University of
Nebraska-Lincoln. 43-56.

139

Tague, G.C. 1946. The Postglacial Geology of the Grand Marais Embayment,
Berrien County, Michigan. Publication 45, Geological Series No. 38. Michigan
Geological Survey, Lansing.
Taylor, R.E. 1987. Radiocarbon Dating: An archaeological perspective.
Academic Press, Orlando.
Thompson, T.A. & S.J. Baedke. 1999. Strandplain evidence for reconstructing
late Holocene lake level in the Lake Michigan basin. In: “Proceedings of the
Great Lakes Paleo-Levels Workshop: The Last 4000 Years” (C.E. Sellinger
and F.H. Quinn, Eds.), NOAA Technical Memorandum ERL GLERL-113, 3034.
Thompson, T.A., S.J. Baedke, & J.W. Johnston. 2004. Geomorphic expression
of late Holocene lake levels and paleowinds in the upper Great Lakes. In:
Hansen, E.C., (ed.), The geology and geomorphology of Lake Michigan’s
coast. Michigan Academician, 35, 355–371.
Thorp, J. & H.T.U. Smith. 1952. Pleistocene eolian deposits of the United
States, Alaska, and parts of Canada. Geological Society of America map,
scale 1:2,500,000. New York.
Trimble, D.E. & M.N. Machette. 1979. Geologic map of the greater Denver
area, Front Range urban corridor, Colorado. U.S. Geological Survey Map I856-H.
Van Oort, M.D., A.F. Arbogast, & E.C. Hansen. 2000. Evolution of a massive
coastal dune along Lake Michigan: Lake level and regional correlations.
Abstracts with Programs, GSA Annual Meeting, 32, A-22.
Van Oort, M., A. Arbogast, E.C. Hansen, & B. Hansen. 2001. Geomorphological
history of massive parabolic dunes, Van Buren State Park, Van Buren county,
Michigan. Michigan Academician XXXIII, 175-188.
von Fischer, J.C., L.L. Tieszen, & D.S. Schimel. 2008. Climate controls on C3
vs. C4 productivity in North American grasslands from carbon isotope
composition of soil organic matter. Global Change Biology, 14, 1-15.
de Vries, H. 1958. Atom Bomb Effect: variations of radiocarbon in plants, shells,
snails in the past four years. Science, 128, 250-251.
Wang, X.L., Y.C. Lu, & A.G. Wintle. 2006. Recuperated OSL dating of finegrained quartz in Chinese loess. Quaternary Geochronology, 1, 89-100.
Warren, A. 1968. Dune volume and trend measurements in the Nebraska Sand
Hills. Nebraska Academy of Science Proc., 78, 23.

140

Warren, A. 1976. Morphology and sediments of the Nebraska Sand Hills in
relation to Pleistocene winds and the development of aeolian bed forms.
Journal of Geology, 84, 685-700.
Watanuki, T., A.S. Murray, & S. Tsukamoto. 2005. Quartz and polymineral
luminescence dating of Japanese loess over the last 0.6 Ma: Comparison with
an independent chronology. Earth and Planetary Science Letters, 240, 774789.
Wendorf, F., A.D. Kreiger, & C.C. Albritton. 1955. The Midland discovery.
University of Texas Press, Austin, Texas. 1-139.
Weninger, B., O. Jöris, & U. Danzeglocke. 2013. CalPal-2007. Cologne
Radiocarbon Calibration & Palaeoclimate Research Package.
http://www.calpal.de/, downloaded 2013-01-28.
Werner, C.M., J.A. Mason, & P.R. Hanson. 2011. Non-linear connections
between dune activity and climate in the High Plains, Kansas and Oklahoma,
USA. Quaternary Research, 75, 267-277.
Wetherald, R. & S. Manabe. 1999. Detectability of summer dryness caused by
greenhouse warming. Climate Change, 43, 495-511.
Williams, A.N. 2012. The use of summed radiocarbon probability distributions in
archaeology: a review of methods. Journal of Archaeological Science, 39,
578-589.
Williams, L. 1964. Regionalization of freeze-thaw activity. Annals of the
Association of American Geographers, 54, 597-611.
Wintle, A.G. 1993. Luminescence dating of aeolian sands: an overview. In:
Pye, K., ed., The Dynamics and Environmental Context of Aeolian
Sedimentary Systems. Geological Society of London Special Publication, No.
72, London. 49-58.
Wolfe, S.A., D.R. Muhs, P.P. David, & J.P. McGeehin. 2000. Chronology and
geochemistry of late Holocene eolian deposits in the Brandon Sand Hills,
Manitoba, Canada. Quaternary International, 67, 61-74.
Wolfe, S.A., D.J. Huntley, & J. Ollerhead. 2002. Optical dating of modern and
late Holocene dune sands in the Brandon Sand Hills, southwestern Manitoba.
Géographie physique et Quaternaire, 56 (2-3), 203-214.
Woodruff, N.P. & F.H. Siddoway. 1965. A wind erosion equation. Soil Science
Society of America Proceedings, 29, 602-608.

141

Wright, H.E., J.C. Almendinger, & J. Gruger. 1985. Pollen diagram from the
Nebraska Sandhills and the age of the dunes. Quaternary Research, 24, 115120.

142