Evidence of sequence and age of ancestral Lake Erie lake-levels, Northwest Ohio.
Fisher, Timothy G. ; Blockland, Joseph D. ; Anderson, Brad 等
ABSTRACT. While the general scheme of the retreat of the Late
Wisconsinan glacier from the ancestral Lake Erie basin is understood,
the sequence and timing of those movements that caused lake-level
changes are not well documented. Ground penetrating radar (GPR),
electrical resistivity (ER), and optically stimulated luminescence (OSL)
techniques were used to analyze beach and sand dune formations in the
Wauseon area in northwest Ohio. The beaches showed no evidence of being
flooded after deposition. Thirteen (13) new OSL ages revealed a short
time of approximately 1000 years (16.9-15.9 ka) for formation of the
Maumee, Arkona and Whittlesey shorelines. Evidence supporting the
Ypsilanti lowstand in the ancestral Lake Erie basin was not observed
within two sequences of glacial lacustrine sediments. The results of OSL
dating of sand dunes indicate a period of older activity (~14-15 ka) and
confirmation of previous work that documented activity during the
Younger Dryas cold period (~13-11.5 ka).
Date of publication: 17 November 2015
INTRODUCTION
The ages and evolution of proglacial lakes in the Lake Erie Basin
(Figs. 1, 2) are known in general, but poorly understood in detail. A
common element in most reconstructions is cyclic lake levels that are
controlled by fluctuating ice margins opening and closing poorly
constrained and undated outlets. Large vertical variability of lake
level through the history of ancestral Lake Erie (ALE) is referred to
here as the 'cyclic' model (black line, Fig. 3).
Alternatively, from detailed sedimentologic and geomorphic work on a
sequence of beaches in northeastern Ohio, Totten (1985) proposed a
regular sequence of lake-level fall, also known as the
'episodic' model (gray line, Fig. 3). The chronologies for
ice-margin positions and for occupation of shorelines are based on
little age data.
The cyclic model was championed by Leverett and Taylor (1915), who
noted that some beaches appeared "washed", implying
submergence after deposition (i.e., beaches were transgressed over). In
this case, a higher-elevation beach may be younger than a
lower-elevation beach. Fine-grained sediments found on some beaches also
have been described and used as evidence for submergence. However, the
lower-elevation outlets required to produce early lake-level low phases
have not always been observed. This lack of direct evidence has
sometimes been explained by subsequent covering of those low outlets by
a glacial readvance (Leverett and Taylor 1915). Totten (1985) suggested
that leaching of clay from weathering of shale clasts may explain the
presence of some fine-grained sediment within the beach deposits. Others
(e.g., Bleuerand Moore 1971) have commented that washed beaches may
instead be offshore bars.
Past work proposing the cyclic model of fluctuating water levels in
the Erie basin also includes a major lake-level fall referred to as the
"Ypsilanti lowstand." The lowstand requires recession of the
glacier from the Lake Erie Basin for lake levels to drop at least 100 m
below the Maumee and Arkona shorelines (Fig. 3). Following a readvance
of ice into the Lake Erie Basin (Port Huron stade), lake level then rose
to the Whittlesey Beach, submerging the Arkona Beach (Figs. 2, 3).
Ice-margin recession in the southern Great Lakes and concomitant
development of ALE has received little attention since the publication
of The Quaternary Evolution of the Great Lakes (Karrow and Calkin 1985).
Additional reviews with little new data include those by Barnett (1992),
Lewis et al. (1994), and Herdendorf (2013). The most recent work has
focused on the age of the Warren Beach in northwest Ohio (Campbell et
al. 2011; Higley et al. 2014). A general lack of age control and
detailed examination of shoreline geomorphology and stratigraphy has
hampered more detailed reconstructions of lake history.
[FIGURE 1 OMITTED]
One reason for reconstructing ice-margin recession and glacial-lake
evolution in the Lake Erie basin is to determine whether there is
synchronicity or a lead-lag relationship with Greenland ice-core climate
records (Rasmussen et al. 2006). While aeolian activity in northwest
Ohio appears to be synchronous with short-lived Greenland stadials
(Campbell et al. 2011), not enough data are currently available to
determine if ice-margin or lake-level changes are similarly synchronous.
The primary purpose of this research is to first test the cyclic
model of the Maumee water levels by comparing a "washed" beach
to an unwashed beach--specifically, the rise in water level from Maumee
II (the washed beach) to Maumee III (Fig. 3). The study took place in
the Wauseon area in northwest Ohio and relies on geomorphological and
geophysical investigations. In addition, optically stimulated
luminescence dating (OSL) was used to date sand dunes associated with
the Maumee beaches.
This paper also reports on preliminary testing for the Ypsilanti
lowstand; the "100m lake-level fall with a subsequent
transgression. The result of the Ypsilanti lowstand would have been
subaerial exposure of the landscape below the Maumee beaches before
being flooded at elevations below the Whittlesey beach (Fig. 3). The
cyclic model is tested by stratigraphically examining two sedimentary
sequences below the Warren Beach, which would have been subaerially
exposed during the Ypsilanti lowstand before being resubmerged.
Study area
The Huron-Erie Lake Plain in northwest Ohio (Fig. 1) was glaciated
repeatedly during the Pleistocene (Mickelson et al. 1983) and was last
covered by the Laurentide Ice Sheet (LIS) during the Late Wisconsinan
Nissourian and Port Bruce Stadials (Calkin and Feenstra 1985, Fig. 2).
During the Nissourian Stadial when the glacier reached its maximum
position in Ohio, a loamy, stony till was deposited and is commonly
referred to as the "Early Woodfordian Drift" (Fig. 2). Its
upper surface is typically armored by boulders and overlain by clay-rich
till generally lacking a high concentration of boulders and cobbles
(Forsyth 1960). The clay-rich till deposited during the Port Bruce
Stadial is observed in sediment exposures throughout the Fduron-Erie
Lake Plain (Forsyth I960) and is referred to as the "Late
Woodfordian Drift" in Ohio (Szabo et al. 2011, Fig. 2). The
Nissourian and Port Bruce Stadials are separated by the Erie Interstade,
explained by the retreat of the glacier northeast of Ohio (Morner and
Dreimanis 1973) and the development of Lake Leverett, a proglacial lake
that predates the sequence of the proglacial lakes discussed here. The
Erie Lobe may have retreated east as far as the Niagara Escarpment
(Dreimanis and Goldthwait 1973). A detailed 1:24,000 surficial geology
map of most of the study area is available (Blockland and Fisher 2015).
[FIGURE 3 OMITTED]
The major phases of ALE were produced as glacial meltwater and
terrestrial runoff ponded between the Fort Wayne Moraine to the west and
receding ice margin to the east. Each phase of ALE is referred to as a
discrete lake, with the oldest being Lake Maumee (Figs. 2, 3). Four
possible levels have been proposed for Lake Maumee between ca.
14,400-13,700 [sup.14]Cyrs B.P. (ca. 17, 500-16,900 cal yrs B.P., Fig.
3), but the sequential order in which they formed is unclear (Calkin
andFeenstra 1985). Forsyth (1959) and earlier workers used the names
Lake Maumee I (ca. 244 m elevation), II (ca. 232 m), III (ca. 238 m),
and IV (reoccupation of the ca. 232 m level) to describe the cyclic
changes in the Maumee levels inferring the sequential order in which
they formed (Fig. 3). Because the sequence of the highstands is unclear,
here they are referred to as upper (ca. 244 m), middle (ca. 238 m) and
lower (ca. 232 m). The level examined in this research (lower Maumee;
II) is interpreted to have been submerged by middle Maumee (Leverett and
Taylor 1915) based on its drowned appearance (Forsyth 1959), and because
Lake Maumee II was reportedly too low to drain through the higher Fort
Wayne and Imlay outlets. However, Bleuer and Moore (1971) later
demonstrated that the Fort Wayne outlet was low enough to have drained
all levels of Lake Maumee, eliminating the need for a buried outlet.
With continued eastward retreat of the Erie Lobe, ice either paused
or readvanced to construct the Defiance Moraine (Fig. 1). However, its
formation and timing remain uncertain, with Howard (2010) suggesting
that the WNW margin of the Defiance Moraine north of the Maumee River
was deposited by the Huron Lobe section of the undifferentiated
Huron-Erie Lobe. The strandlines under study are found on the eastern
side of the Defiance Moraine.
During the Mackinaw Interstade, a post-Arkona, pre-Whittlesey
low-water lake phase, known as Lake Ypsilanti (Figs. 2, 3) formed due to
eastward drainage into the Lake Ontario basin. The case for the
Ypsilanti lowstand is summarized by Calkin and Feenstra (1985), but
evidence supporting it was not observed in northeast Ohio by Totten
(1985). Here, we evaluate additional evidence related to a possible
Ypsilanti lowstand in northwest Ohio.
MATERIALS AND METHODS
Ground penetrating radar (GPR)
A Sensors and Software pulseEKKO 100 GPR unit was used to collect
continuous, high-resolution imagery of primarily sandy, near-surface
sediments. Data collected with both 100-MHz and 50-MHz antennas were
processed with the EKKO View and EKKO View Deluxe programs. Data
processing followed procedures in Higley et al. (2014); for more
information on GPR methodology see Jol and Bristow (2003).
Electrical resistivity (ER) & Schlumberger sounding
To image through clay-rich sediment, an Advanced Geoscience
Incorporated (AGI) Supersting R1/IP resistivity unit with a 28-channel
Swift automatic multi-electrode cable was used to constrain sediment
stratigraphy in the absence of natural exposures. The system was
deployed in the dipole-dipole configuration (see Sheriff 1989, Fig. 9.5)
with an electrode spacing of 3.0 m. One feature of AGI's automatic
switching system is its use of all available larger a-spacings, which
allows deeper penetration. Measurements were inverted using AGI's
Earthlmager 2D software (Version 2.4.0; 2004; Interpex 2010).Auger and
drill borings, and shallow pits allowed us to directly observe sediment
to verify geophysical interpretations for both the GPR and ER. The same
system was used to take Schlumberger soundings.
Outcrop and Lab
The stratigraphy at sites in the Wildwood Metropark (WM) and Oak
Openings Metropark (OM) (Fig. 1B and 8) was logged in the field.
Vertical slices of the exposed sediment were cut from the face,
supported in PVC tubes cut lengthwise, and then examined in the lab.
After partial drying, the cores were photographed and magnified so that
the number of rhythmites could be counted. Particle size was measured by
laser diffraction with a Malvern Mastersizer 2000. The loss-on-ignition
technique to determine organic and carbonate content followed procedures
outlined by Heiri et al. (2001). A Bartington MS2E surface scanning
sensor was used to determine magnetic susceptibility of cores at
one-half cm intervals. Hand augering was used to 3.7 m depth where
diamicton was observed at stream level in WM. At the OM site,
vibracoring (Fisher 2004) was used to collect cores of the rhythmites
below stream level. Basal refusal of the vibracore was assumed to
indicate the contact between diamicton and lake sediments as determined
from hand augering.
LiDAR
The high resolution light detecting and ranging (LiDAR) data for
digital elevation models (DEM) is from the Ohio Geographically
Referenced Information Program (OGRIP). DEMs were given 5X vertical
exaggeration and hillshaded using Natural Scene Designer software, and
lit from 315[degrees] declination at 25[degrees] altitude.
Optically stimulated luminescence
Optically stimulated luminescence (OSL) dating followed the
procedures of Aitken (1998). Field samples were collected from hand-dug
soil pits along the crest of sand dunes or hummocks as shown on LiDAR
images. Samples were processed at the Luminescence Geochronology
Laboratory at the University of Nebraska-Lincoln. For details of
sampling and sample preparation see supplementary information from
Campbell et al. (2011). OSL samples were analyzed on Riso Automated OSL
Dating System Models TL/ OSL-DA-15B/C and TL/OSL-DA-20, equipped with
blue and infrared diodes, using the single aliquot regenerative dose
(SAR) technique (Murray and Wintle 2000) for determination of the
equivalent dose (De) with early background subtraction used (Ballarini
et al. 2007; Cunningham and Wallinga 2010). Preheat (240[degrees]C/10s)
and cutheat (220[degrees]C/0s) temperatures were based upon preheat
plateau tests between 180[degrees] and 280[degrees]C. Dose-recovery and
thermal transfer tests were conducted (Murray and Wintle 2003). Growth
curves were examined to determine whether the samples were below
saturation (D/[D.sub.o] < 2; Wintle and Murray 2006). Equivalent dose
distributions are based upon a minimum of 50 aliquots (Rodnight2008).
Individual aliquots were monitored for insufficient count-rate, poor
quality fits (i.e., large error in the equivalent dose, [D.sub.e]), poor
recycling ratio, strong medium vs. fast component (Durcan and Duller
2011), and detectable feldspar. Aliquots deemed unacceptable based upon
these criteria were discarded from the data set prior to averaging.
Calculation of sample [D.sub.e] values was carried out using the Central
Age Model (Galbraith et al. 1999). However, the D distribution for
UNL3637 (asymmetric distribution; decision table of Bailey and Arnold
2006), indicated that the Minimum Age Model (Galbraith et al. 1999) was
more appropriate. Dose rate was calculated by high-resolution gamma
spectroscopy, and cosmic component of the dose rate following Prescott
and Hutton (1994).
RESULTS AND INTERPRETATION
Geomorphology
Winameg ridge. The lineation that includes Winameg and Lyle's
ridges (Fig. 1D) was most recently mapped as a washed beach on the lower
Maumee shoreline by Forsyth (1959). Winameg ridge is ca. 4 km long and
ca. 200 m wide for most of its length, except at its bulbous southern
end where the spit platform was mined for aggregate (Fig. 4A). The land
surface west of the ridge is generally lower by ~5 m and has hummocky
ridges with <2 m relief. Earlier excavations on the west side of the
ridge revealed a mastodon within pond sediments, but no age control was
available for the site (Camp 1981). Further west, parabolic sand dunes
and thin sheet sands mask the hummocky terrain (Fig. 4A). The relief
east of the ridge is mostly flatter than the hummocky terrain to the
west. Numerous parabolic sand dunes have migrated across this surface,
which also has some gently arcuate ridges composed of sand and gravel
(labeled 'off-shore bars' in Fig. ID). Rectangular segments
revealed by LiDAR imagery at the cemetery and other places along Winameg
ridge indicate some minor (<1 m) modification of the ridge surface.
The combination of ER and GPR surveys provides information about
the thickness and nature of the sediment associated with Winameg ridge
(Blockland 2013). The 81 -m long ER survey parallel to the ridge crest
(ER1, Fig. 4A) revealed ~3 m of highly resistive, flat-lying sediment
sharply overlying very low resistive sediment (Fig. 5A). A GPR transect
parallel to this ER survey (Fig. 4A, G PR 1) revealed flat-lying,
continuous to discontinuous reflections parallel to the surface that
also end abruptly at 2-3 m depth (Fig. 4C). A GPR transect perpendicular
to these surveys shows subparallel, continuous and discontinuous
reflections parallel with the surface slope and a localized area of
scour and fill centered on 75 m distance along the transect (Fig. 4D).
Sand and pebble gravel between 2 and 4 m in thickness overlying
clay-rich diamicton were collected from auger holes adjacent to the
geophysical transects. Another GPR survey (GPR4, Figs. 4A, 5C) along the
east slope of the ridge contained mostly subparallel, hummocky
discontinuous reflections. The basal reflections do not have a sharp
contact, suggesting a depth of sand and gravel of about 4 m.
A 630-m long ER survey eastwards from the north end of Winameg
ridge (Figs. 4A, 5B) shows the more-resistive sand and gravel pinching
out downslope to the east and briefly reappearing beneath an offshore
bar at 165 m distance. At approximately 10 m below the surface, a
more-resistive unit was observed accompanied by a color change from blue
to green. A parallel GPR survey (data not shown) revealed a similar
stratigraphy in the first 300 m (GPR 3a). From 300 to 500 m distance,
the parallel GPR survey was offset 200 m north (GPR3b), and the few
reflections were mostly subparallel, continuous to discontinuous, with
some hummocky reflections suggesting infilling of some depressions.
Three Schlumberger soundings were collected from the general area
(Fig. ID) for evaluating the glacial stratigraphy in detail. At each
site the top of a more-resistive unit was encountered between 10 and 11m
beneath the ground surface. It is hypothesized that this unit is a
coarser diamicton equivalent to the Early Woodfordian till, similar to
the stratigraphy exposed at the Toledo Edison dam site along the
Auglaize River upstream of Defiance, OH (Forsyth I960) and at Fort
Wayne, IN (Bleuer and Moore 1971).
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
Lyle's ridge. Lyle's ridge is part of the same shoreline
lineation and about five km SSW of Winameg ridge. It is approximately 1
km long and 400 m wide, and tapers to a point at its southwest end
(Figs. ID, 4B). Two smaller ridges appear north of Lyle's ridge.
The top of the northern half of Lyle's ridge has been extensively
mined for sand and gravel. The highest point adjacent to the mined-out
area is 236 m. An oval-shaped hill to the south is the same elevation as
Lyle's ridge.
Immediately to the east of Lyle's ridge, a shallow channel
about 1 km wide with a depth of 2 to 4 m below the surrounding land
surface trends parallel to the lower Maumee shoreline. The channel
originates near Winameg ridge and extends to the southwest as a boundary
between the hummocky topography of the Defiance Moraine and the
relatively flat lake plain below the shoreline. The channel ends to the
southwest as a closed depression higher than, and landward of, the
Whittlesey shoreline (Fig. 1C).
Diamicton exists at the ground surface on either side of
Lyle's ridge and 1 m below the crest of the ridge immediately south
of the GPR transect (Fig. 4B). Other auger holes along the geophysical
transects record depths of sand and gravel ranging from 1 m to greater
than 1.5 m. Most of the reflections in the GPR profile across the crest
of the ridge are subparallel, discontinuous and hummocky (Fig. 5D).
Basal reflections end abruptly, presumably at the contact with
diamicton. Some reflections extend more deeply, notably at 40 m and 145
m distance. At the southeast end of the transect, the sediment is
organic-rich, black mud. The parallel ER transect shows an "2 m
thick resistive unit (sandier) over more conductive material (siltier)
(Fig. 5E), similar to the GPR profile. Based on the ER, GPR and auger
holes, Lyle's ridge is interpreted as a veneer of sand and gravel
overlying clay-rich diamicton and/or silty clay.
Geophysical and hand augering data from both Winameg and
Lyle's ridges document a layer of highly resistive, sandy surface
sediment often sharply overlying a low-resistive basal sediment. The
subparallel, continuous to discontinuous hummocky and wavy reflections
from the GPR are interpreted as aggrading sand and gravel with some
cross-bedding, cut and fill sequences, and accumulation of sediment
along the east and west sides of the ridges. Auger holes and shallow
pits expose the sand and gravel overlying the clay-rich diamicton. In
places the thickness of sand and gravel along the ridges is less than
the relief of the ridge, and the core of the ridge is diamicton. This
suggests that initially the ridges may have had the form of a DeGeer
moraine, or hummocks of diamicton. This is perhaps what is observed on
the DEMs west of the ridges, where some hummocks currently are draped by
aeolian sand. At the time of deposition of Winameg and Lyle's
ridges, this area would have been the shallow shoreface of the early
Lake Maumee (Figs. 4A, B). Wave and current erosion of these higher
knobs of glacial sediment are likely to have released sand and gravel
into the littoral system.
However, this diamicton has proportionately little sand and gravel,
so complete reworking of ice-marginal kame deposits may best explain the
location and crescentic shape of these landforms. Similarly, the
offshore bars east and south of Lyle's ridge (Fig. 1D) that consist
of sand and gravel were likely created by reworking of ice-marginal
deposits as well. The spit platform and emergent spits on the south side
of Winameg ridge indicate a dominant current direction from the north.
The Maumee shoreline segments north of Winameg ridge also have
asymmetric spit formation indicating net littoral transport to the south
and southwest (Fig. ID).
Age Control
At present, the different phases of ALE have minimal age control.
Radiocarbon ages are often minimum or maximum ages for a given lake
level, determined from wood stratigraphically above, or within/beneath
the beach, respectively. Here, we present the first OSL ages from the
Maumee and Whittlesey shorelines, which directly date deposition of the
beach. Aeolian deposits post date the strandlines, and 14 new ages
(Table 1) are included to compare with previously collected OSL ages
(Campbell et al. 2011; Higley et al. 2014).
Three OSL ages were determined from Winameg ridge (Fig. 4A). One
date (sample #8) comes from fine-to-medium-grained sand interbedded with
pebbly gravel from a middle position near the ridge crest. An older
(16.7 [+ or -] 0.8 ka; Central Age Model; Galbraith et al. 1999) and
younger (12.9 + 0.8 ka); Minimum Age Model; Galbraith et al. 1999) age
was calculated for UNL3637 because of the equivalent dose distribution
(skewness and kurtosis values; Bailey and Arnold 2006). The older age is
consistent with other ages on strandline sediment in the study area
described later in this section, while the younger age is considered too
young for a strandline age based on all previous work on ALE. Further
south along the ridge, an age of 15.1 [+ or -] 0.8 ka (sample #5) is
from fine-to-medium-grained aeolian sand. At the north end, a small
ridge composed of aeolian sand that is east of and lower than the main
Winameg ridge yielded an age of 11.8 [+ or -] 0.6 ka (sample #1).
The hummocky topography northwest of the Town of Wauseon was also
investigated to determine the origins and ages of the various hummocks
(Figs. ID, 6A, B). The morphology of the hummocks is varied, with some
forming ridges, crescents, and oval-shaped ridges that partially or
completely surround a central depression. OST sample #6 is from the top
of the latter type of hummock, and was collected within dipping,
planar-laminated fine sand (Fig. 6C). The resulting age of 19.5 [+ or -]
1.1 ka is considered to be too old because the current deglaciation
model (Fig. 2) has this site subsequently overrun by Port Bruce-aged
ice. A second sample was collected from the same site (within -5 m,
similar sediments), and resulted in an age of 15.4 [+ or -] 0.7 ka
(sample #10, Fig. 6A). At a second hummock -100 m away (sample #11),
flat-lying, planar-laminated, fine lacustrine sand was dated at 16.3 [+
or -] 0.9 ka.
The 239 m elevation at these sites is within the elevation range of
the higher stands of Take Maumee. However, the low-centered hummocks are
not characteristic of a glacial lake strandline. The land surface south
and west of Wauseon has many small ice walled lake plains; these types
of features are generally understood to develop within small lakes in
stagnant glacial ice. The rims typically contain sandy or gravelly
sediment delivered by meltwater (cf. Curry etal. 2010). The observed
sediment may have been deposited in such an environment, and was then
deformed as the ice-melted to create the dipping beds. If the sediment
was deposited in Lake Maumee, it may have been against stagnant ice that
subsequently melted out. OSL age #6 is considered too old and is not
discussed further. OSL ages 10 and 11 are considered minimum ages for
melting stagnant ice that was presumably coeval with Lake Maumee at this
location.
[FIGURE 6 OMITTED]
Aeolian sand was encountered at two other hummocks further to the
northwest in the same field (Fig. 6A). A sinuous ridge (#13) with
aeolian sand 5 m thick dated to 14.3 [+ or -] 0.7 ka at its eastern end.
A small parabolic dune with sand at least 3.9 m thick (#12) was younger,
dating to 13.2 + 0.6 ka. The somewhat crescentic shape of the northern
hummock (Fig. 6A) may reflect an incipient sand dune, and the aeolian
sand at the northeastern end of the ridge (#13, Fig. 6A) may simply
record deposition of sand in the lee of an older hummock, not unlike
deposition of the sheet sand and parabolic dunes elsewhere on this
landscape (northwest corner of Fig. 4A). While detailed landform
analysis of the hummocks in Figure 6A is necessary to more precisely
determine their origins, the OSL ages can be used to date their final
formation.
Two OSL ages are from a Lake Whittlesey spit southeast of Wauseon
(Fig. 7A). At the schoolyard site (#2) much of the spit has been mined
away, and the soil pit dug at the base of the slope revealed weakly
laminated, lacustrine fine sand over diamicton. The age of 16.8 [+ or -]
1.1 ka is the same as the oldest Winameg ridge OSL age. At the west end
of the Whittlesey spit (Fig. 7A), which was partially disturbed by
excavations, an age of 12.4 [+ or -] 0.7 ka is from sand, but is too
young to be Whittlesey in age and instead may record fluvial deposition
from Turkey Foot Creek at this site. A short distance to the northeast
of Winameg ridge, a soil pit was dug into the north arm of a parabolic
sand dune that exposed a 30-cm thick, weakly developed paleosol of
organic-rich sand (Fig. 7C). The ages from below and above this paleosol
are 12.9 [+ or -] 0.6 ka (#3) and 12.3 + 0.6 ka (#4), respectively. At
Raker Cemetery south of Delta, OFF, an age of 14.3 [+ or -] 0.8 ka comes
from a sand dune (Fig. 7B).
Rhythmite Sequence
At the Wildwood Metropark (WM) exposure (Figs. IB, 8A), rhythmites
are well exposed for _100 m along the south side of a gully (Anderson
2011). Three units in the exposed sequence were measured at a meander
cutbank (Fig. 8B). Un-oxidized diamicton is exposed in the stream at the
base of the ravine (Fig. 8G). At the study site the diamicton has a
minimum thickness of 3.73 m. The diamicton-rhythmite contact undulates
along the creek at 186 m elevation. A pebble count (n = 50) resulted in
64 percent carbonates, 20 percent crystalline, and 16 percent clastic
pebbles. Abruptly overlying the diamicton was 3.32 m of rhythmites. The
basal rhythmites were dipping and gradually flattened out at about 1 m
above the diamicton. Ripple-form sets (Fig. 8FF) were observed at
various heights in the section. Laminae of 38 couplets were analyzed for
particle size. The lighter-colored, moderate pink (2.5R7/6) silt laminae
(>3.9-7.8 [micro]m) alternated with darker grey (10YR6/1) silty clay
laminae (<3.9 [micro]m).
[FIGURE 7 OMITTED]
Particle size analysis indicated that the minor sand and clay at
the base of the exposure graded to more silt up core. The lighter and
coarser laminae have higher values of magnetic susceptibility than the
darker and finer-grained laminae where measured on a short section of
core. Silt laminae contained less carbonate and more organic material
(2.53 percent, 0.22 percent, n = 5, respectively) than clay laminae
(3.55 percent, 0.186 percent, n = 5, respectively). Basal contacts of
some silt laminae were erosional, in which shallow troughs were cut into
underlying silty clay with coarser particles in the base of the trough.
Some of the darker laminae had a light speckled appearance (Figs. 8E, F)
that in plan view revealed elongate to circular inclusions of silt
within clay. The lower contact of the silt was often diffuse or
disturbed while the lower contact of the silty clay was usually sharp.
At the WM exposure 483 rhythmites were counted, averaging 0.86 cm
thick. In the lower half of the exposure the finer laminae were thicker,
and thinned at the expense of the coarser silt laminae in the upper half
of the exposure. Two meters of massive sand unconformably overlie the
rhythmites. The sand was not studied in detail and is assumed to be
lower-shoreface sand associated with Lake Warren or Wayne, or aeolian
sand associated with the parabolic dune observed adjacent to the gully
(Fig. 8A).
[FIGURE 8 OMITTED]
Similar rhythmites were well exposed at the Brown Cut, an ~50-m
long cutbank exposure in the Oak Openings Metropark (OM) (Fig. 8D). The
diamicton recovered from hand augering beneath the rhythmites is
massive, clay rich, and gravelly. From the sediment exposure and
vibracore, a 4.73 m sequence of rhythmites that is overlain by
alternating units of finely laminated planar beds of sand and silt with
beds of ripple drift and planar laminations was observed. The massive
sandy sediments overlying the rhythmites were described recently by
Higley et al. (2014) and interpreted as a lower-beachface deposit
overlain by aeolian sand. The rhythmites are similar to those described
from the WM exposure in that ripple-form sets exist throughout the core
(Fig. 8H), increasing in frequency up the section. Rhythmite thickness
appears to be controlled by the thickness of the coarser laminae, and
507 couplets were counted.
The sedimentary sequences described from the WM and OM exposures
are interpreted as a deglacial sequence of till overlain by
glaciolacustrine deposits. This interpretation is based on the massive,
clay-rich characteristic of the diamicton that contains erratic stones
and is similar to till mapped in northwest Ohio (Larsen et al. 2005) and
southeast Michigan (Howard 2010). Presumably, the till was deposited
while the glacier was receding eastwards across the Lake Erie basin and
in contact with ancestral Lake Erie. While inaccessible at outcrop, some
stratification would be expected at the contact between the till and
rhythmites. Localized sediment focusing and infilling of irregular
topography on the lake floor would explain the undulating contact
between the till and dipping rhythmites at WM. Three criteria suggested
by Smith and Ashley (1985) support an interpretation of varves: 1) the
contact between dark, winter laminae and overlying light (summer laminae
is usually sharp and in places bioturbated); 2) the winter laminae have
a relatively consistent thickness; and 3) a decrease in sand (fining) up
section. Suspension settling of clay in winter during the non-melt
season explains the lower magnetic susceptibility values (Snowball et
al. 1999). The mm- to cm-sized scour deposits and silt inclusions within
clay (Figs. 8E, F) are interpreted as evidence of bioturbation during
the summer season, which would explain the mottled appearance of the
darker clay.
Ripple-form sets have been observed within glaciolacustrine
deposits in south central Michigan (Fisher and Taylor 2002), and are
commonly interpreted as underflow or hyperpycnal deposits (Shaw and
Archer 1978). The geographic setting of the study area requires that
such flows are supplied from sediment from streams on the west side of
the lake and not from meltwater from ice receding to the east. Stream
systems incising into recently exposed till and lacustrine sediment may
explain the high sediment load required to set up hyperpycnal flows.
Alternatively, slumping of sediment around the shoreline may initiate
hyperpycnal flows (Shaw and Archer 1978).
If the rhythmites are indeed varves, the count of 507 varves from
OM is a minimum estimate of years of sedimentation. It is likely that
some couplets may be underflow deposits, which along with truncated
upper contacts at both study sites suggests that the count of 507 is
only an estimate. If not varves, then even fewer years are recorded by
the rhythmic sediments. Absolute age of the varves is unknown, and they
would have been deposited in water depths at WM of ~20 m in Lake Warren,
~25 m in Lake Arkona, or ~38 m in Lake Whittlesey. At the OM site, water
depths would have been ~5 m shallower in each lake. Based on the
apparent conformable contact with underlying diamicton at the WM
exposure, the rhythmites were deposited in Lake Maumee. The
coarsening-upward sequence at the OM exposure indicates that some of the
rhythmites likely are associated with Lake Warren, Lake Whittlesey, and
perhaps older lakes.
DISCUSSION
Detailed examination of two segments of the lower Maumee shoreline
indicates that these landforms developed as spit complexes associated
with a segmented island shoreline rather than as a 'washed'
beach. The littoral deposits of Winameg and Lyle's ridges resulted
from reworking a hummocky and ridged diamicton surface. During the late
Maumee highstand, these glacial deposits were eroded by waves, and sand
was transported by currents downdrift to construct shallow spit
platforms with low-relief emergent spit ridges. The washed appearance is
derived from reworking and smoothing of the initial landforms with
deposition of the spit complex in shallow water. Later in the evolution
of the area, aeolian mobilization of surficial sand sheets and the
migration of parabolic dunes further covered (or draped) the lower
Maumee shoreline features.
Overtopping of the lower Maumee shoreline by a subsequent highstand
(for example, Maumee II at 232 m being flooded by Maumee III at 238 m
and reoccupied by Maumee IV at 232 m) should have produced several
specific geomorphic and stratigraphic results. For example, preservation
of the initial shoreline requires an extremely rapid rise of lake level
by many meters in a few years. Otherwise, shoreline ravinement during
the transgression and erosion within the shoreface down to wave base
during the highstand would bevel off any pre-existing shoreline. Even if
the initial shoreline survived the transgression, the preserved
stratigraphy should be dominated by landward transport of eroded sand by
overwash and other inlet processes. If this had occurred in the lower
Maumee, accommodation space (depressions) in the highstand shoreface
would be mostly filled by deposition.
Geophysical evidence of landward migration of sediment within the
stratigraphy similar to transgressing modern barrier islands (e.g.,
Hayes and FitzGerald 2013) was not observed. Nor are there obvious
geomorphic features that suggest extensive landward transport in the
study area. Instead of overwash lobes and channels, a spit platform with
simple ridges at the southern end of Winameg ridge records a short-lived
lake level. The spits in the same lineation north of Winameg ridge at
238 to 233.5 m elevation (Fig. ID) similarly show dominant littoral
transport of sediment from the north within a falling lake level; these
shoreline features have relief equivalent to that of Winameg ridge.
Classification of Lyle's ridge is more difficult as the thin ridges
at its north end are likely sand and gravel, similar to the narrow ridge
that was examined in this study. The low-relief ridges (of sand and
gravel) just east of Winameg and Lyle's ridges shown on Figure ID
are most likely offshore bars as mapped by Blockland and Fisher (201 5).
From this analysis, the "washed" characteristics of the
lower Maumee shoreline (II, Fig. 3) simply record syndepositional
processes that operated while the lake existed, followed by
post-depositional modification by aeolian processes. Without the need
for a younger transgression to overtop and drown an older strandline,
the reconstruction of Maumee lake levels can be simplified to a sequence
in which successively lower-elevation strandlines are progressively
younger. This simplified lake-level history favors the episodic model,
and is consistent with the findings of Bleuer and Moore (1971) that
incision of the Fort Wayne outlet through time could accommodate all of
the Maumee levels. Previously, Leverett (1902) assigned elevations of
233.2-231.6 m to these lower Maumee shoreline ridges, and Forsyth (1959)
assigned an elevation of 232 m. There has been uncertainty on what
specific landforms were interpreted in the western basin of ALE to
indicate past lake levels. Rather than pursuing detailed discussion with
past results, the next step forward to understanding lake-level
evolution in the western basin of ALE should be based on detailed
mapping of individual shorelines to aid water-plane assignments (cf.
Fisher 2005; Breckenridge 2015) using the recently released LiDAR data.
Four OSL dates are used to assess the ages of the Maumee and
Whittlesey shorelines. The two oldest ages are nearly identical and come
from the Maumee and WTittlesey shorelines. The two younger ages are from
fine sand associated with low-centered hummocks northwest of Wauseon,
but at the same elevation as the Winameg ridge. The precise origin of
these hummocks awaits more detailed study, with current hypotheses
favoring small ice-walled lake plains, or outwash or littoral sediment
deposited on stagnant ice. The hummock ages overlap with the two older
ages at one standard deviation, thus an age of ~16.8 [+ or -] 1.0 ka is
assigned to the younger Maumee strandline. According to the cyclic
model, the Whittlesey shoreline should be much younger if it developed
after the Ypsilanti Low. However, OSL date #2 (Table 1) is the same age
as the lower Maumee ridge, which instead favors the episodic model.
Alternatively, OSL date #2 may be from sand that was deposited in Lake
Maumee in -8 m of water, and lies stratigraphically below the deposits
of the Whittlesey strandline. In that case, Lake Whittlesey remains to
be dated by OSL. However, according to Calkin and Feenstra (1985) Lake
Whittlesey is the best-dated lake in ALE at ~13,000 [sup.14]C yr BP
(~15,900 cal yr BP), which is within one standard deviation of the OSL
ages presented here.
While a Lake Arkona strandline was not dated in this study, Hough
(1958) reports an age of 13,600 [+ or -] 500 [sup.14]C yr BP (W-33) for
it, and Szabo et al. (1988) report an age of "about" 13,400
[sup.14C] yr BP (ISGS-1480), both of which calibrate to ~16,400 and
~16,100 cal yrs BP, respectively. Based on these data of overlapping
ages at one standard deviation, there is little time available to form
the strandlines of lakes Maumee, Whittlesey and Arkona, and the
Ypsilanti lowstand using the cyclic model. A suggested sequence of
events is: Maumee to Whittlesey to Arkona, and possibly then the
Ypsilanti low (Fig. 9). More OSL dates are required from these beaches
to refine the lake-level curve.
Lakes Maumee, Whittlesey and Arkona appear to have developed
between 17 and 15 ka, before the warm Bolling-Allerod interstadial. In
contrast, Lake Warren is well dated at about 14.1 ka (Fig. 9), and
developed during the Bolling-Allerod interstadial. The Warren beaches
are sandy and are strongly developed throughout the basin (Calkin and
Feenstra 1985), which may reflect a longer period of time at a
relatively stable water level.
The approximately 500 varves from both the WM and OM sites provide
some limitations on the deglacial history for the region. The sharp
contact from till to lacustrine sediments observed at WM is interpreted
as deposition into an ice-contact glacial lake at the receding ice
margin. The upper contact of the rhythmites is most likely erosional at
both sites, thus the number of rhythmites is a minimum. However, the
similar count from both sites and the increase in ripple-form sets
towards the top of the sequence at OM suggests shoaling and minimal
erosion. Taken together, the evidence suggests: 1) ice-margin recession
was much slower out of the basin than previously realized; or 2) a
disconformity at the top of the till or within the rhythmite sequence
marks a period of time between ice-margin recession (lower rhythmites)
and resumed rhythmite deposition as lake level dropped and sediment was
supplied from the land (upper rhythmites). Subaerial exposure during the
Ypsilanti lowstand would be expected to have generated a weathering
profile, an unconformity, and possibly a ravinement surface as a record
of the regression-transgression cycle. No evidence of these sedimentary
features was observed along the outcrop or within the core. Supporting
evidence for the Ypsilanti low-stand was not observed at sites expected
to record such dramatic lake-level changes.
[FIGURE 9 OMITTED]
The sand dunes dated in this study were the same age or older than
the dunes dated by Campbell et al. (2011) up to 15 km to the east. With
one or two outliers, the sand dunes are progressively younger at lower
elevations (Fig. 9). This general relationship is expected when the
dunes date back to deglaciation and are concomitant with dropping lake
levels, if sediment supply for the dunes is limited. Progressively
younger dunes would also be expected when they are remobilized, as
indicated by OSL dates 3 and 4 from this study on either side of a
paleosol. Eight of the 16 OSL dates on dunes overlap at one sigma error
with the Younger Dryas stadial (Fig. 9). It appears that dune activation
may be associated with episodic sand availability at the land surface
because of either deglaciation controlling lake-level fall, or abrupt
climate change, specifically times of reduced vegetative cover
associated with regional aridity.
SUMMARY
Cyclic changes in lake levels of ancestral Lake Erie have been
discussed for more than 100 years. The concept of "washed"
Lake Maumee shorelines identified as having been submerged by subsequent
transgressions can be explained more simply by first distinguishing
between different types of shorelines. Geomorphological analysis and
geophysical transects across a "washed" lower Maumee beach and
shoreface provided no support for a later transgression and reworking of
sediment. Instead, interpreting the landform as a spit formed in shallow
water better explains the washed appearance. Extensive mobilization of
aeolian sand on and around the ridge also may explain part of the washed
appearance. The results of this research support an interpretation of
episodic lake-level fall in Lake Maumee with temporary
levels--stillstands--controlled by sill elevation at the Fort Wayne
outlet.
OSL ages obtained from a Maumee spit, fine sand beneath a Lake
Whittlesey spit, and fine sand from low-centered hummocks west of the
Maumee shoreline support an age of approximately 16.8 ka for Lake Maumee
in the study area. The one OSL date from beneath the Lake Whittlesey
spit yields the same age as the Maumee spit, which calls into question
whether the Whittlesey is accurately dated with that OSL sample.
Radiocarbon ages summarized by Calkin and Feenstra (1985) suggest an age
of 15,900 cal yr BP for Lake Whittlesey and ~16,300 cal yr BP for Lake
Arkona. This condensed chronology provides little time for formation of
Lake Arkona beaches at elevations between lakes Maumee and Whittlesey,
and the Ypsiland lowstand. It may be possible that the Ypsilanti
lowstand is actually younger than the Whittlesey shoreline.
The floating chronology of ~500 rhythmites from two exposures
stratigraphically below Lake Warren shoreline deposits were expected to
show unconformities associated with the Ypsilanti lowstand. However,
sedimentary features that would support the Ypsilanti lowstand, such as
a weathering profile, an unconformity, and/or a ravinement surface, were
not observed between the till and rhythmites, or within the rhythmites.
The additional OSL ages on sand dunes agree with results from
Campbell et al. (2011) that many of the dunes were last active during
the Younger Dryas stadial. Additional new ages mostly from
higher-elevation sand dunes record earlier aeolian activity.
ACKNOWLEDGMENTS
Field assistance was provided by Mike Angle, Jake Ardner, Kira
Baca, Joel Banaszak, Samantha DeWald, BiniamEstifanos, Jennifer Horton,
Chris Maike, Susan McCormick, Brendan Poffenbaugh, Steve Ransford, Kyle
Siemer, Heather Stewart and Kirk Zmijewski. Technical assistance was
from Richard Becker, Butch Berger, Jamie Martin-Hayden and Jason Witter.
Karen Menard is thanked for facilitating access to the Toledo Metro Park
sites. Funding was provided from USGS EDMAP grant G11AC20135 to Fisher,
and University of Toledo Department of Environmental Sciences Ruedisili
fund. The paper was improved by the journal reviewers, and especially
from the insight of Dr. Henry Loope.
LITERATURE CITED
Aitken MJ. 1998. An Introduction to optical dating: The dating of
Quaternary sediments by the use of photon-stimulated luminescence. New
York: Oxford University Press. 280 p.
Anderson B. 2011. Estimating the duration of ancestral lake Erie
using varve analysis at and above the Warren stage in northwest Ohio.
[Thesis]. [Toledo (OH)]: University of Toledo.
Bailey RM, Arnold LJ. 2006. Statistical modeling of single grain
quartz De distributions and an assessment of procedures for estimating
burial dose. Quat. Sci. Revs. 25:2475-2502.
Ballarini M, Wallinga J, Wintle AG, Bos AJJ. 2007. A modified SAR
protocol for optical dating of individual grains from young quartz
samples. Radiation Measure. 42:360-369.
Barnett PJ. 1992. Quaternary geology of Ontario. In: Thurston PC,
Williams HR, Sutcliffe RH, Scott GM, editors. Geology of Ontario. Sp Vol
4. Toronto: Ontario Geological Survey. p. 1010-1088.
Bleuer NK, Moore MC. 1971. Glacial stratigraphy of the Fort Wayne
area and the draining of glacial Lake Maumee. Indiana Acad. Sci.
81:195-209.
Blockland JD. 2013. The surficial geology of Fulton County, Ohio:
insight into the Late Pleistocene-Early Holocene glaciated landscape of
the Huron-Erie Lake Plain, Fulton County Ohio, USA. [Thesis], [Toledo
(OH)]: University of Toledo.
Blockland JD, Fisher TG. 2015. The surficial geology of the Delta
7.5 minute series quadrangle, Ohio, USA. J Maps. DOI:
10.1080/17445647.2015.1071289
Breckenridge A. 2015. TheTintah-Campbell gap and implications for
glacial Lake Agassiz drainage during the Younger Dryas cold interval.
Quat. Sci. Revs. 117: 124-134.
Camp MJ. 1981. Pleistocene history and molluscan paleoecology of
the Winameg mastodon site, Fulton County, OH. Ohio J Sci. 81:230-235.
Calkin PE, Feenstra BH. 1985. Evolution of the Erie-basin great
lakes. In Karrow PF, Calkin PE. Eds. Quaternary Evolution of the Great
Lakes, Sp. Pap. 30. St. John's: Geological Survey of Canada, p.
149-170.
Campbell MC, Fisher TG, Goble RJ. 2011. Terrestrial sensitivity to
abrupt cooling recorded by aeolian activity in northwest Ohio, USA:
Quat. Res. 75:411-416.
Cunningham AC, Wallinga J. 2010. Selection of integration time
intervals for quartz OSL decay curves. Quat. Geochron. 5:657-666.
Curry BB, Konen ME, Larson TH, Yansa CH, Hackley KC, Alexanderson
H, Lowell TV. 2010. Ihe DeKalb mounds of northeastern Illinois as
archives of deglacial history and postglacial environments. Quat. Res.
74:82-90.
Dreimanis A, Goldthwait RP. 1973. Wisconsin glaciation in the
Huron, Erie, and Ontario Lobes. Geol. Soc. Am. Mem. 136:71-106.
Durcan JA, Duller GAT. 2011. The fast ratio: A rapid measure for
testing the dominance of the fast component in the initial OSL signal
from quartz. Radiation Measure. 46:1065-1072.
Earthlmager 2004. Earthimager 2D, Version 1.7.7, Advanced
Geosciences, Inc., Austin, Texas 78726.
Fisher, TG. 2004. Vibracoring from lake ice with a lightweight
monopod and piston coring apparatus. J Paleolimn. 31:377-382.
Fisher TG. 2005. Strandline analysis in the southern basin of
glacial Lake Agassiz, Minnesota and North and South Dakota, USA. Geol.
Soc. Am. Bull. 117:1481-1496.
Fisher TG, Taylor LD. 2002. Sedimentary and stratigraphic evidence
for subglacial flooding, south-central Michigan, USA: Quat. Int.
90:87-115.
Forsyth JL. 1959. The beach ridges of northern Ohio. Ohio Div.
Geol. Sur. Info. Circ. 25:1-10.
Forsyth JL. 1960. Correlation of till exposed in the Toledo Edison
Dam cut, Ohio. Ohio J Sci. 60:94-100.
Galbraith RF, Roberts RG, Laslett GM, Yoshida H, Olley JM. 1999.
Optical dating of single and multiple grains of quartz from Jinmium Rock
Shelter, Northern Australia: Part I, experimental design and statistical
models. Archaeometry. 41:339-364.
Hayes MO, FitzGerald DM. 2013. Origin, evolution and classification
of tidal inlets. J Coast. Res. 69:14-33.
Heiri O, Lotter AF, Lemcke G. 2001. Loss on ignition as a method
for estimating organic and carbonate content in sediments:
reproducibility and comparability of results. J Paleolimnol. 25:101-110.
Herdendorf CE, 2013. Research overview: Holocene developments of
Lake Erie: Ohio J Sci. 112:24-36.
Higley MC, Fisher TG, Jol HM, Lepper K, Martin-Hayden JM. 2014.
Stratigraphic and chronologic analysis of the Warren Beach, northwest
Ohio, USA. Can. J Earth Sci. 51:737-749.
Howard JL. 2010. Late Pleistocene glaciolacustrine sedimentation
and paleogeography of southeastern Michigan, USA: Sed. Geol.
223:126-142.
Hough JL. 1958. Geology of the Great Lakes. Urbana, Illinois.
University of Illinois Press. 334 p.
Interpex 2010. Interpex, Inc. Golden, Colorado 80402 user manuals
for the 1X1D at http://www.interpex.com/ixld/ ixld.htm
Jol HM, Bristow CS. 2003. GPR in sediments: advice on data
collection, basic processing and interpretation, a good practice guide.
In Bristow CS, Jol HM, editors. Ground Penetrating Radar in Sediments.
London: Geol. Soc. London. 211:9-27.
Karrow PF, Calkin PE. 1985. Quaternary Evolution of the Great
Lakes. St. John's: Geological Survey of Canada. Sp. Pap. 30. 258 p.
Larsen GE, Shrake L, Swinford EM. 2005. Surficial geology of the
Ohio portion of the Toledo 30 x 60-minute quadrangle. Ohio Division of
Geological Survey Maps SG-2 Toledo, 1:100,000.
Leverett F. 1902. Glacial formations and drainage features of the
Erie and Ohio basins. Washington, USGS Monograph 41. 802 p.
Leverett F, Taylor FB. 1915. The Pleistocene of Indiana and
Michigan and the history of the Great Lakes. U. S. Geological Survey
Monograph 53. 527 p.
Lewis CFM, MooreTCJ, Rea DK, Dettman DL, Smith AJ, Mayer LA. 1994.
Lakes of the Huron basin: their record of runoff from the Laurentide Ice
Sheet. Quat. Sci. Revs. 13:891--922.
Mickelson DM, Clayton L, Fullerton DS, Borns HWJ. 1983. The Late
Wisconsin glacial record of the Laurentide Ice Sheet in the United
States. In: Wright HEJ, editor. Late-Quaternary Environments of the
United States, vol 1. Minneapolis: University of Minnesota Press, p.
3-37.
Morner N-A, Dreimanis A. 1973. The Erie Interstade. Geol Soc Am
Mem. 136:107--134.
Murray AS, Wintle AG. 2000. Luminescence dating of quartz using an
improved single-aliquot regenerative-dose protocol. Radiation Measur.
32:57--73.
Murray AS, Wintle AG. 2003. The single aliquot regenerative dose
protocol: potential for improvements in reliability. Radiation Measur.
37:377-381.
Prescott JR, Hutton JT. 1994. Cosmic ray contributions to dose
rates for luminescence and ESR dating: large depths and long-term time
variations. Radiation Measure. 23:497-500.
Rasmussen SO, Andersen KK, Svensson AM, Steffensen JP, Vinther BM,
Clausen HB, Siggaard-Andersen M-L, Johnsen SJ, Larsen LB, Dahl-Jensen D,
and others. 2006. A new Greenland ice core chronology for the last
glacial termination. J Geophys. Res. 11 LDO6102.
Rodnight H.2008. How many equivalent dose values are needed to
obtain a reproducible distribution? Ancient TL. 26:3-9.
Shaw J, Archer J. 1978. Winter turbidity current deposits in Late
Pleistocene glaciolacustrine varves, Okanagan Valley, British Columbia,
Canada. Boreas. 7:123-130.
Sheriff RF!. 1989. Geophysical Methods. Prentice-Hall. 605 p.
Smith ND, Ashley GM. 1985. Chapter 4 Proglacial lacustrine
environment. In: Ashley GM, Shaw J, Smith ND, editors. Glacial
Sedimentary Environments. Short Course No. 16. Tulsa, OK: Soc. Paleontol
Min. p. 135-216.
Snowball I, Sandgren P, Petterson G. 1999. The mineral magnetic
properties of an annually laminated Holocene lake-sediment sequence in
northern Sweden. Holocene. 9:353-362.
Szabo JP, Angle MP, Eddy AM. 2011. Pleistocene glaciation of Ohio,
USA. In Ehlers J, Gibbard PL, Hughes PD, editors. Quaternary
Glaciations--Extent and Chronology, Elsevier Developments in Quaternary
Science 15:513-519.
Szabo JP, Carter CJ, Bruno PW, Jones EJ. 1988. Glacial and
postglacial deposits of northeastern Ohio. Ohio J Sci. 88:66-74.
Totten SM. 1985. Chronology and nature of the Pleistocene beaches
and wave-cut cliffs and terraces, northwestern Ohio. In Karrow PF,
Calkin PE, editors. Quaternary Evolution of the Great Lakes. St.
John's, Sp Pap 30. Geological Survey of Canada, p. 171-184.
Wintle AG, Murray AS. 2006. A review of quartz optically stimulated
luminescence characteristics and their relevance in single-aliquot
regeneration dating protocols. Radiation Measur. 41:369-391.
TIMOTHY G. FISHER, Department of Environmental Sciences, University
of Toledo, OH, USA; JOSEPH D. BLOCKLAND, Department of Environmental
Sciences, University of Toledo, OH, USA, North Dakota Geological Survey,
Bismarck, ND, USA; BRAD ANDERSON, Department of Environmental Sciences,
University of Toledo, OH, USA, Nyrstar Tennessee Mines, Gordonsville,
TN, USA; DAVID E. KRANTZ, Department of Environmental Sciences,
University of Toledo, OH, USA; DONALD J. STIERMAN, Department of
Environmental Sciences, University of Toledo, OH, USA; RONALD GOBLE,
Department of Earth and Atmospheric Sciences, University of Nebraska at
Lincoln, Lincoln, NE, USA
(1) Address correspondence to Timothy G. Fisher, Mail Stop 604,
University of Toledo, Department of Environmental Sciences, Toledo, OH
43606-3390. E-mail: timothy.fisher@utoledo.edu
Table 1
Optically Stimulated Luminescense (OSL) Dosimetric Data and Results
Ref Lab Burial [H.sub.2]O [K.sub.2]O
# # Depth (%) (1) (%)
(m)
1 UNL3455 0.85 2.1 1.51 [+ or -] 0.04
2 UNL3456 0.90 20.7 1.14 [+ or -] 0.04
3 UNL3457 1.24 3.5 1.41 [+ or -] 0.04
4 UNL3458 1.5 2.5 1.19 [+ or -] 0.04
5 UNL3488 1.0 14.9 1.42 [+ or -] 0.04
6 UNL3489 0.85 10.8 1.38 [+ or -] 0.04
7 UNL3636 1.6 21.8 1.8 [+ or -] 0.05
8 UNL3637 1.1 5.3 1.71 [+ or -] 0.05
Minimum Age model (Galbraith et al. 1999)
9 UNL3638 1.6 8.4 1.69 [+ or -] 0.04
10 UNL3770 1.5 7.9 1.85 [+ or -] 0.05
11 UNL3771 1.83 23.0 1.82 [+ or -] 0.05
12 UNL3772 1.15 1.9 2.06 [+ or -] 0.06
13 UNL3773 1.18 3.3 2.01 [+ or -] 0.06
Ref U Th Cosmic
# (ppm) (ppm) (Gy)
1 1.51 [+ or -] 0.09 3.6 [+ or -] 0.22 0.19
2 1.34 [+ or -] 0.09 2.83 [+ or -] 0.19 0.19
3 0.94 [+ or -] 0.08 2.19 [+ or -] 0.16 0.18
4 0.72 [+ or -] 0.07 2.28 [+ or -] 0.16 0.18
5 1.52 [+ or -] 0.09 2.97 [+ or -] 0.2 0.19
6 1.06 [+ or -] 0.08 2.11 [+ or -] 0.16 0.18
7 1.66 [+ or -] 0.12 3.8 [+ or -] 0.23 0.18
8 2.27 [+ or -] 0.11 3.2 [+ or -] 0.19 0.19
Minimum Age model (Galbraith et al. 1999)
9 0.9 [+ or -] 0.08 2.39 [+ or -] 0.17 0.18
10 1.11 [+ or -] 0.11 2.3 [+ or -] 0.25 0.18
11 1.42 [+ or -] 0.12 3.22 [+ or -] 0.27 0.17
12 1.23 [+ or -] 0.12 2.91 [+ or -] 0.27 0.19
13 1.11 [+ or -] 0.11 2.82 [+ or -] 0.27 0.19
Ref Dose Rate [D.sub.e] No. of
# (Gy/ka) (Gy) (2) Aliquots
1 2.02 [+ or -] 0.07 23.79 [+ or -] 0.81 51
2 1.65 [+ or -] 0.08 27.86 [+ or -] 1.17 54
3 1.71 [+ or -] 0.06 21.95 [+ or -] 0.63 54
4 1.61 [+ or -] 0.06 19.83 [+ or -] 0.64 55
5 1.75 [+ or -] 0.07 26.34 [+ or -] 0.73 50
6 1.58 [+ or -] 0.06 30.85 [+ or -] 1.25 55
7 1.88 [+ or -] 0.09 23.21 [+ or -] 0.69 62
8 2.26 [+ or -] 0.08 37.83 [+ or -] 1.3 54
Minimum Age model (Galbrait29.19 [+ or -] 1.32
9 1.8 [+ or -] 0.07 24.72 [+ or -] 0.75 58
10 1.98 [+ or -] 0.08 30.48 [+ or -] 0.54 63
11 1.78 [+ or -] 0.09 29.04 [+ or -] 0.64 59
12 2.38 [+ or -] 0.09 31.5 [+ or -] 0.79 53
13 2.26 [+ or -] 0.08 32.44 [+ or -] 1.0 62
Ref Age
# (ka) (3)
1 11.8 [+ or -] 0.6
2 16.8 [+ or -] 1.1
3 12.9 [+ or -] 0.6
4 12.3 [+ or -] 0.6
5 15.1 [+ or -] 0.8
6 19.5 [+ or -] 1.1
7 12.4 [+ or -] 0.7
8 16.7 [+ or -] 0.8
Minimu12.9 [+ or -] 0.8
9 14.3 [+ or -] 0.8
10 15.4 [+ or -] 0.7
11 16.3 [+ or -] 0.9
12 13.2 [+ or -] 0.6
13 14.3 [+ or -] 0.7
(1) In-situ Moisture Content; (2) Error on [D.sub.e] is 1 standard
error; (3) Error on age includes random and systematic errors
calculated in quadrature
FIGURE 2. Approximate ages and events associated with
the Huron-Erie Lobe. Modified from Howard (2010).
Age (ka) Age (ka) Glacial
Time cal. yrs [sup.14]C Substage
Divisions BP yrs BP Events
Late Pleistocene (Late Wisconsinan)
Early Woodfordian 13.8 12.5 Port Huron Stade
14.2 12.2
16.0 13.1
15.9 13.0
16.4 13.5 Mackinaw Interstade
Late Woodfordian
17.6 14.5 Port Bruce Stade
18.1 14.8 max. ice extent
18.2 15.0
19.8 16.5 Erie Interstade
23.4 19.6 Nissourian
Stade
max. ice extent
27.6 23.0 Plum Point Interstade
36.0 32.0 Cherrytree Stade
Events in Huron-Erie Strand.
Time Lake Plain Elev.
Divisions (m)
Late Pleistocene (Late Wisconsinan)
Early Woodfordian Lake Lundy -188
Lake Grassmere 190-195
Lake Wayne 201
Lake Warren (I-III) 207-203
Lake Whittlesey 224
Lake Ypsilanti 91
Late Woodfordian
Lake Arkona 210-216
Lake Maumee (I-IV) 232-244
Late Woodfordian Drift
also called, Cary, Hayesville,
Hiram, Lavery, Port-Stanley,
and Lagro Drift by region.
Lake Leverett 178
Early Woodfordian Drift
also called Tazewell. Catfish
Creek, Kent, Millbrook, and
Navarre, and Trafalgar Drift
by region
Older Drift
Time Moraines
Divisions
Late Pleistocene (Late Wisconsinan)
Early Woodfordian East to Mohawk
Valley and/or
west to Grand
Valley
Grand Valley R.
? Niagara R.
Late Woodfordian
Grand Valley R.
Ft. Wayne
Union City-
Defiance
Moraines
Mohawk valley?
Hartwell Moraines
