ARTICLE

Paleobotanical and environmental implications of a buried forest bedin northern Lower Michigan, USAR.J. Schaetzl, C.H. Yansa, and M.D. Luehmann

Abstract: Sediments and stratigraphy at the Tower Buried Forest site provide a glimpse into the immediate post-glacialenvironment of northern LowerMichigan, USA. At this site, sandy glacial outwash is overlain by (1) ≈37 cmof peat associatedwitha wetland, above which are (2) flat-lying spruce and larch logs and branches that date between ≈10 910 and 10 340 cal years BP.Above the woody materials are ≈55–75 cm of sand, interbedded with organic muck laminae, which we interpret as localalluvium. This stratigraphic sequence is overlain by muck soil materials associated with the modern wetland. Pollen and plantmacrofossil analysis of the lower peat indicate that a tundra–boreal parkland had been established here, shortly after finaldeglaciation. Later, Picea glauca (white spruce), Picea mariana (black spruce), Larix laricina (larch), and Abies balsamea (balsam fir)became abundant in an open boreal forest – Spaghnum peatland. Subsequent increases in Pinus (pine) and Typha (cattail) indicatedrying and possibly warming conditions. At ca. 10 340 cal years BP (or a few centuries prior), nearly two millennia after finaldeglaciation, water flowed across the site, possibly knocking over the trees.. Interbedded sand and muck deposits above theflat-lying logs are interpreted as alluvial deposits from this event, after which the wetland became quiescent again.

Résumé : Les sédiments et la stratigraphie dans la localité de la forêt ensevelie de Tower ouvrent une fenêtre sur un milieu établiimmédiatement après le retrait des glaciers, dans la partie nord du Bas-Michigan (États-Unis). À cet endroit, des dépôts d'épandagefluvioglaciaire sableux sont recouverts de (1) ≈ 37 cm de tourbe associée a unmilieu humide, au-dessus de laquelle se trouvent (2) destroncs et branches d'épinettes et de mélèzes reposant a l'horizontale, dont les âges se situent entre ≈10 910 années étalonnées avantaujourd'hui (ans AA) et 10 340 ans AA. Au-dessus du matériel ligneux reposent ≈55–75 cm de sable avec de minces interlits deterre tourbeuse organique que nous interprétons comme représentant des alluvions de source locale. Cette séquencestratigraphique est recouverte de terre tourbeuse associée au milieu humide moderne. L'analyse de pollens et de macrofossilesde végétaux de la tourbe inférieure indique qu'une forêt-parc boréale et de toundra s'est établie a cet endroit peu après ladéglaciation finale. Picea glauca (l'épinette blanche), Picea mariana (l'épinette noire), Larix laricina (le mélèze laricin) et Abiesbalsamea (le sapin baumier) sont par la suite devenus abondants dans un paysage ouvert de forêt boréale et de tourbière desphaigne. L'augmentation subséquente de l'abondance des Pinus (pins) et des Typha (quenouilles) indique une tendance vers desconditions plus sèches et possiblement plus chaudes. Vers 10 340 ans AA (ou quelques siècles plus tôt), près de deux millénairesaprès la déglaciation finale, de l'eau a déferlé sur le site, abattant possiblement les arbres. Les dépôts de sable et de terretourbeuse interlités recouvrant les troncs couchés a l'horizontale sont interprétés comme étant des dépôts alluviauxmis en placedurant cet épisode, après quoi le milieu humide est redevenu calme. [Traduit par la Rédaction]

IntroductionSparse vegetation cover on the recently deglaciated, Late

Pleistocene landscapes of the Midwest, USA (e.g., Webb et al. 1983;Curry and Yansa 2004; Hupy and Yansa 2009; Curry and Petras2011; Rech et al. 2011), coupled with incomplete preservation oforganic materials, has led to a relative paucity of radiocarbondates on wood, charcoal, or other terrestrial plant macrofossilsthat could be used to constrain the glacial and immediate post-glacial geomorphic events of this period. This is especially true forlocations north of ≈44° latitude (Clayton et al. 2001). Notable ex-ceptions include the Two Creeks forest bed in Wisconsin(Broecker and Farrand 1963; Kaiser 1994; Maher and Mickelson1996; Rech et al. 2011), and the Cheboygan bryophyte bed (Larsonet al. 1994) and the Gribben Lake forest (Lowell et al. 1999) inMichigan (Fig. 1). These three sites involve burial of biotic commu-nities by glacigenic sediments associated with a readvancing gla-cier (Johnson et al. 1997). Hence, they help constrain some of thelast glacial events in the region.Minimal data exist, however, withwhich to characterize, evaluate, and constrain processes andevents on the immediate post-glacial landscape. Although pollen

and plant macrofossil research have constrained the temporaland spatial advance of vegetation onto the recently deglaciatedlandscapes of the Upper Midwest, documenting first tundra, thenspruce-dominated parkland, and later, mixed forest (e.g., Maherand Mickelson 1996; Maher et al. 1998; Kapp 1999; Yu 2000;Shuman et al. 2002; Webb et al. 2004; Hupy and Yansa 2009;Gonzales and Grimm 2009; Curry and Petras 2011), little is knownabout the paleoecology or geomorphology of this landscape in theperiod immediately after final deglaciation. Fluvial incision andsediment transfer, slope instability, melting of permafrost, ice-walled lake plain formation and drainage, breaching of sedimentdams and subsequent release of ponded water, and eolian activityof various kinds are but a few of the geomorphic processes thatmust have been occurring on the post-glacial landscapes of theUpper Midwest. It is not known, however, whether these pro-cesses essentially shut down following the establishment of forestvegetation, or if, in fact, landscape instability continued for sometime thereafter, even if only locally.

In this paper, we report on a forest bed in northern LowerMichigan, USA, which we interpret to have been buried by sedi-

Received 19 June 2012. Accepted 16 November 2012.

Paper handled by Associate Editor Timothy Fisher.

R.J. Schaetzl, C.H. Yansa, and M.D. Luehmann. Department of Geography, Michigan State University, East Lansing, MI 48824, USA.

Corresponding author: Randall J. Schaetzl (e-mail: soils@msu.edu).

483

Can. J. Earth Sci. 50: 483–493 (2013) dx.doi.org/10.1139/cjes-2012-0115 Published at www.nrcresearchpress.com/cjes on 5 December 2012.

Can

. J. E

arth

Sci

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Mic

higa

n St

ate

Uni

vers

ity o

n 08

/04/

13Fo

r pe

rson

al u

se o

nly.

ment that was deposited ca. 2000 cal years after final deglaciation.Radiocarbon ages on the wood, much of which consists of spruceand (or) larch trees buried in situ, as well as pollen and plantmacrofossil data recovered at the site, are used to refine the pa-leoenvironmental setting in northern Lower Michigan during theend of the Younger Dryas and earliest Holocene.

Study area

Quaternary geologyGreatlakean ice — the last glacial advance to enter Lower

Michigan — buried the Cheboygan bryophyte bed, 48 km north-west of our study site, at approximately 11 750 14C years BP (13 790–13 400 cal years BP) (Larson et al. 1994; Larson and Kincare 2009;Fig. 1). The outermost extent of this advance is not known withcertainty. Landscape linearity and fabric in this part of the state,however, suggest that the ice flowed southeasterly from the LakeMichigan basin, across the study area (Melhorn 1954; Spurr andZumberge 1956; Burgis 1981; Schaetzl 2001). Widespread stagna-tion and collapse is thought to have ensued, based on soils data(Schaetzl et al. 2000), presence of ponded water, esker and kametopography, and kettled areas (Schaetzl 2001). Large areas of redand pink, silty–clayey sediment in the region, especially east andsouth of the study site have been interpreted as having proglaciallake origins, and in some places associated with a stagnant Great-lakean ice margin (Schaetzl et al. 2000). The immediate post-Greatlakean landscape in northern Lower Michigan probablyconsisted of isolated ice blocks, ponded water and perhaps scat-tered permafrost (Spurr and Zumberge 1956; Schaetzl 2008). Otherthan this cursory information, little else is known about the im-mediate post-glacial landscape of this part of Michigan.

During the retreat of the Greatlakean ice, a series of large pro-glacial lakes formed in the Great Lakes basin, some of whichonlapped the Lower Peninsula (Larson and Schaetzl 2001). GlacialLake Algonquin, the largest of these lakes, maintained high levelsbetween approximately 13 100 and 12 500 cal years BP (Harrison1972; Karrow et al. 1975; Futyma 1981; Larsen 1987; Farrand 1995;Krist and Schaetzl 2001; Schaetzl et al. 2002; Karrow 2004; Kincareand Larson 2009; Drzyzga et al. 2012). Although glacial Lake Algon-quin did not cover the study site per se, Milligan Creek, whichflows near the site, was tributary to it, entering the lake by way ofa narrow embayment (Fig. 2).

The Tower site — the focus of our research — (informal name)is a shrub wetland dominated by speckled alder (Alnus incana spp.rugosa) and various, early successional tree species tolerant of wetconditions. It lies within the Mackinac State Forest in northernLower Michigan and is located at 45°22=10==N, and 84° 21= 35==W, atan elevation of 228 m above sea level. Specifically, it is situated5.4 km west-northwest of the village of Tower, in the N ½, NW¼,NW ¼, Section 6, T 34 N, R 1 E, about 100 m south of Kisser Road,which runs east–west across the northern boundary of the site(Fig. 1). A small stream, Milligan Creek, meanders northerly,across the flat, sandy wetland, ≈230–240 m to the west (Fig. 2); thestream formerly drained into glacial Lake Algonquin, whose high-est shoreline is <2 km away, to the northeast.

Although the site is located within a wetland, standing water isuncommon because of the close vicinity of Milligan Creek, whichhas a small incised valley. Poorly and very poorly drained, sandysoils, commonly with horizontally stratified and interbeddedmuck deposits, and with 10–50 cm of highly decomposed organicsoil materials at the surface, dominate the site.

Fig. 1. Locations of sites, dated and published, in the Midwest USA, where glacigenic or other sediments buried an established vegetationcommunity. Also shown are the location of the study site and various estimates of the location of the Greatlakean ice margin.

484 Can. J. Earth Sci. Vol. 50, 2013

Published by NRC Research Press

Can

. J. E

arth

Sci

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Mic

higa

n St

ate

Uni

vers

ity o

n 08

/04/

13Fo

r pe

rson

al u

se o

nly.

MethodsThe Tower site was first brought to our attention as the area was

being mapped by Natural Resources Conservation Service person-nel in 1997; they had discovered buried wood at ≈1 m depth whileaugering in this area. A visit to the site in 1997 by R.J.S. confirmedthe presence of thewood. At this time, several 1–2 cm size (diameter)wood fragments were recovered from a depth of ≈1 m, using ahand auger. The largest of these fragments was dated by BetaAnalytic Inc. (Miami FL 33155) using standard accelerator massspectrometry (AMS) procedures.

We revisited the site in October 2011 and were able to again findand sample the buried wood at similar depths, after excavating alarge pit about 75 m east of the original site. The wood at thissecond site consisted mainly of horizontally lying tree trunks andbranches. Most of these logs were piled one upon the other, at85–70 cm depth, with most of the small branches having beensnapped from the trunks. We recovered several large pieces ofwood (trunks) and selected wood from the outermost rings for14C dating at the University of Illinois radiocarbon dating lab.Several additional wood samples were sent in for anatomical anal-ysis and identification at the U.S. Forest Products Laboratory at theUniversity of Wisconsin-Madison. One particularly notable piecewas a tree trunk (≈8 cm in diameter) that had been bent over intoan almost horizontal position, but had retained its roots in situ(Figs. 3A, 3B). One of the larger logs is shown in Fig. 3C. Radiocar-bon ages (Table 1) were calibrated (two � ranges and median ages)using Calib version 6.1.0.

The sediments in the exposed pit face were described and sam-pled by genetic horizon. All mineral soil samples, i.e., excludingthe organic (muck) samples, were air-dried, gently disaggregatedwith a mortar and pestle, and passed through a 2 mm sieve to re-move coarse fragments. Particle size analysis on the remaining fine-earth fraction was performed on a Malvern Mastersizer 2000 laser

particle size analyzer, after being shaken for 2 h in a water-basedsolution with [NaPO3]13·Na2O as the dispersant.

Samples for pollen and plant macrofossil analysis were recov-ered from the excavated pit as follows. Samples of the organicmaterial (bracketed by sand), above 73 cm depth, were collectedfrom the profile (Fig. 4), but the fossils from these lenses werelater determined to have been redeposited, as indicated by abra-sion, decay and survival of only the hardiest remains. However,the underlying peat was found to have been in situ and fossilifer-ous; it is the analysis of this peat that we report on. Specifically, wecaptured the completewood/peat deposit from 110 to 73 cmdepth,which bottomed in sand (116–110 cm), by taking a vertical corethat was later split and subsampled in the laboratory. Figure 3Dshows a grab sample of the Sphagnum peat from the Oi2 horizon(97–103 cm), where it is thickest and best preserved.

Methods used in fossil analysis followed established procedures(Faegri and Iverson 1975; Birks 2001). Sampling intervals wereevery 5 cm for pollen (1 cm3 volume) and plant macrofossils(50 cm3). The pollen sum of trees, shrubs and upland herbs (ex-cluding Cyperaceae, members of the sedge family) ranged from339 to 444 grains/sample. Macrofossil data are reported as totalcounts, except for leaves and stems of the moss Sphagnum, whichare reported on a relative scale of 0–10, as per standard practice.Most of the conifer needles were broken; those not whole werecounted as halves, quarters and eighths (based on length) andtheir totals divided by 2, 4, and 8, respectively. Macrofossil countswere entered and pollen percentages were calculated in Tiliaspreadsheets, and the resulting data plotted using a combinationof Tiliagraph, TGView, and Adobe Illustrator. The CONISS clusteranalysis program was used to distinguish the plant macrofossiland pollen zones in the diagrams (Grimm 1987). Plant taxonomyand habitat information are based on Voss (1972).

Fig. 2. Detailed map of the topography and hydrology of the area near the study site. Also shown is the extent of glacial Lake Algonquin,Main phase (after Drzyzga et al. 2012), in light blue in online version.

84°18'0"W

84°18'0"W

84°22'0"W

84°22'0"W

84°26'0"W

84°26'0"W

45°22'0"N45°22'0"N

45°20'0"N

84°18'0"W

84°18'0"W

84°22'0"W

84°22'0"W

84°26'0"W

84°26'0"W

45°22'0"N45°22'0"N

45°20'0"N

Gokee Creek

Weed Creek

Adair Creek

Milligan C

reek

Black River

Milligan Creek

Gokee Creek

Weed Creek

Adair Creek

Milligan C

reek

Black River

Milligan Creek

Study siteStudy site

Lake Algonquin - Main phase

OsmunLake

Osmun Lake esker

Afton Drumlin Field

Black RiverEmbayment

Lake Algonquin - Main phase

OsmunLake

Osmun Lake esker

Afton Drumlin Field

Black RiverEmbayment

N2 km

High

Low

Elevation

Schaetzl et al. 485

Published by NRC Research Press

Can

. J. E

arth

Sci

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Mic

higa

n St

ate

Uni

vers

ity o

n 08

/04/

13Fo

r pe

rson

al u

se o

nly.

Fig. 3. Photographs of wood and peat observed at, and recovered from, the Tower site. (A) A small tree that had been bent over, but hadretained some of its roots. (B) The same tree, in situ, before being removed from the pit. (C) One of the larger logs recovered from the pit. Theend was sawn off to remove it from the burying sediment. (D) Minimally decomposed, laminated peat from the Oi2 horizon (97–103 cm),found below the buried wood layer.

A B

DC

Table 1. Radiocarbon ages on buried wood from the Tower site. All ISGS dates are based on the dating of logs overlying each other within onehand-dug pit, which was situated about 75 m east of the hand auger location.

Sample Sample typeSample depth(cm)

Uncalibrated 14C age(14C years BP)

Calibrated age range(cal years BP)a

Median calibrated age(cal years BP)

ISGS-6840 Log, recovered in situ, from pit 36 2570±70 2430–2790 2630ISGS-6841 Log, recovered in situ, from pit 73–80 9660±70 11 210–10 770 10 910ISGS-6842 Log, recovered in situ, from pit 73–80 9200±70 10 520–10 230 10 340ISGS-6843 Log, recovered in situ, from pit 73–80 9590±70 11 170–10 720 10 880ISGS-6844 Log, recovered in situ, from pit 73–80 9390±70 10 790–10 400 10 620Beta-108470 Wood fragment, recovered using hand

auger from a location near pit≈100 10 108±120 12 220–11 340 11 720

aCalibrations using intcal09.14c after Stuiver and Reimer (1993) and Reimer et al. (2009) and showing two � ranges.

486 Can. J. Earth Sci. Vol. 50, 2013

Published by NRC Research Press

Can

. J. E

arth

Sci

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Mic

higa

n St

ate

Uni

vers

ity o

n 08

/04/

13Fo

r pe

rson

al u

se o

nly.

Results, interpretations, and discussion

Wood radiocarbon dates and analysisMost of the buried logs were found to be lying with a south-

southwest–north-northeast orientation, and many had smallbranches that had been broken off of main trunks. The woodsamples collected at the Tower site are Picea (spruce) or Larix(larch/tamarack), based on the wood anatomical analysis, whichcannot distinguish between these two genera of Pinaceae. Be-cause both spruce and larchwere identified from pollen and plantmacrofossils (Table 2), the wood could be of either or both of thesetaxa.

Of the six large wood pieces that we dated (Table 1), one age(ISGS-6840, 2570 ± 70 14C years BP; ISGS, Illinois State GeologicalSurvey) is considerably younger. This wood fragment was isolatedfrom the others, occurring at 36 cm depth, whereas the otherdated-wood samples were from larger, intact, horizontally lyinglogs at 83–73 cm depth. We argue that this fragment (ISGS-6840)was likely contaminated from the many younger (some probablymodern) rootlets that had penetrated it. Efforts were made toremove penetrating rootlets before submission to the 14C-datinglab, but given that not all were removed, the resulting datemay bea younger-than age. The four other wood samples (ISGS-6841,-6842, -6843, and -6844) are from a “log pile” in the same hand-dugpit, but at greater depth, andwere exceedingly well preserved andyielded, what we interpret to be, accurate ages, all clusteredwithin a relatively narrow age range. The top of the wood/peatdeposit (80–73 cm depth), also captured in the core, is directlydated by these four different logs (ISGS-6841, -6842, -6843, and-6844), which returned four ages with medians that range from10 910 to 10 340 cal years BP (Table 1).

Given this relative close proximity of age (in a geologic sense) and,more significantly, the logs touching and piled upon each other, weinterpret this to indicate that these trees were toppled simultane-ously. Supporting evidence includes the prone position of the logs,most ofwhich occur in one layer, aswell as the lack ofwood and logsin the overlying sediment (except for wood at 36 cm depth thatyielded an anomalously younger age). The event that toppled thetrees occurred between 10 910 and 10 340 cal years BP, based on theoutermostmedian calibrated ages of the four log samples at 80–73 cmdepth (Table 1). These dates provide maximum-limiting ages forthe demise of the trees and their burial. We chose the youngest ofthese median ages, 10 340 cal years BP as the “most probable” agefor this event, given that older wood can be found in wetlandsettings. After this event, relative quiescence ensued for the re-mainder of the Holocene, during which the uppermost mucklayer formed in association with the hydric conditions at the site.

We interpret the oldest age (Beta-108470, 12 220–11 340 cal yearsBP, median ≈11 720 cal years BP; Table 1), for a piece of woodrecovered in the vicinity at about 100 cm depth, with a hand augerseveral years previous, as evidence that woody vegetation hadbeen established at this location by at least 11 720 cal years BP. Thelowermost peat level, in the core analyzed for pollen and plantmacrofossils, most likely correlates to this time. Data from Larsonet al. (1994) indicate that Greatlakean ice advanced onto the LowerPeninsula between about 13 790–13 400 cal years BP. The oldestage from the Tower site (Beta-108470; Table 1) suggests that by≈1680–2070 years later, woody vegetation had become establishedat this site – only about 48 km to the southeast of the Cheboyganbryophyte bed (Larson et al. 1994; Fig. 1).

Fig. 4. Photographs of the pit face. (A) Sandy sediment covering 2–3 stacked and buried logs, laying horizontally at the base of the pit, and(B) detail of the overlying and burying sediments — interbedded sand and muck — above buried wood at the water line. Tape increments incentimetres.

A B

Schaetzl et al. 487

Published by NRC Research Press

Can

. J. E

arth

Sci

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Mic

higa

n St

ate

Uni

vers

ity o

n 08

/04/

13Fo

r pe

rson

al u

se o

nly.

Table 2. Descriptiona of the sediments exposed in the pit face.

Horizon Depth (cm) Composition Interpretation Color (moist) Texture StructureConsistence(moist) Roots

Oa 0–24 Highly decomposed muck Muck that has formed after theburial event, in associationwith the modern wetland

2.5Y 2.5/1 Muck Moderate–strong,coarse, granular

Friable–firm Few, fine and very fineroots, especially inthe upper 10 cm; veryfew roots in the lowerpart

Cg 24–32 Sand with muck (Oa) inclusions;common thin lamellae andcircular inclusions of Oa material

Mainly sandy sediment that waswashed in, burying the forest

10YR 5/3 Sand Single grained Loose ---

Oa= 32–51 Highly decomposed muck with thin,sand lamellae; inclusions of sandand wood fragments

Muck, with some sand, that waswashed in, burying the forest

10YR 2/1 Muck Moderate–mediumand thick, platy,parting to strong,fine and medium,angular–blocky

Firm Very few, very fine roots

Cg= 51–74 Sand with thin lamellae and circularinclusions of organic material

Mainly sandy sediment that waswashed in, burying the forest

10YR 5/3 Sand Single grained Loose ---

Oe 74–88 Muck with abundant small fibers;contains abundant logs inhorizontal position, and woodfragments

A mix of sediment from theoriginal forest floor and thedowned trees

5Y 2.5/1 Muck Strong, very thin andthin, platy

Friable ---

Oi1 88–97 Moderately decomposed, fibrouspeat

The original peat soil of theforest floor

5Y 2.5/1 Mucky peat Strong, very thin andthin, platy

Friable ---

Oi2 97–103 Fibrous and only slightlydecomposed peat

The original peat soil of theforest floor

2.5Y 3/2 Peat Moderate, thin andvery thin, platy

Friable ---

2Cg 103+ Gleyed sand Glacial outwash that underliesthe wetland

Gley 1: 3/5 GY Sand Single grained Very friable ---

aSediment descriptions follow standard NRCS (Natural Resources Conservation Service) protocols (Schoeneberger et al. 2002).

488Can

.J.EarthSci.V

ol.50,2013

Publish

edby

NRCResearch

Press

Can

. J. E

arth

Sci

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Mic

higa

n St

ate

Uni

vers

ity o

n 08

/04/

13Fo

r pe

rson

al u

se o

nly.

The estimated 1380 cal years captured in the fossiliferous part ofthe core, below the wood (110–73 cm), ≈11 720 (Beta-108470) to10 340 (ISGS-6842) cal years BP, equate to an estimate mean or-ganic sedimentation rate of 37.3 years/cm. This sedimentationrate was calculated solely to estimate ages that bracket the pollenand plant macrofossil zones and should not be used for compari-son to other site stratigraphies, because rates of peat formationwere undoubtedly spatially and temporally variable. This intervalwas, nonetheless, a time of landscape stability with little clasticinput to the site, during which a Spaghnum peat layer formed atthe surface. This peat/log deposit was then buried by ≈50 cm ofsand and muck (Table 2: Cg, Oa=, and Cg= horizons). After thispoint, organic materials continued to accumulate at the surface,during a second stable period of little or no clastic sedimentation.

Paleoenvironmental reconstruction for the Tower site:pollen data

The pollen (Fig. 5A) and plant macrofossil (Fig. 5B) records,based on analysis of the peat deposit, are bracketed in age from11 720 to 10 340 cal years BP. Three zones are identified in both ofthese paleobotanical records and are assigned ages according tothe calculated sedimentation rate discussed in the preceding sec-tion. The oldest part of this vegetation sequence correlates to thelatter part of the Younger Dryas (12 900–11 500 cal years BP), whencool conditions are reported for the Midwest (e.g., Yu 2000;Shuman et al. 2002; Leavitt et al. 2006; Gonzales and Grimm 2009;Gonzales et al. 2009; Denton et al. 2010; Gill et al. 2012). However,most of the paleovegetation record here, dates to the earliestHolocene. The climate at this time differed from modern – theLaurentide Ice Sheet to the north altered the atmospheric circu-lation (Krist and Schaetzl 2001; Vader et al. 2012), prevailing inso-lation caused greater seasonal extremes, and carbon dioxideconcentrations in the atmosphere were somewhat lower (Ruddi-man 2007).

Although the sequence of vegetation changes at the Tower sitereflects broader regional floristic patterns driven by climatechange and plant succession, the local wetland setting was animportant determinant of which plants lived at this site. In par-ticular, cool-adapted taxa could persist longer in the cool/moistmicroclimate of this location, than they could have on an upland.Given that the site is situated within a small (ca. 4.2 km2) wetland,it would have captured the pollen rain from a relatively small area(tens of km2), following the logic outlined in Jacobson andBradshaw (1981). Also, the plant macrofossils in the peat wouldhave been deposited from the plants that grew within the wet-land, given the hydrogeologic setting and their excellent physicalcondition.

Zone I (≈11 720–11 460 cal years BP), identified from the bottomof the peat core, is interpreted as a wetland where the vegetationwas transitioning from tundra to a “boreal” parkland during theend of the Younger Dryas (Figs. 5A, 5B). This interpretation issupported by the high pollen values of Betula (birch) and con-firmed by several fruits of this taxon. Unfortunately, the wingshad been removed from these fruits (which were otherwise wellpreserved), preventing species determination. However, the pol-len was probably that of the tundra shrub version (Betula nana)rather than of the tree (B. papyrifera), because birch trees are abun-dant with Pinus (pine) later, during the early Holocene, in Midwestpollen records (e.g., Webb et al. 1983). This “tundra birch” inter-pretation is supported by the co-occurrence of (1) relatively abun-dant Artemisia (wormwood/sage) pollen, a tundra or grasslandindicator, and (2) a few leaf fragments of Salix cf. S. herbacea(snowbed willow), the latter of which is a trademark tundra spe-cies found in northern Illinois, from 21 700 to 16 200 cal years BP(Curry and Yansa 2004; Curry et al. 2010; Curry and Petras 2011).

Taken together, these fossils probably capture the end of thetundra–boreal parkland phase, which is dated earlier at the TwoCreeks site in eastern Wisconsin (14 500–13 900 cal years BP), ap-

proximately 280 km to the southwest (Maher et al. 1998; Rechet al. 2011; Fig. 1). The age for this transitional vegetation (≈11 720–11 460 cal years BP) at the Tower site agrees with the timing ofplant succession in northern Lower Michigan, in that the earliertundra phase was captured at the Cheboygan bryophyte bed from13 790 to 13 400 cal years BP, before the Greatlakean readvance(Larson et al. 1994; Fig. 1). Similarly, Yu (2003) reported on sparsetundra vegetation on the Niagara Escarpment of southern On-tario, Canada, with an estimated age of ≈14 000 cal years BP. Theabsence of a pure tundra assemblage in the bottommost sedi-ments at the Tower site has two possible explanations: (1) thetundra fossil-bearing deposit had been eroded, and (or) (2) thisinitial phase of plant succession did not occur because the rapidlyameliorating climate permitted tree establishment shortly afterfinal deglaciation.

Trees were common at the Tower site from ≈11 720 to11 460 cal years BP, as indicated by the abundant needles and someseeds (which permit species level determinations) of Larix laricina(larch/tamarack), Picea glauca (white spruce), P. mariana (blackspruce), and Abies balsamea (balsam fir). Picea glauca, in particular,is the most common tree in the early postglacial successionwithin the formerly glaciated areas of eastern North America(Yansa 2006). Looking at the modern boreal forest as an analog(Ritchie 1987), black spruce and larch trees probably inhabited thewetland, withwhite spruce and balsamfir trees nearby, where theground was slightly higher and better drained. Larix laricina is anotoriously poor pollen producer and disperser (Faegri andIverson 1975); hence the very high needle count (e.g., 1059 for onesample; Fig. 5B) of this taxon provides amore accurate assessmentof how numerous trees of this species were in this wetland. Themacrofossil abundance (5112) of Chara oogonia (the reproductivestructure of the green algae), within the bottommost sample, andthe one above (2890) point to highly calcareous, shallow water atthe site, probably due to CaCO3 from within the calcareous tillsof the region.

The relatively high water levels at the site from ≈11 720 to11 460 cal years BP (Zone I; Figs. 5A, 5B), may reflect a broaderregional climate of cool and wet conditions during the YoungerDryas chronozone, as exemplified by paleobotanical data fromnortheastern Illinois (e.g., Curry et al. 2007; Gonzales and Grimm2009; Gonzales et al. 2009), and elsewhere in the Great Lakesregion (e.g., Webb et al. 1983, 2004; Kapp 1999; Panyushkina et al.2008). Gonzales et al.'s (2009) synthesis of Midwestern pollen re-cords to the south of our study area identified a brief resurgence ofPicea in the later part of the Younger Dryas, coeval with Zone I ofthe Tower site, which they attributed to cool and wet conditions.A spruce-larch-Sphagnum peatland existed not just in our studyarea, but also at Brewster Creek in northeastern Illinois duringthis time (Curry et al. 2007), supporting the interpretation of fairlyhigh levels of precipitation in the Midwest USA during theYounger Dryas, when compared to the present day. This interpre-tation is supported by tree-ring studies of a buried black sprucestand in northern Indiana dating to the mid-Younger Dryas,which contains stumps with adventitious roots designed to stabi-lize trees in water-logged conditions (Panyushkina et al. 2008).

Zone II (≈11 460–10 660 cal years BP) is interpreted as an openboreal forest with large patches of shallow wetlands dominatedby members of the Cyperaceae family, including at least five spe-cies of Carex (sedge) (Figs. 5A, 5B). The same conifers as beforedominated the arboreal vegetation; Picea glauca, Larix laricina andAbies balsamea were most abundant at the site during the earliestpart of the Holocene. Ulmus (elm) and Alnus incana spp. rugosa(speckled alder) became more common in the area (Fig. 5A); bothare excellent indicators of waterlogged soils. Several buds of Popu-lus tremuloides (trembling aspen; quaking aspen) confirm the pres-ence of this tree at the site (Fig. 5B), which today is a commondeciduous tree in the boreal forest (Ritchie 1987). Peat is welldeveloped in this zone; it exists as an almost pure Sphagnum layer

Schaetzl et al. 489

Published by NRC Research Press

Can

. J. E

arth

Sci

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Mic

higa

n St

ate

Uni

vers

ity o

n 08

/04/

13Fo

r pe

rson

al u

se o

nly.

Fig. 5. (A) Pollen percentage diagram and (B) plant macrofossil count diagram of the most common taxa from analysis of the cored peatdeposit, which ranges in age from (bottom to top) ≈11 720 to 10 340 cal years BP. Notes: (1) Sphagnum moss leaf and stem abundance (B) wasassessed on a relative scale of 0–10, and (2) the location of the buried wood bed is shown graphically within these diagrams (A and B).

75

80

85

90

95

100

105

110

Dep

th (c

m)

Picea gl

auca

Picea m

arian

a

Picea (n

eedles

)

Larix

laric

ina

Larix

(nee

dles)

Abies b

alsam

ea

Abies (

need

les)

Alnus (

strob

iles)

Populu

s trem

uloide

s (bu

ds)

BetulaSali

x cf. S

. herb

acea

e (lea

f)

Mudfla

t herb

s

Carex

Najas f

lexilis

Potamog

eton

Sphag

num (le

aves

/stems)

200 400 2000 4000 20 40 60 20 20 20 20 20 2000 6000

Chara

(oogo

nia)

Analyst: C. Yansa# / 50 cm 3

ZONE

I

II

III

20 20 40 20 20 40 20 20 40 60 80

Fraxinus

nigra

-type

Ambrosia

-type

Trees and Shrubs

Picea

Abies

Larix

Pinus

Betula

Corylus

Populu

s

Quercu

s

Salix

Ulmus

Alnus

Artemisa

Poace

ae

Cypera

ceae

Typha

latifo

lia

5X exaggeration

75

80

85

90

95

100

105

110

Dep

th (c

m)

ZONE

I

II

III

Herbs Aquatics

Trees and Shrubs AquaticsHerbs

A

ety

misa

Poacacea

C

atifii off

N

I

ae

TyTTpyyha

latifi

ZON

II

eae

Cypera

ceae

B

xixx lill s ton m(le

ago

ni

asflff e

x

PoPPt

exixx

otamS

is eto m(

ogo

ZONE

III

amog

et

SpSShag

num

Chararr

(oog

Z

490 Can. J. Earth Sci. Vol. 50, 2013

Published by NRC Research Press

Can

. J. E

arth

Sci

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Mic

higa

n St

ate

Uni

vers

ity o

n 08

/04/

13Fo

r pe

rson

al u

se o

nly.

with minimal or no decay, from 100–90 cm depth (Fig. 3D). Thistype of boreal–temperate forest mix, lacking an exact modernanalog, has been reported elsewhere in the Upper Midwest forthis time, at the beginning of the Holocene (e.g., Delcourt andDelcourt 1991; Kapp 1999; Yu 2000; Williams et al. 2001; Curryet al. 2007; Hupy and Yansa 2009; Gill et al. 2012).

During Zone III (≈10 660–10 340 cal years BP), species dominanceshifted toward Pinus (pine), with lesser amounts of the cold-adapted Picea, Abies, and Larix species (Fig. 5A). The decline inUlmus pollen values, low Chara oogonia counts and correspondingslight pollen increases of Quercus (oak) and Ambrosia-type (rag-weed), all suggest that the water table had fallen by this time atthe Tower site. Sphagnum peat accumulation had also declined, asindicated by lesser amounts of the megaspores, leaves and stemfragments of this bryophyte (Fig. 5B). There was still some stand-ing water at the site, however, as indicated by fruits of the sub-merged aquatics Najas flexilis (naiad) and at least five species ofPotamogeton (pondweed) that imply shallow water (<1.5 m). Thepollen (Fig. 5A) and seed abundance of Typha latifolia (commoncattail) is a good indicator of shoreline habitats that expand whenwetlands drain.

All of these fossil indicators suggest a trend toward drying andpossibly warming conditions, with the shrinking of wetlandcoverage. This transition from Picea- to Pinus-dominated forest,associated with a drying climate causing a water-table loweringduring the earliest Holocene, has been noted in other pollen re-cords throughout the Upper Midwest (Yu 2000; Curry et al. 2007;Gonzales and Grimm 2009; Gill et al. 2012). For example, this shiftoccurred at the Twiss Marl Pond site in southern Ontario (Yu2000), 380 km southeast of the Tower site, at approximately thesame time as at the Tower site. The early Holocene climate recon-structed for the Midwest, based on pollen data, entailed warmerwinters and summers than previous, with summer precipitationamounts similar to modern, with slightly drier winters (Gonzaleset al. 2009). Thewarmer temperatures and reduced snowfall couldaccount for the overall water table lowering of the Tower sitewetland in the early Holocene, so that standing water probablyoccurred just seasonally after snowmelt and heavy rains.

Paleoenvironmental reconstruction for the Tower site:sedimentological data

The sand–muck interstratified sediment that buries and inter-fingers with the flat-lying wood and logs at the Tower site iscrudely water-sorted (Fig. 4). Muck interfingers with this sandysediment in a manner that suggests that it did not form by in situwetland pedogenic processes of surficial organic matter accumu-lation, decomposition and paludification (depths <74 cm; Table 2).Rather, we suggest that this material was deposited fluvially,which seems reasonable, given the nearby location of MilliganCreek, and the study area's regional setting within a through-flowing wetland. Near the top of the sediment column, anotherlayer of muck is currently forming in association with the mod-ern, i.e., post-burial, wetland (Table 2: Oa horizon). This upper,surficial muck horizon is associated with landscape stability, post-burial pedogenesis, and continued paludification (e.g., Kroetschet al. 2011), all of which post-date the toppled logs and wood.

The mixed but often quite sharp stratigraphic contacts suggestthat the depositional event that followed (or coincided with) thedowned wood was associated with fairly rapid and turbulent flow.Small pieces of intact, perhaps then-frozen, muck are embeddedwithin crudely laminated sand (Table 2; Fig. 4). Microscopic anal-ysis of the plant remains from some of these upper muck strataclearly indicate abrasion, with survival of only the hardiest plantfragments observed. Taken collectively, these data strongly sug-gest that the stratified muck deposits are a type of transportedsediment, not one that had formed in situ or one that had formedslowly, by small, periodic additions of clastic material.

The exact mechanism by which the trees at the Tower site be-came toppled and later buried is, however, unclear. High watertables and organic soils in wetlands like this would have forcedthe trees to have been shallowly rooted, rendering them vulnera-ble to tilting and uprooting (Schaetzl et al. 1989; Nicoll and Ray1996; Hautala and Vanha-Majamaa 2006). Alluvial sediment couldhave been deposited on top of a sequence of tree trunks that hadalready been lying on the soil surface, perhaps due to a localwindthrow event, or the force of flowing water may have beensufficient to topple them.

Based on the integration of sedimentological and paleobotani-cal data (Table 2, Figs. 4, 5), we can report only that a suite ofdowned trees that had been growing in a forested coniferwetland,situated within a drier pine-dominated upland forest, were laterburied by alluvium, probably deposited by fast-flowing water. Thewater deposited ripped up, sandy sediment, including consider-able amounts of peat andmuck, probably from the wetland itself.The sediment-laden water may have even knocked downmany ofthe shallowly rooted spruce/larch trees and bent over the smallersaplings, and along with any buried trees already lying on thesurface. In the end, it buried them in several tens of cm of sedi-ment (Figs. 3A, 3B). Wood in the pit showed clear evidence of treefragments and logs all having been laid prone (Fig. 4) in a south-southwest–north-northeast direction, suggestive of paleoflow di-rections. The sandy/mucky sediment and some additional, smallpieces of woodwere all deposited in crudely stratified form, abovethis bed of wood, tree trunks and broken branches. The sedimen-tology, coupledwith the fact that trees— some as large as 8–10 cmin diameter — had their branches broken off, both suggest rapidand aggressive flowduring and (or) immediately after the burial ofthe trees. Stratigraphic evidence from the site further shows that,after this event, the area quickly transitioned to a stable wetlandin which muck accumulated at the surface.

ConclusionsThis study reports in detail on the paleoenvironment of the

early, post-glacial landscape in northern Lower Michigan. Afterfinal deglaciation, the vegetation here went from a tundra – openboreal parkland (�11 720–11 460 cal years BP) at the end of theYounger Dryas, to an open boreal forest – Spaghnum peatland(�11 460–10 660 cal years BP), and finally to a drier, more pine-dominated forest (�10 660–10 340 cal years BP) during the earliestHolocene. Specifically, pine and to a lesser extent balsam fir weresituated in the surrounding uplands, whereas black spruce andlarch persisted in the wetland. Trees in wetlands like this one areprone to toppling, and we report on just such an event that oc-curred here, about 1000–2000 years after initial deglaciation.Spruce and (or) larch trees at the site were knocked down andimmediately buried by flowing water, or were variously lying onthe surface (some possibly by windthrow) and later buried byalluvium.

Data and details about the deglacial Midwestern landscape arefew, largely because of the scarcity of 14C-datable materials fromthis time. We believe it was a land of melting permafrost, rapidfluvial erosion, ponded water, and strong winds (e.g., Claytonet al. 2001, 2008; Krist and Schaetzl 2001; Curry and Petras 2011).However, pollen and plantmacrofossil analysis reveals that plantsdid become established in this cold landscape. The Tower BuriedForest site provides a glimpse into this same, periodically unsta-ble, landscape roughly 2000 years after final deglaciation.

AcknowledgementsWe thank TomWilliams of the Natural Resources Conservation

Service for leading us to the site and assisting with sampling;without him, this study would never have happened. GrahameLarson was an invaluable sounding board and consultant duringthe various phases of this research. Comments by JimKnox, a CJES

Schaetzl et al. 491

Published by NRC Research Press

Can

. J. E

arth

Sci

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Mic

higa

n St

ate

Uni

vers

ity o

n 08

/04/

13Fo

r pe

rson

al u

se o

nly.

Associate Editor, and two anonymous reviewers helped us to re-fine our interpretations and improve the manuscript.

ReferencesBirks, H.H. 2001. Chapter 4, Plant macrofossils. In Tracking Environmental

Change Using Lake Sediments: Volume 3, Terrestrial, Algal, and SiliceousIndicators. Edited by J.P. Smol, H.J.B. Birks, and W.M. Last. Kluwer AcademicPress, Dordrecht, The Netherlands. pp. 49–74.

Broecker, W.S., and Farrand, W.R. 1963. Radiocarbon age of the Two Creeksforest bed, Wisconsin. Geological Society of America Bulletin, 74: 795–802.doi:10.1130/0016-7606(1963)74[795:RAOTTC]2.0.CO;2.

Burgis, W.A. 1977. Late-Wisconsinan history of northeastern lower Michigan.Ph.D. Diss., University of Michigan. p. 396.

Burgis, W.A. 1981. Late-Wisconsinan History of Northeastern Lower Michigan.Guidebook, 30th Midwest Friends of the Pleistocene field trip. University ofMichigan, Ann Arbor. p. 110.

Clayton, L., Attig, J.W., and Mickelson, D.M. 2001. Effects of late Pleistocenepermafrost on the landscape of Wisconsin, USA. Boreas, 30: 173–188. doi:10.1080/030094801750424111.

Clayton, L., Attig, J.W., Ham, N.R., Johnson, M.D., Jennings, C.E., andSyverson, K.M. 2008. Ice-walled-lake plains: Implications for the origin ofhummocky glacial topography in middle North America. Geomorphology,97: 237–248. doi:10.1016/j.geomorph.2007.02.045.

Curry, B.B., and Petras, J. 2011. Chronological framework for the deglaciation ofthe Lake Michigan lobe of the Laurentide Ice Sheet from ice-walled lakedeposits. Journal of Quaternary Science, 26: 402–210. doi:10.1002/jqs.1466.

Curry, B.B., and Yansa, C.H. 2004. Stagnation of the Harvard sublobe (LakeMichigan lobe) in northeastern Illinois, USA, from 24,000 to 17,600 BP andsubsequent tundra-like ice-marginal paleoenvironments from 17,600 to15,700 BP. Géographie Physique et Quaternaire, 58: 305–321.

Curry, B.B., Grimm, E.C., Slate, J.E., Hansen, B.C.S., and Konen, M.E. 2007. Thelate-glacial and early Holocene geology, paleoecology, and paleohydrology ofthe Brewster Creek site, a proposed wetland restoration site, Pratt's WayneWoods Forest Preserve, and James “Pate” Philip State Park, Bartlett, Illinois.Illinois Department of Natural Resources, Illinois State Geological SurveyCircular 571. p. 51.

Curry, B.B., Konen, M.E., Larson, T.H., Yansa, C.H., Hackley, K.C.,Alexanderson, H., and Lowell, T.V. 2010. The DeKalbmounds of northeasternIllinois as archives of deglacial history and postglacial environments. Qua-ternary Research, 74: 82–90. doi:10.1016/j.yqres.2010.04.009.

Delcourt, H.R., and Delcourt, P.A. 1991. Quaternary Ecology: A PaleoecologicalPerspective. Chapman and Hall, London.

Denton, G.H., Anderson, R.F., Toggweiler, J.R., Edwards, R.L., Schaefer, J.M., andPutnam, A.E. 2010. The last glacial termination. Science, 328: 1652–1656.doi:10.1126/science.1184119.

Drzyzga, S.A., Shortridge, A.M., and Schaetzl, R.J. 2012. Mapping the stages ofGlacial Lake Algonquin in Northern Michigan, USA, and nearby Ontario,Canada, using an isostatic rebound model. Journal of Paleolimnology, 47:357–371. doi:10.1007/s10933-011-9550-9.

Faegri, K., and Iverson, J. 1975. Textbook of Pollen Analysis. Hafner Press, NewYork.

Farrand, W.R. 1995. The Pleistocene glacial record in the area of Alpena, Mont-morency and Presque Isle Counties, Michigan. In Karst Geology of the North-east Lower Peninsula, Michigan. Edited by T.J. Black. Michigan BasinGeological Society, Department of Geological Science, Michigan State Uni-versity. pp. 35–41.

Futyma, R.P. 1981. The northern limits of glacial Lake Algonquin in Upper Michi-gan. Quaternary Research, 15: 291–310. doi:10.1016/0033-5894(81)90032-6.

Gill, J.L., Williams, J.W., Jackson, S.T., Donnelly, J.P., and Schellinger, G.C. 2012.Climatic and megaherbivory controls on late-glacial vegetation dynamics: anew, high-resolution, multi-proxy record from Silver Lake, Ohio. QuaternaryScience Reviews, 34: 66–80. doi:10.1016/j.quascirev.2011.12.008.

Gonzales, L.M., and Grimm, E.C. 2009. Synchronization of late-glacial vegetationchanges at Crystal Lake, Illinois, USA with the North Atlantic Event Stratig-raphy. Quaternary Research, 72: 234–245. doi:10.1016/j.yqres.2009.05.001.

Gonzales, L.M., Williams, J.W., and Grimm, E.C. 2009. Expanded response-surfaces: a new method to reconstruct paleoclimates from fossil pollenassemblages that lack modern analogues. Quaternary Science Reviews,28: 3315–3332. doi:10.1016/j.quascirev.2009.09.005.

Grimm, E.C. 1987. CONISS - A Fortran-77 program for stratigraphically con-strained cluster-analysis by the method of incremental sum of squares. Com-puters and Geosciences, 13: 13–35. doi:10.1016/0098-3004(87)90022-7.

Harrison, J.E. 1972. Quaternary Geology of the North Bay-Mattawa Region. Geo-logical Survey of Canada Paper 71-26. p. 36.

Hautala, H., and Vanha-Majamaa, I. 2006. Immediate tree uprooting afterretention-felling in a coniferous boreal forest in Fennoscandia. CanadianJournal of Forest Research, 36(12): 3167–3172. doi:10.1139/x06-193.

Hupy, C.M., and Yansa, C.H. 2009. The last 17,000 years of vegetation history. InMichigan Geography and Geology. Edited by R.J. Schaetzl, J.T. Darden, and D.Brandt. Pearson Custom Publishing, Boston, MA. pp. 91–105.

Jacobson, G.L., and Bradshaw, R.H.W. 1981. The selection of sites for paleoveg-etational studies. Quaternary Research, 16: 80–96. doi:10.1016/0033-5894(81)90129-0.

Johnson, W.H., Hansel, A.K., Bettis, E.A., III, Karrow, P.F., Larson, G.J.,Lowell, T.V., and Schneider, A.H. 1997. Late Quaternary temporal and eventclassifications, Great Lakes Region, North America. Quaternary Research, 47:1–12. doi:10.1006/qres.1996.1870.

Kaiser, K.F. 1994. Two Creeks interstadial dated through dendrochronology andAMS. Quaternary Research, 42: 288–298. doi:10.1006/qres.1994.1079.

Kapp, R.O. 1999. Michigan Late Pleistocene, Holocene and Presettlement Vege-tation and Climate. In Retrieving Michigan's Buried Past: The Archaeology ofthe Great Lakes State. Edited by J.R. Halsey, and M.D. Stafford. CranbrookInstitute of Science, Bloomfield Hills, MI. pp. 30–58.

Karrow, P.F. 2004. Algonquin-Nipissing Shorelines, North Bay, Ontario. Géogra-phie Physique et Quaternaire, 58: 297–304.

Karrow, P.F., Anderson, T.W., Clarke, A.H., Delorme, L.D., and Sreenivasa, M.R.1975. Stratigraphy, paleontology, and age of Lake Algonquin sediments insouthwestern Ontario, Canada. Quaternary Research, 5: 49–87. doi:10.1016/0033-5894(75)90048-4.

Kincare, K., and Larson, G.J. 2009. Evolution of the Great Lakes. In MichiganGeography and Geology. Edited by R.J. Schaetzl, J.T. Darden, and D. BrandtD.Pearson Custom Publishing, Boston, MA. pp. 174–190.

Krist, F., Jr., and Schaetzl, R.J. 2001. Paleowind (11,000 BP) directions derived fromlake spits in northern Michigan. Geomorphology, 38: 1–18. doi:10.1016/S0169-555X(00)00040-4.

Kroetsch, D.J., Geng, X., Chang, S.X., and Saurette, D.D. 2011. Organic soils ofCanada: Part 1. Wetland organic soils. Canadian Journal of Soil Science, 91:807–822.

Larsen, C.E. 1987. Geological history of glacial Lake Algonquin and the upperGreat Lakes. United States Geological Survey Bulletin 1801. p. 36.

Larson, G.J., and Kincare, K. 2009. Late Quaternary History of the Eastern Mid-Continent Region, USA. In Michigan Geography and Geology. Edited byR.J. Schaetzl, J.T. Darden, and D. Brandt. Pearson Custom Publishing, Boston,MA. pp. 69–90.

Larson, G.J., and Schaetzl, R.J. 2001. Origin and evolution of the Great Lakes.Journal of Great Lakes Research, 27: 518–546. doi:10.1016/S0380-1330(01)70665-X.

Larson, G.J., Lowell, T.V., and Ostrom, N.E. 1994. Evidence for the Two Creeksinterstade in the LakeHuron basin. Canadian Journal of Earth Sciences, 31(5):793–797. doi:10.1139/e94-072.

Leavitt, S.W., Panyushkina, I.P., Lange, T., Wiedenhoeft, A., Cheng, L.,Hunter, R.D., Hughes, J., Pranschke, F., Schneider, A.F., Moran, J., andStieglitz, R. 2006. Climate in the Great Lakes region between 14,000 and 4000years ago from isotopic composition of conifer wood. Radiocarbon, 48: 205–217.

Lowell, T.V., Larson, G.J., Hughes, J.D., and Denton, G.H. 1999. Age verificationof the Lake Gribben forest bed and the Younger Dryas Advance of the Lau-rentide Ice Sheet. Canadian Journal of Earth Sciences, 36(3): 383–393. doi:10.1139/e98-095.

Maher, L.J., and Mickelson, D.M. 1996. Palynological and radiocarbon evidencefor deglaciation events in the Green Bay lobe, Wisconsin. Quaternary Re-search, 46: 251–259. doi:10.1006/qres.1996.0064.

Maher, L.J., Miller, N.G., Baker, R.G., Curry, B.B., and Mickelson, D.M. 1998.Paleobiology of the sand beneath the Valders diamicton at Valders, Wiscon-sin. Quaternary Research, 49: 208–221. doi:10.1006/qres.1997.1957.

Melhorn, W.N. 1954. Valders glaciation of the southern peninsula of Michigan.PhD Dissertation, Department of Geology. University of Michigan. p. 177.

Nicoll, B.C., and Ray, D. 1996. Adaptive growth of tree root systems in responseto wind action and site conditions. Tree Physiology, 16: 891–898. doi:10.1093/treephys/16.11-12.891.

Panyushkina, I.P., Leavitt, S.W., Thompson, T.A., Schneider, A.F., and Lange, T.2008. Environment and paleoecology of a 12 kamid-North American YoungerDryas forest chronicled in tree rings. Quaternary Research, 70: 433–441.doi:10.1016/j.yqres.2008.08.006.

Rech, J.A., Nekola, J.C., and Pigati, J.S. 2011. Radiocarbon ages of terrestrial gas-tropods extend duration of ice-free conditions at the Two Creeks forest bed,Wisconsin, USA. Quaternary Research, 77: 289–292.

Reimer, P.J., Baillie, M.G.L., Bard, E., Bayliss, A., Beck, J.W., Blackwell, P.G.,Bronk Ramsey, C., Buck, C.E., Burr, G.S., Edwards, R.L., Friedrich, M.,Grootes, P.M., Guilderson, T.P., Hajdas, I., Heaton, T.J., Hogg, A.G.,Hughen, K.A., Kaiser, K.F., Kromer, B., McCormac, F.G., Manning, S.W.,Reimer, R.W., Richards, D.A., Southon, J.R., Talamo, S., Turney, C.S.M.,van der Plicht, J., and Weyhenmeyer, C.E. 2009. IntCal09 and Marine09 ra-diocarbon age calibration curves, 0–50,000 years cal BP. Radiocarbon, 51:1111–1150.

Ritchie, J.C. 1987. Postglacial Vegetation of Canada. Cambridge University Press,Cambridge, UK.

Ruddiman, W.F. 2007. Earth's Climate: Past and Future, 2nd ed. W.H. Freemanand Company. New York, NY.

Schaetzl, R.J. 2001. Late Pleistocene ice flow directions and the age of glaciallandscapes in northern Lower Michigan. Physical Geography, 22: 28–41.

Schaetzl, R.J. 2008. The distribution of silty soils in the Grayling Fingers region ofMichigan: Evidence for loess deposition onto frozen ground. Geomorphol-ogy, 102: 287–296. doi:10.1016/j.geomorph.2008.03.012.

Schaetzl, R.J., Johnson, D.L., Burns, S.F., and Small, T.W. 1989. Tree uprooting:

492 Can. J. Earth Sci. Vol. 50, 2013

Published by NRC Research Press

Can

. J. E

arth

Sci

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Mic

higa

n St

ate

Uni

vers

ity o

n 08

/04/

13Fo

r pe

rson

al u

se o

nly.

review of terminology, process, and environmental implications. CanadianJournal of Forest Research, 19(1): 1–11. doi:10.1139/x89-001.

Schaetzl, R.J., Krist, F., Rindfleisch, P., Liebens, J., andWilliams, T. 2000. Postgla-cial landscape evolution of northeastern Lower Michigan, interpreted fromsoils and sediments. Annals of the Association of American Geographers, 90:443–466. doi:10.1111/0004-5608.00204.

Schaetzl, R.J., Drzyzga, S.A., Weisenborn, B.N., Kincare, K.A., Lepczyk, X.C.,Shein, K.A., Dowd, C.M., and Linker, J. 2002. Measurement, correlation, andmapping of Glacial Lake Algonquin shorelines in northern Michigan. Annalsof the Association of American Geographers, 92: 399–415. doi:10.1111/1467-8306.00296.

Schoeneberger, P.J., Wysocki, D.A., Benham, E.C., and Broderson, W.D. 2002.Field book for describing and sampling soils, Version 2.0. Natural ResourcesConservation Service, National Soil Survey Center, Lincoln, NE.

Shuman, B., Webb, T., III, Bartlein, P, andWilliams, J.W. 2002. The anatomy of aclimatic oscillation: vegetation change in eastern North America during theYounger Dryas chronozone. Quaternary Science Reviews, 21: 1777–1791. doi:10.1016/S0277-3791(02)00030-6.

Spurr, S.H., and Zumberge, J.H. 1956. Late Pleistocene features of Cheboygan andEmmet Counties, Michigan. American Journal of Science, 254: 96–109. doi:10.2475/ajs.254.2.96.

Stuiver, M., and Reimer, P.J. 1993. Extended 14C data base and revised CALIB 3.014C age calibration program. Radiocarbon, 35: 215–230.

Vader, M.J., Zeman, B.K., Schaetzl, R.J., Anderson, K.L., Walquist, R.W.,Freiberger, K.M., Emmendorfer, J.A., and Wang, H. 2012. Proxy evidence foreasterly winds in Glacial Lake Algonquin, from the Black River Delta in

northern Lower Michigan. Physical Geography, 33: 252–268. doi:10.2747/0272-3646.33.3.252.

Voss, E.G. 1972. Michigan Flora: A Guide to the Identification and Occurrence ofthe Native and Naturalized Seed Plants of the State. Cranbrook Institute ofScience, Bloomfield Hills, MI.

Webb, T., III, Cushing, E.J., and Wright, H.E., Jr. 1983. Holocene Changes in theVegetation of the Midwest. In Late-Quaternary Environments of the UnitedStates, Volume 2. Edited by H.E. Wright, Jr. The Holocene. University of Min-nesota Press, Minneapolis.

Webb, T., III, Shuman, B., and Williams, J.W. 2004. Climatically forced vegeta-tion dynamics in eastern North America during the Late Quaternary period.In The Quaternary Period in the United States. Edited by A.R. Gillespie,S.C. Porter, and B.F. Atwater. Elsevier, Amsterdam, The Netherlands. pp.459–478.

Williams, J.W., Shuman, B.N., and Webb, T., III. 2001. Dissimilarity analyses oflate-Quaternary vegetation and climate in eastern North America. Ecology,82: 3346–3362. doi:10.1890/0012-9658(2001)082[3346:DAOLQV]2.0.CO;2.

Yansa, C.H. 2006. The timing and nature of Late Quaternary vegetation changesin the northern Great Plains, USA and Canada: A re-assessment of the sprucephase. Quaternary Science Reviews, 25: 263–281. doi:10.1016/j.quascirev.2005.02.008.

Yu, Z. 2000. Ecosystem response to Lateglacial and early Holocene climate oscil-lations in the Great Lakes region of North America. Quaternary ScienceReviews, 19: 1723–1747. doi:10.1016/S0277-3791(00)00080-9.

Yu, Z. 2003. Late Quaternary dynamics of tundra and forest vegetation in thesouthern Niagara Escarpment, Canada. New Phytologist, 157: 365–390. doi:10.1046/j.1469-8137.2003.00678.x.

Schaetzl et al. 493

Published by NRC Research Press

Can

. J. E

arth

Sci

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

Mic

higa

n St

ate

Uni

vers

ity o

n 08

/04/

13Fo

r pe

rson

al u

se o

nly.