Timing, provenance, and implications of two MIS 3 advances of the Laurentide Ice Sheet into the Upper Mississippi River Basin, USA
Introduction
Knowledge of glacial extent and timing are necessary to model Laurentide Ice Sheet (LIS) dynamics during its build up, acme, and decline. Unfortunately, terrestrial records of North American glaciation leading up to the Last Glacial Maximum (LGM, 19–26.5 ka, Clark et al., 2009) are sparse (Stokes et al., 2012). Because the LIS was more extensive in most areas during MIS 2 (12–27 ka) than during MIS 3 (27–57 ka) (Lisiecki and Raymo, 2005; Siddall et al., 2008), older records of glaciation tend to be buried or eroded by the younger ice and subsequent sediment deposition. A primary indicator of glacial activity within the Upper Mississippi River Basin during MIS 3 has been the loess record, although the initial glacial source in Illinois, the ‘Altonian’ phase, was shown to not be time corelative with the Roxana (Willman and Frye, 1970; Leigh and Knox, 1993; Leigh, 1994; Curry, 1998; Grimley, 2000; Curry and Grimley, 2006). The timing, distribution, and extent of MIS 3 ice are critical parameters to constrain the initial conditions of the MIS 2 LIS. Therefore, the physical record of MIS 3 ice in North America is important for ice sheet reconstruction and modeling ice dynamics, as well as indicating the glacial correlation with loess records in the Midcontinent.
Understanding ice center development in North America leading up to the LGM is a crucial component of global sea level reconstructions, as Euroasian ice centers during MIS 3 were largely confined to alpine settings (Gowan et al., 2021). Whether the Labradoran dome in the east and the Keewatin dome in the west were connected during MIS 3, i.e. if Hudson Bay was glaciated, is a matter of ongoing debate. Previous studies have put forth LIS configurations that cover most of the Canadian Shield (Vincent and Prest, 1987; Dredge and Thorleifson, 1987; Gowan et al., 2021). However, radiocarbon (14C) and optically stimulated luminescence (OSL) ages from the Hudson Bay lowlands gathered by Dalton et al. (2016) suggest an unglaciated marine-terrestrial transition area during MIS 3 that was later buried by ice. This interpretation suggests that the LIS was not as extensive as previously assumed during this interval (Dalton et al., 2019). Pico et al. (2017) paired this reduced ice reconstruction with additional sea level indicators to conclude that the eastern LIS developed rapidly between 50 and 35 ka. Newly published North American MIS 3 ice sheet margin reconstructions also suggest a significant reduction in ice sheet extent during this interval when compared with previous estimations (Dalton et al., 2019; Batchelor et al., 2019). However, the interpretation of an ice-free Hudson Bay lowland has been questioned due to an abundance of “greater than” 14C ages and discordance between some OSL and finite 14C results within the equivalent units (Miller and Andrews, 2019). Subsequently, additional records of the LIS during MIS 3 are critical for modeling ice sheet buildup to the LGM. This data could help clarify apparent signals of dome asynchroneity between eastern and western sourced lobes advancing to their local last glacial maximum (Heath et al., 2018) and the MIS 3 loess records in the Mississippi and Missouri River Valleys (Leigh and Knox, 1993).
The activity of the Des Moines Lobe (DML, Fig. 1) during MIS 2 and 3 is an important indicator of overall behavior of the LIS during this interval. In this paper, the term DML is used to represent the lobe of ice that descended north to south from Manitoba, Canada (Keewatin Sector) into Iowa at any time during the last glacial period (MIS 5–2). The DML is considered to be an ice stream of the LIS (Clark, 1994; Patterson, 1997), and Margold et al. (2015) ranks it as one of the longest ice streams in the world— it is around 1400 km from its presumed source at the Saskatchewan River to its terminus at Des Moines, Iowa. During MIS 2, Dalton et al. (2020) propose the DML was actively advancing and retreating between 22 and 14 ka, reaching its southern limit around 18 ka. In contrast, Heath et al. (2018) suggest the DML advanced 200 km in under one thousand years to reach its terminus around 15 ka, after which it retreated nearly as rapidly. In both models, the DML reached its maximum extent thousands of years after ice lobes to the east (Heath et al., 2018) (Fig. 1). The low-relief till plain left by the MIS 2 DML also contrasts with land systems formed by lobes to the northeast, which consist of drumlins, high-relief moraines, low-relief hummocks, and ice thrust terrains (Colgan et al., 2005). The difference between the DML's timing and style of glaciation when compared with lobes to the east is likely explained by the DML being an ice stream moving mainly across a poorly lithified to unlithified, deformable bed by ice sourced from a different dome (Clark, 1994; Clayton et al., 1985, Patterson, 1997, 1998; Hooyer and Iverson, 2002; Margold et al., 2015). In addition, the DML does not encounter deep water lakes or basins like the eastern Labrador sourced lobes, e.g. the Superior, Green Bay, or Lake Michigan lobes which may have acted as a reservoir of ice for eastern lobes (Fig. 1; Cutler et al., 2000). Detailing DML activity prior to the LGM may aid in understanding the contrast between eastern and western lobes, driving mechanisms across the LIS, as well as the timing of ice dome growth during MIS 3.
Records from other lobes are also necessary to better understand the chronology and behavior of the LIS. Recent studies have placed ice within Wisconsin during MIS 3 (Carlson et al., 2018; Ceperley et al., 2019), but notably not Illinois, (Curry and Pavich, 1996). Additional evidence for MIS 3 glaciation includes glacial outwash (Carson et al., 2019) and loess in the Mississippi River Valley (Willman and Frye, 1970; Leigh and Knox, 1993; Leigh, 1994 Curry, 1996; Grimley, 2000; Forman and Pierson, 2002; Curry and Grimley, 2006). The loess record in major river valleys is important as it can be used to indicate lobe advance into the basin over time (e.g. Curry and Grimley 2006; Muhs et al., 2018). In the Gulf of Mexico, sediment core evidence such as records of meltwater pulses, chemical weathering proxies, detrital zircon geo-thermo-chronology, and shifts in clay mineralogy suggests that the ice was present somewhere within the Upper Mississippi River or Missouri River basins during MIS 3 (Tripsanas et al., 2007; Sionneau et al., 2013; Fildani et al., 2018).
Presented here are new radiocarbon ages and stratigraphic relationships from drill cores and outcrops that document two glacial diamictons deposited in northern Iowa during MIS 3, called the Sheldon Creek Formation (Fig. 2, Bettis et al., 1996). These ages, together with a synthesis of previously published and new data, provide new insights into the extent and timing of LIS advances and demonstrate the presence of two Keewatin-sourced ice advances within the Upper Mississippi River Basin during MIS 3. These advances represent a glacial source for the Roxana Silt within the Mississippi basin and the Pisgah Formation loess in the Missouri River Valley.
Section snippets
Regional setting and past work
The glaciated portion of the Upper Mississippi River Valley (Fig. 1) has been influenced by multiple advances of the Quaternary LIS. These advances left a palimpsest of deposits and landforms, with the most recent deposits being best preserved, and the older units being progressively difficult to recognize and interpret. This record has been of interest to geologists since the 19th century and remains important for Quaternary research as discoveries from the stratigraphic record in the Central
Lithologic description
This study was conducted using continuous drill cores collected from across north-central Iowa, as well as outcrops and quarry exposures (Fig. 3 inset). The majority of cores were collected in plastic sleeves using either a Central Mine Equipment drill rig or a truck mounted Giddings™ probe. Other cores were retrieved using a trailer mounted Giddings™ in unlined core barrels that were wrapped after collection. Cores were described and subsampled in a laboratory facility under fluorescent
Results
The new and legacy radiocarbon ages from across the study area (Table 1) are shown geographically in Fig. 3 and with stratigraphic context in Fig. 5, Fig. 6, Fig. 7. Fig. 5 shows cores and sections collected on the MIS 2 till plain while Fig. 5 shows cores collected to the east of the Bemis moraine on the IES. Stratigraphic (Fig. 7) and chronologic (Fig. 4) results point to the potential for two advances and will be discussed below. Based on this evidence, ages are be subdivided into four
Discussion
To inform the theoretical ice advance model, the MIS 2 DML is used as an analogue for the behavior of the MIS 3 ice. Since the DML was streaming during parts of MIS 2, it is likely that it behaved similarly during MIS 3. The factors that lead to ice streaming as defined in Winsborrow et al. (2010) – meltwater routing, bed roughness, geothermal heat flux, topographic step, and topographic focusing – were likely also present during MIS 3. By the analogy with the MIS 2 DML, thin ice likely
Conclusions
The synthesis of new and previously published ages presented here demonstrates that the Sheldon Creek Formation in Iowa was deposited during MIS 3. Two distinct ice advances within this interval are separated in some localities by weathering horizons or weakly developed soils. The Fort Dodge and Lehigh phases reached their terminal position around 42 and 30 ka, respectively. The behavior of the DML during MIS 3 is assumed to reflect the fast, streaming style of its MIS 2 iteration (Clark, 1992;
Sources of funding
This work was partially supported by the USGS National Cooperative Geological Mapping Program under STATEMAP award numbers G16AC00196 (2016), G17AC00258 (2017), & G18AC00194 (2018). Funding for undergraduate researchers on this project was partially supported by a National Science Foundation Award: Improving Undergraduate STEM Education Grant GP-IMPACT-1600429.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This work was partially supported by the USGS National Cooperative Geological Mapping Program under STATEMAP award numbers G16AC00196 (2016), G17AC00258 (2017), and G18AC00194 (2018). Funding for undergraduate researchers on this project was partially supported by a National Science Foundation Award: Improving Undergraduate STEM Education Grant GP-IMPACT-1600429. We thank the numerous landowners who allowed us to collect samples over the years, especially Martin Marietta and US Gypsum for
References (126)
- et al.
Mid-Wisconsinan environments on the eastern Great plains
Quat. Sci. Rev.
(2009) - et al.
Landscape evolution, alluvial architecture, environmental history, and the archaeological record of the Upper Mississippi River Valley
Geomorphology
(2008) - et al.
Rapid Laurentide ice-sheet advance towards southern last glacial maximum limit during marine isotope stage 3
Quat. Sci. Rev.
(2018) Unstable behavior of the Laurentide ice sheet over deforming sediment and its implications for climate change
Quat. Res.
(1994)Evidence at Lomax, Illinois, for mid-Wisconsin (∼40,000 yr B.P.) position of the Des Moines lobe and for diversion of the Mississippi river by the lake Michigan lobe (20,350 yr B.P.)
Quat. Res.
(1998)- et al.
Absence of glaciation in Illinois during marine isotope stages 3 through 5
Quat. Res.
(1996) - et al.
Quaternary glaciations in Illinois
- et al.
Constraining the late pleistocene history of the Laurentide ice sheet by dating the missinaibi formation, Hudson Bay lowlands, Canada
Quat. Sci. Rev.
(2016) - et al.
Chronologic evidence for multiple periods of loess deposition during the late pleistocene in the Missouri and Mississippi river valley, United States: implications for the activity of the Laurentide ice sheet
Palaeogeogr. Palaeoclimatol. Palaeoecol.
(1992) - et al.
Modern, Sangamon and Yarmouth soil development in loess of unglaciated southwestern Illinois
Quat. Sci. Rev.
(2003)
Pre-Wisconsin glacial stratigraphy of the central plains region in Iowa, Nebraska, Kansas, and Missouri
Quaternary Science Reviews- Quaternary Glaciations in the Northern Hemisphere
Stratigraphy and correlation of the glacial deposits of the Des Moines and James lobes and adjacent areas in north Dakota, south Dakota, Minnesota, and Iowa
Quaternary Science Reviews- Quaternary Glaciations in the Northern Hemisphere
Rates of erosion and sediment evacuation by glaciers: a review of field data and their implications
Global Planet. Change
Pattern of southern Laurentide ice sheet margin position changes during heinrich stadials 2 and 1
Quat. Sci. Rev.
Pedogenesis of a catena of the farmdale–sangamon geosol complex in the North central United States
Palaeogeogr. Palaeoclimatol. Palaeoecol.
Source and origin of Roxana silt and middle wisconsinan midcontinent glacial activity
Quat. Res.
AMS radiocarbon age of the upper Mississippi Valley Roxana silt
Quat. Res.
The application of radiocarbon age estimates to the dating of glacial sequences: an example from the Miami sublobe, Ohio, USA
Quat. Sci. Rev.
Ice streams in the Laurentide Ice Sheet: identification, characteristics and comparison to modern ice sheets
Earth Sci. Rev.
Paleopedology plus TL, Be-10, and C-14 dating as tools in stratigraphic and paleoclimatic investigations, Mississippi River Valley, USA
Quat. Int.
Transport direction of peoria loess in Nebraska and implications for loess sources on the central Great plains
Quat. Res.
Up in the refrigerator: geomorphic response to periglacial environments in the Upper Mississippi River Basin, USA
Geomorphology
Middle to Late Pleistocene loess record in eastern Nebraska, USA, and implications for the unique nature of Oxygen Isotope Stage 2
Quat. Sci. Rev.
Hudson Bay was not deglaciated during MIS-3
Quat. Sci. Rev.
Geochemical variations in peoria loess of western Iowa indicate paleowinds of midcontinental north America during last glaciation
Quat. Res.
Chronology and provenance of last-glacial (peoria) loess in western Iowa and paleoclimatic implications
Quat. Res.
Geochemistry and mineralogy of late quaternary loess in the upper Mississippi river valley, USA: provenance and correlation with Laurentide ice sheet history
Quat. Sci. Rev.
Southern Laurentide ice lobes were created by ice streams: Des Moines lobe in Minnesota, USA
Sediment. Geol.
An optimized scheme of lettered marine isotope substages for the last 1.0 million years, and the climatostratigraphic nature of isotope stages and substages
Quat. Sci. Rev.
The iowan drift, a review of the evidences of the iowan stage of glaciation
Iowa Geological Survey Annual Report
Relations of the Wisconsin and Kansan drift sheets in central Iowa, and related phenomena
Iowa Geological Survey Annual Report
Mid-Wisconsinan stratigraphy and paleoenvironments at the St. Charles site in South-central Iowa
Geol. Soc. Am. Bull.
Absolute chronology for major pleistocene advances of the Laurentide ice sheet
Geology
The first glacial maximum in North America
Science
Dating Plio-Pleistocene glacial sediments using the cosmic-ray-produced radionuclides 10Be and 26Al
Am. J. Sci.
The configuration of Northern Hemisphere ice sheets through the Quaternary
Nat. Commun.
The Deforest Formation of Western Iowa: Lithologic Properties, Stratigraphy, and Chronology
Complex response of a midcontinent North America drainage system to Late Wisconsinan sedimentation
J. Sediment. Res.
Last Glacial loess in the conterminous USA
Quat. Sci. Rev.
North american pleistocene stages reconsidered in light of probable pliocene-pleistocene continental glaciation
Science
Chronology of some Late Cenozoic deposits from the central United States and the ice ages
Transactions of the Nebraska Academy of Science VI
Iowan drift
Geol. Soc. Am. Bull.
The Iowan drift
J. Geol.
LATE MIS 3 ONSET TO LARGE-SCALE AGGRADATION ON THE UPPER MISSISSIPPI RIVER VALLEY, USA
The role of permafrost on the morphology of an MIS 3 moraine from the southern Laurentide Ice Sheet
Geology
The classification of American glacial deposits
J. Geol.
Supplementary hypothesis respecting the origin of the loess of the Mississippi valley
J. Geol.
Surface form of the southern Laurentide ice sheet and its implications to ice-sheet dynamics
Geol. Soc. Am. Bull.
The last glacial maximum
Science
Cited by (12)
Late Pleistocene Laurentide-source iceberg outbursts in the western Arctic Ocean
2022, Quaternary Science ReviewsCitation Excerpt :In the absence of icebergs from the LIS, those from the Kara and Barents shelves were transported in large quantities to the western Arctic Ocean, with the EAIS re-expanding during the MIS4 (∼65 ka) (Fig. 3; Spielhagen et al., 2004; Svendsen et al., 2004). The LIS may have suffered a retreat during 52–40 ka and was divided into the Keewatin Dome and the Labrador Dome (Kerr et al., 2021). The Keewatin Dome re-expanded at ∼42 ka, as the Labrador Dome continued to shrink (Dalton et al., 2019).
The marine δ<sup>18</sup>O record overestimates continental ice volume during Marine Isotope Stage 3
2022, Global and Planetary ChangeCitation Excerpt :Over North America, there is some indirect proxy evidence suggesting a moderately large continental ice sheet during MIS 3. However, this evidence can be explained by non-glacial processes and/or the brief ice extension noted by Kerr et al. (2021). Notably, the eolian-sourced Roxana Silt (and correlatives, e.g. Gilman Canyon Formation) is spatially ubiquitous in the mid-continental United States and was deposited from 55 ka to 30 ka (Bettis et al., 2003; Follmer, 1996; Forman and Pierson, 2002; Leigh, 1994).
Evolution of the Laurentide and Innuitian ice sheets prior to the Last Glacial Maximum (115 ka to 25 ka)
2022, Earth-Science ReviewsCitation Excerpt :Our maximum ice extent also considers evidence suggesting a short-lived, dramatic extension of the Keewatin Dome southward into the continental United States via a low-lying topographic corridor in Manitoba. The timing and extent of this advance (between ~34 ka and ~30 ka; known as the Lehigh Advance) is based on radiocarbon dating of wood contained within, overlying and underlying this till unit associated with the Keewatin Dome (Kerr et al., 2021). Along the Atlantic coastline, the ice margin is drawn at midway between the present-day coastline and the continental shelf edge.