Elsevier

Quaternary Science Reviews

Volume 248, 15 November 2020, 106590
Quaternary Science Reviews

Subtill nonglacial deposits and their climatic implications for the Last Interglacial (MIS 5e), Hudson Bay Lowlands, Canada

https://doi.org/10.1016/j.quascirev.2020.106590Get rights and content

Highlights

  • First documentation of subtill nonglacial deposits for the Winisk River, Hudson Bay Lowlands (HBL).

  • Only site of the HBL where multiple luminescence dates (OSL) indicate a Last Interglacial or MIS 5e affinity.

  • First report on S-rich ferromagnetic nodules and their development interpreted as a result of significant warming.

  • Climatic conditions inferred from multiple lines of evidence including palaeobotany, sedimentology and geochemistry.

  • Provide new insights into the stratigraphy and climatic conditions associated with the Missinaibi Formation of the HBL.

Abstract

In Hudson Bay Lowlands (HBL), the Wisconsinan tills are often underlain by organic-bearing nonglacial deposits formed during episodes of significant reduction of ice volumes or complete ice-free conditions in this region. The subtill nonglacial deposits, often correlated with the Missinaibi Formation, are beyond the method limit of radiocarbon dating in general. As such, they have been a subject of debates for their stratigraphy and associated climatic conditions owing to poor age constraints. Extensive studies were undertaken on a subtill argillaceous lacustrine deposit uncovered recently on the Winisk River. This 9 m thick organic-rich deposit, referred to as the Webequie Beds, was dated at 121 ± 9 ka, 126 ± 24 ka and 105 ± 11 ka using optically stimulated luminescence (OSL) dating method and correlated with the Last Interglacial (LIG) or Marine Isotope Stage (MIS) 5e. Pollen and plant macrofossils suggest spruce boreal forests along with extensive wetlands, not unlike what exists in the HBL today. Of particular note is the sporadic rise of birch in the upper part of the pollen sequence, interpreted as a response to the availability of frequent forest clearance by fires under a warmer climate. The birch zone is noteworthy because it corresponds to the development of conspicuous S-rich, ferromagnetic nodules (0.125–1.5 mm diameter) in the lake sediment. The nodules comprise native sulphur and greigite (Fe3S4), along with iron oxyhydroxide goethite and lepidocrocite. They likely resulted from significant warming during this interval that enhanced microbial sulphate reduction and increased the release of H2S, an essential component for the formation of the S minerals in the lake. Besides the strengthened microbial metabolization, a prolonged summer lake stratification and development of an anoxic bottom water also likely promoted the sulphate reduction. Although confirmation is needed, the warming probably linked to the LIG thermal maximum (∼120 ka) when a climate with a higher summer temperature than today prevailed. A similar subtill nonglacial deposit on the Little Current, a tributary of the Albany River, was also examined for comparison. Apart from a similar pollen flora, S-rich nodules were also recovered near a horizon OSL dated at 118 ± 13 ka comparable to the LIG thermal maximum. Results of this study provide new insights into the stratigraphy and climatic conditions of the LIG in the HBL but also help better understand the behavior and dynamics of the Laurentide Ice Sheet during the Late Pleistocene.

Introduction

Hudson Bay Lowlands (HBL) constitute a poorly drained, low-relief area along Hudson and James bays (Fig. 1). During the recent geological past, the HBL was located near the geographic centre of the Pleistocene Laurentide Ice Sheet (LIS). The Quaternary stratigraphic records of this region can thus provide key information for better understanding of the behavior and dynamics of the LIS. In the HBL, the Wisconsinan tills are often underlain by a suite of organic-bearing nonglacial deposits that have been used as the marker beds in regional stratigraphic correlations (Fig. 1). Because these subtill nonglacial deposits are, in general, beyond the method limit of radiocarbon dating, dating of them has been proved very challenging. Consequently, these deposits have been a subject of debates for their age and climatic conditions, i.e., interglacial vs. interstadial as highlighted recently by Miller and Andrews (2019) and Dalton et al. (2016).

The subtill nonglacial deposits accumulated in lakes, rivers and inland seas during episodes of significant reduction of ice volumes or complete ice-free conditions in this region. They occur primarily as a single lithostratigraphic unit in the sedimentary succession and are commonly correlated with the Missinaibi Formation of the Last Interglacial as defined by Skinner (1973) in the Moose River basin in the southern part of the HBL. The pollen from these deposits indicates a vegetation very similar to today in this region (Terasmae and Hughes, 1960; Netterville, 1974; Mott and DiLabio, 1990). Because the Last Interglacial was warmer than today such as seen in the Don Formation in southern Ontario (Terasmae, 1960; Karrow, 1990; Occhietti et al., 2016), the lack of significant compositional change in the pollen assemblages has led some to question its interglacial status (e.g., Terasmae and Hughes, 1960). Furthermore, radiometric dating, which includes radiocarbon, luminescence and U/Th as well as amino acid epimerization and racemization (AAR) studies, has returned inconsistent results, with ages assigned to various Marine Isotope Stages (MIS), including MIS 3, MIS 5e (the Last Interglacial), MIS 5, and MIS 7 (Andrews et al., 1983; Nielsen et al., 1986; Forman et al., 1987; Berger and Nielsen, 1990; Wyatt, 1990; Thorleifson et al., 1993; Roy, 1998; Allard et al., 2012; Dubé-Loubert et al., 2013; Dalton et al., 2016, 2017, 2018; Miller and Andrews, 2019). Lastly, the Missinaibi Formation at its type site is a composite derived from fragmentary records of marine, fluvial, peat and glaciolacustrine deposits (Skinner, 1973). While the sedimentary model derived from this formation has been widely used in the study of the subtill nonglacial deposits in the HBL interior, it may not be applicable for the margin of the HBL where different deposits and sedimentary processes, e.g., lacustrine, may have predominated.

Recent Quaternary geological mapping by the Ontario Geological Survey in the HBL has resulted in the finding of additional subtill nonglacial deposits on the Winisk, Attawapiskat and Albany rivers and their tributaries (Fig. 1) (Gao, 2011; Gao et al., 2015; Bajc et al., 2014). Of the major rivers in the HBL, the Winisk River is the only one where deposits of this type have not been previously documented. Therefore, their discovery on this river provides a unique opportunity with which to better understand the stratigraphy and climatic conditions associated with the nonglacial deposits. The objectives of this study are 1) to constrain the age of the deposit, and 2) to discuss the associated climatic conditions using multiple proxies derived from palaeobotany, sedimentology and geochemistry. Previous work in the adjacent lower stream of the Winisk River provides important information on glacial lithostratigraphy for correlation (Thorleifson et al., 1993). In addition, for comparison, a subtill nonglacial deposit of lacustrine origin on the Little Current, a tributary of the Albany River, was also examined (Fig. 1). To avoid confusion, unless otherwise stated in the following discussions, the Last Interglacial (LIG) is used in a strict sense, roughly correlating with MIS 5e (132-116 ka) (Shackleton et al., 2003; Otvos, 2015).

Section snippets

Regional setting

The study site is located on the middle stream of the Winisk River near the western margin of the HBL over Precambrian granitic bedrock (53°42′57″N, 87°2′25″W) (Fig. 1; Appendix A). Downstream to the north, Palaeozoic carbonate bedrock underlies the HBL interior. Except for occasional esker and moraine ridges, this region has a flat, featureless topography with extensive open to treed wetlands (Appendix A). The Winisk River incises deeply into the lowlands, exposing along its banks over 10 m of

Methods

Fieldwork was conducted with the support of a helicopter. Large sections exposed along riverbanks were cleaned and logged. Samples were obtained from freshly exposed sections and drill cores using a soil auger with extension rods each 1m long. Total carbonate content and calcite-to-dolomite ratios in the <0.074 mm fraction were determined using a Chittick apparatus at the Geoscience Laboratories of the Ontario Geological Survey (the Geo Labs). Bulk geochemistry, e.g., S and P, of the <0.063 mm

Geological sections and lithostratigraphy

The geological sections are exposed along the western bank of the Winisk River, immediately downstream from a series of rapids known as the Seashell Rapids that pass over a granite bedrock ridge (Appendix A). The river here stands at 116 m above the sea level (asl), ∼13 m below the valley shoulder (∼129 m asl). The sections exposed a thick, nonglacial silt deposit referred to as the Webequie Beds below several tills and Holocene sediments (Fig. 2, Fig. 3b) (Gao and Crabtree, 2016). In the

Pollen

Pollen from the subtill nonglacial Webequie Beds is dominated by spruce (Picea) at 57% (21%–76%) and pine (Pinus) at 38% (22%–51%) (Fig. 8). Of the total pine pollen, jack pine accounts for 47% on average (21%–76%), with the rest presumably being white pine (Pinus strobus) (Appendix B). Birch rarely exceeds 5% but can reach 18% in some upper samples (Fig. 8). Other trees are minor, each represented by a couple of pollen grains, including larch (Larix), fir (Abies), cedar (Juniperus or Thuja)

Ferromagnetic nodules

The ferromagnetic nodules occur primarily in the upper 1 m of the Webequie Beds, at concentrations exceeding 1000 grains in one of the samples (Table 2). Fresh magnetite grains are common in the lower samples but become rare to absent in the upper section, exhibiting an inverse relationship with the ferromagnetic nodules (Table 2). The nodules are irregular, ovoid, platy, bullet-to rod-shaped, with a dark, submetallic warty surface stained with rust and coated locally with sand-to silt-sized

Magnetic susceptibility

A low magnetic susceptibility of 0.04 × 10−3 SI (0.03–0.05 × 10−3 SI) is recorded in the Webequie Beds (Fig. 6). Little difference exists in this parameter value, regardless of whether the samples are magnetite-bearing in the lower part of the sequence or ferromagnetic-nodules-bearing in the upper part. Since magnetite is rare to absent in the upper part of the sequence, it can be concluded that the magnetic susceptibility values there must arise primarily from the contributions by the

Radiometric dating

Radiocarbon dating by the AMS method on wood from the Webequie Beds returned dates of >48,500 and > 49,500 14C BP, providing a minimum age for the deposit (Table 3). OSL dating on feldspar grains yielded dates of 121 ± 9 ka, 126 ± 24 ka and 105 ± 11 ka at 2.1 m, 1.5 m and 0.6 m depths, respectively (Table 4). The reason for the date reversal at 2.1 m and 1.5 m depths is unclear. However, the 2 dates overlap at the 1σ level and, statistically, they are indistinguishable. As such, the dating

Sedimentary conditions

The argillaceous texture and laminated character of the Webequie Beds indicate a deposition within a deep lake. The well-developed lamination suggests the accumulation of the beds below the wave base. While the effective wave-base depth capable of moving fine sand on the lake floor remains unknown, this depth is ∼7 m on open ocean shelves dominated by waves and storms under fair-weather conditions (Plint, 2010). Based on this observation, the lake is estimated to have had a depth of similar

Geochemical conditions

The development of the S-rich, ferromagnetic nodules is concomitant with the rise of total S in the sediment matrix in the upper part of the Webequie Beds (Fig. 10). The elevated S precipitation is considered to arise through enhanced microbial sulphate reduction and H2S production in the lake during this interval. In response, total organic matter declined in the sediment, interpreted as a result of increased decomposition or remineralization of the organic material (Fig. 2d). The enhanced

Geochronology and stratigraphy

The OSL dates of 121 ± 9 ka, 126 ± 24 ka and 105 ± 11 ka indicate a LIG or MIS 5e age for the subtill nonglacial Webequie Beds on the Winisk River (Table 4). The only other site in the HBL wherein multiple radiometric dates are available for a LIG age is on the Nottaway River, where U/Th dating on wood from a subtill nonglacial deposit returned consistent dates of 100–120 ka (Allard et al., 2012). A comparable OSL date of 95 ± 7 ka on feldspar grains from the same deposit supports the

Conclusions

Extensive studies were undertaken to constrain the age of the subtill nonglacial Webequie Beds and to understand the climatic conditions under which they developed. The Webequie Beds, which consist of 9 m of organic-rich, argillaceous lacustrine deposit, were OSL dated at 121 ± 9 ka, 126 ± 24 ka and 105 ± 11 ka and, therefore, correlated to the LIG or MIS 5e. Pollen and plant macrofossils suggest spruce boreal forests along with extensive wetlands, not unlike what exists in the HBL today.

Author contributions

C.G. Conceptualization, Resources, Investigation, Formal analysis, Methodology, Writing- Original draft preparation, Writing - Review & Editing, Data Curation, Visualization and Funding acquisition. S.H. Investigation, Methodology, Writing - Review & Editing, Validation and Formal analysis. A.M.D. Investigation, Methodology, Writing - Review & Editing, Validation and Formal analysis. D.C.C. Investigation, Formal analysis, Methodology, Validation and Data Curation. C.L.T. Investigation,

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 study was supported by the Ontario Geological Survey and, partly, by a Natural Sciences and Engineering Research Council of Canada (NSERC) Discovery grant to AMM. John H. McAndrews is thanked for the analysis of the initial pollen data and generation of an early pollen diagram. The A. E. Lalonde AMS Laboratory, University of Ottawa and the Radiocarbon Dating Laboratory, the Illinois State Geological Survey conducted radiocarbon dating. Paul Coulson, Trevor Jones, Victoria Lee, Jordan

References (96)

  • W.T. Jiang et al.

    Contradictory magnetic polarities in sediments and variable timing of neoformation of authigenic greigite

    Earth Planet Sci. Lett.

    (2001)
  • J. Kleman et al.

    North American ice sheet build-up during the last glacial cycle, 115-21 kyr

    Quat. Sci. Rev.

    (2010)
  • M. Margold et al.

    Reconciling records of ice streaming and ice margin retreat to produce a palaeogeographic reconstruction of the deglaciation of the Laurentide Ice Sheet

    Quat. Sci. Rev.

    (2018)
  • D.R. Muhs et al.

    Timing and warmth of the Last Interglacial period: new U-series evidence from Hawaii and Bermuda and a new fossil compilation for North America

    Quat. Sci. Rev.

    (2002)
  • E.G. Otvos

    The last interglacial stage: definitions and marine highstand, North America and Eurasia

    Quat. Int.

    (2015)
  • A. Picard et al.

    Sulfate-reducing bacteria influence the nucleation and growth of mackinawite and greigite

    Geochem. Cosmochim. Acta

    (2018)
  • D. Rickard et al.

    Botanical constraints on pyrite formation

    Chem. Geol.

    (2007)
  • A.P. Roberts

    Magnetic mineral diagenesis

    Earth Sci. Rev.

    (2015)
  • M. Roy et al.

    Insights on the events surrounding the final drainage of Lake Ojibway based on James Bay stratigraphic sequences

    Quat. Sci. Rev.

    (2011)
  • N.J. Shackleton et al.

    Marine Isotope Substage 5e and the Eemian Interglacial

    Global Planet. Change

    (2003)
  • J.J. Veillette et al.

    Ice-flow evolution of the Labrador sector of the Laurentide ice sheet: a review, with new evidence from northern Quebec

    Quat. Sci. Rev.

    (1999)
  • S.J. Wu et al.

    Sulfur cycling in freshwater sediments: a cryptic driving force of iron deposition and phosphorus mobilization

    Sci. Total Environ.

    (2019)
  • M. Adler et al.

    Quantifying rates of methanogenesis and methanotrophy in Lake Kinneret sediments (Israel) using pore-water profiles

    Limnol. Oceanogr.

    (2011)
  • S. Al-Zuhair et al.

    Sulfate inhibition effect on sulfate reducing bacteria

    J. Biochem. Technol.

    (2008)
  • T.W. Anderson

    Forest changes in the Great Lakes region at 5-7 ka BP

    Géogr. Phys. Quaternaire

    (1995)
  • A.F. Bajc et al.

    Field studies in support of terrain mapping and aggregate resource assessment, James Bay Lowland, Ontario

  • P.J. Barnett

    Surficial Geology of the Winisk Lake Area, Northern Ontario

    (2008)
  • G.W. Berger et al.

    Evidence from thermoluminescence dating for Middle Wisconsinan deglaciation in the Hudson Bay Lowland of Manitoba

    Can. J. Earth Sci.

    (1990)
  • R.K. Bezys

    Geology of gypsum deposits in the James Bay Lowland. Ontario Geological Survey Open File Report

  • H.H. Bostock

    Precambrian sedimentary rocks of the Hudson Bay Lowlands

  • I. Candy et al.

    How warm was Britain during the Last Interglacial? A critical review of Ipswichian (MIS 5e) palaeotemperature reconstructions

    J. Quat. Sci.

    (2016)
  • A.S. Dalton et al.

    Pollen and macrofossil inferred paleoclimate at the Ridge site, Hudson Bay Lowlands, Canada: evidence for a dry climate and significant recession of the Laurentide ice sheet during Marine Isotope Stage 3

    Boreas

    (2017)
  • A.S. Dalton et al.

    An updated radiocarbon-based ice margin chronology for the last deglaciation of the North American Ice Sheet Complex

    Quat. Sci. Rev.

    (2020)
  • L.A. Dredge et al.

    Sangamon and pre-Sangamon interglaciations in the Hudson Bay Lowlands of Manitoba

    Géogr. Phys. Quaternaire

    (1990)
  • H. Dubé-Loubert et al.

    Glacial and nonglacial events in the eastern James Bay lowlands, Canada

    Can. J. Earth Sci.

    (2013)
  • L.D. Farley-Gill

    Contemporary pollen spectra in the James Bay Lowland, Canada, and comparison with other forest tundra assemblages

    Géogr. Phys. Quaternaire

    (1980)
  • E.J. Fee et al.

    Effects of lake size, water clarity, and climatic variability on mixing depths in Canadian Shield lakes

    Limnol. Oceanogr.

    (1996)
  • S.L. Forman et al.

    Thermoluminescence properties and age estimates for Quaternary raised marine sediments, Hudson Bay Lowland, Canada

    Can. J. Earth Sci.

    (1987)
  • C.F. Fu et al.

    Occurrence of greigite in the Pliocene sediments of Lake Qinghai, China, and its paleoenvironmental and paleomagnetic implications

    G-cubed

    (2015)
  • R.F. Galbraith et al.

    Optical dating of single and multiple grains of quartz from Jinmium rock shelter, northern Australia: part I, experimental design and statistical models

    Archaeometry

    (1999)
  • C. Gao

    Surficial geology mapping and drift sediment sampling in the McFaulds Lake (“Ring of Fire”) area, northern Ontario

  • C. Gao

    Till Sample and Indicator Mineral Data for the Detour Lake and Burntbush Area, Northern Ontario

    (2015)
  • C. Gao

    Indicator Mineral and till Bulk Sample Geochemistry Data, Chapleau Area, Northern Ontario

    (2016)
  • C. Gao et al.

    Glaciation of North America in the James Bay Lowland, Canada, 3.5 Ma

    Geology

    (2012)
  • C. Gao et al.

    Results of Regional Till and Modern Alluvium Sampling in the McFaulds Lake (“Ring of Fire”) Area, Northern Ontario

    (2016)
  • C. Gao et al.

    Indicator Mineral and Geochemistry Data for the till and Alluvium Sampling Survey in the McFaulds Lake (“Ring of Fire”) Area, Northern Ontario

    (2016)
  • C. Gao et al.

    Field studies in support of remote predictive mapping in the Missisa Lake area, Far North of Ontario. Summary of Field Work and Other Activities 2015

  • C. Gao et al.

    Ipswichian (Eemian) floodplain deposits and terrace stratigraphy in the lower Great Ouse and Cam valleys, southern England, UK

    Boreas

    (2011)
  • View full text