The post-glacial history of northern Lake of the Woods: A multi-proxy perspective on climate variability and lake ontogeny

https://doi.org/10.1016/j.jglr.2018.04.002Get rights and content

Abstract

Lake of the Woods (LOW) is a large, morphologically and hydrologically complex lake of international importance, located in the provinces of Ontario and Manitoba and the state of Minnesota. A high-resolution sedimentary sequence retrieved near Kenora, Ontario, and spanning at least the past ~11,000 cal yr BP (calibrated years before present), was analysed for multiple environmental proxies with an emphasis on diatom assemblage composition and spectrally-inferred chlorophyll a. These biological proxies indicate that northern LOW was relatively nutrient-rich soon after its isolation from glacial Lake Agassiz ~10,000 cal yr BP. The post-glacial hydrological and environmental history of LOW was found to be controlled by both climate and isostatic rebound. During the low water phase of the mid-Holocene dry and warm period, abrupt and synchronous shifts across all proxies suggest that the northern basin had a relatively deep and well-mixed water column that experienced increases in nutrients and whole-lake algal production. This differs from recent limnological changes associated with warming since the late-1970s, where primary production increased concurrently with large shifts in diatoms indicative of increased thermal stability, but with little change in nutrients. The millennial-scale context of this study provides evidence that climate has long played an important role in algal dynamics in LOW, with implications for lake management strategies concerning recent increases in nuisance algal blooms on LOW.

Introduction

The Lake of the Woods (LOW) is a large remnant of glacial Lake Agassiz, the largest lake in North America during the last glacial retreat, that covered much of the LOW basin from about 12,000 to 10,000 cal yr BP (Yang and Teller, 2005). The developmental history, areal extent and volume of glacial Lake Agassiz has been linked to its proximity to the margin of the Laurentide Ice Sheet (LIS), the elevation and location of its overflow channels, and to differential isostatic rebound (e.g. Teller and Leverington, 2004). In the Lake of the Woods region of Ontario, Manitoba, and Minnesota, the fluctuating level of Lake Agassiz controlled the level of water across the LOW basin during its early stages. Initially, after the LIS had retreated from the region, water depths exceeded the confining margins of the LOW basin, and water covered a vast area south of the ice sheet. When the level of Lake Agassiz declined, waters eventually became confined by the topography surrounding the LOW basin, and the depth and extent of LOW was controlled mainly by climate and the effects of differential isostatic rebound (Teller, 1987; Teller and Leverington, 2004; Yang and Teller, 2005; Teller et al., 2018).

Today LOW is a large, hydrologically and morphologically complex freshwater system, the majority of which is located in Ontario (Canada) but straddling the boundary of the province of Manitoba and the state of Minnesota (U.S.A.) (Fig. 1). The southern region of LOW (e.g. Big Traverse Bay, which is largely located in Minnesota) is a shallow basin that is relatively uniform in depth, is well-mixed and is mesotrophic to eutrophic (Anderson et al., 2017). The northern region of LOW is distinctly different in that it is hydrologically and morphologically heterogeneous with numerous deep sub-basins and bays that thermally stratify, over 14,000 islands, and is generally less productive (Pla et al., 2005). The LOW region is adjacent to two ecotonal boundaries – the deciduous forest-boreal forest ecotone and the prairie-forest ecotone – making this a particularly sensitive area for tracking the limnological effects of past changes in climate (Frelich and Reich, 2010). This region of northwestern Ontario has reported some of the highest rates of temperature increases in North America since the mid-twentieth century (Schindler, 1997; Chiotti and Lavender, 2008) with increases in average annual temperature of 1.4 °C, which are projected to further increase by 1.5 to 2.5 times over the next 25 to 50 years (McKenney et al., 2010), particularly during the winter (Chiotti and Lavender, 2008; McDermid et al., 2015). In addition to increasing air temperatures, summers in northwestern Ontario are projected to become drier over the next ~50 years (McDermid et al., 2015), and will likely be punctuated by extreme precipitation events.

Over the past few decades, lake users have raised concerns that there has been an increase in the intensity and frequency of algal blooms on LOW that may be indicative of a deterioration in lake water quality (Chen et al., 2009; DeSellas et al., 2009; Clark and Sellers, 2014), in spite of substantial reductions in external nutrient loading from the Rainy River since the 1970s (Hargan et al., 2011). However, historical reports suggest that algal blooms have been a part of the history of the southern basins of LOW for at least the past ~200 years (Robertson and McCracken, 2003; McElroy and Riggs, unpublished data; Anderson et al., 2017). A warmer and drier climate can potentially exacerbate water quality concerns such as cultural eutrophication (Adrian et al., 1995; Smol, 2010), enhance the release of phosphorus from lake sediments (Edlund et al., 2017; James, 2017; Reavie et al., 2017), and, together with a longer open water season, extend the period for whole-lake primary production (Michelutti et al., 2010; Paterson et al., 2017) and for algal bloom formation (Jeppesen et al., 2007; Paerl and Huisman, 2008).

Paleolimnological studies exploring the effects of climate change on water quality more often focus on a lake's recent history (e.g. past two centuries) than on longer-term Holocene-scale trends (Smol and Cumming, 2000; Moos et al., 2009). For example, diatom-based paleolimnological records tracking changes over the past ~150 years from the Ontario portion of the LOW indicate that lake water total phosphorus concentrations declined over the past few decades (Rühland et al., 2010) and, with the exception of a few oligotrophic bays, most studied portions of northern LOW are currently mesotrophic (Pla et al., 2005). In addition, recent warming, an increase in the ice-free period, and indirect changes in nutrient cycling were found to be the main driver of algal assemblage composition (Rühland et al., 2010; Hyatt et al., 2011) and overall increases in whole-lake primary production (Paterson et al., 2017). These previous paleolimnological studies focused on the past ~150 years, and thus provided insights into baseline (reference) conditions and the limnological effects of recent anthropogenic disturbances.

The detailed diatom and chlorophyll a record from a northern location in LOW, which we present here was part of a broader assessment of LOW post-glacial history detailed by Teller et al. (2018) spanning >11,000 years. Their study focused primarily on sedimentological, mineralogical, and stratigraphic analyses, and included a study of ostracodes from six cores across this large, complex system. Based predominantly on sedimentological characteristics, Teller et al. (2018) provided evidence that differential isostatic rebound resulted in substantial hydrological differences in LOW that were superimposed on the region's response to the dry and warm Hypsithermal period, with the deeper, more isostatically depressed northern regions remaining relatively deep whereas the shallower southern basins likely dried up completely. This conclusion supports interpretations about changes in lake extent and depth in LOW based on the isostatic modeling of Yang and Teller (2005). Whilst a cursory summary of diatom changes was presented in Teller et al. (2018), here we provide a detailed examination of a suite of environmental proxies (focusing on diatoms and sedimentary chlorophyll a) preserved in a dated sediment core retrieved in the northernmost region of LOW near the city of Kenora, Ontario. We focus on the effects that large-scale changes in post-glacial climate from the time of glacial Lake Agassiz to modern-day LOW have had on aquatic biota, primary production, trophic status, and physical lake properties (e.g. water column mixing and thermal stratification). Building on previous algal-based paleolimnological work on LOW that focused on the past ~150 years (Rühland et al., 2010; Hyatt et al., 2011; Paterson et al., 2017), this millennial-scale study aims to provide new perspectives on current water quality concerns presently affecting LOW by improving our understanding of the role that accelerated warming has on nutrient availability and algal production.

Section snippets

Core retrieval

Using a Kullenburg rig and gravity corer, a 4.6-metre long sediment core (hereafter referred to as LOW 7A) was retrieved from the northeastern part of LOW (49.72984 N, 94.5130 E) near the city of Kenora (Fig. 1) in August 2006 with the aid of personnel and equipment from the Limnological Research Centre (LRC) at the University of Minnesota. The sediment core was stored at LRC, where it was sub-sampled approximately every 5.0 cm, resulting in ~80 intervals available for dating and sedimentary

Radiocarbon dating

AMS radiocarbon dates on 11 sedimentary intervals were obtained for LOW 7A; the material used for dating and the calibrated years are described in Teller et al. (2018). The youngest estimated date at the 2 cm sedimentary interval yielded a date of 234 ± 45 cal yr BP and the oldest date, analysed from sediment at 247 cm, yielded a date of 9827 ± 59 cal yr BP. About 1.5 m below the 9827 cal yr BP date, there is a pink clay lamina at a depth of 398 cm that has been interpreted as a marker for a

Discussion

Paleoecological proxies analysed in LOW 7A reflect several changes over the post-glacial period that correspond to the ontogeny and paleoenvironmental history of this large lake from glacial Lake Agassiz to modern-day Lake of the Woods. Isostatic rebound and climate changes appear to have been most important in driving the shifts in all paleo proxy indicators throughout the post-glacial history. Disentangling the effects of glacial retreat, watershed stability, isostatic rebound, and

Acknowledgements

This project was supported by an Ontario Ministry of the Environment Best in Science grant to JPS (79518) and a Natural Sciences and Engineering Research Council of Canada grant to JTT (32350). We thank Trevor Mellors and the team at the Limnological Research Center (LRC) at the University of Minnesota in Minneapolis for helping JTT retrieve the sediment core. We thank Neal Michelutti for aiding with chlorophyll a analysis, Carsten Meyer-Jacob for insightful comments on an earlier draft of the

References (69)

  • M.R. Twiss et al.

    Diatoms abound in ice-covered Lake Erie: an investigation of offshore winter limnology in Lake Erie over the period 2007 to 2010

    J. Great Lakes Res.

    (2012)
  • S.C. Zoltai et al.

    Holocene climatic change and the distribution of peatlands in western interior Canada

    Quat. Res.

    (1990)
  • R. Adrian et al.

    A long-term study of the Heilingenese (1975–1992): evidence for effects of climatic change on the dynamics of eutrophied lake ecosystems

    Arch. Hydrobiol.

    (1995)
  • J.P. Anderson et al.

    An introduction to Lake of the Woods – from science to governance in an international waterbody

    Lake Reserv. Manage.

    (2017)
  • R.W. Battarbee et al.

    Diatoms

  • B.F.N. Beall et al.

    Ice cover extent drives phytoplankton and bacterial community structure in a large north-temperate lake: implications for a warming climate

    Environ. Microbiol.

    (2016)
  • C.E. Binding et al.

    Time series analysis of algal blooms in Lake of the Woods using the MERIS maximum chlorophyll index

    J. Plankton Res.

    (2011)
  • C.J.F. ter Braak et al.

    CANOCO Reference Manual and CanoDraw for Windows User's Guide: Software for Canonical Community Ordination (Version 4.5)

    (2002)
  • L. Bunting et al.

    Increased variability and sudden ecosystem state change in Lake Winnipeg, Canada, caused by 20th century agriculture

    Limnol. Oceanogr.

    (2016)
  • W. Cannon et al.

    Effects of Holocene climate change on mercury deposition in Elk Lake, Minnesota: the importance of eolian transport in the mercury cycle

    Geology

    (2003)
  • H. Chen et al.

    Cyanobacteria and microcystin-LR in a complex lake system representing a range in trophic status: Lake of the Woods, Ontario, Canada

    J. Plankton Res.

    (2009)
  • Q. Chiotti et al.

    Ontario

  • B.J. Clark et al.

    Rainy-Lake of the Woods State of the Basin Report

    (2014)
  • W.E. Dean et al.

    Regional aridity in North America during the middle Holocene

    The Holocene

    (1996)
  • A.M. DeSellas et al.

    State of the basin report for the Lake of the Woods and rainy river basin

  • M.B. Edlund et al.

    Historical phosphorus dynamics in a large transboundary lake (Lake of the Woods) – do legacy loads still affect the southern basin?

    Lake Reserv. Manage.

    (2017)
  • L.E. Frelich et al.

    Will environmental changes reinforce the impact of global warming on the prairie-forest border of central North America?

    Front. Ecol. Environ.

    (2010)
  • E. Grimm

    TILIA and TILIA-GRAPH

    (1991)
  • H. Håkansson

    A compilation and evaluation of species in the general Stephanodiscus, Cyclostephanos and Cyclotella with a new genus in the family Stephanodiscaceae

    Diatom Res.

    (2002)
  • S. Heiskary et al.

    Remote sensing: does it have a role?

    N. Am. Lake Manag. Soc. Lakeline. Spring

    (2006)
  • M.O. Hill et al.

    Detrended correspondence analysis – an improved ordination technique

    Plant Ecol.

    (1980)
  • A. Hyodo et al.

    Variations in the oxygen-isotope composition of ancient Lake Superior between 10,600 and 8,800 cal BP

    J. Paleolimnol.

    (2012)
  • W. James

    Diffusive phosphorus fluxes in relation to the sediment phosphorus profile in Big Traverse, Lake of the Woods

    Lake Reserv. Manage.

    (2017)
  • E. Jeppesen et al.

    Shallow lake restoration by nutrient loading reduction – some recent findings and challenges ahead

    Hydrobiologia

    (2007)
  • Cited by (5)

    View full text