Enhanced algal abundance in northwest Ontario (Canada) lakes during the warmer early-to mid-Holocene period

Quaternary Science Reviews 123 (2015) 168e179

Contents lists available at ScienceDirect

Quaternary Science Reviews journal homepage: www.elsevier.com/locate/quascirev

Enhanced algal abundance in northwest Ontario (Canada) lakes during the warmer early-to mid-Holocene period Moumita Karmakar a, *, Peter R. Leavitt b, Brian F. Cumming a, ** a

Paleoecological Environmental Assessment and Research Laboratory (PEARL), Department of Biology, Queen's University, Kingston, Ontario, K7L 3N6, Canada b Limnology Laboratory, Department of Biology, University of Regina, Laboratory Building, Saskatchewan, S4S 0A2, Canada

a r t i c l e i n f o

a b s t r a c t

Article history: Received 18 February 2015 Received in revised form 21 June 2015 Accepted 22 June 2015 Available online 13 July 2015

This study investigates regional changes in primary producers in boreal head-water lakes during the warmer early-to-mid-Holocene (EMH) period, across the present-day boreal forest in northwest Ontario, a region that is adjacent to the prairie-forest ecotone. We quantified changes in algal abundance and composition over the Holocene period using pigments, spectrally-inferred chlorophyll a and diatom assemblages in well-dated sediment cores from three lakes. All three indicators showed a coherent pattern of enhanced primary producers in two of the study lakes (Gall Lake and Lake 239) during the EMH, whereas only diatom assemblages suggested higher levels of nutrients in Meekin Lake. Overall, this study supports a regional pattern of enhanced primary producers during the EMH, likely as a function of lower water-levels and warmer temperatures. Elevated concentrations of cyanobacterial pigments also occurred in two of the three lakes during the EMH, whereas pigments from purple-sulphur bacteria provide evidence of enhanced deep-water anoxia in one lake. These findings suggest that future climatic warming in boreal regions could include regional eutrophication and associated increases in cyanobacteria. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Northwest Ontario Boreal lakes Holocene Multiproxy Phytoplankton Algae Cyanobacteria

1. Introduction Temperatures have increased an average of 0.8
C (1.4
F) across the northern hemisphere since 1900, and climate models predict an additional increase between 0.3 and 4.8
C by the end of the 21st century relative to the 1986e2005 reference period (Collins et al., 2013). Under future warming scenarios, scientists, governments and industry have been tasked to understand the risk of climatic warming on water resources. The direct and indirect mechanisms that could cause changes to lakes in a warmer world are numerous. For example, changes in the surface-water temperature of lakes can result in changes to mixing regimes and thermal stability (Adrian et al., 2009; Read et al., 2014). Changes in physical characteristics may also have cascading effects on nutrient, oxygen and biotic assemblages (Adrian et al., 2009). Further, it is well known that

* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (M. Karmakar), brian.cumming@ queensu.ca (B.F. Cumming). http://dx.doi.org/10.1016/j.quascirev.2015.06.025 0277-3791/© 2015 Elsevier Ltd. All rights reserved.

climate influences the length of the ice-free periods (Schindler et al., 1996a; Rodriguez et al., 2001; Diehl et al., 2002; Smol et al., 2005) and many studies have demonstrated an increase in icefree conditions as air temperatures have increased (Mcknight et al., 1996; Livingstone and Dokulil, 2001; Magnusson et al., 2000; Williams and Stefan, 2006). Climatic warming can also increase water-residence times in lakes, through decreases in stream flow (Schindler et al., 1996a; Rippey et al., 1997). Declines in stream flow can result in lower lake-levels, but also a lower inputs of nutrients and dissolved organic carbon (DOC) from the catchment, with resultant increases in lake-water transparency, and changes in thermocline depth (Schindler et al., 1996a, 1996b). Nutrient concentration can also be affected by the internal processes related to changes in the thermal structure and/or primary production (Jeeppesen et al., 2005; Wilhelm and Adrian, 2008). Warmer climatic conditions also favour cyanobacterial blooms (Paerl and Huisman, 2008), a change that is widely acknowledged as a threat to water quality in many regions (O'Neil et al., 2012), because many species are toxic (Carmichael et al., 1988). The growth rates of many cyanobacteria are higher relative to eukaryotic

M. Karmakar et al. / Quaternary Science Reviews 123 (2015) 168e179

phytoplankton at higher temperatures (Robarts and Zohary, 1987), which also increase the buoyancy of many cyanobacteria (Kromkamp et al., 1988; Reynolds, 2006; Carey et al., 2012). In addition to temperature, high concentrations of nutrients are also € hnk et al., 2008; important predictors of cyanobacteria blooms (Jo Kosten et al., 2012; Taranu et al., 2012), which can also change in a warmer climate. In summary, many physical, chemical and biological factors can influence the abundance and composition of primary producers, often in complex and unexpected ways (Findlay et al., 2001), and the response of lakes to climatic change can vary between lakes with different characteristics (Adrian et al., 2006; Read et al., 2014). Given this complexity, it is often hard to predict how primary producers in lakes will change under a warmer climate. Long-term studies are important to understand how past periods of warmer conditions have influenced algal abundance. Unfortunately, long-term records of changes in algal abundance are rare, especially during prolonged periods of enhanced warmth. Fortunately, records of past periods that were warmer than present do exist, thereby allowing the opportunity to examine how primary producers changed in the past under warmer conditions. Such changes can be investigated using paleolimnological approaches, by examining changes in sedimentary pigments, inferences of chlorophyll a, and changes in diatom assemblages in well-dated sediment cores (e.g. Hall et al., 1999; Garrison and Wakeman, 2000; Reavie et al., 2006). To date, climate-related changes in lake trophic status during warmer periods on the Holocene scale are available from few lakes (e.g., Fietz et al., 2007; Lake Baikal; Kirilova et al., 2009; central European Lake; Hickman et al., 1990; Baptiste Lake, Alberta), with a particular absence of relatively small boreal lakes. In this paper, we quantify past changes in the abundance and composition of algae in sediment cores from three boreal lakes to evaluate how the warmer early-to-mid-Holocene (EMH) period affected primary producers. This study builds on previous research from a core from Lake 239, that indicated nutrient-rich planktonic diatom taxa were most abundant during the EMH period (c. 8500 e 4500 cal yr BP) (Moos et al., 2009). The EMH period in northwest Ontario was characterized as a more open boreal forest € rck, 1985; Lewis et al., 2001; Moos and (McAndrews, 1982; Bjo Cumming, 2011) based on a low abundance of spruce pollen, and increases in non-arboreal pollen including Cupressaceae, Artemesia, and Ambrosia. Pollen-based inferences of the EMH period suggest that temperatures were warmer than present conditions by approximately 2e3
C (Moos and Cumming, 2012). During the EMH, lake levels were regionally low in comparison with the last 4e5 thousand years (Karmakar et al., 2015), and in conjunction with the pollen, suggests both warm and arid conditions (Karmakar et al., 2015). We use these established changes to address two questions: a) is enhanced algal abundance in lakes a common phenomenon in head-water lakes in northwest Ontario during the EMH period in comparison with earlier and later in the Holocene; and b) did cyanobacterial assemblages become more abundant during the EMH? 2. Study sites The boreal region of northwest Ontario is located approximately 200-km east of the modern-day prairie-forest ecotone, a climatically-sensitive ecotone (Umbanhowar et al., 2006). All of our study lakes are located within the Winnipeg River Drainage Basin (WRDB), a large catchment (150,000 km2) located primarily in northwest Ontario (Fig. 1). Instrumental records from the WRDB are limited, but show a warming trend of 1e2
C over the past century, and spatially-asynchronous droughts on a sub-decadal

169

scale (Laird et al., 2012). We selected three lakes to study within the WRDB that span a spatial transect of over ~200 km (Fig. 1) across the boreal region. This region consists of rolling topography of hills and valleys located on crystalline bedrock with shallow soils (McAndrews, 1982). The easternmost site, Gall Lake is located in the English River Watershed (50
140 N, 91
270 W) whereas the westernmost site, Meekin Lake (49
490 N, 94
460 W) is located ~100 km from the modern-day prairie-forest ecotone. Lake 239 (49
400 N, 93
440 W) was included as the changes in the diatom assemblages during the EMH was the motivation for this study (Moos et al., 2009) and is located centrally between the other two study sites. The study lakes are all relatively small (Gall Lake: surface area ¼ 19 ha, maximum depth ¼ 18 m, fetch ¼ 0.85 km; Lake 239: surface area ¼ 56 ha, maximum depth ¼ 31 m, fetch ¼ 1.1 km; Meekin Lake: surface area ¼ 78 ha, maximum depth ¼ 13 m, fetch ¼ 1.5 m), slightly acidic (pH ¼ 5.9 e 6.5), dimictic, first-order lakes, with gentle sloping bathymetry in at least one basin (Kingsbury et al., 2012). These lakes represent presently oligotrophic (Meekin Lake and Lake 239) to slightly mesotrophic nutrient condition (Gall Lake) (Kingsbury et al., 2012), and contain few macrophytes. The vegetation near Gall Lake is dominated by black spruce (Picea mariana), jack pine (Pinus banksiana), and poplar (Populus spp.), along with white birch (Betula papyrifera), balsam fir (Abies balsamea), and larch (Larix spp.). Similarly, the boreal forest around Lake 239 consists mainly of black spruce, jack pine and some poplar. Further to the west, the vegetation around Meekin Lake consists of balsam fir, poplar, white birch with some black ash (Fraxinus nigra) and red maple (Acer rubrum). A synthesis of available pollen-based records, which span the Canadian prairie-boreal forest ecotone, clearly indicate EMH warmth (Moos and Cumming, 2011). However, changes in pollen species were dependent on lake location, with the prairie lakes showing increases in grasses and Ambrosia spp. during the EMH, whereas changes in Cupressaceae and Ambrosia are more apparent in the boreal region.

3. Materials and methods In June 2011, a piston core was collected from the deepest region of the central basin in Gall and Meekin lakes using a Livingstone square-rod piston corer, with an internal diameter of 5.1 cm (Glew et al., 2001). A piston core was also removed from the deep central basin of Lake 239 in July 2004 (Moos et al., 2009). The total core lengths were ~5, 11 and 9 m of sediment from Gall Lake, Lake 239 and Meekin Lake, respectively. In the lab, each core was sectioned at 0.5 cm intervals into 10 oz Whirlpak® bags. A total of 10 (Gall Lake), 8 (Lake 239, Moos et al., 2009) and 13 intervals (Meekin Lake) were sampled, and analysed along the length of both cores, for radiocarbon dates (Table 1). AMS 14C age estimates were based on analysis of carbon from pollen isolated at the Limnological Research Center at the University of Minnesota using the procedures of Brown et al. (1989), after which they were analysed for 14C activity at the Lawrence Liverpool National Laboratory. Radiocarbon dates were calibrated (using IntCal, Reimer et al., 2009) and age-depth models for sediment cores from Gall and Meekin lakes were created using ‘classical’ age-depth modelling (clam; Blaauw, 2010). The age-depth model for the Lake 239 sediment core was based on a simple polynomial (Fig. 3 in Moos et al., 2009). To determine the age at the top interval of the piston core for cores from Gall and Meekin lakes, total 210Pb and 137Cs were measured in the uppermost sediment in the first section of the piston cores from each lake and matched to the activities measured in a dated gravity core retrieved at the same depth as the piston core (e.g., Laird et al., 2012; Haig et al., 2013; Ma et al., 2013).

170

M. Karmakar et al. / Quaternary Science Reviews 123 (2015) 168e179

Fig. 1. Map showing the location of the three study lakes (Gall Lake, ELA Lake 239 and Meekin Lake) within the Winnipeg River Drainage Basin. Lake 239 and Gall Lake are located in the English River Watershed, whereas Meekin Lake is located in the Lake of the Woods/Rainy River Watershed.

3.1. Estimates of primary producer abundance and composition Sedimentary pigment analysis was undertaken following standard procedures in Leavitt and Hodgson (2001). Briefly, pigments from a subsample of ~50 mg of freeze-dried sediment were extracted using a mixture of acetone, methanol, and water (80:15:5

by volume). Pigment extracts were then filtered through a 0.22 mm PTFE filter and after 24 h were dried with inert N2 gas. The dried pigment residues on the filters were then re-dissolved in a standard injection solution (combination of methanol, Sudan II stock solution and IPR stock solution) before introduction into an Agilent 1100 High Performance Liquid Chromatography (HPLC) system,

Fig. 2. A) Ageedepth model for the sediment cores from Gall Lake and B) for Meekin Lake, based on calibrated 14C dates (Table 1), obtained by the classical age-depth model under “clam” package in R software (Blaauw, 2010). Cumulative depth vs calendar years BP dates and errors are shown in blue. The grey area shows 95% confidence intervals of the inferred dates based on 1000 iterations. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.).

M. Karmakar et al. / Quaternary Science Reviews 123 (2015) 168e179

171

Table 1 Summary of the14C-dating results for the sediment cores from Gall and Meekin lakes. All analyses were performed based on pollen isolated at the Limnological Research Center at the University of Minnesota, and dated at Lawrence Livermore National Laboratory. Lake name

Sample depth (cm)

Cumulative depth (cm)

Material dated

14

Gall P1S1 Gall P1S2 Gall P1S2 Gall P1S3 Gall P1S3 Gall P1S4 Gall P1S4 Gall P1S5 Gall P1S5 Meekin P1S1 Meekin P1S2 Meekin P1S2 Meekin P1S3 Meekin P1S3 Meekin P1S4 Meekin P1S4 Meekin P1S5 Meekin P1S5 Meekin P1S6 Meekin P1S6 Meekin P1S7 Meekin P1S9

30e30.5 32.5e33 82e82.5 28e28.5 77.5e78 27.5e28 77e77.5 29.5e30 80.5e81 75e76 25e26 75e76 25e26 75e76 25e26 75e75.5 25e25.5 75e76 25e25.5 75e76 25e26 25e26

30 80 130 180 230 280 330 379 430 75 121 171 216 266 316 366 416 468 516 566 616 811

Isolated Isolated Isolated Isolated Isolated Isolated Isolated Isolated Isolated Isolated Isolated Isolated Isolated Isolated Isolated Isolated Isolated Isolated Isolated Isolated Isolated Isolated

445 1200 2035 2695 3750 4785 6355 8140 8970 1755 2350 3105 3830 4525 5085 5865 6505 7415 8145 8875 10120 10945

pollen pollen pollen pollen pollen pollen pollen pollen pollen pollen pollen pollen pollen pollen pollen pollen pollen pollen pollen pollen pollen pollen

C Age (±1 SD)

equipped with a photo diode array detector. Pigments were differentiated on the basis of chromatographic position and light absorbance characteristics using detailed monographs and comparison with authentic standards (Leavitt and Hodgson, 2001). Spectral analysis of chlorophyll a and associated degradation products in sediment samples was conducted following Michelutti et al. (2010). Sediments were freeze dried and then sieved through a 125 mm mesh, and placed into glass vials. All reflectance spectra were obtained using a Model 6500 series Rapid Content Analyzer (FOSS NIRSystems Inc.). The sedimentary chlorophyll a concentration was inferred using an equation (chlorophyll

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

30 30 30 30 35 30 30 30 220 25 30 30 25 30 30 30 30 30 25 40 170 40

Calibrated

14

C Age (±2 sigma) calendar years BP (clam)

466e533 1056e1184 1922e2064 2755e2849 4062e4164 5470e5561 7246e7333 9006e9134 9538e10587 1598e1725 2327e2459 3237e3381 4148e4299 5053e5189 5748e5832 6633e6750 7412e7476 8181e8324 9008e9133 9886e10175 11242e12241 12709e12912

No. of CAMS 157,800 157,801 157,802 157,804 157,803 157,805 157,806 157,799 157,807 162,171 162,172 162,173 162,174 162,175 162,176 162,177 162,178 162,179 162,180 162,181 162,182 162,184

a þ derivatives ¼ 0.0919  peak area650e700 nm þ 0.0011) derived based on a 35-lake calibration dataset by Michelutti et al. (2010). Analysis of diatom assemblages was conducted using standard strewn mounts following the procedure of Wilson et al. (1996). Briefly, a sub-sample of ~0.2e0.3 g of wet sediment was digested in 20 ml glass vials using a 50:50 M solution of concentrated sulphuric and nitric acids. Following the removal of the acid by aspiration, and daily rinses with distilled water every 24 h until the sample was acid free. The diatom slurry was reduced to a volume of ~5 ml, to which a known aliquot of microsphere solution (0.8 ml of a 2.0  107 spheres/ml) was added to allow the calculation of diatom concentrations (Battarbee and Kneen, 1982). Permanent slides

Fig. 3. Concentration profile for selected fossil pigments in Gall Lake sediment core over time (cal yr BP). Pigments includes: chlorophyll a (Plantae, algae), chlorophyll b (Plantae, Chlorophyta, Eugluenophyta), b-carotene (all plants), diatoxanthin (Bacillariophyceae, Dinophyta, Chrysophyta), myxoxanthophyll (colonial cyanobacteria), canthaxanthin (colonial cyanobacteria), echinenone (all cyanobacteria), okenone (purple sulphur bacteria), chlorophyll a to Pheophytin A (an index of preservation), and organic matter (%) are also shown. The sub-zones (A1, A2, B1, B2) are shown as defined by the constrained cluster analysis.

172

M. Karmakar et al. / Quaternary Science Reviews 123 (2015) 168e179

were made by following the procedure of slide preparation outlined in Moos et al. (2005). For most diatom samples a minimum of 400 valves were identified and enumerated using a Leica DMRB microscope with a 100x Fluotar objective (NA ¼ 1.3), and differential interference contrast optics at 1000x magnification. For samples from the bottom section of the Meekin Lake sediment core (below cumulative depth of 608 cm), only ~200 valves were counted due to low diatom concentrations. Diatoms were typically identified down to the species level or lower using the standard taxonomic references (Krammer and Lange-Bertalot, 1986, 1988, 1991a, 1991b; Cumming et al., 1995; Lange-Bertalot and Melzeltin, 1996; Camburn and Charles, 2000). Diatom data from Lake 239 was available from a previous study (Moos et al., 2009), but is summarized in the results.

sedimentation rates from the ageedepth models, the analysis of fossil pigment, spectrally-inferred chlorophyll a, and diatom assemblages at 8 cm intervals (in Gall Lake) represents an integrated sample of ~12 years sampled at an approximate temporal resolution of every 190 year over the Holocene. Similarly, for the sediment core from Meekin Lake, the analysis of three proxies at a 16 cm interval in the core represents an integrated sample of ~8 years sampled at an approximate 250 year temporal resolution. For the sediment core from Lake 239, the fossil pigment and spectrallyinferred chlorophyll a analyses at a 16 cm intervals represents an integrated sample of ~7 years at an approximate resolution of every 220 years.

3.2. Numerical analysis

Cluster analysis on the concentrations of fossil pigments in Gall Lake sediments (Fig. 3) identified four main zones during the Holocene: i) an early-Holocene period, circa 11,200 to 9400 cal yr BP (Zone A1, Fig. 3); ii) an EMH period, circa 9400 to 7400 cal yr BP (Zone A2); iii) a mid-Holocene period, circa 7400e4300 cal yr BP (Zone B1); and iv) a late-Holocene period from circa 4300 cal yr BP to present (Zone B2). In the early Holocene, the organic content of the sediment was low <10% with corresponding low concentrations of pigments and spectrally-inferred chlorophyll a. Pigment concentrations of chlorophyll, b-carotene, and spectrally-inferred chlorophyll a, achieved their highest concentrations during the EMH period, and remain high until approximately 4200 cal yr BP. The ratio of chlorophyll a: pheophytin a, an indicator of preservation, remained high until circa 4200, where it drops and remains relatively constant for the remainder of the Holocene. Myxoxanthophyll, canthaxanthin, and echinenone initially increase in the early Holocene and continue at similar or higher levels (echinenone) throughout the EMH (Zone A2) when organic matter increases to almost 40%, thereafter it remained relatively stable for the remainder of the Holocene. The pigment okenone, unique to purple-sulphur bacteria, becomes dominant in the EMH (Zone A2), where after it declined to below detection limits by 6800 cal yr BP. In the mid-Holocene period, myxoxanthophyll declined below detection, while canthaxanthin and echinenone persist, but at slightly lower concentrations than in the EMH. In the late-Holocene period, concentrations of echinenone dropped to levels similar to those in the early Holocene and myxoxanthrophyll declined to below detection limits in the most recent sediments. The HPLC measured chlorophyll a and spectrally-inferred chlorophyll a are significantly correlated (r ¼ 0.54, n ¼ 55, p ¼ <0.0001). In the sediment core from Lake 239, three major time periods based on changes in the concentrations of fossil pigments (Fig. 4) were identified: i) an early-Holocene period, circa 10,000 to 8400 cal yr BP (Zone A, Fig. 4); ii) an EMH period, circa 8400 to 4100 cal yr BP (Zone B1); and iii) a late-Holocene period, circa 4100 to present (Zone B2). During the early-Holocene period the concentrations of all pigments were low (Fig. 4). During the EMH period, the concentration of chlorophyll a and b-carotene increased, as does diatoxanthin and myxoxanthophyll. Similarly, the spectrally-inferred chlorophyll a was highest from circa 8500e4000 cal yr BP (Fig. 4), as was the ratio between chlorophyll a and pheophytin a. Organic matter increased from approximately 5% to around 20% by 4000 cal yr BP, where it slowly increased to a maximum of between 25% and 30% by the late Holocene. Myxoxanthophyll decreased sharply at circa 4000 cal yr. BP, but canthaxanthin remains high. The HPLC measured chlorophyll a and spectrally-inferred chlorophyll a are significantly correlated (r ¼ 0.5, n ¼ 48, p ¼ <0.0002). In the sediment core from Meekin Lake, three major time periods based on changes in the pigment concentrations were

Fossil pigment and spectrally-inferred chlorophyll a data were graphed using the computer program Tilia v.2.0.2 (Grimm, 2004). Diatom species with greater than 5% abundance in at least one sample were included in the plots. The correlation between HPLC measured chlorophyll a and NIRS chlorophyll a was based on interpolated data, as the same intervals were not measured for all samples (Chlorophyll a concentration was converted to mg/g organic matter for comparison to HPLC). A depth-constrained cluster analysis (CONISS, Grimm, 1987) was used to identify zones to define broad-scale trends in the sedimentary records. For diatoms, both percent relative abundance and concentration data were calculated at a resolution of 8 cm intervals for Gall Lake, and 16 cm intervals for Meekin Lake. The chrysophyte scale to diatom index (C to D index) was calculated using the formula: scales/ (diatoms þ scales)  100 (Moos et al., 2005). Diatom-inferred values for Total Phosphorus (TP) were developed based on a 268lake calibration set of freshwater lakes from British Columbia, from which surface samples and measured water chemistry were available. The range of TP of the British Columbia model was 5e322 mg/l. Log TP was reconstructed with a simple weighted average model with inverse deshrinking (bootstrapped r2 ¼ 0.51, RMSEP ¼ 0.34, Moos et al., 2009) using the computer program C2 (Juggins, 2003). To determine the suitability of using the modern dataset from British Columbia at reconstructing TP from lakes from northwest Ontario, we assessed how well the dominant taxa in the cores (Supplemental Figs. 1 and 3, and Fig. 3 in Moos et al., 2009) were represented by the taxa in the calibration dataset. For each of the dominant taxa, we assessed if the maximum abundance in the calibration dataset surpassed the maximum values observed in the cores. To help determine if the main directions of variation in the diatom assemblages followed the TP reconstructions in each of the cores, correlations were assessed between the axis-1 scores of a  PCA ordination (Ter Braak and Smilauer, 2012) and the TP reconstructions. 4. Results The cumulative length of the sediment cores from Gall, Lake 239 and Meekin lakes are 450, 1135, and 870 cm, respectively. In all three cores, grey clays were present in the bottommost sections. The activities of 210Pb in the upper intervals of the piston cores from Meekin and Gall lakes correspond to the activities in 210Pb-dated gravity cores taken from similar locations that were dated to the years 1995 for Meekin Lake, and 1977 for Gall Lake. In all lakes, the calibrated ages show a consistent change, almost linear, to older samples with increasing cumulative depth in the cores (Fig. 2; Fig. 2 in Moos et al., 2009), with errors typically <100 years, but with higher uncertainty prior to ~9000 cal yr BP (Fig. 2). Based on the

4.1. Fossil pigments and spectrally-inferred chlorophyll a

M. Karmakar et al. / Quaternary Science Reviews 123 (2015) 168e179

173

Fig. 4. Concentration profile for selected fossil pigments in ELA Lake 239 sediment core over time (cal yr BP). Pigments includes: chlorophyll a (Plantae, algae), b-carotene (all plants), diatoxanthin (Bacillariophyceae, Dinophyta, Chrysophyta), myxoxanthophyll (colonial cyanobacteria), canthaxanthin (colonial cyanobacteria), echinenone (all cyanobacteria), Chlorophyll a to pheophytin A, and organic matter (%) are also shown. The sub-zones (A, B1, B2) are shown as defined by the constrained cluster analysis.

identified (Fig. 5): i) a late-glacial to early-Holocene period, circa 13,600 to 9500 cal yr BP (Zone A, Fig. 5); ii) an EMH period, circa 9500 to 5400 cal yr BP (Zone B1); and iii) a late-Holocene period, circa 5400 cal yr BP to present (Zone B2). During Zone A, pigment concentrations were low. At this time organic matter was also low, but the chlorophyll a to pheophytin a ratio was high. During the EMH, pigments including myxoxanthophyll, echinenone, canthaxanthin, chlorophyll a and b-carotene all increased, as does the percent organic matter from 10 to 30 %. Interestingly, myxoxanthophyll only achieves a sustained concentrations between 10,000 and 8000 cal yr BP. During the late-Holocene period, the concentration of both chlorophyll a and b-carotene, as well as many carotenoids including canthaxanthin, increased, achieving levels higher than in the EMH. The spectrally-inferred chlorophyll a increased quickly during the EMH followed by slower increases over the last 5000 years (Fig. 5). A similar pattern is seen in the percent organic matter. The two measures of chlorophyll a are significantly correlated (r ¼ 0.85, n ¼ 52, p ¼ <0.0001). 4.2. Diatom composition and inferred total phosphorus concentration The cores from all three lakes span at least the last 10,000 years, and all cores underwent large changes in diatom species composition over this period (summarized in Fig. 6A). The detailed diatom assemblages in the cores from Gall and Meekin lakes, for both percent abundance and concentrations, are shown in Supplemental Figs. 1e4, and the stratigraphy of the dominant diatom taxa from Lake 239 is available in Moos et al. (2009). The high degree of similarity between the relative abundance data and the concentration data of all cores suggests that similar, but not identical

interpretations, would result if made on interpretations of relative abundance, concentration or accumulation data (i.e., sedimentation rates are approximately linear for all cores). However, we include the concentration data as it allows interpretation of some important differences, especially following deglaciation and the EMH period. Briefly, in each core from the three lakes, a depth-constrained cluster analysis identified four-to-five major zones based on changes in the diatom assemblages (Fig. 6A, Supplemental Figs. 1,3; Fig. 3 in Moos et al., 2009). For the sediment core from Gall Lake there are four zones including: i) an early-Holocene period, circa 10000 to 8200 cal yr BP (Zone A1); ii) an EMH period, circa 8200 to 5400 cal yr BP (Zone A2); iii) a mid-to-late-Holocene period, circa 5400 to 2000 cal yr BP (Zone B1); and iv) a late-Holocene period, circa 2000 to present (Zone B2). The early Holocene was dominated mainly by benthic diatom taxa including Staurosirella pinnata (Ehrenberg) Williams and Round, 28 species of Achnanthes spp., and 32 species of Navicula and many other benthic taxa, as well as intermittent but high abundance of the planktonic taxon Cyclotella bodanica v. lemanica (O. Müll. Ex Schrot) H. Bachmann (Fig. 6A), all of which occurred at very low concentrations (Supplemental Fig. 2). During the EMH period, Cyclotella michiganiana Skv. increases in both relative abundance and concentrations, and the concentrations of S. pinnata increases over 20x, along with many other benthic and planktonic taxa (Fig. 6A). The mid-to late-Holocene period was characterized by an increase in many planktonic taxa including Discostella stelligera (Cleve and Grunrew) Houk and Klee, Aulacoseira tenella (Nygaard) Simonsen, Asterionella formosa (Hassall) Houk and Klee, and C. bodanica v. lemanica (Fig. 6A), as well as many other planktonic taxa that occur at lower abundances (e.g. Tabellaria flocculosa (Roth) Kutz, Fragilaria tenera (W. Smith) Lange-

174

M. Karmakar et al. / Quaternary Science Reviews 123 (2015) 168e179

Fig. 5. Concentration profile for selected fossil pigments in the Meekin Lake sediment core over time (cal yr BP). Pigments includes: chlorophyll a (Plantae, algae), b-carotene (all plants), diatoxanthin (Bacillariophyceae, Dinophyta, Chrysophyta), myxoxanthophyll (colonial cyanobacteria), canthaxanthin (colonial cyanobacteria), echinenone (all cyanobacteria), chlorophyll a to pheophytin A, and organic matter (%) are also shown. The sub-zones (A, B1, B2) are shown as defined by the constrained cluster analysis.

Bertalot). During the late-Holocene, circa 2000 cal yr BP to present, the planktonic diatom taxon D. stelligera became consistently more dominant (Fig. 6A). The C to D index increased at circa 6500 cal yr BP, reaching stable but variable levels by circa 5000 cal yr BP (Supplemental Fig. 1). The changes in the diatom assemblages from the Lake 239 core are summarized into five periods based on major changes in the diatom assemblages (Fig. 6A, Moos et al., 2009). These periods include: i) a post-glacial period (Zone A1), ii) an early-Holocene period, circa 10,000 to 8600 cal yr BP (Zone A2); iii) an EMH period, circa 8600 to 4500 cal yr BP (Zone A3); iv) a mid-to-lateHolocene period, circa 4500 to 3200 cal yr BP (Zone B1); and v) a late-Holocene period, circa 3100 cal yr BP to present (Zone B2). Zone A1 is dominated by low abundances of planktonic taxa including D. stelligera, Cyclotella ocellata Pantocsek, and the benthic taxon S. pinnata. The early Holocene is defined by a decline in the planktonic taxa, and increases in the relative abundance of benthic taxa including S. pinnata and many species of Achnanthes (Fig. 6A). The EMH is characterized by increases in the relative abundance and concentrations of a number of planktonic taxa including D. stelligera, Aulacoseira subarctica (O Müller) Haworth, and Fragilaria crotonensis, as well as sustained abundances of benthic S. pinnata (Fig. 6A and Fig. 3 in Moos et al., 2009). Zones B1 and B2 are characterized by increasing abundances of D. stelligera, as well as increases in Fragilaria nanana Meister (Moos et al., 2009). The C to D index increased at circa 8600 cal yr BP with another stepwise increase c. 4600 cal yr BP, reaching relatively stable levels thereafter (Moos et al., 2009). Four major periods were also defined in the core from Meekin Lake based on large changes in the diatom assemblages: i) a postglacial period, circa 13,800 to 11, 200 cal yr BP (Zone A1); ii) an early-Holocene period, circa 11,200 to 8000 cal yr BP (Zone A2); iii) a mid-Holocene period, circa 8000 to 5000 (Zone B1); and iv) a lateHolocene period, circa 5000 to present (Zone B2). Detailed diatom

stratigraphies based on relative abundance data and concentration data can be found in Supplemental Figs. 3 and 4. During the postglacial period, Meekin Lake was dominated by low concentrations of both benthic (S. pinnata and Achnanthes spp.) and planktonic taxa (A. subarctica, Stephanodiscus medius Håkasson, and Aulacoseira islandica (O. Müller) Simonsen). The early-Holocene period was dominated by high concentrations and abundance of A. subarctica, as taxa including S. medius and A. islandica declined in both relative abundances and concentrations (Supplemental Figs. 3 and 4). During the mid-Holocene period A. subarctica was still abundant, but declined throughout this period as D. stelligera and C. michiganiana increased to above 20% relative abundance (Fig. 6A), and a number of other planktonic diatoms increased. The late-Holocene period was characterized by the dominance of D. stelligera (Fig. 6A) and Discostella pseudostelligera Houk and Klee, while other planktonic taxa (Cyclotella and Tabellaria) remained subdominant. The C to D index within this lake was low in comparison with the other two study lakes, although during the past ~4000 years this index was higher compared to the early- and midHolocene periods. The dominant diatom taxa found in the cores (i.e., those in Supplemental Figs. 1 and 3; Fig. 3 in Moos et al., 2009) were well represented by the taxa in the modern-day calibration samples, with the modern samples generally achieving a higher maximum abundance than seen in the cores. In the Gall Lake core, all 14 dominant taxa (Supplemental Fig. 1) achieved a greater maximum abundance in the modern calibration dataset than in the core samples. A number of the less abundant benthic taxa that collectively were more abundant in the early Holocene in Gall Lake (i.e., rare benthic taxa, Navicula spp., Achnanthes spp.) were not well represented in the calibration dataset, using a cut-off for inclusion of at least a species evenness (N2) of 5. In the core from Lake 239 (Fig. 3 in Moos et al., 2009), all but one taxon, Navicula farta, was well represented by the modern calibration dataset. Similarly, 8 of

M. Karmakar et al. / Quaternary Science Reviews 123 (2015) 168e179

175

Fig. 6. A) Summary of the diatom taxa in the Gall Lake sediment core, that occurred in greater than 20% relative abundance in the zones and subzones defined by a depth constrained cluster analysis, as well as diatom-inferred total phosphorus (DI-TP) values over the Holocene for Gall, ELA Lake 239 and Meekin Lake. The pigment zones from the cores from each of the lakes from Fig. 3e5, are shown next to the DI-TP values for each lake, and the zones with high pigment concentrations are shaded. The study sites are arranged in geographic position from east to west. B) PCA axis-1 scores derived from the diatom assemblages data from the cores from the three study lakes for the last 10,000 years. See supplemental Fig. 1e4 for the details on changes in the dominant diatom taxa, and Fig. 3 (in Moos et al., 2009) for the diatom stratigraphy for Lake 239.

the 11 taxa in the Meekin Lake core (Supplemental Fig. 3) achieved higher abundances in the modern dataset than in the core samples. The three taxa that surpassed the relative abundances in the modern samples included A. subarctica, S. medius, and A. islandica. However, the maximum abundance (evenness) of A. subarctica and S. medius, in the calibration dataset was 39% (N2 ¼ 15), and 16% (N2 ¼ 11), lower than the maximum abundance in the Meekin Lake core (~60% and 35%, respectively). The trends in diatom-inferred TP (DI-TP) values for all three cores over the last 10,000 years (Fig. 6A) are highly and significantly correlated to the main directions of variation in the diatom assemblages as summarized by PCA axis-1 scores (Fig. 6B; correlation coefficients of 0.93, 0.8 and 0.97, for Gall Lake, Lake 239 and Meekin Lake, respectively). The high-inferred DI-TP values correspond to the periods of high pigments in Gall and Lake 239 cores, but not the Meekin Lake core (Fig. 6A). The highest and most stable DI-TP values occurs prior to 6000 (Gall Lake), 7000 (Lake 239) and 9000 (Meekin Lake) cal yr BP years for the sediment cores from these lakes (Fig. 6A). The lowest DI-TP values of the post-glacial/ Holocene occur over the last few millennia (Fig. 6A). In all three sites, the change to the dominance of D. stelligera also appears time transgressive, occurring later in the east, than in the west (Fig. 6, Zone B2 in all three lakes). A similar time transgressive pattern is seen in the scaled chrysophyte to diatom index (C to D index), with

the first large increase in this ratio occurring at circa. 5000 cal yr BP in the Gall Lake core (Supplemental Fig. 1), c. 8600 cal yr BP in Lake 239 (Fig. 3 in Moos et al., 2009), and circa. 9000 in Meekin Lake (Supplemental Fig. 3). A similar time-transgressive pattern is also present in the ratio of chlorophyll a to pheophytin a, with higher ratio between circa 9000 and 4000 cal yr BP in Gall Lake, 8500 and 4200 cal yr BP in Lake 239, and 13,000 and 7000 cal yr BP in Meekin Lake. The high and stable DI-TP values show similar patterns to the high concentrations of sedimentary pigments in the core from Gall Lake (Fig. 6A, pigment zones A2 and B1) and Lake 239 (Fig. 6A, pigment zone B1), but not in the core from Meekin Lake (Fig. 6A). 5. Discussion 5.1. Enhanced primary producers during the warmer early-to midHolocene (EMH) period Northwest Ontario has been shown to be sensitive to climate change. For example, pollen-based estimates of temperature from Lake 239 in northwest Ontario suggest that the EMH period was 2e3
C warmer than today (Moos et al., 2009), and that lake-levels were much lower than today in all three study lakes (Karmakar et al., 2015). Deep-water sediment cores from three lakes across an east-to-west transect across northwest Ontario was used to

176

M. Karmakar et al. / Quaternary Science Reviews 123 (2015) 168e179

quantify past changes in algal primary producers during the warmer EMH period. High concentrations of sedimentary pigments and spectrally-inferred values of chlorophyll a in the sediment cores from Gall Lake and Lake 239 during the EMH period, in conjunction with diatom-based evidence for higher nutrient levels in all three study lakes, suggests enhanced algal abundance at this time. However, the sediment core from Meekin Lake did not show a consistent pattern of enhanced algal abundance between the diatoms and the pigment data. In both Gall Lake and Lake 239, elevated concentrations of chlorophyll a and b-carotene suggested higher abundance of algae in the EMH, occurring between 9400 and 4300 cal yr BP in Gall Lake (Fig. 3), and between 8400 and 4100 cal yr BP in Lake 239 (Fig. 4). However, in Gall Lake, there was also an indication of different limnological conditions between 9000 and 7000 years ago, as indicated by the consistent presence of deep-blooming purplesulphur bacteria, an obligate anaerobe, suggesting either permanent stratification and anoxia, or seasonal anoxia at this time. The diatom assemblages in the cores from Gall Lake and Lake 239 are consistent in that the assemblages suggest high production of nutrient-rich diatoms during the EMH, consistent with the pigment data (Fig. 6A), and that the main direction of variation in the diatom assemblages is highly correlated to the TP inferences. Changes in both water temperature, as well as inferred declines in lake level (Karmakar et al., 2015), can cause both direct and indirect changes to the limnology of a lake, that could lead to changes in nutrients and mixing regimes resulting in enhanced algal abundance and changes to community structure (Huisman et al., 2004; Adrian et al., 2006; Winder and Hunter, 2008). The core from Meekin Lake suggests that algal abundance over the past 5000 years was higher than in the EMH period, a trend opposite to that observed in the other sites (Fig. 6A). However, the diatom assemblage data does not support this interpretation, and in contrast to the pigments suggests that during the EMH period the nutrient conditions were high (Fig. 6A). As shown in the cores from Gall Lake and Lake 239 the inferences of enhanced algal abundance from pigments and diatoms during the warmer EMH period agreed. Pigment accumulations in lake sediments are often correlated with algal biomass, often as a result of increased production. However the concentrations of pigments found in lake sediments are the result of the difference between production and degradation (Leavitt, 1988), with less degradation associated with rapid burial, and/or enhanced preservation under anoxic conditions (Fritz, 1989). Rapid burial of pigments could also be related to lower lake levels. However, lower lake levels could also result in enhanced photo degradation, if sufficient light penetrates to the lake bottom. Like the other lakes, the ratios between chlorophyll a to pheophytin a, an indicator of pigment preservation, is high during the EMH period, suggesting either enhanced abundance of algae and/or preservation (Leavitt, 1988), a finding that is difficult to reconcile with the pigment trends in this core and the cores from Gall Lake and Lake 239. The dominant diatom in Meekin Lake during the EMH period is A. subarctica, which achieved a relative abundance close to 60%. Subdominant planktonic taxa include S. minutulus, F. crotonenesis, and A. ambigua. All of these taxa have a high optima to lake-water total phosphorus (Wilson et al., 1996; Ginn et al., 2007). A. subarctica is an important spring and/or hypolimnetic bloomer (Interlandi et al., 1999), is a diatom that is a highly silicified often benefiting from turbulent conditions (Gibson et al., 2003). Similarly, A. ambigua has been interpreted as an indicator of more turbulent conditions, requiring moderate to high light levels and high silica levels (Bradbury and Dieterich-Rurup, 1993). A possible explanation for the conflict between the diatoms and the pigments is perhaps the low lake levels during the EMH, combined with warm waters and turbulent conditions due to the relatively large

fetch of Meekin Lake. This may have resulted in favourable conditions for diatoms, but not for pigment preservation. 5.2. Cyanobacteria during the warmer early-to-mid-Holocene (EMH) period In all three lakes, cyanobacterial pigments were present throughout the Holocene period, but were much more pronounced during the EMH in two lakes (e.g. Gall Lake and Lake 239). Among the three study lakes, Gall Lake (~9000e4000 cal yr BP) and Lake 239 (~8500e4000 cal yr BP) showed a similar pattern of increasing cyanobacterial pigment concentration during the EMH period. Elevated abundance of cyanobacterial populations is not surprising, € hnk et al., because warm and stable waters favours cyanobacteria (Jo 2008), and increases in cyanobacteria at higher temperatures has € hnk been reported in lakes of very different mixing regimes (Jo et al., 2008). High temperature can also influence buoyancy and delay sinking (Kromkamp et al., 1988; Reynolds, 2006; Carey et al., 2012). In the United States, models based on modern surveys show that nitrogen, phosphorus and water temperature are the best predictors of cyanobacteria biomass (Beaulieu et al., 2013). In the Gall Lake core, the abundance of cyanobacteria was high in the EMH period in comparison with the late Holocene. This pattern is consistent with the enhanced algal abundance at this time. The pigment myxoxanthophyll, which represents colonial cyanobacteria was present in the early Holocene within Gall Lake and then disappeared between 7500e5000 cal yr BP, when okenone, a pigment from purple-sulphur bacteria, increased. Presence of purple-sulphur bacteria, an obligate anaerobe, indicates anoxic conditions possibly of excess nutrients and could result in additional internal loading of nutrients. Similarly, in Lake 239, myxoxanthophyll was present throughout the EMH period (~8000e4000 cal yr BP) but disappeared after ~4000 cal yr BP. Myxoxanthophyll was also present in the Meekin Lake sediment core, but only during the very early Holocene (~10,500e8000 cal yr BP). In studies of lake eutrophication, increases in myxoxanthophyll and/or oscillaxanthin in sediments have been interpreted as a signal of increased algal abundance (e.g. Griffiths et al., 1969; Gorham and Sanger, 1975; Griffiths, 1978; Züllig, 1981, 1982; Guilizzoni et al., 1982, 1983; Swain, 1985). 5.3. Time transgressive patterns across the study sites There are broad-scale patterns of enhanced algal abundance during the EMH period inferred from fossil pigment, diatom assemblage and spectrally-inferred chlorophyll a, changes across the region are either climatically-driven time transgressive trends or due to lake-specific differences, or associated with core chronology. The highest DI-TP inference occurred before 8000 years in Meekin Lake, by 7000 years in Lake 239, and by 6000 years in Gall Lake. Changes in the late Holocene also suggest a timetransgressive trend across the region as shown by the dominance of the oligotrophic diatom D. stelligera at by 5000 cal yr BP in Meekin Lake, 3000 cal yr BP in Lake 239, and 2000 cal yr BP in Gall Lake. Given the strong chronology of the age-depth models, it would seem unlikely that the observed temporal lags between lakes are due to imprecise dating. However, it is possible that climatic conditions varied across our west-to-east transect of sites, with an earlier warming in the west and a delayed response in the east. Inferences of changes in climatic conditions during the midHolocene period has been undertaken across the prairie-forest boundary in northwest Ontario based on changes in pollen assemblages (Moos and Cumming, 2011). The cores from the prairie lakes show some evidence of an earlier increase in non-arboreal pollen during the EMH, whereas the more eastern lakes, show a

M. Karmakar et al. / Quaternary Science Reviews 123 (2015) 168e179

more muted response; however, with the exception of the core from Lake 239, the chronologies of these cores were not well dated (Moos and Cumming, 2011). The cores from the prairie lakes also exhibited a larger inferred temperature change in comparison to the present boreal sites (Moos and Cumming, 2011). Evidence that supports a climate-driven time transgressive pattern in the mid-tolate Holocene comes from reconstructions of changes in lake level from these three lakes, based on the selection of sensitive nearshore coring locations (Laird et al., 2012). Based on this approach, Karmakar et al. (2015) showed increases in lake level in Meekin Lake between circa 8000 cal yr BP and 5000 cal yr BP., whereas in the Lake 239 and Gall Lake, lake levels began to increase between circa 6000 and 3000 cal yr BP (Karmakar et al., 2015), suggesting a much later penetration of moisture into the boreal region. Future research on the pollen from the cores from Meekin and Gall Lake should provide further information as to the timing of climatic change across this region. 5.4. Relevance of findings to regional climatic change in NW Ontario Warm temperatures over the last several decades in northwest Ontario (Laird et al., 2012), combined with future projections for a further increase of over 2
C over the next 20 years (Chiotti and Lavender, 2008) have led studies that help address the risk of climate change on water quality and quantity. The northwest region of Ontario is fortunate to have had observations on the physical, chemical and biological changes in a number lakes since 1969, with the Experimental Lakes Area (ELA) being located in this region. Scientists have been able to document changes in limnological variables over the past 46 years as climate changed including measurements of stream flow, lake level, water retention times, dissolved organic carbon (DOC), water transparency, and thermal structure of the lake (Schindler et al., 1990, 1992, 1996a, b), as well as biological change (e.g., Findlay et al., 2001). Instrumental records from northwest Ontario also show that droughts have occurred during the 1930s, 1950s, and 1980s, but these droughts were spatially asynchronous across climate stations (Laird et al., 2012). Based on this and other available information, Chiotti and Lavender (2008) concluded that the central region of Ontario is at minimal risk from climatic change to water quantity, and 20e50 years away from any negative impacts on water quality related to climate warming (Chiotti and Lavender, 2008). However, this assessment was admittedly based on limited data, over a very large region. Paleolimnological approaches have now provided a more detailed assessment of the vulnerability of this region to climate change. Using the longer temporal perspective, provided by proxy data in well-dated sediment cores from this region, it can be concluded that this region is sensitive to climatic change, in terms of water availability and water quality. These studies have shown that the instrumental record of the last 100 years is not representative of the climatic variability that has been experienced over the past two millennia (Laird et al., 2012), and that changes of the past two millennia are not representative of changes over the last 10,000 years. In a study of a network of six northwest Ontario lakes over the past 2000 years, Laird et al. (2012) concluded that a regional-scale period of aridity occurred between 900 and 1400 CE, a time period commonly referred to as the Medieval Climate Anomaly. Finally, changes over the Holocene show that the currently high water levels in this region are a phenomenon of the last several thousand years (Karmakar et al., 2015) and that lower water levels in the past are linked to climate, and both water quantity and quality have changed substantially over the Holocene. This long-term perspective clearly shows that the boreal region in northwest Ontario, is indeed sensitive to regional-scale changes in

177

moisture balance. Given that the estimated aridity of the midHolocene was of a similar magnitude to the projected changes over the next 20 years, and is dwarfed by projected changes in temperature by 2050 (Chiotti and Lavender, 2008), it can be argued that the changes in water quantity and quality that occurred when the climate was warmer and more arid than present in the EMH may have relevance as potential climatic scenarios that we may face in the future. 6. Conclusions The prairie-forest ecotone in northwest Ontario is a climaticallysensitive region. Past changes in head-water lakes in this region have undergone large changes in lake hydrology in response to changing climatic conditions over the Holocene. Most importantly, this study has shown that boreal lakes can experience significant changes in algal abundance associated with warmer and more arid climatic conditions. Finally, changes across the boreal region of northwest Ontario were complex, with potentially-important timetransgressive changes related to climate, or lake-specific differences. If the EMH period is a good analog for understanding future climate change, this study suggests enhanced algal production with associated cyanobacteria blooms, could become a more common phenomenon in the future under a warmer climate. Acknowledgements We would like to thank Brendan Wiltse, and Donya Danesh for assistance with field work. Many thanks to Zoraida QuinonesRivera at the University of Regina in HPLC training. We thank Lac Core at the University of Minnesota for preparing pollen samples for AMS dating, and Tom Brown at Lawrence Livermore National Laboratory for supervising the preparation and analyses of the pollen samples for the AMS dates. This project was funded by an NSERC Discovery Grant RGPIN/170321-2011 to BFC and a Graduate Dean Doctoral travel award to Moumita Karmakar from Queen's University. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.quascirev.2015.06.025. References Adrian, R., Wilhelm, S., Gerten, D., 2006. Life history traits of lake plankton species may govern their phonological response to climate warming. Glob. Change Biol. 12, 652e661. Adrian, R., O' Reilly, C.M., Zagarese, H., Baines, S.B., Hessen, D.O., Keller, W., Livingstone, D.M., Sommaruga, R., Straile, D., Van Donk, E., Weyhenmeyer, G.A., Winder, M., 2009. Lakes as sentinels of climate change. Limnol. Oceanogr. 54, 2283e2297. Battarbee, R.W., Kneen, M.J., 1982. The use of electronically counted microspheres in absolute diatom analysis. Limnol. Oceanogr. 27, 184e188. Beaulieu, M., Pick, F., Gregory-Eaves, I., 2013. Nutrients and water temperature are significant predictors of cyanobacterial biomass in a 1147 lakes data set. Limnol. Oceanogr. 58, 1736e1746. €rck, S., 1985. Deglaciation chronology and revegetation in northwestern Ontario. Bjo Can. J. Earth Sci. 22, 850e871. Blaauw, M., 2010. Methods and codes for ‘classical’ age-modelling of radiocarbon sequences. Quat. Geochronol. 5, 512e518. Bradbury, J.P., Dieterich-Rurup, K., 1993. Holocene diatom paleolimnology of Elk Lake, Minnesota. In: Bradbury, J.P., Dean, W.E. (Eds.), Elk Lake, Minnesota: Evidence for Rapid Climate Change in the North-Central United States. Geological Society of America, Boulder, CO, pp. 215e237. USA Special paper 276. Brown, T.A., Nelson, D.E., Mathewes, R.W., Vogel, J.S., Southon, J.R., 1989. Radiocarbon dating of pollen by accelerator mass spectrometry. Quat. Res. 32, 1205e1212. Camburn, K.R., Charles, D.F., 2000. Diatoms of Low-alkalinity Lakes in Northwestern United States. Academy of Natural Sciences, Philadelphia.

178

M. Karmakar et al. / Quaternary Science Reviews 123 (2015) 168e179

Carmichael, W.W., Beasley, V., Bunner, D.L., Eloff, J.N., Falcorner, I., Gorham, P., Harada, K., Krishnamurthy, T., Min-Juan, Y., Moore, R.E., Rinehart, K., Runnegar, M., Skulberg, O.M., Watanabe, M., 1988. Naming of cyclic heptapeptide toxins of cyanobacteria (cyanobacteria algae). Toxicon 26, 971e974. Carey, C.C., Ibelings, B.W., Hoffmann, E.P., Hamilton, D.P., Brookes, J.D., 2012. Ecophysiological adaptations that favour freshwater cyanobacteria in a changing climate. Water Res. 46, 1394e1407. Chiotti, Q., Lavender, B., 2008. In: Lemmen, D.S., Warren, F.J., Lacroix, J., Bush, E. (Eds.), From Impacts to Adaptation: Canada in a Changing Climate 2007. Government of Canada, Ottawa, pp. 28e56. Collins, M., Knutti, R., Arblaster, J., Dufresne, J.L., Fichefet, T., Friedlingstein, P., Gao, X., Gutowski, W.J., Johns, T., Krinner, G., Shongwe, M., Tebaldi, C., Weaver, A.J., Wehner, M., 2013. Long-term climate change: projections, commitments and irreversibility. In: Stocker, T.F., Qin, D., Plattner, G.K., Tignor, M., Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V., Midgley, G.F. (Eds.), Climate Change 2013: the Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, pp. 1029e1136. Cumming, B.F., Wilson, S.E., Hall, R.I., Smol, J.P., 1995. Diatoms from British Columbia (Canada) Lakes and Their Relationship to Salinity, Nutrients, and Other Limnological Variables.Gebruder. Borntrager, Berlin. Diehl, S., Berger, S., Ptacnik, R., Wild, A., 2002. Phytoplankton, light, and nutrients in a gradient of mixing depths: field experiments. Ecology 83, 399e411. Fietz, S., Nicklisch, A., Oberhansli, H., 2007. Phytoplankton response to climate changes in Lake Baikal during the Holocene and Kazantsevo interglacials assessed from sedimentary pigments. J. Paleolimnol. 37, 177e203. Findlay, D.L., Kasian, S.E.M., Stainton, M.P., Beaty, K., Lyng, M., 2001. Climatic influences on algal populations of boreal forest lakes in the experimental lakes area. Limnol. Oceanogr. 46, 1784e1793. Fritz, S.C., 1989. Lake development and limnological response to prehistoric and historic land-use in Diss, Norfolk, U.K. J. Ecol. 77, 182e202. Garrison, P.J., Wakeman, R.S., 2000. Use of paleolimnology to document the effect of shoreland development on water quality. J. Paleolimnol. 24, 369e393. Gibson, C.E., Anderson, J.,N., Haworth, E.Y., 2003. Aulacoseirasubarctica: taxonomy, physiology, ecology and palaeoecology. Eur. J. Phycol. 38, 83e101. Ginn, B.K., Cumming, B.F., Smol, J.P., 2007. Diatom-based environmental inferences and model comparisons from 494 northeastern North American lakes. J. Phycol. 43, 647e661. Glew, J.R., Smol, J.P., Last, W.M., 2001. Sediment core collection and extrusion. In: Last, W.M., Smol, J.P. (Eds.), Tracking Environmental Change Using Lake Sediments, Basin Analysis Coring and Chronological Techniques, vol. I. Kluwer Academic Publishers, Dordrecht, pp. 73e105. Gorham, E., Sanger, J.E., 1975. Fossil pigments in Minnesota lake sediments and their bearing upon the balance between terrestrial and aquatic inputs to sedimentary organic matter. Verh. Des. Int. 19, 2267e2273. Griffiths, M., Perrot, P.S., Edmonson, W.T., 1969. Oscillaxanthin in the sediments of Lake Washington. Limnol. Oceanogr. 14, 317e326. Griffiths, M., 1978. Specific cyanobacteria algal carotenoids in the sediments of Esthwaite water. Limnol. Oceanogr. 23, 777e784. Grimm, E.C., 1987. CONISS, a Fortran77 program for stratigraphically constrained cluster analysis by the method of incremental sum of squares. Computers and Geosciences 13, 13e35. Grimm, E.C., 2004. TILIA and TILIA GRAPH Computer Programs Version 2.0.2. Illinois State Museum Research and Collections Center. Guilizzoni, P., Bonomi, G., Galanti, G., Ruggiu, D., 1982. Basic trophic status and recent development of some Italian lakes as revealed by plant pigment and other chemical components in sediment cores. Mem. dell' Ist. Idrobiol. 40, 79e98. Guilizzoni, P., Bonomi, G., Galanti, G., Ruggiu, D., 1983. Relationship between sedimentary pigments and primary production: evidence from core analyses of twelve Italian lakes. Hydrobiologia 103, 103e106. Haig, H.A., Kingsbury, M.V., Laird, K.R., Leavitt, P.R., Laing, R., Cumming, B.F., 2013. Assessment of drought over the past two millennia using near-shore sediment cores from a Canadian boreal lake. J. Paleolimnol. 50, 175e190. Hall, R.I., Leavitt, P.R., Quinlan, R., Dixit, A.S., Smol, J.P., 1999. Effects of agriculture, urbanization and climate on water quality in the northern Great Plains. Limnol. Oceanogr. 43, 739e756. Hickman, M., Schweger, C.E., Klarer, D.M., 1990. Baptiste Lake, Alberta, A Late Holocene history of changes in a lake and its catchment in the southern boreal forest. J. Paleolimnol. 4, 253e267. Huisman, J., Sharples, J., Stroom, J.M., Visser, P.M., Kardinaal, W.E.A., Verspagen, J.M.H., Sommeijer, B., 2004. Changes in turbulent mixing shift competition for light between phytoplankton species. Ecology 85, 2960e2970. Interlandi, S.J., Kilham, S.S., Theriot, E.C., 1999. Responses of phytoplankton to varied resource availability in large lakes of the greater Yellowstone ecosystem. Limnol. Oceanogr. 44, 668e682. Jeppesen, E., Sondergaard, M., Jensen, J.P., Havens, K., Anneville, O., Carvalho, L., Coveney, M.F., Deneke, R., Dokulil, M., Foy, B., Gerdeaux, D., Hampton, S.E., Kangur, K., Kohler, J., Korner, S., Lammens, E., Lauridsen, T.L., Manca, M., Miracle, R., Moss, B., Noges, P., Persson, G., Phillips, G., Portielje, R., Romo, S., Schelske, C.L., Straile, D., Tatrai, I., Willen, E., Winder, M., 2005. Lake responses to reduced nutrient loadingdan analysis of contemporary long-term data from 35 case studies. Freshw. Biol. 50, 1747e1771.

€hnk, K.D., Huisman, J., Sharples, J., Sommeijer, B., Visser, P.M., Stroom, J.M., 2008. Jo Summer heatwaves promote blooms of harmful cyanobacteria. Glob. Change Biol. 14, 495e512. Juggins, S., 2003. C2 Software for Ecological and Palaeoecological Data Analysis and Visualization. User Guide Version 1.3. University of Newcastle, Newcastle, UK. Karmakar, M., Laird, K.R., Cumming, B.F., 2015. Diatom-based evidence of regional aridity during the mid-Holocene period in boreal lakes from northwest Ontario (Canada). Holocene 25, 166e177. Kingsbury, M.V., Laird, K.R., Cumming, B.F., 2012. Consistent patterns in diatom assemblages and diversity measures across water-depth gradients from eight Boreal lakes from north-western Ontario (Canada). Freshw. Biol. 57, 1151e1165. Kirilova, E., Heiri, O., Enters, D., Cremer, H., Lotter, A.F., Zolitschka, B., Hubener, T., 2009. Climate-induced changes in the trophic status of a Central European lake. J. Limnol. 68, 71e82. Kosten, S., Huszar, V.L.M., Becares, E., Costa, L.S., Van Donk, E., Hansson, L.A., Jeppesen, E., Kruk, C., Lacerot, G., Mazzeo, N., Meester, L.De, Moss, B., Luerling, M., Noges, T., Romo, S., Scheffer, M., 2012. Warmer climates boost cyanobacterial dominance in shallow lakes. Glob. Change Biol. 18, 118e126. Krammer, K., Lange-Bertalot, H., 1986. Bacillariophyceae 1: Teil: Naviculaceae. In: €rtner, G., Gerloff, J., Heynig, H., Mollenhauer, D. (Eds.), Süßwasserflora Ettl, H., Ga von Mitteleuropa, Band 2/1. Gustav Fischer Verlag, Stuttgart/New York, 876 pp. Krammer, K., Lange-Bertalot, H., 1988. Bacillariophyceae 2: Teil: Bacillariaceae, Epithmiaceae, Surirellaceae. In: Ettl, H., G€ artner, G., Gerloff, J., Heynig, H., Mollenhauer, D. (Eds.), Süßwasserflora von Mitteleuropa, Band 2/2. Gustav Fischer Verlag, Stuttgart/New York, 611 pp. Krammer, K., Lange-Bertalot, H., 1991a. Bacillariophyceae 3: Teil: Centrales, Fragilariaceae, Eunotiaceae. In: Ettl, H., G€ artner, G., Gerloff, J., Heynig, H., Mollenhauer, D. (Eds.), Süßwasserflora von Mitteleuropa, Band 2/3. Gustav Fischer Verlag, Stuttgart/New York, 576 pp. Krammer, K., Lange-Bertalot, H., 1991b. Bacillariophyceae. 4: Teil: Achnanthaceae. In: Ettl, H., G€ artner, G., Gerloff, J., Heynig, H., Mollenhauer, D. (Eds.), Süßwasserflora von Mitteleuropa, Band 2/4. Gustav Fischer Verlag, Stuttgart/ New York, 437 pp. Kromkamp, J., Botterweg, J., Mur, L.R., 1988. Buoyancy regulation in microcystis aeruginosa grown at different temperatures. FEMS Mircobiol. Lett. 53, 231e237. Laird, K.R., Haig, H.A., Ma, S., Kingsbury, M.V., Brown, T.A., Lewis, M., Oglesby, R., Cumming, B.F., 2012. Expanded spatial extent of the medieval climate anomaly revealed in lake-sediment records across the boreal region in northwest Ontario. Glob. Change Biol. 18, 2869e2881. Lange-Beralot, H., Melzeltin, D., 1996. Indicators of Oligotrophy. Koeltz Scientific € nigstein, Germany. Books, Ko Leavitt, P.R., 1988. Experimental determination of carotenoid degradation. J. Paleolimnol. 1, 215e228. Leavitt, P.R., Hodgson, D.A., 2001. Sedimentary pigments. In: Smol, J.P., Birks, H.J.B.B., Last, W.M. (Eds.), Tracking Environmental Change Using Lake Sediments, Terrestrial, Algal, and Siliceous Indicators, vol. 3. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 295e326. Lewis, C.F.M., Forbes, D.L., Todd, B.J., Nielsen, E., Thorleifson, L.H., Henderson, P.J., McMartin, I., Anderson, T.W., Betcher, R.N., Buhay, W.M., Burbidge, S.M., Schroder-Adams, C.J., King, J.W., Moran, K., Gibson, C., Jarrett, C.A., King, H.J., Lockhart, W.L., Last, W.M., Matile, G.L.D., Risberg, J., Rodrigues, C.G., Telka, A.M., Vance, R.E., 2001. Uplift-driven expansion delayed by middle Holocene desiccation in Lake Winnipeg, Manitoba, Canada. Geology 29, 743e746. Livingstone, D.M., Dokulil, M.T., 2001. Eighty years of spatially coherent Austrian lake surface temperatures and their relationship to regional air temperature and the North Atlantic oscillation. Limnol. Oceanogr. 46, 1220e1227. Ma, S., Laird, K.R., Kingsbury, M.V., Lewis, C.F.M., Cumming, B.F., 2013. Diatominferred changes in effective moisture during the late Holocene from nearshore cores in the southeastern region of the Winnipeg river drainage basin (Canada). Holocene 23, 568e578. Magnuson, J.J., Robertson, D.M., Benson, B.J., Wynne, R.H., Livingstone, D.M., Arai, T., Assel, R.A., Barry, R.G., Card, V., Kuusisto, E., Granin, N.G., Prowse, T.D., Stewart, K.M., Vuglinski, V.S., 2000. Historical trends in lake and river ice cover in the Northern Hemisphere. Science 291, 254. McAndrews, J.H., 1982. Holocene environment of a fossil bison from Kenora, Ontario. Ont. Archaeol. 37, 41e51. Mcknight, D.D., Brakke, F., Mulholand, P.J., 1996. Freshwater ecosystems and climate change in North America. Limnol. Oceanogr. 41, 815e1149. Michelutti, N., Blais, J.M., Cumming, B.F., Paterson, A.M., Rühland, K., Wolf, A.P., Smol, J.P., 2010. Do spectrally inferred determinations of chlorophyll a reflect trends in lake trophic status? J. Paleolimnol. 43, 205e217. Moos, M.T., Laird, K.R., Cumming, B.F., 2005. Diatom assemblages and water depth in Lake 239 (Experimental Lakes Area, Ontario): implications for paleoclimatic studies. J. Paleolimnol. 34, 217e227. Moos, M.T., Laird, K.R., Cumming, B.F., 2009. Climate-related eutrophication of a small boreal lake in northwestern Ontario: a paleolimnological perspective. Holocene 19, 359e367. Moos, M.T., Cumming, B.F., 2011. Changes in the parkland-boreal forest boundary in northwestern Ontario over the Holocene. Quat. Sci. Rev. 30, 1232e1242. Moos, M.T., Cumming, B.F., 2012. Climate-fire interactions during the Holocene: a test of the utility of charcoal morphotype in a sediment core from the boreal region of north-western Ontario (Canada). Int. J. Wildland Fire 21, 640e652. O'Neil, J.M., Davis, T.W., Burford, M.A., Gobler, C.J., 2012. The rise of harmful cyanobacteria blooms: the potential roles of eutrophication and climate change. Harmful Algae 14, 313e334.

M. Karmakar et al. / Quaternary Science Reviews 123 (2015) 168e179 Paerl, H.W., Huisman, J., 2008. Climatedblooms like it hot. Science 320, 57e58. Read, J.S., Winslow, L.A., Hansen, G.J.A., van den Hoek, J., Hanson, P.C., Bruce, L.C., Markfort, C.D., 2014. Simulating 2368 temperate lakes reveals weak coherence in stratification phenology. Ecol. Model. 291, 142e150. Reavie, E.D., Neill, K.E., Little, J.L., Smol, J.P., 2006. Cultural eutrophication trends in three southeastern Ontario lakes: a paleolimnological perspective. Lake Reserv. Manag. 22, 44e58. Reimer, P.J.P., 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., Weyhenmeyer, C.E., 2009. INTCAL09 and MARINE09 radiocarbon age calibration curves, 0e50000 years cal BP. Radiocarbon 51, 1111e1150. Reynolds, C.S., 2006. Ecology of Phytoplankton (Ecology, Biodiversity and Conservation). Cambridge University Press, Cambridge, UK. Rippey, B., Anderson, N.J., Foy, R., 1997. Accuracy of diatom-inferred total phosphorus concentration and the accelerated eutrophication of a lake due to reduced flushing and increased nutrient loading. Can. J. Fish. Aquat. Sci. 54, 2637e2646. Robarts, R.D., Zohary, T., 1987. Temperature effects on photosynthetic capacity, respiration, and growth rates of bloom forming cyanobacteria. N. Z. J. Mar. Freshw. Sci. 21, 391e399. Rodriguez, J., Tintore, J., Allen, J.T., Blanco, J.M., Gomis, D., Reul, A., Ruiz, J., Rodriguez, V., Echevarria, F., Jimenez-Gomez, F., 2001. Mesoscale vertical motion and the size structure of phytoplankton in the ocean. Nature 410, 360e363. Schindler, D.W., Beaty, K.G., Fee, E.J., Cruikshank, D.R., DeBruyn, E.D., Findlay, D.L., Linsey, G.A., Shearer, J.A., Stainton, M.P., Turner, M.A., 1990. Effects of climatic warming on lakes of the central boreal forest. Science 250, 967e970. Schindler, D.W., Bayley, S.E., Curtis, P.J., Parker, B.R., Stainton, M.P., Kelly, C.A., 1992. Natural and man-caused factors affecting the abundance and cycling of dissolved organic substances in Precambrian Shield lakes. Hydrobiologia 229, 1e21. Schindler, D.W., Bayley, S.E., Parker, B.R., 1996a. The effects of climatic warming on the properties of boreal lakes and streams at the experimental lakes area, northwestern Ontario. Limnol. Oceanogr. 41, 1004e1017.

179

Schindler, D.W., Curtis, P.J., Parker, B., Stainton, M.P., 1996b. Consequences of climate warming and lake acidification for UV-B penetration in North American boreal lakes. Nature 379, 705e708. Smol, J.P., Wolfe, A.P., Birks, H.J.B., Douglas, M.S.V., Jones, V.J., Korhola, A., Pienitz, R., Ruhland, K., Sorvari, S., Antoniades, D., Brooks, S.J., Fallu, M.A., Hughes, M., Keatley, E., Laing, T.E., Michelutti, N., Nazarova, L., Nyman, M., Paterson, A.M., Perren, B., Quinlan, R., Rautio, M., Saulnier-Talbot, E., Siitoneni, S., Solovieva, N., Weck-strom, J., 2005. Climate-driven regime shifts in the biological communities of arctic lakes. Proc. Natl. Acad. Sci. U. S. A. 102, 4397e4402. Swain, E.B., 1985. Measurement and interpretation of sedimentary pigments. Freshw. Biol. 15, 53e75. Taranu, Z., Zurawell, R., Pick, F., Gregory-Eaves, I., 2012. Predicting cyanobacterial dynamics in the face of global change: the importance of scale and environmental context. Glob. Change Biol. 18, 3477e3490.  Ter Braak, C.J.F., Smilauer, P., 2012. Canoco Reference Manual and User's Guide: Software for Ordination (Version 5.0). Microcomputer Power, Ithaca, NY, USA, 496 pp. Umbanhowar, C.E., Camill, P., Geiss, C.E., Teed, R., 2006. Asymmetric vegetation responses to mid-Holocene aridity at the prairie-forest ecotone in south-central Minnesota. Quat. Res. 66, 53e66. Wilhelm, S., Adrian, R., 2008. Impact of summer warming on the thermal characteristics of a polymictic lake and consequences for oxygen, nutrients and phytoplankton. Freshw. Biol. 53, 226e237. Williams, S.G., Stefan, H.G., 2006. Modeling of lake ice characteristics in North America using climate, geography, and lake bathymetry. J. Cold Region Eng. 20, 140e167. Wilson, S.E., Cumming, B.F., Smol, J.P., 1996. Assessing the reliability of salinity inference models from diatom assemblages: an examination of a 219-lake data set from western North America. Can. J. Fish. Aquat. Sci. 53, 1580e1594. Winder, M., Hunter, D.A., 2008. Temporal organization of phytoplankton communities linked to chemical and phys- ical forcing. Oecologia 156, 179e192. Züllig, H., 1981. On the use of carotenoid stratigraphy in lake sediments for detecting past developments of phytoplankton. Limnol. Oceanogr. 26, 970e976. Züllig, H., 1982. Untersuchungenüber die Stratigraphie von Carotinoidenimgeschichteten Sediment von 10 Schweizer Seen zurErkundungfrüherer Phytoplankton-Entfaltungen. Schweiz. Z. Hydrol. 44, 1e98.