Patterns in the crustacean zooplankton community in ...

Journal of Great Lakes Research 43 (2017) 80–90

Contents lists available at ScienceDirect

Journal of Great Lakes Research journal homepage: www.elsevier.com/locate/jglr

Patterns in the crustacean zooplankton community in Lake Winnipeg, Manitoba: Response to long-term environmental change Brenda Hann a,⁎, Alex Salki b a b

Department of Biological Sciences, University of Manitoba, Winnipeg, MB, Canada Salki Consultants, Inc., Winnipeg, MB, Canada

a r t i c l e

i n f o

Article history: Received 28 June 2016 Accepted 28 October 2016 Available online 20 December 2016 Keywords: Zooplankton Climate change Eutrophication Large lakes Lake Winnipeg Non-indigenous species

a b s t r a c t Changes in the crustacean zooplankton community composition and abundance in Lake Winnipeg (1969–2006) provide a rare opportunity to examine their response to environmental changes in the largest naturally eutrophic lake on the Canadian prairies. Since 1929, zooplankton species composition in Lake Winnipeg has changed little except for the addition of the invasive cladoceran, Eubosmina coregoni in 1994. The dominant taxa in the lake in summer include: Leptodiaptomus ashlandi, Acanthocyclops vernalis, Diacyclops thomasi, Daphnia retrocurva, Daphnia mendotae, Diaphanosoma birgei, Eubosmina coregoni, and Bosmina longirostris. Climate-accelerated nutrient loading to southern Lake Winnipeg over the last two decades has led to increased phytoplankton abundance and higher frequency of cyanobacterial blooms especially in its northern basin. Crustacean zooplankton have likewise increased especially in the North Basin, but less so in the more nutrient rich South Basin, possibly as a consequence of higher densities of pelagic planktivorous fish and light-limited primary production compared with the more transparent North basin (Brunskill et al., 1979, 1980). Calanoid copepods play a larger role in the South basin food web in contrast to cyclopoid copepods and Cladocera in the North basin. The study begins to fill the recognized gap in understanding of Lake Winnipeg's food web structure and provides a baseline for evaluating ongoing changes in the zooplankton community with the arrival of new non-indigenous taxa, e.g. Bythotrephes longimanus and Dreissena polymorpha. It reinforces previous work demonstrating that zooplankton provide valuable indices toward evaluating the health of an ecosystem. © 2016 International Association for Great Lakes Research. Published by Elsevier B.V. All rights reserved.

Introduction Zooplankton are essential components of the pelagic food web of aquatic ecosystems, comprising a vital intermediate trophic link, subject to both top-down predation as well as bottom-up dietary factors driven by nutrient availability. Changes in predators and food quality or quantity combine to influence the abundance and composition of the zooplankton community that reflect changes occurring in the lake ecosystem. Zooplankton can serve as indicators of climate change in Canadian lakes as regional climate is known to be the main driver of variation in crustacean community structure (Pinel-Alloul et al., 2013). Zooplankton can also be an excellent, cost-efficient indicator for assessment of eutrophication in lake ecosystems (Ejsmont-Karabin, 2012; Ejsmont-Karabin and Karabin, 2013; Jeppesen et al., 2011; Caroni and Irvine, 2010; Mimouni et al., 2015). Lakes and their biotic communities act as sentinels of climate change (Adrian et al., 2009; Leavitt et al., 2009). Climate change has influenced Lake Winnipeg in several ways, including increased mean annual water ⁎ Corresponding author. E-mail address: [email protected] (B. Hann).

temperatures and elevated precipitation in its southern Red River catchment. This has led to higher flow rates in the Red River over the last two decades (particularly in 1997 and 2005), more severe floods and droughts, altered sediment, detritus, nutrient, and contaminant loading (Stewart et al., 2000, McCullough et al., 2012). Variation in water residence time of Lake Winnipeg has occurred, as well as increased duration of the open water period (Environment Canada and Manitoba Water Stewardship, 2011; Cushing, 1998). All of these factors have interacted to affect the water chemistry and biota in the lake. Several non-indigenous species have entered Lake Winnipeg in recent years, including white bass (Roccus chrysops) in the 1960s (Stewart and Watkinson, 2004), rainbow smelt (Osmerus mordax), and the cladoceran, Eubosmina coregoni, in the 1990s (Franzin et al., 1994; Salki, 1996; Suchy et al., 2010). Lake Winnipeg, like many prairie lakes, has been mildly naturally eutrophic for centuries (Bunting et al., 2011; Kling et al., 2011). Accelerated nutrient loading over the last two decades has led to increased phytoplankton abundance (Kling et al., 2011) and higher frequency, extent, and duration of cyanobacterial blooms (McCullough et al., 2012). In 1969, cyanobacteria constituted only 56% of the total heterogeneous phytoplankton community; however, since that time, cyanobacteria

http://dx.doi.org/10.1016/j.jglr.2016.10.015 0380-1330/© 2016 International Association for Great Lakes Research. Published by Elsevier B.V. All rights reserved.

B. Hann, A. Salki / Journal of Great Lakes Research 43 (2017) 80–90

have increased to comprise 90% of the total phytoplankton biomass, especially during mid-summer (Kling et al., 2011; Environment Canada and Manitoba Water Stewardship, 2011). In order to develop effective adaptive lake management strategies, climate indices must be teased apart from confounding and interacting biotic factors, e.g. eutrophication, fishery management, and invasion of non-indigenous species, for which an extensive temporal record is required. Long-term datasets for extant biotic communities and their associated environmental variables in lakes are few (Shurin et al., 2010), but even more rare for large lakes. Substantial changes have been documented in the zooplankton community of some larger, shallow lakes such as Lake Champlain, Lake Balaton, and especially Lake Erie among the Laurentian Great Lakes (Davis, 1969; Fahnenstiel et al., 1998; Madenjian et al., 2002; Stewart et al., 2010; Johannsson et al., 2000; Makarewicz, 1993; Barbiero and Rockwell, 2008; Kane et al., 2004; Miller et al., 2010). These changes have been correlated with effects on commercial fisheries and water quality from prolonged nutrient loading. However, identification of primary and secondary drivers of such changes remains a daunting task. Whereas the cladoceran component of the zooplankton community is amenable to analysis over the long term using sedimentary remains (Alric et al., 2013), copepods are much less well preserved, and both are subject to differential preservation in the sediment record. Hence, decades-long records of extant crustacean zooplankton communities that encompass both Cladocera and Copepoda provide an invaluable opportunity to investigate the impact of the interactions among species and environmental variables over a time span that encompasses preand post-disturbance periods (Battarbee, 2000). Lake Winnipeg is unique in providing that opportunity. There is a critical need for a detailed study of temporal changes in the zooplankton community of Lake Winnipeg, particularly to serve as a baseline in light of the quite recent invasion of spiny waterfleas (Bythotrephes longimanus) and zebra mussels (Dreissena polymorpha) and their potential impact on the food web. The main question to be addressed is how has the overall crustacean zooplankton community changed in response to environmental stressors such as nutrient loading (leading to cultural eutrophication) and climate change in Lake Winnipeg over the last 40 years? Methods Study site: Lake Winnipeg Lake Winnipeg is the largest remnant of glacial Lake Agassiz (Patalas, 2006), and is now a eutrophic prairie lake, 10th largest by surface area in the world. The lake is relatively shallow (mean depth = 12 m) with a long fetch (Brunskill et al., 1980); hence it is well mixed and rarely stratifies during the open water period. The North basin is deeper (mean depth = 13.3 m) than the South basin (mean depth = 9.7 m) (Brunskill et al., 1980). Limnological and water quality characteristics that differentiate the two basins are well established: the North basin is cooler, more transparent, and less nutrient enriched than the South basin (Brunskill et al., 1979; Lumb et al., 2012; McCullough et al., 2012). Although records are sparse, there was no evidence of bottom water hypoxia (b 2 mg O2/L) during the period 1964–2002; in 2003 hypoxia was detected in summer in the North basin (Wassenaar, 2012). There is growing evidence from forage fish and cormorant diets obtained via stable isotopes (Hobson et al., 2012; Ofukany et al., 2015) that the North and South basins of Lake Winnipeg are biologically distinct. Hence, the zooplankton communities characterizing the two basins were analyzed separately. The Lake Winnipeg watershed (one million km2) is 40 times larger than the lake surface area, the highest ratio among the world's great lakes. The lake receives water and nutrients via N 60 rivers, but the Nelson River is the exclusive outflow (Rosenberg et al., 2005). Half of the water inflow to the South Basin comes from the Winnipeg River

81

draining the Precambrian Shield; however, the majority of nutrients come from the Red River and agricultural regions to the south (68% of total phosphorus; Environment Canada and Manitoba Water Stewardship, 2011). Since 1990, an escalation in Red River nutrient loading, caused by climate-related increased precipitation, runoff, and flooding of farmland and urban areas, is mainly responsible for the intensification of eutrophication of Lake Winnipeg (McCullough et al., 2012). The main water inflow into the North Basin is the Saskatchewan River. It carries only a modest nutrient load despite draining a much larger watershed than the more southern rivers, largely as a consequence of the construction of the Cedar Lake-Grand Rapids impoundment that reduced sediment inputs and increased transparency of the North basin in the 1960s (Patalas and Salki, 1992). Water levels in Lake Winnipeg have been regulated since 1976 (see McCullough et al., 2012). Because Lake Winnipeg zooplankton community diversity is generally highest during the summer period (Patalas and Salki, 1992) and summer samples were collected in each of the 8 survey years, this investigation focussed on changes in the summer zooplankton community. We hypothesised that the summer crustacean zooplankton community would have increased in abundance as a consequence of eutrophication (McCauley and Kalff, 1981; Hanson and Peters, 1984; Patalas and Salki, 1992). Whereas an increase in nutrient loading may be expected to lead to higher primary producer biomass, the consequences of possible changes in N:P ratio and predominance of possibly inedible and/or toxic cyanobacteria may offset the benefits of increased nutrient loading for zooplankton (Haney, 1987; Ferrao-Filho et al., 2007; Bednarska et al., 2011, Kling et al., 2011). Lake Winnipeg also supports the second largest inland commercial fishery in North America with average round weight production, largely walleye, of 12.8 million kg/yr during the 2000s (Manitoba Conservation and Water Stewardship, Fisheries Branch, 2012). Walleye catch rates recently (1998–2008) have been at an unprecedented high level (Ayles et al., 2011), an order of magnitude higher than in 1969 (W. Lysack, Manitoba Water Stewardship, Fisheries Branch), exerting substantial predation pressure on the ecosystem. Environmental variables To evaluate potential drivers of patterns of change in zooplankton species composition and abundance over the years, a suite of relevant environmental variables was assessed. Basin-volume weighted total phosphorus (TP) data, chlorophyll a concentrations, phytoplankton biomass (whole lake), and plankton (phytoplankton + zooplankton) biovolume (whole lake) from mid-summer surveys were reported in McCullough et al. (2012). Surface water temperatures (SWTemp), Secchi transparency (Secchi), total nitrogen (TN), cyanobacterial biomass (whole lake), and TP loadings from the Red River (into the South basin) were summarized from the State of Lake Winnipeg 1999 to 2007 Report (Environment Canada and Manitoba Water Stewardship, 2011). Planktivorous fish biomass (whole lake) was reported in Lumb et al. (2012) and walleye landings (whole lake) were extracted from Ayles et al. (2011). TP and TN showed collinearity (see ‘Results’), so only TP was used in analyses. Similarly, chlorophyll a, phytoplankton biomass, and cyanobacteria biomass showed collinearity, and only cyanobacteria biomass was used in the analysis. Zooplankton datasets To examine long-term summer patterns in zooplankton species composition and abundance in Lake Winnipeg, historical datasets were compared with samples collected over the last decade: 1) 1928–29: Semi-quantitative sampling of the lake by Bajkov (1934) included both pelagic and littoral samples. 2) 1969: Six lake-wide field surveys conducted at approximately monthly intervals throughout the open water season at 83

82


occurred among the datasets included in this study, and the most recent classification was used (see McLaughlin et al., 2005). Data analyses

Fig. 1. Lake Winnipeg showing sampling stations in North and South basins demarcated by the thick black line.

3)

4)

5)

6)

predetermined stations (not all stations sampled for zooplankton) (Fig. 1). Patalas and Salki (1992) published the first detailed description of spatial and seasonal variation in the pelagic crustacean zooplankton community of Lake Winnipeg. Data from only the 3 summer surveys were included in our analyses. 1994: Lake Winnipeg Project by the Geological Survey of Canada, focused on geological analysis of bottom sediments, coupled with biological and water sampling (Todd et al., 1996, 1998). 1998: The International Joint Commission investigation of the effects of the 1997 Red River Flood on the South Basin of Lake Winnipeg (Stewart et al., 2000, 2003). 1999: An inaugural July–August survey by Lake Winnipeg Research Consortium Inc. (LWRC), formed to facilitate and co-ordinate research by provincial, federal and non-profit agencies. 2002–2006: LWRC conducted open-water surveys in spring, summer, and fall of each year except for 2005 when only fall samples were collected. Only summer data were included in the analyses.

Field sampling Zooplankton samples (1969–2006) were collected using consistent methods, i.e. an integrated zooplankton haul, from 1 m above the bottom sediments to the surface with the use of a single Wisconsin net, 1 m long, with 73 μm mesh and 25-cm mouth diameter. Samples were made up to a standard volume of 125 mL and preserved with 10% formalin. Zooplankton abundance was expressed as numbers of individuals per litre, assuming 100% net filtration efficiency (discussed in more detail in Patalas and Salki (1992)). Laboratory analyses All zooplankton samples were analyzed using methods outlined in Salki and Patalas (1992). Each zooplankton sample was examined at 63–160× magnification and at least 200 individuals per sample were identified at the species level. Several taxonomic revisions have

All stations were not sampled in every survey. Summer sampling periods in 1969 (July 9–16, July 24–August 1, September 2–10) were averaged for comparison with summer surveys from other years which were not synchronous, but encompassed the overall sampling periods in 1969. Densities (mean number of individuals per litre ± standard error) were calculated for the major crustacean zooplankton groups (cladocerans, calanoid and cyclopoid copepods, and nauplii) in summer in each basin (North Basin, South Basin) in each year. Inter-annual summer patterns in total zooplankton abundance, total Cladocera, Calanoida, and Cyclopoida in the whole lake, and in North Basin and South Basin separately, were examined using one-way ANOVAs with post hoc Tukey's HSD tests (using JMP®12). Within-year variance was based on spatial variation in densities (untransformed) among stations. Zooplankton species heterogeneity was evaluated (1969–2006) using several measures each of which assesses different ecological qualities about the status of the community: (1) species richness, i.e. number of species, which is sensitive to sample sizes, (2) Shannon's diversity index (H′), and Brillouin's index, a modification of the Shannon index that takes into account unequal sample sizes, both of which are sensitive to changes in the importance of rare species, and (3) Simpson's index , a measure of dominance that is most sensitive to changes in common species (Peet, 1974; Lande, 1996). Diversity measures were not calculated for 1929 data (treated as qualitative only) as sampling methods and intensity were substantially different from all subsequent years. Other structural criteria included three commonly used density ratios (not including nauplii): (1) Calanoida: Cyclopoida, (2) large Cladocera species: total Cladocera, where large cladocerans included Daphnia spp. and Diaphanosoma birgei, and (3) Calanoida: (Cyclopoida + Cladocera) (Gannon and Stemberger, 1978). The structure and composition of the summer zooplankton communities in Lake Winnipeg incorporated the rich information contained in the species-specific densities of all taxa of Cladocera and Copepoda. Variation in community structure was examined using multivariate ordination techniques, e.g. non-metric multi-dimensional scaling (NM-MDS) using PRIMERv7 (Plymouth Routines in Multivariate Ecological Research, Clarke and Gorley, 2006). NM-MDS ordination maps the samples as points in 2-dimensional space such that the distances among points are of the same rank order as the relative (dis)similarities of the samples, measured by the Bray-Curtis resemblance matrix on the square-root transformed data. The algorithm is iterative, here repeated 25 times to converge on an optimal solution. A measure of goodness-offit or “stress” (dimensionless) is provided as a criterion of how well the multi-dimensional relationships among samples are represented by the 2-dimensional ordination plot. The lower the stress value, the more optimal the solution obtained (Clarke and Warwick, 2001). Values of stress b0.05 correspond to an excellent ordination with minimal prospect for misinterpretation; stress values between 0.1 and 0.2 indicate a useful 2dimensional representation of the structure of the groups (Clarke and Warwick, 2001). To take advantage of the species-level identification of the zooplankton community, the structure was also examined using a non-parametric permutation procedure, called analysis of similarities (ANOSIM), applied to the Bray-Curtis similarity matrix, to test the null hypothesis of no assemblage differences between years in each basin (Clarke and Green, 1988). The test statistic, R, ranges from − 1 to 1 but typically falls between zero and 1, and higher values indicate greater discrimination among groups. To further examine community change within a basin across years, we used CLUSTER analysis based on Bray-Curtis similarities derived from annual mean community composition, followed by permutation tests to produce a “similarity profile” (SIMPROF) used


to detect significant differences among clusters (Clarke and Gorley, 2006). Percentage similarity between groups was evaluated using the algorithm SIMPER within PRIMERv7 which permits identification of percentage contribution to changes between groups. The relationship between zooplankton species and environmental variables in each basin among years was examined using redundancy analysis (RDA). This is a form of canonical analysis in which the ordination vectors are linear combinations of the response variables and is an extension of linear regression to modeling multivariate response data (Legendre and Legendre, 1998, Chapter 11). RDA was executed on the mean annual multi-year zooplankton species abundance data (square-root transformed) and environmental variables (TP, surface water temperature, Secchi transparency, cyanobacterial biomass) for each basin separately using interactive forward selection of environmental predictor variables according to the amount of variation captured in the species data using CANOCO v5 (ter Braak and Šmilauer, 2012).

Results Environmental variation Substantial change in many environmental variables has occurred in Lake Winnipeg (1969–2006) as summarized in Table 1. TP and TN were significantly correlated (Pearson: n = 26, r = 0.77, p b 0.0001). Both indicators of nutrient status were significantly higher in the South basin than in the North basin (ANOVA: TP: n = 25, F = 30.0, p b 0.0001; TN: n = 25, F = 8.9, p b 0.0001). Secchi transparency was significantly greater in the North basin than the South basin (ANOVA: n = 11, F = 38.7, p b 0.0001). Significant collinearity was found among chlorophyll a, phytoplankton biomass, and cyanobacterial biomass (chla vs phytoplankton biomass: n = 7, Pearson's r = 0.89, p = 0.007; chla vs cyanobacteria biomass: n = 7, Pearson's r = 0.84, p = 0.02; phytoplankton biomass vs cyanobacteria biomass: n = 7, Pearson's r = 0.83, p = 0.02).

Table 1 Mean values for environmental variables for the open water season in each basin (where available) or for the whole lake. Year a

TP (μg/L) TNb (μg/L) Surface water temperatureb (Celsius) Secchib (m) Chlorophyll ab (μg/L) Phytoplanktona (mg/m3) Cyanobacteriaa (mg/m3) Planktona Phyto + Zoo (mm3/L) Planktivorous fishc (kg) Walleye landingsb (kg × 1000) TP load from Red Riverb (tonnes/yr)

1969

1994 1999

2002

2003

2004

2006

NB SB NB SB NB

28 73 0.395 0.532 17.2

20 88 0.3 0.9 18

45 86 0.778 0.838 18

27 100 0.267 0.48 18.7

34 101 1.023 0.929 20.3

35 70 0.83 0.896 16.5

36 125 0.677 1.214 20.1

SB NB SB NB SB Whole lake Whole lake Whole lake

18.9 1.9 0.8 2.2 0.8 1637

21.5 2 ND 6 ND 4155

22 2 ND 16 ND 5299

21 1.8 0.7 3.5 5.9 2000b

21.5 2.1 0.65 12.5 9.3 7210

22 1.2 0.3 13.5 ND 6800b

22.7 1.3 0.76 34.5 ND 12000b

923

3650 4886

1000b 6406

2200b 12000b

5.9

11.5

ND

14.4

12.8

13.7

ND

ND

211.8

552.4 374.5

210.3

1950 2750

3100

3400

3700

4500

2661 5425

3082

2050

6800

7044

Whole ND lake Whole ~100 lake SB ND

8.7

Sources: aMcCullough et al. (2012); bEnvironment Canada and Manitoba Water Stewardship (2011); cLumb et al. (2012).

83

Structural stability of the core Lake Winnipeg zooplankton species community, 1969–2006 The composition of pelagic species of crustacean zooplankton reported in Lake Winnipeg (Electronic Supplementary Material (ESM) Table S1) showed little change over the period of study. Dominant species in both the North basin and South basin over the course of the study were: Cladocera: Daphnia retrocurva, D. mendotae, Bosmina longirostris, Diaphanosoma birgei; Cyclopoida: Acanthocyclops vernalis, Diacyclops thomasi; Calanoida: Leptodiaptomus ashlandi. All of these were included in the suite of “core” species designated by Patalas and Salki (1992) with lake-wide distributions based on the 1969 surveys. The other core species, though not dominants, but also with relatively unchanged lakewide distribution and densities were: Limnocalanus macrurus, Epischura lacustris, E. nevadensis (see Fig. 8a in Patalas and Salki, 1992, mistakenly presented as E. lacustris), E. oregonensis, and Leptodora kindti. However, there were taxa whose distribution over the years consistently showed overwhelming prevalence in particular basins: North basin—Eubosmina coregoni, Daphnia longiremis, Chydorus cf. sphaericus, Leptodiaptomus sicilis, L. minutus; South basin—L. siciloides and Tropocyclops prasinus mexicanus. E. coregoni, a cladoceran invader, expanded its range in the North basin. It has not yet successfully established in the South basin despite its likely original point of entry being via the Winnipeg River into the South basin. Species richness of pelagic taxa (criteria according to Dodson (1992)) (Table 2), and species diversity in Lake Winnipeg (1969– 2006) showed divergent patterns in the North and South Basins, despite similar changes in the number of stations sampled. Both Shannon's index and Brillouin's formulation (which accounts for different sample sizes) showed a trend of increasing diversity in the North basin, and a decrease in the South basin from 1969 to 2006 (Table 2). In contrast, Simpson's λ, which weights dominant species more heavily, showed relative stability in the North basin and an increase in the South basin. Thus, in the North basin, where zooplankton species numbers changed little, evenness among taxa increased. However, in the South basin, species numbers declined along with evenness, so that fewer dominant species characterized the assemblage. The ratios of Calanoida: Cyclopoida, Calanoida: (Cyclopoida + Cladocera) and large Cladocera: total Cladocera were consistently higher in the South basin than in the North basin (Table 2). No temporal trends were detected (via linear regression) in any of these indices. Species composition of zooplankton communities in Lake Winnipeg in summer (1969–2006) Calanoida Leptodiaptomus ashlandi was overwhelmingly the dominant calanoid throughout the lake, with Skistodiaptomus oregonensis a distant second in abundance (Fig. 2). The North basin calanoid community was more diverse than in the South basin and includes also: L. macrurus, L. minutus, L. sicilis, E. lacustris, and E. nevadensis. L. macrurus, L. sicilis, and L. minutus occurred rarely in the South basin where instead L. siciloides was found. Calanoids occurred in significantly higher densities in the South basin than in the North basin across years (F1,318 = 56.98, p b 0.0001) and comprised a higher proportion of the zooplankton community in the South basin (ESM Fig. S1). In each basin, there was significant variation in densities of L. ashlandi among years (Fig. 2; South basin: F7,157 = 6.84, p b 0.0001; North basin: F6,167 = 19.72, p b 0.0001), and densities in 1969 and 1994 were significantly lower than in all subsequent years. Cyclopoida In the North basin, both A. vernalis and D. thomasi were consistently dominant species (Fig. 3). There were significant inter-annual differences in density of all cyclopoids: A. vernalis (F6, 167 = 8.38, p b 0.0001), D. thomasi (F6, 167 = 12.65, p b 0.0001), and Mesocyclops

84


Table 2 Zooplankton species heterogeneity and structural indices in Lake Winnipeg (1969–2006). Basin

Index

1969

1994

North basin

# samples # species Shannon (H′’) Brillouin Simpson's λ Calanoida/Cyclopoida Large Cladocera/total Calanoid/Cycl + Clad # samples # species Shannon (H′’) Brillouin Simpson's λ Calanoida/Cyclopoida Large Cladocera/total Calanoid/Cycl + Clad

68 19 1.49 1.00 0.33 1.6 0.46 1.22 56 23 1.44 1.18 0.34 5.7 0.77 2.79

14 22 1.66 1.41 0.29 0.49 0.55 0.37 19 17 1.40 1.08 0.36 11 0.75 3.55

South basin

1999

2002

2003

2004

2006

18 16 1.24 1.14 0.41 1.4 0.97 0.99

31 21 1.62 1.37 0.33 4.2 0.62 2.01 15 18 0.99 0.74 0.57 10.8 0.87 4.09

13 22 1.71 1.51 0.28 1.1 0.6 0.68 11 19 1.08 0.83 0.52 6.4 0.94 3.58

15 19 1.91 1.64 0.21 1.5 0.63 0.76 11 17 0.98 0.88 0.52 2.8 0.85 1.96

12 19 1.59 1.40 0.31 0.9 0.58 0.53 9 14 0.87 0.68 0.60 9.6 0.9 4.19

10 18 1.82 1.57 0.22 1.7 0.9 0.69 6 14 1.02 0.87 0.51 17.8a 0.97 2.26

One outlier removed (value with outlier included = 65.8).

D. retrocurva, D. mendotae, and D. birgei have been the most abundant cladoceran taxa in the South basin since 1969. B. longirostris was the most common taxon among the smaller cladocerans in the South 0.5

25

North Basin

Density (#/L)

0.4

Cladocera Daphnia retrocurva, D. mendotae (the largest daphniids in Lake Winnipeg), D. longiremis, Diaphanosoma birgei, Chydorus cf. sphaericus, Eubosmina coregoni and Bosmina longirostris were the most abundant cladocerans in the North basin (Fig. 4), especially in the last decade. D. longiremis, species preferring deeper water, was found in the North basin almost exclusively. Cladocera and Cyclopoida together comprised a larger percentage of the zooplankton community than Calanoida in the North basin over time (ESM Fig. S1).

20

0.35 0.3

15

0.25 0.2

10

0.15 0.1

5

0.05 0

0 1969

1994

1999

M. edax 4.5

25

0.4 20 15

2.5 2

10

1.5 1

5

0.5 1994

1999

D. minutus

2002

S. oregonensis*

2003

2004

L. sicilis

Density (#/L)

3 30 2

20

1

10

0

0 1994

1998

1999 L. siciloides

2002

2003

20 15

0.15 0.1

10

0.05

5 0

2004

E. nevadensis

1994

1998

D. thomasi 6

60

40

S. oregonensis

25

0.2

70

4

35 30

1969

50

1969

40

South Basin

L. ashlandi*

5

A. vernalis*

0.25

2006

South Basin

2006

0

5

Density (#/L)*

1969

D. thomasi*

2004

0.3

0

0

6

Density (#/L)

3

2003

0.35

Density (#/L)

Density (#/L)

3.5

2002

0.45

North Basin Density (#/L)*

4

1999

2002

2003

M. edax

2004

2006

A. vernalis* 30

North Basin

25

4

20

3

15

2

10

1

5

0

0 1969

1994

1999

2002

2003

2004

2006

2006

D. retrocurva

D. mendotae

D. birgei

L. ashlandi*

E. coregoni

D. longiremis*

B. longirostris*

Fig. 2. Density of Calanoida across years in summer based on station averages. Species indicated with an asterisk (*) are plotted in units on the right Y-axis in both NB and SB figures. All other species are plotted in units on the left Y-axis.

Density (#/L)*

0.45

Density (#.L)*

edax (F6, 167 = 4.97, p b 0.0001); for all cyclopoids, 1969 density was noticeably lower than all other years. In the South basin, the community was overwhelmingly dominated by A. vernalis (Fig. 3) and significant differences existed across years without any consistent trend (One-way ANOVA F7, 157 = 5.05, p b 0.0001). Densities of D. thomasi declined over the last decade.

Density (#/L)*

a

1998

Fig. 3. Densities of Cyclopoida in summer based on station averages. Species indicated with an asterisk (*) are plotted in units on the right Y-axis in both NB and SB figures. All other species are plotted in units on the left Y-axis.

B. Hann, A. Salki / Journal of Great Lakes Research 43 (2017) 80–90 12

25

(a)

South Basin 8 15 6 10 4

Zooplankton Group Mean(Total Zoop) Mean(TotalCycl) Mean(TotalCal) Mean(TotalClad)

20

Density (#/L)*

Density (#/L)

10

85

5

2

0

0 1969 D. birgei

1994

1998

1999

B. longirostris

2002

2003

D. mendotae

2004

2006 D. retrocurva*

Fig. 4. Densities of Cladocera in summer based on station averages. Species indicated with an asterisk (*) are plotted in units on the right Y-axis in both NB and SB figures. All other species are plotted in units on the left Y-axis.

basin. E. coregoni, a non-indigenous cladoceran, was found at only 2 stations in 1994: one station in the South basin near the Winnipeg River mouth (the presumed entry point into Lake Winnipeg) and one station in the North basin near Warren Landing near the Nelson River the outlet of the lake. By 1999, E. coregoni was widespread in the North basin and remains widely distributed throughout the North basin, but has proven largely unsuccessful in establishing itself in the South basin.

(b)


Lake Winnipeg summer zooplankton community abundance 1969–2006 Whole lake Crustacean zooplankton lake mean density (Fig. 5) showed significant variation among all years in summer (ANOVA: F6 = 7.6, p b 0.0001); in particular, 1969 was significantly lower (Tukey's HSD, p b 0.05) than other years. Major taxonomic groups of zooplankton also displayed significant variation among years in summer (Calanoida: F6 = 6.55, p b 0.0001; Cyclopoida: F6 = 3.53, p b 0.002; Cladocera: F6 = 5.70, p b 0.0001). For each group, density was lowest in 1969 (Tukey's HSD, p b 0.05). Whole-lake zooplankton density varied significantly with surface water temperature (ANOVA: F16 = 17.9, p b 0.001) among all years in summer, but not with TP, or Secchi transparency. North and South basins There were significant differences among years within each basin for all taxonomic groups determined using one-way ANOVA. In the North basin (Fig. 6a), for Total Zooplankton (F6 = 17.11, p b 0.0001) as well as for Total Cladocera (F6 = 7.88, p b 0.0001), Total Cyclopoida (F6 = 9.39, p b 0.0001), and Total Calanoida (F6 = 26.43, p b 0.0001), densities


Fig. 5. Crustacean zooplankton density (mean ± SE) for the whole lake in summer. Summer densities for the major zooplankton groups (Calanoida, Cyclopoida, Cladocera) are also indicated. First red arrow on the left indicates the invasion of rainbow smelt in 1990; the second arrow shows the first record of Eubosmina coregoni in 1994. The third red arrow indicates the 1997 Red River basin flood.

Fig. 6. Zooplankton densities (mean number of individuals/L ± SE) in (a) North basin and (b) South basin.

were significantly lower in 1969 than in all subsequent years (Tukey's HSD, p b 0.05), with no differences among them. In the South basin (Fig. 6b) for Total Zooplankton (F7 = 7.80, p b 0.0001), Total Cladocera (F7 = 5.62, p b 0.0001), Total Cyclopoida (F7 = 8.51, p b 0.0001), and Total Calanoida (F7 = 2.73, p b 0.01), there were again significant differences among years, but 1998 and 2003 were significantly higher than all other years which did not differ among themselves (Tukey's HSD, p b 0.05). Thus, it appears that the zooplankton community in the North basin is now fluctuating around a new higher mean density, whereas the South basin has not responded similarly.

Measures of interannual and interbasin zooplankton community variation Amalgamation of patterns of variation in species-level community composition using multivariate approaches reinforced conclusions derived from data for individual groups, i.e. Calanoida, Cyclopoida, and Cladocera. In addition, zooplankton species were clearly identified that contributed to patterns of change over the years. There were significant differences among years in the North basin (ANOSIM, R = 0.344, p b 0.001). Pairwise comparisons indicated that the assemblage for 1969 was significantly demarcated from those of all subsequent years (ESM Table S2). Average Bray-Curtis similarities of the North basin communities, based on inter-station variation within each year, ranged from 62% to 67% (SIMPER, Supplementary Table 3), indicating considerable homogeneity in zooplankton community composition among stations within a basin in summer in each year. Major contributors to the inter-annual dissimilarity (SIMPER, ESM Table S3) were higher densities of L. ashlandi and D. thomasi after 1969. In the North basin, ordination plots of zooplankton community composition among stations within years showed clear-cut separation between 1969 and all other years (NM-MDS, Fig. 7a). Among years based on average community composition, 1969 was again significantly differentiated from all other years (Fig. 7b, SIMPROF test, p b 0.001). This result was replicated using cluster analysis (Fig. 7c).


Year

2D Stress: 0.16

(a)

1969 1994 1999 2002 2003 2004 2006

1.0

86

TP

1999

EDAX EUBOSMIN DIAPHANS ASHLANDI

MENDOTAE CyanBiom

CHYDORUS

OREGON*

SWTemp SICILIS LACUSTRS 2003 2006

LEPTODOR

THOMASI*

Secchi LIMNO VERNALIS 2004 1969

LONGIREM RETROCUR 2002

2D Stress: 0

1999 2006

-1.0

2002 2003

1994

Group average

(c)

70

Similarity

80

1994

2004

2006

2002

1999

2003

90

1969

1.0

Fig. 8. RDA for species (abbreviations in species list (ESM Table S1)) and environmental variables (TP, SWTemp, Secchi, cyanobacterial biomass) in the North basin of Lake Winnipeg based on mean summer densities across years.

2004

100

BOSMINA*

-1.0

1969

60

NEVADENS SICILOID 1994

Year pairwise tests

(b)

MINUTUS

Samples

Fig. 7. (a) NM-MDS ordination of zooplankton species composition based on Bray-Curtis similarities of square-root transformed densities at stations sampled 1969–2006 in the North basin of Lake Winnipeg. Ellipses fit by eye. Stress = 0.16. (b) NM-MDS ordination of zooplankton community structure based on densities averaged over all stations sampled per year (1969–2006) in the North basin. Groups are distinct at 63% (solid line) and 77% (dashed line) similarities, and solid lines identify groups that are significantly different (p = 0.001, SIMPROF test). (c) CLUSTER analysis (group average) using BrayCurtis resemblance on annual mean densities. Black lines identify groups that are significantly different (p = 0.001, SIMPROF test).

In RDA, explanatory variables accounted for 79.1% of cumulative variation in the first 2 axes with forward selection of 4 variables (TP, Secchi, SWTemp, cyanobacterial biomass) (Fig. 8). Several minor species (e.g. Limnocalanus macrurus, E. nevadensis, L. siciloides, and L. minutus), that comprised a larger proportion of the zooplankton community in 1969 and 1994, were strongly associated with those years. The triplot identified 2003 and 2006 as the warmest years with increased densities of S. oregonensis, A. vernalis, and Diaphanosoma birgei associated with this vector. In the South basin, there were significant differences among years (ANOSIM, R = 0.347, p b 0.001); however, no single year was clearly distinguished (ANOSIM, ESM Table 2). Zooplankton community

composition among stations within years in ordination plots showed considerable overlap (NM-MDS, Fig. 9a). Among years based on average community composition, 1998 and 2003 were grouped together, but they were not significantly separated from other years (Fig. 9b, SIMPROF test). This result was replicated using cluster analysis (Fig. 9c). Average similarities of the South basin communities within each year confirmed substantial homogeneity in community composition; similarities ranged from 66% (1998) to 73% (2002) with 2006 being only 55% (SIMPER, ESM Table S4). The major contributors to the similarity in these years were higher densities of L. ashlandi, A. vernalis, M. edax, D. mendotae and D. birgei (SIMPER, ESM Table S4). RDA on South basin data showed that explanatory variables accounted for 44.8% of cumulative variation in the first two axes (Fig. 10) with forward selection to include three variables (TP, SWTemperature, Secchi transparency). Zooplankton assemblages characteristic of the North basin and South basin were influenced by the same three primary environmental variables, but individual species were impacted differently in each basin. This merits further detailed examination with respect to their seasonal biology. Discussion The current study revealed that seven dominant zooplankton species (Leptodiaptomus ashlandi, Acanthocyclops vernalis, Diacyclops thomasi, Daphnia mendotae, Daphnia retrocurva, Bosmina longirostris, Diaphanosoma birgei) have been present throughout Lake Winnipeg for over almost 80 years, and this suite of species are similar to the dominant taxa in the Laurentian Great Lakes (Barbiero and Rockwell, 2008; Barbiero et al., 2009; Stewart et al., 2010) prior to invasion of Bythotrephes longimanus and dreissenid mussels. Together with five additional subdominant species (L. macrurus, E. oregonensis, E. lacustris, E. nevadensis, and L. kindti), they constitute the same suite of 12 core species proposed by Patalas (1981), (Patalas and Salki, 1992) with the addition of the invader, Eubosmina coregoni, which has gained a substantial presence in the North basin community. The continuing presence of as many as eight species of Calanoida (6 diaptomids) in Lake Winnipeg embodies a high species richness characteristic of the prairie eco-province (Pinel-Alloul et al., 2013). The overwhelming dominance throughout the lake of Leptodiaptomus ashlandi,

2D Stress: 0.16

(a)

Year 1969 1994 1998 1999 2002 2003 2004 2006

87

1.0


1969

Secchi THOMASI

LACUSTRIS

1998

OREGON SICILIS SICILOIDES NEVADENSIS RETROCURVA* BOSMINA 2002 PARVULA LIMNO LEPTODORA DIAPHANOSOMA VERNALIS*

2003 1994

2006

MEXICAN

TP

1999

EUBOSMINA

EDAX

MENDOTAE

2004

2D Stress: 0.08

(b)

ASHLANDI*

-1.0

1969

SWTemp

-1.0 2002 1994

1999

1.0

1998

Fig. 10. RDA for species (see species list (ESM Table S1)) and environmental variables (TP, SWTemp, Secchi) in the South basin of Lake Winnipeg across years.

2003

2004

2006

Group average

(c) 70 75

Similarity

80 85 90

2004

2002

1999

2006

1994

1969

1998

100

2003

95

Samples

Fig. 9. (a) MDS ordination of zooplankton species composition based on Bray-Curtis similarities of square-root transformed densities at stations sampled 1969–2006 in the South basin of Lake Winnipeg. Ellipse fit by eye. Stress = 0.16. (b) MDS ordination of zooplankton community structure based on densities averaged over all stations sampled per year (1969–2006) in the South basin where 1969 stations are for August survey only. Groups are distinct at 71% (dashed line) similarities, and are not significant (SIMPROF test). (c) CLUSTER analysis (group average) using Bray-Curtis resemblance on square-root transformed annual mean densities. No groups were significantly different (SIMPROF test).

in particular, with Skistodiaptomus oregonensis a distant second, characterizes a calanoid community similar to that found in the offshore waters of Lake Michigan (Torke, 1975) before the invasion of dreissenids. Species numbers have declined in both basins, but less in the North than have those in the South. Other measures of heterogeneity in the zooplankton community consistently showed relatively stable diversity in the North basin versus decreasing diversity in the South basin over time. This discrepancy may be a consequence of the South basin containing only 10% of the total lake volume, yet since the early 1990s it has been impacted by a variety of environmental stressors, e.g. warmer water temperatures, increasing nutrient loads, and sediment loading

from flooding events accompanied by agricultural chemicals, pharmaceuticals, and other contaminants from the Red River (Carlson et al., 2013). It is likely that zooplankton diversity in the South Basin will decline further as dreissenid populations become established and exert their tropho-dynamic influence (Higgins et al., 2014). Consequences for the much larger North Basin are unclear. Regional climate has been shown to be the main driver of variation in crustacean community structure (Pinel-Alloul et al., 2013). Cultural eutrophication of Lake Winnipeg has accelerated with the onset of climate-induced flooding and surface water warming, coupled with agricultural intensification (Schindler et al., 2012). The result has been elevated overall lake productivity to a higher baseline level as substantiated by higher lake-wide zooplankton densities in the last decade. Temperature has an overwhelming impact on zooplankton because they are poikilotherms, so all their metabolic processes, e.g. ingestion, respiration, reproduction, as well as population growth rates and life cycles are directly affected (Vadadi-Fülöp et al., 2012). Patalas (1972) predicted that the main factors controlling zooplankton abundance in lakes were temperature and food, interacting to increase population growth rate. Beginning in the 1990s, zooplankton densities were routinely higher in the North basin than in the South basin, correlated with higher summer chlorophyll a concentrations (Environment Canada and Manitoba Water Stewardship, 2011) and likely more available food resources, despite lower surface water temperatures compared to the South basin. In the South basin in 1998 and 2003, zooplankton densities were significantly elevated relative to long term mean densities. Unusual conditions developed in 1998 following the 1997 Red River Flood of the Century which delivered large quantities of sediment and organic detritus to the South basin (Stewart et al., 2000, 2003). Cyclopoids (A. vernalis, D. thomasi, M. edax) were the major constituents of the zooplankton increase in 1998, and these normally predatory species possibly utilized protozoan and bacterial food resources typically associated with decomposing detrital materials from flooded watershed sources. In 2003, a year of minimum Red River flows and above average temperatures, the South basin zooplankton species assemblage reflected littoral habitat conditions (Walseng et al., 2006; Mimouni et al., 2015) likely strongly influenced by the extensive Netley-Libau Marsh that stretches along the entire southern margin of the South basin. These responses in the zooplankton community provide strong evidence of how

88


zooplankton communities react to extreme climate related events, e.g. the influence of Red River discharges associated with either floods or droughts on the South Basin of the Lake Winnipeg ecosystem. These short-term changes could only be detected because of the short residence time and shallow nature of the South basin of Lake Winnipeg. Planktivory may also be contributing to inter-basin differences in Lake Winnipeg zooplankton species composition and abundance. Total biomass of all species of pelagic planktivorous fishes (2002–2008), predominantly emerald shiner, rainbow smelt, and cisco, was significantly higher in the South basin than in the North basin (Lumb et al., 2012). In the absence of comparative data prior to 2002 and before the invasion of smelt in the North basin, we suggest that the pattern remained consistent, i.e. predation pressure from planktivorous fishes has been higher in the South than in the North basin. Emerald shiner biomass was higher than any other species in the South Basin and smelt biomass was highest in the North basin (Lumb et al., 2012). All of these species feed heavily on large cladocerans (Olynyk et al. in review; Pothoven et al., 2009). Differential planktivory likely contributed to the consistently higher zooplankton density, as well as the larger contribution of Cladocera, in the North basin. In some eutrophic systems, zooplankton densities and diversity have declined through the effects of trophic cascades (McQueen et al., 1986, 1989). Hence, crustacean zooplankton trophic indices and indicator species (Ejsmont-Karabin and Karabin, 2013) have not been consistently useful in Lake Winnipeg, although rotifers (not studied in the lake), less impacted by fish predation, may prove more indicative of high trophy (Ejsmont-Karabin, 2012). Calanoida (especially Leptodiaptomus ashlandi) were more prevalent in the more eutrophic, turbid South basin, whereas Cyclopoida and Cladocera prevailed in the less eutrophic, more transparent North basin. Copepods have much more forceful escape jumps to avoid predators than cladocerans have, as well as superior abilities to remotely detect prey (Kiørboe, 2011). Both behavioural adaptations may work to the advantage and, ultimately, numerical dominance of calanoids (in the South basin) and cyclopoids (in the North basin) of Lake Winnipeg as predation pressure from planktivorous fish is an order of magnitude more intense in the South basin than in the North basin (Lumb et al., 2012). Thus, the relative success of calanoid copepods in the South basin of Lake Winnipeg likely results from the combination of a highly selective feeding mechanism, superior escape behaviour, and lower visibility to planktivorous fish in the more turbid waters. The increasing dominance of the herbivore Leptodiaptomus ashlandi throughout the lake, and the declining proportions of Leptodiaptomus minutus and Limnocalanus macrurus in the North basin over the last decade, parallel the growing predominance of cyanobacteria among the primary producers (Kling et al., 2011). Large calanoids, e.g. species of Heterocope, Epischura, and Limnocalanus were never abundant in Lake Winnipeg. Limnocalanus macrurus, a large calanoid (female body size 2.2–3.2 mm; Balcer et al. 1984), declined in parallel with cultural eutrophication and increased abundance of rainbow smelt, similar to the pattern observed in Lake Erie (Kane et al., 2004). Among Cyclopoida, Acanthocyclops vernalis and Diacyclops thomasi comprised the dominant taxa in both basins of Lake Winnipeg. In all of the Laurentian Great Lakes, D. thomasi is the most abundant cyclopoid copepod; only in the shallow western basin of Lake Erie is A. vernalis present in higher abundance than D. thomasi (Patalas, 1972). Prior to the invasion of dreissenids, both species were more common in the warmer, more eutrophic of the Laurentian Great Lakes, especially the western basin of Lake Erie, and rare in Lake Superior (Robertson and Gannon, 1981; Barbiero et al., 2001). The species of cyclopoid copepods in Lake Winnipeg represent a considerable range of adult body size: A. vernalis (0.8–1.8 mm) and M. edax (1.3–1.7 mm) are large body-sized cyclopoids, D. thomasi (1.0– 1.4 mm) is of medium size, and Tropocyclops prasinus mexicanus (0.5– 0.9 mm) is small (Balcer et al., 1984). The range of cyclopoid copepod body sizes present in Lake Winnipeg likely mirrors the spectrum of food ingested and their role in the pelagic food web, with increasing

body size paralleling increased proportional consumption of invertebrate prey in the diet. Detailed examination using stable isotope techniques could resolve food niches in even closely related copepod taxa (Santer et al., 2006). Cladocera are consistently more abundant in the North basin with its higher transparencies compared with the South basin. Schulze (2011) showed that even in turbid environments, fish predation can be a main driver in shaping the zooplankton community. The higher total biomass of planktivores in the South basin, predominantly emerald shiners found in significantly higher abundance in surface trawls (Lumb et al., 2012), undoubtedly feed heavily on cladocerans and may explain their relatively lower densities and lower proportional representation in the zooplankton community there compared to the North basin. However, all “large cladocerans” are not equal in terms of planktivore predatory behaviour, reaction distances, and prey choice (O'Brien et al., 1976), suggesting that such choices by the dominant emerald shiners in the South basin and Rainbow smelt in the North basin may contribute differentially to the varying densities of the large Daphnia spp. found in the two basins of Lake Winnipeg. Large cladocerans, e.g. Daphnia spp., have constituted the focal interest in grazing models; however, small cladocerans and copepods frequently coexist with cyanobacterial blooms and with intense predation pressure from planktivores. Chydorus cf. sphaericus and Eubosmina coregoni, found exclusively in the North basin, were more abundant in years with extensive cyanobacterial blooms. C. sphaericus is typically found in ponds and littoral zones of lakes, and its morphological adaptations and feeding limb structure are not those of a pelagic organism (Fryer, 1968); nevertheless, it has frequently been reported in the open water of lakes in association with cyanobacterial filaments (Birge, 1897; Scourfield, 1898, cited in Fryer, 1968). Most studies have observed little direct effect of these small cladocerans on cyanobacterial populations (Schoenberg, 1989; Christoffersen et al., 1990; Ventelä et al., 2002); however, impact has been demonstrated to vary seasonally and with different levels of fish predation (Urrutia-Cordero et al., 2015). Diaphanosoma birgei densities were also elevated in both basins in years with higher surface water temperatures and/or with cyanobacterial blooms, especially 2003 and 2006 (Environment Canada and Manitoba Water Stewardship, 2011). Arts et al. (1992) suggested that D. birgei as well as Chydorus may feed on bacterial populations that were elevated concurrently with cyanobacterial blooms in hypereutrophic Humboldt Lake. This further reinforces the need for detailed studies of the diet of major zooplankton species in order to trace energy flow in Lake Winnipeg. Concluding thoughts Application of molecular and stable isotope approaches to clarify diverse and frequently ontogenetically variable diet constituents, for copepods in particular (Santer et al., 2006; Craig et al., 2014; Doubek and Lehman, 2014; Hu et al., 2014), may contribute greatly to the resolution of details of food web structure in Lake Winnipeg. Spatial and seasonal variation in the phytoplankton community in the lake remains unknown. Similarly, the microzooplankton (ciliates, flagellates, rotifers) are unstudied. Unless the diets of the major taxa comprising the zooplankton community in Lake Winnipeg are more clearly defined, the response of the ecosystem to ongoing nutrient enrichment will remain speculative, and the network of species interactions elusive (Hulot et al., 2000). Many studies have found that zooplankton provide valuable indices whereby the health of the ecosystem may be efficiently and effectively evaluated (Gannon and Stemberger, 1978; Jeppesen et al., 2011; Kane et al., 2009; Haberman and Haldna, 2014; Munawar et al., 2012; Ejsmont-Karabin, 2012; Ejsmont-Karabin and Karabin, 2013). Whereas sampling for zooplankton is relatively inexpensive, identification to species level is expensive and time-consuming, and the pool of experts with sufficient taxonomic knowledge available to assess accurately the


community to species-level identification is rapidly shrinking. Although sufficiently detailed zooplankton data are available to employ the “battery of tests” approach suggested by Munawar et al. (2012) or the Planktonic Index of Biotic Integrity (P-IBI) developed by Kane et al. (2009) to assess the health of Lake Winnipeg ecosystem, much more comprehensive information is required for the phytoplankton community for this approach to be employed for Lake Winnipeg. Furthermore, study and understanding of the Lake Winnipeg ecosystem has been almost exclusively from the “bottom-up” point of view, i.e. nutrient-driven dynamics. We need a more balanced, inclusive, whole ecosystem approach that will ultimately contribute to formulation of regulatory policies and management decision-making strategies. Acknowledgments We thank the captain and crew of M.V. Namao and LWRC for assistance with collection of samples. The clarity of the manuscript was greatly improved by comments from Mike Paterson. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jglr.2016.10.015. References Adrian, R., et al., 2009. Lakes as sentinels of climate change. Limnol. Oceanogr. 54, 2283–2297. Alric, B., Jenny, J.-P., Berthon, V., Arnaud, F., Pignol, C., Reyss, J.-L., Sabatier, P., Perga, M.-E., 2013. Local forcings affect lake zooplankton vulnerability and response to climate warming. Ecology 94, 2767–2780. Arts, M.T., Evans, M.S., Robarts, R.D., 1992. Seasonal patterns of total and energy reserve lipids of dominant zooplanktonic crustaceans from a hyper-eutrophic lake. Oecologia 90, 560–571. Ayles, G.B., Campbell, K., Gillis, D., Saunders, L., Scott, K.J., Tallman, R., Traverse, N., 2011. Technical assessment of the status, health and sustainable harvest levels of the Lake Winnipeg fisheries resource. Lake Winnipeg Quota Review Task Force Report (196 pp.). Bajkov, A.D., 1934. The plankton of Lake Winnipeg drainage system. Int. Rev. Gesamten Hydrobiol. 31, 239–272. Balcer, M.D., Korda, N.L., Dodson, S.I., 1984. Zooplankton of the Great Lakes; a Guide to the Identification and Ecology of the Common Crustacean Species. The University of Wisconsin Press. Barbiero, R.P., Rockwell, D.C., 2008. Changes in the crustacean communities of the central basin of Lake Erie during the first full year of the Bythotrephes longimanus invasion. J. Great Lakes Res. 34, 109–121. Barbiero, R.P., Little, R.E., Tuchman, M.L., 2001. Results from the U.S. EPA's biological open water surveillance program of the Laurentian Great Lakes: III. Crustacean zooplankton. J. Great Lakes Res. 27 (2), 167–184. Barbiero, R.P., Bunnell, D.B., Rockwell, D.C., Tuchman, M.L., 2009. Recent increases in the large glacial-relict calanoid Limnocalanus macrurus in Lake Michigan. J. Great Lakes Res. 35, 285–292. Battarbee, R.W., 2000. Palaeolimnological approaches to climate change, with special regard to the biological record. Quat. Sci. Rev. 19, 107–124. Bednarska, A., Los, J., Dawidowicz, P., 2011. Temperature-dependent effect of filamentous cyanobacteria on Daphnia magna life history traits. J. Limnol. 70 (2), 353–358. Birge, E.A., 1897. Plankton studies on Lake Mendota. II. The Crustacea of the plankton. Trans. Wis. Acad. Sci. Arts Lett. 11, 274–448. Brunskill, G.J., Schindler, D.W., Elliott, S.E.M., Campbell, P., 1979. The attenuation of light in Lake Winnipeg. Canadian Fisheries and Marine Science Manuscript Report, 1522 (79 pp.). Brunskill, G.J., Elliott, S.E.M., Campbell, P., 1980. Morphometry, hydrology, and watershed data pertinent to the limnology of Lake Winnipeg. Canadian Fisheries and Marine Science Manuscript Report, 1556 (32 pp.). Bunting, L., Leavitt, P.R., Wissel, B., Laird, K.R., Cumming, B.F., St. Amand, A., Engstrom, D.R., 2011. Sudden ecosystem state change in Lake Winnipeg, Canada, caused by eutrophication arising from crop and livestock production during the 20th century. Report to the Manitoba Department of Water Stewardship (72 pp. http://www.gov.mb.ca/ waterstewardship/water_quality/lake_winnipeg/pdf/report_lake_wpg_ paleolimnology_2011.pdf). Carlson, J.C., Anderson, J.C., Low, J.E., Cardinal, P., MacKenzie, S.D., Beattie, S.A., Challis, J.K., Bennett, R.J., Meronek, S.S., Wilks, R.P.A., Buhay, W.M., Wong, C.S., Hanson, M.L., 2013. Presence and hazards of nutrients and emerging organic micropollutants from sewage lagoon discharges into Dead Horse Creek, Manitoba, Canada. Sci. Total Environ. 445–446. Caroni, R., Irvine, K., 2010. The potential of zooplankton communities for ecological assessment of lakes: redundant concept or political oversight? Biol. Environ. Proc. R. Ir. Acad. 110B, 35–53.

89

Christoffersen, K., Riemann, B., Hansen, L.R., Klysner, A., Sørensen, H.B., 1990. Qualitative importance of the microbial loop and plankton community structure in a eutrophic lake during a bloom of cyanobacteria. Microb. Ecol. 20, 253–272. Clarke, K.R., Gorley, R.N., 2006. PRIMER v7. PRIMER-E Ltd., Plymouth, UK. Clarke, K.R., Green, R.H., 1988. Statistical design and analysis for a ‘biological effects’ study. Mar. Ecol. Prog. Ser. 46, 213–226. Clarke, K.R., Warwick, R.M., 2001. Change in Marine Communities: An Approach to Statistical Analysis and Interpretation. second ed. PRIMER-E, Plymouth, UK. Craig, C., Kimmerer, W.J., Cohen, C.S., 2014. A DNA-based method for investigating feeding by copepod nauplii. J. Plankton Res. 36, 271–275. Cushing, C.E., 1998. Freshwater Ecosystems and Climate Change in North America: A Regional Assessment. Wiley (270 pp.). Davis, C.C., 1969. Seasonal distribution, constitution, and abundance of zooplankton in Lake Erie. J. Fish. Res. Board Can. 26, 2459–2476. Dodson, S., 1992. Predicting crustacean zooplankton species richness. Limnol. Oceanogr. 37, 848–856. Doubek, J.P., Lehman, J.T., 2014. Historical trophic position of Limnocalanus macrurus in Lake Michigan. J. Great Lakes Res. 40, 1027–1032. Ejsmont-Karabin, J., 2012. The usefulness of zooplankton as lake ecosystem indicators: rotifer trophic index. Pol. J. Ecol. 60, 339–360. Ejsmont-Karabin, J., Karabin, A., 2013. The suitability of zooplankton as lake ecosystem indicators: crustacean trophic state index. Pol. J. Ecol. 61, 561–573. Environment Canada & Manitoba Water Stewardship, 2011. State of Lake Winnipeg: 1999 to 2007 (209 pp.). Fahnenstiel, G.L., Krause, A.E., McCormick, M.J., Carrick, H.J., Schelske, C.L., 1998. The structure of the planktonic food-web in the St. Lawrence Great Lakes. J. Great Lakes Res. 24 (3), 531–554. Ferrao-Filho, A., Tessier, A.J., DeMott, W.R., 2007. Sensitivity of herbivorous zooplankton to phosphorus-deficient diets: testing stoichiometric theory and the growth rate hypothesis. Limnol. Oceanogr. 52 (1), 407–415. Franzin, W.G., Barton, B.A., Remnant, R.A., Wain, D.B., Pagel, S.J., 1994. Range extension, present and potential distributions, and possible effects of rainbow smelt in Hudson Bay drainage waters of northwestern Ontario, Manitoba, and Minnesota. N. Am. J. Fish Manag. 14, 65–76. Fryer, G., 1968. Evolution and adaptive radiation in the Chydoridae (Crustacea: Cladocera): a study in comparative functional morphology and ecology. Philos. Trans. R. Soc. Lond. B Biol. Sci. 254, 221–385. Gannon, J.E., Stemberger, R.S., 1978. Zooplankton (especially crustaceans and rotifers) as indicators of water quality. Trans. Am. Microsc. Soc. 97 (1), 16–35. Haberman, J., Haldna, M., 2014. Indices of zooplankton community as valuable tools in assessing the trophic state and water quality of eutrophic lakes: long term study of Lake Võrtsjärv. J. Limnol. 73, 263–273. Haney, J.F., 1987. Field studies on zooplankton-cyanobacteria interactions. N. Z. J. Mar. Freshw. Res. 21, 467–475. Hanson, J.M., Peters, R.H., 1984. Empirical prediction of crustacean zooplankton biomass and profundal macrobenthos biomass in lakes. Can. J. Fish. Aquat. Sci. 41 (3), 439–445. Higgins, S.N., Althouse, B., Devlin, S.P., Vadeboncoeur, Y., Vander Zanden, M.J., 2014. Potential for large-bodies zooplankton and dreissenids to alter the productivity and autotrophic structure of lakes. Ecology 95, 2257–2267. Hobson, K.A., Ofukany, A., Soto, D.X., Wassenaar, L.I., 2012. An isotopic baseline (δ13C, δ15N) for fishes of Lake Winnipeg: implications for investigating impacts of eutrophication and invasive species. J. Great Lakes Res. 38, 58–65. Hu, S., Gao, Z., Li, T., Carpenter, E.J., Liu, S., Lin, S., 2014. Detecting in situ copepod diet diversity using molecular technique: development of a copepod/symbiotic ciliate-excluding eukaryote-inclusive PCR protocol. Public Libr. Sci. One 9, e103528. Hulot, F.D., Lacroix, G., Lescher-Moutoué, F., Loreau, M., 2000. Functional diversity governs ecosystem response to nutrient enrichment. Nature 405, 340–344. Jeppesen, E., Noges, P., Davidson, T.A., Haberman, J., Noges, T., Blank, K., Lauridsen, T.L., Sondergaard, M., Sayer, C., Laugaste, R., Johansson, L.S., Bjerring, R., Amsinck, S.L., 2011. Zooplankton as indicators in lakes: a scientific-based plea for including zooplankton in the ecological quality assessment of lakes according to the European Water Framework Directive (WFD). Hydrobiologia 676, 279–297. Johannsson, O.E., Dermott, R., Graham, D.M., Dahl, J.A., Millard, E.S., Myles, D.D., LeBlanc, J., 2000. Benthic and pelagic secondary production in Lake Erie after the Invasion of Dreissena spp. with implications for fish production. J. Great Lakes Res. 26 (1), 31–54. Kane, D.D., Gannon, J.E., Culver, D.A., 2004. The status of Limnocalanus macrurus (Copepoda: Calanoida: Centropagidae) in Lake Erie. J. Great Lakes Res. 30 (1), 22–30. Kane, D.D., Gordon, S.I., Munawar, M., Charlton, M.N., Culver, D.A., 2009. The planktonic index of biotic integrity (P-IBI): an approach for assessing lake ecosystem health. Ecol. Indic. 9 (6), 1234–1247. Kiørboe, T., 2011. What makes pelagic copepods so successful? J. Plankton Res. 33, 677–685. Kling, H.J., Watson, S.B., McCullough, G.K., Stainton, M.P., 2011. Bloom development and phytoplankton succession in Lake Winnipeg: a comparison of historical records with recent data. J. Aquat. Ecosyst. Health Manag. 14 (2), 219–224. Lande, R., 1996. Statistics and partitioning of species diversity, and similarity among multiple communities. Oikos 76, 5–13. Leavitt, P.R., et al., 2009. Paleolimnological evidence of the effects on lakes of energy and mass transfer from climate and humans. Limnol. Oceanogr. 54, 2230–2248. Legendre, P., Legendre, L., 1998. Numerical Ecology. second ed. Elsevier, Amsterdam (853 pp.). Lumb, C.E., Franzin, W.G., Watkinson, D.A., 2012. Temporal and spatial patterns in pelagic trawl fish catches in Lake Winnipeg. J. Great Lakes Res. 38, 14–25. Madenjian, C.P., Fahnenstiel, G.L., Johengen, T.H., Nalepa, T.F., Vanderploeg, H.A., Fleischer, G.W., et al., 2002. Dynamics of the Lake Michigan food web, 1970–2000. Can. J. Fish. Aquat. Sci. 59, 736–753.

90


Makarewicz, J.C., 1993. A lakewide comparison of zooplankton biomass and its species composition in Lake Erie, 1983–87. J. Great Lakes Res. 19 (2), 275–290. Manitoba Conservation and Water Stewardship, Fisheries Branch, 2012n. A Profile of Manitoba's Commercial Fishery (14 pp.). McCauley, E., Kalff, J., 1981. Empirical relationships between phytoplankton and zooplankton biomass in lakes. Can. J. Fish. Aquat. Sci. 34 (4), 458–463. McCullough, G.K., Page, S.J., Hesslein, R.H., Stainton, M.P., Kling, H.J., Salki, A.G., Barber, D., 2012. Hydrological forcing of a recent trophic surge in Lake Winnipeg. J. Great Lakes Res. 38, 95–2012. McLaughlin, P.A., et al., 2005. Common and scientific names of aquatic invertebrates from the United States and Canada. 31. American Fisheries Society, Bethesda, Maryland, p. 545 Special Publication. McQueen, D.J., Post, J.R., Mills, E.L., 1986. Trophic relationships in freshwater pelagic ecosystems. Can. J. Fish. Aquat. Sci. 43, 1571–1581. McQueen, D.J., Johannes, M.R.S., Post, J.R., 1989. Bottom-up and top-down impacts on freshwater pelagic community structure. Ecol. Monogr. 59, 289–309. Miller, D.H., Kreis Jr., R.G., Huang, W., Xia, X., 2010. Application of a lower food web ecosystem productivity model for investigating dynamics of the invasive species Bythotrephes longimanus in Lake Michigan. Biol. Invasions 12, 3513–3524. Mimouni, E., Pinel-Alloul, B., Beisner, B.E., 2015. Assessing aquatic biodiversity of zooplankton communities in an urban landscape. Urban Ecosyst. 18, 1353–1372. Munawar, M., Fitzpatrick, M., Munawar, I.F., Niblock, H., Kane, D., 2012. Assessing ecosystem health impairments using a battery of ecological indicators: Bay of Quinte, Lake Ontario example. Aquat. Ecosyst. Health Manag. 15, 430–441. O'Brien, W.J., Slade, N.A., Vinyard, G.L., 1976. Apparent size as the determinant of prey selection of bluegill sunfish (Lepomis macrochirus). Ecology 57, 1304–1310. Ofukany, A.F.A., Hobson, K.A., Wassenaar, L.I., Bond, A.L., 2015. Prey consumption and trace element concentrations in double-crested cormorants (Phalacrocorax auritus) from Lake Winnipeg, Canada. J. Great Lakes Res. 41, 643–651. Patalas, K., 1972. Crustacean plankton and the eutrophication of St. Lawrence Great Lakes. J. Fish. Res. Board Can. 29, 1451–1462. Patalas, K., 1981. Spatial structure of the crustacean planktonic community in Lake Winnipeg, Canada. Verh. Int. Ver. Limnol. 21, 305–311. Patalas, K., 2006. Lake Winnipeg, remnant of glacial lake Agassiz, an efficient dispersal route of planktonic crustaceans through central and Northern Canada. Aquat. Ecosyst. Health Manag. 9, 285–306. Patalas, K., Salki, A., 1992. Crustacean plankton in Lake Winnipeg: variation in space and time as a function of lake morphology, geology and climate. Can. J. Fish. Aquat. Sci. 49, 1035–1059. Peet, R.K., 1974. The measurement of species diversity. Annu. Rev. Ecol. Syst. 5, 285–307. Pinel-Alloul, B., Adrien, A., Legendre, P., Cardille, J., Patalas, K., Salki, A., 2013. Large-scale geographic patterns of diversity and community structure of pelagic crustacean zooplankton in Canadian lakes. Glob. Ecol. Biogeogr. 22, 784–795. Pothoven, S.A., Vanderploeg, H.A., Ludsin, S.A., Höök, T.O., Brandt, S.B., 2009. Feeding ecology of emerald shiners and rainbow smelt in central Lake Erie. J. Great Lakes Res. 35, 190–198. Robertson, A., Gannon, J.E., 1981. Annotated checklist of the free-living copepods of the Great Lakes. J. Great Lakes Res. 7, 382–393. Rosenberg, D.M., Chambers, P.A., Culp, J.M., Franzin, W.G., Nelson, P.A., Salki, A.G., Stainton, M.P., Bodaly, R.A., Newbury, R.W., 2005. Nelson and Churchill River Basins. In: Benke, A.C., Cushing, C.E. (Eds.), Rivers of North America. Elsevier (1144 pp.). Salki, A.G., 1996. The crustacean plankton community of Lake Winnipeg in 1929, 1969 and 1994. In: Todd, B.J., et al. (Eds.), Geological Survey of Canada Open File 3113, pp. 319–344 (1996). Salki, A.G., Patalas, K., 1992. The crustacean plankton of Lake Winnipeg, June to October 1969. Canadian Data Report 868. Fisheries and Aquatic Sciences (22 pp.).

Santer, B., Sommerwerk, N., Grey, J., 2006. Food niches of cyclopoid copepods in eutrophic Plußsee determined by stable isotope analysis. Arch. Hydrobiol. 167, 301–316. Schindler, D.W., Hecky, R.E., McCullough, G.K., 2012. The rapid eutrophication of Lake Winnipeg: greening under global change. J. Great Lakes Res. 38, 6–13. Schoenberg, S.A., 1989. Effects of algal concentration, bacterial size and water chemistry on the ingestion of natural bacteria by cladocerans. J. Plankton Res. 11, 1273–1295. Schulze, P.C., 2011. Evidence that fish structure the zooplankton communities of turbid lakes and reservoirs. Freshw. Biol. 56, 352–365. Scourfield, D.J., 1898. A very common water-flea (Chydorus sphaericus). Illus. Ann. Microsc. 62–67. Shurin, J.B., Winder, M., Adrian, R., Keller, W., Matthews, B., Paterson, A.M., Paterson, M.J., Pinel-Alloul, B., Rusak, J.A., Yan, N.D., 2010. Environmental stability and lake zooplankton diversity—contrasting effects of chemical and thermal variability. Ecol. Lett. 13, 453–463. Stewart, K.W., Watkinson, D., 2004. The freshwater fishes of Manitoba. University of Manitoba Press, Winnipeg, MB (278 pp.). Stewart, A.R., Stern, G.A., Salki, A., Stainton, M.P., Lockhart, W.L., Billeck, B.N., Danell, R., Delaronde, J., Grift, N.P., Halldorson, T., Koczanski, K., MacHutcheon, A., Rosenberg, B., Savoie, D., Tenkula, D., Tomy, G., Yarchewski, A., 2000. Influence of the 1997 Red River flood on contaminant transport and fate in southern Lake Winnipeg. Department of Fisheries and Oceans Report to the International Red River Task Force of the International Joint Commission (190 pp.). Stewart, A.R., Stern, G.A., Lockhart, W.L., Kidd, K.A., Salki, A.G., Stainton, M.P., Koczanski, K., Rosenberg, G.B., Savoie, D.A., Billeck, B.N., Wilkinson, P., Muir, D.C.G., 2003. Assessing trends in organochlorine concentrations in Lake Winnipeg fish following the 1997 Red River flood. J. Great Lakes Res. 29, 332–354. Stewart, T.J., Johannsson, O.E., Holeck, K., Sprules, W.G., O'Gorman, R., 2010. The Lake Ontario zooplankton community before (1987–1991) and after (2001–2005) invasioninduced ecosystem change. J. Great Lakes Res. 36, 596–605. Suchy, K.D., Salki, A., Hann, B.J., 2010. Investigating the invasion of the nonindigenous zooplankter, Eubosmina coregoni, in Lake Winnipeg, Manitoba, Canada. J. Great Lakes Res. 36, 159–166. ter Braak, C.J.F., Šmilauer, P., 2012. Canoco reference manual and user's guide. Software for Ordination Version 5. Lake Winnipeg Project: cruise report and scientific results. In: Todd, B.J., Lewis, C.F.M., Thorleifson, L.H. (Eds.), Geological Survey of Canada Open File 3113 (656 pp.). Todd, B.J., Lewis, C.F.M., Thorleifson, L.H., Nielsen, E., Last, W.M., 1998. Paleolimnology of Lake Winnipeg—introduction and overview. J. Paleolimnol. 19, 211–213. Torke, B., 1975. The Population Dynamics and Life Histories of Crustacean Zooplankton at a Deep-water Station in Lake Michigan. (Ph.D. Thesis). University of Wisconsin (91 pp.). Urrutia-Cordero, P., Ekvall, M.K., Hansson, L.-A., 2015. Responses of cyanobacteria to herbivorous zooplankton across predator regimes: who mows the bloom? Freshw. Biol. 60, 960–972. Vadadi-Fülöp, C., Sipkay, C., Mészáros, G., Hufnagel, L., 2012. Climate change and freshwater zooplankton: what does it boil down to? Aquat. Ecol. 46, 501–519. Ventelä, A.-M., Wiackowski, K., Moilanen, M., Saarikari, V., Vuorio, K., Sarvala, J., 2002. The effect of small zooplankton on the microbial loop and edible algae during a cyanobacterial bloom. Freshw. Biol. 47, 1807–1819. Walseng, B., Hessen, D.O., Halvorssen, G., Schartau, A.K., 2006. Major contribution of littoral crustaceans to zooplankton species richness in lakes. Limnol. Oceanogr. 51, 2600–2606. Wassenaar, L.I., 2012. Dissolved oxygen status in Lake Winnipeg: spatio-temporal and isotopic (δ18O–O2) patterns. J. Great Lakes Res. 38, 123–134.