8
Investigating the invasion of the nonindigenous zooplankter, Eubosmina coregoni, in Lake Winnipeg, Manitoba, Canada Karyn D. Suchy a,1 , Alex Salki b , Brenda J. Hann a, a Department of Biological Sciences, University of Manitoba, Winnipeg, MB, Canada R3T 2N2 b Freshwater Institute, Fisheries and Oceans Canada, 501 University Cres, Winnipeg, MB, Canada R3T 2N6 abstract article info Article history: Received 18 June 2009 Accepted 1 December 2009 Communicated by Joseph C. Makarewicz Index words: Eubosmina coregoni Nonindigenous species NIS Invasion Lake Winnipeg The spread of nonindigenous species (NIS) over land and via interconnecting water bodies is threatening aquatic ecosystems worldwide. This study examines the invasion of the rst known NIS zooplankter, Eu- bosmina coregoni, into Lake Winnipeg, Manitoba, Canada. Analyses of cladoceran microfossils from a sediment core collected in the North Basin of the lake indicate this species rst appeared in sediments dated to the late 1980s. An increase in total cladoceran accumulation rates coupled with increasing N, C, P, and chlorophyll a over the last 40 years provides evidence of eutrophication. Extant samples from fall 2002-2005 indicate that E. coregoni is mainly restricted to the North Basin while Bosmina longirostris is present throughout the lake. Results from this study provide baseline data regarding the invasion and establishment of E. coregoni, a precursor to future NIS that may have substantial ecological and economic impacts on the Lake Winnipeg ecosystem. © 2009 Elsevier B.V. All rights reserved. Introduction The introduction of nonindigenous species (NIS) is an ongoing threat faced by aquatic ecosystems worldwide (Vitousek et al., 1997; Hall and Mills, 2000; Schindler, 2001). Over the last century, water bodies such as the Laurentian Great Lakes have received increasing numbers of NIS due to the discharge of ballast water (Mills et al., 1993; Ricciardi and MacIsaac, 2000). In contrast, lakes in Central Canada are more likely to receive NIS as a result of dispersal over land and via interconnecting river systems. The likelihood of NIS spreading to inland lakes has been modeled for the zebra mussel, Dreissena polymorpha (Buchan and Padilla, 1999; Allen and Ramcharan, 2001), and the spiny waterea, Bythotrephes longimanus (MacIsaac et al., 2004). Many of the present invaders of the Laurentian Great Lakes are potential invaders of Lake Winnipeg. As Lake Winnipeg is used extensively for commercial and recreational purposes, it is essential to learn more about aquatic NIS in this water body and the vulnerability of this ecosystem to future NIS that have already had major economic impacts and have permanently changed the food web dynamics of the Laurentian Great Lakes. Cladocerans (water eas) such as Bythotrephes longimanus and Cercopagis pengoi are among the many successful NIS that have spread to inland water bodies after establishing in the Laurentian Great Lakes (MacIsaac et al., 1999; Makarewicz et al., 2001; MacIsaac et al., 2004). Cladocera, a major component of the microcrustacean fauna in fresh- water lakes and ponds (Hann, 1989), possess a number of character- istics that enable them to be successful invaders. They mature rapidly and spend a large proportion of their lifetime as reproductive adults (Allan and Goulden, 1980), allowing them to establish quickly in aquatic ecosystems. Furthermore, most cladoceran species produce several generations of unfertilized eggs via parthenogenesis (Hann and Hebert, 1982) and diapausing, or resting, eggs via sexual reproduction during periods of environmental stress (Hann and Hebert, 1982; Hairston et al., 1995; Reid et al., 2000). As a result, both living individuals and diapausing eggs may be dispersed to new water bodies via transport by humans, animals, surface water (Frey, 1982; Shurin and Havel, 2002) or by wind (Cáceres and Soluk, 2002, Vanschoenwinkel et al., 2008). Eubosmina coregoni Baird 1857 is the rst known nonindigenous zooplankter to invade Lake Winnipeg. Common in water bodies of its native Eurasia (Lieder, 1991), E. coregoni rst colonized the Laurentian Great Lakes in North America in the mid-1960s, likely transported in the ballast water of transoceanic vessels (Deevey and Deevey, 1971; Lieder, 1991). By the late 1960s, E. coregoni had dispersed to inland lakes within 100 km of the Laurentian Great Lakes (Deevey and Deevey, 1971) but was thought to be restricted to this region, potentially due to habitat constraints (De Melo and Hebert, 1994a). Mabee (1988) reported a southern range extension in Missouri for E. coregoni, and this species was detected in sediments dated to the early 1990s from Lake of the Woods, Ontario (Suchy and Hann, 2007). Salki (1996) found low numbers of E. coregoni (as E. longispina) in the 1994 Journal of Great Lakes Research 36 (2010) 159166 Corresponding author. Tel.: +1 204 474 7450. E-mail addresses: [email protected] (K.D. Suchy), [email protected] (B.J. Hann). 1 Department of Biology, University of Victoria, PO Box 3020, Station CSC, Victoria, BC, Canada V8W 3N5. Tel.: 1-250-472-5098. 0380-1330/$ see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jglr.2009.12.004 Contents lists available at ScienceDirect Journal of Great Lakes Research journal homepage: www.elsevier.com/locate/jglr

Investigating the invasion of the nonindigenous zooplankter, Eubosmina coregoni, in Lake Winnipeg, Manitoba, Canada

Embed Size (px)

Citation preview

Page 1: Investigating the invasion of the nonindigenous zooplankter, Eubosmina coregoni, in Lake Winnipeg, Manitoba, Canada

Journal of Great Lakes Research 36 (2010) 159–166

Contents lists available at ScienceDirect

Journal of Great Lakes Research

j ourna l homepage: www.e lsev ie r.com/ locate / jg l r

Investigating the invasion of the nonindigenous zooplankter, Eubosmina coregoni, inLake Winnipeg, Manitoba, Canada

Karyn D. Suchy a,1, Alex Salki b, Brenda J. Hann a,⁎a Department of Biological Sciences, University of Manitoba, Winnipeg, MB, Canada R3T 2N2b Freshwater Institute, Fisheries and Oceans Canada, 501 University Cres, Winnipeg, MB, Canada R3T 2N6

⁎ Corresponding author. Tel.: +1 204 474 7450.E-mail addresses: [email protected] (K.D. Suchy), hann

1 Department of Biology, University of Victoria, PO BBC, Canada V8W 3N5. Tel.: 1-250-472-5098.

0380-1330/$ – see front matter © 2009 Elsevier B.V. Adoi:10.1016/j.jglr.2009.12.004

a b s t r a c t

a r t i c l e i n f o

Article history:Received 18 June 2009Accepted 1 December 2009

Communicated by Joseph C. Makarewicz

Index words:Eubosmina coregoniNonindigenous speciesNISInvasionLake Winnipeg

The spread of nonindigenous species (NIS) over land and via interconnecting water bodies is threateningaquatic ecosystems worldwide. This study examines the invasion of the first known NIS zooplankter, Eu-bosmina coregoni, into Lake Winnipeg, Manitoba, Canada. Analyses of cladoceran microfossils from asediment core collected in the North Basin of the lake indicate this species first appeared in sediments datedto the late 1980s. An increase in total cladoceran accumulation rates coupled with increasing N, C, P, andchlorophyll a over the last 40 years provides evidence of eutrophication. Extant samples from fall 2002-2005indicate that E. coregoni is mainly restricted to the North Basin while Bosmina longirostris is presentthroughout the lake. Results from this study provide baseline data regarding the invasion and establishmentof E. coregoni, a precursor to future NIS that may have substantial ecological and economic impacts on theLake Winnipeg ecosystem.

© 2009 Elsevier B.V. All rights reserved.

Introduction

The introduction of nonindigenous species (NIS) is an ongoingthreat faced by aquatic ecosystems worldwide (Vitousek et al., 1997;Hall and Mills, 2000; Schindler, 2001). Over the last century, waterbodies such as the Laurentian Great Lakes have received increasingnumbers of NIS due to the discharge of ballast water (Mills et al., 1993;Ricciardi and MacIsaac, 2000). In contrast, lakes in Central Canada aremore likely to receive NIS as a result of dispersal over land and viainterconnecting river systems. The likelihood of NIS spreading toinland lakes has been modeled for the zebra mussel, Dreissenapolymorpha (Buchan and Padilla, 1999; Allen and Ramcharan, 2001),and the spiny waterflea, Bythotrephes longimanus (MacIsaac et al.,2004). Many of the present invaders of the Laurentian Great Lakes arepotential invaders of Lake Winnipeg. As Lake Winnipeg is usedextensively for commercial and recreational purposes, it is essential tolearn more about aquatic NIS in this water body and the vulnerabilityof this ecosystem to future NIS that have already had major economicimpacts and have permanently changed the food web dynamics of theLaurentian Great Lakes.

Cladocerans (water fleas) such as Bythotrephes longimanus andCercopagis pengoi are among themany successful NIS that have spreadto inland water bodies after establishing in the Laurentian Great Lakes

@cc.umanitoba.ca (B.J. Hann).ox 3020, Station CSC, Victoria,

ll rights reserved.

(MacIsaac et al., 1999; Makarewicz et al., 2001; MacIsaac et al., 2004).Cladocera, a major component of the microcrustacean fauna in fresh-water lakes and ponds (Hann, 1989), possess a number of character-istics that enable them to be successful invaders. They mature rapidlyand spend a large proportion of their lifetime as reproductive adults(Allan and Goulden, 1980), allowing them to establish quickly inaquatic ecosystems. Furthermore, most cladoceran species produceseveral generations of unfertilized eggs via parthenogenesis (Hannand Hebert, 1982) and diapausing, or resting, eggs via sexualreproduction during periods of environmental stress (Hann andHebert, 1982; Hairston et al., 1995; Reid et al., 2000). As a result,both living individuals and diapausing eggs may be dispersed to newwater bodies via transport by humans, animals, surface water (Frey,1982; Shurin and Havel, 2002) or by wind (Cáceres and Soluk, 2002,Vanschoenwinkel et al., 2008).

Eubosmina coregoni Baird 1857 is the first known nonindigenouszooplankter to invade Lake Winnipeg. Common in water bodies of itsnative Eurasia (Lieder, 1991), E. coregoni first colonized the LaurentianGreat Lakes in North America in the mid-1960s, likely transported inthe ballast water of transoceanic vessels (Deevey and Deevey, 1971;Lieder, 1991). By the late 1960s, E. coregoni had dispersed to inlandlakes within 100 km of the Laurentian Great Lakes (Deevey andDeevey, 1971) but was thought to be restricted to this region,potentially due to habitat constraints (De Melo and Hebert, 1994a).Mabee (1988) reported a southern range extension in Missouri for E.coregoni, and this species was detected in sediments dated to the early1990s from Lake of the Woods, Ontario (Suchy and Hann, 2007). Salki(1996) found low numbers of E. coregoni (as E. longispina) in the 1994

Page 2: Investigating the invasion of the nonindigenous zooplankter, Eubosmina coregoni, in Lake Winnipeg, Manitoba, Canada

160 K.D. Suchy et al. / Journal of Great Lakes Research 36 (2010) 159–166

zooplankton community of Lake Winnipeg at two of 30 samplingsites: the Traverse Bay region (Winnipeg River inflow) in the SouthBasin and at the Nelson River outflow in the North Basin. Previouscomprehensive zooplankton sampling throughout Lake Winnipeg in1929 (Bajkov, 1934) and 1969 (Patalas and Salki, 1992) failed touncover E. coregoni. From 1969 to 1994, zooplankton sampling wasnot undertaken on Lake Winnipeg, and it is conceivable that E.coregoni may have invaded the lake during this surveillance gap.

The present study uses microfossil remains from a Lake Winnipegsediment core to reconstruct the invasion history of E. coregoni in

Fig. 1. Sampling stations for LakeWinnipeg, 2002–2005. Triangles indicate extant zooplanktoInset map shows location of Lake Winnipeg (star) within the province of Manitoba, Canada

order to address the gap in the zooplankton record. This paleolimno-logical approach provides a historical account of the arrival of E.coregoni and its subsequent development in the region. To examinethe current spatial distribution of E. coregoni, zooplankton sampleswere collected by the Lake Winnipeg Research Consortium Inc. fromlakewide sites during the 2002 to 2005 period. In addition, dataconcerning limnological conditions and other biological factors werecollected at all sampling stations to examine whether a relationshipexists between E. coregoni and these parameters. Ultimately, resultsfrom this study will provide a framework for future studies

n sampling stations, circle indicates the location of the sediment core collected in 2003..

Page 3: Investigating the invasion of the nonindigenous zooplankter, Eubosmina coregoni, in Lake Winnipeg, Manitoba, Canada

161K.D. Suchy et al. / Journal of Great Lakes Research 36 (2010) 159–166

concerning the invasion of E. coregoni and other potential NIS that aredispersing to lakes in Central Canada.

Materials and methods

Study site

Lake Winnipeg (50°0'–53°50' N, 96°15′–99°15′ W), Manitoba,Canada, is the 10th largest freshwater lake in the world with a surfacearea of 24,500 km2 (Fig. 1) (LakeWinnipeg Stewardship Board, 2006).Lake Winnipeg has a large watershed of 953,250 km2 spanning fourCanadian provinces and four US states (Lake Winnipeg StewardshipBoard, 2006). The lake is divided into distinct South and North basinsthat are separated by a region only a few kilometers wide called ‘TheNarrows’ (Todd et al., 1997). LakeWinnipeg rests on a unique geologicboundary with its eastern half underlain by Precambrian rock and itswestern side by Palaeozoic limestone (Brunskill et al., 1980; Todd etal., 1997). As a result, water inflows from southern, western, andnorthwestern drainages have higher alkalinity, mineral and nutrientcontent than eastern river inflows (Brunskill et al., 1980). LakeWinnipeg receives water from the south and west via the Red andSaskatchewan Rivers, respectively, and from the east mainly by theWinnipeg, Pigeon, Berens, and Poplar Rivers (Patalas and Salki, 1992).The only outflow of LakeWinnipeg, the Nelson River, flows northwardfrom the North Basin to Hudson Bay (Brunskill et al., 1980). Maximumdepth of the lake is approximately 36 m, while the mean depths of theNorth and South Basins are 13.3 m and 9.7 m, respectively (Patalasand Salki, 1992). Residence time of the water in LakeWinnipeg is 2.9–4.3 years (Brunskill et al., 1980).

Sediment analysis

The sediment core used in this study was collected in August 2003in the North Basin of Lake Winnipeg (Fig. 1) using a Kajak-Brinkhurstgravity corer with an internal diameter of 10 cm. The total length ofthe core was approximately 80 cm. The top 10 cm of the core wassectioned at 0.5-cm intervals while the remainder of the core wassectioned at 1-cm intervals. Slices were stored in individual WhirlPacks™ and placed in plastic boxes at 4–8 °C prior to radiometric

Fig. 2. Sediment core profile from the North Basin of Lake Winnipeg showing excess 210PbConstant Rate of Supply (CRS) model is shown on the right axis.

210Pb dating at the Freshwater Institute, Winnipeg, Manitoba. Total210Pb activity showed a general decrease with depth except between15 and 25 cm where a flattening out of the 210Pb profile occurred,indicating a significant disturbance to the sediments (e.g., slumping,sediment mixing, or sediment focusing). Despite this irregularity,plotted on a logarithmic scale with depth, 210Pb expressed ascumulative dry mass (g/cm2) decreased almost linearly (Fig. 2).Using the slope of this profile, sedimentation rates and chronologieswere calculated based on the Constant Rate of Supply (CRS) model(Appleby and Oldfield, 1983). 210Pb dates are supported by themeasurement of a 137Cs peak in sediments dated to 1964, a result ofglobal fallout, and all sedimentation rates were corrected for sedimentfocusing (Oldfield and Appleby, 1984). Sediment core chemistry (C, N,P, and chlorophyll a) was determined on the same core slices usingstandard methods in the water chemistry laboratory at the Freshwa-ter Institute, Winnipeg, MB (Stainton et al., 1974).

Sediment material for analysis of cladoceran remains wasprepared following methods outlined in Frey (1960). The 12uppermost sediment layers of the core (corresponding to a depth of6 cm and spanning 1983–2003 as determined by 210Pb dating) wereanalyzed, followed by every fourth layer until the bottommost layerfor which a 210Pb date was available (1860). Subsamples of a knowndry weight were taken from the sediment layers, heated gently in 10%KOH for approximately 30min on amagnetic stirring plate, and sievedusing a 53-μm mesh screen to concentrate the remains. Subsampleswere made up to a known volume (5 ml) and a few drops of formalinwere added. Quantitative slides were prepared by transferring a 100-μL aliquot from a well-mixed subsample to a microscope slide, usingglycerin jelly as the mounting medium. Identification and countingwas completed at 100×magnification using a compound microscopewith phase contrast optics. Entire coverslips were enumerated forcarapaces as they were the most abundant exoskeletal fragments(Hann, 1989). Remains were identified following Deevey and Deevey(1971), De Melo and Hebert (1994b), and Lieder (1983), and by usingextant reference samples from Lake Winnipeg. Three groups wereenumerated: Bosmina longirostris, E. coregoni, and “other” cladocerans(typically chydorids). Bosmina longirostris and E. coregoni weredistinguished on the basis of carapace morphology and the presence(B. longirostris) or absence (E. coregoni) of a mucrone. Except for a few

and 137Cs activity (Bq/g) versus accumulated dry weight (g/cm2). The time-depth or

Page 4: Investigating the invasion of the nonindigenous zooplankter, Eubosmina coregoni, in Lake Winnipeg, Manitoba, Canada

162 K.D. Suchy et al. / Journal of Great Lakes Research 36 (2010) 159–166

samples near the bottom of the core that had very few remains, aminimum of 100 intact carapaces was counted for each sedimentlayer, which was determined to be sufficient in detecting cladoceranremains in this region (see Suchy and Hann, 2007). This conservativeapproach of counting only intact carapaces was taken to ensurecorrect identification of remains, as opposed to also enumeratingfragments of carapaces and other exoskeletal remains, which can leadto incorrect or biased identifications (Hann, 1989). Data are presentedas accumulation rates (number of remains/cm2/year). Although notpresented, concentrations (individuals/g dry sediment) of cladoceranmicrofossils varied in parallel with accumulation rates. Statistical testswere performed using SYSTAT Version 12 (SYSTAT Software, Inc., USA2007).

Extant zooplankton

Zooplankton sampling was carried out onboard M.V. NAMAO inSeptember and October 2002–2005 as part of a comprehensivelimnological study of 60 LakeWinnipeg sites (Fig. 1). At each samplingstation during daylight hours, a Wisconsin net with a 25-cm mouthdiameter, 72-μm mesh size, and a filtering cone length of 1 m washauled vertically from just above (0.25 m) the lake bottom to thesurface. Zooplankton samples were preserved with 8% formalin.Approximately 10% of each sample was examined and Bosminalongirostris, E. coregoni, and “other” cladocerans were identified andcounted. Densities (individuals/L) of Cladocera at each station in thefour fall surveys were estimated and mapped using ArcGIS 9.0 (ESRI,Redlands, CA) to illustrate the distribution of cladocerans throughoutthe lake. Other limnological data for each site, e.g., chlorophyll a,settled volume (total phyto- and zooplankton, see Patalas and Salki,1992) were collected simultaneously with extant zooplanktonsamples. Mantel tests based on Pearson's product-moment correla-

Fig. 3. Accumulation rates (number of remains/cm2/year) for Bosmina longirostris (a), EubosLake Winnipeg, Manitoba.

tions were performed using R statistical software (R DevelopmentCore Team, 2008) in order to measure the association between thedensities of E. coregoni and these limnological data (Manly, 1997).

Results

Sediment cores

In general, cladoceran remains were abundant and well preserved.Few remains were present in sediment layers dated from 1860 to1924, insufficient to satisfy the minimum count of 100 individuals foranalysis (Fig. 3). The cladoceran species assemblage was dominatedby B. longirostris throughout all sediment layers analyzed (Fig. 3). E.coregoniwas first detected in sediments dated to 1988. Between 210Pbdates 1988 and 1995, accumulation rates of E. coregoniwere very low,i.e., fewer than 15 individuals/cm2/year (Fig. 3). In sediments datedpost-1997, the E. coregoni accumulation rate increased substantiallywith a peak of approximately 480 individuals/cm2/year in sedimentsdated to 2003. Although E. coregoni continues to be present insediments deposited recently, the accumulation rates of this speciesare much lower in comparison to B. longirostris accumulation rates.Accumulation rates of B. longirostris peaked at 210Pb dates 1986 to1988 and 1993 to 1999, with the highest accumulation rate ofapproximately 1700 individuals/cm2/year in sediments dated to1993. Phosphorus and chlorophyll a accumulation rates remainedrelatively constant throughout all sediment layers with the exceptionof the most recently deposited sediments dated to 2002 and 2003(Fig. 4). Results from these layers should be interpreted with cautiondue to the mobility of P and the process of chlorophyll a diagenesis inthe surface sediments. Nitrogen and carbon, on the other hand,showed a slight increase in accumulation since sediments dated to1960 (Fig. 4). Correlations between sediment core chemistry and

mina coregoni (b), and “other” cladocerans (c) in the 2003 core from the North Basin of

Page 5: Investigating the invasion of the nonindigenous zooplankter, Eubosmina coregoni, in Lake Winnipeg, Manitoba, Canada

Fig. 4. (a) Total cladoceran accumulation rates (number of remains/cm2/year) and (b) sediment core chemistry accumulation rates (μg/cm2/year) for N, P, and C (bottom axis);chlorophyll a (top axis) in the 2003 core from the North Basin of Lake Winnipeg, Manitoba.

163K.D. Suchy et al. / Journal of Great Lakes Research 36 (2010) 159–166

cladoceran accumulation rates showed that E. coregoni was positivelycorrelated with N, P, C, and chlorophyll a, and these correlations weresignificantly different from zero (Table 1). Correlations between bothB. longirostris and total cladoceran accumulation rates, on the otherhand, had a positive correlation with only N and C; however, none ofthese correlations were significantly different from zero (Table 1).

Extant zooplankton

In all four study years, E. coregoni was found predominantly in theNorth Basin of Lake Winnipeg (Table 2, Fig. 5). Average densities of E.coregoni were similar in Fall 2002 and 2003 at 3.85 and 4.27individuals/L, respectively. In 2004, the average density of E. coregonidecreased substantially in the North Basin (0.81 individuals/L),whereas in 2005, E. coregoni densities reestablished to an average of2.44 individuals/L. Densities of E. coregoni in the South Basin, on the

Table 1Pearson's correlation coefficient (r) for cladoceran accumulation rates of Eubosminacoregoni, Bosmina longirostris and total cladocerans (number of remains/cm2/year) andsediment core chemistry for N, C, P, and chlorophyll a (μg/cm2/year) from a sedimentcore collected from the North Basin of Lake Winnipeg, Manitoba, Canada. P-values arepresented to test whether r was significantly different from zero (α=0.05). Note:sediment layers below 15 cm have been excluded from analysis due to a lack of remains.

B. longirostris E. coregoni Total cladocerans

r P-value r P-value r P-value

N 0.223 0.425 0.700 0.004⁎ 0.381 0.161C 0.346 0.206 0.597 0.019⁎ 0.497 0.060P −0.088 0.756 0.592 0.020⁎ 0.016 0.954Chlorophyll a 0.074 0.792 0.571 0.026⁎ 0.167 0.551

⁎ Indicates significant results.

other hand, were low each year with the highest average densityoccurring in 2002 at 0.89 individuals/L (Fig. 5).

Similarly, the highest B. longirostris densities were found atstations in the North Basin in all 4 years; however, this species wasconsistently present at most stations throughout the lake (Table 2,Fig. 5). As with E. coregoni, average densities of B. longirostris in theNorth Basin were highest in 2002 and 2003 at 3.58 and 6.98 indi-viduals/L, respectively. After 2003, B. longirostris densities declined inthis region, yielding an average density of 2.63 individuals/L in 2004and 1.51 individuals/L in 2005. In the South Basin, the average densityof B. longirostris reached a maximum in 2003 (1.88 individuals/L) anda minimum in 2004 (0.26 individuals/L).

In 2002 and 2004, densities of E. coregoni showed positive spatialautocorrelation indicating there was a significant correlation be-tween the densities of E. coregoni in reference to its spatial location inthe lake (Table 3). There was no significant correlation betweenE. coregoni and either chlorophyll a or station depth in any of thesampling. In terms of settled volume (total phytoplankton and

Table 2Fraction (and %) of stations containing Eubosmina coregoni and Bosmina longirostris inthe North and South basins from Fall extant samples 2002–2005 in Lake Winnipeg,Manitoba, Canada.

Year

2002 2003 2004 2005

E. coregoniNorth 24/29 (83%) 23/30 (77%) 21/27 (78%) 22/27 (81%)South 4/24 (17%) 8/29 (28%) 2/25 (8%) 6/25 (24%)

B. longirostrisNorth 29/29 (100%) 30/30 (100%) 27/27 (100%) 26/27 (96%)South 23/24 (96%) 26/29 (90%) 24/25 (96%) 25/25 (100%)

Page 6: Investigating the invasion of the nonindigenous zooplankter, Eubosmina coregoni, in Lake Winnipeg, Manitoba, Canada

Fig. 5. Fall 2002–2005 density patterns of Eubosmina coregoni and Bosmina longirostris from extant samples from LakeWinnipeg, Manitoba, Canada. Darker shading indicates higherdensities of E. coregoni and B. longirostris.

164 K.D. Suchy et al. / Journal of Great Lakes Research 36 (2010) 159–166

zooplankton volume), there was a significant correlation between E.coregoni and settled volume only in 2003, while in 2004 and 2005there was a trend towards a significant correlation between thesetwo variables with significance values near α=0.05.

Discussion

Key historical information provided by the sediment core indicatesE. coregoni first appeared in the North Basin of Lake Winnipeg in thelate 1980s. It is possible, however, that this species was present inLake Winnipeg prior to this time as cladoceran populations may besmall and/or isolated when first dispersed into a new habitat, therebyremaining undetected (MacIsaac et al., 2001; Havel and Medley,2006). Factors that may hinder the detection of a rare species includeits conspicuousness, spatial arrangement in the water column, andbehaviour (Harvey et al., 2009). Harvey et al. (2009) determined thatthe probability of detecting Cercopagis pengois was predominantlyinfluenced by population density and the location of sampling sites.Further, in a study examining the invasion of Bythotrephes cederstroe-mii (=longimanus), first reported in the Laurentian Great Lakes by Buret al. (1986), Sprules et al. (1990) calculated the number of years itwould take a seed population of B. cederstroemii to reach the densitiesdetected in 1986 and estimated that the invasion likely took placeduring the late 1970s or early 1980s. While the possibility of E.coregoni being present in the lake earlier than the late 1980s shouldnot be ruled out, this species was similarly detected in sedimentsdated to the early 1990s in Lake of the Woods, a water body that ishydrologically connected to Lake Winnipeg via the Winnipeg River(Suchy and Hann, 2007). These parallel results confirm that thisspecies likely arrived in the Lake Winnipeg watershed within the last20 years.

Table 3Mantel statistic (r) and significance results for Mantel's tests performed on the density (indproduct-moment correlation, 1000 permutations (α=0.05).

2002 2003

n r p-value n r

E. coregoni and UTM Coordinates 53 0.17 b0.001⁎ 59E. coregoni and hlorophyll chlorophyll a 53 −0.04 0.65 59E. coregoni and settled volume 36 0.07 0.16 57E. coregoni and station depth 52 −0.002 0.42 49 −

⁎ Indicates significant result.

Once dispersed into a new water body, establishment of a NIS willoccur only if conditions are suitable for that particular species (Kolarand Lodge, 2001). Even though the reproductive strategies employedby Cladocera allow for rapid establishment in new water bodies, lowabundances in the sediments following E. coregoni's arrival occurreduntil the late 1990s when accumulation rates of E. coregoni began torepresent a substantial proportion (approximately 25%) of the totalcladoceran microfossil population. Although E. coregoni accumulationrates remained lower than those of B. longirostris in the sediments, by1999 E. coregoni was the dominant cladoceran in extant samplescollected from the North Basin (Salki unpublished). Results from theextant samples used in this study showed that the average densities ofE. coregoni were higher than B. longirostris in 2002 and 2005.Therefore, over the last 10 years, conditions have allowed E. coregonito periodically dominate the crustacean plankton community in theNorth Basin of Lake Winnipeg.

Given the complexity of aquatic ecosystems, the specific para-meters that govern the establishment of E. coregoni may never bedeciphered. One of the factors that may have contributed to thesuccessful establishment of E. coregoni in Lake Winnipeg relates totrophic status, which has been shown to influence the success ofinvading zooplankters in other systems (Lennon et al., 2003). Shifts inbosminid species have long been associated with changing trophiclevels in studies of lake sediments (Hofmann, 1978; Boucherle andZüllig, 1983; Szeroczynska, 2002; Gasiorowski and Szeroczynska,2004). Due to the taxonomic inconsistency associated with bosminidsin North America, however, the literature concerning a link betweenshifts in bosminid species and changes in the trophic state of a lakemust be interpreted with caution.

Nevertheless, changes in the accumulation rates of cladoceranmicrofossils in Lake Winnipeg are apparent in sediments from the

ividuals/L) of Eubosmina coregoni with other limnological variables based on Pearson's

2004 2005

p-value n r p-value n r p-value

0.02 0.29 52 0.16 0.001⁎ 52 0.06 0.090.11 0.12 45 0.19 0.09 N/A N/A N/A0.65 b0.001⁎ 52 0.16 0.06 48 0.17 0.060.05 0.66 35 −0.01 0.5 52 −0.01 0.5

Page 7: Investigating the invasion of the nonindigenous zooplankter, Eubosmina coregoni, in Lake Winnipeg, Manitoba, Canada

165K.D. Suchy et al. / Journal of Great Lakes Research 36 (2010) 159–166

North Basin with total cladoceran rates increasing dramatically overthe last 25 years. Although the virtual absence of remains prior to1920 must be interpreted with caution due to the possibility ofsediment mixing or slumping between 15 and 25 cm, higher accumu-lation rates of total cladocerans from sediments dated to approxi-mately 1960 to the most recent sediments could be indicative of asystem that is becoming more eutrophic. Further evidence to supportthis comes from the increase in nutrients in the top 10 cm of the coreand the positive correlations observed between cladoceran accumu-lation rates, particularly in terms of E. coregoni, and sediment corechemistry. Similarly, analysis of algal microfossils in a sediment coretaken in 1994 in the North Basin showed that shifts in diatom speciesfrom oligo-meso trophic taxa to more eutrophic taxa occurred at thetop of the core, thereby indicating an increase in anthropogeniceutrophication in this region of the lake (Kling, 1998). The increasedeconomic activity related to agricultural crop production, intensivelivestock operations, wetland removal, drainage network expansion,and urbanization, particularly in the Red River sub-basin of the LakeWinnipeg watershed, has contributed to rising nutrient loads to LakeWinnipeg (Bourne et al., 2003).

In comparison to the 150-year time span of the sediment core, theextant samples represent only a short time period and, as a result, areless likely to show strong relationships between variables, especially ifa system is already in a eutrophic state (as evidenced by the sedimentcore chemistry). Even though algal biomass, inferred by chlorophyll aconcentrations, can subsequently be used as an indicator of the trophicstatus of a lake, no significant correlations were found betweenchlorophyll a and the densities of E. coregoni in the extant samplesfrom 2002 to 2005. Although a pattern is suggested by the extantdensity figures, it is important to recognize that understanding thepatchiness of zooplankton distribution requires much more detailedstudy. Interannual variation in E. coregoni densitymay bemore relatedto the different weather patterns observed during each of the studyyears as parameters such as temperature, rainfall, and amount ofavailable sunlight all have an effect on the primary producers of thelake and thus indirectly affect the zooplankton community.

Regardless, it is crucial to assess the types of phytoplankton presentin a given system. Analysis of the composition of the phytoplanktoncommunity in Lake Winnipeg is currently underway (Hedy Kling,personal communication) and will provide insight relevant to theshifts observed in the cladoceran community. Another small cladoc-eran species, Chydorus sphaericus, has been found to dominate thezooplankton community when nutrient enrichment increased zoo-plankton biomass (Szeroczynska, 2002; Lennon et al., 2003). Similarly,Vijverberg and Boersma (1997) found that C. sphaericus and othersmall-bodied species decreased in biomass with increasing chloro-phyll a concentration. Small cladoceran species such as C. sphaericusand E. coregoni have a narrow carapace gap and are less likely to beinhibited by the presence of large filamentous algae (Vijverberg andBoersma, 1997). Chydorus sphaericus has, in fact, been increasing inextant samples from2002 to 2005 in theNorthBasin of LakeWinnipeg,thereby succeeding in eutrophic conditionswhere dense cyanobacter-ial (blue-green algal) blooms occur (Salki, unpublished data).

Future studies should explore the association between E. coregoniand other trophic levels. A number of studies have examined therelationship between large-bodied and small-bodied species in res-ponse to predation (Kerfoot, 1981; Sanford, 1993; Sarmaja-Korjonen,2002) and in response to habitat changes (Kerfoot, 1974; Grigorovichet al., 1998). Consequently, changes in predation pressuremay have aninfluence on the successful establishment of NIS. As a result, futurestudies should investigate the response of E. coregoni and othercladocerans to planktivory by fish by examining changes in morpho-logical features within a given species. Similarly, the appearance of E.coregoni in Lake Winnipeg coincides with the first appearance ofrainbow smelt (Osmerus mordax) in 1990 (Franzin et al., 1994),although it is possible that smelt appeared as early as 1975 (Campbell

et al., 1991). An examination of smelt gut contents could indicatewhether or not predation by smelt is a vector for transporting E.coregoni into Lake Winnipeg. Moreover, the exploration and docu-mentation of food web structure and dynamics of this lake are, ingeneral, in a very preliminary stage and more attention must be givento understanding the ecosystem as a whole.

In summary, the invading zooplankter E. coregoni is now asubstantial component of the zooplankton community in the NorthBasin of Lake Winnipeg. In all likelihood, E. coregoni will persist inLake Winnipeg and this species' abundance will continue to fluctuatebased on annual limnological conditions. The establishment of E.coregoni has likely been facilitated by both a variable climate andhuman activities. In the near future, other aquatic invaders willundoubtedly enter the LakeWinnipeg ecosystem, some of which mayhave enormous ecological and economic impacts. For a lake thatprovides substantial environmental services via recreational, com-mercial fishing, and energy generation opportunities, the conse-quences of a deteriorating ecosystem due to the invasion of aquaticspecies, in addition to other anthropogenic and climatic factors, couldbe economically significant. More effort and time must be dedicatedto study and protect Lake Winnipeg before the conditions of thisvaluable ecosystem degenerate.

Acknowledgments

We thank Paul Wilkinson andMike Stainton for 210Pb dating of thecore. Mike Stainton also supplied water chemistry data and sedimentcore material. Mike Paterson, Darren Gillis, and two anonymousreviewers provided valuable comments on the manuscript. KenSandilands and Lu Guan assisted in the production of ArcMaps. Wethank the crew of the M.V. NAMAO and all individuals who helped inthe collection of samples used in this study. This research was fundedby the University of Manitoba, Fish Futures Inc., and Fisheries andOceans Academic Subvention Program.

References

Allan, J.D., Goulden, C.E., 1980. Some aspects of reproductive variation among fresh-water zooplankton. In: Kerfoot, W.C. (Ed.), Evolution and ecology of zooplanktoncommunities. University Press, London, pp. 388–410.

Allen, Y.C., Ramcharan, C.W., 2001. Dreissena distribution in commercial waterways ofthe US: using failed invasions to identify limiting factors. Can. J. Fish. Aquat. Sci. 58,898–907.

Appleby, P.G., Oldfield, F., 1983. The assessment of 210Pb data from sites with varyingsediment accumulation rates. Hydrobiologia 103, 29–35.

Bajkov, A., 1934. The plankton of Lake Winnipeg drainage system. Int. Rev. Ges.Hydrobiol. Hydrogr. 31, 239–272.

Boucherle, M.M., Züllig, H., 1983. Cladoceran remains as evidence of change in trophicstate in three Swiss lakes. Hydrobiologia 103, 141–146.

Bourne, A., Armstrong, N., Jones, G., 2003. A preliminary estimate of total nitrogen andtotal phosphorus loading to streams in Manitoba, Canada. Manitoba ConservationReport 2002-04. 49pp.

Brunskill, G.J., Elliot, S.E.M., Campbell, P., 1980. Morphometry, hydrology, andwatershed data pertinent to the limnology of Lake Winnipeg. Can. MS. Rep. Fish.Aquat. Sci. 1556, v + 23p.

Buchan, L.A.J., Padilla, D.K., 1999. Estimating the probability of long-distance overlanddispersal of invading aquatic species. Ecol. Appl. 9, 254–265.

Bur, M.T., Klarer, D.M., Krieger, K.A., 1986. First records of a European cladoceran,Bythotrephes cederstroemi, in Lake Erie and Huron. J. Great Lakes Res. 12, 144–146.

Cáceres, C.E., Soluk, D.A., 2002. Blowing in the wind: a field test of overland dispersaland colonization by aquatic invertebrates. Oecologia 131, 402–408.

Campbell, K.B., Derksen, A.J., Remnant, R.A., Stewart, K.W., 1991. First specimens of therainbow smelt, Osmerus mordax, from Lake Winnipeg, Manitoba. Can. Field-Nat.105, 568–570.

De Melo, R., Hebert, P.D.N., 1994a. Founder effects and geographical variation in theinvading cladoceran Bosmina (Eubosmina) coregoni Baird 1857 in North America.Heredity 73, 490–499.

De Melo, R., Hebert, P.D.N., 1994b. A taxonomic reevaluation of North AmericanBosminidae. Can. J. Zool. 72, 1808–1825.

Deevey Jr., E.S., Deevey, G.B., 1971. The American species of Eubosmina Seligo(Crustacea, Cladocera). Limnol. and Oceanogr. 16, 201–218.

Franzin, W.G., Barton, B.A., Remnant, R.A., Wain, D.B., Pagel, S.J., 1994. Range extension,present and potential distribution, and possible effects of rainbow smelt in HudsonBay drainage waters of northwestern Ontario, Manitoba, and Minnesota. N. Am. J.Fish. Manage. 14, 65–76.

Page 8: Investigating the invasion of the nonindigenous zooplankter, Eubosmina coregoni, in Lake Winnipeg, Manitoba, Canada

166 K.D. Suchy et al. / Journal of Great Lakes Research 36 (2010) 159–166

Frey, D.G., 1960. The ecological significance of cladoceran remains in lake sediments.Ecology 41, 684–699.

Frey, D.G., 1982. Questions concerning cosmopolitanism in Cladocera. Arch. Hydrobiol.93, 484–502.

Gasiorowski, M., Szeroczynska, K., 2004. Abrupt changes in Bosmina (Cladocera,Crustacea) assemblages during the history of the Ostrowite Lake (northernPoland). Hydrobiologia 526, 137–144.

Grigorovich, I.A., Pashkova, O.V., Gromova, Y.F., van Overdijk, C.D.A., 1998. Bythotrepheslongimanus in the Commonwealth of Independent States: variability, distributionand ecology. Hydrobiologia 379, 183–198.

Hairston Jr., N.G., Van Brunt, R.A., Kearns, C.M., 1995. Age and survivorship ofdiapausing eggs in a sediment egg bank. Ecology 76, 1706–1711.

Hall, S.R., Mills, E.L., 2000. Exotic species in large lakes of the world. Aquat. Ecosyst.Health Manag. 3, 105–135.

Hann, B.J., 1989. Methods in quaternary ecology No. 6. Cladocera. Geosci. Can. 16,17–26.

Hann, B.J., Hebert, P.D.N., 1982. Re-interpretation of genetic variation in Simocephalus(Cladocera, Daphniidae). Genetics 102, 101–107.

Harvey, C.T., Qureshi, S.A., MacIsaac, H.J., 2009. Detection of a colonizing, aquatic, non-indigenous species. Diversity Distrib. 15, 429–437.

Havel, J.E., Medley, K.A., 2006. Biological invasions across spatial scales: interconti-nental, regional, and local dispersal of cladoceran zooplankton. Biol. Invasions 8,459–473.

Hofmann, W., 1978. Bosmina (Eubosmina) populations of Grosser Plöner See andSchöhsee lakes during late-glacial and postglacial times. Pol. Arch. Hydrobiol. 25,167–176.

Kerfoot, W.C., 1974. Net accumulation rates and the history of cladoceran communities.Ecology 55, 51–61.

Kerfoot, W.C., 1981. Long-term replacement cycles in cladoceran communities: ahistory of predation. Ecology 62, 216–233.

Kling, H.J., 1998. A summary of past and recent plankton of Lake Winnipeg, Canadausing algal fossil remains. J. Paleolimnol. 19, 297–307.

Kolar, C.S., Lodge, D.M., 2001. Progress in invasion biology: predicting invaders. TREE16, 199–204.

Lake Winnipeg Stewardship Board, 2006. Reducing nutrient loading to Lake Winnipegand its watershed. Our Collective Responsibility and Commitment to Action. Reportto the Minister of Water Stewardship. (www.lakewinnipeg.org).

Lennon, J.T., Smith, V.H., Dzialowski, A.R., 2003. Invasibility of plankton foodwebs alonga trophic state gradient. Oikos 103, 191–203.

Lieder, U., 1983. Revision of the genus Bosmina BAIRD, 1845 (Crustacea, Cladocera). Int.Revue. Ges. Hydrobiol. 68, 121–139.

Lieder, U., 1991. The Bosmina kessleri-like morphotype of Eubosmina in Lake Muskoka,Ontario, Canada, as putative interspecific hybrids. Hydrobiologia 225, 71–80.

Mabee, W.R., 1988. A note on the occurrence of Bosmina (Eubosmina) coregoni(Crustacea, Cladocera) in Missouri: a new record and range extension. Southwest.Nat. 43, 95–96.

MacIsaac, H.J., Grigorovich, I.A., Hoyle, J.A., Yan, N.D., Panov, V.E., 1999. Invasion of LakeOntario by the Ponto Caspian predatory cladoceran Cercopagis pengoi. Can. J. Fish.Aquat. Sci. 56, 1–5.

MacIsaac, H.J., Grigorovich, I.A., Ricciardi, A., 2001. Reassessment of species invasionsconcepts: the Great Lakes basin as a model. Biological Invasions 3, 405–416.

MacIsaac, H.J., Borberly, J.V.M., Muirhead, J.R., Graniero, P.A., 2004. Backcasting andforecasting biological invasions of inland lakes. Ecol. Appl. 14, 773–783.

Makarewicz, J.C., Grigorovich, I.A., Mills, E., Damaske, E., Cristescu, M.E., Pearsall, W.,LaVoie, M.J., Keats, R., Rudstam, L., Hebert, P., Hallbritter, H., Kelly, T., Matkovich,C., MacIsaac, H.J., 2001. Distribution, fecundity, and genetics of Cercopagis pengoi

(Ostroumov) (Crustacea, Cladocera) in Lake Ontario. J. Great Lakes Res. 27,19–32.

Manly, B.F.J., 1997. Randomization, bootstrap and Monte Carlo methods in biology, 2nded. Chapman & Hall, New York.

Mills, E.L., Leach, J.H., Carlton, J.T., Secor, C.L., 1993. Exotic species in the Great Lakes: ahistory of biotic crises and anthropogenic introductions. J. Great Lakes Res. 19,1–54.

Oldfield, F., Appleby, P.G., 1984. Empirical testing of 210Pb dating models. In: Haworth,E.Y., Lund, J.G. (Eds.), Lake sediments and environmental history. LeicesterUniversity Press, pp. 93–124.

Patalas, K., Salki, A., 1992. Crustacean plankton in LakeWinnipeg: variation in space andtime as a function of lake morphology, geology, and climate. Can. J. Fish. Aquat. Sci.49, 1035–1059.

R Development Core Team, 2008. R: A language and environment for statisticalcomputing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org.

Reid, V.A., Carvalho, G.R., George, D.G., 2000. Molecular genetic analysis of Daphnia inthe English Lake District: species identity, hybridization and resting egg banks.Freshwater Biol. 44, 247–253.

Ricciardi, A., MacIsaac, H.J., 2000. Recent mass invasion of the North American GreatLakes by Ponto-Caspian species. TREE 15, 62–65.

Sanford, P.R., 1993. Bosmina longirostris antennule morphology as an indicator ofintensity of planktivory by fishes. Bull. Mar. Sci. 53, 216–227.

Salki, A., 1996. The crustacean plankton community of Lake Winnipeg in 1929, 1969,and 1994. In: Todd, B.J., Lewis, C.F.M., Thorleifson, L.H., Nielson, E. (Eds.), LakeWinnipeg Project: Cruise Report and Scientific Results, pp. 319–344.

Sarmaja-Korjonen, K., 2002. Multi-proxy data from Kaksoislammi Lake in Finland:dramatic changes in the late Holocene cladoceran assemblages. J. Paleolimnol. 28,287–296.

Schindler, D.W., 2001. The cumulative effects of climate warming and other humanstresses on Canadian freshwaters in the newmillennium. Can. J. Fish. Aquat. Sci. 58,18–29.

Shurin, J.B., Havel, J.E., 2002. Hydrologic connections and overland dispersal in an exoticfreshwater crustacean. Biol. Invasions 4, 431–439.

Sprules, W.G., Riessen, H.P., Jin, E.H., 1990. Dynamics of the Bythotrephes invasion of theSt. Lawrence Great Lakes. J. Great Lakes Res. 16, 346–351.

Stainton, M.P., Capel, M.J., Armstrong, F.A.J., 1974. The chemical analysis of freshwater.Can. Fish. Mar. Serv. Misc. Spec. Publ. 25, 120.

Suchy, K.D., Hann, B.J., 2007. Using microfossil remains in lake sediments to examinethe invasion of Eubosmina coregoni (Cladocera, Bosminidae) in Lake of the Woods,Ontario, Canada. J. Great Lakes Res. 33, 867–874.

SYSTAT Software, Inc., 2007. SYSTAT 12. Richmond, CA.Szeroczynska, K., 2002. Human impact on lakes recorded in the remains of Cladocera

(Crustacea). Quaternary International 95-96, 165–174.Todd, B.J., Lewis, C.F.M., Nielsen, E., Thorleifson, L.H., Bezys, R.K., Weber, W., 1997. Lake

Winnipeg: geological setting and sediment seismostratigraphy. J. Paleolimnol. 19,215–244.

Vanschoenwinkel, B., Gielen, S., Seaman, M., Brendonck, L., 2008. Any way the windblows—frequent wind dispersal drives species sorting in ephemeral aquaticcommunities. Oikos 117, 125–134.

Vitousek, P.M., D’Antonio, C.M., Loope, L.L., Rejmánek, M., Westbrooks, R., 1997.Introduced species: a significant component of human-caused global change. N.Z.J. Ecol. 21, 1–16.

Vijverberg, J., Boersma, M., 1997. Long-term dynamics of small-bodied and large-bodied cladocerans during the eutrophication of a shallow reservoir, with specialattention to Chydorus sphaericus. Hydrobiologia 360, 233–242.