Quagga mussels (Dreissena bugensis) as biomonitors of metal contamination: A case study in the upper St. Lawrence River

Journal of Great Lakes Research 37 (2011) 140–146

Contents lists available at ScienceDirect

Journal of Great Lakes Research j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / j g l r

Quagga mussels (Dreissena bugensis) as biomonitors of metal contamination: A case study in the upper St. Lawrence River Carolyn Johns ⁎ Environmental Studies Department, St. Lawrence University, Canton, NY 13617, USA

a r t i c l e

i n f o

Article history: Received 8 April 2010 Accepted 24 September 2010 Available online 3 December 2010 Communicated by David Barton Index words: Quagga mussel Bioaccumulation Metal contamination Biomonitor St. Lawrence River

a b s t r a c t In this study, the utility of quagga mussels (Dreissena bugensis) as biomonitors was investigated by measuring total concentrations of three trace metals, cadmium, copper, and zinc, in soft tissues. Quagga mussels were sampled from five sites along the upper St. Lawrence River, including one industrially influenced site, from 1999 through 2007. Mussels were collected from near-shore areas, divided into 5 size classes based on maximum shell length, and tissues were pooled for analysis of each size group. Two-way analysis of variance and a posteriori range tests were used to test for differences among sites along a distance gradient from the outflow of Lake Ontario and to examine inter-annual variability within and among sites. Cadmium concentrations were higher nearer the outflow of the lake. Copper concentrations varied among sites and years, but were generally highest near the industrial site. Zinc concentrations were relatively uniform, possibly reflecting internal regulation. Animal size measured as shell length was not an important factor in this section of the river, but warrants further consideration in a wider range of ecosystems and contaminant exposure levels. In general, concentrations of the three metals were not high compared to reports in the published literature for dreissenid mussels in contaminated environments. However, few studies have utilized quagga mussels rather than zebra mussels. The two species may differ in bioaccumulation patterns and may not be interchangeable as biomonitors. Further studies of bioaccumulation of contaminants by quagga mussels in a wider range of contaminant exposures would be useful particularly as quagga mussels displace zebra mussels in the Laurentian Great Lakes and the St. Lawrence River. © 2010 Published by Elsevier B.V. on behalf of International Association for Great Lakes Research.

Introduction Benthic macro-invertebrates have played a major role in evaluating contamination of both marine and freshwater ecosystems worldwide. Mussels, especially, have been used to delineate spatial distribution patterns of organo-chlorine and metal contaminants and to monitor changes in conjunction with remediation activities (Brown and Luoma, 1995; Metcalfe and Charlton, 1990; Rainbow et al., 2002; Richman and Somers, 2010; Tessier et al., 1984; Wang et al., 2002). These organisms assimilate a proportion of contaminants available in dissolved, particulate, biotic (prey), or sedimentary forms, depending on their particular microhabitats, diet, metabolism, and life histories (Luoma and Rainbow, 2005, 2008). Many species in these groups are useful as biomonitoring organisms as they are relatively long-lived and sedentary, are widespread and easy to collect, accumulate contaminants in proportion to length and level of exposure yet are relatively tolerant of these contaminants, and are large enough to provide suffi cient tissue to analyze easily (Phillips and

⁎ Tel.: + 1 315 229 5814. E-mail address: [email protected].

Rainbow, 1994). Dreissenid mussels in North America fulfill many of these requirements for biological monitoring organisms. Quagga mussels (Dreissena bugensis) were first found in Lake Erie in 1989, a year after the zebra mussels (D. polymorpha) had been identified in Lake St. Clair (Mills et al., 1999; New York Sea Grant, 2002). A highly invasive species, by 1993 quagga mussels had spread eastward to Quebec City along the St. Lawrence River and by 1995 had become more abundant in many areas of Lake Ontario than the zebra mussel (Mills et al., 1996, 1999). While very similar to the zebra mussel in terms of ecological niche, the quagga mussel may better tolerate lower temperatures and may be able to outcompete the zebra mussel in shallow waters even during summer months (Stoeckmann, 2003; Vanderploeg et al., 2002). Zebra mussels function as ecosystem engineers in shallow waters by selectively grazing, altering nutrient cycles, contributing organic matter in the form of feces and pseudofeces to benthic areas, increasing water clarity, and changing the physical habitat by producing clumps of shells in their colonies or druses (Limburg et al., 2010; Vanderploeg et al., 2002). Presumably, as quagga mussels increase in numbers in shallow waters, their role will be similar. Zebra mussels bioaccumulate various contaminants, including PCBs and metals, through their extensive filter feeding (Comba et al., 1996; Kwan et al., 2003; Kraak et al., 1991; Mills et al.,

0380-1330/$ – see front matter © 2010 Published by Elsevier B.V. on behalf of International Association for Great Lakes Research. doi:10.1016/j.jglr.2010.11.002

C. Johns / Journal of Great Lakes Research 37 (2011) 140–146

1993). Because quagga mussels feed in a similar manner and on similar food items, they are also likely to bioaccumulate contaminants in their tissues. Compared to zebra mussels, only a few studies have been published on bioaccumulation by quaggas and their potential as biomonitors of contaminants (Johns and Timmerman, 1998; Richman and Somers, 2005, 2010; Roseman et al., 1994; Rutzke et al., 2000). Given their increasing prevalence in the Laurentian Great Lakes and across North America (e.g., into Arizona, Nevada and California; Stokstad, 2007), more understanding of the quaggas' ability for bioaccumulation would be useful to environmental managers. The overall goal of this study is to examine bioaccumulation of selected trace metals by quagga mussels using the international section of the St. Lawrence River (SLR) as a case study. The upper SLR is an understudied ecosystem compared to Lake Ontario or to the river downstream of Cornwall, Ontario (Twiss, 2007). In this section, most of the river's water is from the outflow of Lake Ontario with few major point discharges until the Cornwall-Massena Area of Concern below the Moses Saunders Power Dam (Fig. 1). Clean-up activities over the past decades, required by the Great Lakes Water Quality Agreement (International Joint Commission (1978)) have improved water quality in Lake Ontario and the St. Lawrence River by reducing inputs of phosphorus and toxic contaminants (Stevens and Neilson, 1987). Previous work at Cape Vincent, NY, near the outflow of the lake, found elevated concentrations of cadmium, copper and zinc in both zebra and quagga mussels sampled in 1993 from May to October (Johns and Timmerman, 1998). Subsequently, zebra mussels showed spatially variable tissue concentrations of these metals at six sites along the international portion of the river with higher bioaccumulation near the outflow of Lake Ontario and near an industrialized site (Johns, 2001). Numbers of quagga mussels began increasing in the upper SLR in the mid-1990s. Beginning in 1999, quagga mussels were sampled at five of these same sites. In this study quagga mussels are hypothesized to bioaccumulate metals in the upper SLR corresponding to proximity to Lake Ontario and to show declines over time as water quality has improved in the lake. Specific objectives of this study are to (1) delineate spatial variability in bioaccumulation of three metals:

141

copper, cadmium, and zinc, (2) examine trends in inter-annual variability of metal concentrations, and (3) elucidate any influence of size of the mussels on extent of bioaccumulation. Methods Study area, water characteristics and sample sites The St. Lawrence River (SLR) has the second largest discharge of rivers in North America (Hudon and Carignan, 2008) with a mean flow of 7500–9200 m3/s at the Moses Saunders hydropower dam spanning the river at Cornwall, Ontario, the lower end of the study area, 110 km downstream of Lake Ontario (Basu et al., 2000; Hudon, 2000; Morin et al., 2003). Approximately 95–99% of the flow at the hydropower dam consists of outflow from the lake. The SLR is a hard-water river with calcium carbonate buffering; alkalinity is approximately 90 mg/L CaCO3 (Kleinschmidt Assoc. 1996). Water hardness ranged from 110 to 130 mg/L CaCO3 from 1990 to 1996 and from 2007 to 2008 (USGS NWIS). River waters are slightly basic with pH averaging 7.5–8.19 (USGS NWIS). The SLR contains unusually clear and low nutrient waters for a river of its magnitude (Hudon and Carignan, 2008). Levels of dissolved organic carbon in the upper river averaged 5.5 ± 0.2 mg/L in 1997 with no spatial trend in concentrations between Lake Ontario and Trois Rivieres, Quebec (Basu et al., 2000). Dissolved oxygen, measured at the power dam during the ice free months, reached 12–14.1 mg/L in March, decreased to approximately 8 mg/L in summer months and rose to approximately 10 mg/L in October and November. Water temperatures range from 0.2 °C in February to 22–23 °C in July and August (USGS NWIS). Sample sites were established in an earlier biomonitoring study of zebra mussels (Johns, 2001) in shallow, littoral areas. Mussels could be accessed by wading in less than 1-m depth of water when river levels decreased by 0.3–0.5 m in fall months. Sites are designated 1–5 in this study (Fig. 1), corresponding to sites 1, 3, 6, 8, and 10 from the previous study (Johns, 2001). Site 5 is adjacent to an industrial site and landfill located within the IJC's Massena-Cornwall Area of Concern. While

Fig. 1. Map of the international section of the St. Lawrence River and locations of quagga mussel sampling sites.

142

C. Johns / Journal of Great Lakes Research 37 (2011) 140–146

primarily a site of concern for PCBs, metal contaminants were also present. The other four sites were chosen for a combination of proximity to the dominant current, public accessibility, distance from lake outflow, and abundance of mussels. None of the sites was presumed, a priori, to be an uncontaminated reference site. Local tributaries do not yet have resident populations of either zebra or quagga mussels and mostly drain through agricultural (dairy) areas, limiting their potential utility as reference areas even if dreissenids were present. Collection and handling of quagga mussels Quaggas were hand-picked from 10 to 12 hard substrates along a 5-m length of shoreline at each site annually in October. Depending on the strength of the current, at some sites mussels were collected from upper surfaces of rocks (2, 3); at other sites they were collected along rock edges or from the undersides (1, 5, and 4). Only mussels with shell lengths approximately 12 mm or greater were collected in the field due to difficulty in dissecting tissues from smaller animals without breaking their shells. Animals were transported to the lab in river water in clean plastic containers. In the lab, the mussels were placed in clean glass aquaria in commercially available spring water and held at 10 °C for 48–72 h to depurate. Depuration helps eliminate possible bias in estimates of metal concentrations in tissues associated with undigested material in the digestive system (Lobel et al., 1991). After depuration, animals were removed from their containers, drained, and sorted into five (occasionally four or six) size classes based on maximum shell length. Lengths were measured to the nearest 0.1 mm with a digital caliper and recorded. To obtain sufficient biomass for analysis, soft tissues, dissected from the shells of animals in each size class, were pooled, placed into clean borosilicate scintillation vials, and dried at 60 °C for 48 h. Dry weights were recorded to the nearest 0.0001 g. The number of individuals in each size class varied depending on the shell lengths with fewer (five or six) in larger size classes and more (17–20) in the smallest groups. Mean lengths and metals concentrations reported are based on, generally, five size classes per site, each size class containing 5–20 individuals. Determination of metal concentrations Dry tissue samples were digested by reflux in hot 16 N HNO3 (Johns and Timmerman, 1998). When solutions cleared, digestion was considered complete and the acid was allowed to evaporate to dryness. The residues were reconstituted to a final volume of 7.5 mL in 3 N HCl. After a 48-h period to allow time for the residue to fully dissolve, sample solutions were filtered to pass through a 0.45-μm membrane. Samples from 2001 were analyzed by inductively coupled plasma absorption spectrophotometry (ICAPS). Samples from 1999 and 2000 and 2002 through 2007 were analyzed by flame atomic absorption spectrophotometry (AAS) on a Perkin Elmer Analyst 800. Quality control/quality assurance activities were comprised of regular inclusion and analysis of reagent blanks in the sample stream and digestion of aliquots of NIST SRM 1566a Oyster Tissue along with the sets of mussel tissue samples at a rate of 5–10%. For samples analyzed by AAS, a mid-range calibration standard was periodically checked during sample analysis and full standard recalibration was performed if the value of the standard deviated by more than 10%. Replicate analyses of samples and analyses of SRM Oyster Tissue were made for every analytical run at a level of 10% of the total number of samples analyzed. For analyses by ICAPS, samples were coded for analysis. Reagent blanks, digested SRM tissue samples, and SRM 1643d Trace Metals in Water were interspersed in the ICAPS sample sequence. Recoveries of SRM Oyster Tissue averaged 109.8%, 98.2%, and 108.7% for cadmium, copper, and zinc, respectively. Recoveries from SRM water averaged 91.2%, 95.2%, and 105.3%. For samples from 1999, 2000 and 2002–2007, analyzed by flame AAS, recoveries averaged 107.7%, 98.2% and 103.5% of SRM 1566b certified values

for cadmium, copper and zinc. All reported concentrations of metals for mussels have been adjusted for metal concentrations in the reagent blanks, but not for recoveries from standard reference materials. Limits of detection were lower for the 2001 samples analyzed by ICAPS. Statistical analyses Simple linear regressions of mean shell length and metal concentration in each size pool were performed using SPSS (16.0) for each site in each year in order to test for the effect of animal size on metal concentration. If significant linear relationships were not regularly found, the metal concentration of each size pool was averaged with the others to calculate a mean metal concentration. A two-way analysis of variance (ANOVA) was performed on shell lengths using year and site as fixed effects in order to determine whether animal sizes varied significantly over the course of the study period. Results were deemed significant if p ≤ 0.05. In order to investigate possible inter-annual variability in total metal concentrations and animal size, two-way analysis of variance (general linear model, type IV) was performed. Mussels were collected from most sites in most years, but a few are missing (2007 at site 1 and 2005 at site 5). The type IV model accommodates unbalanced cell design (unequal numbers of size pools of animals per collection and missing cells). Variances for means of metal concentrations and average shell lengths were examined for heteroscedascity by use of Levene's statistic which does not assume an underlying normal distribution in the data (SPSS, Inc. (16.0); Zar, 1999). If sample variances were not homogenous, data were converted using a log10 transformation and the variances were retested. Transformation decreased heteroscedascity for zinc, but did not eliminate it. For copper and cadmium, log10 transformation increased the F ratio in the Levene's test so untransformed data were used. The general linear model is relatively robust to departures from normality and homogeneity of variances (Zar, 1999). If any ANOVA found that means differed significantly (p≤0.05), a post-hoc range test was performed, Fisher's LSD if variances were homogeneous(shell length). Otherwise Dunnett's T3, was performed using p≤ 0.05 as the level of significance (Zar, 1999). This test does not assume variances are equal. Friedman's test does not incorporate more than one data point per cell so could not be utilized. Overall, the approach taken to analysis was assumed to be conservative in accepting any differences as significant. Results Spatial and inter-annual variability in bioaccumulation of metals Total cadmium concentrations in quagga mussels varied significantly among sites (Fig. 2a, Table 1) Levels of Cd were significantly higher at sites 1 and 5 compared to the other sites. Overall, means of total copper also varied spatially along the upper river. Levels at sites 5 and 3 were similar but exceeded those at sites 2 and 4 (Fig. 2b, Table 1). The grand mean of copper at site 1 was intermediate: lower than site 5 but higher than sites 2 and 4. The overall mean concentrations of total zinc varied little among the five sites. The only significant difference occurred at site 4 where levels were lower compared to the other four sites (Fig. 2c, Table 1). Concentrations of all three metals varied significantly in the mussels over the 9-year period studied. However, no general trend toward lower concentrations over time was apparent. For cadmium, tissue levels in the year 2000 were significantly higher than for all other years except 2001 and 2006 (Table 1, Fig. 2a). Mean concentration in 2007 was lower than in 2000, 2001, and 2006 due to low variability in 2007 (small standard error). For copper, the mean concentrations in 2003 were highest and were greater than in 1999, 2000, 2001 and 2007 (Table 1, Fig. 2b). Copper levels were not significantly dissimilar between 2002 and 2006. Levels in 2007 were significantly lower than for most years after 2001. The grand mean

C. Johns / Journal of Great Lakes Research 37 (2011) 140–146

143

occurred at sites 1 and 5, where the mean size and range of sizes of the mussels were smaller and larger, respectively, on average (Fig. 2d, Table 1). Location of the sampling site along the river accounted for only 15.8% of variability in mussel size (ANOVA table not shown). Average animal size was consistent at each site for the study period (Table 1, Fig. 2d). Simple linear regressions using concentrations of each of the three metals as the dependent variable and with length as the fixed variable were performed for each year at each site. Very few significant regressions emerged. For copper, 6 of 42 regressions were significant (p ≤ 0.05) ranging from zero to two for any given site. For cadmium and zinc, only 3 of 42 regressions analyses were significant (p ≤ 0.05). As a result of these analyses, animal size was not further considered as an important factor for quagga mussels' uptake of biologically available metals. Discussion

Fig. 2. Total metal concentrations in quagga mussel tissues and mean shell lengths are shown for each year and for each sampling site. Each column represents the mean with standard error bar for a collection. Columns are clustered to show results of annual collections at the given site. a (Cd); b (Cu); c (Zn); d (length).

zinc concentrations showed a broader trend: levels in 1999, 2000, 2001 and 2004 were similar to each other but were significantly lower than those in 2002 and 2003 (Table 1, Fig. 2c). The ANOVA interaction terms for site and year were highly significant (p ≤ 0.0001) for all three metals indicating occurrence of significant differences amongst years at the various sites. Again, clear trends were hard to decipher. However, overall, year of sampling and location along the river, combined, accounted for 75% of the variability of cadmium, 64.7% of variability of copper, and 63.7% for zinc in quagga tissues (ANOVA tables not shown). Variability of animal size and correlation of size with metal concentrations For grand means of shell lengths, mussels were similar in average size across sites 2, 3, and 4 (Fig. 2d, Table 1). Significant differences

Metal concentrations in quagga mussels showed neither strong downstream nor temporal declines as would be expected if the outflow of Lake Ontario provided a significant, but declining, source of metals to the upper SLR during this study. Except at sites 1 and 5, cadmium concentrations were generally below 1.5 μg/g, relatively low compared to bioaccumulation in dreissenid mussels affected by industrial contamination or urban runoff (Table 2; De Kock and Bowmer, 1993; Kwan et al., 2003; Kraak et al., 1991; Richman and Somers, 1995, Richman and Somers, 2010). However, compared to the early to mid-1990s, average cadmium concentrations in quagga mussels at site 1 declined from a range of 4.5–8.7 μg/g to 0.98–3.22 μg/ g (Table 2). This decline, coupled with current similarity of levels at sites 2, 3 and 4, may reflect decreased loading of cadmium to the river between the early and late 1990s with slightly elevated cadmium levels remaining only near the Lake's outflow and the industrial site. The lack of strong temporal trends of cadmium, despite some annual fluctuations, suggests reduction of cadmium loading to the river by the year 2000. The ranges of cadmium concentrations in mussels were small and the means low; thus, inter-annual differences still present may not be toxicologically meaningful. Mean copper levels in quaggas varied more than cadmium or zinc, both spatially and annually. Mean concentrations were consistently highest near the industrialized facility (site 5; Fig. 2b) and lowest mid-river (sites 2 and 4; Fig. 2b). Tissue copper spiked in 2003 compared to earlier years. Only in 2007 were copper levels in mussels significantly lower than in the 2000s (Table 1, Fig. 2b). Copper concentrations in mussels from sites 1, 3, and 5 were comparable to those from contaminated sites in the lower Niagara River (Table 2, Richman and Somers, 2010). With a few exceptions, mean zinc concentrations in quaggas were uniform between years and similar amongst sites. Both copper and zinc are required micro-nutrients in animals, important for formation of enzymes and hemocyanin. Some studies indicate that zinc may be subject to internal regulation by mollusks except in highly contaminated environments (Luoma, 1989; Luoma and Rainbow, 2008; Kraak et al., 1994a, 1994b). Environmental exposures to biologically available zinc in the upper SLR may not be high enough to overwhelm these internal mechanisms. Many biomonitoring studies use mussel size as an indication of age and thus as a proxy for length of exposure to contaminants in the environment. Dreissenid mussels do not form clear annual growth rings on their shells making it difficult to definitively determine their age. Animal size was similar across sites 2–5 so they were assumed to be of similar age. Mussels at site 1 were smaller, on average, regardless of year of sampling (Fig. 2d, Table 1). Mean shell lengths of quaggas at site 1 declined from 34.7 mm in 1993–27.7 mm in 1995 and then to 17.85 mm for 1999–2007 (Table 2, Johns and Timmerman, 1998), possibly related to higher animal densities and intensified competition for food. For this

144

C. Johns / Journal of Great Lakes Research 37 (2011) 140–146

Table 1 Grand means (standard error) of metal concentrations in quagga tissues (μg/g dry wt.) and shell length (mm) from years 1999–2007 for all sites in the upper St. Lawrence River. Significant differences in mean concentrations or mean lengths of mussels among sites are designated by letters. Within each column, means followed by the same letters are not significantly different (Dunnett's T3 multiple range test with p ≤ 0.05). Sampling site (all years)

Cadmium mean (se)

1 2 3 4 5

2.31 1.08 1.06 1.31 3.39

Year of (all sites)

Mean (se)

1999 2000 2001 2002 2003 2004 2005 2006 2007

1.50 2.27 4.12 0.99 1.28 1.32 1.29 1.67 1.56

(0.14) (0.12) (0.1) (0.06) (0.75)

(0.14) (0.16) (1.23) (0.11) (0.14) (0.21) (0.11) (0.19) (0.10)

Copper mean (se) A B B BC AC

21.1 18.2 25.1 16.3 30.4

A B AB AD A A A AB AC

19.4 20.7 17.2 23.2 28.6 23.8 25.8 22.3 14.1

Zinc mean (se)

(1.45) (0.96) (1.56) (0.68) (1.75)

B CD AB CD A

75.2 77.2 73.8 64.0 80.2

B BC B ABC AC ABC ABC ABC BD

73.3 66.8 63.4 76.8 79.5 68.8 70.8 82.8 84.3

Mean (se)

(1.03) (3.78) (1.75) (1.70) (2.83)

Length mean (se) A A A B A

17.85 21.21 21.86 22.83 23.63

AB B BC AC AC B AB ABD AB

21.82 22.01 22.74 22.06 21.71 21.14 21.32 20.74 20.94

Mean (se)

(1.25) (1.45) (1.76) (2.3) (1.66) (2.42) (2.52) (1.27) (1.18)

study, mussels at site 1 were sampled from shallower waters than previously. Higher summer water temperatures may have increased respiration and decreased rates of growth (Stoeckmann, 2003). Smaller sized quaggas collected in the present study were likely younger with shorter exposure periods than mussels sampled in 1993–1995. Nevertheless, cadmium in mussels from site 1 remained elevated compared to downstream sites. Limited correlations of metal concentrations in tissues with shell size could result from several interacting factors. If overall biologically available concentrations of metals in the upper SLR are low, or only slightly enriched, little bioaccumulation or correlation would be expected. Second, the strength and direction of any size–concentration relationship can vary with season and over the lifetime of individuals with changes in metabolic rates (Strong and Luoma, 1981). Also, shell length may be a less effective proxy for animal age (length of exposure) for quagga mussels given reports that their growth rates are greater than rates of zebra mussels (Baldwin et al., 2002; Mills et al., 1999; Stoeckmann, 2003). If so, a growth-induced dilution effect may diminish or obscure metal bioaccumulation. Data from a wide range of contaminated sites would be useful in determining how important the size may be in assessing bioaccumulation. If significant correlations between animal size and tissue concentrations occur under more heavily contaminated conditions, regression models could be utilized to

(2.3) (1.6) (3.9) (1.8) (1.5) (1.6) (2.7) (4.1) (6.7)

(0.42) (0.58) (0.53) (0.66) (0.71)

A B B BC C

Mean (se) (0.84) (0.90) (0.77) (0.83) (0.81) (0.98) (1.03) (0.90) (0.87)

A A A A A A A A A

estimate the tissue concentrations for standard sized mussels to facilitate inter-site comparisons (Brown and Luoma, 1995). Relatively low bioaccumulation of these three trace metals by quaggas can be attributed to several factors. Water hardness may reduce uptake of dissolved species of the metals, although most bioaccumulation of cadmium by zebra mussels is from dietary uptake (Roditi and Fisher, 1999; Roditi, et al., 2000b). In high DOC rivers, such as the lower Hudson River in New York state, zebra mussels can utilize DOC as an energy source, ingesting any metals bound in this fraction (Roditi et al., 1996, 2000a). The relatively low DOC levels in the SLR could limit availability of some ligand bound species of these metals (Gillis et al., 2008). Cold temperatures present for half of each year could reduce feeding rates, thereby limiting uptake and assimilation of metals (Garton and Johnson, 2000; Johns, 2001). Metal inputs from local discharges are small and few. As discharges within the Lake Ontario basin have been reduced since the 1980s, cleaner sediments have been deposited on the lake bottom, covering older, more contaminated sediment and reducing re-suspension due to scouring or bioturbation. Implementation of strategies from the Lake Ontario Management Plan should be further reducing load of nutrients and contaminants, lowering levels in the outflow of the lake. As a result, less of the remaining trace metal contaminants in Lake Ontario may reach mussels in the upper SLR.

Table 2 Sizes and total metal concentrations of cadmium, copper, or zinc in quagga mussels sampled in the Laurentian Great Lakes Basin as reported in published literature compared to the present study. Location

Year collected

Size (mm) range

Cadmium (μg/g)

Copper (μg/g)

Zinc (μg/g)

Reference

Cape Vincent, NY, Lake Ontario Dunkirk, NY, Lake Erie Dunkirk, NY, Lake Erie Dunkirk, NY, Lake Erie Fort Erie, Ont., Lake Erie Upper Niagara River, Ont. & NY Lower Niagara River, Ont & NY Cape Vincent, NY, Lake Ontario+ Cape Vincent, NY, Lake Ontario+ Cape Vincent, NY, Lake Ontario+ Southern Lake Ontario, NY Upper Niagara River, Ont. & NY Lower Niagara River, Ont & NY Cape Vincent, NY (site 1) Site 4, St. Lawrence River Site 5, St. Lawrence River

June 1997 June 1997 June 1997 June 1997 1995 1995 1995 1992 1993 1994 1992 2003 2003 1999–2007 1999–2007 1999–2007

25–30 15–20 18–22 25–30 16–25 16–25 16–25 34.7(0.12)⁎ 31.1(0.27)⁎ 27.7(0.34)⁎

8.5 3.9 3.0 2.5 8.9, 10 0.8–4.9 3.9–6.5 5.7 (0.89)⁎ 8.7 (1.49)⁎ 4.5 (1.37)⁎

25 21 36 61 19, 16 15–21 15–20 16.2 (0.74)⁎ 14.6 (1.54)⁎ 16.0 (1.39)⁎

134 63 92 85 71, 68 58–630 72–260 64.3 (0.12)⁎ 62.1 (3.47)⁎ 69.9 (0.34)⁎

Mixed 10–25 10–25 16.4–19.3 19.8–24.8 23–26.9

3–6.4 2.5–6.9⁎⁎ 1.4–2.1⁎⁎ .98–3.22 .75–1.8 1.4–2.4

5–15 15–27⁎⁎ 20–60⁎⁎

71–236 57–64⁎⁎ 58–102⁎⁎

9.2–29.1 10.9–22.7 26.4–40.4

69.4–82.5 44.1–72.4 67.2–93

Rutzke et al., 2000 Rutzke et al., 2000 Rutzke et al., 2000 Rutzke et al., 2000 Richman and Somers, 2005 Richman and Somers, 2005 Richman and Somers, 2005 Johns and Timmerman, 1998 Johns and Timmerman, 1998 Johns and Timmerman, 1998 Mills et al., 1993 Richman and Somers, 2010 Richman and Somers, 2010 Present study Present study Present study

⁎ Mean and standard error. ⁎⁎ Means from several sample sites. Standard errors for each site are given by Richman and Somers, 2010. + Same as in site 1 in the present study.

C. Johns / Journal of Great Lakes Research 37 (2011) 140–146

Heavy colonization of Lake Ontario by dressenid mussels in the late 1980s and 1990s may have effectively reduced contaminant loads in the outflow of Lake Ontario in addition to reductions due to management activities. Dreissenids act as ecosystem engineers causing significant shifts in ecosystem processes including benthification of the food web and energy flow, “oligotrophication” of the overall water column and increased water clarity in littoral areas by means of their extensive filtering actions (Dobiesz and Lester, 2009; Mills et al., 2003; Limburg et al., 2010; Vanderploeg, et al., 2002). Biogenic particles from Lake Ontario's epilminion probably comprise most of the suspended particulate in the upper SLR (Lum et al., 1991). Zebra mussels obtain a large proportion of their assimilated cadmium from their diet not water (Roditi and Fisher, 1999; Roditi et al., 2000a). Given similarities in dietary preferences (Baldwin et al., 2002), presumably quaggas mussels do too. Reduced seston levels in the outflow due to dreissenid filtration coupled with deposition of contaminants with feces and pseudofeces into the benthic environment could result in lower levels of particulate bound metal exiting the lake. Elevated cadmium levels at found site 1 would be consistent with this explanation. Mussels at sites downstream would be exposed to much lower levels of metals and would bioaccumulate less. Quagga mussels are potentially useful biomonitors. They meet the established general criteria for suitable organisms: sedentary, abundant, easily identified, accessible for sampling, large enough for tissue analysis, relatively tolerant of contaminants (Phillips and Rainbow, 1994). Richman and Somers (2005, 2010) found that quaggas could differentially bioaccumulate organochlorines and metals relative to contamination levels. In the present study, quaggas bioaccumulated more cadmium and copper near an industrialized site. The present study lacked a definitive background site. Published studies examining metal concentrations in quagga mussels are few. More information is needed on bioaccumulation over a range of uncontaminated and contaminated sites. Quagga mussels inhabit substrates in all or parts of four of the Laurentian Great Lakes, occur in the St. Lawrence River as far as Quebec City, and have spread to parts of the western U.S. including Lake Mead, Nevada (Stokstad, 2007). Their wide distribution puts them in the category of “cosmopolitan biomonitors” (Luoma and Rainbow, 2008), allowing for comparisons among a wide array of sites. Measurements of trace metals in dissolved forms, suspended particulate or bed sediment give a partial picture of contaminant levels but do not represent concentrations available to biota (Luoma and Carter, 1991). Estimates of biologically available metals can be made from observations of metals bioaccumulated in tissues of biomonitors as these represent a “time integrated proxy measurement for total bioavailability of metals” (Luoma and Rainbow, 2008). Quagga mussels, if successful as biomonitors, would be important components in assessing ecological risk of contaminated systems, especially because dreissenids should no longer be viewed as “ecological dead-ends” in Great Lakes food webs (Madenjian et al., 2010).

Acknowledgments The author gratefully acknowledges financial support from the Environmental Studies Department and St. Lawrence University for costs associated with collection and analysis of mussel samples, including a faculty development grant in 2002. Eric Clark assisted with analysis of samples from 2006 and 2007. D. Cain and C. Brown at the USGS in Menlo Park facilitated analysis of samples from 2001 by ICAPS. M. Skeels of the Chemistry Department at St. Lawrence University provided crucial assistance in analysis of samples from 2004 to 2007 by flame AAS. The author very much appreciates the constructive comments from two anonymous reviewers of the manuscript and the Journal's associate editor. Carol Cady, GIS specialist at St. Lawrence University, constructed the map of the sampling area and sites.

145

References Baldwin, B.S., Mayer, M.S., Dayton, J., Pau, N., Mendilla, J., Sullivan, M., Moore, A., Ma, A., Mills, E.L., 2002. Comparative growth and feeding in zebra and quagga mussels (Dreissena polymorpha and Dreissena bugensis): implications for North American lakes. Can. J. Fish. Aquat. Sci. 59, 680–694. Basu, B.K., Kalff, J., Pinel-Alloul, B., 2000. Midsummer plankton development along a large temperate river: the St. Lawrence River. Can. J. Fish. Aquat. Sci. 57 (Suppl. 1), 7–15. Brown, C.L., Luoma, S.N., 1995. Use of the euryhaline bivalve Potamocorbula amurensis as a biosentinel species to assess trace metal contamination in San Francisco Bay. Mar. Ecol. Progr. Series. 124, 129–142. Comba, M.E., Metcalfe-Smith, J.L., Kaiser, K., 1996. Zebra mussels as biomonitors for organic contaminants in the lower Great Lakes. Water Qual. Res. J. Canada 2, 411–430. De Kock, W.C., Bowmer, C.T., 1993. Bioaccumulation, biological effects and food chain transfer of contaminants in the zebra mussel (Dreissena polymorpha). In: Nalepa, T., Schloesser, D.W. (Eds.), Zebra Mussels: Biology, Impacts, and Control. Lewis Publishers, Boca Raton, pp. 503–533. Dobiesz, N.E., Lester, N.P., 2009. Changes in mid-summer water temperature and clarity across the Great Lakes between 1968 and 2002. J. Great Lakes Res. 35, 371–384. Garton, D.W., Johnson, L.E., 2000. Variation in growth rates of the zebra mussel, Dreissena polymorpha, within Lake Wawasee. Freshw. Biol. 45, 443–451. Gillis, P.L., Mitchell, R.J., Schwalb, A.N., McNichols, K.A., Mackie, G.L., Wood, C.M., Ackerman, J.D., 2008. Sensitivity of the glochidia (larvae) of freshwater mussels to copper: assessing the effect of water hardness and dissolved organic carbon on the sensitivity of endangered species. Aquat. Toxicol. 88, 137–145. Hudon, C., 2000. Phytoplankton assemblages in the St. Lawrence River, downstream of its confluence with the Ottawa River, Quebec. Canada. Can. J. Fish. Aquat. Sci. 57, 16–30. Hudon, C., Carignan, R., 2008. Cumulative impacts of hydrology and human activities on water quality in the St. Lawrence River (Lake Saint-Pierre, Quebec, Canada). Can. J. Fish. Aquat. Sci. 65, 1165–1180. Agreement, with annexes and terms of references, between the United States of America and Canada signed at Ottawa November 22, 1978, and Phosphorus Load Reduction Supplement signed October 16, 1983. International Joint Commission (IJC), Windsor, Ontario. Johns, C., 2001. Spatial distribution of total cadmium, copper and zinc in the zebra mussel (Dreissena polymorpha) along the upper St. Lawrence River. J. Great Lakes Res. 27, 354–366. Johns, C., Timmerman, B., 1998. Total cadmium, copper, and zinc in two Dreissenid mussels, Dreissena polymorpha and Dreissena bugensis, at the outflow of Lake Ontario. J. Great Lakes Res. 24, 55–64. Kleinschmidt Associates and Beak Consultants, Inc., 1996. New York Power Authority SLR-FDR Power project. Initial consulting package for Relicensing, April, 1996. Kraak, M.H.S., Scholten, W.H.M., Peeters, W.H.M., de Kock, W.C., 1991. Biomonitoring of heavy metals in the Western European rivers Rhine and Meuse using the freshwater mussel Dreissena polymorpha. Environ. Pollut. 74, 101–104. Kraak, M.H.S., Toussaint, M., Lavy, D., Davids, C., 1994a. Short term effects of metals on filtration rate of the zebra mussel, Dreissena polymorpha. Environ. Pollut. 84, 139–143. Kraak, M.H.S., Wink, Y.A., Stuijfzand, S.C., Buckest-DeJong, M.C., Degro, C.J., Admiraal, W., 1994b. Chronic ecotoxicity of Zn and Pb to the zebra mussel Dreisenna polymorpha. Aquat. Tox. 30, 77–89. Kwan, K.H.M., Chan, H.M., de LaFontaine, Y., 2003. Metal contamination in zebra mussels (Dreissena polymorpha) along the St. Lawrence River. Environ. Mon. Assess. 88, 193–219. Limburg, K.E., Luzadis, V.A., Ramsey, M., Schulz, K.L., Mayer, C.M., 2010. The good, the bad, and the algae: perceiving ecosystem services and disservices generated by zebra and quagga mussels. J. Great Lakes Res. 36, 86–92. Lobel, P.B., Belkholde, S.P., Jackson, S.E., Longerich, H.P., 1991. Sediment in the intestinal tract: a potentially serious source of error in aquatic biological monitoring programs. Mar. Environ. Res. 31, 163–174. Lum, K.R., Kaiser, K.L.E., Jaskot, C., 1991. Distribution and fluxes of metals in the St. Lawrence River from the outflow of Lake Ontario to Quebec City. Aquatic Sciences 53, 1–19. Luoma, S.N., 1989. Can we determine the biological availability of sediment-bound trace elements? Hydrobiologia 176 (177), 379–396. Luoma, S.N., Carter, J.L., 1991. Effects of trace metals on aquatic benthos. In: Newman, M.C., McIntosh, A.W. (Eds.), Metal ecotoxicology—concepts and applications. Lewis Publishers, Inc., Chelsea, pp. 261–300. Luoma, S.N., Rainbow, P.S., 2005. Why is metal bioaccumulation so variable? Biodynamics as a unifying concept. Environ. Sci. Techn. 39, 1921–1931. Luoma, S.N., Rainbow, P.S., 2008. Metal contamination in aquatic environments—science and lateral management. Cambridge University Press, New York. Madenjian, C.P., Pothoven, S.A., Schneeberger, P.J., Ebener, M.P., Mohr, L.C., Nalepa, T.F., Bence, J.R., 2010. Dreissenid mussels are not a “dead end” in Great lakes food webs. J. Great Lakes Res. 36, 73–77. Metcalfe, J.L., Charlton, M.N., 1990. Freshwater mussels as biomonitors for organic industrial contaminants and pesticides in the St. Lawrence River. Sci. Tot. Environ. 97/98, 595–615. Mills, E.L., Roseman, E.F., Rutzke, M., Gutemann, W.H., Lisk, D.J., 1993. Contaminant and nutrient element levels in soft tissues of zebra and quagga mussels from waters of southern Lake Ontario. Chemosphere 27, 1465–1473. Mills, E.L., Rosenberg, G., Spidle, A.P., Ludyanskiy, M., Pligin, Y., May, B., 1996. A review of the biology and ecology of the quagga mussel (Dreissena bugensis), a second

146

C. Johns / Journal of Great Lakes Research 37 (2011) 140–146

species of freshwater dreissenid mussel introduced to North America. Amer. Zool. 36, 271–286. Mills, E.L., Chrisman, J.R., Baldwin, B., Owens, R.W., O'Gorman, R., Howell, T., Rosemand, E.F., Raths, M.K., 1999. Changes in the dreissenid community in the lower Great Lakes with emphasis on southern Lake Ontario. J. Great Lakes Res. 25, 187–197. Mills, E.L., Casselman, J.M., Dermott, R., Fitzsimons, J.D., Gal, G., Holeck, K.T., Hoyle, J.A., Johannsson, O.E., Lantry, B.F., Makarewicz, J.C., Millard, E.S., Munawar, I.F., Munawar, M., O'Gorman, R., Owens, R.W., Rudstam, L.G., Schaner, T., Stewart, T.J., 2003. Lake Ontario: food web dynamics in a changing ecosystem (1970–2000). Can. J. Fish. Aquat. Sci. 60, 471–490. Morin, J., Mingelbier, M., Bechara, J.A., Champoux, O., Secretan, Y., Jean, M., Frenette, J.-J., 2003. Emergence of new explanatory variables for 2D habitat modeling in large rivers: the St. Lawrence experience. Can. Water Res. J. 28, 249–272. New York Sea Grant, 2002. National Aquatic Nuisance Species Clearinghouse (maps). http://cce.cornell.edu/aquaticinvaders/nan_maps.cfm2002. Phillips, D.J.H., Rainbow, P.S., 1994. Biomonitoring of trace aquatic contaminants, second edition. Chapman and Hall, London. Rainbow, P.S., Smith, B.D., Lau, S.S., 2002. Biomonitoring of trace metal availabilities in the Thames estuary using a suite of littoral biomonitors. J. Mar. Biol. Assoc. UK 82, 793–799. Richman, L., Somers, K., 2010. Monitoring metal and persistent organic contaminant trends through time using quagga mussels (Dreissena bugensis) collected from the Niagara River. J. Great Lakes Res. 36, 28–36. Richman, L., Somers, K., 2005. Can we use zebra and quagga mussels for biomonitoring contaminants in the Niagara River? Water Air Soil Pollut. 167, 155–178. Roditi, H.A., Caraco, N.F., Cole, J.J., Strayer, D.L., 1996. Filtration of Hudson River water by the zebra mussel (Dreissena polymorpha). Estuaries 19, 824–832. Roditi, H.A., Fisher, N.S., 1999. Rates and routes of trace element uptake in zebra mussels. Limnol. Oceanogr. 44, 1730–1749. Roditi, H.A., Fisher, N.S., Sanudo-Wilhelmy, S.A., 2000a. Uptake of dissolved organic carbon and trace elements by zebra mussels. Nature (London) 407, 78–80.

Roditi, H.A., Fisher, N.S., Sanudo-Wilhelmy, S., 2000b. Field testing a bioaccumulation model for zebra mussels. Environ. Sci. Techn. 34, 2817–2825. Roseman, E.F., Mills, E.L., Rutzke, M., Gutenmann, W.H., Lisk, D., 1994. Absorption of cadmium from water by North American zebra and quagga mussels (Bivalvia: Dreissenide). Chemosphere 28, 737–743. Rutzke, M.A., Gutenmann, W.H., Lisk, D.J., Mills, E.L., 2000. Toxic and nutrient element concentrations in soft tissues of zebra and quagga mussels from Lakes Erie and Ontario. Chemosphere 40, 1353–1356. Stevens, R.J.J., Neilson, M.A., 1987. Response of Lake Ontario to reductions in phosphorus load, 1967–1982. Can. J. Fish. Aquat. Sci. 44, 2059–2068. Stoeckmann, A., 2003. Physiological energetics of Lake Erie dreissenid mussels: a basis for displacement of Dreissena polymorpha by Dreissena bugensis. Can. J. Fish. Aquat. Sci. 60, 126–134. Stokstad, E., 2007. Feared quagga mussel turns up in western United States. Science 315, 453. Strong, C.R., Luoma, S.N., 1981. Variations in the correlation of body size with concentrations of Cu and Ag in the bivalve Macoma balthica. Can. J. Fish. Aquat. Sci. 38, 1059–1064. Tessier, A., Campbell, P.G.C., Auclair, J.C., Bisson, M., 1984. Relationships between the partitioning of trace metals in sediments and their accumulation in the tissues of the freshwater mollusk Elliptio camplanata in a mining area. Can. J. Fish. Aquat. Sci. 41, 1463–1471. Twiss, M., 2007. Wither the Saint Lawrence River? J. Great Lakes Res. 33, 693–698. United States Geological Survey, National Water information system; web interface. USGS 04264331 St. Lawrence River at Cornwall, Ont. near Massena NY. Accessed July 11, 2010. Vanderploeg, H.A., Nalepa, T.F., Jude, D.J., Mills, E.L., Holeck, K.T., Liebig, J.R., Grigorovich, I.A., Ojaveer, H., 2002. Dispersal and emerging ecological impacts of Ponto-Caspian species in the Laurentian Great Lakes. Can. J. Fish. Aquat. Sci. 59, 1209–1228. Wang, W.-X., Yan, Q.-L., Fan, W., Xu, Y., 2002. Bioavailability of sedimentary metals from a contaminated bay. Mar. Ecol. Progr. Ser. 240, 27–38. Zar, J., 1999. Biostatistical analysis, 4th. Prentice Hall, New York.