Dreissena population status in nearshore Lake Ontario

Journal of Great Lakes Research 38 (2012) 161–170

Contents lists available at SciVerse ScienceDirect

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

Dreissena population status in nearshore Lake Ontario C.M. Pennuto a,⁎, E.T. Howell b, T.W. Lewis c, 1, J.C. Makarewicz c, 2 a b c

Great Lakes Center & Biology Department, Buffalo State College, Buffalo, NY 14222, USA Environmental Monitoring and Reporting Branch, Ontario Ministry of the Environment, Etobicoke, ON, Canada M9P 3V6 Department of Environmental Science and Biology, The College at Brockport, State University of New York, Brockport, NY 14420, USA

a r t i c l e

i n f o

Article history: Received 17 May 2011 Accepted 17 April 2012 Available online 24 May 2012 Communicated by David Barton Keywords: Zebra mussel Quagga mussel Benthic correlates Lake Ontario

a b s t r a c t Dreissenid mussels are ecosystem engineers in the Great Lakes, affecting benthic and water column communities and production. We surveyed mussel populations at four Canadian and three U.S. locations in summer 2008 to update population status and examine correlations with water column data. We measured mussel length, density, shell-free dry mass (SFDM), condition index, and phosphorus content of both shells and mussel tissue. The water column variables of chlorophyll a, turbidity, and total phosphorus (TP) were correlated with each other lake-wide, but exhibited only a few correlations with mussel metrics within seasons or shorelines. Quagga mussels (Dreissena rostriformis bugensis) represented ~ 99% of the mussel community in nearshore collections. Mussel length declined in a west-to-east direction and increased with depth in both U.S. and Canadian nearshore waters. Mussel density declined west-to-east in U.S. water, but exhibited no difference among sites in Canadian waters. Mussel condition index and phosphorus concentrations were correlated and increased west-to-east within the U.S. nearshore. There were significant declines in both tissue and shell P content with season in U.S. mussels, but no clear patterns in Canadian mussels. We estimated there were 9.7 × 1012 mussels (mean = 3402.9/m 2) in the Lake Ontario nearshore totalling 1.2 × 10 5 mT of mussel tissue which could filter the entire Lake Ontario nearshore volume (0–20 m depth = 30.9 km 3) in roughly 1 to 7 days. It appears that mussel density has declined since the last large surveys of 5 or 10 years ago (Canadian nearshore or U.S. nearshore, respectively), however the data were either only slightly supportive of, or showed no support for, food limitation or goby predation as the most parsimonious explanation for the decline in mussel abundance. © 2012 International Association for Great Lakes Research. Published by Elsevier B.V. All rights reserved.

Introduction Dreissenid mussels [zebra mussel (Dreissena polymorpha) and quagga mussel (D. rostriformis bugensis)] function as ecosystem engineers in the Great Lakes. Since their arrival in the mid 1980s, these mussels have been implicated in altering substrate habitat conditions (Botts et al., 1996; Ricciardi et al., 1997; Vanderploeg et al., 2002), increasing water clarity via their filtering capacity (Barbiero et al., 2006; Strayer et al., 1999), facilitating the invasion success of subsequent Ponto-Caspian invaders (Ricciardi, 2001), altering the composition of the benthic community (Barton et al., 2005; Haynes et al., 2005; Kuhns and Berg, 1999; Lederer et al., 2006), and redirecting pelagic nutrients to the nearshore benthic zone (Hecky et al., 2004). Recent survey data from across the Great Lakes suggest that Dreissena populations may have reached their

⁎ Corresponding author. Tel.: + 1 716 878 4105. E-mail addresses: [email protected] (CM. Pennuto), [email protected] (ET. Howell), [email protected] (TW. Lewis), [email protected] (JC. Makarewicz). 1 Tel.: + 1 585 395 5746. 2 Tel.: + 1 585 395 5747.

peak densities in some locations, but expansion continues into deeper waters in others (Nalepa et al., 2010). Additionally, data indicate quagga mussels have nearly replaced zebra mussels in most locations throughout the Great Lakes (Barton et al., 2005; Jarvis et al., 2000; Mills et al., 1999; Nalepa et al., 2010; Wilson et al., 2006). In particular, Wilson et al. (2006) surveyed the north shore of Lake Ontario in 2003 and reported a nearly 100% quagga mussel community, representing a 50% change in the percent composition of dreissenids in just 8 years (Kilgour et al., 2000). Thus, although dreissenids remain an important structuring force of the benthic zones of the Great Lakes, there are still community changes occurring, and sometimes on a rapid time scale. As part of a large, coordinated international sampling effort, the water quality and benthic community of the Lake Ontario nearshore zone was assessed in summer 2008 (Makarewicz et al., 2012a). In particular, dreissenid mussel community composition, density, size distribution, and biomass were determined across both the north and south shores for comparison with 2003 and 1995 collections, respectively. In 1995, zebra and quagga mussels each represented about 50% of the mussel community along the north shore (Marvin et al., 2000), whereas by 2000, quagga mussels accounted for nearly 100% of the mussels in the southern nearshore zone (Haynes et al.,

0380-1330/$ – see front matter © 2012 International Association for Great Lakes Research. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.jglr.2012.04.002

162

CM. Pennuto et al. / Journal of Great Lakes Research 38 (2012) 161–170

Table 1 Water column characteristics of sampling polygons. Values are means (1st. err.) from a combination of CTD profiles (Makarewicz et al., 2012c) and fluoroprobe tows (Pavlac et al., 2012) throughout each polygon in the U.S. nearshore, and as described in Howell et al. (2012) in the Canadian nearshore. Date

Site

Surf temp (°C)a

Bott temp (°C)b

Turbidity (NTU)c

TP (μg/L)d

Chl a (μg/L)c

May/June

Oak Orchard Rochester Mexico Bay Grimsby Torontoe Ajax Cobourg Oak Orchard Rochester Mexico Bay Grimsby Torontoe Ajax Cobourg

12.3 (0.67) 12.9 (1.66) 18.2 (1.02) 13.2 (0.8) 10.4 (1.1) 10.0 (1.7) 8.7 (1.2) 22.8 (0.59) 23.5 (0.36) 24.8 (0.65) 20.2 (0.7) 19.0 (0.6) 16.3 (1.6) 21.6 (0.1)

8.2 (2.68) 12.0 (0.31) 16.1 (0.24) 12.0 (2.6) 7.1 (1.8) 6.8 (2.6) 6.4 (1.4) 22.7 (0.69) 22.9 (0.17) 24.3 (0.82) 14.5 (4.5) 13.1 (4.6) 11.4 (3.6) 18.7 (4.5)

2.7 (0.72) 4.3 (0.86) 3.5 (0.91) 1.2 (1.1) 1.9 (1.6) 2.0 (3.1) 0.9 (0.4) 2.2 (0.12) 4.0 (1.04) 4.1 (0.70) 2.0 (1.6) 2.3 (2.9) 3.4 (7.2) 1.6 (1.8)

12.8 (1.4) 17.1 (4.0) 15.7 (2.3) 7.1 (9.8) 10.4 (15.3) 8.0 (7.3) 5.2 (1.0) 5.6 (0.7) 12.3 (1.4) 13.2 (2.2) 7.9 (4.1) 16.1 (18.0) 7.8 (1.4) 7.1 (4.3)

3.2 (0.3) 10.0 (5.2) 11.7 (8.5) 1.3 (1.2) 2.2 (0.75) 1.6 (1.0) 0.6 (0.3) 2.8 (0.2) 3.4 (0.9) 6.4 (2.2) 1.2 (0.7) 4.7 (0.6) 2.3 (0.6) 2.9 (1.1)

July/August

a Surface temperature, turbidity and chl a total are for full polygons extending from a depth of ~ 3 m to 5 km offshore. Estimates are means over the area derived from kriged surface‐interpolated from field surface measurements over survey tracks. b Bottom temp is the average of the minimum temperatures over profiles (averaged over profiles collected during a survey). c The May/June values are for a single survey (CA water quality survey 2) collected in early June; estimates for July/August are for (CA water quality survey 3) conducted in late July/early August. d TP is average concentration among discrete surface samples collected over the full polygon. e The Toronto polygon is split between two sub-polygons over the Greater Toronto area (GTA Centre and GTA West: refer to 6). The reported values are the averages over the two polygons.

2005). This transition in dominance is similar to observations made in southern Lake Michigan by Nalepa et al. (2010) and has been attributed to differences in filtering efficiency at low food levels (Baldwin et al., 2002; Diggins, 2001), lower respiration requirements (Stoeckmann, 2003), greater tolerance for cold water by D. r. bugensis (Diggins, 2001; Vanderploeg et al., 2010), or a slightly larger size attained by quagga mussels relative to zebra mussels (Patterson et al., 2005; Wilson et al., 2006). Regardless of the mechanism(s) which have allowed D. r. bugensis to displace D. polymorpha, mussel populations are still very large and management decisions linked to ecosystem changes require up-to-date assessments of population size, size structure, and distribution. Large spatial scale assessments of dreissenid distribution and abundance, coupled with information on chemical and biological properties (e.g., phosphorus concentrations and seston abundance) may provide new insights into mechanisms important in regulating mussel populations. Predation by round gobies (Neogobius melanostomus), food limitation, and possibly upwelling events were all proposed by Wilson et al. (2006) as likely mechanisms controlling mussel populations in Lake Ontario. Multiple lines of evidence, both manipulative and correlative, indicate that round gobies are capable of altering the local size distribution of mussels via direct predation (Barton et al., 2005; Campbell et al., 2009; Djuricich and Janssen, 2001; Johnson et al., 2005; Lederer et al., 2006; Ray and Corkum, 1997). This benthic-feeding fish exhibits a shift to nearly strict molluscivory somewhere between 7 and 10 cm TL (Barton et al., 2005; Campbell et al., 2009; Ghedotti et al., 1995; Janssen and Jude, 2001; Lederer et al., 2006; Ray and Corkum, 1997), as evidenced by gut content and stable isotope analyses. The preferred prey is dreissenid mussels. In contrast, there is little evidence for lake-wide control of mussels by round gobies. Bunnell et al. (2005) and Johnson et al. (2005) have suggested that round gobies consume a very small fraction of available mussel tissue in Lake Erie, and Pennuto et al. (2012) indicated there were only weak correlations between round goby abundance or size and dreissenid mussel benthic density, valve length, or biomass across the nearshore of Lake Ontario. Dreissenid mussels have been implicated in whole lake nutrient dynamics. Nearshore and offshore nutrient environments are linked by lake hydrodynamics and biological interactions like fish movements, and possibly mussel activity. Hecky et al. (2004) proposed a nearshore shunt hypothesis to explain the redirection of

water column nutrients to the benthic zone as a function of dreissenid mussel filtering and recycling. This hypothesis provides a mechanism to explain why nearshore nutrient levels have not declined as rapidly as offshore nutrient levels, while also providing a mechanism for a recent resurgence in the abundance and coverage of the benthic alga Cladophora sp. in the Great Lakes (Depew et al., 2011; Higgins et al., 2008). Several authors have indicated that dreissenid mussels effectively increase light penetration depth as a result of their feeding activity (Barbiero et al., 2006; Strayer et al., 1999), provide new attachment site substrate for algae (Botts et al., 1996; Hecky et al., 2004), and excrete high levels of dissolved nutrients (Arnott and Vanni, 1996; Ozersky et al., 2009), all contributing to conditions favorable for Cladophora growth. The combination of mussels and benthic algae might serve to intercept nutrients prior to their reaching open waters. The nearshore shunt hypothesis presumes that waters and nutrients entering a lake get mixed evenly with waters throughout the lake (Hecky et al., 2004). If so, it suggests a nearshore–offshore gradient in water column conditions should result since particulates and nutrients arrive to the lake via tributaries and erosion in the nearshore and then they are removed via dreissenid mussel filtering before mixing with offshore waters. In particular, total phosphorus (TP), turbidity, and chlorophyll a should decline in an offshore direction due to mussel filtering (from assumptions 1–5 in Table 1 of Hecky et al., 2004). However, even in the absence of mussel filtering, at some scale a nearshore-to-offshore gradient in water column nutrients and particulates should be excepted since they do arrive to the lake near shore before mixing with offshore waters. If lake hydrodynamics or nearshore-offshore density differences create along-shore currents or otherwise prevent or reduce nearshore waters from mixing with the offshore, there will still be a nearshore-to-offshore gradient in nutrients and particulates (e.g., Makarewicz and Howell, 2009; Neilson and Stevens, 1987; Rao et al., 2004). Hecky et al. (2004) also suggest it is possible for mussel populations to self-control at high densities as a result of food limitation because of their high filtration efficiency. Thus, as one moves offshore mussels might exhibit signs of food limitation, potentially manifested as a reduction in body mass or condition index. Collectively, we expect to see some correlative evidence of round goby predation or food limitation effects on mussel populations in the nearshore zone of Lake Ontario. Here we quantify

CM. Pennuto et al. / Journal of Great Lakes Research 38 (2012) 161–170

the density, size distribution, biomass, and condition index of dreissenid mussels for comparisons with past estimates of mussel abundance and examine correlations between the nearshore mussel population and environmental conditions. We define the nearshore zone as that area of the lake from the shoreline to a depth of 20 m (see Makarewicz et al., 2012a). Methods Field protocols As part of the Lake Ontario Nearshore Nutrient Study (LONNS; see Makarewicz et al., 2012c), dreissenid mussel abundance, biomass, and size distribution were estimated at four Canadian and three U.S. locations (CAN: Grimsby, Toronto, Ajax, and Cobourg; U.S.: Oak Orchard, Rochester, and Mexico Bay) (Fig. 1) in two seasons, spring (June) and summer (July/August), 2008. Each Canadian sampling area was defined as approximately 5 × (13–23) km polygons whereas U.S. sampling polygons were 5 × 20 km. Duplicate shoreline-tooffshore transects were sampled in each CAN polygon and triplicate transects were sampled in U.S. polygons at fixed depth stations (CAN: 3, 6, 9, and 18 m; U.S.: 2, 5, 10, and 20 m). Transects in Canadian waters were focused on sites with hard substrates following maps of Rukavina (1976). U.S. transects were sited based on an expectation of hard substrates from aerial imaging and professional knowledge. However, aerial images did not accurately reflect substrates at depth and thus a range of substrate types was encountered and sampled. At each depth station, replicate air-lift (CAN: n = 3, 0.15 m 2 quadrat; U.S.: n = 5, 0.25 m 2 quadrat) samples were collected by divers into fine-meshed bags to estimate Dreissena population density (#/m 2), size distribution, and biomass (g/m 2). Divers scraped the surface of all areas within the quadrat while applying air-lift pressure to suck up dislodged material. Air-lift samplers were powered by a SCUBA tank connected to a 10-cm diameter, Schedule 40 PVC standpipe (1.5-m tall) with a 0.5-mm mesh Nitex bag attached. Shipside, sample bags were emptied into white sorting trays where Cladophora was removed using forceps and floatation, placed in sealable plastic bags, and frozen. The remaining mussels, invertebrates, and sediment were placed in sealable plastic bags and frozen until further analysis. A separate sample was collected to estimate phosphorus content in mussels. Divers scraped a random mussel sample (~ 25 mussels)

163

from an area adjacent to each quadrat, pooled them, and placed in pre-labeled plastic bags. These were frozen shipside and later sent to the Water Chemistry Laboratory at The College at Brockport, State University of New York (NELAC–EPA Lab Code # NY01449) (U.S. samples) or to Ontario Ministry of the Environment laboratory (Canadian samples). Total phosphorus in mussel tissue and mussel shells was determined. General water quality features were collected by Makarewicz et al. (2012b) and Pavlac et al. (2012) for U.S. waters, and by Howell et al. (2012) for Canadian waters. Parameters included surface and bottom temperature, turbidity, total phosphorous (TP), and chlorophyll a. Laboratory methods Mussel samples from U.S. waters were thawed and empty shells, invertebrates, and debris were removed. Samples were distributed evenly into gridded sorting trays and random grid locations were selected, counting and measuring (nearest 1 mm, maximum valve length, calipers) every mussel within the grid under visual inspection. This procedure was repeated until 100 animals or the entire sample, if less than 100, was counted. Mussel density (#/m 2) and size frequency distributions were determined from these data. Biomass of mussel soft tissue (shell-free dry mass, SFDM, g/m 2) was determined after counting and measuring was completed. Measured mussels were placed into four size categories (b5, 5–10, 10–20, >20 mm) as they were measured. A random subsample consisting of five of the 10–20 and >20-mm mussels were weighed individually and a random subsample of up to 10 individuals in the b5 and 5–10 mm size categories were pooled and weighed. Shell-free dry mass (SFDM) was then obtained. All tissue was removed from the shell and placed in a pre-tared weigh boat, weighed (tissue wet weight), and freezedried (− 40 °C) to a constant weight. SFDM was found by subtraction. For Canadian samples, mussels were freeze-dried from a frozen state using a Labconco FreeZone 6Plus with bulk tray dryer prior to enumeration and chemical analysis. Mussel samples were sub-sampled by weight for determination of numbers of individuals and whole mussel biomass when numbers exceeded 700 per sample by spreading samples over a tray and randomly selecting quadrants; a minimum of 500 individuals were counted. Samples were processed with the aid of a dissecting microscope. Of the 500 counted for each site, at least 100 randomly selected individuals were measured to the nearest 1 mm for determination of size distribution. Of the 100 mussels used for

Fig. 1. Sampling polygons for assessing dreissenid populations in nearshore Lake Ontario, 2008.

164

CM. Pennuto et al. / Journal of Great Lakes Research 38 (2012) 161–170

determining length distributions, a minimum of 30 random individuals per sample were dissected to determine soft tissue to shell mass ratios used to estimate sample soft tissue and shell biomass from whole mussel biomass. For phosphorus analysis in U.S. mussels, whole mussels were first cleaned of debris and external algae and weighed. Shells were scrubbed with a soft bristle brush, placed in an ultrasonic bath to remove any attached algae, rinsed again with de-ionized water, dried at 60 °C in a drying oven, and weighed. Dried shells were then ground to a fine powder with a mortar and pestle. Mussel tissue was removed from the shell and rinsed in three successive baths of de-ionized water, freeze-dried (Virtis Freeze, Model 10-145MR-BA) to a constant weight, and ground in a mortar and pestle prior to digestion for phosphorus analysis. Phosphorus analysis followed the automated ascorbic acid reduction method (APHA 4500-P F) after tissues and shells were digested by the sulfuric acid–nitric acid digestion procedure (Method 4500-P, APHA 1999). Total phosphorus and Kjeldahl nitrogen analyses for Canadian mussels were determined by colorimetry using OMOE method E3116 (OMOE 2011) on acid-digested samples of mussel soft tissue and shells subsampled randomly from the quantitative freeze-dried samples above. Shells were cleaned of debris and rinsed in de-ionized water prior to analysis. Loss on ignition of freeze-dried samples was determined using OMOE method E3139 (OMOE 2010). Mussel health in U.S. waters was assessed using a condition index based on the relationship between mussel tissue mass and shell size: CI = tissue dry mass (g) ∗ 1000 ÷ (whole animal wet mass − shell dry mass) (Nalepa et al., 2010). Condition index values were not generated for CAN samples. Data handling Date, site, and depth differences in mussel benthic density, length, condition index (CI), and phosphorus content were examined using ANOVA procedures on means after pooling transects within sites, resulting in a total of 32 Canadian and 24 U.S. data points. Density data were either 4th-root (U.S.) or log-transformed (CAN) to meet variance and normality assumptions (Levene's test and Shapiro– Wilk W tests P b 0.05). Mussel length data for both shores and condition index data for U.S. mussels met variance and normality assumptions, requiring no transformations prior to analyses. Canadian data for phosphorous content in mussel tissue met variance and normality assumptions whereas U.S. data for tissue P content required a 4th-root transformation. Shell phosphorous data for both CAN and U.S. mussels met variance and normality assumptions following the use of Harter's rank order transformation (Harter 1961). For the mussel length analysis in U.S. nearshore waters, the Mexico Bay, 2-m depth values in both June and August, as well as the Oak Orchard, 20-m depth value in June were estimated using the technique of Anderson (1946) to ensure a balanced design. Similarly, techniques of Anderson were used to estimate the Grimsby, 3-m mussel density value in July for design balance. For all analyses, U.S. and Canadian data were examined separately due to differences in depth collections and quadrat sizes. Country differences in mussel density, length, and P content were examined using a t-test on pooled site, depth, and season data. Analyses were conducted using Statistix 9.0 software with an α = 0.05. The correlations between P content of tissues and CI were examined using Pearson product moment correlations. Relationships between mussel metrics of mean length and mean density and water column variables (TP, turbidity, and chl a) also were examined with Pearson product moment correlations. Estimates of total biomass of mussels throughout the nearshore environment were generated by determining the average depth-specific biomass (per m 2) across all Canadian sites and all U.S. sites, and multiplying by the area bound by the 0-to-20 m depth contour. Estimates of nearshore clearance

rates by mussels were determined using the mean filtering rate of 0.02 to 0.125 L/mussel/h as reported for quagga mussels in Baldwin et al. (2002) for mussels obtained in the St. Lawrence River and Lake Erie. Contour area and the respective volume for the 0-to-20 m depth zone were obtained from Virden et al. (2000). Results Water chemistry results reflected land use features adjacent to each sampling polygon (see Makarewicz et al. (2012a, 2012b, 2012c) and Howell et al. (2012) for details). On the U.S. side of the lake, turbidity, TP, and chl a generally increased from west-to-east, whereas they decreased along the Canadian nearshore. Mean chl a steadily increased west-to-east in June and August in U.S. waters, whereas CAN data exhibited a general decrease in June and an increase in August. Chl a levels were roughly 7× higher along the U.S. nearshore in the spring and about 2× higher in the late summer compared to the CAN nearshore. This coincided with both higher surface and bottom temperatures and higher TP concentration in surface waters in U.S. waters relative to CAN waters in both sampling periods (Table 1). Mean TP increased from west-to-east on the U.S. shore in both June and August, with a spike occurring in June in the Rochester polygon. In Canadian waters, TP values exhibited a decline from west-to-east in both June and July with a spike in the Toronto polygon (Table 1). Turbidity levels also showed an increase from west-to-east in U.S. waters in both months, with a June spike within the Rochester polygon, whereas turbidity declined west-to-east in the Canada nearshore with a spike near Toronto (Table 1). The U.S. nearshore exhibited a higher mean turbidity value in both sampling periods (3.5 vs 1.5 in June and 3.4 vs 2.3 in July/August for U.S. and CAN waters, respectively). The three water column metrics all were correlated with each other on a lake-wide basis. Chlorophyll a levels were correlated with turbidity levels (r = 0.689, n = 14, P = 0.006) and with TP concentrations (r = 0.684, n = 14, P = 0.007), and

Table 2 Substrate type among sampling polygons. Each ‘X’ represents approximately 25% coverage. The data are substrate types of the highest coverage for each transect and date pooled within polygons. Location

Depth (m)

Oak Orchard

2 5 10 20 2 5 10 20 2 5 10 20 3 6 10 18 3 6 10 18 3 6 10 18 3 6 10 18

Rochester

Mexico Bay

Grimsby

Toronto

Ajax

Cobourg

Notes: 1 — hard packed clay.

Sand

Clay/mud/silt

X

Gravel/cobble/ boulder

Bedrock

XX XXX XXX

XX X

XXXX XXX XX XX X XXX XXX XX X

X XX XX XXX X X X

XX X

X1

XX XXXX XX X X X XXX XX XXXX XXXX

X1X1X1

X

X X XX XX

X X XX XX XX X XX XX X XXXX X

X XXX XXX XX X

CM. Pennuto et al. / Journal of Great Lakes Research 38 (2012) 161–170

165

Fig. 2. Mean (± s.e.) dreissenid seasonal valve length (mm), density (no./m2), and shell-free dry mass (SFDM) (g/m2) at different depths in the Lake Ontario nearshore environment, 2008. Panels A, C, and E = United States and panels B, D, and F = Canada. Means without error bars indicate a single sample.

turbidity and TP were correlated (r = 0.707, n = 14, P = 0.005). There also were differences in physical habitat conditions between the CAN and U.S. nearshore environments. U.S. and CAN polygons differed in the amount of hard substrate (i.e., cobble, boulder, and bedrock) encountered and sampled (Table 2). Bedrock was the most prevalent substrate encountered in CAN waters (dominant at 24 of 36 depth stations), whereas sand and cobble were roughly co-dominant substrates in U.S. waters. CAN sampling targeted primarily hard substrates for mussels, whereas U.S. sampling included a broader array of substrate types, including sand. Mussel density was correlated with percent hard substrates in U.S. waters (r = 0.576, n = 12, P = 0.05). Quagga mussels dominated the mussel community at all depths on both the north and south shores of Lake Ontario, comprising >99% of the mussels collected. In both Canada and the U.S., mussel length was significantly different among the sample polygons (CAN: F3,9 = 16.22, P = 0.001; U.S.: F2,6 = 17.67, P = 0.003; Fig. 2), and mussels declined in length from west-to-east throughout the nearshore. Grimsby mussels (western-most Canada site) were twice as long as Cobourg (eastern-most Canada site) mussels (15.0 vs 7.5 mm, respectively; P b 0.0125, adjusted Bonferroni). Similarly, Oak

Orchard mussels (western-most U.S. site) were significantly longer than Mexico Bay mussels (18.5 vs 12.6 mm, respectively; P b 0.017, adjusted Bonferroni). The overall mean mussel length was larger in U.S. waters compared to CA waters (16.3 vs 11.3 mm) when pooling mean lengths from each site × depth combination over both seasons (t = 4.36, df = 54, P b 0.001). Mussels also had longer shell lengths at the deeper sites compared to the shallow sites on both shores (CAN: F3,9 = 10.72, P = 0.003; U.S.: F2,6 = 11.33, P = 0.007; Fig. 2a,b). Mussels from 18-m depth were significantly longer than those from 6 m (14.6 vs 9.2 mm, respectively; P b 0.025, adjusted Bonferroni) in Canadian waters, and mussels from 20 m were longer than mussels from 5 m (19.1 vs 12.1 mm, respectively; P b 0.05; adjusted Bonferroni) in U.S. waters. There also was a significant season effect on mussel length in the U.S., but not the Canadian nearshore (F1,6 = 7.11, P = 0.037; Fig. 2a,b). In June, mussels were longer compared to August (17.5 vs 15.1 mm, respectively; P b 0.05, adjusted Bonferroni). Mussel length was not correlated with water column chl a in June in either nearshore habitat (r= 0.77 and 0.73 for U.S. and CAN, respectively, both P > 0.05), but there were significant negative correlations in late summer (August for U.S. data and July for CAN data; r = −0.991 and −0.989,

166

CM. Pennuto et al. / Journal of Great Lakes Research 38 (2012) 161–170

Fig. 3. Tissue and shell phosphorus content for Dreissena mussels collected in the Lake Ontario nearshore, 2008. Values are means (± s.e.). Panel A and C = United States and panels B and D = Canada. Means without error bars indicate a single sample.

respectively). There were no significant mussel length:turbidity or mussel length:TP correlations for either season on either shore (all P > 0.05). There was a significant site effect on mussel density in the U.S. nearshore environment, but not in the Canadian nearshore (CAN: F3,9 = 3.53, P > 0.05; U.S.: F2,6 = 20.35, P = 0.002; Fig. 2c,d). In U.S. waters, mussels were most dense in the Oak Orchard polygon (1421.2/m 2) and least dense in the Mexico Bay polygon with all substrates included (616.0/m 2, P b 0.017, adjusted Bonferroni). Depth had a significant effect on U.S. but not Canadian mussel density (CAN: F3,9 = 0.75, P > 0.05; U.S.: F3,6 = 15.43, P = 0.003; Fig. 2c,d) with mussels being less abundant at 2 m compared to deeper sites (P b 0.025). Mussel density showed a significant seasonal change along the CAN nearshore (F1,9 = 5.41, P = 0.045), whereas there was no season effect on mussel density in U.S. waters (F1,6 = 1.74, P > 0.05). There were no correlations between mussel density and chl a on either shore or in either season or between mussel density and TP or turbidity within the north shore (all P > 0.05). Within the southern nearshore, both turbidity and TP were significantly negatively correlated with mussel density in August and overall (August: r = − 0.998 and −0.999, n = 3, P = 0.040 and 0.028 for turbidity and TP, respectively, and Overall: r = −0.930 and −0.901, n = 6, P = 0.007 and 0.014 for turbidity and TP). Mussels were approximately 6× more dense in the Canadian nearshore compared to the U.S. nearshore (5848 vs 903 per m 2, CAN vs U.S., respectively; t = 7.38, df = 54, P b b 0.001) when combining all sites and dates. Since SFDM is a measure combining length and density of mussels, we did not examine differences statistically. SFDM trends were generally consistent with observations on abundances and shell length. In June, there was a west-to-east decline in SFDM/m 2 for most depths of the Canadian nearshore, but only at the 10-m depth sites on the U.S. side. Along the southern shore, SFDM declined from a high of 63.5 g/m 2 at the 10-m depth in the Oak Orchard polygon to 0.3 g/m 2 at the 2-m depth station in the Mexico Bay polygon. Values declined from 145.5 g/m2 at the Grimsby, 18-m station to 22.1 g/m2 at the 3-m Cobourg station (Fig. 2e,f). In August the patterns were not as clear for either shore. The U.S. data generally showed an

increase in SFDM with depth within each sampling polygon, except for a spike at 5 m in the Rochester site. SFDM in the Canadian nearshore showed a consistent increase with depth within each sampling polygon (Fig. 2e,f). There did not appear to be a consistent change in SFDM with location along either shore in August. Mussel tissue phosphorus concentrations in Canadian nearshore waters showed no seasonal, location, or depth differences, but there was a significant season effect on tissue P in U.S. nearshore waters (F1,6 = 14.44, P = 0.009; Fig. 3a,b). Mussels contained a higher concentration of P in their tissues in June compared to August (4.36 vs 3.21 mg/g, P b 0.025). There were no significant depth or site differences in tissue-P levels among the U.S. polygons. For U.S. data, phosphorus content in mussel shells also showed a significant date effect (F1,6 = 55.40, P b 0.001; Fig. 3c). June shell-P was

Fig. 4. Relationship between the condition index of dreissenid mussels and tissue phosphorus concentration for nearshore habitats along the U.S. shore of Lake Ontario in summer 2008.

CM. Pennuto et al. / Journal of Great Lakes Research 38 (2012) 161–170

Fig. 5. Mussel condition index (CI) in Lake Ontario nearshore waters of the United States in summer 2008.

significantly higher than August shell-P (861 vs 354 μg/g, P b 0.025). The Canadian data for shell-P content had two significant interaction terms (season × location: F3,3 = 12.24, P = 0.034; depth × location: F3,3 = 25.58, P = 0.012, respectively; Fig. 3d) making interpretation of any pattern difficult. Shell-P concentrations in mussels from the Grimsby and Toronto polygons (western-most locations) exceeded those at the Ajax and Cobourg polygons, and shell-P concentrations increased with depth in the western polygons but not the eastern polygons (Fig. 3d). Both tissue and shell phosphorous content exhibited a significant difference between U.S. and CAN nearshore zones. CAN mussels had higher tissue P (7.4 vs 3.9 mg/g; t = 6.57, df= 38, P ≪ 0.001), but lower shell P (278 vs 588 μg/g; t = 3.89, df= 38, P b 0.001) relative to U.S. mussels when all sites and dates were combined. Condition index values in U.S. waters were significantly correlated with tissue phosphorus levels in June (r = 0.679, n = 9, P = 0.044), but not in August (r = 0.368, P > 0.05; Fig. 4). There was a significant site effect on CI within the U.S. nearshore (F2,6 = 6.08, P = 0.036; 3way ANOVA on season × site × depth means), with the Rochester and Mexico Bay mussels having CI values significantly higher compared to the Oak Orchard mussels (P b 0.05, Bonferroni post-hoc test). CI values did not differ by date or depth within the U.S. nearshore (Fig. 5). We estimated that there were 9.7 × 10 12 mussels (mean = 3402.9 mussels/m 2 × 2850 km 2) in the Lake Ontario nearshore totalling 1.2 × 10 11 g (1.2 × 10 5 mT) of dry mussel tissue. On average, CAN mussels occurred at 5966.4 ± 563.9 (s.e.)/m 2 and U.S. mussels occurred at 839.4 ± 160.1 (s.e.)/m 2. The mean valve length of mussels throughout the nearshore was 13.6 ± 0.68 mm with a mean SFDM of 40.8 ± 4.66 g/m 2. Both of these values differed in CAN and U.S. waters. Mussel length in the U.S. nearshore had a mean dimension of 16.7 mm and a mean SFDM of 23.4 ± 4.23 g/m 2 whereas CAN mussels had a mean valve length of 11.5 mm and a mean SFDM of 54.3 ± 6.29 g/m 2. If we assume filtering rates of 0.02 to 0.125 L/mussel/h (Baldwin et al. 2002), then mussels covering the nearshore at our mean density could filter the entire Lake Ontario nearshore volume (0–20 m depth = 30.9 km 3) in 24.5 to 159.3 h or about 1 to 7 days. Discussion Dreissenid mussels were present throughout the entire nearshore environment of Lake Ontario and our samples were consistent with other recent large-scale sampling efforts which indicated dominance by quagga mussels (D. rostriformis bugensis) over zebra mussels (D. polymorpha) (e.g., Haynes et al., 2005; Nalepa et al. 2010; Wilson et al., 2006). Our abundance estimates support earlier research showing

167

a reduction in total numbers of dreissenids throughout the nearshore of Lake Ontario. Wilson et al. (2006) noted a decline in total dreissenid abundance from 1993 to 2003 in the Port Dalhousie area in western Lake Ontario, declining from 14,494 (Chase and Bailey, 1999) to 3984/m 2. The Grimsby site was our closest station to Port Dalhousie and it exhibited a density (5136/m 2) very similar to the 2003 value reported by Wilson et al. (2006). A station near Toronto declined from 12,111/m 2 in 2003 (Wilson et al., 2006) to 9620/m 2 in this study. Similarly, a site on the southern shore within the Oak Orchard polygon was sampled in '91, '99, and in our study in 2008 and showed a consistent decline in mean abundance on cobble or reef substrates (7398, 1835, and 1582/m 2, respectively) (Haynes et al., 2005; Stewart and Haynes, 1994). One bit of caution should be used when interpreting the Stewart and Haynes (1994) and Haynes et al. (2005) results for the U.S. nearshore. Those collections were made with a battery-powered pump providing the suction to a sampler which enclosed a standard area. Haynes et al. (2005) commented that this device probably underestimated dreissenid abundance since it did not remove all dreissenids attached to the substrate. Thus, their density estimates should be viewed as relative, not absolute abundances. This means our 2008 estimates are potentially an even greater decline from 1999 estimates, and collectively, suggests there has been a decline in mussel abundance but not necessarily a decline in every location. The large differences in mussel density between the north and south shores might reflect the higher percentage of sand substrate along the southern shoreline. Sand substrates are not conducive to mussel colonization or attachment. Food limitation, the shunt hypothesis, and Dreissena populations Although local nutrient inputs can have important implications for local benthic algal (e.g., Higgins et al., 2012), and invertebrate populations (e.g., Peterson et al., 2007), lake trophic state is a function of lake-wide nutrient concentrations and watershed conditions. Tributary nutrient inputs are generally the dominant contributor for lake-wide nutrient budgets in drainage lakes like Lake Ontario. Since the mid-1970s, offshore phosphorus concentrations (both TP and SRP) have declined dramatically (e.g., Malkin et al., 2010), prompting investigations into nearshore dynamics to help explain the decline in open water nutrients (e.g., Hecky et al., 2004; Makarewicz and Howell 2009). Lake Ontario receives approximately 9056 mT/year of TP from surface water sources (Makarewicz et al., 2012d), of which ~75% arrives via the Niagara River with Canadian tributaries contributing 9.4% and U.S. tributaries contributing the remaining 15.6%. Thus, the cumulative annual inputs of nutrients from tributary streams other than the Niagara River deliver a very small fraction of the TP to Lake Ontario, arriving along its northern or southern shores. That is not to say that tributary nutrient inputs are not significant to local regions within the lake, especially the nearshore. Makarewicz et al. (2012a, 2012b, 2012c) and Howell et al. (2012) show a combination of lake hydrodynamics, season, and tributaries exhibit significant control over water column nutrient conditions within the nearshore zone and Depew et al. (2011) showed a combination of local land cover, mussel abundance, and local water quality explained about 95% of the variability in benthic algal standing crop in the nearshore environment. Malkin et al. (2010) suggest that the north and south nearshore waters in Lake Ontario interact with offshore waters in a very different fashion. They posit that north shore waters mix efficiently with offshore waters whereas south shore waters mix to a lesser extent with offshore waters owing to a combination of prevailing winds, upwelling frequencies, and a central gyre. Since phytoplankton and suspended seston are the dominant food for filtering dreissenids, the decline in open water TP and turbidity relative to the nearshore in Lake Ontario might suggest mussels are redirecting these suspensoids to the benthic zone. However, if along

168

CM. Pennuto et al. / Journal of Great Lakes Research 38 (2012) 161–170

shore currents are significant and if offshore productivity is minimal, simple dilution effects might lead to the same gradients. The result for mussels might be reduced growth rates or other evidence of food limitation in an offshore or down-current direction. On both the north and south shores, mussel length declined from west-to-east, supporting a possible food limitation or nearshore trapping scenario coupled with dominant along shore currents described by Beletsky et al. (1999). However, mussel size and SFDM increased with depth. This observation does not support a food limitation hypothesis. The correlative nature of these observations does not allow a causative mechanism to be established. However, other investigations have documented an increase in Dreissena valve length with depth (e.g., Nalepa et al., 2010) and colder temperatures associated with deeper waters, reduced hydrologic sheer forces, or high food availability associated with deep chlorophyll peaks (e.g., Pemberton et al. 2007) have been proposed as potential mechanisms behind this observation. In Lake Ontario, no previous surveys have shown as large a spatial difference in mussel length as reported here. We suspect differences in mussel counting and measuring likely led to the observations reported. Canadian technicians scanned samples with the aid of a microscope whereas U.S. technicians counted mussels visually. Thus, U.S. counters may have missed the smallest size class (i.e., ≤1 mm), resulting in a larger apparent mussel length. Surprisingly, mussel condition index (CI) values were highest in the eastern-most sampling polygons in U.S. waters, suggesting that a food limitation hypothesis is not defensible for this part of the Lake Ontario nearshore. However, since the CI is basically a tissue mass:shell volume ratio (Nalepa et al., 2010), absolute mussel size will be important in determining calculated index values. Shell mass:tissue mass ratios are maximized at intermediate lengths, and minimized at the smallest and largest size classes (Karatayev et al., 1997). Thus, the larger mussels in the western end of the lake relative to the eastern end would be expected to show a lower CI all else equal. Tissue and shell P concentrations were significantly correlated with CI, increasing from west-to-east, although the correlation was only significant for shell P-level. Many organisms exhibit a sizedependent relationship with internal P content since smaller, more rapidly growing individuals tend to have higher P values associated with more active cell division compared to older individuals (Sterner and Elser, 2002). Although we did not age our mussels, the small mean size of mussels in eastern nearshore waters on both the U.S. and Canadian side of Lake Ontario suggests there should be an increase in P content in tissues of mussels to the east, which there was. The higher chl a and turbidity concentration observed in Mexico Bay relative to Oak Orchard also might be suggestive of a reduction in mussel filtering capacity. Collectively, the CI data and P in tissue data do not support a food limitation hypothesis for Lake Ontario dreissenids. The positive correlation between mussel CI and tissue P concentration in June probably reflects pre-reproductive condition. Naddafi et al. (2009) also found that tissue P content was correlated with CI in zebra mussels in the spring and a decline in CI in August and September was associated with an increase in both N:P and C:P ratios. Tissue and shell P content patterns are best explained by size differences in mussels and spawning. Both tissue P and shell P declined from June to August in U.S. waters and this is similar to the later summer P-content decline reported by Arnott and Vanni (1996) for mussels in Lake Erie and for the intertidal Modiolus demissus in a Georgia estuary (Kuenzler, 1961). Only shell-P declined with season in CAN waters. Claxton and Mackie (1998) reported a minimum spawning temperature of 9.0 °C for quagga mussels in eastern Lake Erie. Lake surface temperatures in Lake Ontario did not reach 9 °C until after the first of June in 2008, but remained greater than 9 °C into late October. Thus it is most parsimonious to explain the decline in tissue P as a reallocation to eggs/sperm associated with spawning after this

temperature was reached. The seasonal decline in shell P might reflect seasonal changes in shell production. Unionidae shell inner and outer periostracum layers begin as proteinacious deposits, especially conchiolin, prior to aragonite deposition near the shell edge (Kuenzler, 1961). These protein layers can also be secreted into the extrapallial fluid (Checa, 2000). Since many proteins have P atoms (Sterner and Elser, 2002), this may be a mechanism underlying the seasonal shell P loss. Alternatively, if shell growth exhibits a seasonal change in the proportion of various compounds other than P content (e.g., higher aragonite content), the concentration of P will appear to decline in concentration due to ‘dilution’. The increase in water column TP, turbidity, and chl a from westto-east in the southern nearshore and from east-to-west on the north shore might be explained by a combination of nearshore-wide mussel excretion, local nutrient inputs from tributary sources, and reduced mixing with offshore waters. Dreissenid mussels excrete significant quantities of dissolved nutrients (Arnott and Vanni, 1996; Ozersky et al., 2009). Ozersky et al. (2009) indicate that dreissenids excreted enough P to support local Cladophora growths along regions of the north shore and Naddafi et al. (2009) suggest dreissenids are less homeostatic in their tissue C:N:P ratios than many other organisms for which data are available. This indicates that food quality might not limit population sizes of these mussels. But, food quality and tissue ratios of N to P will affect the amount and nutrient ratios of excretion by-products (Naddafi et al., 2008, 2009; Sterner and Elser, 2002). Thus, the down-current increase in TP and chl a might reflect mussel excretion. In addition, local tributary inputs seem to have a significant effect on nearshore water column conditions. This is especially evident along the north shore, where all water column parameters examined (turbidity, chl a, and TP) spiked in the Toronto polygon. For a richer discussion of tributary impacts on local and nearshore water column chemistry see Makarewicz et al. (2012c) and Howell et al. (2012). In any event, these three variables were all correlated with each other throughout the nearshore. There were no significant correlations between any of these variables and mussel metrics within the northern nearshore, but within the U.S. nearshore, August chl a was negatively correlated with mussel size and both turbidity and TP were negatively correlated with mussel density, suggesting a possible mussel-influenced water column effect. Round gobies and Dreissena populations Round gobies larger than 70–100 mm TL are considered molluscivores (Ghedotti et al., 1995; Ray and Corkum, 1997; Skora and Rzeznik, 2001), and have been shown to prefer dreissenid mussels in the 4–10 mm size range. Size frequency distributions of nearshore mussels (data not shown) showed a reduction in this size class of mussels within the Canadian nearshore, but not in the U.S. nearshore. Pennuto et al. (2012) provide a fuller description of round goby population abundance and size structure throughout the U.S. nearshore, but report few correlations between dreissenid metrics and round goby metrics. Small crevices within the substrate, spaces between large mussels, and the undersides of rocks would all be likely predation refuges for small mussels sought by round gobies (e.g., Djuricich and Janssen, 2001). Thus, it is unlikely that round gobies will eliminate dreissenids from the Lake Ontario nearshore, although they may have some local impacts. This mirrors observations made by Bunnell et al. (2005) and Johnson et al. (2005) who indicate round gobies consume only a small fraction of Dreissena tissue throughout Lake Erie. Although population densities of dreissenid mussels were lower than recent regional surveys in Lake Ontario (e.g., Haynes et al., 2005; Wilson et al., 2006), they are still substantial. Lake-wide, we estimate that mussels occur at a density of ~3400 per m 2 when averaged across all habitat types. Collectively, these mussels are capable of filtering the nearshore volume in roughly 1–7 days, and

CM. Pennuto et al. / Journal of Great Lakes Research 38 (2012) 161–170

the entire lake volume in 58–365 days. This is well within the time reported for dreissenids to filter other large lakes around the world. Karatayev et al. (1997 and references therein) summarized the existing reports of dreissenid filtering in large lakes and reservoirs, and found populations of mussels filtered entire lake volumes in 5–45 days. With phytoplankton maximum growth rates ranging from a few hours to days (Kalff, 2002), it is unlikely that dreissenids will eliminate any but the slowest growing algal species in the nearshore environment. Additionally, the nearshore water column response to mussel filtering is likely to be affected by the timing and extent of offshore water exchange and tributary inputs. Nearshore waters may experience dilution events when there is high offshore exchange, or increased seston following storm run-off events. Thus, the nearshore response to mussel filtering may be stronger or weaker depending on circulation and weather-related conditions. We anticipate mussel filtering will continue to have water column implications (i.e., reduced seston abundance, increased water clarity) with potential impacts to higher trophic levels via reductions in energy availability. Acknowledgments Allyse Fischer and Caleb Basiliko were invaluable mussel collectors. Thanks to the BSC Great Lakes Center for ship time. This manuscript benefited from discussions with other members of the 2008 LONNS crew: J. Atkinson (University at Buffalo), G. Boyer (ESF), B. Edwards (Niagara University). Funding for this project was provided by a New York Department of Environmental Conservation (NYSDEC) grant to JCM. Comments from two anonymous reviewers greatly improved the manuscript. References Anderson, R.L., 1946. Missing-plot techniques. Biom. Bull. 2, 41–47. APHA, 1999. Standard Methods for the Examination of Waste and Wastewater. American Public Health Association, New York. Arnott, D.L., Vanni, M.J., 1996. Nitrogen and phosphorous recycling by the zebra mussel (Dreissena polymorpha) in the western basin of Lake Erie. Can. J. Fish. Aquat. Sci. 53, 646–659. Baldwin, B.S., Mayer, M.S., Dayton, J., Pau, N., Mendilla, J., Sullivan, M., Moore, A., Ma, M., 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. Barbiero, R.P., Tuchman, M.L., Millard, S.E., 2006. Post-dreissenid increases in transparency during summer stratification in the offshore waters of Lake Ontario: is a reduction in whiting events the cause? J. Great Lakes Res. 32, 131–141. Barton, D.R., Johnson, R.A., Campbell, L., Petruniak, J., Patterson, M., 2005. Effects of round gobies (Neogobius melanostomus) on dreissenid mussels and other invertebrates in eastern Lake Erie, 2002–2004. J. Great Lakes Res. 31 (Suppl. 2), 252–261. Beletsky, D., Saylor, J.H., Schwab, D.J., 1999. Mean circulation in the Great Lakes. J. Great Lakes Res. 25, 78–93. Botts, P.S., Patterson, B.A., Schloesser, D.W., 1996. Zebra mussel effects on benthic invertebrates: physical or biotic? J. N. Am. Benthol. Soc. 15, 179–184. Bunnell, D.B., Johnson, T.B., Knight, C.T., 2005. The impact of introduced round gobies (Neogobius melanostomus) on phosphorus cycling in central Lake Erie. Can. J. Fish. Aquat. Sci. 62, 15–29. Campbell, L.M., Thacker, R., Barton, D., Muir, D.C.G., Greenwood, D., Hecky, R.E., 2009. Re-engineering the eastern Lake Erie littoral food web: the trophic function of non-indigenous Ponto-Caspian species. J. Great Lakes Res. 35, 224–231. Chase, M.E., Bailey, R.C., 1999. The ecology of the zebra mussel (Dreissena polymorpha) in the lower Great Lakes of North America: I. Population dynamics and growth. J. Great Lakes Res. 25, 107–121. Checa, A., 2000. A new model for periostracum and shell formation in Unionidae (Bivalvia, Mollusca). Tissue Cell 32, 405–416. Claxton, W.T., Mackie, G.L., 1998. Seasonal and depth variations in gametogenesis and spawning of Dreissena polymorpha and Dreissena bugensis in eastern Lake Erie. Can. J. Zool. 76, 2010–2019. Depew, D.C., Houben, A.J., Guilford, S.J., Hecky, R.E., 2011. Distribution of nuisance Cladophora in the lower Great Lakes: patterns with land use. J. Great Lakes Res. 37, 656–671. Diggins, T.P., 2001. A seasonal comparison of suspended sediment filtration by quagga (Dreissena bugensis) and zebra (D. polymorpha) mussels. J Great Lakes Res 27, 457–466. Djuricich, P., Janssen, J., 2001. Impact of round goby predation on zebra mussel size distribution at Calumet Harbor, Lake Michigan. J. Great Lakes Res. 27, 312–318.

169

Ghedotti, M.J., Smihula, J.C., Smith, G.R., 1995. Zebra mussel predation by round gobies in the laboratory. J. Great Lakes Res. 21, 665–669. Harter, H.L., 1961. Expected values of normal order statistics. Biometrika 48, 151–165. Haynes, J.M., Tisch, N.A., Mayer, C.M., Rhyne, R.S., 2005. Benthic macroinvertebrate communities in southwestern Lake Ontario following invasion of Dreissena and Echinogammarus: 1983 to 2000. J. N. Am. Benthol. Soc. 24, 148–167. Hecky, R.E., Smith, R.E.H., Barton, D.R., Guildford, S.J., Taylor, W.D., Charlton, M.N., Howell, T., 2004. The near shore phosphorus shunt: a consequence of ecosystem engineering by dreissenids in the Laurentian Great Lakes. Can. J. Fish. Aquat. Sci. 61, 1285–1293. Higgins, S.N., Malkin, S.Y., Howell, E.T., Guilford, S.J., Campbell, L., Hiriart-Baer, V., Hecky, R.E., 2008. An ecological review of Cladophora glomerata (Chlorophyta) in the Laurentian Great Lakes. J. Phycol. 44, 839–854. Higgins, S.N., Pennuto, C.M., Howell, E.T., Lewis, T.W., Makarewicz, J.C., 2012. Urban influences on Cladophora blooms in Lake Ontario. J. Great Lakes Res. 38 (Suppl. 4), 116–123. Howell, E.T., Chomicki, K.M., Kaltenecker, G., 2012. Patterns in water quality on Canadian shores of Lake Ontario: Correspondence with proximity to land and level of urbanization. J. Great Lakes Res. 38 (Suppl. 4), 32–46. Janssen, J., Jude, D.J., 2001. Recruitment failure of mottled sculpin Cottus bairdi in Calumet harbor, Southern Lake Michigan, induced by the newly introduced round goby Neogobius melanostomus. J. Great Lakes Res. 27, 319–328. Jarvis, P., Dow, J., Dermott, R., Bonnell, R., 2000. Zebra (Dreissena polymorpha) and quagga mussel (Dreissena bugensis) distribution and density in Lake Erie, 1992–1998. Can. Tech. Rep. Fish. Aquat. Sci. 2304 46 pp. Johnson, T.B., Bunnell, D.B., Knight, C.T., 2005. A potential new energy pathway in Central Lake Erie: the round goby connection. J. Great Lakes Res. 31 (Suppl. 2), 238–251. Kalff, J., 2002. Limnology. Prentice Hall, Upper Saddle River, New Jersey. 592 pp. Karatayev, A.Y., Burlakova, L.E., Padilla, D.K., 1997. The effects of Dreissena polymorpha (Pallas) invasion on aquatic communities in Eastern Europe. J. Shellfish Res. 16, 187–203. Kilgour, B.W., Bailey, R.C., Howell, E.T., 2000. Factors influencing changes in the nearshore benthic community on the Canadian side of Lake Ontario. J. Great Lakes Res. 26, 272–286. Kuenzler, E.J., 1961. Phosphorus budget of a mussel population. Limnol. Oceanogr. 6, 400–415. Kuhns, L.A., Berg, M.A., 1999. Benthic invertebrate community responses to round goby (Neogobius melanostomus) and zebra mussel (Dreissena polymorpha) invasion in Lake Michigan. J. Great Lakes Res. 25, 910–917. Lederer, A., Massart, J., Janssen, J., 2006. Impact of round gobies (Neogobius melanostomus) on dreissenids (Dreissena polymorpha and Dreissena bugensis) and the associated macroinvertebrate community across an invasion front. J. Great Lakes Res. 32, 1–10. Makarewicz, J.C., Howell, E.T., 2009. Lake Ontario intensive year — 2008: the Lake Ontario coastal zone — status and assessment. Developing a Cooperative Monitoring Strategy for Lake Ontario: 2008 Intensive Year and Long-term Sampling Design. White Paper. LOLA Workshop. International Joint Commission, Kingston, Ontario. Makarewicz, J.C., Lewis, T.W., Boyer, G.L., 2012a. Nutrient enrichment and depletion on the shoreside of the spring thermal front. J. Great Lakes Res. 38 (Suppl. 4), 72–77. Makarewicz, J.C., Lewis, T.W., Boyer, G.L., Edwards, W.J., 2012b. The influence of streams on nearshore water chemistry, Lake Ontario. J. Great Lakes Res. 38 (Suppl. 4), 62–71. Makarewicz, J.C., Lewis, T.W., Pennuto, C.M., Atkinson, J.F., Edwards, W.J., Boyer, G.L., Howell, E.T., Thomas, G., 2012c. Physical and chemical characteristics of the nearshore zone of Lake Ontario. J. Great Lakes Res. 38 (Suppl. 4), 21–31. Makarewicz, J.C., Booty, W.G., Bowen, G.S., 2012d. Tributary phosphorus loading to Lake Ontario, J. Great Lakes Res. 38 (Suppl. 4), 14–20. Malkin, S.Y., Dove, A., Depew, D.C., Smith, R.E., Guildford, S.J., Hecky, R.E., 2010. Spatiotemporal patterns of water quality in Lake Ontario and their implications for nuisance growth of Cladophora. J. Great Lakes Res. 36, 477–489. Marvin, C.H., Howell, E.T., Reiner, E.J., 2000. Polychlorinated dioxins and furans in sediments at a site colonized by Dreissena in western Lake Ontario, Canada. Environ. Toxicol. Chem. 19, 344–351. Mills, E.L., Chrisman, J.R., Baldwin, B.S., 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. Naddafi, R., Eklöv, P., Pettersson, K., 2009. Stoichiometric constraints do not limit successful invasers: zebra mussels in Swedish lakes. PLoS One 4 (4), e5345. http://dx.doi.org/10.1371/journal.pone.0005345. Naddafi, R., Pettersson, K., Eklöv, P., 2008. Effects of the zebra mussel, an exotic freshwater species, on seston stoichiometry. Limnol. Oceanogr. 53, 1973–1987. Nalepa, T.F., Fanslow, D.L., Pothoven, S.A., 2010. Recent changes in density, biomass, recruitment, size structure, and nutritional state of Dreissena populations in southern Lake Michigan. J. Great Lakes Res. 36 (Suppl. 3), 5–19. Neilson, M.A., Stevens, R.J.J., 1987. Spatial heterogeneity of nutrients and organicmatter in Lake Ontario. Can. J. Fish. Aquat. Sci. 44, 2192–2203. OMOE, 2010. The determination of moisture content, RST, RSTA and LOI in solids by gravimetry. Method PHYSOLID-E3139 Ontario Ministry of the Environment, Laboratory Services Branch. OMOE, 2011. The Determination of Total Kjeldahl Nitrogen and Total Phosphorus in Soils Sediment and Sludge and Vegetation by Colourimetry. Method NPSED-E3116. PHYSOLID-E3139 Ontario Ministry of the Environment, Laboratory Services Branch. Ozersky, T., Malkin, S.Y., Barton, D.R., Hecky, R.E., 2009. Dreissenid phosphorous excretion can sustain C. glomerata growth along a portion of Lake Ontario shoreline. J. Great Lakes Res. 35, 321–328.

170

CM. Pennuto et al. / Journal of Great Lakes Research 38 (2012) 161–170

Patterson, M.W.R., Ciborowski, J.J.H., Barton, D.R., 2005. The distribution and abundance of Dreissena species (Dreissenidae) in Lake Erie, 2002. J. Great Lakes Res. 31 (Suppl. 2), 223–237. Pavlac, M.M., Smith, T.T., Thomas, S.P., Makarewicz, J.C., Edwards, W.J., Pennuto, C.M., Boyer, G.L., 2012. Assessment of phytoplankton distribution in the nearshore zone using continuous in situ flourometry. J. Great Lakes Res. 38 (Suppl. 4), 78–84. Pemberton, K.L., Smith, R.E.H., Silsbe, G.M., Howell, T., Watson, S.B., 2007. Controls on phytoplankton physiology in Lake Ontario during the late summer: evidence from new fluorescence methods. Can. J. Fish. Aquat. Sci. 64, 58–73. Pennuto, C.M., Howell, E.T., Makarewicz, J.C., 2012. Relationships among round gobies, Dreissena mussels, and benthic algae in the south nearshore of Lake Ontario. J. Great Lakes Res. 38 (Suppl. 4), 154–160. Peterson, G.S., Sierszen, M.E., Yurista, P.M., Kelly, J.R., 2007. Stable nitrogen isotopes of plankton and benthos reflect a landscape-level influence on Great Lakes coastal ecosystems. J. Great Lakes Res. 33 (Special Issue 3), 27–41. Rao, Y.R., Skafel, M.G., Charlton, M.N., 2004. Circulation and turbulent exchange characteristics during the thermal bar in Lake Ontario. Limnol. Oceanogr. 49, 2190–2200. Ray, W.J., Corkum, L.D., 1997. Predation of zebra mussels by round gobies, Neogobius melanostomus. Environ. Biol. Fishes 50, 267–273. Ricciardi, A., 2001. Facilitative interactions among aquatic invaders: is an “invasional meltdown” occurring in the Great Lakes? Can. J. Fish. Aquat. Sci. 58, 2513–2525. Ricciardi, A., Whoriskey, F.G., Rasmussen, J.B., 1997. The role of zebra mussels (Dreissena polymorpha) in structuring macroinvertebrate communities on hard substrates. Can. J. Fish. Aquat. Sci. 54, 2596–2608. Rukavina, N.A., 1976. Nearshore sediments of Lakes Ontario and Erie. Geosci. Can. 3, 185–190.

Skora, K.E., Rzeznik, J., 2001. Observations on diet composition of Neogobius melanostomus Pallas 1811 (Gobiida, Pisces) in the Gulf of Gdansk (Baltic Sea). J. Great Lakes Res. 27, 290–299. Sterner, R.W., Elser, J.J., 2002. Ecological Stoichiometry: The Biology of Elements from Molecules to the Biosphere. Princeton University Press, Princeton, NJ. 439 pp. Stewart, T.W., Haynes, J.M., 1994. Benthic macroinvertebrate communities of southwestern Lake Ontario following invasion of Dreissena. J. Great Lakes Res. 20, 479–493. Stoeckmann, A., 2003. Physiological energetic of Lake Erie dreissenid mussels: a basis for the displacement of Dreissena polymorpha by Dreissena bugensis. Can. J. Fish. Aquat. Sci. 60, 126–134. Strayer, D.L., Caraco, N.F., Cole, J.J., Findlay, S., Pace, M.L., 1999. Transformation of freshwater ecosystems by bivalves — a case study of zebra mussels in the Hudson River. Bioscience 49, 19–27. Vanderploeg, H.A., Nalepa, T.F., Jude, D.J., Mills, E.L., Holeck, K.T., Liebig, J.R., Gregorovich, 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. Vanderploeg, H.A., Liebig, J.R., Nalepa, T.F., Fahnenstiel, G.L., Pothoven, S.A., 2010. Dreissena and the disappearance of the spring phytoplankton bloom in Lake Michigan. J. Great Lakes Res. 36 (Suppl. 3), 50–59. Virden, W.T., Warren, J.S., Holcombe, T.L., Reid, D.F., 2000. Bathymetry of Lake Ontario CD-ROM, volume G2, version 1. Data Announcement 2000-MGG-01. National Geophysical Data Center, World Data Center for Marine Geology and Geophysics, Boulder. Wilson, K.A., Howell, E.T., Jackson, D.A., 2006. Replacement of zebra mussels by quagga mussels in the Canadian nearshore of Lake Ontario: importance of substrate, round goby abundance, and upwelling frequency. J. Great Lakes Res. 32, 11–28.