Hamilton Harbour is an Area of Concern in western Lake Ontario, long stressed by cultural eutrophication, urbanization and invasive species. Despite high nutrient levels leading to hypolimnetic hypoxia and contaminated sediment, it is a highly productive environment. To better understand zooplankton dynamics in Hamilton Harbour, we conducted biweekly May to October sampling of zooplankton and rotifer composition at open water harbour sites from 2002–2014. May to October zooplankton density, dry biomass and total production averaged 265 ± 16 animals l−1, 306 ± 19 mg m−3 and 4131 ± 359 mg m−3, (±SE), respectively. These values are among the highest reported in the Great Lakes, with biomass two to seven times greater than in other eutrophic embayments. Zooplankton populations and taxonomic seasonality have remained relatively stable in Hamilton Harbour since 2002. Biomass is often dominated by smaller taxa such as Bosmina, Eubosmina and juvenile copepods, suggestive of high fish planktivory, but Daphnia retrocurva and D. galeata mendotae are also dominant during the summer, indicating improvements in the zooplankton community since the 1970s when Daphnia and cyclopoids were uncommon. Conversely, rotifers have declined over the last 40 years, though while still numerically dominant, now comprise <4% of total biomass and production compared to 40% in the 1970s. Both adult Dreissenid Mussels and their veliger larvae are less abundant in Hamilton Harbour compared to nearshore Lake Ontario. Zooplankton appear to be effectively utilizing high production rates of edible algae and microorganisms in the harbour. More work is needed to explore trophic interactions in this eutrophic ecosystem and the effects of hypolimnetic hypoxia on the zooplankton community.
Hamilton Harbour (HH) is a eutrophic, urbanized embayment located at the western end of Lake Ontario. It is bordered by the cities of Hamilton and Burlington, Ontario, Canada, with a combined regional population of about 700,000 people. It covers 2150 ha and measures approximately 8 km by 4 km, with a maximum depth of about 23 m and a mean depth of 13 m (Barica, 1989). Cootes Paradise, a shallow eutrophic coastal wetland, drains into the west end of the harbour. Although oscillating currents and water exchange occurs between Lake Ontario and HH through the narrow Burlington ship canal (Kohli, 1979), water residence time is relatively long (95 to 145 days) (Haffner et al., 1982).
HH was identified as an Area of Concern (AOC) in 1987 as part of the Great Lakes Water Quality Agreement (GLWQA), and a Remedial Action Plan (RAP) was prepared to address these environmental degradation issues (Rodgers et al., 1992). For example, heavy industrialization, predominantly from the steel industry in Hamilton, sewage effluent discharge from four treatment plants, and urban and agricultural runoff have resulted in degraded sediment and water quality, including high nutrient loadings, poor water clarity, elevated chlorophyll and frequent algal blooms (Charlton and LeSage, 1996; Hiriart-Baer et al., 2009; Painter et al., 1990; Rodgers et al., 1992). Severe hypolimnetic oxygen depletion during the stratified season was identified as a stressor for fish and zooplankton as early as the 1970s (Charlton, 1993; Polak and Haffner, 1978; Harris, 1976). Upgrades to sewage treatment plants began in 1988, and both total phosphorus (TP) and chlorophyll a concentrations in surface water have significantly decreased and Secchi disc depth has increased over this time period (Hiriart-Baer et al., 2009). Summer epilimnetic chlorophyll has averaged 14.1 µg l-1 between 1987 and 2007, and summer TP is now typically around 35 to 38 µg l-1 (Hiriart-Baer et al., 2009), meeting the interim RAP goal of 34 µg l-1. However, hypolimnetic oxygen levels have not significantly improved in HH, and severe hypoxia is still a regular occurrence during the stratified season (Dermott et al., 2007).
Despite extensive research on water quality and eutrophication in HH since the 1970s, few studies have focused on the status of zooplankton. Zooplankton may be influenced by contaminants, including elevated ammonia, hypoxia of the deeper water, periodic blooms of inedible or potentially toxic phytoplankton and domination of the fish community by planktivores. Harris (1976) and Piccinin (1977) indicated that rotifers and the small cladoceran Bosmina were abundant at the central station, but copepods and other species of cladocerans such as Daphnia were less common than might be expected. Koenig (1992) showed an increase in Daphnia and cyclopoid copepods in 1990 relative to the 1970s, but the system was still dominated by Bosmina and rotifers. Little published information is available about the current state of zooplankton and rotifers, yet the RAP identifies the “Degradation of zooplankton and phytoplankton populations” as one of the Beneficial Use Impairments (BUIs) that requires further assessment (Environment Canada and Ontario Ministry of Environment, 2011). We undertook a zooplankton monitoring program from 2002 to 2014 at several open water stations to serve as a foundation for the assessment of this BUI. The objectives of this work were to: (1) investigate temporal trends in May-Oct. mean zooplankton and rotifer densities, biomass, production and species richness and compare to other Great Lakes embayments, (2) determine zooplankton seasonal succession patterns and (3) develop species lists for zooplankton and rotifers in HH.
In this study, the term zooplankton includes both crustacean zooplankton and veligers, but not rotifers, which are treated separately. Biweekly zooplankton sampling was carried in out HH from the beginning of May to the end of October (2002–2014), resulting in 13 cruises per year. However, no sampling was conducted in 2005, 2010 or 2011. One to three stations were examined each year as indicated in Figure 1. These included mid-harbour station HH258 (43° 17.241′ N, 79° 50.446′ W; 23 m deep), western station HH908 (43° 16.768′ N, 79° 52.443′ W; 14 m deep) and shallow station HH6 along the north shore (43° 18.133′ N, 79° 50.300′ W; 6 m deep) (Dermott et al., 2007). Zooplankton were collected during the day at discrete depths every 2 m, from 1 m-below surface to 1-m off bottom, using a 41-litre Schindler-Patalas trap fitted with a 64-µm mesh sock and preserved in 4% sugar buffered formalin solution. For analysis, a single composite sample was constructed for each station-date by combining 50% of the sample from each depth. In 2013, sampling frequency was reduced to 9 times through the season, and zooplankton were collected using a metered 64-µm mesh, 40-cm diameter Wisconsin net hauled from 2-m off bottom to the surface. Additional sampling was carried out in mid- to late April, in 2008 and 2012 by DFO, and by the Ontario Ministry of Environment and Climate Change Index Program in 2003 and 2009 at the mid-harbour station only (Currie et al., 2015).
Zooplankton and loose eggs were enumerated using a minimum of 400 individuals. These enumeration methods, production calculations and length-weight regression equations are described in Johannsson et al. (2000) and Bowen and Johannsson (2011). Taxonomic references include Balcer et al. (1984), Dodson and Frey (1991) and Dussart and Defaye (2001).
To explore seasonal succession patterns in HH zooplankton, mean density and biomass of dominant groups were determined for each cruise, representing the mean of 24 samples collected between 2002 and 2014. Cruise 1 represents early May and Cruise 13 represents late October. Cruise 0 represents mid to late April and was represented by only 6 samples. The groups used were Bosmina, Eubosmina, Daphnia, “other cladocerans,” Cyclopoids (excluding nauplii), Calanoids (excluding nauplii), Copepod nauplii and veligers.
Rotifers were not collected in 2013 or at station HH6 in 2014. From 2002 to 2004, total water column samples for rotifers were collected using a Guzzler® diaphragm hand pump as described in Dermott et al. (2007). From 2006 to 2014, epilimnetic water was collected using a 3-litre van Dorn bottle. At the deeper stations (HH258 and HH908), 1 litre of water was pooled from each of six depths spread evenly through the epilimnion during the stratified season. When unstratified, water was collected between 1 and 14 m at HH258 and between 1 and 11 m at HH908. At shallower station HH6, 2 litres of water were used from each of 1, 3 and 5 m depths. The resulting 6 litres of sample water was sieved through 20-µm mesh to avoid loss of small individuals, rotifers were narcotized with soda water and preserved with 4% sugar-buffered formalin. For analysis, a May to October annual composite for each station and year was made by pooling 50% of the sample from each date. Rotifer enumeration used a counting method similar to that used for zooplankton, to a maximum of 400 animals or 25% of the sample by volume (Bowen and Johannsson, 2011). All rotifers were identified by the same taxonomic specialist (Dr. Claudiu Tudorancea), using resources such as Stemberger (1979).
When determining the number of zooplankton and rotifer taxa present each year, coarser taxonomic groupings such as order (e.g. cyclopoid nauplii) or genus were excluded if those taxa were already represented by individuals identified to the species level. Annual mean density data were used to calculate the transformed Shannon-Wiener Diversity (exp H′) index for both zooplankton and rotifers (Molles, 1999). Copepod nauplii were not included in these calculations, but copepodids were allocated to species according to the proportion of adults that occurred for each station-year. Diversity was not calculated at HH6 or HH908 in 2014 as the zooplankton were not identified to species.
Zooplankton mean density, biomass and production
Over the 2002 to 2014 study period, May to October zooplankton densities ranged from 175 to 440 animals l-1 in HH, with an overall mean of 265 ± 16 animals l-1 (Figure 1a). Annual dry zooplankton biomass averaged 306 ± 19 mg m-3, with a range of 182 to 533 mg m-3 (Figure 1b). Cladocerans comprised 50.6% of the total by density and 64.6% by biomass (Appendix 1, available in the online supplementary material). Bosmina (primarily B. longirostris) was numerically the most common zooplankton taxon in the harbour, averaging 83 animals l-1 (32.4%). Bosminids (Bosmina and Eubosmina coregoni) tended to be most numerous at the shallower stations (110 ± 17 and 121 ± 27 animals l-1 at HH6 and HH908, respectively), and averaged only 78 ± 15 animals l-1 at deep station HH258. Bosminids were most abundant in 2002, 2006 and 2008, and least abundant in 2007 and 2012. When this group was excluded, annual mean zooplankton densities were generally quite similar among stations for a given year.
Daphnia was the other abundant cladoceran genus in HH, with annual averages ranging from 8 to 52 animals l-1, and a mean of 24 ± 3 animals l-1. D. retrocurva and D. galeata mendotae represented 5.6% and 3.5% of total zooplankton density, respectively. Their contributions to biomass were greater due to their large relative size, together averaging 80 ± 9 mg m-3 or 26.8% (Appendix 1). Numerically, other common species were Chydorus sphaericus (2.8%), Diaphanosoma birgeii (0.2%) and Ceriodaphnia sp. (0.1%).
The invasive predator Cercopagis was usually only found for a short period during July and August. During this summer period, density and biomass averaged 0.16 ± 0.05 animals l-1 and 0.98 ± 0.29 mg m-3, respectively. This species was usually more abundant at epilimnetic HH6 (Figure 2). The highest Cercopagis level occurred in August 2008 at HH6 (3.3 animals l-1 or 18.6 mg m-3), and were >1.0 animal l-1 at HH6 July 2006 and 2007, and at HH908 August 2003. Cercopagis dropped to very low levels in 2009, 2012 and 2013, but recovered in 2014. Only nine individuals of the spiny water flea Bythotrephes longimanus were found in the harbour, all in October 2009.
Cyclopoid copepods were also dominant in HH, with density and biomass averaging 117 ± 11 animals l-1 (44.0%) and 93 ± 9 mg m-3 (30.6%), respectively. These were mostly juveniles, with copepodids and nauplii larvae comprising 22.1% and 20.2% of the total density, respectively. The most abundant species were Mesocyclops edax, Diacyclops thomasi and Acanthocyclops vernalis. Cyclopoid densities were unusually low in 2002; in contrast they dominated the zooplankton community in 2012. Calanoid copepod density averaged 8.8 ± 0.7 animals l-1 (3.4%). The most numerous calanoid in HH was Leptodiaptomus siciloides. The mean dreissenid veliger larvae density in HH was 4.2 ± 1.4 animals l-1 (1.6%) and in terms of biomass, only 1.5 ± 0.7 mg m-3 (0.5%). Veligers were most abundant at HH6 in 2002 and at HH908 in 2006 and 2014.
Annual (1 May to 31 October) total zooplankton production in HH averaged 4131 ± 359 mg m-3 over the study period. It tended to be highest at epilimnetic HH6 (mean = 4853 ± 597 mg m-3), intermediate at HH908 (mean = 4203 ± 1121 mg m-3), and lowest at deep station HH258 (mean = 3537 ± 410 mg m-3) (Figure 1c). These differences among stations were largely driven by bosminids; when they were removed, production values were fairly similar across the depth gradient. On average, cladocerans and cyclopoids accounted for 77.7% and 19.0% of production, respectively. Cladoceran production was unusually low at HH258 in 2006 and at both HH258 and HH6 in 2012, whereas the highest levels were seen at HH6 in 2007 and HH908 in 2014.
Seasonal succession patterns in zooplankton
When all HH stations were averaged across the study period, zooplankton density and biomass peaked in early June, with secondary peaks in early July and late August (Figure 3). The timing of maximum biomass was variable, ranging from early May to mid-September. Over the study period, peak biomass ranged from only 417 mg m-3 in June 2004 to 2020 mg m-3 in July 2002, and was frequently over 1000 mg m-3. Biomass in April and early May was usually dominated by juvenile cyclopoid copepods, most of which were probably D. thomasi. This spring cyclopoid peak, which was largely absent in the early 2000s, was especially prevalent from 2009 onwards. The highest cyclopoid biomass occurred at HH6 at 1009 mg m-3 in late May 2014. Cyclopoid nauplii densities of over 400 animals l-1 have been observed from mid-April to mid-May in recent years, although they contributed little to biomass due to their small size. M. edax generally replaced D. thomasi as the dominant adult cyclopoid in HH from July to October.
The large zooplankton peaks in June and early July were generally driven by Bosmina, often reaching 300 mg m-3 or greater, especially at the shallower stations. The highest Bosmina biomass values occurred at HH6 in early July 2002 and early June 2006, reaching 1940 mg m-3 and 1117 mg m-3, respectively. Corresponding Bosmina densities were 2350 and 1103 animals l-1. Eubosmina tended to replace Bosmina by early August, and it usually remained at moderate, relatively constant levels until early October (e.g. ≤100 mg m-3). Daphnia were usually present in low numbers in May, began increasing by early June and generally peaked from August to early September. D galeata mendotae tended to be found in spring and October, and the more numerous D. retrocurva usually dominated in the summer. The highest Daphnia biomass occurred in August 2007 and June 2014 (around 700 mg m-3), although summer values more typically ranged between 50 and 300 mg m-3. Calanoids, particularly D. siciloides and herbivorous cladocerans such as Ceriodaphnia sp. and Chydorus sphaericus tended to appear in late summer and early fall, and veliger larvae in mid-summer. The highest veliger biomass observed in HH occurred at HH6 in August 2002 (70.8 mg m-3), although they rarely exceeded 10 mg m-3 or 30 animals l-1 on any given date.
Between 2002 and 2014, rotifer annual densities in HH averaged 302 ± 41 animals l-1, with a range of 18–699 animals l-1 (Figure 1a). Biomass averaged 4.6 ± 0.6 mg m-3, with a range of 0.3–10.8 mg m-3 (Figure 1b). Mean biomass was unusually low at HH258 in 2008. Numerically, Keratella cochlearis was dominant (57%), followed by Polyarthra vulgaris (13%), Polyarthra dolychoptera (5%) and Pompholyx sulcata (5%) (Appendix 2). Other common rotifers were Keratella quadrata, Synchaeta pectinata and Synchaeta kitina. The genus Asplanchna dominated biomass (32%) due to their large size relative to other rotifers. Although mean rotifer density was approximately equal to that of crustacean zooplankton and veligers, they are generally much smaller than zooplankton. When both groups were added together, rotifers represented on average only 1.9% ± 0.2% of total biomass. Mean total rotifer production was estimated to be 143 ± 23 mg m-3 (3.5% ± 0.4%) for the May to October period (Figure 1c).
Taxa richness and evenness
A total of 41 zooplankton taxa were identified in HH during the study period (Appendix 1). On average, 30 taxa each comprised <0.1% of the zooplankton community by density, and 10 taxa were considered very rare, with 10 or fewer individuals identified over the study period. Between 16 and 22 zooplankton taxa were found each year (Figure 4), with an overall mean of 19.0 ±0.4 taxa. No strong temporal or spatial trends in taxa richness were evident. From 2002 to 2014, 48 rotifer taxa were identified in the harbour, with 25 rare taxa each comprising <0.1% by density (Appendix 2). The mean rotifer taxa richness was 15.5 ± 0.6 taxa. Taxa richness was unusually low at HH258 in 2008, 2009 and 2012, with only 10 to 12 rotifer taxa identified in the annual composites (Figure 4). Taxa evenness, as expressed by the transformed Shannon-Wiener Index (exp H′) averaged 5.73 ± 0.30 for zooplankton and 4.30 ± 0.28 for rotifers (Figure 4).
The zooplankton community in HH is often numerically dominated by small-bodied zooplankton, including Bosmina, Eubosmina and juvenile copepods, although Daphnia are important during the summer. Bosmina longirostris, the dominant species in HH, is opportunistic and can reproduce rapidly when conditions are good. As it is widespread and can tolerate a range of conditions, it is not considered a good trophic indicator (Gannon and Stemberger, 1978). However, it can be very abundant in shallow, eutrophic environments (Bowen and Johannsson, 2011; Sager and Richman, 1991). Bosmina prefer to feed on algae, flagellates and bacteria 1 to 3 µm in size (DeMott and Kerfoot, 1982). The small cladoceran Chydorus sphaericus can also be numerous, particularly during algal blooms as it can cling to filamentous algae and feed on attached diatoms and detritus (Fryer, 1968). Unlike the larger Daphnia, both species are more resilient to high rates of fish planktivory due to their small size (Mills et al., 1987). L. siciloides is the dominant calanoid in HH, which unlike most calanoids, tends to prefer warm, shallow, productive waters (Torke, 2001). Both C. sphaericus and L. siciloides have been used as indicators of eutrophication (Gannon, 1972; Gannon and Stemberger, 1978). Cyclopoids characteristic of HH include the large, predatory species M. edax and the eutrophic nearshore species A. vernalis and Eucyclops agilis. The cladoceran Moina brachiata that was identified as characteristic of Cootes Paradise and HH to a lesser extent by Rodgers et al. (1992) was not found in the present study. Another species noteworthy in its absence is Polyphemus pediculus, a predatory cladoceran that can be abundant in nearshore Lake Ontario during the summer (Rudstam et al., 2015).
Despite relative stability in zooplankton populations between 2002 and 2014, there have been dramatic changes since the 1970s. Between 1975 and 1979, rotifers and Bosmina numerically dominated the open waters of the harbour, and copepods and other cladocerans such as Daphnia were uncommon (Harris, 1976; Piccinin, 1977; Piccinin and Harris 1980). Although Bosmina is still the dominant genus in HH, its densities appear to have declined since the 1970s (Table 1). The authors of the early studies concluded that the community was typical of a detrital food chain. Bosmina distribution was positively correlated with oxygen concentration during the summer, as Bosmina avoided the hypoxic deeper waters. Further, hypolimnetic anoxia was identified as a severe stress for all species of cladocerans in the harbour, and this was used to explain the lack of Daphnia and copepods at that time. However, both Daphnia and cyclopoids are now dominant taxa in the harbour, despite ongoing hypolimnetic oxygen depletion for much of the stratified season (Hiriart-Baer et al., 2009). Many zooplankton taxa, including Daphnia and copepods, can tolerate oxygen levels between 1 and 2 mg l-1 (Weider and Lampert, 1985), although they may avoid the hypolimnion at these concentrations (Vanderploeg et al., 2009; Roman et al., 1993). Factors other than dissolved oxygen may have been limiting crustacean zooplankton in the 1970s. However, vertical distribution of zooplankton and possible implications of hypoxia in HH need further exploration.
Daphnia populations appeared to be recovering in HH as early as 1990, when an offshore Daphnia biomass of 139 mg m-3 was observed in late August (adapted from Koening, 1992). Peak summer values are now often two or more times higher, and HH Daphnia populations are now among the highest observed in the Great Lakes (Table 1). Cyclopoid copepods have also increased dramatically since the 1970s, when densities averaged across the season were <20 animals l-1 (Harris, 1976; Piccinin, 1977).
Conversely, rotifer densities and biomass appear to have declined by an order of magnitude or more since the earlier studies. In the 1970s, rotifers comprised 30 to 40% of the biomass in HH (adapted from Harris  and Piccinin ), compared to <3% in the 2000s. The current rotifer proportion is similar to that observed in the Bay of Quinte in northeastern Lake Ontario (Bowen and Rozon, 2015) and in Western Lake Erie (Barbiero and Warren, 2011). Open waters of the Great Lakes tend to have higher proportions of rotifers, averaging 4 to 17% of total biomass, excluding veligers (Barbiero and Warren, 2011). In the 1970s, the June peak in abundance was predominantly the rotifer K. quadrata, and the dominant species during the stratified period was the eutrophic species Brachionus angularis. In the present study, K. quadrata was still numerous in the spring, B. angularis was only rarely encountered, and the most common species was C. cochlearis. Rotifers dominant in Lake Ontario but less numerous or absent in HH included Kellicottia longispina, Conochilus unicornis and Collotheca sp. (Barbiero and Warren, 2011; Makarewicz and Lewis, 2015). However, the dominant HH rotifers also tend to be common throughout the Great Lakes, including Lake Ontario, and their usefulness as trophic indicators is questionable.
HH is now an extremely productive environment for zooplankton as it supports among the highest production, densities and biomass reported in the Great Lakes, including other eutrophic embayments (Table 1). For example, May to October total zooplankton production was 1.3 times higher in HH than in the shallow productive upper Bay of Quinte in northeastern Lake Ontario, 6 times higher than the mesotrophic lower bay of Quinte (Bowen and Rozon, 2015), and about 3-fold higher than western Lake Erie (1993 data, Johannsson et al., 2000). Similarly, May–October zooplankton biomass was about 2.3 times higher in HH than both the upper Bay of Quinte (Bowen and Johannsson, 2011) and eutrophic Sodus Bay, New York (excluding veligers) (Makarewicz and Lewis, 2005). May–October. HH biomass was also 4.4 to 5 times higher than western Lake Erie in the 1990s (Dahl et al., 1995; MacDougall et al., 2001) and about 7 times higher than in Saginaw Bay, Lake Huron in 2009 and 2010 (Pothoven et al., 2013; excluding nauplii and veligers) and Lake St Clair in 2000 (David et al., 2009).
Differences between HH and nearshore waters of Lake Ontario are even more dramatic, especially in recent years when zooplankton biomass in HH can be an order of magnitude or more higher. Lakewide monitoring surveys in 2003 and 2008 indicated that nearshore whole water column crustacean biomass values were <20 mg m-3 in spring and early summer, and about 31 mg m-3 in September 2008 (Rudstam et al., 2015). Biomass in HH averaged 429 mg m-3 in May–June and 276 mg m-3 in July–September. Zooplankton populations, and cyclopoids, in particular, have remained relatively stable in HH since the early 2000s, unlike the precipitous declines observed in Lake Ontario. Rudstam et al. (2015) reported that between 2003 and 2008, offshore epilimnetic crustacean zooplankton density declined by a factor of 12 and biomass by a factor of 5 in the summer, especially for cyclopoids and cladocerans including Daphnia, although decreases were less in the fall. Makarewicz and Lewis (2015) also showed a dramatic loss of the cyclopoid D. thomasi between 1984 and 2007 at a 100 m deep southern Lake Ontario site. This decline in cyclopoids and cladocerans has been largely attributed to increasing predation by the invasive cladoceran Bythotrephes, along with Cercopagis (Makarewicz and Lewis, 2015; Rudstam et al., 2015; Barbiero and Tuchman, 2004). However, Bythotrephes is rarely encountered in HH and probably has had little impact on zooplankton. This may partially explain the persistence of large populations of cyclopoids (31% of total biomass) and Daphnia (27%) in HH in the last decade. Cercopagis, another invasive cladoceran predator, can become abundant in HH during the summer months, with a mean July–August biomass of 0.98 mg m-3. Their densities were in the same range or lower than summer values reported from Lake Ontario (Warner et al., 2006; Rudstam et al., 2015; Makarewicz and Lewis, 2015). Cercopagis is thought to impact rotifers and small zooplankton such as Bosmina and nauplii (Benoit et al., 2002; Warner et al., 2006), although there is no strong evidence that predation by Cercopagis is reducing crustacean zooplankton abundances in the harbour. Densities of crustacean zooplankton prey may be sufficiently high in HH such that moderate populations of Cercopagis have a minimal impact. However, it is possible that the observed declines in rotifers since the 1970s may in part be due to Cercopagis predation. Other reasons for loss of rotifers may include phosphorus controls and improved sewage treatment, and competition with now abundant Daphnia. Daphnia, who are more efficient grazers of phytoplankton, may also physically damage rotifers while feeding (Gilbert, 1988).
The primary reason for the tremendously productive zooplankton community within the harbour likely lies with the high rates of primary and microbial production, especially of edible and nutritious forms including the flagellates Rhodomonas and Cryptomonas, the green Algae Scenedesmus and the Diatom Stephanodiscus (Dermott et al., 2007; Munawar et al., 2017). They noted that although primary production rates were high, phytoplankton biomass was lower than expected given the harbour's eutrophic state. This high production to biomass (P:B) ratio could be caused by strong grazing pressure exerted by herbivorous zooplankton. It is also likely that the microbial community, including bacteria, and allochthonous material associated with sewage discharge may be important to harbour zooplankton, particularly bosminids.
Fish planktivory is likely another important factor influencing zooplankton biomass, species composition and size structure in HH. Zooplankton in the harbour appear to support high populations of planktivorous fish, including Emerald Shiners (Notropis atherinoides), White Perch (Morone americana), Alewife (Alosa pseudoharengus) and Gizzard Shad (Dorosoma cepedianum) (Leisti, 2011; Boston et al., 2016). HH is also an important fish nursery area, and juveniles of many species feed on zooplankton.
Grazing by Dreissenid Mussels (Dreissena bugensis and D. polymorpha) can also cause substantial reductions in phytoplankton biomass (Fahnenstiel et al., 1995; Makarewicz et al., 1999) and there are many examples where zooplankton populations, particularly nauplii and rotifers, have declined following Mussel invasion (Bowen and Johannsson, 2011; David et al., 2009; Pace et al., 1998, MacIsaac et al., 1995). However, adult dreissenid mussels are generally unable to survive at depths greater than 8 m in HH due to hypoxic conditions (Dermott et al., 2007), but they may be abundant on hard nearshore substrates. Given the absence of mussels and correspondingly low veliger numbers in open water areas of HH, food competition and predation on microzooplankton by dreissenids would presumably be less than other locations where mussels are more abundant. The minimal dreissenid effect in HH also likely contributes to the system's high rate of zooplankton productivity. Veliger larvae biomass may be an order of magnitude or more higher in other Great Lakes systems (Table 1), although veligers are often not reported in many Great Lakes zooplankton studies (Rudstam et al., 2015; Pothoven et al., 2013; Makarewicz and Lewis, 2005; Barbiero and Tuchman, 2004). They typically comprise less than 1.5 mg m−3 (<1%) of zooplankton biomass in HH, compared to about 10% to 31% in the Bay of Quinte (Bowen and Rozon, 2015), 17% in Lake Erie (Dahl et al., 1995; MacDougall et al., 2001) and 48% in Lake St Clair (David et al., 2009).
Despite eutrophication-related hypoxia, zooplankton populations are robust in Hamilton Harbour. While community composition and abundance are consistent with eutrophic conditions, there have been dramatic improvements since the 1970s when the system was dominated by Bosmina and rotifers tolerant of eutrophic conditions. Unlike adjacent nearshore areas in Lake Ontario in recent years, HH supports stable populations of Daphnia and cyclopoid copepods, relatively few veligers and even fewer Bythotrephes. Furthermore, extremely high levels of zooplankton production in HH appear to be effectively utilizing the abundant, highly edible algae and microorganisms produced in the harbour, and in turn, zooplankton are supporting an abundance of planktivorous fish. Further study on the role of fish planktivory and the drivers of system production is needed to explore these trophic interactions in depth and ultimately assess the status of plankton populations within Hamilton Harbour.
Foremost, we acknowledge our indebtedness to the Fisheries and Oceans Canada field and laboratory crews who have made this work possible: J. Gerlofsma, R. Bonnell, A. Bedford, M. Frank, H. Niblock, M. Fitzpatrick, R. Rozon, T. Hollister, K. Bonnell and others too countless to name. Thanks to C. Tudorancea, D. Geiling and A. Conway for zooplankton and rotifer sample enumeration and taxonomy. We acknowledge the support of the Hamilton Harbour RAP Restoration council and other scientific leads over the years, including O. Johannsson, M. Koops, M. Munawar, R. Randall, R. Dermott and others. Thanks also to Environment Canada's Technical Operations Services for assistance with sampling and moorings.
Financial support was provided by the Great Lakes Action Plan and by Fisheries and Oceans Canada.
Supplemental data for this article can be accessed on the publisher's website.