The Bay of Quinte, Hamilton Harbour and Toronto Harbour are all coastal regions of Lake Ontario that have experienced eutrophication and all have been designated as ‘Areas of Concern’ under the terms of the Great Lakes Water Quality Agreement. An assessment of the phytoplankton communities in relation to nutrient (P,N,Si) regimes was undertaken during 2015 (Bay of Quinte) and 2016 (Hamilton Harbour and Toronto Harbour) in order to compare and contrast the dynamics of eutrophication in the three ecosystems. Bay of Quinte was found to be phosphorus and silica enriched, but nitrogen limited which resulted in a phytoplankton community dominated by both filamentous diatoms and diazotrophic (N–fixing) cyanobacteria. Hamiton Harbour was phosphorus and nitrogen enriched, but silica depleted with a community dominated by small and large phytoflagellates in addition to experiencing cyanobacteria blooms. Toronto Harbour, by contrast, showed only moderate phosphorus enrichment and no nitrogen limitation, but some silica depletion; phytoplankton was dominated by smaller flagellates and pennate diatoms. Our findings suggest that while phosphorus was a key factor causing cultural eutrophication, other nutrients including nitrogen and silica also had important roles in determining the biomass and composition of the algal standing crop. Future management activities need to consider how the interactions of phosphorus with other nutrients (nitrogen, silica) affect the dynamics of the phytoplankton community in order to promote the recovery of eutrophic ecosystems.
Eutrophication has been a major stressor affecting the health of the Great Lakes. Public concern over algal blooms during the 1960s particularly in Lakes Erie and Ontario prompted action by the governments of Canada and the United States which included an international agreement to protect all of the lakes from pollution. The Great Lakes Water Quality Agreement (1972) provided the framework for protecting, rather than exploiting, this shared resource (IJC 1989; Vallentyne, 1993). Among the provisions, both countries agreed to develop and implement a phosphorus management strategy to combat eutrophication, which was fully implemented in the revised agreement of 1978 (IJC, 1989). In later revisions to the agreement, the two governments identified 43 Areas of Concern (AOCs): ecosystems that were highly degraded and would require additional remediation efforts. Among the 14 stressors highlighted as Beneficial Use Impairments (Hartig and Zarull, 1992) was Eutrophication or undesirable algae. In the Canadian waters of Lake Ontario, Hamilton Harbour, Toronto Harbour and the Bay of Quinte were all listed as AOCs with this impairment (among others).
Eutrophication is generally described as excess algal production caused by elevated nutrient loadings, primarily anthropogenic sources of phosphorus (e.g. Vollenweider, 1968). More recent work has started to consider other factors such as iron and nitrogen, and their role in shaping the composition of the algal community, particularly the formation of cyanobacteria blooms (e.g. Molot et al., 2014; Gardner et al., 2017); others largely consider the problem to be one of toxin producing cyanobacteria (e.g. Pick, 2016). Few of these studies however, delve directly into the structure and function of the entire phytoplankton community. In past work, we have shown that algal blooms can support diverse assemblages of phytoplankton species and maintain high rates of primary production (Munawar et al., 2017, 2018a), and that blooms can be mixtures of multiple taxa (including Cryptophyceae, Diatomeae and Dinophyceae) with different physiological requirements.
In the current study, we consider changes in major nutrient concentrations (P, N, Si) and relate them to phytoplankton community composition and biomass over the growing season (May–October) at three Canadian AOCs listed as ‘eutrophic’. The goal is to offer a comparative assessment of phytoplankton and nutrient dynamics in order to provide insights into the nature of eutrophication and provide managers with guidance for interpreting this Beneficial Use Impairment. While comparative ecology is an imperfect science, there is considerable utility in comparing ecosystems that face similar stresses in order to uncover common principles and development sound management strategies (e.g. Leach et al., 1987). For Great Lakes AOCs with the eutrophication impairment, understanding the nature of the problem at each individual AOC will be helpful in developing strategies towards further remediation and ultimately delisting.
Materials and methods
Three nearshore regions of Lake Ontario were examined during this study: the Bay of Quinte, Hamilton Harbour, and Toronto Harbour. All are designated as AOCs under the terms of the Great Lakes Water Quality Agreement of 1987 and are formally listed with the Beneficial Use Impairment of “Eutrophication or Undesirable Algae”. Locations and sampling sites are shown in Figure 1. The Bay of Quinte is a ‘Z’ shaped embayment of 254 km2 located on the northeastern shores of Lake Ontario. Five major tributaries flow into the Bay and it is a largely riverine system. Depth ranges from 4–8 m, but approaches 30 m near the outflow into Lake Ontario. One station in the upper bay (B) was sampled for this study biweekly from May 6 to October 20, 2015 (n = 13). Hamilton Harbour is a shallow embayment on the western end of Lake Ontario covering 21.5 km2 with a maximum depth of 24 m. One centrally located offshore station (258) was sampled for this study biweekly from May 10 to October 25, 2016 (n = 13), however only monthly phytoplankton data were available (n = 6). Toronto Harbour, formally the Toronto and Region Area of Concern, spans 42 km of waterfront along the northwestern shores of Lake Ontario (Toronto and Region Remedial Action Plan, 2016). One site from each of the inner harbour (IH 4), influenced by the Don River, and the outer harbour (TH 1), influenced by the offshore waters of Lake Ontario, was sampled approximately monthly from May 26 to November 2, 2016 (n = 6).
On site, temperature profiles were measured with a YSI ExoSonde. Integrated water samples were collected and stored in darkened, insulated carbuoys and kept on ice for transport to the lab. At the lab, subsamples were drawn for nutrients, chlorophyll a, and size-fractionated primary productivity assays and preserved for microscopic analyses of phytoplankton. Hamilton Harbour samples were processed the same day, Toronto Harbour and Bay of Quinte samples were stored overnight in a walk-in cooler (≈ 8 °C) and processed the next morning.
Nutrient analysis, including total phosphorus (TP), nitrate + nitrite (NO3+NO2) and silica (SiO2), followed the standard protocol of the National Laboratory for Environmental Testing (NLET, 1997). Chlorophyll a was determined by acetone pigment extraction (Strickland and Parsons, 1968).
Phytoplankton samples were fixed with acidified Lugol’s iodine upon collection. Identification, enumeration and measurement followed the HPMA (2-hydroxypropyl methacrylate) technique described by Crumpton (1987) and is broadly compatible with the Utermöhl (1958) inverted microscope technique.
With respect to the various data sets, significant differences among AOCs were assessed using a one way Analysis of Variance. Unless specifically noted, individual means were compared using post hoc Tukey-Kramer tests.
The results presented below are organized by AOC and summarize various physical, chemical, and biological properties including: surface temperature, nutrients, chlorophyll a, phytoplankton biomass and taxonomic composition. Comprehensive results are outlined in Tables 1, 2 and Figures 2-4.
Bay of Quinte, May 6 to October 20, 2015
Surface temperatures in the Bay of Quinte (stn B) ranged from 11.1 to 25.2 °C over the course of the 2015 survey. Nutrient concentrations were variable: TP ranged from 15.3–69.5 μg l−1; NO3+NO2 from <0.005 (i.e. below detection) to 0.13 mg l−1, and SiO2 from 1.8 –9.2 mg l−1. During the study period, chlorophyll a ranged from 4.0–35.0 μg l−1. Complete results are shown in Table 1a.
Phytoplankton biomass ranged from 0.5–9.5 g m−3 and was composed of Cyanophyta (1.4–81.1%), Chlorophyta (0.7–13.4%), Chrysophyceae (0.7–48.2%), Diatomeae (4.2–63.7%), Cryptophyceae (2.0–28.7%) and Dinophyceae (0–13.8%) (Fig. 2a). Among the dominant species on each sample date were: Chrysochromulina parva, Fragilaria crotonensis, Dolichospermum lemmermannii, Aulacoseira granulata, A. ambigua, Microcystis aeruginosa, D. Crassa, and Cryptomonas erosa (Table 2a).
Algal blooms (biomass >3 g m−2) were observed on 8 (of 13) sampling dates and included three diatom blooms (Fragilaria crotonensis, Aulacoseira granulata, A. ambigua), three cyanobacteria blooms (Microcystis aeruginosa, Dolichospermum lemmermannii, D. crassa) and two that were mixtures of diatoms and cyanobacteria. On the remaining dates, Diatomeae were most prevalent (≈60% total biomass) on two of those (June 30, July 28), while Chyrsophyceae were most prevalent in early May and the others were mixtures of multiple taxa (Fig. 2a, Table 2a).
Hamilton Harbour, May 10–October 25, 2016
During our survey, surface temperatures in Hamilton Harbour ranged from 11.3 °C (May 10) to 25.2 °C (Aug. 15). With respect to nutrients, TP ranged from 24.6–59.7 μg l−1; NO3+NO2 from 1.5–2.7 mg l−1, and SiO2 ranged from 0.06–1.37 mg l−1. Chlorophyll a concentrations varied from 5.8–22.3 μg l−1 over the course of the sampling season. Complete results are shown in Table 1b.
Phytoplankton biomass ranged from 1.2–4.7 g m−3 during the growing season of 2016. The relative composition of the major taxonomic groups fluctuated widely throughout the year including: Cyanophyta (0.4–54.7%), Chlorophyta (3.4–32.9%), Chrysophyceae (0.5–72.6%), Diatomeae (1.2–20.4%), Cryptophyceae (1.5–20.3%), and Dinophyceae (0–41.4%) (Fig. 2a). The major species associated with each taxon were also highly variable and included Gymnodinium helveticum, Tribonema sp., Lyngbya birgei, Ceratium hirundinella, and Microcystis aeruginosa (Table 2b).
There were only two observations where a single taxon composed more than 50% of the biomass. On June 8, Chrysophyceae represented 72.6% of the total phytoplankton biomass, primarily Tribonema sp. Similarly on October 12, Cyanophyta accounted for 54.7% of the biomass and included Microcystis aeruginosa, M. wesenbergii, Phormidium sp., Dolichospermum planktonica, Cuspidothrix issatschenkoi, Psuedanabaena limnetica, and P. galeata (Fig. 2a, Table 2b).
Algal blooms (biomass >3 g m−3) were observed on two occasions during 2016, July 19 and September 14. On July 19, the bloom contained a mixture of Cyanophyta (39.7%, Lyngbya birgei), Chlorophyta (32.9%, Oocystis lacustris, O. parva) and Diatomeae (20.4%, Fragilaria crotonensis). Likewise, the September 14 bloom contained a slightly higher proportion of Cyanophyta (46.6%, L. birgei, Aphanizomenon flos-aquae, Microcystis aeruginosa) in addition to Dinophyceae (30.3%, Ceratium hirundinella) and Diatomeae (17.0%, F. crotonensis) (Fig. 2a, Table 2b).
Toronto Harbour, May 26–November 2, 2016
During the Toronto Harbour survey, surface temperatures in the inner harbour (IH4) ranged from 11.3 °C on May 10 to a high of 19.8 °C on August 18, compared to 10.1–22.3 °C in the outer harbour (TH1). Nutrient concentrations were variable, TP ranged from 8.9–30.1 μg l−1 in the inner harbour and 9.9–19.6 μg l−1 in the outer harbour. NO3+NO2 ranged from 0.2–0.4 mg l−1 in both the inner and outer harbour. SiO2 in the inner harbour ranged from 0.7–1.1 mg l−1 and in the outer harbour from 0.6–0.8 mg l−1. Additionally, chlorophyll a varied from 1.7–9.1 μg l−1 in the inner harbour and 1.3–6.3 μg l−1 in the outer harbour. Complete results for the inner and outer harbour sites are shown in Table 1c.
Phytoplankton biomass in the inner harbour ranged from 0.4–2.2 g m−3. The relative composition of the major taxa, including Cyanophyta (0.5–15.9%), Chlorophyta (2.2–68.2%), Chrysophyceae (9.7–34.8%), Diatomeae (5.1–46.6%), Cryptophyceae (11.2–54.6%) and Dinophyceae (0–6.7%), was highly variable throughout the course of the study (Fig.2b). Among the most dominant species contributing to total biomass were: Synedra ulna, Cryptomonas erosa, Fragilaria crotonensis, and Pandorina morum (Table 2c).
No algal blooms were observed during this survey. There were two occasions where a single taxon contributed >50% of the total biomass. On September 19, Chlorophyta (mostly Pandorina morum) accounted for 68.2% of the total biomass, and on November 2, Cryptophyceae (C. Erosa, Rhodomonas minuta) represented 54.6% of the biomass (Fig. 2b, Table 2c). On another two occasions, Diatomeae accounted for nearly half the biomass: May 26 (45.5%) and July 20 (46.6%).
Phytoplankton biomass in the outer harbour ranged from 0.3–0.9 g m−3. The relative composition of the major taxa, including Cyanophyta (0.5– 11.8%), Chlorophyta (2.0–20.6%), Chrysophyceae (6.5–29.7%), Diatomeae (15.7–55.1%), Cryptophyceae (11.1–47.4%) and Dinophyceae (0.7–44.7%), was highly variable throughout the study period (Fig. 2b). Among the most dominant species contributing to total biomass were: Cryptomonas erosa, Fragilaria crotonensis, Chrysochromulina parva, Ceratium hirundinella and Stephanodiscus binderanus (Table 2c).
Algal blooms were not observed at the outer harbour. On one occasion (Nov. 2), Diatomeae accounted for 55.1% of the total biomass; otherwise no single taxon contributed more than 50% of the total biomass (Fig. 2b, Table 2c). There were however, three dates where a single taxon was close to that threshold: May 26 (45.8% Diatomeae), June 24 (47.4% Cryptophyceae) and September 19 (47.4% Dinophyceae).
Comparison of sites
All of the major parameters were compared among sites using a one-way ANOVA with post hoc Tukey Kramer tests. Significant differences (p < 0.05) among sites were observed for total phosphorus, chlorophyll a, nitrate + nitrite, and silica (Fig. 3). Significant differences among sites were also observed for total phytoplankton biomass, Diatomeae biomass and Dinophyceae biomass (Fig. 4). Significant differences were not observed for Cyanophyta biomass (Fig. 4c), as well as Chlorophyta, Chrysophyceae and Cryptophyceae biomass (not shown).
Hamilton Harbour (HH), Toronto Harbour (TH) and the Bay of Quinte (BQ) are all Areas of Concern located in the coastal waters of Lake Ontario and all have been formally designated with the impairment of “Eutrophication or undesirable algae”. All have distinct geographical, hydrological, limnological and biological characteristics (e.g. Johnson and Owen, 1971, Harris et al., 1980, Haffner et al., 1982) raising questions as to the nature of eutrophication in each ecosystem. In the current exercise, we compare these ecosystems based on nutrient regimes, phytoplankton biomass and taxonomic composition. The goal is to improve our understanding of the biological consequences of eutrophication with respect to phytoplankton ecology so that long-term adaptive management strategies for this impairment can be developed and implemented in the Great Lakes AOCs and elsewhere.
Eutrophication management in the Great Lakes has relied on phosphorus abatement to regulate the amount of primary production and the size of the algal standing crop (Vollenweider et al., 1974); a paradigm which was enshrined in the Great Lakes Water Quality Agreement beginning in 1978 (IJC, 1989). Underlying this paradigm is the assumption that phosphorus enriched systems become nitrogen limited resulting in excess growth of nitrogen-fixing (diazotrophic) cyanobacteria (e.g. Vollenweider, 1968; Schindler, 1978). One of the unintended consequences of the paradigm has been that total phosphorus and chlorophyll a concentrations became the de facto measures for assessing the nutrient regime and algal standing crop respectively in eutrophic ecosystems while other factors have simply been assumed and all too often ignored. Accordingly, fundamental differences in nutrient and phytoplankton dynamics may be missed.
Phosphorus and chlorophyll a
In the current study, we observe that total phosphorus concentrations are elevated in the Bay of Quinte and Hamilton Harbour (TP ≈ 35–40 μg l−1), and significantly higher (p = 0.0001) than those observed in Toronto Harbour (TP ≈ 15 μg l−1) (Fig. 3a). Similarly, chlorophyll a was also elevated in BQ and HH (Chl ≈ 13–15 μg l−1) and significantly higher (p = 0.0005) than TH (Chl ≈ 4–5 μg l−1) (Fig. 3b). Taken at face value, these measures alone suggest that BQ and HH are eutrophic while TH is mesotrophic using standard empirical benchmarks (e.g. Vollenweider et al., 1974; Carlson, 1977). But they provide no insight into other nutrients which may affect the growth and composition of the algal standing crop (e.g. nitrogen, silica) nor the composition of the algal standing crop.
Excess phosphorus concentrations are certainly a precondition of cultural eutrophication, however there is increasing concern that nitrogen enrichment in some eutrophic ecosystems–in addition to phosphorus enrichment–is promoting the development of harmful (i.e toxigenic) algal blooms (e.g. Glibert and Burford, 2017; Paerl et al., 2016; Monchamp et al., 2014). In this scenario, neither phosphorus nor nitrogen are limiting the algal standing crop during the growing season. There is some evidence for this in Hamilton Harbour where nitrogen (NO3+NO2) concentrations (N ≈ 2 mg l−1) are significantly higher (p < 0.0001) than Toronto Harbour (N = 0.3 mg l−1, both sites) and Bay of Quinte (N = 0.03 mg l−1) (Fig. 3c). HH has seen steady increases in NO3+NO2 concentrations since 2003 as a consequence of changes in wastewater treatment (Hiriart-Baer et al., 2016). In this regard, HH appears to be both nitrogen and phosphorus enriched; N:P ratios of 26–94 were well above the Redfield (1958) value of 16 which suggests that neither is restricting phytoplankton growth. More important, however, is how excess nitrogen may shape the composition of the phytoplankton community.
Although NO3+NO2 concentrations in Toronto Harbour and the Bay of Quinte were not found to be significantly different from each other, there is a critical ecological difference. BQ is the only site where NO3+NO2 falls below detection during the peak growing season, providing strong evidence for nitrogen limitation of the algal standing crop as a consequence of phosphorus enrichment. As a result, the phytoplankton community is often dominated by eutrophic species that either tolerate low nitrogen or are able to fix nitrogen (see Munawar et al., 2018a for more discussion). In contrast, NO3+NO2 concentrations in Toronto Harbour never fell below 0.2 mg l−1 at either the inner or outer harbour sites and are consistent with values reported from Lake Ontario (e.g. Munawar et al., 2018b; Dove and Chapra, 2015). NO3+NO2 values in this study are similar to those reported 30 years ago by Haffner et al. (1982) for the inner harbour with the notable exception that at the lowest ebb in the late summer, the minimum value we observed, 0.24 mg l−1, was approximately 2.5X greater than the 0.09 mg l−1 observed in 1977. Overall, this is indicative of less biological (phytoplankton) demand for nitrate at the peak of the growing season and suggests other factors may be affecting the growth of the phytoplankton community in Toronto Harbour.
Silica (SiO2) is another nutrient that has significant implications for the structure and dynamics of the phytoplankton community. Diatoms in particular require dissolved silicate for growth and in eutrophic environments, silica may become limiting (e.g. Schelske et al., 1986; Conley et al., 1993). Silica concentrations below 0.8 mg l−1 can be limiting for larger filamentous diatoms like Aulacoseira (Lund, 1954) with concentrations ≤0.39 mg l−1 representing severe depletion (Schelske et al., 1986). The Bay of Quinte has significantly higher (p < 0.0001) silica concentrations (S = 5.7 mg l−1) compared to Hamilton Harbour (S = 0.6 mg l−1) or the inner (S = 0.8 mg l−1) and outer (S = 0.7 mg l−1) harbour sites of Toronto Harbour (Fig. 3d). In Toronto Harbour, there is evidence for silica limitation (< 0.8 mg l−1) at the inner harbour site (4/6 observations) and especially at the outer harbour site (6/6). In Hamilton Harbour, there was evidence for silica depletion on 9 (of 13) observations with severe limitation observed on 4 of those occasions. In contrast, silica in the Bay of Quinte is well above these thresholds.
Phytoplankton composition and nutrient interactions
Despite being designated as eutrophic by the remedial action plans, these three AOCs have different nutrient regimes. To give a sense of how different these ecosystems are, we plotted mean nutrient concentrations on a logarithmic scale in Figure 5. The Bay of Quinte is characterized by high TP, N limitation and high silica; Hamilton Harbour has high TP, high N and low silica, and both Toronto Harbour sites have moderately high TP, moderately high N and low silica. The net result is that the phytoplankton assemblages these ecosystems support have different characteristics in terms of community structure driven by these different nutrient regimes. Notably, we found significant (p < 0.05) differences among the AOCs with regards to: total phytoplankton biomass (BQ), diatom biomass (BQ) and dinoflagellate biomass (HH) (Fig. 4).
As eutrophication is fundamentally a problem of excess algal growth, the size of the algal standing crop (and not proxy measures) is the important consideration. In recent work, Munawar et al. (2017) defined phytoplankton biomass >3 g m−3 as being the threshold for algal bloom conditions with persistent observations above that threshold being characteristic of a eutrophic environment. This threshold is a refinement of the original biomass scale developed by Munawar and Munawar (1982) wherein single observations >3 g m−3 were considered to be eutrophic conditions. Phytoplankton biomass is highest in the Bay of Quinte (P = 4.0 g m−3) characteristic of a eutrophic environment with frequent algal blooms (Fig. 4a). This is not significantly different from Hamilton Harbour (P = 2.6 g m−3, p > 0.4), but is significantly higher than Toronto Harbour (p < 0.02) (Fig. 4a). However, phytoplankton biomass in Hamilton Harbour falls slightly below the empirical threshold of 3 g m−3 expected in eutrophic environments. At TH, both inner (P = 1.0 g m−3) and outer (P = 0.6 g m−3) harbour sites were well below this threshold and consistent with oligotrophic environments.
The composition of the phytoplankton community also reveals important information about the nature of ecosystem impairment in the three Canadian AOCs. Diatom biomass in the Bay of Quinte (D = 1.4 g m−3) is significantly higher (p < 0.001) than the other sites (D = 0.2–0.4 g m−3) (Fig. 4c). In BQ, the phytoplankton community includes blooms of Aulacoseira granulata and A. ambigua which are large filamentous diatoms that thrive in P - enriched, N - depleted and high silica environments (e.g. Reynolds et al., 2002). It is also worth noting that during the course of our survey, some of the highest concentrations of diatoms observed in BQ (2–3 g m−3) occurred during Cyanobacteria blooms. Similar findings in Lake Erie were reported by Reavie et al. (2016) and also attributed to increasing silica loads supporting the growth of filamentous diatoms. We conclude that diatoms are significant part of the eutrophication problem in the bay as a result of high levels of silica and high phosphorus concentrations. Cyanobacteria biomass was also highest in BQ (C = 1.7 g m−3) but only significantly greater than TH (p < 0.02, Wilcoxon Pairs, Fig 4b). The blooms in BQ were composed largely of diazotrophic species including Dolichospermum lemmermannii and D. crassa; cyanobacteria blooms began in August when Nitrate + Nitrite concentrations fell below detection. Nitrogen limitation did not appear to inhibit the success of diatom populations in BQ but it did allow opportunities for diazotrophic cyanobacteria to thrive as well.
The nutrient regime in HH is characterized by high phosphorus and nitrogen, and low silica. These physicochemical conditions in HH differ substantially from those in BQ and are also reflected in phytoplankton community structure. First, HH typically lacks large filamentous diatoms which results in less biomass. The diatom community is composed of pennate, centric and elongate forms including Fragilaria crotonensis, Cyclotella atomus, and Tabellaria flocculosa. More importantly, these are not the largest contributors to the phytoplankton assemblage. Our study suggests that the diatom community may be affected by severe silica limitation which in turn limits the size and composition of the diatom community. Likewise, an independent study by Ridenour (2017) conducted during 2016 found that diatoms assimilated ≈0.6 moles Si yr−1, an amount similar to ultra-oligotrophic Lake Superior and the open ocean. One of the paradoxes of HH is its relatively low phytoplankton biomass, at least for a eutrophic environment. This was originally described by Haffner et al., (1980) and attributed to the harsh physical environment. Later work made similar observations, but also suggested strong zooplankton grazing as a contributing factor (Munawar and Fitzpatrick, 2017; Bowen and Currie, 2017). A lack of silica to support diatom growth in the summer months is another factor that may also contribute to this paradox.
The second point with respect to the comparison of HH with the other AOCs is that cyanobacteria biomass in HH is not significantly different from BQ or TH (Fig. 4c), but the composition of the dominant species, especially during bloom conditions, is quite different. In July, there is a bloom of Limnoraphis birgei (syn. Lyngbya birgei), a filamentous cyanobacterium characteristic of turbid, meso- to eutrophic environments (c.f. Komarek et al., 2013). In addition to L. birgei, colonial forms of cyanobacteria including Aphanizomenon flos-aquae and Microcystis aeruginosa are also prevalent during the late-summer but do not reach bloom concentrations. Third, perhaps the most striking difference is the comparatively large biomass of Dinophyceae, including Gymnodinium helveticum and Ceratium hirundinella. These large flagellates rely on phagotrophy for essential nutrients; an advantage in an environment subject to extreme physical disturbances (Haffner et al., 1980; Munawar et al., 2017). Other flagellates, notably Cryptomonas erosa (Cryptophyceae), are also prevalent in the spring and fall.
In this comparison of three Canadian AOCs, TH stands out for its overall lack of eutrophic characteristics in the pelagic zone. Generally speaking, the inner harbour contained a mixture of small phytoflagellates including Chrysochromulina parva, Cryptomonas minuta and Rhodomonas minuta, as well as pennate diatoms including Synedra ulna, Diatoma tenuis, and Fragilaria crotonensis. The outer harbour site contained a similar mixture of smaller flagellates in addition to larger dinoflagellates including Peridinium umbonatum, and Ceratium hiriundinella. Likewise, in addition to the pennate diatoms found in the inner harbour, Tabellaria flocculosa and Asterionella formossa were also found during spring in the outer harbour. On the whole, the phytoplankton standing crop, both in terms of biomass and composition, is consistent with oligotrophic conditions. While there is some evidence of moderate phosphorus enrichment in the inner harbour (Table 1), there is little evidence of eutrophication in either the inner or outer harbour. As observed in HH, the large number of flagellates suggests a phytoplankton community well adapted to physical disturbances. Equally important, we did not observe blooms of diatoms and/or cyanobacteria that are prevalent in the other AOCs. It is likely that the potential development of eutrophic conditions in TH is restricted by the frequent exchange of water with oligotrophic Lake Ontario.
Long term trends
Improvements in sewage treatment and storm water management have led to significant phosphorus load reductions in the BQ, HH and TH AOCs since the 1970s. In Table 3, we provide a brief contrast of pre-phosphorus abatement conditions with results of the current study. During 1973 for example, BQ had total phosphorus concentrations of 36–160 µg l−1, phytoplankton biomass of 2–34 g m−3 and experienced blooms of diatoms (Aulecoseira spp.) and cyanobacteria (Dolichospermum spp.) (e.g. Nicholls et al., 2002). In 2015, we observed TP concentrations of 15.3–69.5 µg l−1, biomass of 0.5–9.5 g m−3 and blooms of Aulecoseira and Dolichospermum. Overall, the decline in TP has resulted in less biomass of both diatoms and cyanobacteria which in turn has lessened the biological demand for silica and nitrate although N limitation is still common. BQ is less eutrophic than it was in 1973, however, it is still eutrophic.
Hamilton Harbour has also seen dramatic reductions in total phosphorus concentrations, from 28–128 µg l−1in 1977 to 24.6–59.7 µg l−1 in 2016 (Table 3). However, silica, nitrate and phytoplankton biomass all have similar values in 2016 compared to 1977. Further, the dominant phytoplankton assemblages reported for 1977 by Haffner et al. (1980), primarily diatoms (Stephanodiscus hantzschii, Aulacoseira islandica), Cryprophyceae (Cryptomonas ovata, C. erosa and Rhodomonas minuta) and Chlorophyta (Chlamydomonas spp.) were different from the dinoflagellate (Gymnodinium helveticum, Ceratium hiriundinella) and cyanobacteria (Lyngbya birgei, Microcystis aeruginosa) dominated assemblages observed in our study in 2016. Phosphorus load reductions in Hamilton Harbour do not appear to have affected phytoplankton biomass (see also Munawar et al., 2017), but there does appear to be a shift in the composition of the phytoplankton community due at least in part to changes in nutrient ratios.
Total phosphorus concentrations in the inner harbour area of the Toronto Harbour AOC were in the range of 25–33 µg l−1 in 1977 compared to 8.9–30.1 µg l−1 during 2016 (Table 3). Nitrogen concentrations showed little change and phytoplankton biomass was considerably lower in 2016 (0.4–2.2 g m−3) compared to 1977 (1–5 g m−3). Major phytoplankton species in 1977 included Aulacoseira islandica, Stephanodiscus hantzschii, Cryptomonas ovata, C. erosa and Rhodomonas minuta (Haffner et al., 1982). During 2016, Synedra ulna, Fragilaria crotonensis, Cryptomonas erosa and Pandorina morum were the most common species. Overall, this suggests a general improvement in conditions since the 1970s but also a shift in limiting nutrients.
One of the unique characteristics of the Remedial Action Plan program was the ability of each AOC to set its own listing and delisting criteria for the relevant Beneficial Use Impairments. Unfortunately, this has also led to a lack of consistent standards being applied across AOCs (Krantzberg, 2004; George and Boyd, 2007). The delisting criteria for BUI 8: Eutrophication or undesirable algae for each of BQ, HH and TH are given in Table 4. There is considerable variability in the choice of parameters; the use of specific target values or more general guidelines; and the applicability to open waters vs sewage discharges.
While the overarching goal of eutrophication management is to reduce phosphorus loading in order to reduce noxious algal blooms, the size and composition of the algal standing crop is often overlooked in the process. Only the Bay of Quinte has a specific target for phytoplankton biomass (4–5 g m−3) under BUI 8; the others do not. Likewise, none of the AOCs have targets for phytoplankton composition under BUI 8 but BQ does have one under BUI 13: Degradation of phytoplankton and zooplankton populations (Bay of Quinte Remedial Action Plan, 2018).
The results of this study show that phytoplankton biomass and composition are important considerations in assessing the health and status of eutrophic ecosystems. Moreover, we showed that nitrogen and silica, in addition to phosphorus, also help to regulate phytoplankton biomass and composition. Managing eutrophication within each AOC requires a detailed understanding of how the nutrient regime affects the structure and dynamics of the algal standing crop. We recommend that both phytoplankton and nutrient dynamics be given careful consideration when making decisions regarding delisting and/or future abatement strategies.
Summary and conclusions
Eutrophication is generally managed by phosphorus abatement with little consideration given to how other nutrients might affect the size and composition of the phytoplankton community. Our experience in the Bay of Quinte, Hamilton Harbour and Toronto Harbour Areas of Concern represent three unique examples of phytoplankton–nutrient interactions. BQ experiences phosphorus enrichment coupled with late summer nitrogen depletion and elevated silica concentrations that results in a community characterized by filamentous diatoms co-existing with diazotrophic cyanobacteria. On the other hand, HH has high concentrations of both phosphorus and nitrogen but low silica which results in cyanobacteria blooms in the late summer, as well as an abundance of phytoflagellates (Dinophyceae and Cryptophyceae) throughout the growing season. TH presents a different scenario since it does not have the same degree of phosphorus enrichment as the other AOCs which is reflected by the abundance of pennate diatoms and smaller flagellates characteristic of an oligotrophic environment.
This comparison of the three AOCs demonstrates that the base of the foodweb in each ecosystem is unique and driven by different nutrient regimes. While phosphorus abatement is an established tool for controlling eutrophication, our data shows it is not the only factor that determines the nature and characteristics of phytoplankton communities in a eutrophic ecosystem. Therefore, it is suggested that eutrophication studies include other major nutrients like nitrogen and silica. Management strategies aimed at controlling eutrophication need to be tailored to the environment in question and cannot continue to rely on one simple paradigm of phosphorus abatement.
The authors would like to thank H. Niblock, R. Rozon, R. Bonnell, K. Bowen and W. Currie of DFO as well as the staff and students who assisted with sample collection and laboratory analysis during 2015 and 2016. We also thank J. Lorimer (DFO) for help with the figures and tables. Phytoplankton enumeration and taxonomic identification was conducted by Phytcotech (L. St. Amand and staff). We appreciate the constructive comments provided by the referees which improved the manuscript. Funding was provided by Environment Canada under the Great Lakes Action Plan.