Bottom sediment quality in Hamilton Harbour was assessed as part of a long-term research and monitoring program over a period of three decades in order to support remedial activities. Sampling locations reflected a range of shoreline activities and sources of chemical contamination to the harbour. An assessment of temporal trends in metals, polycyclic aromatic hydrocarbons and polychlorinated biphenyls indicate that concentrations of all three classes of contaminants have decreased in sediments in most areas of the harbour since the period 1990–2000; however, the Windermere Arm area impacted by historical industrial activities along the southeastern shoreline area of the harbour was an exception, as trends in some metals and polychlorinated biphenyls showed overall increases. Assessment of spatial distributions of contaminants and the associated polycyclic aromatic hydrocarbon and polychlorinated biphenyl profiles showed that Randle Reef and Windermere Arm continue to be significant contributors to harbour-wide contamination by polycyclic aromatic hydrocarbons and polychlorinated biphenyls, respectively. Continuation of the program after remedial activities should provide an assessment of the overall efficacy of management actions to improve environmental quality in Hamilton Harbour.
Hamilton Harbour was identified as an Area of Concern (AOC) in 1987 by the governments of Canada and the United States as a result of beneficial use impairments, some of which are related to the presence of toxic chemicals. Great Lakes AOCs are geographical areas identified pursuant to the Great Lakes Water Quality Agreement as having a high level of environmental degradation. Sources of contaminants to Hamilton Harbour are the result of at least 150 years' worth of industrial emissions, vehicular emissions, urban runoff treated sewage, and waste disposal from a 600 km2 drainage basin. The Harbour is and continues to be a hub for steel making. A range of contaminants including polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs), organochlorine pesticides and metals have been detected in sediments in Hamilton Harbour at levels that exceed provincial and/or federal sediment quality guidelines (Mayer and Manning, 1990; Marvin et al., 2000; Murphy et al., 1990; Poulton, 1987; Zeman and Patterson, 2003). Some areas of the harbour are characterized by highly-contaminated sediments, including the areas of Randle Reef and Windermere Arm (Figure 1). Total PAH concentrations in sediments from the Randle Reef area exceed 1,000 µg g−1 (Murphy et al., 1990; Graham et al., 2012), while total PCB levels from the Windermere arm exceed 150 µg g−1 (Zeman and Patterson, 2003); Randle Reef is one of Canada's largest contaminated sediment sites and is being subjected to a large-scale sediment remediation project (Graham et al., 2012). Historically, chemical contaminants entered the harbour through tributary inflows, wastewater treatment plants, runoff, industrial emissions, atmospheric loadings, and inflow from Lake Ontario (Ling et al., 1993).
In its efforts to support the Remedial Action Plan (RAP) for the Hamilton Harbour AOC, Environment and Climate Change Canada monitors the occurrence, distribution and fate of harmful pollutants in Hamilton Harbour in a variety of matrices, including bottom sediment. While contamination of bottom sediment itself is not a beneficial use impairment (BUI) as it pertains to the designation of Hamilton Harbour as an AOC, the matrix is inextricably linked to BUIs including restrictions on fish and wildlife consumption, fish tumours and other deformities, and degradation of benthos (Hamilton Harbour RAP, 1992). In addition, bottom sediments themselves can serve as a source of contaminants to the harbour water column (Ling et al., 1993). We previously reported on trends of contaminants associated with suspended sediments (Burniston et al., 2016). In this article, we present the temporal trends and spatial distributions of PAHs, PCBs and metals including mercury, associated with bottom sediments in Hamilton Harbour over the past several decades, and discuss the association of contaminants with various source types and historical industrial activities in the watershed.
Materials and methods
Surficial (0–10 cm) sediment samples were collected in 1990, 2000 and 2014 throughout the harbour as part of comprehensive surveys. The selection of sampling stations was based on the variety of conditions that are dictated by water exchange with Lake Ontario and point- and non-point source discharges (Mayer and Nagy, 1992). A total of 117 stations were sampled; 48 in 1990, 44 in 2000 and 25 in 2014. From each of these 3 studies, co-located stations were selected for temporal comparison which included 13 sites over all 3 years and 25 sites for 2000 and 2014; these 25 sites included the 13 sites sampled over the 3 studies plus an additional 12 (Figure 1). Details of sediment collections for 2000 are provided in Milani and Grapentine (2006), and were similar for all three sampling years. Site coordinates were obtained using a differentially-corrected GPS receiver. Sediment samples were collected using either a petite Ponar grab (area = 255 cm2) or a mini-box corer (40 cm × 40 cm = 1,600 cm2), and represented a homogenized sample taken from a Ponar grab or from the surface (0–10 cm) of the box corer. In 2011, surficial (0–3 cm) sediment samples were collected harbour-wide as part of a recurring monitoring study; 11 sampling sites were chosen to represent all sections of the harbour including 4 in WA (WA1-WA4), 3 in RR (RR1 – RR4), and one in each of the NE (NE1), OH (OH1), and WE (WE1) (Figure 1). All equipment was thoroughly rinsed in the site water prior to sampling. Sediment samples were collected from a mini-box core using a 5-cm long (6.5 cm diameter) acrylic tube. All sediment samples were passed through a 250 µm mesh screen into a glass bowl and stirred for 2 min to further homogenize the sample. Separate sample jars were filled for the analysis of metals, and TOC and grain size (125 ml polyethylene containers), and for organic contaminants and archiving (250-mL amber glass containers with Teflon-lined lids). Samples were transported on ice and frozen until analysis.
For spatial and temporal comparisons, Hamilton Harbour was divided into five areas; these areas were selected based on shoreline activities, including industrialization, tributary inputs, physical processes in the harbour, and for comparison with areas less impacted by sources of chemical contamination. These areas were identified by the following boundaries: (1) Windermere Arm (WA); (2) Randle Reef (RR); (3) Open Harbour (OH); (4) Northeast Harbour (NE) and (5) West End (WE) (Figure 1). Rao et al. (2009, 2016) have modeled circulation patterns in Hamilton Harbour and described their potential impacts on contaminant distributions; two main counter-rotating eddies in the harbour correspond to the deep water area of the OH area and the NE area, which further influenced our clustering of sampling locations within the five zones.
For the 1990, 2000 and 2014 sample sets, sediment concentrations of PAHs (16 parent compounds) were determined by GC/MS by USEPA method SW846-8270, while total PCBs were determined by gas chromatography with electron capture detection (GC/ECD) using USEPA method SW846-8082. Trace metals (hot aqua regia extraction) were analyzed using American Public Health Association (APHA, 1992) Standard Method (SM) 3120 by inductively coupled plasma – atomic emission spectroscopy (ICP-AES), while mercury was determined using SM 3112 (APHA, 1992) based on cold-vapour atomic absorption spectroscopy (AAS).
For the 2011 sample set, sediment samples were dried (air-dried or freeze-dried) and spiked with surrogate PAHs (deuterated derivatives) and PCBs 30 and 204. Sediment samples were extracted in dichloromethane using a Soxhlet or by Accelerated Solvent Extraction (ASE) followed by an open-column silica gel or florosil cleanup resulting in conventional A and B fractions. Fraction A contained PCBs while Fraction B contained the PAHs. Extracts were analyzed for PAHs by gas chromatography – mass spectrometry (GC/MS) and for PCBs using dual-column GC/MS analysis with quantitation based on 209 individual congeners.
Data analyses and presentation
Metal (chromium, copper, nickel and lead) and organic (total PAHs, PCBs) contaminant data are presented spatially for all sites sampled within a year. Temporal trends in concentrations of the 4 metals plus zinc were examined from 1990 to 2014 at 13 co-located sites; there were no data for Hg, PAHs and PCBs in 1990. From 2000 to 2014, temporal trends in concentrations of total PAHs, total PCBs and all metals including Hg were examined at 25 co-located sites. The number of sampling sites used in the analysis varied by area and year, and ranged from 1 to 10 (Table S1, available in the online supplementary information [SI]). Total PAHs were based on the sum of 16 individual compounds for the 1990, 2000 and 2014 studies; naphthalene (N), acenaphthylene, acenaphthene, fluorene, phenanthrene (Ph), anthracene, fluoranthene (Fl), pyrene (Pyr), benzo[a]anthracene (BaA), chrysene (Ch), benzo[b]fluoranthene, benzo[k]fluoranthene (sum of both benzofluoranthene compounds identified as B[b/j/k]F), benzo[a]pyrene (BaP), indeno[1,2,3-c,d]pyrene Ind[1,2,3cd]P), dibenzo[a,h]anthracene (DB[a,h]A) and benzo[g,h,i]perylene (B[ghi]P). For the 2011 study, total PAHs were based on the sum of 18 compounds including benzo[e]pyrene (BeP) and perylene (Pery).
To avoid detecting differences in contaminant concentrations between sampling years that could be attributed to spatial variation within an area of the harbour, only the co-located stations from 1990–2014 were selected for statistical analysis. For contaminant data from co-located sites sampled in 1990, 2000 and 2014 (metals excluding mercury), a one-way analysis of variance (ANOVA) was performed for the OH area only (N = 8 sites); tests were not conducted for all other areas of the harbour which had N = 1 or N = 2 sampling sites (Table S1, available in the SI). If the data were not normally distributed, as determined by the Shapiro-Wilk test, data were log(x) transformed; if the normality test failed after transformation, a Kruskal-Wallis One-Way ANOVA on Ranks was performed. If the ANOVA p value was significant, a post hoc multiple comparison test (Tukey or Holm Sidak test) was performed at p < 0.05.
For all contaminant data collected from co-located sites in 2000 and 2014 (25 sites), paired t-tests were conducted for 3 areas of the harbour: WA (6 sites), RR (6 sites) and OH (10 sites) (Table S1, available in the SI); analyses were not performed for the WE and NE areas of the harbour which had minimal data points across years (N = 1 or N = 2). The Wilcoxon Signed Rank Test was performed if the data were not normally distributed after log(x) transformation. All statistical procedures were performed using SigmaPlot version 12.5, from Systat Software, Inc. (San Jose, California, USA, www.sigmaplot.com).
Where applicable, sediment contaminant concentrations were compared to Canadian sediment quality guidelines (CCME, 1999), and for nickel and total PAHs to Ontario Provincial sediment quality guidelines (MacDonald et al., 2000).
Results and discussion
Spatial distributions and temporal trends in metals in Hamilton Harbour Sediment–Chromium, Copper, Mercury, Nickel, Lead and Zinc (1990–2014)
Metal concentrations have remained stable or have decreased overall since 1990 in all areas of the harbour except in WA for copper, lead and zinc, and in RR for copper (Figure 2; Figure S1, available in the SI). In the OH area of the harbour, significant decreases occurred for 4 of the 5 metals in 2000 and/or 2014, compared to 1990; chromium, nickel and lead concentrations were significantly lower in both 2000 (Cr: p = 0.006, t = 3.327; Ni: p = 0.022, t = 2.785; Pb: p = 0.017, t = 2.889) and 2014 (Cr: p < 0.001, t = 4.471; Ni: p = 0.002, t = 3.98; Pb: p < 0.001, t = 4.397) versus 1990, and significantly lower zinc concentrations Zn (p = 0.022, Q = 3.900) were observed in 2014 versus 1990. Copper showed overall increases since 1990 in WA as well as RR (Figure 2; Figure S1, available in the SI), although there were no significant differences detected between years (p = 0.133, F = 2.221).
Over the period of 2000–2014, metal concentrations remained stable or decreased in all areas of the harbour except in WA, where chromium, lead and zinc showed increases; these increases were not significant due to the high inter-year variability (Figure S2, available in SI). In 2014, metals were overall most elevated in WA (Table S2, available in the SI), due to one site that exhibited the highest concentrations for all metals. In the OH, there were significantly decreased metal concentrations in 2014 versus 2000 for chromium (p = 0.025, t = 2.680), nickel (p = 0.027, t = 2.632), lead (p = 0.009, t = 3.308), zinc (p = 0.035, t = 2.473) and mercury (p = 0.027, z = −2.191). In assessing median concentrations, mercury was fairly similar across 4 of the 5 areas of the harbour (WA, RR, OH, NE; median values ranging from 0.22–0.35 µg g−1) and nickel was similar across all 5 areas of the harbour (median values ranging from 22–65 µg g−1; Table S2, available in the SI). Chromium and copper concentrations were highest in WA (median concentrations of 95 and 99 µg g−1, respectively), compared to other areas of the harbour (Cr 40–75 µg g−1 and Cu 60–86 µg g−1), whereas lead and zinc were highest at RR (median concentrations of 153 and 1096 µg g−1, respectively), compared to other areas of the harbour (Table S2, available on the publisher's website). Metal concentrations were lowest in the WE for 4 of the 6 metals; copper and nickel were similar to other areas of the harbour (Table S2, available in the SI).
Metal concentrations (means) have decreased in some cases to below the Canadian Sediment Quality Guidelines Probable Effect Level (PEL), e.g. mercury in all areas, chromium in all areas except WA and lead in the WE (Figure 2). Except for lead and zinc, concentrations were between the PEL and the Threshold Effect Level (TEL), or for nickel between the Lowest and Severe Effect Levels (LEL and SEL), the range where adverse effects may occasionally occur (CCME, 1999; MacDonald et al., 2000). Lead and zinc have remained above the PEL in most areas of the harbour, although concentrations were approaching the PEL in 2014 except for WA and perhaps RR (Figure 2c). Exceedances of the PEL guidelines indicate the potential for adverse impacts on benthic community health, and as a result a barrier to restoration of beneficial uses prior to delisting the harbour as an AOC. However, it should be noted in the case of Hamilton Harbour that high sediment oxygen demand has been implicated as an additional factor in considering the causes of sediment toxicity (Krantzberg and Boyd, 1992; Krantzberg, 1994).
The metals datasets for 1990, 2000 and 2014 represent the most spatially comprehensive information available for Hamilton Harbour sediment. A comparison of these data sets provides insight into sources of metals contamination in the harbour over the last three decades. The spatial distribution of metals in 1990 (Figure 2a) is quite homogeneous, compared to 2000 and 2014. In 2000 and 2014, the highest metals concentrations were observed in the RR and WA areas (Figures 2b and c), and illustrate the continuing impact of legacy industrial contamination on sediment quality. However, the relatively lower concentrations of metals in the other areas of the harbour (WE, NE and OH) also indicate that loadings of metals to the harbour decreased substantially over the period of 1990–2000 (Figures 2b and c). Temporal trends in contaminants in Hamilton Harbour bottom sediment for the period of 1990–2014 have been summarized in tabular form (Table 1). There were two options for investigating temporal trends over the periods 1990–2000–2014 and 2000–2014 as per Figure 2. Including all clustered sites within the five designated harbour zones sampled for each period (1990, 2000 and 2014) would have resulted in higher N values, and presumably more definitive statistical trends; however, this approach would also have introduced the issue of considerable inter-station variability in contaminant concentrations, as is common throughout Hamilton Harbour. As a result, we chose to base our statistical trends solely on data for co-located sites for all three sampling years.
Spatial distributions and temporal trends in polycyclic aromatic hydrocarbons in Hamilton Harbour sediment (2000–2014)
Total PAHs showed decreasing concentrations throughout the harbour from 2000 to 2014 in all areas of the harbour (Figure 3), with significantly lower concentrations observed in the OH (p = 0.012, t = 3.140). Overall, total PAHs were highest at RR (11.8–499 µg g−1), followed by WA (1.0–81.5 µg g−1) and the OH (1.4–56.9 µg g−1) and have remained above the LEL (4 µg g−1) in all areas of the harbour (Table S2, available in the SI). Lower PAH concentrations were found in the NE (8.3–12.6 µg g−1) and WE (4.3–11.5 µg g−1) areas of the harbour. PAH distribution is consistent with sources from both the RR and WA. From 1977 to 1987, loadings of suspended sediments to Hamilton Harbour decreased from an average of 140,000 kg d−1 to 50,150 kg d−1 (Hamilton Harbour RAP, 1989); correspondingly, contaminants associated with particulate loadings would concomitantly be reduced. Control measures implemented by industry and wastewater treatment plants also reportedly contributed to decreased loadings of PAHs (Hamilton Harbour RAP, 1989). Presumably, decreased particulate loadings continued to influence the rate of decline of contaminant concentrations in the past 20 years as wastewater plant and stormwater management best practices have continued to evolve and find implementation in the surrounding watershed. According to Sofowote et al. (2008) mobile sources are important contributors to PAH in the harbour and its associated watershed that can enter the harbour through a variety of vectors, including direct atmospheric entry and urban runoff via creeks, storm sewers and wastewater treatment plants. However, urban expansion in the watershed is presumably resulting in continuing production of particulates and their associated contaminants, which presents a challenge for Hamilton Harbour in development of strategies to reduce inputs from mobile sources. Mobile sources remain particularly important given that reductions in other particulate loadings have been achieved through improved industrial standards and wastewater treatment technology (Sofowote et al., 2008).
Historically, the highest concentrations of PAHs in Hamilton Harbour, both in bottom sediments and suspended sediments, have been associated with stations near Randle Reef (RR) and in the WA area (Murphy et al., 1990). The maximum depth of water in the OH areas is roughly 20 m; this area reflects both ongoing loadings of suspended sediment entering the harbour, and resuspension and subsequent circulation and deposition of bottom sediment (Nriagu et al., 1983). As a result, contaminant concentrations in OH are generally representative of harbour-wide contamination from industrial loadings, sewage treatment plants, urban non-point sources and sediment resuspended and redistributed from Randle Reef. Consistently high PAH concentrations have also been observed in WA in the southeast area of harbour. This area can be influenced by sedimentary exchange with Lake Ontario, but the high PAH concentrations were primarily contributed by contaminated sediments in Windermere Arm and the Dofasco boat slip (Labencki, 2008) originating from areas in east Hamilton characterized by intensive historical industrial activity.
Relatively lower PAH concentrations have historically been observed at WE and to a lesser degree NE. The shallower area of the northeastern corner of the harbor (NE) and the western end of the harbour (WE) are generally outside areas of the harbour typically subjected to general particulate circulation patterns. The area ranging from the northeast corner of the harbour to the Burlington Canal is a broad littoral zone. This area was impacted both by Indian Creek, which discharges an urban watershed characterized by major highways, and by suspended sediments from resuspension events. Prior to the mid-1990s, transport of suspended sediment was driven by a combination of prevailing winds and wave-generated orbital currents (Brassard and Morris, 1997). In the mid-1990s (completed in 1996), artificial islands were constructed as bird habitat, with the result that the previously predominant current regimes, and correspondingly the distribution and deposition of suspended sediments, in this area of the harbor may have been disrupted (Coakley et al., 2002; Rao et al., 2009).
In order to study the influence of different source types on PAH contamination, individual compound profiles were assessed spatially according to the five designated areas of the harbour. The relative contributions of selected individual PAH compounds are shown in Figure 4. Definitive differences in PAH profiles between areas of the harbor are difficult to distinguish; however, the RR profile exhibited the greatest contributions from lower molecular weight compounds including naphthalene (N), phenanthrene (Ph) and fluoranthene (Fl); these compounds are associated with coal combustion (Heit, 1985; Gogou et al., 1996). Correspondingly, the RR PAH profile exhibited the lowest contribution from the highest molecular weight compounds; dibenzo[a,h]anthracene (Db[a,h[A), indeno[1,2,3-cd]pyrene (Ind[1,2,3-cd]P) and benzo[g,h,i]perylene (B[ghi]P) (Figure 4). In addition, Sofowote et al. (2008) determined using two factor analysis methods that coal-derived sources of PAHs were the primary contributors to contamination of suspended sediments in the areas near Randle Reef. The WA sediment profile exhibited the lowest contributions from the volatile low molecular weight PAHs including naphthalene and phenanthrene; we attribute this observation to weathering of PAHs originating in the watershed and discharged to the harbor via Red Hill Creek that would result in volatilization of these compounds, with an associated decrease in their relative contributions to the overall PAH profile. The general similarity between the RR and OH profiles would appear to support the implication of Randle Reef as a significant source of PAH contamination to far field areas of the harbor through the continual resuspension and transport of contaminated sediment.
Spatial distributions and temporal trends in polychlorinated biphenyls in Hamilton Harbour sediment (2000–2014)
Concentrations of total PCBs have remained relatively stable or showed slight declines in all areas of the harbour except WA (Figure 5) which exhibited greater inter-station variability than other areas and where no statistically significant differences could be determined between years (Table S2, available in the SI). Total PCBs have remained above the PEL (0.277 µg g−1) in all areas of the harbour except in the WE in 2014 and were overall highest throughout the time period in WA (range of 0.07–5.27 µg g−1), compared to the RR (0.23–1.70 µg g−1), OH (0.08–1.1 µg g−1), and NE and WE (≤0.60 µg g−1) areas of the harbour (Table S2, available in the SI); these observations were consistent with a source originating in WA. There is a paucity of information on PCB loadings to Hamilton Harbour over the last 15 years. In 1982 and 1983, loadings of PCBs to the harbour were approximately 50 kg yr−1 but decreased over the period until 1989 to approximately 9.9 kg yr−1 (Hamilton Harbour RAP, 2008). Labencki et al. (2011) reported surficial sediment PCB concentrations in 2000 that ranged from 600 ng g−1 200 ng g−1. Historical contamination at depth in Windermere Arm sediments is as high as 150 µg g−1 (Zeman and Patterson, 2003). Windermere Arm is discharged by Windermere Basin, a receiving body that historically received inputs from both the Hamilton Sewage Treatment Plant, combined sewer overflows (CSOs), and a primary tributary, Red Hill Creek. It has been reported that PCB loadings to Windermere Arm are approximately four times more than the loadings to the rest of the harbour (Hamilton Harbour RAP, 2009). These PCB levels are reflective of historical contamination from industrial activities along the southern shoreline, and ongoing resuspension and redistribution of contaminated sediment as a result of weather events, navigation and subsequent transport to open water areas of the harbour. Recently, a source of PCBs discharging into Windermere Arm from the Strathearne Avenue boat slip was identified and a follow-up study initiated (Hamilton Harbour RAP, 2009; Labencki, 2008). Although additional data characterizing the apparent ongoing source will be required to confirm the hypothesis, this source may contribute to PCB congener profiles observed throughout the harbor (Burniston et al., 2016).
With the implementation of the Randle Reef contaminated sediment remediation project, research and monitoring programs in Hamilton Harbour have placed additional focus on other potential priority areas, most notably Windermere Arm. Our previously reported study on Hamilton Harbour suspended sediments (Burniston et al., 2016) determined that some areas of the harbour appeared to be associated with distinct PCB profiles. A number of general determinations were made regarding PCB profiles and potential sources:
PCB homolog profiles were highest in contributions from the penta and hexa homologs at stations in the OH; these homolog profiles most resembled Aroclor 1254/1260, the Aroclor associated from those time periods of highest historical suspended sediment PCB contamination in the harbour, i.e. the late 1980s;
Relative to the OH area, sampling stations in WA exhibited a greater contribution from the hexa and hepta homologs;
Inter- and intra-annual variability in PCB homolog profiles at stations in WA was minimal, but monthly variability in homolog profiles in the Strathearne slip, particularly for the tri and tetra homologs, was considerable. The Strathearne slip area of WA exhibited high PCB concentrations and congener profiles enriched in tri and tetra homologs;
The WA area was implicated as a source of PCB contamination to other areas of Hamilton Harbour.
The 2011 work on bottom sediments has corroborated some of these determinations. For stations in WA, the greatest inter-station variability in PCB homolog profiles was observed for the tri and tetra compounds (Figure 6a), which was also observed for PCB homolog profiles in suspended sediments (Burniston et al., 2016). In addition, the hexa and hepta homologs represented the greatest relative contributions to total PCBs for the WA samples (Figure 6a). In general, the PCB homolog profiles were quite similar for WA (Figure 6a), OH (Figure 6b), WE (data not shown) and NE (data not shown). However, the homolog profile for the RR samples showed little inter-station variability, and was dominated by the penta and hexa congeners (Figure 6c), which more closely approximated the profile in suspended sediments for the OH area of the harbour (Burniston et al., 2016). The distinctness of the RR area PCB homolog profile, compared to all other areas of the harbour, and its similarity to the profile of suspended sediments in the OH area, provides further indications of the influence of Randle Reef contaminated sediments on open water parts of the harbour and further justifies ongoing remedial activities.
Temporal trends in metals, PAHs and PCBs exhibited declines in most areas of the harbour over the period of 1990–2000 to 2014. There have been statistically significant (p ≤ 0.05) declines in metals (Zn, Pb, Cr, Ni, Hg) and PAHs in the OH zone only, although the data also demonstrate a qualitative reduction in concentrations of some metals and PAHs for the WE, NE and RR zones, and concentrations of PCBs exhibited a qualitative decline in zones WE, NE and RR. Some metals (Zn, Cu, Pb, Cr) and PCBs showed a qualitative increase in zone WA, although the variability was high and the zone characterization was sensitive to the inclusion of certain individual sampling locations. We attribute these reductions to management actions including environmental stewardship, wastewater treatment plant upgrades and stricter industrial emission standards.
Windermere Arm, an area impacted by historical industrial activities along the southeastern shoreline area of the harbour, was an exception as temporal trends in some metals and PCBs showed overall increases. After the apparent initial reductions, the harbour was then subjected to loadings from historically-contaminated areas, i.e. Randle Reef and Windermere Arm, and additional ongoing sources including mobile and industrial emissions and urban run-off. Due to the nature of the harbour as a confined embayment discharging a large urban/industrial watershed, sediment quality in Hamilton Harbour is expected to continue to be reflected in relatively high pollutant concentrations, compared to those considered to be ambient background levels in sediments in other areas of the Great Lakes. This study also showed the impact of areas of contaminated bottom sediment and point sources on sediment quality in other areas of the harbour. However, it is also expected that remedial measures with respect to legacy contaminated sediment sites (Randle Reef) and mitigation/remediation of ongoing sources (Strathearne slip in Windermere Arm) will ultimately result in improvements in bottom sediment quality throughout the harbour. In summary, the synthesis of sediment data from the past 25 years indicated various management actions over the past several decades have generally resulted in the desired long-term outcomes, i.e. reductions on contaminant loadings to the harbour.
The general downward trajectories in temporal trends for contaminants in most areas of the harbour indicate progress toward restoration of beneficial uses and subsequent delisting. With respect to metals and sediment toxicity, sediment oxygen demand and periodic hypolimnetic anoxia will continue to present significant constraints to recovery. The encouraging reduction in PAH concentrations in areas outside of Randle Reef indicate that the sediment remediation project currently underway in this area could result in further declines in PAH contamination throughout the harbour, and a corresponding reduction in fish tumours and other deformities that would represent further progress toward delisiting. The issue of PCB contamination, particularly in Windermere Arm, clearly remains a concern with respect to restrictions on fish consumption. Further monitoring in this area following abatement actions at a current suspected ongoing source could provide an additional indication of positive progress toward delisting.
The authors thank Jake Kraft, Craig Logan, Jennifer Webber and Sherri Thompson of the Water Science and Technology Directorate of Environment and Climate Change Canada, and Technical Operations of the Water Science and Technology Directorate for technical support.
This work was conducted with support from Environment and Climate Change Canada's Great Lakes Action Plan.
Supplemental data for this article can be accessed on the publisher's website.