During construction of the trans-Canada railway in the 1870s, rich deposits of copper and nickel were found in the Sudbury basin located in northeastern Ontario (Canada). This discovery led to concentrated mining and smelting activities in the area (Winterhalder 1995). In 1888, open pit roasting of these ores began, which led to the opening of Nickel Co. (currently known as Vale Inco, est. 1902), the Coniston smelter (1913), and Falconbridge Limited (currently known as Xstrata Nickel, est. 1928) (Tropea et al. 2010). By the 1960s, the Sudbury smelters were emitting over 2.5 million tonnes of sulphur dioxide per year, making the area one of the world’s largest sources of acid-generating pollution in the world. Due to concerns over the widespread environmental damage from mining operations, emissions were reduced by ~ 90% by 1994 (Snucins and Gunn 1998), and even further to the present day.
Particulate emissions from smelters and mining operations contaminated much of the Sudbury region and surrounding areas (Pearson et al. 2002). Early smelting emissions resulted in the loss of large amounts of vegetation, and severe acidification of terrestrial and aquatic environments, with soil acidity and metal concentrations highest in areas near smelters (Winterhalder 1995). The loss of vegetation left Sudbury soils exposed, and they became severely eroded. Atmospheric deposition of contaminants (i.e., strong acid, metals) altered water and sediment quality, and, in the case of the metals, were toxic to aquatic organisms (Fleeger et al. 2003). Moreover, in the 1960s, increases in cyanobacterial and algal blooms were observed in many Sudbury lakes because of high nutrient inputs linked to soil erosion from industrial activities and urbanization.
Whilst all lakes and catchments within KPP and Sudbury were likely affected to some degree by acidifying emissions, there is much variation in the magnitude, timing, and extent of impacts from acidification. For example, lake acidification and metal contamination in Sudbury peaked in the 1960s, while the acidification of lakes in KPP peaked in the 1970s and 1980s. This is likely because the KPP lakes are located farther from the mining and smelting activities, and therefore received longer range atmospheric deposition of SO2, as opposed to being directly exposed (Keller et al. 2008). In addition, this variation is due, in part, to differences in natural buffering capacity, which is especially notable in lakes that were thought to be acidic prior to the onset of smelting activities [i.e., Ruth-Roy Lake (KPP) and (Sudbury)] (Belzile et al. 2004). Meanwhile, delayed acidification in some lakes in KPP was believed to be due to pockets of carbonate-containing rocks in the watersheds (i.e., Lumsden and Acid lakes; Hindar and Henriksen 1998).
Given the lack of direct, long-term monitoring data for most lakes, the effects of acidification were clearly demonstrated in many Sudbury lakes using paleolimnological approaches, conducted mainly in the 1980s and early-1990s. For example, Baby Lake, which is located ~ 1 km away from the Coniston smelter in Sudbury, showed marked changes in acid sensitive algae (i.e., diatoms and chrysophytes) that were coincident with the onset of major smelting operations and acidification, as well as subsequent recovery noted especially in chrysophyte assemblages (Dixit et al. 1992a,b). In a study of 151 lakes in the Sudbury region, Dixit et al. (2002) concluded that 77% of modern diatom species variation could be attributed to the combined effects of various measured environmental variables (such as lake catchment characteristics, lake size, elevation, pre-industrial pH, metals). Dixit et al. (2002) also applied a diatom-based pH inference model to conclude that peak acidification occurred in the ~ 1970s in KPP and ~ 1960s in Sudbury, but also that the large reductions in sulphur dioxide emissions had resulted in early signs of recovery. Collectively, these studies, conducted more than three decades ago, demonstrated the effectiveness of diatom and chrysophyte assemblages to track changes in limnological variables related to acidification (Dixit et al. 1992a,b).
Some re-assessments of acidification and recovery patterns in select Sudbury lakes, using a variety of paleolimnological indicators, have shown that biological recovery trajectories can be varied (Tropea et al. 2010). Although pH is still identified as the strongest spatial variable shaping diatom assemblages across Sudbury lakes (Cheng et al. 2022), biological assemblages in many lakes show limited recovery, and have not returned to their pre-industrial condition. Recent warming and legacy metal contamination have been cited as possible factors complicating biological recovery (Tropea et al. 2010; Simmatis et al. 2021; Cheng et al. 2023).
Recent climate warming has impacted Sudbury-area and KPP lakes, especially over the last few decades. Snucins and Gunn (2000) used temperature profiles to examine the thermal structure of 86 relatively small (< 550 ha) lakes in KPP to show that mixing regimes were strongly influenced by water clarity and lake depth. Snucins and Gunn (2000) found that 84% of the lakes that failed to thermally stratify were shallow (depth < 5 m) and had low DOC concentrations (< 1 mg/L). This is relevant to lakes in KPP as the small lakes of the La Cloche Mountain range are some of the clearest lakes in North America. These findings suggest that DOC decreases due to a changing climate and/or acidification will have a significant effect on various physical and chemical lake characteristics. As DOC concentrations slowly return to pre-industrial levels (Meyer-Jacob et al. 2019), some of these thermal changes may be reversed.
In addition to the impact to and recovery of lake chemistry and biological communities from industrial activities, urbanization is complicating ecological trajectories. Tropea et al. (2011) studied four Sudbury lakes (Ramsey, Nepahwin, McFarlane and Richard lakes) to investigate the impact of urbanization and showed that diatom community composition has changed drastically over the past ~ 150 years as a result of sewage effluent, excess nutrients from fertilizers, shoreline alteration, and altered hydrology. Similar distinctions among urban and non-urban lakes in the Sudbury region have also been found in cladoceran taxa, driven by differences in aqueous specific conductance and the indirect effects of nutrient enrichment (Simmatis et al. 2022). Urban Sudbury lakes highlight the potential for altered ecological recovery trajectories due to a combination of legacy and modern stressors.
Despite the many earlier studies on the degradation and subsequent recovery of aquatic ecosystems in KPP and Sudbury, little is known about how more recent conditions (i.e., last three decades) relate to background (pre-industrial) conditions, or how biological recovery trajectories may be affected by a rapidly changing climate. To develop this long-term perspective, we selected five study lakes for multi-proxy paleolimnological analyses, including subfossil remains of algae (diatoms), inferred whole-lake chlorophyll-a (VRS-Chl a) and lake water dissolved organic carbon (DOC) concentrations. We examine information preserved in lake sediment to obtain a long-term (past ~ 200 years) record of ecological change, including assessing how recent warming has affected recovery trajectories. The specific study objectives are to: (1) reconstruct fossil diatom assemblages, sediment Chl a and lake-water DOC concentrations over the past several decades to assess trends in recovery from acidification, and, in particular, to compare present-day with pre-industrial conditions of five lakes in KPP and Sudbury; (2) explore patterns of potential ecological recovery from acidification over the past several decades; and (3) collectively use paleolimnological, long-term monitoring and meteorological data to explore the potential drivers and trajectories of KPP and Sudbury ecosystem changes.
Study region
Sudbury (46.48959° N, 80.99011° W) is located near the southern margin of the Precambrian Shield in northeastern Ontario, on the margins of the Sudbury Basin, a topographic low attributable to an ancient (1.8 mya) meteor impact and subsequent deformation of the Earth’s crust. The Sudbury basin is ~ 60 km in length and ~ 27 km in width; its structural axis is aligned northeast to southwest (Dirszowsky 2020; Tropea 2008). The rugged topography of the Shield, and basin structure modified by long-term denudation and glacial activity, has resulted in the highest number of lakes contained within a city for any urban area within Canada. The Sudbury region has a humid continental climate [Climate normal (1981–2010): mean daily minimum temperature of -17.9°C in January; mean daily maximum temperature of 24.8°C in July; mean monthly maximum precipitation of 101.1 mm in September; mean monthly minimum precipitation of 51.1mm in February; total annual precipitation of 903.2 mm; average annual wind speed of 14 km/hr; and an average frost-free period of 136 days (Sudbury station A ID 71730; ECCC 2024)].
Killarney Provincial Park (46.1333° N, 81.4167° W), located ~ 60 km southwest of Sudbury (Fig. 1), contains over 200 lakes and ponds. These freshwater ecosystems exhibit large differences in size, elevation, water chemistry, and catchment characteristics. The majority of KPP is underlain by the Lorrain formation, which is comprised of feldspathic and kaolinitic sandstone and orthoquartzite (Hindar and Henriksen 1998). The folds of quartzite rock form the ancient geological formation that characterize the La Cloche Mountain range. Located in the Ontario Shield Ecozone, KPP’s mean minimum temperature is approximately − 14°C in January, with a mean maximum temperature of approximately 24°C in July. KPP’s mean monthly precipitation reaches a maximum of over 75 mm in April and October, and a minimum of approximately 35 mm in February (Belzile et al. 2004).
Study lakes
Study lakes were chosen to reflect select limnological gradients and varying levels of impacts. This work builds on previous paleolimnological work from the 1980s and 1990s in KPP and the Sudbury region. Sediment cores for diatom analyses have not been collected from these lakes in ~ 30 years, which include lakes in KPP (Ruth-Roy and Johnnie lakes) and Sudbury (Tillie, Crooked, and Baby lakes; Fig. 1).
Ruth-Roy Lake (46.0961° N, 81.2467° W): Located in KPP, ~ 45.2 km from the Copper Cliff smelter (Fig. 1; Table 1), Ruth-Roy Lake has a maximum depth of 18 m and surface area of 54.5 ha. Ruth-Roy Lake has undergone numerous paleolimnological assessments. Sediment cores collected between 1975 and 1980 were analyzed for diatom and chrysophyte assemblages in core ‘tops’ (surface sediments representing current conditions) and core ‘bottoms’ (generally from > 30 cm deep, representing pre-industrial conditions) (Dixit et al. 1992c). Earlier analyses showed that Ruth-Roy was one of the most naturally acidic lakes in the region prior to smelting (inferred pH ~ 4.9; Dixit et al. 2002), possibly because the lake drains quartzite ridges with little to no soil. Although being naturally acidic, Dixit et al. (2002) noted that the lake acidified further (to pH ~ 4.58) due to acid precipitation. Simmatis et al. (2023) recently re-assessed long-term changes in diatoms, cladocerans, inferred whole-lake primary production, and inferred lake-water DOC in Ruth-Roy Lake. Pre-industrial assemblages confirmed that Ruth-Roy Lake was naturally acidic but began to respond to smelting emissions as early as ~ 1920, which was attributed to its naturally low buffering capacity and, therefore, high sensitivity to acidifying emissions (Simmatis et al. 2023).
Table 1
Summary of lake characteristics and recent measured water chemistry for the five focal study lakes.
Parameter | Units | Baby | Crooked | Ruth-Roy | Johnnie | Tillie |
Distance from smelter | km | 1 S (Coniston) | 6.3 S (Copper Cliff) | 45.2 SW (Copper Cliff) | 46.2 S (Copper Cliff) | 62.1 N (Coniston) |
Elevation | masl | 224 | 310 | 217 | 213 | 404 |
Surface area | ha | 11.9 | 26.3 | 54.5 | 342.3 | 76.7 |
Max. Depth | m | 22.5 | 8 | 18 | 33.6 | 11 |
Mean Depth | m | 9.6 | 3.8 | 4.1 | 7.9 | 4.7 |
DOC (2018)1 | mg/L | 1.6 | 8.4 | 2.5 | 4.5 | 9.4 |
pH (2018)1 | pH units | 7.3 | 6.7 | 5.8 | 6.5 | 5.9 |
SO4 (2018)1 | mg/L | 6.70 | 5.25 | 2.95 | 3.55 | 3.45 |
Source: 1MECP dataset (Sudbury full water chemistry - used with permission) |
Johnnie Lake (46.0915° N, 81.2400° W): Johnnie Lake, located in KPP, ~ 42.6 km from the Copper Cliff smelter (Fig. 1; Table 1), has a maximum depth of 33.6 m and surface area of 342.3 ha. Previous studies done on Johnnie Lake include zooplankton surveys, precipitation chemistry and wet deposition data collection, paleolimnological analyses, and water chemistry analysis (Szkokan-Emilson et al. 2010; Keller et al. 2003; Snucins et al. 2001). Keller et al. (2003) collected lake and precipitation chemistry, and wet deposition data, between 1978 and 1998 from Johnnie Lake showing the lake experienced significant declines in calcium (measured from lake water precipitate; 140 eq/L in 1980 to ~ 80 eq/L in 2001) as well as increases in alkalinity (-7 eq/L in 1980 to 11 eq/L in 2001). Keller et al. (2003) also found that DOC concentrations increased significantly between 1988 and 2001 and that lake water pH closely tracked precipitation chemistry. Meanwhile, Snucins et al. (2001) collected water samples (from 1996–1997) for chemical analyses from several KPP lakes, including Johnnie Lake, and found that lakes with catchments that are made up of more easily weathered bedrock formations acidified the least, relative to pre-industrial diatom-inferred pH, and showed the largest increases in pH between 1981 and 1999. A recent paleolimnological top-bottom comparison of cladoceran and chironomid assemblages demonstrated that subfossil invertebrate assemblages were somewhat similar between pre-1880 and post-2010 sediments (Simmatis et al. 2021, 2022, 2023). Modern invertebrate assemblages in Johnnie Lake are most similar to lakes far from the original smelters (e.g., Whitepine Lake in Lady Evelyn-Smoothwater Provincial Park; Simmatis et al. 2022, 2023). Currently, Johnnie Lake is slightly acidic (pH of ~ 6.3) and slightly above its inferred pre-industrial pH level (pH of 6.1).
Baby Lake (46.4612° N, 80.8651° W): Located ~ 1 km from the Coniston smelter (Fig. 1, Table 1), Baby Lake has a maximum depth of 22.5 m and a surface area of 11.9 ha. Baby Lake has been subject to multiple studies, including water chemistry analysis (i.e., metals, sulphate and lake water pH), zooplankton identification, and paleolimnological analyses (Hutchinson and Havas 1986; Havas et al. 1995; Smol et al. 1998; Simmatis 2021).
Hutchinson and Havas (1986) collected water chemistry data to show that when the Coniston smelter closed (1972), lake-water sulphate concentrations decreased by 50% (from 60 mg/L in 1968 to 30 mg/L in 1983), lake-water copper concentrations decreased by 92% (from 0.78 mg/L in 1971 to 0.06 mg/L in 1984), and nickel concentrations decreased by 87% (from 3.2 mg/L in 1970 to 0.41 mg/L in 1984). Trends in lake-water metal concentrations collected by Hutchinson and Havas (1986) were reflected by trends in more recent sedimentary metal concentrations (Simmatis 2021). Notably, concentrations of copper, nickel, lead, and zinc have not returned to pre-industrial concentrations (Simmatis 2021). Havas et al. (1995) additionally sampled Baby Lake for water chemistry and zooplankton to track biological changes over decadal timescales (1970-1990s). They found the lake had much lower phytoplankton biomass and species richness than other Shield lakes of a similar pH (6.8) in 1986.
Dixit et al. (1992c) completed a paleolimnological analysis and showed that long-term pH, inferred from diatom assemblages, recorded dramatic acidification of Baby Lake, changing from an inferred pH of 6.5 in 1940 to 4.2 in 1975. Beginning in 1972 (following the closure of the Coniston smelter) diatoms and especially chrysophyte assemblages in Baby Lake showed significant recovery (Dixit et al. 1992b). Despite recovery in algal indicators, cladoceran taxa showed limited recovery, especially littoral taxa (Simmatis 2021). Sedimentary cladoceran assemblages have indicated reduced littoral diversity and increased abundance of generalist taxa (i.e., Chydorus brevilabris and Bosmina spp.) after ~ 1925 without recovery to pre-industrial assemblage composition (Simmatis 2021). Further, few chironomid head capsules were found in Baby Lake, which may be related to sediment type and high inorganic content (Simmatis 2021). Baby Lake was amongst the most acidic of our study lakes during the period of peak emissions from Sudbury smelters and is therefore an interesting contrast to KPP lakes that are located more distant from smelter emissions.
Tillie Lake (47.0425° N, 81.0008° W): Among the study sites, Tillie Lake is located the furthest (62.1 km) from the nearest smelter (Fig. 1, Table 1). It has a maximum depth of 11.0 m and a large surface area of 76.7 ha. Tillie Lake is also located at the highest elevation of any of the study lakes (404 masl), had the highest DOC value of 5.5 mg/L in 1975, and showed recent increases in total phosphorus (Keller et al. 2019). A recent paleolimnological top-bottom comparison of cladoceran assemblages showed some similarities between pre-1880 and post-2010 sediments (Simmatis 2021). Chironomid counts were inadequate for analysis and interpretation (Simmatis 2021). Tillie Lake is not well-researched compared to the other study lakes, and thus a thorough paleolimnological analysis was needed to place recent limnological data within a long-term context.
Crooked Lake (46.4185º N, 81.0331º W)
Crooked Lake is located 6.3 km from the Copper Cliff smelter (Fig. 1, Table 1), has a maximum depth of 8.0 m, and a surface area 26.29 ha. Dramatic declines in acidity have been observed in severely damaged urban lakes close to the Sudbury smelters, such as Crooked Lake (Keller et al. 2019). Crooked Lake had the most acidic pre-industrial diatom inferred pH values (4.0) of the selected study lakes (Dixit et al. 1992c), and one of the most acidic pH values in 1975 (4.3) (Snucins and Gunn, 1998). Simmatis et al. (2022) completed top-bottom analysis of cladocerans, finding that modern assemblages had higher relative abundances of generalist taxa (i.e., Chydorus brevilabris) and reduced littoral diversity compared to pre-industrial conditions. Relative to other Sudbury and KPP lakes, Crooked Lake experienced a moderate amount of change (Simmatis et al. 2022). Inadequate chironomid remains were recovered to allow for further analysis or interpretation (Simmatis 2021). Crooked Lake is an interesting example of a naturally acidic Sudbury lake.