Introduction

Lake sediment cores contain a large amount of information about the state of lakes and their catchments over time, which can be linked to climatic, land-use, and other environmental and anthropogenic changes, including mining, smelting, agriculture, forestry, transport, and other industries. Aquatic environments often accumulate metals and other anthropogenic pollutants that have negative impacts. High concentrations of metals and metalloids (‘metals’ henceforth refers to both metal and metalloid species) such as arsenic (As), cadmium (Cd), copper (Cu), mercury (Hg), lead (Pb), and zinc (Zn) are toxic to aquatic life (Sharma and Agrawal 2005, Tchounwou et al. 2012), and elevated concentrations have been found in remote and sparsely inhabited locations, such as Antarctica (Magesh et al. 2021) and remote parts of Australia (Roach 2005, Schneider et al. 2019). Measuring past concentrations and identifying sources can provide context to assess modern concentrations. This is particularly important for the Southern Hemisphere where geochemical sediment core data are sparse.

The Central Highlands region of Tasmania, Australia (Fig. 1), contains many natural and anthropogenic lakes that not only generate hydro-electric power and supply irrigation, but also support biodiverse ecosystems and popular recreational fisheries. Despite their importance, minimal attention has been given to their biogeochemistry and sedimentary composition. These lakes may have been affected by Aboriginal land management practices, European settlement, sheep grazing, agriculture, infrastructure, damming, atmospheric transport of pollutants from mining and smelting, aerosols and particulates from bushfires, and lead shot from animal hunting. Many of these activities involve the use or mobilisation of metals that may be delivered to these lakes and could adversely affect their ecosystems.

Fig. 1
figure 1

Map of Tasmania, Australia, showing the locations of major mining operations (Queenstown–purple triangle, and Rosebery–blue triangle), the five sediment cores analysed in this study (purple circles), sediment cores analysed in previous Tasmanian metal studies (Harle et al. 2002, Saunders et al. 2013, Schneider et al. 2019) (red, green, and blue circles, respectively), and ‘Tyler’s Line’ denoting the approximate interface between major climatic, geologic, vegetation, and limnologoical differences in Tasmania (Tyler 1992) (orange line)

Full size image

Mining is a major potential source of metals to the environment in Tasmania. The largest and most productive mines were centred in the west of the state. In the late 19th, and the first half of the twentieth century, this region was the Southern Hemisphere’s largest copper producer (Schneider et al. 2020). These smelting and mining operations have affected the immediate neighbouring environments including sediments, streams, lakes, and terrestrial ecosystems (Harle et al. 2002, Augustinus et al. 2009, Schneider et al. 2019). The majority of this contamination is through direct run-off and short-range aerial transport, although Saunders et al. (2013) identified metal contamination in Bathurst Harbour (150 km southeast of the major mining sites) potentially due to long-range atmospheric transport. Similarly, Schneider et al. (2019) identified elevated metal concentrations in lakes up to 130 km southeast of major mining sites and contended that these derived from long-range atmospheric transport.

This study further explores the hypothesised long-range atmospheric transport by investigating metal (and metalloid) concentrations in sediment cores from five lakes (‘lakes’ henceforth refers to both lakes and lagoons) that lie between 86 and 125 km directly to the east of the West Coast mining belt (Fig. 1), to:

  1. 1.

    Identify metal sources, particularly to identify the impact of long-range atmospheric deposition from major Tasmanian mining sites.

  2. 2.

    Assess how damming and impoundment has affected the sediment composition and metal concentrations.

These lakes were chosen for four reasons. Firstly, they are within the potential atmospheric contamination range (150 km) of mining operations (Schneider et al. 2019). Secondly, these lakes are home to threatened and endemic species including Paragalaxias eleotroides McDowall and Fulton, 1978 (yingina/Great Lake) and Galaxias tanycephalus Fulton, 1978 (Woods Lake). Thirdly, they are popular and valuable recreational fisheries, meaning that human health is at risk if metal contamination is high within the lakes. Finally, as dammed lakes they have high sedimentation rates (Kondolf et al. 2014), thus providing better opportunities to study recent anthropogenic impacts than other, natural Tasmanian lakes, which have low sedimentation rates (e.g., 0.43 mm year−1 in Lake Johnston (Anker et al. 2017) and 0.5 mm year−1 in Lake Vera (Beck et al. 2019)). Due to these increased sedimentation rates, short time-scale dating techniques such as 210Pb become more feasible, providing a reliable method of dating sediments deposited within the last 100–150 years (Krishnaswamy et al. 1971). Combining sediment age, geochemical data, enrichment factors, and historical information about external metal sources, it is possible to reconstruct the recent history of metal deposition in these five lakes to help contextualise management decisions.

Methods

Study site

Sediment cores were taken from five lakes (yingina/Great Lake, Woods Lake, Lagoon of Islands (LOI), Little Pine Lagoon (LPL), and Penstock Lagoon) in the Central Highlands region of Tasmania, Australia (Fig. 1). The prevailing winds in Tasmania, on average, travel in an east/south-east direction, meaning that these lakes are downwind of the Queenstown and Rosebery mine sites (Schneider et al. 2019). A major difference between the selected lakes in this study and the lakes selected in previous Tasmanian studies, is the size of the catchment areas. Harle et al. (2002) and Schneider et al. (2019) both targeted lakes with minor fluvial input and small catchments (2.41 km2, and 1.32 km2 average for six lakes, respectively) to minimise major geochemical influence from the drainage basin. However, the lakes in this study have much larger catchments (Table 1) meaning that other sources of heavy metals should be considered, such as run-off from the drainage basin. Another difference is that all five lakes in this study are dammed, and their water-levels are manipulated for hydro-electric power generation and irrigation, whereas all other lakes highlighted in Fig. 1 are natural lakes. The final important difference between the sites of this study and other sites shown in Fig. 1, relates to the incongruities between the west and east of Tasmania, denoted by ‘Tyler’s Line’ (Fig. 1). The west of Tasmania is much older geologically, is dominated by rainforest and sedgeland, and has much higher rates of rainfall. The east (including the Central Highlands) is younger, dominated by sclerophyll forests, and has much less rainfall (Tyler 1992). There are also limnological differences between these provinces. The western lakes are predominantly dark, humic, and red coloured, with high light attenuation and high (5–20 mg L−1) dissolved organic carbon. Eastern lakes typically have a blue-green light climate, have less light attenuation, and lower dissolved organic carbon (Tyler 1992).

Table 1 Sampling information, lake attributes, damming history, and distances from major mining sites
Full size table

Yingina/Great Lake was first dammed in 1918 via the construction of the Miena dam across the major outflow. In 1967, a rock-fill dam was constructed just downstream of the Miena dam. In 1982, the rock-fill dam was raised a further 6 m to increase storage capacity of the lake (Hydro 1999). Woods Lake was first dammed in 1911 via the construction of a levee across the upper Lake River. In 1962, a subsequent 7 m-high rock-fill dam was constructed to increase storage capacity. Lagoon of Islands was first dammed in 1964 via the construction of a 6 m-high earth-wall dam. The flooding in 1964 led to the death of the unique LOI vegetation, which led to diminished water quality downstream. In 2013, the dam was completely removed. Little Pine Lagoon was created in 1954 by flooding several natural lagoons fed by the Little Pine River. Penstock Lagoon was created as an artificial water storage in 1916 (Hydro 1999).

The underlying geology of all five lakes is the same, with dolerite and locally developed granophyre dominating the region (MRT 2023). Dolerite consists of plagioclase feldspars and pyroxenes, meaning that major elements include silicon (Si), aluminium (Al), sodium (Na), calcium (Ca), plus interchangeable metals, commonly iron (Fe), magnesium (Mg) or titanium (Ti). Granophyre consists of quartz and alkali feldspars meaning that major elements include Si, potassium (K), Na, and Al. Therefore natural, geogenic elements will be dominated by Si, Al, Na, K, Ca, Fe, Mg, and Ti.

There are several mineral deposits close to the study sites, including Quamby Brook copper mine (~ 24 km north of the yingina/Great Lake), bauxite drilling (~ 30 km away from the yingina/Great Lake), Little Den Goldfield (~ 15 km upstream of Woods Lake), as well as some minor quarries close to the yingina/Great Lake, such as Sandbanks Tier Quarry (8 km to the North–East). The Quamby Brook copper mine has been assessed for profitability a few times but no significant production of metallic minerals is recorded from either alluvial or hard rock mining activities in the Deloraine district (Evans 2013). The Little Den Goldfield is located on private land and was only intermittently worked for alluvial gold that was found on and near the ground surface, meaning that the emission of particulates is unlikely. These proximal sites would likely have minimal impact on the surrounding environment because of their modes of operation and extremely small scale.

Coring

Five sediment cores were collected in polycarbonate tubes with a Uwitec hammer corer, with three collected in 2015 and two collected in 2021 (Table 1). Elizabeth Bay was chosen as the exact location of the yingina/Great Lake core because it was previously a separate waterbody (Elizabeth Lake) and damming was expected to change lacustrine sediment geochemistry. Woods Lake was cored in the centre of the lake. LOI was cored near the dam and is shallow throughout. LPL was cored in the centre of the lagoon. Penstock Lagoon was cored in the centre of the southern end. The yingina/Great Lake and Penstock Lagoon cores were sectioned at 0.5 cm intervals in the top 5 cm, and 1 cm thickness for the remainder. The LPL core was sectioned at 0.5 cm in the top 10 cm, and 1 cm thickness for the remainder. Woods Lake and LOI were sectioned at 0.5 cm throughout. All samples were stored in plastic bags in a freezer at −18 °C, before being freeze dried and ground using a Retsch mixer mill 200 (25/s for 60 s).

210Pb dating

The cores were dated at the Australian Nuclear Science and Technology Organisation (ANSTO) by 210Pb dating, following standard methods for alpha spectrometry described by (Appleby 2001). All samples were digested on a hot plate according to ANSTO Method I-3331 (Lead-210 dating sample preparation). Each sample was further prepared using Method I-3329 (Polonium Chemical Isolation) and I-3330 (Radium Chemical Isolation). Polonium fractions were auto-deposited onto silver disks, radium fractions were co-precipitated with BaSO4 and collected on fine resolution filter papers ready for spectroscopy analyses. The filter samples containing 133Ba and 226Ra were first analysed by gamma ray spectroscopy to measure 133Ba activities and determine the 226Ra recoveries. Finally, each prepared sample was analysed for 210Po and 226Ra activities using alpha particle spectroscopy. 210Po is the granddaughter of 210Pb and is in secular equilibrium with 210Pb, allowing the activity of total 210Pb to be determined. 226Ra is the grandparent of 210Pb which can be used as the proxy of supported 210Pb. Unsupported 210Pb activity was calculated by subtracting the supported activity from the total 210Pb activity for each sample. The constant rate of supply model, which assumes a constant atmospheric flux of 210Pb and variable influx rate of sediments, was applied to the determined unsupported 210Pb activities (Appleby and Oldfield 1978).

X-ray fluorescence core scanning

Prior to subsampling, three cores (yingina/Great Lake, Woods Lake, and LOI) were split in half and analysed at the ANSTO using an Itrax™ X-ray fluorescence (XRF) core scanner. This non-destructive technique produces semi-quantitative data expressed as total counts. Cathode-ray tube settings were constant for all cores—voltage of 40 kV and current of 30 mA. Step size varied, with 1000 microns for yingina/Great Lake, and 200 microns for Woods Lake and LOI. Optical images of these cores were also taken.

XRF scanning also measured the incoherent scattering to coherent scattering ratio (Inc/Coh ratio). The Inc/Coh ratio is commonly used as an organic matter proxy in sediment cores (Guyard et al. 2007, Chagué-Goff et al. 2016, Jacq et al. 2019), although it can also reflect changes to water content or sediment density down-core (Evans et al. 2019). Incoherent scattering is more likely to occur when the energy of incident X-rays is much greater than the energy binding electrons to the atom nucleus, such as that for elements with a lower atomic mass (Duvauchelle et al. 1999). Sediments rich in organic matter (containing abundant H, C, N, and O), or water (H2O), tend to have increased incoherent scattering (Woodward and Gadd 2019).

Sedimentary metal concentrations

Yingina/Great Lake, Woods Lake, and LOI samples were submitted to Analytical Services Tasmania for the analysis of metals (Al, As, Cd, Cu, Fe, Pb, Zn), and phosphorous (P), in soils/sediment by ICP-AES (inductively coupled plasma atomic emission spectroscopy). Twenty-two samples were analysed for the yingina/Great Lake and Woods Lake cores, while twenty samples were analysed for the LOI core. Samples were analysed at 2 cm resolution in the upper 20 cm of each core, and remaining samples were chosen based on key dates, such as damming, peak mining activity, and pre-mining. One gram of dried sediment was digested in a mixture of hot aqua regia. Samples were then diluted to 50 mL with purified water before being analysed on an Agilent 730-ES ICP-AES. Method reporting limits (AST 2020) were reached for some samples for three elements—cadmium (0.5 mg kg−1), arsenic (1.0 mg kg−1), and lead (1.0 mg kg−1).

LPL and Penstock Lagoon samples were submitted to the ANSTO for analysis by ICP-AES (Al, Fe, and P) and inductively coupled plasma mass spectrometry (As, Cu, Pb, and Zn) according to their standard procedure. Approximately 0.2 g of each sample was weighed out and microwave digested in aqua regia according to ANSTO Method VI 2995. Every subsample was analysed. The method reporting limit was not reached for any measured element.

Enrichment factors

Enrichment factors (EF) were calculated to minimise the effect of varying sediment composition on the concentration of metals (Abrahim and Parker 2008). Metal concentrations are often strongly correlated with grain size, where smaller grain sizes tend to correlate with higher metal concentrations and vice versa, and to organic matter concentration, where higher organic matter tends to correlate with higher metal concentrations (Martinčić et al. 1990, Salomons and Förstner 2012). EFs also highlight the magnitude of anthropogenic impacts because they can differentiate between metals from natural processes and those from human activities (Yongming et al. 2006). EFs are calculated by normalising the measured metal concentration with respect to a reference element, typically lithogenic elements such as Ti, Fe, or Al (Cevik et al. 2009, Karageorgis et al. 2009). This study used Al because it has not been strongly impacted by mining activities, it had a consistent down-core profile in each of the five studied lakes, and it allowed for a direct comparison of EFs with six previously studied Tasmanian lakes (Owen Tarn, Basin Lake, Lake Dove, Lake Cygnus, Lake Dobson, and Perched Lake) by Schneider et al. (2019). It is important to note that different lakes were cored in different years, meaning the number of data points available for the post-1980 period varied between sites. For example, Basin Lake, Lake Cygnus, and Lake Dobson all contain only two data points for this period because they have low sedimentation rates and were sampled in the years 2011, 2000, and 2000, respectively. Lakes in this study have more data points (yingina/Great Lake (n = 7), Woods Lake (n = 6), LOI (n = 4), LPL (n = 14), and Penstock Lagoon (n = 5)) for this period.

EFs were calculated by comparing the average concentration of a particular metal during a certain time period (such as ‘peak’ mining: 1930–1980), with the average background concentration (pre-1880), both with respect to the average Al concentration during the respective time period:

$$ {\text{EF}} = \left( {\left[ {{\text{M}}_{{{\text{peak}}}} } \right]/\left[ {{\text{Al}}_{{{\text{peak}}}} } \right]} \right) \, / \, \left( {\left[ {{\text{M}}_{{{\text{bkgrd}}}} } \right]/\left[ {{\text{Al}}_{{{\text{bkgrd}}}} } \right]} \right) $$

Results

210Pb dating

The yingina/Great Lake, Woods Lake, LOI, LPL, and Penstock Lagoon cores were dated back to c. 1919, 1915, 1893, 1858, and 1863, respectively (Table S1). The 210Pb activity decreased exponentially for all cores (except yingina/Great Lake), indicating that the sediments were intact and undisturbed (Figures S2, S3, S4, and S5). yingina/Great Lake had one minor anomaly, with unsupported activity increasing from 69 ± 3 Bq kg−1 at 8.25 cm to 136 ± 6 at 12.25 cm (Figure S1). yingina/Great Lake and LPL generally had the highest mass accumulation rates, followed by Penstock Lagoon, Woods Lake, and then LOI.

Sedimentary metal concentrations

Metal concentrations varied considerably through time in many of the cores (Figs. 2, 3, 4, 5, and 6). Notably, the first damming event (denoted by the red, dashed line) often coincided with drastic concentration changes for many of the analysed elements. This is most noticeable in Great Lake (Fig. 2), Woods Lake (Fig. 3), LPL (Fig. 5), and Penstock Lagoon (Fig. 6).

Fig. 2
figure 2

yingina/Great Lake down-core metal (and P) concentrations (mg kg−1). Red dots indicate values below the LOD and are displayed on the plot at 0.5 times the reporting limit for that element. The Inc/Coh line was determined from Itrax™ XRF count data. The leftmost column is an optical image of the sediment core. The red horizontal dashed lines indicate approximate depth of damming events (1918, 1967, 1982), and the green shaded area represents the period of peak mining (1930–1980). Sediment ages (and corresponding depths) were calculated using 210Pb data and the constant rate of supply model for all sites

Full size image
Fig. 3
figure 3

Woods Lake down-core metal (and P) concentrations. The Inc/Coh line was determined from Itrax™ XRF count data. The leftmost column is an optical image of the sediment core. The red horizontal dashed lines indicate approximate depth of damming events (1911 and 1962), and the green shaded area represents the period of peak mining (1930–1980)

Full size image
Fig. 4
figure 4

Lagoon of Islands (LOI) down-core metal (and P) concentrations. Red dots indicate values below the LOD and are displayed on the plot at 0.5 times the reporting limit for that element. The Inc/Coh line was determined from Itrax™ XRF count data. The leftmost column is an optical image of the sediment core. The red horizontal dashed line indicates the approximate depth of damming (1964), the blue horizontal dashed line indicates the approximate depth of dam removal (2010), and the green shaded area represents the period of peak mining (1930–1980)

Full size image
Fig. 5
figure 5

Little Pine Lagoon down-core metal (and P) concentrations. The red horizontal dashed line indicates the approximate depth of damming (1954), and the green shaded area represents the period of peak mining (1930–1980). No Itrax™ XRF scanning was performed on this core, meaning the optical image and Inc/Coh columns are omitted

Full size image
Fig. 6
figure 6

Penstock Lagoon down-core metal (and P) concentrations The red horizontal dashed line indicates the approximate depth of damming (1916), and the green shaded area represents the period of peak mining (1930–1980). No Itrax™ XRF scanning was performed on this core, meaning the optical image and Inc/Coh columns are omitted

Full size image

Arsenic concentrations were low across all lakes, with values ranging from below the limit of detection (LOD) (1 mg kg−1) to a maximum of 5 mg kg−1 (Woods Lake). Copper, lead, and zinc levels were low across all five lakes, except for zinc in the yingina/Great Lake core which reached 139 mg kg−1. Cu ranged from 6 mg kg−1 (Penstock Lagoon) to 94 mg kg−1 (LOI). Pb ranged from below the LOD (1 mg kg−1) in yingina/Great Lake and LOI, to a maximum of 15 mg kg−1 in LPL. Zn ranged from 3 mg kg−1 (Penstock Lagoon) to 139 mg kg−1 (yingina/Great Lake). Al concentrations ranged from 3770 (Penstock Lagoon) to 71,400 mg kg−1 (yingina/Great Lake). yingina/Great Lake and Penstock Lagoon showed the largest down core Al variations, ranging from  ~ 25,000 mg kg−1 to ~ 70,000 mg kg−1, and ~ 4000 mg kg−1 to ~ 46,000 mg kg−1, respectively. Fe also had high concentrations, ranging from 12,500 (LOI) to 133,000 mg kg−1 (Woods Lake).

Enrichment factors

EFs of metals in the lakes were generally low, with most values around 1 (Table 2). Pb showed the largest enrichment of the analysed elements, with factors of 3.1, 1.9, 12.5, 2.1, and 2.4 in the peak mining period (Table 2A), and 2.0, 1.8, 24.0, 1.5, and 1.9, in the post-1980 period (Table 2B).

Table 2 A and B: Enrichment factors for eleven Tasmanian lakes. Enrichment factor categories are based on those used in Schneider et al. (2019). * Indicates lakes previously studied by Schneider et al. (2019). C: The difference between EFs from different time periods (peak mining EF—recent times EF = ΔEF). Green therefore indicates an element that has decreased in enrichment in recent times compared to during peak mining (1930–1980) and red indicates that an element is more enriched in recent times
Full size table

Discussion

Yingina/Great Lake

Yingina/Great Lake had a consistent sediment composition in the deep portion of the core (pre-damming) with generally low concentrations for all measured elements (Fig. 2). At approximately 28 cm depth (25.25 cm is the oldest dated depth, 1919 ± 10, Table S1), the sediment became a deep brown colour, indicating an organic-rich Sect. (23–28 cm depth). Sharp increases in TC and TN also occurred around this time (Figure S6). The large change in sediment composition that began around 28 cm depth is likely related to the construction of the Miena dam in 1918 (approximately 25.5 cm depth). Dams have been shown to increase total P and N within a reservoir compared to levels upstream and levels in the water bodies prior to damming (Vaikasas et al. 2015). Correspondingly, downstream levels are often drastically decreased after the construction of a dam because of the trapped particles and nutrients (Zhou et al. 2013, Maavara et al. 2015, Vaikasas et al. 2015). The construction of the Miena dam likely increased trapped particulates and associated nutrients such as P and N (Figure S6). This effect was likely exacerbated by sheep grazing in the catchment, which reached its peak in the late nineteenth century and early twentieth century, and can lead to nutrient run-off via sediment mobilization (Shepherd 1972, Elliott and Carlson 2004, Cournane et al. 2011). The dam also changed the dynamics of the system. The sediment core was taken from Elizabeth Bay, which was previously a separate lake (Elizabeth Lake) and not a part of Yingina/Great Lake. These major changes from damming resulted in increasing nutrient concentrations in the sediments, and secondarily, may have stimulated primary production in the lake, leading to further increases in organic matter content. Organic matter has a high sorption affinity for metals and can play a significant role in metal retention within sediment (Yang et al. 2010, Zhang et al. 2014). Thus, the visible shift post-damming to many of the measured metals (Fig. 2).

Woods lake

There were greater differences between the different metals in Woods Lake compared to those observed in yingina/Great Lake (Fig. 3). However, Cu and Pb showed a similar pattern, with low concentrations in the deepest part of the core, and an increase after damming (1911) and during the time of peak mining (1930–1980). In the deepest part of the Woods Lake core (pre-damming, 25–45 cm depth), there were large variations in Fe and Al concentrations (Fig. 3), however, without accurate age estimates of this depth, it is difficult to determine the reasons for these fluctuations. Following damming in 1911 (approximately 21.5 cm depth), there were similar responses to those observed in the Yingina/Great Lake. Organic matter content rose, as shown by TC and TN concentrations (Figure S6), and visual inspection of the core (Fig. 3). The Inc/Coh ratio also increased but it may represent sediment density or water content changes, rather than organic matter changes, because of the varied dry bulk density throughout this core (Table S3 and Figure S8). The rise in organic matter was likely associated with damming, as was the case for yingina/Great Lake. Cu and Pb showed increasing concentrations post-damming, likely associated with the increase in organic matter and its high sorption affinity for metals, however, Zn concentrations decreased slightly during this time.

Lagoon of islands

LOI had a much lower sedimentation rate compared to Woods Lake and yingina/Great Lake (Table S1). This makes interpretation more difficult because the age of most of the core is unknown. In the top 6.5 cm (1893 ± 11), Pb, Zn, As, and Fe all showed minor increases towards the surface (Fig. 4). This increase begins before damming took place in 1964 (approximately 3.75 cm). However, it is important to note that all measured metals (except for Pb) reached a higher concentration in the deepest part of the core (~ 23 cm) than in the surface region of core. This shows that the concentrations remain in the range of what can be assumed to be natural levels. Assuming that sedimentation rate pre-damming is linear, the extrapolated age at 23 cm depth significantly predates European settlement in Tasmania (~ 1800). After damming in 1964, primary production increased and by 1989 the waterbody became eutrophic and visible algae blooms could be seen (Maxwell 2007). This stemmed primarily from internal P loading from the decay of the extensive plant community that was flooded after damming. This period coincides with relatively high concentrations of Fe, Zn, Pb, and As, likely associated with the increased sediment organic matter content, and increased metal retention.

Little pine lagoon

Little Pine Lagoon was created in 1954 by flooding several natural lagoons fed by the Little Pine River (Hydro 1999). It is difficult to determine how often the study site was submerged prior to damming. Fe and Al concentrations remained consistent prior to damming, with only minor increases observed for Al between 1901 ± 10 and damming (Fig. 5). After damming, both Fe and Al showed marked increases, with Fe more than doubling and Al almost doubling. Zn, Cu, Pb, and As concentrations all began to slowly increase above background levels around 20 cm depth, or 1870 ± 15, with Zn and Cu concentrations continuing to increase up the remainder of the core. The Fe, Al, Zn, and Cu profiles appeared tied to the increasing P concentrations, and likely increasing organic matter content, that occurred concurrently. The increase in P may be related to run-off from agricultural practices (such as sheep grazing) that became more widespread around this time, similar to the pattern identified in Lagoon of Islands and yingina/Great Lake. The Pb profile is dissimilar to the Zn, Cu, and Al profiles, with a maximum occurring before damming took place, and a slow decrease towards the cores surface. The source of Pb to this lake is likely independent to the source of the other metals, and is discussed in a later section (‘Potential Pb sources’).

Penstock lagoon

Penstock Lagoon was created as an artificial water storage in 1916 (Hydro 1999). The drastic changes to most measured elements around this time (Fig. 6) are likely associated with the shift from existing as a wetland or periodically submerged location, to a permanent lagoon. Similar to yingina/Great Lake, elements were relatively low pre-damming, increased rapidly around the time of initial damming, and remained at this level in the top portion of the core. The metal increases are likely related to the increases in organic matter content associated with damming and the transformation to a lagoon ecosystem.

Impact of mining on distant lakes

Despite the concentrations of many analysed metals increasing around the time of peak mining (Figs. 2, 3, 4, 5, and 6), mining does not appear to be a major contributor. Enrichment factors, which take into account Al concentrations, remained close to 1 for most elements (except Pb) in both the period of peak mining (Table 2A) and the post-1980 period (Table 2B). Additionally, most elements showed no change between the peak mining period and the recent surface period, with a ΔEF (ΔEF = EFPeak–EFrecent) close to 0 (Table 2C). Assuming other metal inputs remained the same over the whole study period, and assuming mining was the major source of metals, EFs would be expected to decrease after the decline in mining production post 1980 (and thus have a positive ΔEF), which they do not. Schneider et al. (2020) showed the concentration and depositional flux of metals entering lakes near the mining sites (Owen Tarn and Basin Lake) peak between 1930 and 1970, and decrease towards the surface of each core, with some elements returning to background concentrations (Se and As), and some decreasing but not yet fully returning to background levels (Cu and Pb). Therefore, it may be expected that metal concentrations would decrease after 1980, but this is not evident in the cores of this study.

One factor limiting aerial mining pollution from depositing into the lakes of the Central Highlands is distance, but another important factor is the lack of rainfall. Western Tasmania is geographically, geologically, and climatically distinct from the east, separated by what is referred to as ‘Tyler’s Line’ (Tyler 1992). The Central Highlands are located to the east of Tyler’s Line, whereas the previously studied Tasmanian lakes lie to the west (Fig. 1). Between 1981 and 2010, average annual rainfall for the west ranged from 1500 to 3000 mm, while the Central Highlands received 600–1200 mm (BOM 2020). Precipitation increases the rates of wet deposition of trace metals (Hu and Balasubramanian 2003, Kim et al. 2012, Cheng et al. 2021, He et al. 2023). Both wet and dry deposition contribute to the deposition of atmospheric particles, but wet deposition typically dominates for metals because they are more commonly adsorbed to smaller particles, which are most affected by wet deposition (Weerasundara et al. 2017, He et al. 2023). For this reason, lakes in regions of reduced average annual rainfall, such as the Central Highlands, may be less susceptible to long range atmospheric transport of metals, particularly if the metal source is situated in a region of high rainfall, where particles are more likely to be deposited, and less likely to remain in the atmosphere for long time periods or distances.

The largest anthropogenic impact on the lakes in this study was most likely damming, and the effect this had on system dynamics and the subsequent raising of nutrients and organic matter content. This was likely exacerbated by increased nutrient and sediment run-off from intense sheep grazing in the region. The increase in organic matter was evident in all five cores, as indicated by increases in the Inc/Coh ratios (Figs. 2, 3 and 4), P increases, and considerable TC and TN increases in both yingina/Great Lake and Woods Lake (Figure S6). This, in-turn, likely caused the increase to the concentrations of metals post-damming, due to the known relationship between organic matter and metals. The one metal that may have been anthropogenically enriched is Pb. Pb EFs are consistently higher than other metals, and in LOI the EF falls into the severely enriched category (12.5 and 24 in the peak mining period and post-1980 period, respectively, Table 2). The potential sources of Pb are discussed below.

Potential Pb sources

Mining is a potential source of Pb, however, there are two indications that this is unlikely to be the primary source. Firstly, Pb EFs remain similar between the period of peak mining and recent times (Table 2C), with the EF in LOI almost doubling in recent times (24.0, Table 2B) compared to the period of peak mining (12.5, Table 2A). This is not consistent with atmospheric deposition from mining aerosols being the primary source, as EFs would be expected to be considerably higher during the period of peak mining. Secondly, in every core, the enrichment factors for Pb, Cu, and Zn are inconsistent (Table 2). If mining operations were a significant source of metals, then there would likely be agreement between Pb, Cu, and Zn trends because these metals are commonly released from mining operations and have been found to increase concurrently in sediments impacted by mining elsewhere in Tasmania (Harle et al. 2002, Augustinus et al. 2009, Schneider et al. 2019). Instead, Pb enrichment factors were consistently higher than both Cu and Zn (Table 2).

Contamination from lead shot is another potential Pb source. Hunting with firearms is a pastime that began soon after European arrival in Tasmania (early 1800s) and continues today. Waterfowl (ducks), deer, wallaby, mutton-bird, pheasant, brown quail, and rabbit are common targets for recreation, commercial harvest, and pest control. Lead is toxic to humans and other animals, and it is highly persistent in the environment (Tong et al. 2000, Wani et al. 2015). Lead shot was phased out for waterfowl hunting in Tasmania and was officially banned over public wetlands in 2005 and from all wetlands in 2006 (DoNRaET 2021). However, the use of lead shot to hunt everything except for waterfowl is still permitted. LOI was a popular duck hunting location, whereas the other sites in this study were less so, with more of a focus on trout fishing. It is possible that lead shot is contributing to Pb levels in the sediment, as pellets can be mobilised and distributed across the environment (Romano et al. 2016). Romano et al. (2016) found that soil Pb concentrations in hunting sites in Argentina (approximately 10–20 mg/kg) were only slightly higher than control sites (approximately 5–10 mg/kg). Lead shot may be a source of Pb in Central Highland lakes, particularly in LOI, although it is difficult to confirm because the timeline of lead shot use is similar to that of other Pb sources such as industrial processes and leaded petrol.

Industrial processes, such as fossil fuel burning and metal production, have contributed heavily to metal loads (including Pb) in the earth’s atmosphere and ecosystems since the 1850s (Marx et al. 2016). Marx et al. (2016) assessed Pb records from lakes, peat mires, and ice cores from Europe, North America, South America, Asia, Australia, and the polar regions, and found that Pb enrichment is often over 100 times greater than background levels in Europe and North America, and between 4 and 15 times background levels in Australia. However, the Australian data were extremely sparse, and only one lake sediment core was included (Club Lake, New South Wales). Specifically, Club Lake Pb data showed only very minor enrichment (1.3 EF) in the post-1850 period (Stromsoe et al. 2013). However, similar to the case for local mining, the increase in Pb from industrial processes is typically accompanied by increases in other metals such as Cr, Hg, Mn, As, Cd, Cu, and Zn, which is only partially observed in the Club Lake sediments (Pacyna and Pacyna 2001).

Leaded petrol was introduced into Australia in 1932 and was completely phased out by 2002 (Kristensen 2015). Many environmental archives and proxies show that there were marked shifts in lead concentrations and isotopic compositions during the time of leaded petrol use (Kristensen et al. 2017). One major problem with leaded petrol emissions was that more than 90% of airborne lead was found within fine particulates (< 2.5 µm), meaning that widespread contamination was possible, despite most of the emissions arising from urban areas where the density of vehicles was highest. The long-range impact of global lead emissions was confirmed by multiple studies that found concentration increases in Antarctic ice cores (Planchon et al. 2003, McConnell et al. 2014). Similarly, a study of lead concentrations in lichen and fungi around central and greater Victoria, Australia, found that leaded petrol emissions impacted sites up to 150 km away from central Melbourne (Wu et al. 2016). Kristensen, (2015) estimated that between 1958 and 2002, more than 240,500 tonnes of lead were emitted Australia-wide, with 5769 tonnes originating from Tasmania alone.

Ultimately, Pb concentrations remain well below Australian interim sediment quality guidelines. The minor increases are likely related to the general increase in lead use globally in the twentieth century. Specifically, lead shot, leaded petrol, and worldwide industrial processes (fossil fuel burning and metal production) may all be sources of Pb to Tasmanian lakes. Evidence includes the similar timing between increases in Pb concentrations in the sediment (Figs. 2, 3 and 4) and the rise of industrial processes, the use of leaded petrol, and the use of lead shot for hunting. The phasing-out of Pb use in the twenty-first century is not yet reflected in the LOI surface sediment due to the low sedimentation rate, but it is important that future work is conducted to identify if changes occur.

Sediment quality guidelines

Australian interim sediment quality guideline (ISQG) values (Table S2) provide a guideline for concentrations of elements within sediment in Australia and New Zealand (AGI 2019). There are two values assigned to each element, a low value below which there is a low risk of unacceptable effects occurring in aquatic ecosystems, and a high value, which provides an indication of levels that are expected to already have toxicity related adverse effects (Long et al. 1995). The five lakes in this study all fell below the low ISQG level for every metal (As, Cd, Pb, Zn) except for Cu, which exceeded the low threshold (65 mg/kg) in eight LOI samples. However, this is of low concern because the maximum Cu concentration was 94 mg/kg (significantly lower than the ISQG high value 270 mg/kg), and it occurred deep down in the sediment, not near the surface.

The enrichment of Pb in LOI falls into the “severely enriched” category (Table 2), but the maximum Pb concentration (12 mg/kg) falls well below the low ISQG level (50 mg/kg) and imminent concern is low. Overall, the aquatic life in these lakes is most likely not facing adverse effects from the metals contained within the sediment.

Conclusion

A recent study (Schneider et al. 2019) indicated that remote Tasmanian Wilderness World Heritage Area lakes have been negatively impacted by long-distance atmospheric metal deposition from major mining operations on the West Coast. However, there is no strong evidence in our study indicating that lakes a similar distance away and east of Tyler’s Line in the Central Highlands have been impacted by long-distance atmospheric deposition. Overall metal concentrations are low, and even those elements that show enrichment (e.g. Pb in the surface of LOI) are well below guideline values. This may be related to the lower average rainfall in the east of Tasmania and the resultant decrease in wet deposition of aerosols compared to western Tasmania. The anthropogenic activity that appears to have had the largest impact on the sediment geochemistry and composition is the damming of the natural lakes. Damming typically leads to an increase in organic matter content either directly, by the trapping of particles and nutrients, or indirectly, by stimulating primary production in the lake. Increasing organic matter content generally increases the retention of metals in the sediments. Despite the small increase of metal concentrations after damming, all samples remain below the Australian interim sediment quality guidelines low-risk threshold (except for Cu in some LOI samples, which exceeds the low threshold) and therefore pose a low risk of unacceptable effects occurring in these aquatic ecosystems.