Unravelling a 'miner's myth' that environmental contamination in mining towns is naturally occurring

Blood lead increases and haemoglobin decreases in urban birds along a soil contamination gradient in a mining city

Lead contaminated soil is a persistent global threat to the health of animal populations. Nevertheless, links between soil lead and its adverse effects on exposed wildlife remain poorly understood. Here, we explore local geographic patterns of exposure in urban birds along a gradient of lead contamination in Broken Hill, an Australian mining city. Soil lead concentrations are linked to co-located blood lead measurements in rock pigeons (Columba livia), house sparrows (Passer domesticus), crested pigeons (Ocyphaps lophotes) and white-plumed honeyeaters (Lichenostomus ornatus). Median blood lead levels were highest in crested pigeons (59.6 μg/dL), followed by house sparrows (35.2 μg/dL), rock pigeons (35.1 μg/dL), and white-plumed honeyeaters (27.4 μg/dL). Blood lead levels in all species declined away from mining areas, the primary source of lead contamination in Broken Hill. Blood lead increased significantly and at the greatest rate relative to soil lead in the three ground foraging species (crested pigeons, house sparrows, rock pigeons). For these species, soil lead concentrations below 200 mg/kg and 900 mg/kg were needed to maintain a median blood lead concentration under the lower threshold of the subtoxic (20-50 μg/dL) and toxic (≥50 μg/dL) effect ranges previously identified for some bird species. We also investigated the effects of lead exposure on blood haemoglobin levels as a general measure of physiological condition in birds exposed to different levels of soil lead contamination. Overall, for every 1 μg/dL increase in blood lead, haemoglobin decreased by 0.11 g/L. The rate of this decrease was not significantly different between species, which supports the measurement of haemoglobin as a consistent though insensitive measure of physiological condition in chronically lead exposed birds. Our findings reflect the importance of lead contaminated soil as a widespread source of elevated blood lead and supressed haemoglobin levels in birds inhabiting urbanised and mining impacted environments.

Peppered with lead: An environmental forensics approach to identify the source of rising blood lead levels

Despite the phase-out of lead-based products, lead contamination can still present a contemporary risk to public health. In situations where elevated blood lead cannot be attributed to common sources, detailed environmental investigation is needed to identify more elusive sources and manage harmful exposure pathways. We apply a forensics approach to assess common and elusive sources of lead in the home environment of two individuals with fluctuating blood lead levels in Sydney, Australia. Using multiple analytical lines of evidence (portable X-Ray Fluorescence spectrometry (pXRF), inductively coupled-plasma mass spectrometry (ICP-MS), lead isotopic compositional analysis (PbIC) and haematological assessment) a pewter pepper grinder containing lead (>6000 mg/kg; 70% bioavailable) was identified as a potential source. After removing the pepper grinder from the home, the couple's blood lead decreased to below the Australian intervention level of 5 μg/dL within a year (Person A: from 12.5 μg/dL in August 2020 to 4.4 μg/dL in March 2022; and Person B: 15.4 μg/dL in August 2020 to 2.1 μg/dL in July 2021). This case study demonstrates how environmental science investigations can play a crucial role in supporting people to take evidence-based action to improve their health.

Tracing nickel smelter emissions using European honey bees

This study investigated trace element contamination in honey bees inhabiting urban areas around the South Pacific's largest and longest operating nickel smelter in Nouméa, New Caledonia. There remains a paucity of research on the environmental impact of nickel smelting, and to date, there has been no assessment of its effects on the popular practice of beekeeping, or whether honey bees are a suitable tracer for nickel smelting emissions. Honey bees and honey were sampled from 15 hives across Nouméa to ascertain linkages between nickel smelter emissions, environmental contamination, and trace element uptake by bees. Comparison of washed and unwashed bees revealed no significant difference in trace element concentrations, indicating trace elements bioaccumulate within the internal tissues of bees over time. Accordingly, trace element concentrations were higher in dead bees than those that were sampled live, with smelter related elements chromium, cobalt and nickel being significantly different at p < 0.05. Except for boron, trace element concentrations were consistently higher in bees than in honey, suggesting that the transfer of trace elements from bees during honey production is negligible. Elevated concentrations of potentially toxic trace elements including cobalt, chromium and nickel in bees declined with distance from smelting operations (Spearman's Rho, p < 0.05), indicating the relationship between environmental contamination and the uptake of trace elements by bees. The findings of this study emphasise potential environmental and human health risks associated with trace element contamination from nickel smelting operations and affirm the use of honey bees as a biomonitor of potentially harmful nickel smelting emissions.

Spatial distribution and composition of mine dispersed trace metals in residential soil and house dust: Implications for exposure assessment and human health

Trace metal exposure from environmental sources remains a persistent global problem, particularly in communities residing adjacent to metal extraction and processing industries. This study examines front yard soil and house dust from 62 residences throughout the Australian Ag-Pb-Zn mining city of Broken Hill to better understand spatial variability in metal distributions, compositions and exposures across an industrially polluted urban environment. X-ray fluorescence analysis of paired soil/dust samples indicated that geomean concentrations (mg/kg) of Cu (32/113), Zn (996/1852), As (24/34) and Pb (408/587) were higher in house dust while Ti (4239/3660) and Mn (1895/1101) were higher in outdoor soil. Ore associated metals and metalloids (Mn, Zn, As, Pb) in soil and house dust were positively correlated and declined in concentration away from mining areas, the primary source of metalliferous emissions in Broken Hill. The rate of decline was not equivalent between soil and house dust, with the indoor/outdoor concentration ratio increasing with distance from mining areas for Zn/Pb (geomean = 1.25/1.05 (<1 km); 2.14/1.52 (1-2 km); 2.54/2.04 (>2 km)). House dust and Broken Hill ore Pb isotopic compositions (²⁰⁶Pb/²⁰⁷Pb; ²⁰⁸Pb/²⁰⁷Pb) were more similar in homes nearest to mining areas than those further away (geomean apportioned ore Pb = 88% (<1 km); 76% (1-2 km); 66% (>2 km)), reflecting spatial shifts in the balance of sources contributing to indoor contamination. Incorporation of house dust Pb reduced overestimation of IEUBK modelled blood Pb concentrations compared to when only soil Pb was used. These findings demonstrate that even in contexts where the source and environmental burden of metals are relatively apparent, geochemical relationships and exposures between outdoor and indoor environments are not always predictable, nor easily disaggregated.

Plumbojarosite Remediation of Soil Affects Lead Speciation and Elemental Interactions in Soil and in Mice Tissues

Lead (Pb) contamination of soils is of global concern due to the devastating impacts of Pb exposure in children. Because early-life exposure to Pb has long-lasting health effects, reducing exposure in children is a critical public health goal that has intensified research on the conversion of soil Pb to low bioavailability phases. Recently, plumbojarosite (PLJ) conversion of highly available soil Pb was found to decrease Pb relative bioavailability (RBA <10%). However, there is sparse information concerning interactions between Pb and other elements when contaminated soil, pre- and post-remediation, is ingested and moves through the gastrointestinal tract (GIT). Addressing this may inform drivers of effective chemical remediation strategies. Here, we utilize bulk and micro-focused Pb X-ray absorption spectroscopy to probe elemental interactions and Pb speciation in mouse diet, cecum, and feces samples following ingestion of contaminated soils pre- and post-PLJ treatment. RBA of treated soils was less than 1% with PLJ phases transiting the GIT with little absorption. In contrast, Pb associated with organics was predominantly found in the cecum. These results are consistent with transit of insoluble PLJ to feces following ingestion. The expanded understanding of Pb interactions during GIT transit complements our knowledge of elemental interactions with Pb that occur at higher levels of biological organization.

Trace element contamination of soil and dust by a New Caledonian ferronickel smelter: Dispersal, enrichment, and human health risk

Metallurgical industries remain a considerable source of trace element contamination and potential human health risk. Determination of sources is a key challenge. With respect to the South Pacific's largest and longest operating metallurgic smelter in Nouméa, New Caledonia, determining the environmental impact and subsequent human health risk associated with local ferronickel smelting is complicated by natural geological enrichment of Ni and Cr. This study applies a multi-method and multi-matrix approach to disentangle smelter emissions from geogenic sources and model the consequent health risk from industrial activity. Dust wipes (n = 108), roadside soil (n = 91), garden soil (n = 15) and household vacuum dust (n = 39) were assessed to explore geospatial trace element (As, Cr, Cu, Fe, Mn, Ni, Pb, S, V and Zn) variations across outdoor and indoor environments. Enrichment factors (EF) identified elevated levels of smelter-related trace elements: S (EF = 7), Ni (EF = 6) and Cr (EF = 4), as well as Zn (EF = 4). Smelter-related elements in soil and dust deposits were negatively correlated with distance from the facility. Similarity of Pb isotopic compositions between dust wipes, surface soil and vacuum dust indicated that potentially toxic trace elements are being tracked into homes. Non-carcinogenic health risk modelling (Hazard Index, HI) based on 15 spatial nodes across Nouméa revealed widespread exceedance of tolerable risk for children (0-2 years) for Ni (HI 1.3-15.8) and Mn (HI 0.6-1.8). Risk was greatest near the smelter and to the north-west, in the direction of prevailing wind. Given the elevated cancer risk documented in New Caledonia, disentanglement of environmental from industrial sources warrants further attention to ensure community health protection. Our analysis illustrates how the confounding effects from complex environmental factors can be distilled to improve the accuracy of point source apportionment to direct future mitigation strategies.

Novel Application of Machine Learning Algorithms and Model-Agnostic Methods to Identify Factors Influencing Childhood Blood Lead Levels

Blood lead (Pb) poisoning remains a global concern, particularly for children in their early developmental years. Broken Hill is Australia's oldest operating silver-zinc-lead mine. In this study, we utilized recent advances in machine learning to assess multiple algorithms and identify the most optimal model for predicting childhood blood Pb levels (BLL) using Broken Hill children's (<5 years of age) data (n = 23,749) from 1991 to 2015, combined with demographic, socio-economic, and environmental influencing factors. We applied model-agnostic methods to interpret the most optimal model, investigating different environmental and human factors influencing childhood BLL. Algorithm assessment showed that stacked ensemble, a method for automatically and optimally combining multiple prediction algorithms, enhanced predictive performance by 1.1% with respect to mean absolute error (p < 0.01) and 2.6% for root-mean-squared error (p < 0.01) compared to the best performing constituent algorithm (random forest). By interpreting the model, the following information was acquired: children had higher BLL if they resided within 1.0 km to the central mine area or 1.37 km to the railroad; year of testing had the greatest interactive strength with all other factors; BLL increased faster in Aboriginal than in non-Aboriginal children at 9-10 and 12-18 months of age. This "stacked ensemble + model-agnostic interpretation" framework achieved both prediction accuracy and model interpretability, identifying previously unconnected variables associated with elevated childhood BLL, offering a marked advantage over previous works. Thus, this approach has a clear value and potential for application to other environmental health issues.

A 25-year record of childhood blood lead exposure and its relationship to environmental sources

Broken Hill, the oldest silver (Ag)-zinc (Zn)-lead (Pb) mining community in Australia, has a legacy and ongoing problem of environmental Pb exposure that was identified as early as 1893. To reduce Pb exposure risks, identifying potential exposure pathways and related factors is a critical first step. This study examined blood lead (PbB) levels of children ≤60 months old (n = 24,106 samples), along with Pb concentrations in corresponding soil (n = 10,160 samples), petri-dish dust (n = 106 houses) and ceiling dust (n = 80 houses) over a 25-year period from 1991 to 2015. Regression analysis was used to examine the relationships between environmental Pb sources and children's blood lead (PbB) outcomes. Analysis of the dataset showed Aboriginal children in Broken Hill had a geometric mean PbB of 7.4 μg/dL (95% CI: 6.7-7.4) being significantly higher (p < 0.01) than non-Aboriginal children (PbB 6.2 μg/dL, 95% CI: 6.2-6.3) for all years between 1991 and 2015. Children at the age of 24-36 months had a higher PbB compared with other age groups. Higher PbB levels were also statistically associated with lower socio-economic status and children living in houses built before 1940 (p < 0.01). Blood Pb was also significantly correlated with both soil Pb and indoor petri-dish dust Pb loadings, confirming that these are important pathways for exposure in Broken Hill. A 100 mg/kg increase in soil Pb was associated with a 0.12 μg/dL increase in childhood PbB. In addition, PbB concentrations increased with indoor petri-dish dust Pb loadings (i.e., 0.08 μg/dL per 100 μg/m²/30 days). The 25-year data show that the risk of exposure at ≥ 10 μg/dL was seemingly unavoidable irrespective of residential address (i.e., children of all ages presenting with a ≥10 μg/dL across the whole city area). In terms of moving forward and mitigating harmful early-life Pb exposures, all children aged 24-36 months should be prioritised for feasible and effective intervention practices. Primary intervention must focus on mitigating contemporary ongoing dust emissions from the mining operations and the associated mine-lease areas along with household soil remediation, to help prevent recontamination of homes. Additional practices of dust cleaning using wet mopping and wiping techniques, vacuuming of carpets and furnishings, ongoing monitoring of children and household dust remain important but short-lived abatement strategies. Overall, the key goal should be to eliminate risk by removing contamination in the wider environment as well as in individual homes.

Lead content in soils and native plants near an abandoned mine in a protected area of south-western Spain: an approach to determining the environmental risk to wildlife and livestock

A study of the impact of an abandoned lead (Pb) mine ("Las Musas"), located in SW Spain, on the contamination of the surface soil and pastures in its vicinity revealed the presence of widely distributed, high levels of Pb contamination. The total Pb concentrations in soils sampled at distances from 3 to 998 m from the mine ranged between 129 and 1053 mg/kg, when it has been reported that non-polluted soils have concentrations of 29-40 mg/kg. These exceed the maximum tolerable levels in agricultural soils for the protection of environmental and human health as established in international and regional regulations. While the concentrations of potentially bioavailable Pb in the soils also surpassed the regulatory levels, the effective bioavailable fractions were low. The Pb concentrations measured in native plants ranged from 1.70 to 129 mg/kg dry weight, with Cynosurus echinatus, Philadelphus coronarius, and Fraxinus angustifolia being the species that bioaccumulated the greatest concentrations of this metal. Estimation of the environmental risk to wildlife and livestock grazing in the studied area showed no potential toxicity for these animals.

Signs of adaptation to trace metal contamination in a common urban bird

Metals and metalloids at elevated concentrations can be toxic to both humans and wildlife. In particular, lead exposure can act as a stressor to wildlife and cause negative effects on fitness. Any ability to adapt to stress caused by the negative effects of trace metal exposure would be beneficial for species living in contaminated environments. However, mechanisms for responding adaptively to metal contamination are not fully understood in free-living organisms. The Australian populations of the house sparrow (Passer domesticus) provides an excellent opportunity to study potential adaptation to environmental lead contamination because they have a commensal relationship with humans and are distributed broadly across Australian settlements including many long-term mining and smelting communities. To examine the potential for an evolutionary response to long-term lead exposure, we collected genomic SNP data using the house sparrow 200 K SNP array, from 11 localities across the Australian distribution including two mining sites (Broken Hill and Mount Isa, which are two genetically independent populations) that have well-established elevated levels of lead contamination as well as trace metals and metalloids. We contrast these known contaminated locations to other lesser-contaminated environments. Using an ecological association genome scan method to identify genomic differentiation associated with estimates of lead contamination we identified 60 outlier loci across three tests. A total of 39 genes were found to be physically linked (within 20 kbps) of all outliers in the house sparrow reference genome. The linked candidate genes included 12 genes relevant to lead exposure, such as two metal transporters that can transport metals including lead and zinc across cell membranes. These candidate genes provide targets for follow up experiments comparing resilience to lead exposure between populations exposed to varied levels of lead contamination.

The effect of contemporary mine emissions on children's blood lead levels

BACKGROUND

Broken Hill is home to Australia's oldest silver-zinc-lead mine. However, the precise source of childhood blood lead (PbB) exposures has been subject to considerable debate. Lead sources include natural soil Pb enrichment, legacy deposition, contemporary mining emissions, and Pb-based paint.

OBJECTIVE

To test whether contemporary mining emissions independently affect childhood PbB in Broken Hill.

METHODS

Children's (<5 years old) PbB measures from 2011 to 2015 (n = 4852), obtained from Broken Hill Child & Family Health Centre, were analyzed using generalised linear regression models, including covariates of household soil Pb, city dust Pb concentrations (PbD), demographic factors and Pb ore production. Two natural experiments involving wind direction and the 2009 dust storm were examined to test whether the PbB-distance gradient from the mining operations was influenced by contemporary emissions. The influence of contemporary emissions was further interrogated by examining the effect of ore production on PbB and PbD.

RESULTS

Children living downwind and proximate to the mine had substantially higher PbB outcomes than children similarly distant but upwind. Dust Pb deposition increased significantly with proximity to mining operations as well to Pb production (1991-2013). Average annual PbB correlated with Pb ore production (p < 0.01) with all subsets of children PbB levels responding with near unit elasticity to Pb ore production (p < 0.01). Pre- and post-analysis of the dust storm showed the PbB-distance gradient remained statistically unaltered further confirming contemporary emissions as a source of exposure.

CONCLUSIONS

Contemporary mining emissions influence children's PbB measures independent of other sources and need to be remediated to facilitate reductions in harmful exposure.

Tracing natural and industrial contamination and lead isotopic compositions in an Australian native bee species

This study investigates trace element concentrations (arsenic (As), manganese (Mn), lead (Pb) and zinc (Zn)) and Pb isotopic compositions in an Australian native bee species, Tetragonula carbonaria, and its products of honey and wax. Co-located soil and dust samples were simultaneously analysed with the objective of determining if the bees or their products had potential application as a proxy for monitoring environmental contamination. The most significant relationships were found between Pb concentrations in honey (r = 0.814, p = 0.014) and wax (r = 0.883, p = 0.004) and those in co-located dust samples. In addition, Zn concentrations in honey and soil were significantly associated (r = 0.709, p = 0.049). Lead isotopic compositions of native bee products collected from background sites adjacent to Sydney national parks (²⁰⁶Pb/²⁰⁷Pb = 1.144, ²⁰⁸Pb/²⁰⁷Pb = 2.437) corresponded to local geogenic rock and soil values (²⁰⁶Pb/²⁰⁷Pb = 1.123-1.176, ²⁰⁸Pb/²⁰⁷Pb = 2.413-2.500). By contrast, inner Sydney metropolitan samples, including native bees and wax (²⁰⁶Pb/²⁰⁷Pb = 1.072-1.121, ²⁰⁸Pb/²⁰⁷Pb = 2.348-2.409), co-located soil and dust (²⁰⁶Pb/²⁰⁷Pb = 1.090-1.122, ²⁰⁸Pb/²⁰⁷Pb = 2.368-2.403), corresponded most closely to aerosols collected during the period of leaded petrol use (²⁰⁶Pb/²⁰⁷Pb = 1.067-1.148, ²⁰⁸Pb/²⁰⁷Pb = 2.341-2.410). A large range of Pb isotopic compositions in beehive samples suggests that other legacy sources, such as Pb-based paints and industrials, may have also contributed to Pb contamination in beehive samples. Native bee data were compared to corresponding samples from the more common European honey bee (Apis mellifera). Although Pb isotopic compositions were similar in both species, significant differences in trace element concentrations were evident across the trace element suite, the bees and their products. The statistical association between T. carbonaria and co-located environmental contaminant concentrations were stronger than those in European honey bees, which may be attributable to its smaller foraging distance (0.3-0.7 km versus 5-9 km, respectively). This implies that T. carbonaria may be more suitable for assessing small spatial scale variations of trace element concentrations than European honey bees.

Identifying Sources of Environmental Contamination in European Honey Bees (Apis mellifera) Using Trace Elements and Lead Isotopic Compositions

Trace element concentrations (As, Mn, Pb, and Zn) and Pb isotopic compositions were analyzed in honey bees, wax, and honey along with co-located soil and dust samples from Sydney metropolitan and Broken Hill, Australia. Compared with the other trace elements, Pearson correlations show that Pb concentrations in soil and dust had the strongest relationship to corresponding values in honey bees and their products. Dust Pb was not only highly correlated to corresponding soil values (r = 0.806, p = 0.005), it was the strongest predictor of Pb concentrations in honey bees, wax, and honey (p = 0.001, 0.007, 0.017, respectively). Lead isotopic compositions (²⁰⁶Pb/²⁰⁷Pb and ²⁰⁸Pb/²⁰⁷Pb) showed that honey bees and their products from Broken Hill were nearly identical (95-98%) to the composition of the local ore body. Samples of honey bees and their products collected from background sites adjacent to national parks in Sydney had Pb isotopic compositions (²⁰⁶Pb/²⁰⁷Pb = 1.138-1.159, ²⁰⁸Pb/²⁰⁷Pb = 2.417-2.435) corresponding to local geogenic values (²⁰⁶Pb/²⁰⁷Pb = 1.123-1.176, ²⁰⁸Pb/²⁰⁷Pb = 2.413-2.500). By contrast, honey bees and their products from Sydney metropolitan (²⁰⁶Pb/²⁰⁷Pb = 1.081-1.126, ²⁰⁸Pb/²⁰⁷Pb = 2.352-2.408) were similar to aerosols measured during the period of leaded petrol use (²⁰⁶Pb/²⁰⁷Pb = 1.067-1.148, ²⁰⁸Pb/²⁰⁷Pb = 2.341-2.410). These measurements show Pb concentrations and its isotopic compositions of honey bees, and their products can be used to trace both legacy and contemporary environmental contamination, particularly where sources are well documented. Moreover, this study demonstrates that legacy Pb emissions continue to be remobilized in dust, contaminating both food and ecological systems.

Reducing risk and increasing confidence of decision making at a lower cost: In-situ pXRF assessment of metal-contaminated sites

This study evaluates the in-situ use of field portable X-ray Fluorescence (pXRF) for metal-contaminated site assessments, and assesses the advantages of increased sampling to reduce risk, and increase confidence of decision making at a lower cost. Five metal-contaminated sites were assessed using both in-situ pXRF and ex-situ inductively coupled plasma mass spectrometry (ICP-MS) analyses at various sampling resolutions. Twenty second in-situ pXRF measurements of Mn, Zn and Pb were corrected using a subset of parallel ICP-MS measurements taken at each site. Field and analytical duplicates revealed sampling as the major contributor (>95% variation) to measurement uncertainties. This study shows that increased sampling led to several benefits including more representative site characterisation, higher soil-metal mapping resolution, reduced uncertainty around the site mean, and reduced sampling uncertainty. Real time pXRF data enabled efficient, on-site decision making for further judgemental sampling, without the need to return to the site. Additionally, in-situ pXRF was more cost effective than the current approach of ex-situ sampling and ICP-MS analysis, even with higher sampling at each site. Lastly, a probabilistic site assessment approach was applied to demonstrate the advantages of integrating estimated measurement uncertainties into site reporting.

Geochemical sources, forms and phases of soil contamination in an industrial city

This study examines current soil contamination in an Australian industrial city, Newcastle. Public (roadside verges and parks) and private (homes) surface soils (n=170) contained metal(loid)s elevated above their respective Australian Health Investigation Levels (HIL). Lead (Pb), the most common contaminant in the city, exceeds the HIL for residential soils (HIL-A, 300mg/kg) in 88% of private soils (median: 1140mg/kg). In-vitro Pb bio-accessibility analysis of selected soils (n=11) using simulated gastric fluid showed a high affinity for Pb solubilisation (maximum Pb concentration: 5190mg/kg, equating to 45% Pb bio-accessibility). Highly soluble Pb-laden Fe- and Mn-oxides likely contribute to the bio-accessibility of the Pb. Public and private space surface soils contain substantially less radiogenic Pb (range: ²⁰⁸Pb/²⁰⁷Pb: 2.345-2.411, ²⁰⁶Pb/²⁰⁷Pb: 1.068-1.312) than local background soil (²⁰⁸Pb/²⁰⁷Pb: 2.489, ²⁰⁶Pb/²⁰⁷Pb: 1.198), indicating anthropogenic contamination from the less radiogenic Broken Hill type Pb ores (²⁰⁸Pb/²⁰⁷Pb: 2.319, ²⁰⁶Pb/²⁰⁷Pb: 1.044). Source apportionment using Pb isotopic ratio quantification and soil mineralogy indicate the city's historic copper and steel industries contributed the majority of the soil contaminants through atmospheric deposition and use of slag waste as fill material. High-temperature silicates and oxides combined with rounded particles in the soil are characteristic of smelter dust emissions. Additionally, a preliminary investigation of polycyclic aromatic hydrocarbons in soils, sometimes associated with ferrous metal smelting, coal processing or burning of fossil fuels, shows that these too pose a health exposure risk (calculated in comparison to benzo(a)pyrene: n=12, max: 13.5mg/kg, HIL: 3mg/kg).