Spatial extent and age-related bioaccumulation of fluoride in the bones of house sparrows around aluminium smelting operations.

4. Discussion

In this study, we examined the distribution and extent of fluoride exposure in populations of house sparrows inhabiting urban areas near two aluminium smelters. We found evidence for elevated bone fluoride concentrations which were spatially related to aluminium smelting emissions and persisted over a large geographic area.

Across both locations in this study, there was only one site where bone fluoride concentrations in adult house sparrows (380–490 mg/kg, n = 2) were consistently below levels representing background concentrations (600 mg/kg) in mature passerine birds from uncontaminated areas (Table 1; Kay et al., 1975; Robertson and Lock, 1994; Stewart et al., 1974). This site was 11.5 km from the Portland Aluminium Smelter and is outside the range of significant emissions from this facility (Fig. 1a). In New Zealand, adult house sparrows inhabiting unindustrialised locations without significant emission sources had mean bone fluoride concentrations of 626 ± 412 mg/kg (Robertson and Lock, 1994). At an uncontaminated site in North America, similar concentrations (536 ± 311 mg/kg) have been reported for European magpies (Pica pica; Kay et al., 1975). Along with our findings, these studies indicate high variability in background bone fluoride concentrations within and between bird species which are likely related to differences in age and diet. The concentrations reported for these species, however, are consistently lower than those of birds inhabiting fluoride contaminated environments.

At catch sites within 5 km (the overall median distance of sites) of the smelting facilities in Portland and George Town, most house sparrows (90 %) had bone fluoride levels exceeding 600 mg/kg, and almost half of adults (43 %) within this distance exceeded 1200 mg/kg. This suggests that emissions from local aluminium smelting facilities are contributing to elevated bone fluoride levels in nearby house sparrows. The highest recorded concentration from adults in Portland (1200–1400 mg/kg, n = 2) and George Town (1400–1800 mg/kg, n = 2) were from the sites closest to aluminium smelting operations (Fig. 2). Similarly, in a survey of Canada geese (Branta canadensis) in Norway, the highest bone fluoride concentrations (1844–2156 mg/kg) were found in individuals sampled closest to an aluminium smelting facility (Vikøren and Stuve, 1995). Based on a limited number of samples (n = 9), similarly elevated fluoride levels (1094–3500 mg/kg) were reported in a population of house sparrows inhabiting areas around an aluminium smelter in New Zealand (Robertson and Lock, 1994). While these studies do not explore local-scale variability in exposure, they indicate that proximity to aluminium smelting operations is a primary determinant of elevated levels of fluoride in birds.

Few studies have investigated how the spatial distribution of industrial fluoride emissions influences exposure in wildlife at a site-specific scale. Here, bone fluoride concentrations in house sparrows decreased significantly with distance to the nearest aluminium smelter (Table 3, Fig. 2). The rate of decline indicates that background bone fluoride concentrations (600 mg/kg) in adult house sparrows would be reached at a distance of approximately 9.8 km and 18.1 km from the aluminium smelters in Portland and George Town, respectively. This relationship is based on sites located between 1.9 and 11.5 km from smelting facilities and does not account for higher fluoride deposition anticipated closer to the operations (Davis et al., 2016). Surveys indicate that fluoride concentrations in vegetation drop steeply within a 1 km radius of either smelter and more gradually outside this range (Bell Bay Aluminium, 2024; Davis et al., 2016). In other contexts, elevated fluoride concentrations in vegetation have been reported upwards of 10 km from the emission source (Afanasyeva et al., 2021), with impacts on herbivorous animals reported over similar distances (Kierdorf and Kierdorf, 1999). Our findings suggest that in house sparrows, levels of fluoride exposure associated with aluminium smelting emissions may remain elevated over a similar geographic extent.

Adjusting for site distance, mean bone fluoride levels in house sparrows from George Town were significantly higher than in Portland (Table 3). Data from the National Pollution Inventory (NPI) indicate that fluoride emissions from the Bell Bay Aluminium Smelter, near George Town, were approximately 14 % higher than emissions from the Portland Aluminium Smelter over the same time period (National Pollutant Inventory, 2024). Given that the distribution of sites was not significantly different between these two locations, higher fluoride emissions from the Bell Bay Aluminium Smelter may explain the higher bone fluoride levels detected in house sparrows from George Town. Other sources of contamination may also contribute to the observed differences between the populations. In George Town, for example, consistently elevated bone fluoride levels were measured in house sparrows from a peri-urban site approximately 7.5 km north of the smelter (Fig. 1b). The levels seen here could relate to the reported application of fluoride enriched phosphate fertilisers to pasture around this site (Fuge, 2019). This is corroborated by a spike in fluoride concentrations measured in vegetation at nearby monitoring sites over the house sparrow sampling period (Bell Bay Aluminium, 2024).

Age was also a significant predictor of variability in bone fluoride (Table 3). Mean bone fluoride concentrations in adult house sparrows (1062 mg/kg) were higher than in juveniles (603 mg/kg). We also identified a positive relationship between morphometric measurements and bone fluoride, suggesting that the rate of fluoride uptake into mineralised tissue is related to an individual’s body size, which often correlates with age. Growth rates are fastest during the early post-natal stages, and house sparrows, like most songbirds, leave their nest with limbs and bones that are still developing (Aldredge, 2016). This growth is supported by a protein rich invertebrate diet (Klva?ová et al., 2012). Since fluoride bioaccumulates in invertebrates (Aguirre-Sierra et al., 2013), the concentrations detected at this early life stage follow expectations. Bone fluoride in juveniles (aged 0–4 months) reached almost half of the concentration detected in adults (aged > 1 year) within at most four months of development. These findings are consistent with other studies examining age-related fluoride exposure in a range of animals, including birds. In black-crowned night herons (Nycticorax nycticorax) sampled downstream of a phosphate production plant, bone fluoride concentrations increased twice as much between hatchling and 1 year old birds than between 1 year old and 2 year old birds (Henny and Burke, 1990). This suggests that while overall levels in juveniles are lower than in mature individuals of the same species, juveniles are more susceptible to fluoride exposure due to its rapid accumulation in mineralised tissue during the early stages of development.

Sex-related differences in fluoride bioaccumulation in birds are less clear. Average bone fluoride concentrations were marginally higher in adult male than in female house sparrows, though this difference was not significant (Table 3). In experimentally dosed screech owls (Megascops asio), Pattee et al. (1988) found higher bone fluoride concentrations in females than in males. A similar difference was identified between male and female European herring gulls (Larus argentatus) from a colony adjacent to an aluminium smelter, but not from another colony at an uncontaminated site (Vikøren and Stuve, 1996). Similarly, no significant difference was found in the bone fluoride concentration of male and female goosanders (Mergus merganser) inhabiting uncontaminated areas in Poland (Kalisinska et al., 2014). Further research is needed to resolve sex-related differences in fluoride accumulation in wild birds.

Chronic fluoride exposure can lead to changes in bone mineral composition. In adult house sparrows, higher bone fluoride levels were associated with decreased bone phosphorus concentrations while calcium remained stable (Fig. 3a). In bone tissue, fluoride ions substitute for hydroxyl groups in hydroxyapatite, forming the more stable crystalline mineral fluorapatite (Ciosek et al., 2021). This process may alter normal phosphorus incorporation during bone mineralisation, contributing to reduced phosphorus content in the mineral matrix (Wang et al., 2025). In contrast, bone calcium is regulated by parathyroid hormone and other feedback mechanisms (e.g., vitamin D activity, renal reabsorption, and intestinal absorption) that may maintain calcium in homeostasis despite fluoride-induced changes in bone mineral composition (Simon et al., 2014; Wang et al., 2015). Phosphorus is essential for energy metabolism, cellular signalling, and maintaining bone strength, and so its apparent mobilisation from adult bone tissue has potential consequences on the fitness of these birds (Ciosek et al., 2021; Wagner, 2024). For example, reduced phosphorus availability could compromise flight performance through weakened bone structure and impair reproductive success by limiting phosphorus transfer to developing eggs and offspring (Bird and Massari, 1983; Pattee et al., 1988).

Despite evidence for changes in bone mineral composition, we did not observe any effect of fluoride exposure on body condition. Research on fluoride exposure in birds has demonstrated a high tolerance to doses above levels typically encountered around industrial sources. Xie and Sun (2003) found no evidence of skeletal fluorosis in Adelie penguins (Pygiscelis adeliae) and South Polar skua (Stercorariusmaccormicki) despite highly elevated bone fluoride concentrations (832–7187 mg/kg). Similarly, doses of 1120–2240 mg/kg of dietary fluoride did not disrupt organ development in American kestrels (Falco sparverius), though did lead to a reduction in bone strength (Bird and Massari, 1983). At lower and more environmentally realistic concentrations, screech owls (Megascops asio) dosed with 200 mg/kg of dietary fluoride were observed to have lower hatching success than control birds (Pattee et al., 1988). Fluoride concentrations exceeding 200 mg/kg are reported in vegetation around the Portland and George Town aluminium smelters, though these levels are largely confined to the vegetated buffer zones surrounding either operation (Bell Bay Aluminium, 2024; Davis et al., 2016). Additionally, much of the fluoride taken up by plants is stored in the foliage, not in the fruit or seed which is targeted by granivorous and frugivorous birds (Panda, 2015). It is therefore unlikely that the house sparrows studied here are at risk of exposure to fluoride at levels linked to negative reproductive outcomes in avian dosing studies. However, given the importance of phosphorus to egg development, its decline points to a mechanistic pathway whereby reproductive outcomes could be affected in fluoride exposed birds (Sinclair-Black et al., 2023).

Our findings indicate that the vegetated buffer zones established around aluminium smelters near Portland and George Town are insufficient to prevent elevated bone fluoride accumulation in surrounding wildlife. Within the buffer zone surrounding the Portland Aluminium Smelter, exposure to fluoride enriched vegetation has been attributed to skeletal and dental fluorosis in a range of marsupial species (Death et al., 2015; Hufschmid et al., 2015). Other studies have reported similar impacts in herbivorous mammals sampled at much greater distances from fluoride emission sources. For example, a high incidence of dental fluorosis was identified in red deer (Cervus elaphus) living over 10 km from power plants in the Czech Republic (Kierdorf and Kierdorf, 1999). Here, the large geographic extent of elevated fluoride exposure in house sparrows indicates heightened risk to the health of susceptible wildlife foraging within a similar distance of aluminium smelting facilities. The extent of exposure was not captured by existing vegetation surveys from either Portland or George Town (Bell Bay Aluminium, 2024; Davis et al., 2016), which underscores the value of sedentary species in monitoring fluoride emissions. The long-term bioaccumulation of fluoride in the bones of such species can reveal patterns of contamination not as apparent from more temporally variable vegetation data.

In this study, we investigated fluoride exposure in two populations of house sparrows inhabiting areas impacted by aluminium smelting emissions. Bone fluoride concentrations in adult house sparrows remained elevated compared to background levels up to a distance of at least 9.8 km from the emission source and well beyond the designated buffer zones surrounding each smelter. Our findings reveal the potentially large geographic extent of exposure to industrial fluoride emissions in wildlife surrounding aluminium smelting operations. They further show the susceptibility of sedentary animals to the age-related bioaccumulation of fluoride and its possible influence on bone mineral composition. Further research is needed to understand subtoxic effects in birds and other wildlife exposed to environmentally realistic levels of fluoride, such as that relevant to the contexts examined here. This will become increasingly important as a growing extent and diversity of ecosystems are impacted by fluoride emissions in response to rising global demand for aluminium and its production.