Interspecific variation in the spatially-explicit risks of trace metals to songbirds

Abstract

Many wild animals can be adversely affected by trace metals around point sources but little is known about the risks to birds across their ranges. Trace metals in the soil are ubiquitously, if heterogeneously distributed, across the world due to natural and anthropogenic sources. Here, we built, parameterized and applied a spatially explicit modelling framework to determine the risks of soil-associated metals to 30 invertebrate-consuming passerine species across their spatial distribution in England and Wales. The model uses a risk characterization approach to assess the risks of soil-associated metals. Various monitoring datasets were used as input parameters: soil metal concentrations in England and Wales, bird spatial distribution; bird diet, bioaccumulation and toxicity data were extracted from the literature. Our model highlights significant differences in toxicity risks from Cd, Cu, Pb and Zn across the UK distributions of different species; Pb and Zn posed risks to all species across most of species' distributions, with more localised risks to some species of conservation concern from Cd and Cu. No single taxa of invertebrate prey drove avian exposure to metal toxicity. Adults were found to be at higher risk from Pb and Zn toxicity across their distributions than nestlings. This risk was partially driven by diet, with age differences in diets identified. Our spatially explicit model allowed us to identify areas of each species' national distribution in which the population was at risk. Overall, we determined that for all species studied an average of 32.7 ± 0.2%, 8.0 ± 0.1%, 86.1 ± 0.1% and 93.2 ± 0.1% of the songbird spatial distributions in the UK were characterized at risk of Cd, Cu, Pb and Zn, respectively. Despite some limitations, our spatially explicit model helps in understanding the risks of metals to wildlife and provides an efficient method of prioritising areas, contaminants and species for environmental risk assessments. The model could be further evaluated using a targeted monitoring dataset of metal concentration in bird tissues. Our model can assess and communicate to stakeholders the potential risks of environmental contaminants to wildlife species at a national and potentially international scale.