[en] Highlights: • Rodenticide exposure increased with the contribution of urban habitat. • Red kites and urban goshawks were at greatest risk for rodenticide poisoning. • Rodenticide exposure may not be limited to terrestrial food webs. • Ibuprofen and fluoroquinolones were the most frequently detected medicinal products. • Most analysed and currently used plant protection products were not detected. Intensification of agricultural practices has resulted in a substantial decline of Europe's farmland bird populations. Together with increasing urbanisation, chemical pollution arising from these land uses is a recognised threat to wildlife. Raptors are known to be particularly sensitive to pollutants that biomagnify and are thus frequently used sentinels for pollution in food webs. The current study focussed on anticoagulant rodenticides (ARs) but also considered selected medicinal products (MPs) and frequently used plant protection products (PPPs). We analysed livers of raptor species from agricultural and urban habitats in Germany, namely red kites (MIML; Milvus milvus), northern goshawks (ACGE; Accipiter gentilis) and Eurasian sparrowhawks (ACNI; Accipiter nisus) as well as white-tailed sea eagles (HAAL; Haliaeetus albicilla) and ospreys (PAHA; Pandion haliaetus) to account for potential aquatic exposures. Landscape composition was quantified using geographic information systems. The highest detection of ARs occurred in ACGE (81.3%; n = 48), closely followed by MIML (80.5%; n = 41), HAAL (38.3%; n = 60) and ACNI (13%; n = 23), whereas no ARs were found in PAHA (n = 13). Generalized linear models demonstrated (1) an increased probability for adults to be exposed to ARs with increasing urbanisation, and (2) that species-specific traits were responsible for the extent of exposure. For MPs, we found ibuprofen in 14.9% and fluoroquinolones in 2.3% in individuals that were found dead. Among 30 investigated PPPs, dimethoate (and its metabolite omethoate) and thiacloprid were detected in two MIML each. We assumed that the levels of dimethoate were a consequence of deliberate poisoning. AR and insecticide poisoning were considered to represent a threat to red kites and may ultimately contribute to reported decreased survival rates. Overall, our study suggests that urban raptors are at greatest risk for AR exposure and that exposures may not be limited to terrestrial food webs.

[en] Highlights: • Hair THg concentration is strongly correlated with liver and moderately with kidney THg concentrations • Equations are provided to extrapolate THg values from hair to soft tissues • Mean mercury levels in Icelandic arctic foxes are higher than reported values of this species from other arctic regions Monitoring organic pollutants in wildlife is a common approach to evaluate environmental health, chemical exposure and to make hazard assessments. However, pollutant concentrations measured from different tissue types among studies impede direct comparisons of levels and toxicity benchmarks among species and regions. For example, mercury (Hg) is a metal of both natural and anthropogenic origin which poses health risks for marine and arctic biota in particular. Although hair is recognized as the least invasive sample type for Hg exposure measurement in wildlife, measurements in previous studies have used different tissues among individuals and species. This lack of tissue type consistency hinders cross study comparisons. Therefore to systematically evaluate the use of hair in ecotoxicological studies, total mercury (THg) concentrations measured from hair were compared to values obtained from liver and kidney in 35 Icelandic arctic foxes (Vulpes lagopus). THg concentrations varied considerably among tissues with hair and kidney levels generally lower than in liver. Nevertheless, significant correlations among tissue types were observed. THg values in hair were predictive for liver (R² = 0.61) and kidney THg levels (R² = 0.51) and liver values were a good predictor of THg in kidney (R² = 0.77). We provide further evidence that non-invasively collected hair samples reflect the THg levels of internal tissues. We present equations derived from multiple linear regression models that can be used to relate THg levels among tissue types in order to extrapolate THg values from hair to soft tissues. Using these equations, we compare the results of previous studies monitoring THg levels in different tissues of arctic foxes from various regions of the Arctic. Our findings support that hair is a suitable sample matrix for ecotoxicological studies of arctic predators and may be applied in both wildlife welfare and conservation contexts for arctic vulpine species.