Perfluoroalkyl substances in soft tissues and tail feathers of Belgian barn owls (Tyto alba) using statistical methods for left-censored data to handle non-detects

Perfluoroalkyl substances (PFASs) were investigated in tail feathers and soft tissues (liver, muscle, preen gland and adipose tissue) of barn owl (Tyto alba) road-kill victims (n = 15) collected in the province of Antwerp (Belgium). A major PFAS producing facility is located in the Antwerp area and levels of PFASs in biota from that region have been found to be very high in previous studies. We aimed to investigate for the first time the main sources of PFASs in feathers of a terrestrial bird species. Throughout this study, we have used statistical methods for left-censored data to cope with levels below the limit of detection (LOD), instead of traditional, potentially biased, substitution methods.Perfluorooctane sulfonate (PFOS) was detected in all tissues (range: 11 ng/g ww in muscle–1208 ng/g ww in preen oil) and in tail feathers (< 2.2–56.6 ng/g ww). Perfluorooctanoate (PFOA) was measured at high levels in feathers (< 14–670 ng/g ww), but not in tissues (more than 50% < LOD). Perfluorohexane sulfonate (PFHxS) could only be quantified in liver and preen oil, while other PFASs were sporadically detected in liver. PFOS levels in feathers and liver were highly correlated (r = 0.78, p < 0.01), in contrast to PFOA (r = ? 0.11, p = 0.78). Combined with high PFOA levels in feathers this suggests that PFOA may be present on the external surface of feathers, due to external contamination originating from the air in the vicinity of point sources. Therefore the possibility of using feathers as a passive air sampler for high PFOA levels should be investigated in the future.