Trends in polychlorinated biphenyl residues in three British predatory bird species.

Liver total PCBs have been monitored in British sparrowhawks Accipiter nisus, kestrels Falco tinnunculus and herons Ardea cinerea since the 1960s. Concentrations do not appear to have declined in sparrowhawks or kestrels, despite bans and restrictions on PCB manufacture and disposal, but residues vary markedly between individuals which may obscure long-term trends. In this study measurements were made of PCB liver concentrations in sparrowhawks, kestrels and herons collected from locations throughout Britain between 1992 and 1997: (i) to determine the major causes of intraspecies variation in liver concentrations, and (ii) to characterise and quantify liver concentrations of individual congeners and their associated toxicity. The long-term PCB data were also reanalysed to determine if there was evidence of declines in liver PCBs once intra-species variation was accounted for. Body condition, and to a lesser extent age and sex, affected liver PCB congener, sum PCB and Toxic Equivalent Quotient (TEQ) concentrations. Concentrations were higher in starved than in non-starved birds, largely because of remobilisation of residues (sparrowhawks and herons) and starvation-induced liver wastage (all species). Adults generally had higher liver PCB concentrations than first-year birds, and male sparrowhawks and kestrels lower concentrations than females, although this varied with body condition. Congener profiles were similar across species, and concentrations of individual congeners were primarily determined by their potential to be metabolised. PCBs 138, 153, 180 and 170 accounted for up to 76% £PCB concentrations. Non-ortho, coplanar PCBs (primarily 118 and 169) were mostly detected in starved birds. Liver TEQ concentrations were below Lowest Observable Effect Concentrations. Intra-year variation in body condition obscured a decline in liver PCB concentrations in sparrowhawks, but liver PCB concentrations have not declined in kestrels. The results obtained provide insights into which factors should be taken into account when using tissue concentrations in biota to monitor changes in environmental exposure to contaminants