Effects of polyhalogenated aromatic hydrocarbons and related contaminants on common tern reproduction: integration of biological, biochemical, and chemical data.

In eight Dutch or Belgian common tern (Sterna hirundo) colonies, breeding biology and food choice were determined, and 15 second eggs were collected from three-egg clutches for artificial incubation, biochemical analysis and analysis of yolk-sac polyhalogenated hydrocarbon (PHAH) levels. Results from these analyses were combined with biological data from the eggs remaining in each clutch. In some breeding colonies severe flooding, rainy and cold weather, and extreme predation caused extensive losses of eggs and chicks. A relationship was found between yolksac mono-ortho polychlorinated biphenyl (mo-PCB) levels and main food species (fish or insects) of the adult terns before egg-laying. Colony average breeding data differed only slightly, and were difficult to relate to PHAH-levels. When the colonies were grouped after yolksac PHAH-patterns and main food species, significant differences in average egg laying date, incubation period, egg volume and chick weight could be related to differences in yolksac PHAH and retinoid levels, and hepatic ethoxyresorufin-O-deethylase (EROD) activity. The data from all colonies also were combined into one data-set and correlated with the biochemical parameters and PHAH levels. In summary higher yolksac PHAH levels or hepatic EROD-activity correlated with and later egg laying, prolonged incubation period and smaller eggs and chicks. Lower yolksac retinoid- and plasma thyroid hormone levels, and a higher ratio of plasma retinol over yolksac retinoids correlated with later egg laying, prolonged incubation periods and smaller chicks and eggs. The dynamic environment of the terns had more obvious detrimental effects on breeding success than PHAHs. However, the more subtle effects observed for PHAHs could still be of importance during specific stress circumstances. To monitor site-specific reproduction effects, tree-nesting birds feeding on relatively big and non-migrating fishes would be most suitable. The use of specific biomarkers for exposure and effect is recommended to establish a causal relationship between a certain class of pollutants and an adverse biological effect.