Biases and best approaches for assessing debris ingestion in sea turtles, with a case study in the Mediterranean.

In a sample of 567 loggerhead turtles (Caretta caretta) from the central Mediterranean, debris occurrence varied according to methods and turtle source, and was up to 80% in pelagic turtles. Frequencies of plastic types, size and color are also reported. These results and a critical review of 49 studies worldwide indicate that: (i) the detected occurrence of plastic (% turtles) is affected by several factors (e.g., necropsy/feces, ecological zone, type and date of finding, captivity period for feces collection), (ii) mixed dataset and opportunistic approaches provide results which are biased , not comparable, and ultimately of questionable value, (iii) only turtles assumed to have had a normal feeding behaviour at the time of capture or death should be considered, (iv) turtle foraging ecology and possible selectivity may undermine the use of turtles as indicator species for monitoring marine litter, as recently proposed for the Mediterranean.