Environmental DNA for detecting marine fish species: testing the volume of sampled water, markers and primers in a public oceanarium.

To maximize the efficiency of marine fish detection using environmental DNA (eDNA) metabarcoding, methodological factors such as PCR amplification bias and sampled water volume must be carefully considered. In this study, we evaluated the influence of different genetic markers (COI, 12S rRNA, and 16S rRNA) and water sample volumes (0.5, 1.0, 2.5, 5.0, and 10 L) on eDNA-based detection of 60 marine fish species maintained at the Oceanário de Lisboa. Five primer sets were used to generate amplicon libraries from triplicate water samples per tested volume. Overall, the combined use of all markers enabled the detection of 81.7% of the fish species present in the tank (n=49), as well as 37 additional exogenous taxa. Detection success varied among primer sets, ranging from 31.7% for COI FishATL_Cocktail1 to 73.3% for Fish16S. When accounting for reference sequence availability, relative detection success reached 82.4% for 12S and 77.2% for 16S. Increasing water volume consistently led to higher species detection across all markers, with detection percentages approaching saturation at the highest volumes tested. When pooling all markers, up to 76.7% of the species present were detected using a water volume of 5 L. These results highlight the limitations of relying on a single genetic marker and demonstrate the strong influence of sampling volume on detection efficiency, a factor likely to be even more critical in open marine environments.