Tropical seagrass habitats support diverse fish assemblages and provide critical nursery habitat for many ecologically and economically important species. However, biodiversity assessments in these systems are challenging due to turbidity, suspended sediments, and dense vegetation, which can restrict both traditional capture- and visual-based surveys, as well as emerging molecular approaches. In this study, we compared fish assemblages detected using beam trawl sampling and multi-assay environmental DNA (eDNA) metabarcoding, including two fish-targeted assays, 16S-Fish/D and MiFish-U, and the broader metazoan COI-Leray assay in a turbid tropical seagrass meadow in Trinity Inlet, north Queensland, Australia. Environmental DNA metabarcoding detected greater genus-level richness than beam trawling, recovering 112 genera compared with 27 genera detected by beam trawl surveys. Beam trawls detected a smaller subset of taxa associated with local, trawl-accessible fishes, including four taxa not recovered by eDNA metabarcoding. Among the metabarcoding assays, MiFish-U and 16S-Fish/D recovered substantially more fish taxa than COI-Leray, which produced limited fish detections. Univariate analyses indicated that site-level richness patterns varied among sampling approaches, with MiFish-U detecting higher richness at Ellie Point, and beam trawl sampling detecting higher richness at the Esplanade. Multivariate analyses revealed a significant Method x Site interaction, suggesting fish assemblages detected by the two sampling techniques respond to local conditions in different ways. Incomplete overlap in taxonomic detections between MiFish-U and 16S-Fish/D further highlighted the strong influence of marker choice on fish detection. These findings demonstrate that eDNA metabarcoding and beam trawl surveys capture complementary components of seagrass-associated fish assemblages within the same habitat. Integrating molecular and capture-based methods therefore provides a more comprehensive and reliable assessment of biodiversity in turbid tropical seagrass ecosystems. • eDNA metabarcoding detected a broader range of fish genera than beam trawling. • Beam trawls confirmed local occurrence for a smaller subset of taxa. • Fish richness patterns varied among methods and sites. • Primer selection influenced fish community composition inferred from eDNA metabarcoding. • Integrated sampling improves biodiversity assessment in turbid tropical seagrass systems.
Philpott et al. (Fri,) studied this question.