Reliable and accurate biodiversity census methods are essential for monitoring ecosystem health and assessing potential ecological impacts of future development projects. Although metabarcoding is increasingly used to study biodiversity across ecological research, morphology‐based identification remains the preferred approach for marine ecological impact assessments. Comparing metabarcoding to morphology‐based protocols currently used by ecological surveyors is essential to determine whether this DNA‐based approach is suitable for the long‐term monitoring of marine ecosystems.
We compared metabarcoding and morphology‐based approaches for the analysis of invertebrates in low diversity intertidal marine sediment samples. We used a recently developed bioinformatics pipeline and two taxonomic assignment methods to resolve and assign amplicon sequence variants (ASVs) from Illumina amplicon data. We analysed the community composition recovered by both methods and tested the effects, on the levels of diversity detected by the metabarcoding method, of sieving samples prior to DNA extraction.
Metabarcoding of the mitochondrial marker cytochrome c oxidase I (COI) gene recovers the presence of more taxonomic groups than the morphological approach. We found that sieving samples results in lower alpha diversity detected and suggests a community composition that differs significantly from that suggested by un‐sieved samples in our metabarcoding analysis. We found that whilst metabarcoding and morphological approaches detected similar numbers of species, they are unable to identify the same set of species across samples.
Synthesis and Applications We show that metabarcoding using the cytochrome c oxidase I (COI) marker provides a more holistic, community‐based, analysis of benthic invertebrate diversity than a traditional morphological approach. We also highlight current gaps in reference databases and bioinformatic pipelines for the identification of intertidal benthic invertebrates that need to be addressed before metabarcoding can replace traditional methods. Ultimately, with these limitations taken into consideration, resolving community‐wide diversity patterns with metabarcoding could improve the management of non‐protected marine habitats in the U.K.