A multi-method approach for monitoring deep-sea fishes on a seamount in a marine protected area within the West Mariana Ridge

Management of marine resources is necessary as anthropogenic pressures increase, particularly for vulnerable deep-sea communities. Utilizing non-invasive methodologies (including eDNA sampling and visual surveys) are ideal for surveying protected areas as they do not damage the environment. Previous studies which utilize both eDNA and visual surveys to examine fish communities have noted that employing multiple methods increases species richness and diversity overall, though these comparisons typically only utilize single platforms for each method and sample at a single depth strata. We conducted two surveys (in 2020 and 2024) at Ritto Seamount within one of Japan’s offshore marine protected areas at two depth strata within the benthic boundary layer: shallow (500–700 m) and deep (1, 800–2, 100 m). Multiple platforms (including ROV, DSV, AUV, baited cameras, CTD Niskin, and in situ water filtration) were used, with data combined by method to assess community-level differences at each depth strata. Species richness per deployment was pooled into four categories (shallow/deep and eDNA/visual), and differences in species richness, community composition, detection probabilities, and the habitat of species detected were compared. Community overlap between methods was low (7.7% similarity for shallow, 1.4% for deep), though this is similar to previous surveys from shallow reef fishes. eDNA and visual surveys observed different community composition and species richness for the shallow site, with eDNA detecting more pelagic species compared to visual surveys. Certain groups (families Etmopteridae Pentanchidae, Setarchidae, and Triacathodidae) commonly observed from visual surveys were absent from eDNA metabarcoding and occupancy models for a subset of species revealed different detection probabilities between methods, possibly due to different eDNA shedding rates and the need for additional eDNA sampling replicates. While the amount of eDNA in the environment roughly correlates to fish abundance, offshore currents, vertical mixing, and colder water temperatures within the deep sea could proliferate eDNA further from the source which may have contributed to the differences in community composition between methods. Utilizing multiple methodologies increased overall species richness, though more work is needed to determine the full fish community and to examine differences in community composition between survey platforms.