Affiliation:
1. School of Molecular and Life Sciences Curtin University Bentley WA Australia
2. Australian National Fish Collection National Research Collections Australia, CSIRO Hobart TAS Australia
3. Western Australian Fisheries and Marine Research Laboratories Department of Primary Industries and Regional Development Government of Western Australia Hillarys WA Australia
Abstract
Abstract
Medium‐ to large‐scale marine protected areas (MPAs) are playing an increasingly important role in global marine biodiversity conservation. A key question is “how do we collect relevant data on biodiversity and population trends to inform their design and measure success?”
This question is particularly relevant for marine taxa that are difficult to survey, such as sharks and rays, and where populations may occur over vast areas. Environmental DNA (eDNA) metabarcoding has proven to be an effective sampling method and may provide a solution to these challenges; however, it remains unclear how its sampling effectiveness compares with traditionally used methods for elasmobranch surveying.
Here, we directly compared the efficacy of eDNA metabarcoding and demersal longline deployments to survey elasmobranchs across 31 sites over a 55,000 km2 area of the Kimberley and Roebuck Australian Marine Parks in Australia's North‐west Marine Parks Network.
In total, we documented 49 unique elasmobranch taxa, 36 of which were detected by eDNA and 32 (from a total of 815 captured sharks) by longline. A combined approach yielded over 34% more elasmobranch taxa than either method alone.
Site community compositions varied between the two survey methods; notably eDNA was able to detect species from outside of the immediate sampling area, although this was still consistent with a detection radius of a few kilometres, highlighting a particular use in rugose reef habitats where it is difficult to deploy longlines.
In investigating the quantitative use of eDNA data, we report that eDNA metabarcoding read abundance (count) data was poorly correlated with longline aggregate catch (count and biomass) data across raw, relative and rank abundance measures. However, we found that in multivariate analyses, both binary (presence–absence) and abundance (after square‐root transformation) datasets produced highly similar ordination patterns, largely segregating method (eDNA vs. longline), followed by latitude and depth.
Lastly, we identified required levels of eDNA replication and longline deployments to maximize captured elasmobranch diversity.
This study sets an integrated, georeferenced baseline and long‐term monitoring approach for the management and conservation of elasmobranch diversity within this unique marine park network.