Fish diversity assessment and semi-quantitative biomass estimation through metabarcoding of environmental DNA

• Computational model shows up to 70% correlation between eDNA-based predictions and trawl-derived biomass data. • eDNA metabarcoding provides both presence/absence data and robust quantitative information. • 81% overlap in species diversity assessments between eDNA and bottom-trawl analyses. A sustainable management of marine fisheries resources requires the continuous development and adaptation of monitoring and tracking techniques, both in terms of species diversity and abundance. In order to find alternatives to common invasive catch-based biodiversity assessments, we tested a fish-specific environmental DNA (eDNA) metabarcoding approach, targeting the mitochondrial 12S rRNA gene. Analysis of 64 seawater and sediment samples revealed a high detectability of fish DNA in the North Sea, with more reliable results in water compared to sediment samples. Species diversity assessments overlapped by 81 % between eDNA and bottom trawl catch analyses, with more than 32 % of the species detected only by eDNA compared to bottom trawling as reference value (100 %). With regard to quantitative biomass estimates for common dab ( Limanda limanda) , the eDNA-based estimates revealed similar results as compared to bottom-trawl catches. Modelling of trawl outputs as a function of read counts and sampling depth yielded up to 70% correlation between the model and the observed data for common dab. The outcomes of this study again highlight the potential of eDNA for marine biodiversity monitoring, not only for presence/absence assignments of species, but also for biomass estimates, with a high degree of reliability as compared to reference methods. The model reveals the importance of including environmental data to correct the bias linked to the persistence of eDNA in seawater.