Combining eDNA metabarcoding and morphological-based approaches to increase resolution of biodiversity assessments

Abstract

Environmental DNA (eDNA) metabarcoding has the potential to improve aquatic biodiversity assessments by improving detection capability, taxonomic resolution and cost-effectiveness when compared to morphological- based approaches. Advances have been made in the field of metabarcoding, but challenges remain, including sensitivity of results to DNA extraction method and marker choice, and its dependency on often-incomplete reference databases. Also, it does not easily provide quantitative estimates for surveyed species, whereas abundance and biomass data are often required. Here, we present a framework to combine datasets from eDNA metabarcoding- and morphological-based approaches in an attempt to address some of these limitations. Biodiversity surveys were conducted over time and space across the Iberian Pond Network, covering a biogeographic gradient, from semi-arid to alpine environments. Different markers were used to amplify DNA from different trophic levels (from bacteria to macroinvertebrates). Organisms were also collected with conventional methods, and were identified and enumerated under the microscope. Preliminary results showed that there were differences when comparing directly both datasets. However, both approaches identified similar environmental gradients, with regions exposed to greater environmental filters, e.g., higher temperatures in southern regions or colder temperatures in mountain tops, having fewer numbers of species than temperate regions. Combining eDNA metabarcoding- and morphological-based datasets improved the recovery of different taxa. These results indicate that eDNA-based assessments may not always directly replace morphological-based identification and biomass quantification, but might certainly complement and enhance current approaches when combined, especially for those groups that are perceived as difficult to identify, allowing more consistent and faster identifications.