bioRxiv2017
Freshwater aquatic ecosystems provide a wide range of ecosystem services, yet provision of these services is increasingly threatened by human activities. Directly quantifying freshwater biotic assemblages has long been a proxy for assessing changing environmental conditions, yet traditional aquatic biodiversity assessments are often time consuming, expensive, and limited to only certain habitats and certain taxa. Sequencing aquatic environmental DNA via metabarcoding has the potential to remedy these deficiencies. Such an approach could be used to quantify changes in the relative abundances of a broad suite of taxa along environmental gradients, providing data comparable to that obtained using more traditional bioassessment approaches. To determine the utility of metabarcoding for comprehensive aquatic biodiversity assessments, we sampled aquatic environmental DNA at 25 sites that spanned the full length of the Potomac River from its headwaters to the Potomac estuary. We measured dissolved nutrient concentrations and also sequenced amplified marker genes using primer pairs broadly targeting four taxonomic groups. The relative abundances of bacteria, phytoplankton, invertebrate, and vertebrate taxa were distinctly patterned along the river with significant differences in their abundances across headwaters, the main river, and the estuary. Within the main river, changes in the abundances of these broad taxonomic groups reflected either increasing river size or a higher degree of eutrophication. The larger and more eutrophic regions of the river were defined by high total dissolved phosphorus in the water, a unique suite of bacteria, phytoplankton such as species of the diatom Nitzschia, invertebrates like the freshwater snail Physella acuta, and high abundance of fish including the common carp (Cyprinus carpio). Taxonomic richness of phytoplankton and vertebrates increased downriver while it consistently decreased for bacteria. Given these results, multi-assemblage aquatic environmental DNA assessment of surface water quality is a viable tool for bioassessment. With minimal sampling effort, we were able to construct the equivalent of a freshwater water quality index, differentiate closely-related taxa, sample places where traditional monitoring would be difficult, quantify species that are difficult to detect with traditional techniques, and census taxa that are generally captured with more traditional bioassessment approaches. To realize the full potential of aquatic environmental DNA for bioassessment, research is still needed on primer development, a geographically broad set of reference sites need to be characterized, and reference libraries need to be further developed to improve taxonomic identification.