Macroinvertebrate community diversity and habitat quality relationships along a large river from targeted eDNA metabarcode assays

Abstract

Abstract Species compositions and diversity levels of aquatic macroinvertebrate communities provide important indicators of ecosystem health. However, such community analyses typically are limited by time and effort of sampling, sorting, and identifications, as well as morphological character uncertainty for some taxa, especially at early life stages. Our objective was to evaluate a suite of targeted metabarcode high‐throughput sequencing assays to characterize the macroinvertebrate communities (specifically targeting Annelida, Bryozoa, Crustacea, Insecta, and Mollusca) from environmental (e)DNA water samples along 160 km of the Maumee River, OH (a major Lake Erie, Laurentian Great Lakes tributary). Multiple alpha (richness, Shannon Index, and Simpson's Index) and beta (Bray–Curtis and Sørensen dissimilarities) diversity metrics from the metabarcode assays were compared with an Invertebrate Community Index (ICI) metric calculated from traditional morphological sampling surveys conducted by the Ohio Environmental Protection Agency. The 15 Maumee River sites varied in their ICI scores (ranging from 4 to 46), with seven sites rated as “poor” or “very poor” and eight scoring “fair” or “good.” Metabarcoding assays yielded greater gamma richness, delineating 181 Operational Taxonomic Units versus 172 taxa from morphology (the latter often limited to family or genus‐level identifications). Both datasets supported similar river‐wide trends, with comparable gamma, alpha, and beta diversity patterns and community compositions across habitat types and habitat quality scores. Metabarcode assays revealed similar detection of important bioindicator Insecta, but missed most Trichoptera (caddisflies). eDNA identified eight aquatic invasive species on the GLANSIS (Great Lakes Aquatic Nonindigenous Species Information System) list, including three missed by the morphological surveys ( Branchiura sowerbyi , Potamothrix bedoti , and Skistodiaptomus pallidus; with Lophopodella carteri , Faxonius rusticus , Corbicula fluminea , Dreissena polymorpha , and Valvata piscinalis in both datasets). Findings illustrate the utility of eDNA sampling and targeted metabarcode assays to enhance and complement environmental assessments of aquatic ecosystems.