AUTHORS: David D. Duvernell, Missouri S&T; Veronica M. Lee, Missouri S&T; Eric J. Ludwig, Missouri S&T; Aaron D. Geheber, University of Central Missouri; Brett Landwer, Missouri Department of Conservation; Leah K. Berkman, Missouri Department of Conservation
ABSTRACT: Biodiversity monitoring in freshwater habitats provides essential data for assessing and quantifying spatiotemporal patterns and long-term changes. Fisheries biologists use a variety of capture-based methods for biodiversity sampling. Non-capture-based methods have come into common use that rely on the detection of environmental DNA (eDNA), which is released by organisms into their environment. Metabarcoding involves the filtering and extraction of eDNA from samples, followed by sequencing and alignment to a reference database to infer species presence. We worked closely with the Missouri Department of Conservation (MDC) to perform comparative biodiversity assessments using traditional capture-based methods along side eDNA metabarcoding methods. In support of our efforts, we submitted 189 new reference sequences to GenBank, to bring statewide databased coverage of Missouri fish species to over 90%. We teamed with the Resource Assessment and Monitoring (RAM) program to sample sites in six drainages of the Missouri Ozarks, and we worked with Cape Girardeau fisheries biologists to sample lowland habitats in Black Island Conservation Area along the Mississippi River. In Ozark streams we were able to detect approximately double the number of species compared to standard RAM sampling methods, while in the aquatic habitats at Black Island we increased the number of documented species by more than a third relative to MDC survey records dating back to 1940. eDNA sampling missed or failed to discriminate among very few species detected by RAM sampling or documented at Black Island. These missed species were either rare (often single specimens) or were pairs or trios of closely related species with highly similar sequences that were indiscernible using eDNA sequencing methods. Spatial analyses demonstrated that eDNA sampling was sensitive to changes in fish assemblages along Ozark stream habitat gradients, and habitat complexity at Black Island.
