PS 37-44 - Comparison of eDNA metabarcoding markers to detect aquatic macroinvertebrates in reference streams within the Colorado River Basin

Abstract: Environmental DNA (eDNA) metabarcoding is widely growing in the field of ecology to better understand the effect that pollution and/or climate change have on a diverse range of ecosystems, to detect and monitor a variety of species, and inform management decisions. The Colorado River basin is one of the largest drainages in the western United States and biotic communities are increasingly subject to pressures such as fire and drought. Macroinvertebrates in streams can serve as useful indicators of environmental quality and climate change impacts. As a goal of this project, we are assessing the ability of eDNA metabarcoding to efficiently monitor aquatic macroinvertebrate communities, beginning with surveys at higher quality reference sites. Successful eDNA metabarcoding can be used to eliminate or lessen the difficulty of traditional macroinvertebrate sampling techniques, such as, the need to collect multiple samples, extensive sorting of invertebrates, and manual identification of taxa. Here, we used samples at two sites (in Colorado and Utah) with previous physical sampling to help validate our techniques and determine the best primers to be used in the metabarcoding of macroinvertebrates. Each eDNA sampling event consisted of triplicate collections of water filter samples at three different locations in each stream, along with the collection of 3 negative controls at each site (N = 24). We filtered water through a 5-micron self-desiccating filter until filters clogged (average 1-3 L per filter) and preserved each sample on wet ice until DNA could be extracted. During metabarcoding library preparation we tested four different primer pairs from the COI mitochondrial region, sequenced libraries on a MiSeq system, and compared taxa from kick-net samples to those identified from each primer pair to determine the best marker(s) to be used in our protocol. We identified macroinvertebrates down to the lowest taxonomic level possible via DNA sequences from metabarcoding results and compared these to visually identified macroinvertebrates from kick-net samples. Preliminary results suggest that macroinvertebrate taxonomic diversity detected varies greatly depending on the metabarcoding primer, and certain taxonomic groups are amplified preferentially when compared to kick-net sample results. In particular, two primer sets appear to provide the most comprehensive taxa identification for macroinvertebrates in these streams. Our next steps will be using these metabarcoding markers to detect macroinvertebrates from samples collected at reference streams across the entire basin.