Comparison of American Fisheries Society (AFS) Standard Fish Sampling Techniques and Environmental DNA for Characterizing Fish Communities in a Large Reservoir

Abstract

Recently, methods involving examination of environmental DNA (eDNA) have shown promise for characterizing fish species presence and distribution in waterbodies. We evaluated the use of eDNA for standard fish monitoring surveys in a large reservoir. Specifically, we compared the presence, relative abundance, biomass, and relative percent composition of Largemouth Bass Micropterus salmoides and Gizzard Shad Dorosoma cepedianum measured through eDNA methods and established American Fisheries Society standard sampling methods for Theodore Roosevelt Lake, Arizona. Catches at electrofishing and gillnetting sites were compared with eDNA water samples at sites, within spatial strata, and over the entire reservoir. Gizzard Shad were detected at a higher percentage of sites with eDNA methods than with boat electrofishing in both spring and fall. In contrast, spring and fall gillnetting detected Gizzard Shad at more sites than eDNA. Boat electrofishing and gillnetting detected Largemouth Bass at more sites than eDNA; the exception was fall gillnetting, for which the number of sites of Largemouth Bass detection was equal to that for eDNA. We observed no relationship between relative abundance and biomass of Largemouth Bass and Gizzard Shad measured by established methods and eDNA copies at individual sites or lake sections. Reservoirwide catch composition for Largemouth Bass and Gizzard Shad (numbers and total weight [g] of fish) as determined through a combination of gear types (boat electrofishing plus gillnetting) was similar to the proportion of total eDNA copies from each species in spring and fall field sampling. However, no similarity existed between proportions of fish caught via spring and fall boat electrofishing and the proportion of total eDNA copies from each species. Our study suggests that eDNA field sampling protocols, filtration, DNA extraction, primer design, and DNA sequencing methods need further refinement and testing before incorporation into standard fish sampling surveys.

Received October 26, 2016; accepted June 9, 2017Published online August 10, 2017

ACKNOWLEDGMENTS

We thank the Arizona Game and Fish Department (AZGFD), USGS, and University of Arizona (UA) for project funding. Funding for this work was primarily provided through U.S. Fish and Wildlife Service Cooperative Agreement 843208J477. We are particularly grateful to the USGS–UMESC personnel for their collaboration on this project. We thank Tony Robinson, Matt O’Neil, and Esther Rubin (AZGFD) for assistance with project coordination and advice; Joe Sayer (AZGFD), Sam Rector, Kristina Ertzner, Chelsea Powers, Colleen Svancara, Travis Wagner, Ryan Schaefer, Ian Harding, Dave Daniels (AZGFD), Kendal Perez, and Ronald John Perez for assistance with fieldwork on Lake Roosevelt; and Lindsey Fera (UA) for administrative assistance. All research was conducted in accordance with Protocol Number 09-088 approved by the Institutional Animal Care and Use Committee at UA. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.