Effects of Electrofishing Gear Type on Spatial and Temporal Variability in Fish Community Sampling

Michael R. Meador¹ and Julie P. McIntyre²

¹U.S. Geological Survey, 12201 Sunrise Valley Drive, MS 413, Reston, Virginia, USA
²North Carolina State University, Department of Statistics, Raleigh, North Carolina, USA

ABSTRACT.—Fish community data collected from 24 major river basins between 1993 and 1998 as part of the U.S. Geological Survey’s National Water-Quality Assessment Program were analyzed to assess multiple-reach (three consecutive reaches) and multiple-year (3 consecutive years) variability in samples collected at a site. Variability was assessed using the coefficient of variation (CV) of species richness, Jaccard Index (JI), and Percent Similarity Index (PSI). Data were categorized by three electrofishing sample-collection methods—backpack, towed barge, and boat. Overall, multiple-reach CV values were significantly lower than those for multiple years, whereas multiple-reach JI and PSI values were significantly greater than those for multiple years. Multiple-reach and multiple-year CV values did not vary significantly among electrofishing methods, although, JI and PSI values were significantly greatest for backpack electrofishing across multiple reaches and multiple years. The absolute difference between mean species richness for multiple-reach samples and mean species richness for multiple-year samples was 0.8 species (9.5% of total species richness) for backpack samples, 1.7 species (10.1%) for towed barge samples, 4.5 species (24.4%) for boat-collected samples. Review of boat-collected fish samples indicated that representatives of four taxonomic families—Catostomidae, Centrarchidae, Cyprinidae, and Ictaluridae—were collected at all sites. Of these, catostomids exhibited greater inter-annual variability than centrarchids, cyprinids, or ictalurids. Caution should be exercised when combining boat-collected fish community data from different years because of relatively high inter-annual variability, primarily as a result of certain relatively mobile species. Such variability may obscure detection of any longer-term trends.