Comparison of Catch per Effort and Removal Procedures for Sampling Stream Fish Assemblages

Methods used to estimate fish abundance in streams should be chosen based on the precision required by the study, the available time, and the number and kinds of species targeted. When entire assemblages of predominantly small, nongame fishes are to be sampled, most existing procedures have limitations. We compared estimates of species richness, abundance, and assemblage structure based on catch per effort (CPE) during a single tow-barge electrofishing sample versus intensive tow-bare electrofishing removal sampling procedures with block nets in paired, contiguous stations on nine streams in southern Wisconsin. Use of block nets had little effect on CPE during single upstream electrofishing passes; for stations approximately 35 times the mean stream width in length, the overall influence of fish entering and leaving the station appeared to be negligible. Estimates of abundance, based on total catch and based on the removal model, were higher in removal stations than in CPE stations. However, estimates of abundance between stations were correlated, and estimates of species richness and assemblage structure were similar. Relative to removal sampling, a single upstream CPE pass adequately assessed fish species richness, abundance, and assemblage structure in small streams.