Optimizing a Standard Sampling Program to Evaluate Fish Community Structure for Non-wadeable Rivers in Alabama

Abstract

Alabama’s non-wadeable rivers support exceptionally high biodiversity of fishes but no formal sampling program has been developed to monitor the fish community in these systems. Recent developments in Alabama have caused increased interest by Alabama Department of Conservation and Natural Resources biologists to develop such a sampling program, but studies determining adequate sampling effort are lacking. This study was developed to compare two different boat-based electrofishing methods (continuous bank-line and point sampling) to sample the fish community and determine the effort necessary to accurately represent the fish communities present. Furthermore, a comparison of day and night electrofishing was done in two rivers with the continuous bank-line method. Four rivers of various sizes and locations in Alabama were sampled along two 100-mean-stream-width transects. Because habitat complexity can affect sampling effort, substrate was mapped using side-scan sonar within select reaches of each river, and low and high complexity transects were identified for electrofishing sampling. Sampling was done in summer and fall of 2015 and 2016. This study had three objectives: 1) Determine the optimal amount of sampling required to collect 50% and 75% of the species expected to be encountered, with 95% confidence, using bank-line and point sampling in each river, 2) Assess the effects of season and habitat complexity on sampling optimization for bank-line and point sampling in each river, and 3) Determine the effect of day/night sampling on sample optimization for bank-line sampling in two rivers. At 95% confidence, the bank-line method caught 50% of species estimated within 30 to 50 mean-stream-widths, and 75% of species estimated within 90 to 130 mean-stream-widths during fall. Point sampling was the least effective method and did not capture enough species to determine effort necessary to meet sampling objective. Fall was generally a more efficient season than summer to meet sampling objectives. Species richness and sampling effort was typically not affected by habitat complexity category. Night shocking was more efficient than day in the larger river, but similar in the smaller river.