Inter-replicate variance and statistical power of electrofishing data from low gradient streams

摘要

Electrofishing data from streams on the coastal plain of South Carolina were used to assess the relationship between inter-replicate variance and the mean for catch per unit effort expressed as number of fish per 100 m(sup 2) of stream surface area. Variance and mean were strongly and positively related. This relationship was unaffected by type and quantity of instream structure and was consistent among schooling and nonschooling species. Computations of statistical power indicated that large numbers of samples (>25) were required to detect small (<20%) differences among means and that the required number of samples was greater when catch per unit effort was low. Inter-replicate variance increased as the reach length of individual replicates decreased, with the greatest increases occurring at replicate reach lengths under 60 m. Inter-replicate variance can be decreased by increasing sampling effort (i.e., multiple passes) and by employing replicate reaches that are long. However, the most cost-effective approach may be to employ single passes and replicates of moderate size.