Estimating Sampling Effort for Biomonitoring of Nearshore Fish Communities in Small Central Minnesota Lakes

Abstract

AbstractI estimated the sampling effort required to sufficiently describe the species richness and composition of fish inhabiting the nearshore (wadeable) zone of lakes and compared 10 species richness estimators in 56 similarly sized (small) central Minnesota lakes. A systematic sampling design was employed wherein 10 stations were equally spaced around each lake and the first station was selected randomly. Sampling methods consisted of a combination of backpack electrofishing and shoreline seining. Using sample‐based rarefaction curves, eight sampling stations produced at least 90% of the sampled species in 96% of the study lakes. Less sampling effort (four to six stations) was needed to describe the composition of the nearshore fish assemblage, as measured by similarity of species composition or species relative abundances in cumulative samples. The effort needed to sufficiently describe nearshore species richness was not related to observed species richness but was negatively correlated with species evenness among sampling stations. Estimates of total nearshore species richness were made using an asymptotic accumulation function, a log‐linear model, two abundance‐based nonparametric estimators, and six incidence‐based nonparametric estimators. All estimates of total species richness were correlated with observed species richness. Estimated total species richness exceeded observed species richness; however, this difference was smallest when species evenness was high. Nonparametric methods produced unrealistically high richness estimates if many species were encountered infrequently. Observed richness varied among lakes and was positively related to measures of lake integrity. Observed richness, estimated total richness, and sufficient effort levels were similar for lakes sampled twice, suggesting that both the sampling method and estimated levels of sufficient effort were robust. The results of this study suggest that more intensive nearshore sampling with seines and backpack electrofishing will be of limited benefit unless the detection of rare species is the major goal of sampling; alternative sampling gears may be needed to detect new species.