Fish Assemblages and Substrates in the Middle Wabash River, USA
We collected fishes at 28 sites of the middle Wabash River, Indiana, using a boat electrofisher and at ten additional sites using a 10-m beach seine. We used Canonical Correspondence Analysis to test for fish assemblage variation that was explained by variation in water depth and substrate frequency. The multivariate analyses resulted in a longitudinal gradient in water depth and frequency of occurrence of substrate size categories that explained a large component of variation among fish assemblages for both sampling regimes. Our results illustrate that substrate composition and water depth variation are significant environmental predictors of fish assemblage composition for a large river such as the Wabash River.Abstract

Collection locations on the middle Wabash River, Indiana. Open circles represent sites that were electrofished and closed triangles represent sites that were seined in 2005.

The first and second axes of CCA triplots for 2005 collections by electrofisher. F refers to substrate frequency and D refers to depth of substrate. Open circles on the upper plot (A) are July collections and closed circles are June collections. The middle plot (B) shows species, and environmental vectors are represented by arrows on the bottom plot (C). The arrow length represents strength of environmental vectors, and the square in plot C represents changes with July collections. Abbreviations are from Table 1. Note that scales of axes vary among plots.

The second and third axes of CCA triplots for 2005 collections by seine. F refers to substrate frequency and D refers to depth of substrate. Open circles on the upper plot (A) are July collections and closed circles are June collections. The middle plot (B) shows species, and environmental vectors are represented by arrows on the bottom plot (C). The arrow length represents strength of environmental vectors, and the square in plot C represents changes with July collections. Abbreviations are from Table 1. Note that the scales axes vary among plots.
Contributor Notes
Associate Editor: J. W. Snodgrass.