Fish condition and community structure in an agricultural landscape

The potential impacts of suspended sediment from potato agriculture were examined in small New Brunswick streams using the severity of ill effects (SEV) model. Proposed by Newcombe and Jensen (1996), the SEV is derived from exposure duration and concentration of suspended sediment. In late August 2007, 100 m of eight streams were electrofished for brook trout ( Salvelinus fontinalis) and slimy sculpin (Cottus cognatus ) to test the SEV model. Total suspended sediments (TSS) measured hourly from May to August at each site were used to calculate the SEV using two different methods. The first method used hourly duration and the second method used percentiles to distinguished peaks and troughs in the natural variability of the TSS data. Regression analysis suggested poor relationships between the SEV index and measures of fish of condition and density for either method of calculation. To assess the potential impacts of landuse on the fish communities, seven additional streams were electrofished and environmental variables describing the landscape and local habitat were measured for the stream catchments in 2006 and 2007. Multivariate analyses showed that maximum summer temperature was the best predictor of fish density and best explained differences in fish community composition for both years. My results indicate the complexity of interpreting stressor impacts and challenges to develop simple indices for stream fishes in agricultural regions.