Advances in Understanding Landscape Influences on Freshwater Habitats and Biological Assemblages

<i>Abstract.</i>—We document a simple electrofishing-only monitoring program for assessing fish assemblages across large spatial extents. First, we demonstrate the justification for using only electrofishing for the monitoring. Second, we demonstrate the usefulness of having a well-designed surveillance-monitoring program in place to demonstrate the effect of landscape disturbances. Implementing electrofishing alone means that multiple sites can be sampled in a single day and there is no need to return to clear nets or traps within a sampling site. Whereas electrofishing alone does not return full species lists within sampled sites, we demonstrate that when data are aggregated up to the watershed or catchment extent, more than 90% of species are included. Analyses that do not require a census of species, such as bioassessment of river health can be readily carried out using electrofishing data. The Murray–Darling basin, Australia, was sampled with the recommended large-extent electrofishing program between 2004 and 2012, a period that saw the region subjected to large-scale variations in river flow levels spatially and temporally. We fit generalized additive models to the electrofishing data in conjunction with river flow data to document large-extent relationships between fish species occurrence and relative flow levels for the previous 3 d, 3 months, or 3 years. We found that several small-bodied species, Eastern Mosquitofish <i>Gambusia holbrooki</i>, Flathead Gudgeon <i>Philypnodon grandiceps</i>, and Australian Smelt <i>Retropinna semoni</i>, were more likely to be collected when conditions were drier in the past 3 d to 3 months, whereas common medium and large-bodied species were less likely to be collected when flow was lower over the previous 3 months to 3 years.