Investigating hydrologic alteration as a mechanism of fish assemblage shifts in urbanizing streams

Stream biota in urban and suburban settings are thought to be impaired by altered hydrology; however, it is unknown what aspects of the hydrograph alter fish assemblage structure and which fishes are most vulnerable to hydrologic alterations in small streams. We quantified hydrologic variables and fish assemblages in 30 small streams and their subcatchments (area 8–20 km²) in the Etowah River Catchment (Georgia, USA). We stratified streams and their subcatchments into 3 landcover categories based on imperviousness (<10%, 10–20%, >20% of subcatchment), and then estimated the degree of hydrologic alteration based on synoptic measurements of baseflow yield. We derived hydrologic variables from stage gauges at each study site for 1 y (January 2003–2004). Increased imperviousness was positively correlated with the frequency of storm events and rates of the rising and falling limb of the hydrograph (i.e., storm “flashiness”) during most seasons. Increased duration of low flows associated with imperviousness only occurred during the autumn low-flow period, and this measure corresponded with increased richness of lentic tolerant species. Altered storm flows in summer and autumn were related to decreased richness of endemic, cosmopolitan, and sensitive fish species, and decreased abundance of lentic tolerant species. Species predicted to be sensitive to urbanization, based on specific life-history or habitat requirements, also were related to stormflow variables and % fine bed sediment in riffles. Overall, hydrologic variables explained 22 to 66% of the variation in fish assemblage richness and abundance. Linkages between hydrologic alteration and fish assemblages were potentially complicated by contrasting effects of elevated flows on sediment delivery and scour, and mediating effects of high stream gradient on sediment delivery from elevated flows. However, stormwater management practices promoting natural hydrologic regimes are likely to reduce the impacts of catchment imperviousness on stream fish assemblages.