Changes in fish assemblage structure with variability of flow in two different channel types

Despite the fact that flow and channel morphology are two critical factors that need to be taken into account when considering the conservation or restoration of river environments, there are no reports on the response of fish assemblages to changes involving both variables in combination. This study examined the response of fish assemblages to artificially induced changes in flow discharge in an experimental stream that had simple (a glide reach) and complex (pool–riffle reach) morphologies. When the flow of the glide reach was increased, the fish assemblage structure changed from one dominated by the demersal species Misgurnus anguillicaudatus and Cobitis sp. to one dominated by the water-column species Zacco platypus and Gnathopogon elongatus. In the pool–riffle reach, the fish assemblage structure changed from one dominated by Cobitis sp. and Z. platypus to one dominated by Plecoglossus altivelis altivelis. When the flow discharge was low, the fish assemblages of both reach types resembled each other in being dominated largely by demersal species, but when flow was increased these similarities faded and distinct assemblages emerged. In the glide reach, increasing the flow volume caused a linear increase in both water depth and velocity and a gradual increase in the diversity of water-column species, their relative abundance, and size. The number of Z. platypus increased fivefold whereas the number of M. anguillicaudatus decreased to less than a quarter of their original number. In the pool–riffle reach, the number of P. altivelis altivelis grew conspicuously although increased flow produced no clear increase in depth or velocity. Calculation of the availability of physical habitat environments suitable for Z. platypus and P. altivelis altivelis in each reach type indicated that preferable habitat for both species was more available in the glide reach. The lower abundance of P. altivelis altivelis in the glide reach was attributed to the relatively low availability of algal food resource due to the sand-predominated substrate whereas pebbles predominated in the pool–riffle reach providing good conditions for algal growth. Our findings suggest the need for a framework for considering environmental flow that takes into account variables such as channel morphology and food resource conditions.