Influence of Streamflow and Predators on Habitat Choice by Trout

DOCTORATE DISSERTATION: I examined the effect of flow and presence of large brown trout on habitat choice by juvenile rainbow trout and brook charr. Experiments were conducted in a large stream enclosure with a pool-riffle-pool configuration. Large brown trout, when present, were confined to the upstream pool. Juvenile trout and charr tended to leave the riffle and occupy the pools when flow was low. They also were less likely to be found in the upstream pool when large brown trout were present; juvenile fish remaining in the upstream pool avoided the deep area with cover favored by the large brown trout especially at sunset and at night at high flow. Low flow apparently caused juvenile trout and charr to inhabit pools with large brown trout that they would otherwise avoid. Observations of marked individuals indicated that juvenile fish increased their preference for shallower water of intermediate velocity when flow changed from low to high, even after adjusting for different depth and velocity availabilities. This suggests that depth and velocity preferences were flow dependent. Furthermore, velocity choice of aggressive fish increased more than for non-aggressive individuals, and velocity choice of rainbow trout increased more than for brook charr. Feeding rates of juvenile rainbow trout and brook charr declined both when flow was low and when large brown trout were present, although aggressive fish had higher overall feeding rates than non-aggressive individuals. Feeding rates were likely affected by changes in food availability related to both flow level and changes in local fish density, as well as the aggressiveness of individual fish. This result indicates that microhabitats having water of the same velocity may not always be of the same value to drift-feeding juvenile salmonids. Overall, this study has shown that flow and predator effects on juvenile salmonid habitat choice are extremely complex. Both habitat use and feeding rate were strongly affected by flow, predator presence, diel period, and the aggressiveness level of individual fish. Interactions among these variables were common. Models used to recommend instream flows for regulated streams may not adequately evaluate stream fish habitat.