Seasonal Movement and Habitat Use of Wild Brook Trout (Salvelinus fontinalis)

Open Access
Mollenhauer, Robert
Graduate Program:
Wildlife and Fisheries Science
Master of Science
Document Type:
Master Thesis
Date of Defense:
October 25, 2011
Committee Members:
  • Tyler Wagner, Thesis Advisor
  • discrete choice models
  • brook trout
  • radio telemetry
  • thalweg profiles
  • GAMMs
Eastern brook trout Salvelinus fontinalis populations face a myriad of threats throughout the species’ native range in the eastern United States. Anthropogenic alterations to the landscape, in particular, have the potential to degrade and fragment brook trout habitat. Therefore, understanding wild brook trout movement patterns and habitat requirements is essential for conserving existing populations, and for restoring habitats that no longer support self-sustaining populations. Radio biotelemetry is one approach commonly used to quantify fish movements and habitat use. Until recently, however, telemetry research targeting small stream-dwelling salmonids has been limited by the relatively large size of transmitters and short battery life (i.e., only larger fish were tagged and tracked over relatively short time frames). Recent advances in radio biotelemetry allows for smaller fishes to be tracked for longer time periods (i.e., for months rather than weeks), and thus provides the opportunity to gain additional insight into the ecology of stream-dwelling salmonids. To address uncertainties related to wild eastern brook trout movements and habitat use, I radio-tagged and tracked 36 fish in a headwater stream system in central Pennsylvania during the fall and early winter of 2010 – 2011. I used generalized additive mixed models to identify relationships between brook trout movement and covariates. I also surveyed thalweg profiles of the study area to quantify available stream habitat and used discrete choice models with random effects to evaluate habitat use, a novel modeling approach for fish habitat use studies. There was a large amount of among-fish variability in movement patterns; however, most of the activity was associated with the onset of the spawning season (late September – early October), including several long-distance (> 1 km) movements. Some of the among-fish variability could be explained by the size of individual fish, because the larger main-stem brook trout were more active, on average. In addition to a positive relationship between fish size and movement, modeling results also indicated that movement was positively correlated with higher stream flow. However, these effects of fish size and stream flow stage varied seasonally. Although movement during the spawning season was variable, little movement was observed across all brook trout once stream temperatures dropped below 7 °C in early December. Lastly, the larger brook trout that were tagged in the main-stem utilized lower regions of only one of the tributaries during spawning season, despite other tributaries being nearby and accessible. These results suggest that not all tributaries provided equal-quality habitat and individual tributaries may contribute disproportionally to reproduction and sustainability of brook trout in this system. Although there was a general preference for pool versus non-pool habitats, there was a large amount of among-fish heterogeneity in habitat use. Specifically, there was variability in preference for intermediate (0.26 – 0.44 m deep) and deep (0.44 – 1.06 m deep) residual pools, while all fish showed a similar selection for shallow (0.10 – 0.26 m) residual pools compared to non-pool habitats. In addition, habitat use was nonlinear, both seasonally and in relation to the length of specific habitat types. Seasonally, there was a selection for shallow residual pools during spawning season, followed by a selection for deep residual pools as winter approached. This nonlinear seasonal selection likely reflected brook trout spawning and overwintering behavior. Brook trout were also found to show a threshold-effect for pool use with respect to pool length: habitat use for pools tended to increase as the average length of pools increased up to a point and then use declined rapidly for very long pool habitats. The heterogeneity and nonlinear dynamics of movement and habitat use of the wild brook trout observed in this study underscores two important points: (1) linear models may not provide an accurate description of movement and habitat use, which can have implications for management, and (2) the importance of maintaining stream connectivity when managing self-sustaining populations. These results also demonstrate the importance of diversity, in terms of the physical stream environment, when protecting or restoring instream habitat. A way of preserving stream connectivity and providing quality instream habitat is to use a watershed approach when managing resident brook trout populations.