Using Multiple Approaches to Evaluate the Effects of Flow Regulation on Fishes in the Tallapoosa River, Alabama.
Type of DegreeMaster's Thesis
School of Fisheries, Aquaculture, and Aquatic Sciences
Restriction TypeAuburn University Users
MetadataShow full item record
Dams alter many aspects of riverine environments and generally have broad effects both upstream and downstream. Despite the variety of potential benefits of dams to humans (recreation, flood control, navigation, etc.), they can also negatively affect riverine ecosystems. In particular, hydropeaking dams affect downstream fish habitats by increasing water velocity and altering water temperature regimes. Since 1983, the discharge and water temperatures in the Tallapoosa River, Alabama, have been regulated by Harris Dam. In 2005, a group of stakeholders successfully petitioned to implement an adaptive management plan (the Green Plan) on the operation of Harris Dam which set limits on the timing, schedule, and duration of water releases. To assess the effects of Harris Dam operating under the Green Plan, I collected fish from four sites on the Tallapoosa River, three of which were downstream of and regulated by Harris Dam, and one site upstream and unregulated by Harris Dam. I used multiple approaches to quantify patterns in fish assemblage structure, and diet composition and movement patterns of four recreationally important species - Channel Catfish Ictalurus punctatus, Redbreast Sunfish Lepomis auritus, Alabama Bass Micropterus henshalli, and Tallapoosa Bass Micropterus tallapoosae. First, I quantified the fish assemblage structure using Shannon’s H, nonmetric multidimensional scaling (NMDS), a multiresponse permutation procedure (MRPP), and indicator species analysis to determine if the distribution of species varied among sites along a gradient downstream of Harris Dam. Shannon’s H varied little across my sites, whereas NMDS and MRPP revealed significant assemblage differences among sites. The tailrace fish assemblage was distinct from the other downstream sites and was characterized by higher number of fluvial specialist species. This suggests that the tailrace assemblage may favor species that are able to persist through higher flows at the expense of other native species that are not. Additionally, target species’ diets were quantified, revealing spatially variable diet compositions. Diets (proportion by weight) from fish collected in the tailrace were distinct from those from the other sites in that they contained some prey types (e.g., amphipods, isopods) not found in diets at the other sites. Trace element ratios in otolith edges (i.e., recently incorporated material) and water samples were weakly positively related. In addition, water trace element ratios did not vary seasonally for any element except barium. Elemental signatures in both water and otoliths varied across sites with the largest difference occurring between the upstream unregulated site and the three downstream regulated locations. Differences between the three downstream sites were less apparent, and Sr:Ca ratios were identified as the most informative of the four elemental ratios analyzed (Sr:Ca, Ba:Ca, Mn:Ca, Mg:Ca). Using linear discriminant analysis, otoliths were correctly assigned to capture region with overall accuracy of 39.5-82.7% depending on the otolith region being considered (core, edge, or entire otolith transect). Variation in classification accuracy among otolith segments indicated potential ontogenetic shifts in site fidelity, although interpretation was limited by low variation across downstream site element ratios. Strontium ratios across the entire otolith ablation transect suggested three predominant movement patterns: 1) individuals spent their entire life at the capture location, 2) individuals recruited to the capture location from a different river section, and 3) individuals moved away from, then returned to, the capture location. Distributions of these three patterns indicated that fish in the tailrace recruited from a limited area whereas fish at downstream sites recruited from, and continued to use, a broader area of the river. To further quantify the effects of Harris Dam on fish movement, combined acoustic and radio tags were surgically implanted into (13) Alabama Bass and (3) Tallapoosa Bass individuals. An acoustic array of 10 stationary receivers was deployed in the area immediately downstream of the Harris Dam tailrace, and fish were also manually tracked with a radio receiver. Telemetry data revealed that longitudinal movement of black basses was minimal in response to the Harris Dam operation, with a maximum net longitudinal movement over the course of the study being 6.3 km. Overall, considering fish assemblages, diets, and trace element analyses, my results from the tailrace consistently differed from the other downstream sites. These results suggest that although the Green Plan may have mitigated some of the initial effects of peaking hydropower flows from Harris Dam, the current operation of the dam continues to affect downstream fishes.