River temperature is projected to increase in the southeastern United States (SEUS) due to climate change, exacerbating the invasion of warm-water species and reducing suitable habitats for cold- and cool-water species. However, the response of river thermal regimes to climate change is also influenced by human activities, especially dam construction and operation. Large dams impound deep reservoirs, expand water surface area and prolong water residence time, modifying the interaction of surface meteorology with river systems. During warm seasons, surface energy fluxes can only heat the top layer (epilimnion) in deep reservoirs with bottom layer (hypolimnion) remaining cold. This vertical temperature gradient is called thermal stratification. Cold hypolimnetic releases from stratified reservoirs changes downstream thermal regimes that can expel indigenous warm-water species yet provide an ideal habitat for introduced cold-water species. For example, multiple species of trout (Family: Salmonidae) have been introduced to tailwaters downstream of multiple dams operated by the Tennessee Valley Authority, which has become a popular and lucrative recreational fishing location in the SEUS. Previous research has shown that reservoir thermal stratification will be retained under climate change, but stronger surface energy fluxes warm downstream river temperature, suggesting there will be a future decline in cold-water species habitat and a corresponding increase in local warm-water species habitat. In this study, we used a physically-based modeling method to simulate river temperatures, explicitly considering the impact of thermal stratification. The SEUS has a highly regulated river system and diverse freshwater fish species. We mapped the suitable habitats for selected cold-water and warm-water fish species by comparing the simulated river temperature against their physiological constraints. Model experiments were designed to quantify the impacts of dam operation by simulating river temperature for both regulated and unregulated scenarios. Potential ecological consequences under climate change were analyzed through projected changes in river thermal regimes, e.g., shrinking habitats for cold-water species and restoring local warm-water species.

Over 270 major dams have been constructed in the Southeastern United States (SEUS) during the past century, changing natural flow patterns and affecting stream temperatures. Projected increases in air temperature combined with changes in precipitation may result in water scarcity and affect maximum stream temperatures during the summer for some regions in the SEUS. Currently existing reservoirs mitigate water shortages during drought by releasing more water but reducing residence time, the ratio of reservoir volume to inflow. Regulating stream temperature in the summer can be done by either increasing residence time or releasing more water. In this study, we investigate the extent to which the current reservoir infrastructure can be used to mitigate the impacts of climate change under current reservoir regulations as well as the range of operating rules that could minimize climate change impacts on both streamflow and river temperature. We use the Variable Infiltration Capacity (VIC) hydrological model to simulate runoff, which is then used as input to a large-scale river routing-reservoir model (MOSART-WM) to simulate reservoir operations and produce regulated streamflow. VIC and MOSART-WM outputs are then used as input to a stream temperature model that accounts for thermal stratification in reservoirs (RBM-res). Climate change projections are based on two representative concentration pathways (RCPs) and multiple global climate models from the Coupled Model Intercomparison Project Phase 5 (CMIP5). We compare modeled changes with those from a model implementation that does not include any reservoirs and which therefore lacks any flow regulation (VIC->MOSART-RBM) to evaluate the resilience of current reservoir infrastructures. We also evaluate different reservoir operating rules (residence time versus low flow mitigation) to investigate the extent to which the current reservoir system can be used to mitigate the impacts of climate changes on both streamflow and stream temperature.