High river temperatures, or ‘thermal extremes’, can cause fish mortality and thermoelectric powerplant derating. Under climate change, projected higher air temperature and stronger surface energy fluxes will lead to increased water temperatures, exacerbating thermal extremes. However, cold hypolimnetic releases from thermally stratified reservoirs can depress tailwater temperatures and therefore alleviate thermal extremes. Thermal extremes are more harmful when they coincide with low flows, which we refer to as ‘hydrologic hot-dry events’. To assess multi-sectoral impacts of climate change over large regions, we evaluate thermal events according to three impact attributes: duration (D), intensity (I), and severity (S). We apply an established model framework to simulate streamflow and stream temperature over the southeastern US regulated river system. We quantify climate change impacts (by the 2080s under RCP8.5) by comparing historical and future periods and quantify regulation impacts by comparing unregulated and regulated model setups. We find that climate change will exacerbate thermal extremes (all three metrics) in both unregulated and regulated model setups, albeit less in the regulated setup. Thermal mitigation from reservoir regulation will be stronger under climate change, decreasing the three metrics compared to the unregulated case. Even so, thermal extremes in the regulated setup will still be more severe under climate change, and only 12.2%, 19.7%, and 26.0% of D, I, and S can be mitigated by reservoirs. Despite stronger reservoir stratification, the number of regulated river segments that experience simultaneous high temperature and low flow events (hydrologic hot-dry events) will increase by 21.4% by the 2080s under RCP8.5. These events will have a median annual duration of 10.3 day/year, over 10 times the historical value.
Panels (b), (c), and (d) respectively show the distribution of mean annual duration, intensity, and severity of thermal extreme events for different river sizes separated according to the horizontal axis at the bottom of the figure. Left (orange and red) and right (blue and navy) sides of the vertical distributions denote unregulated and regulated model setups, respectively. Orange and blue are for the historical period; red and navy are for the 2080s under RCP8.5. Thermal extremes will increase in the regulated system under climate change. The climate-induced increases are only slightly lower in the regulated model setup than those in the unregulated model setup. Furthermore, by the 2080s under RCP8.5, only 12.2%, 19.7%, and 26.0% of duration, intensity, and severity, respectively, that would have occurred in the unregulated case can be mitigated by reservoir regulation, assuming all releases come from the hypolimnion.
For detailed results, please refer to Cheng et al. (2020).