In 2020, renewables became the second-largest source of electricity generation in the United States after natural gas (US EIA, 2021). In recent years, wind energy generation has overtaken hydropower as the dominant source of renewable generation in the United States, but hydropower continues to offer advantages, in particular large-scale storage, that makes it particularly valuable as a complement to other weather-driven renewables. This storage, in the form of reservoirs, is rarely managed exclusively to optimize hydropower generation. Instead, reservoirs are operated for flood control, ecosystem services, irrigation, water supply, navigation, and recreation as well as hydropower. Managing these competing demands in a changing climate with existing infrastructure creates difficult challenges, because all these demands are themselves subject to change as is the electricity demand itself. Yet many climate change impact studies continue to treat rivers as entirely natural systems and water resources infrastructure is ignored or treated as an afterthought. In this presentation, we will discuss recent climate change impact studies in both the northwestern and southeastern United States in which we quantified the effects of regulation on discharge and other variables. We will make the case that to develop new strategies for mitigating and adapting to climate change, it is paramount to account for humans as active agents in the hydrologic cycle. The first study focuses on the Columbia River Basin in the Pacific Northwest, the main hydropower producing region in the United States, and examines the effect of accounting for regulation on changes in high and low flow extremes. The second study focuses on the southeastern United States and evaluates the effects of regulation on estimated changes in flow, stream temperature, and habitat suitability. US EIA, 2021: Monthly Energy Review, July 2021. www.eia.gov/mer [Last accessed on 8/3/2021].

Alaska and the Yukon are a challenging area to develop observationally based spatial estimates of meteorology. Complex topography, frozen precipitation undercatch, and extremely sparse observations all limit our capability to accurately estimate historical conditions. In this environment it is useful to develop probabilistic estimates of precipitation and temperature that explicitly incorporate spatiotemporally varying uncertainty and bias corrections. In this paper we exploit recently-developed ensemble Climatologically Aided Interpolation (eCAI) systems to produce daily historical observations of precipitation and temperature across Alaska and the Yukon territory at a 2 km grid spacing for the time period 1980-2013. We extend the previous eCAI method to include an ensemble correction methodology to address precipitation gauge undercatch and wetting loss, which is of high importance for this region. Leave-one-out cross-validation shows our ensemble has little bias in daily precipitation and mean temperature at the station locations, with an overestimate in the daily standard deviation of precipitation. The ensemble has skillful reliability compared to climatology and significant discrimination of events across different precipitation thresholds. Comparing the ensemble mean climatology of precipitation and temperature to PRISM and Daymet v3 show large inter-product differences, particularly in precipitation across the complex terrain of SE and northern Alaska. Finally, long-term mean loss adjusted precipitation is up to 36% greater than the unadjusted estimate in windy areas that receive a large fraction of frozen precipitation.