Long-term temperature changes can affect building heating and cooling demands and the corresponding electric sector capacity and capital investments. This study explores the effects of long-term temperature changes and variability on electric sector capacity in the U.S. by accounting for monthly day/night profiles of electricity use, associated electric sector investments, and power plant operations within a long-term, multisectoral model with state-level detail (GCAM-USA). The study shows that future peak temperature changes could drive increases in capital investments by 3-22% across the U.S. relative to a future scenario that assumes temperature has no impact on electricity demand. These temperature-induced capital investments are highly sensitive to socioeconomic assumptions, which underscores the importance of using a multisectoral approach.

The results of this study underline the need for electric sector capacity expansion planning and modeling to account for impacts of temperature changes on the peak electricity loads (rather than the mean, the focus of many prior efforts) since peak loads drive capacity and investment requirements. The study also underscores the need for such planning and modeling to account for broader socioeconomic drivers and the development and interactions of other sectors with the electric sector. Simultaneously accounting for these factors will help ensure a long-term electricity system that is reliable with sufficient capacity to meet demands at all times of the year.

This work sets the stage for future research on how a variety of factors, like electricity storage, electric vehicle penetration, and battery charging patterns, might impact the future evolution of the electric sector.

The improved version of GCAM-USA calculates electricity loads by service in the residential and commercial building sectors in 25 sub-annual (24 monthly day/night plus a “superpeak” segment) time-segment units, thus endogenously and dynamically capturing the impacts of future temperature changes on sub-annual load profiles. The model performs a merit-order based routine to dispatch capacity to meet energy loads, prioritizing the cheapest energy first. Simultaneously, it calculates the retirement of existing capacity and necessary capacity investments, where investments depend on superpeak demand, existing stock, and retirements.

 The study uses the improved model to examine the implications of long-term temperature change and spatial and temporal variability on electric sector capacity and capital investment requirements. Under middle-of-the road socioeconomic assumptions combined with future climate change consistent with 8.5 W/m2 forcing by 2100, mean temperature changes drive increases in national annual electricity demands by about 5%. However, peak temperature changes drive increases in installed capacity and capital investments by about 15%. This difference in temperature impacts suggests that planning for future electricity systems based only on annual impacts on electricity supplies and demands could grossly underestimate the economic implications of long-term temperature changes. These temperature-induced increases in electric capacity and investments vary spatially across the U.S. based on local conditions.