No abstract

In the last decade, the transition away from coal and to fossil gas and biomass in the U.S. has had a major influence on greenhouse gas emissions, especially from electricity generation. However, the effect of this transition on the public health burden of air pollution is not well understood. We use three reduced complexity models (RCMs) and emissions inventory data to reconstruct the changes in health impacts due to PM2.5 exposure from stationary fuel combustion sources in the U.S., from 2008 to 2017. In 2008, the health impacts of air pollution from stationary sources was largely driven by coal combustion. By 2017, the contribution of coal has dropped precipitously, and the health burden of stationary air pollution sources is shared among a mixture of source types and fuels—largely gas and biomass in buildings and industry, and the remaining coal-fired electricity generation. Nationwide, in 2017, health impacts of biomass and wood combustion are higher than combustion of coal and gas individually. Industrial boilers had the highest emissions and health impacts, followed by residential buildings, electricity, and then commercial buildings. All three RCMs indicate that biomass and wood are the leading sources of stationary source air pollution health impacts in 24 states, and that the total health impacts of gas surpass that of coal in 19 states and the District of Columbia. We develop a projection method using state-level energy consumption data for 2018 and show that these trends likely continued. The RCMs had high agreement for 2008 emissions, when sulfur dioxide emissions from coal-fired power plants were the predominant air pollution source. However there was substantial disagreement between the three RCMs on the 2017 health burden, likely due to pollutants less well-characterized by the RCMs having a higher proportionate share of total impacts.

Building electrification is essential to many full-economy decarbonization pathways. However, current decarbonization modeling in the United States (U.S.) does not incorporate seasonal fluctuations in building energy demand, seasonal fluctuations in electricity demand of electrified buildings, or the ramifications of this extra demand for electricity generation. Here, we examine historical energy data in the U.S. to evaluate current seasonal fluctuation in total energy demand and management of seasonal fluctuations. We then model additional electricity demand under different building electrification scenarios and the necessary increases in wind or solar PV to meet this demand. We found that U.S. monthly average total building energy consumption varies by a factor of 1.6×—lowest in May and highest in January. This is largely managed by fossil fuel systems with long-term storage capability. All of our building electrification scenarios resulted in substantial increases in winter electrical demand, enough to switch the grid from summer to winter peaking. Meeting this peak with renewables would require a 28× increase in January wind generation, or a 303× increase in January solar, with excess generation in other months. Highly efficient building electrification can shrink this winter peak—requiring 4.5× more generation from wind and 36× more from solar.