The downstream air pollution impacts of the transition from coal to natural gas in the United States
By studying what happens to local pollution when old electric power units are shut down (or new ones are turned on) we can better understand pollution's impacts on people, plants, and regional climate.
[Image: Paradise Fossil Plant in Muhlenberg County, KY. Credit: Tennessee Valley Authority (TVA) and WKU FM.]
Background: Over the past two decades, technological and economic factors have dramatically reshaped the United States’ electric power sector. Fracking technology and the ability to access new natural gas resources has lowered the price of natural gas and made gas-fired power plants economically competitive with coal-fired plants. As a result, in recent years an increasing fraction of the nation’s electric power has been generated using natural gas as a feedstock. This shift has been a hot topic for many different communities, since it raises issues of:
- Energy security: A new accessible fossil resource within national borders changes both domestic and international political considerations.
- Indigenous and landholder rights: The Dakota Access Pipeline (DAPL) and wide-ranging variance in mineral rights contracts for people living over shale gas deposits have shown how local considerations and voices may be either elevated or suppressed.
- Socio-economic inequality: Overall, coal-rich regions have suffered and natural gas-rich regions have benefitted from this transition, with far-reaching consequences for local populations.
- Environmental impacts: Scientific studies have documented water and seismic risks associated with fracking, as well as methane leakage from well heads and transport infrastructure.
- Climate change: Natural gas-fired units emit less carbon dioxide per power produced than coal-fired units.
These important issues are part of a vibrant series of conversations about the future of the U.S. energy sector, but I was interested in another angle: in addition to emitting carbon dioxide, fossil-fuel burning plants emit a host of other compounds that evolve in the atmosphere and stay closer to their emissions source. The resulting pollutant mix (including primary aerosols, secondary aerosols, and ozone) is known to impact human health, crop growth, and regional climate. These impacts would ideally inform energy infrastructure decisions; unfortunately, however, it is often difficult to study the relationship between emissions changes and these kinds of outcomes, because: (a) emissions and ambient pollution data can be sparse, both geographically and in terms of chemical detail; and (b) accounting for other factors that might confound a statistical relationship between emissions and (e.g.) crop yields can be notoriously tricky.
This paper: In this project (published this week in Nature Sustainability), I leveraged the fact that this major energy sector transition was actually comprised of hundreds of local observable changes: Roughly a fifth of electric power infrastructure across the continental U.S. was geographically redistributed over the study period (2005-2016), as coal-fired power plant units were taken offline and – mostly in different locations – hundreds of new gas-fired units came online. Because units were fully shut down (or conversely, turned on where there had been none prior), this coal-to-natural gas transition provides a unique opportunity for scientific discovery. It is a study environment that minimizes the two problems listed above: (a) emissions go from high to zero (or vice versa) so it is possible to cleanly understand the local impacts of power plant units in operation, and (b) because this takes place at many sites and many different moments over time, the threat of confounding factors becomes much less worrisome.
I looked at local air pollution at earth's surface and in the full thickness of the atmosphere, as well as human mortality and crop yields, before and after coal-fired units were decommissioned, and compared those changes to locations that did not have coal-unit shutdowns. This gives the clearest estimate to date of the true impact of coal plant-related pollution on local populations. I find that, on average across all of these locations, when a coal-fired unit shuts down, local pollution (including particulate matter) levels drop, mortality rates drop, and crop yields of major staple crops rise. Aggregating these numbers over the study period suggests that the impacts have been large: tens of thousands of lives, and hundreds of millions of bushels of crops, were saved by shutting down coal-fired plants since 2005.
New natural gas-fired plants do result in higher local pollution levels (and changes in atmospheric column burdens of pollution), but the spatial pattern and chemical composition is (perhaps unsurprisingly) different from coal. Although these new units are not statistically associated with excess deaths or crop impacts in this analysis, they will require more study in the future to understand their full impacts.
Finally, the impact exerted by coal-related pollution on climate that is clearly visible in this study. The clearer atmosphere from coal plant decommissioning has changed the balance between incoming sunlight and outgoing radiation (instantaneous radiative forcing at the top of the atmosphere). Locally, aerosol pollution “masks” some of the longer-term greenhouse gas-driven warming by blocking a fraction of incoming sunlight and cooling the surface (like wearing a hat on a hot sunny day to keep your face cool). As the air clears, there have been (and will be) tremendous benefits to human and crop health, but we will also start to feel the full warming caused by greenhouse gases.
What stood out to me: The aggregate numbers are big, but are largely in agreement with other estimates made using different methods. The fact that coal-fired plants (or natural gas-fired plants) create local pollution, which has deleterious effects downstream, is not in and of itself new or surprising. But it is really interesting to see who has benefitted from this large scale infrastructure shift and, conversely, to think about who continues to be negatively impacted by coal-fired plants still in operation. The geographic distribution of this set of impacts may be something that sets these findings apart from other studies.
This is important because energy infrastructure decisions are often made based on a political and economic benefit-cost analysis, where economic, political, and some welfare benefits are compared to capital and operational costs. Ideally, the full set of benefits of any technology would be weighed against the full set of costs associated with it, including climate, environment, and pollution impacts. At present, these are not usually considered in the calculus, in part because they are difficult to pin down. This study provides some hard numbers on where and how the power plant-related pollution has affected communities around the country. This kind of information can be used by communities in their decision-making, and may change the balance in favor of cleaner energy sources.