By 2100, Over 100 Million Americans Could Face Unhealthy Air Each Summer

Right now, roughly 14 million Americans live in areas where summer air quality regularly reaches levels considered unhealthy for sensitive groups. By the end of the century, that number could surpass 100 million, nearly one in three Americans, if greenhouse gas emissions go unchecked. A new study mapping how climate change will reshape air quality across the United States lays out a future where bad air days are no longer exceptions but seasonal norms for a vast share of the population.

Published in Environmental Science & Technology, the research comes from scientists at the University of Waterloo, Harvard University, North Carolina State University, and the University of California, Davis. Working with a climate and air quality modeling system developed at MIT, the team projected how rising temperatures will affect the two pollutants responsible for more than 99% of all air quality alerts in the United States: ground-level ozone and fine particulate matter, known as PM2.5.

At the heart of the problem is what scientists call the “climate penalty.” Even without any increase in direct industrial emissions, warming temperatures accelerate the chemical reactions that produce ground-level ozone, intensify wildfires that generate fine particle pollution, and extend the stagnant, hot stretches of weather when pollutants build up and linger. As the climate penalty grows, so does the number of days when the Air Quality Index, the federal government’s standard measure of outdoor air quality, tips into territory that poses real health risks.

How Researchers Modeled Future Air Quality Alerts

To project how air quality could shift over the century, scientists ran hundreds of simulations under three scenarios: no climate policy, warming capped at 2 degrees Celsius above preindustrial levels, and warming limited to 2.5 degrees Celsius. Running 150 simulations per time period let the team separate genuine climate-driven trends from normal year-to-year swings in weather.

100 Million Americans Could Face Unhealthy Air Quality by 2100

At the start of this century, about 5% of Americans lived in areas where average summer air quality reaches the “Unhealthy for Sensitive Groups” range, the point at which health officials recommend that children, older adults, and people with heart or lung conditions reduce outdoor activity. Under a no-policy scenario, that share could grow to include more than 100 million Americans by 2100. Parts of the Eastern United States and California could see up to two additional months’ worth of alert days spread across the summer.

Air quality alerts for sensitive groups could double over that span, and days when both ozone and PM2.5 spike simultaneously could roughly quadruple, climbing from around two such days per summer to about eight. Limiting global warming to 2 to 2.5 degrees Celsius could reduce additional alert days by roughly 30%, but even the most ambitious climate targets modeled here would not eliminate the problem.

Seniors Carry a Disproportionate Air Quality Health Burden

Not everyone in that 100 million faces the same risk. According to the study, adults 65 and older account for roughly two-thirds of air pollution-related deaths, despite making up just 12% of the population. As the population ages and the smog season extends, that concentration of harm among the oldest Americans is only likely to grow.

When seniors heed air quality alerts and stay indoors, the health benefits, measured in avoided illness and premature death, are more than 45 times greater per day on average than those received by adults between 18 and 35. As the researchers write, “This disproportionate impact requires targeted messaging and guidance, especially as climate-related risks rise.” Most older adults get air quality information through television rather than apps or online sources, a gap that could leave them less informed on the days it matters most.

For the average senior, the health value of staying indoors on an “unhealthy” air day outweighs the relatively modest daily cost of doing so. For younger adults, the calculus looks very different.

Nearly half of seniors report spending no time outside on a typical day to begin with, meaning that on a bad air day, even a small adjustment in routine carries real protective value. Children, people with asthma, and those with cardiovascular conditions face similarly elevated risks, making the rise in “Unhealthy for Sensitive Groups” days a direct threat to tens of millions of Americans who may not fully realize how exposed they are.

Staying Indoors Helps, But It Is Not Enough

Individual responses to air quality alerts, primarily staying indoors when warnings are issued, do reduce exposure and lower associated health risks. But even if every at-risk American followed every alert perfectly, the health benefits would offset only about 15% of the added risk under a no-policy scenario. Under stricter climate policies, that fraction rises as the overall number of dangerous days is brought down. Personal behavior change alone is no substitute for the emissions cuts needed to address the underlying cause.

One structural quirk in the current alert system compounds the problem. Because the Air Quality Index reports only the single worst pollutant on a given day, it doesn’t fully capture the combined risk when both ozone and PM2.5 are elevated simultaneously. Moving indoors cuts exposure to both at once, so the real health benefit of following an alert is likely greater than what the current system conveys.

Getting to a future where those 100 million Americans are not spending their summers breathing unhealthy air requires more than better alert systems and individual precautions. Substantial cuts to the emissions driving the climate penalty are the only thing with enough force to change that trajectory.

---

Disclaimer: This article is based on a peer-reviewed modeling study. Projections reflect modeled scenarios under specific climate and emissions assumptions and should not be interpreted as precise predictions. Results are subject to uncertainty in climate modeling, health impact functions, and economic valuation. Findings may differ from other studies using different methodologies or assumptions.

---