A surfer cools off during a heat wave in southern California in October 2017.Mike Blake / Reuters

For almost two centuries now, scientists have noticed a place’s suicide rate bears troubling links to the changing of the seasons and the friendliness of its climate.

In 1881, the Italian physician Enrico Morselli noted that suicide rates peak in the summer, deeming the effect “too great for it to be attributed to chance of the human will.” Two decades later, the French sociologist Emile Durkheim noticed the same effect—though he also found the suicide rate was higher in Scandinavian countries.

Even today, CDC data confirms that suicides peak in the United States in the early summer.

Now, scientists have identified one more way that climate shapes suicide—and, worryingly, they have projected that it will only become more pronounced as suicide rates rise in a rapidly warming world.

Unusually hot days cause the suicide rate to rise, according to a study published Monday in Nature Climate Change. If a month is 1 degree Celsius warmer than normal, then its suicide rate will increase by 0.7 percent in the United States and 2.1 percent in Mexico.

“It’s sort of a brutal finding,” says Marshall Burke, a professor of earth science at Stanford University and one of the authors of the paper.

The finding has anxious implications for a world whose climate is rapidly changing. The authors project that roughly 14,000 people—and as many as 26,000—could die by suicide in the United States by 2050 if humanity does not reduce its emissions of greenhouse-gas pollution.

This is roughly the same change in suicide rates as are driven by “the estimated impact of economic recessions, suicide-prevention programs, or gun-restriction laws.”

It also concludes that humans can do little about this suicide-climate link beyond developing better medical care to address suicide specifically. The normal ways that people adapt to high temperatures generally—by installing air conditioners, for instance—do not seem to affect the suicide rate.

Suicide is the second most-common cause of death among Americans between 10 and 34 years of age. It is also one of the few leading causes of death in the United States where the age-adjusted mortality rate is not falling. In other words, more people are dying by suicide than used to.

“I found the results compelling and the methodology strong,” says Gregory Tasian, a professor of epidemiology and biostatistics at the University of Pennsylvania who was not related to the study.

“It fits very well into the overall narrative about the effect of climate change and high temperatures on human health—and it incorporates another aspect of human health which is often neglected, which is mental health,” he told me.

The new paper provides unusually robust evidence that high temperatures will significantly aggravate the problem of suicide.

First, the new paper makes a causative claim: Unusually high temperatures really do seem to cause suicide rates to increase. The authors have controlled for every other major variable that might affect suicide rates. The effect doesn’t seem to change if the victims are men or women, if the month is January or June, if they live in an American city or the Mexican countryside. Nor does it vary in areas where guns are plentiful.

Second, it finds that the effect shows up everywhere: in Saint Paul and San Antonio, in Ohio and Oaxaca. “We see the exact same effect size in hot places versus cold places,” Burke told me.

Previous studies of climate change and mortality have found that global warming’s effects on mortality will be mixed. In a landmark study last year, for instance, a team of economists found that warmer temperatures will grant modest economic benefits to the upper Midwest and New England, since they will blunt those regions’ deadly winters. (The Sunbelt will suffer far worse costs during the summer, however.)

Not so for the suicide rate, which will increase ubiquitously under global warming.

“‘Climate change is going to generate winners and losers’—this is a phrase you hear all the time,” Burke said. “But for this outcome, it’s all losers. There are no winners. We find these strong linear relationships everywhere when you crank up the temperature.”

Finally, the paper finds there are few ways to blunt this effect. In previous studies, economists have found that air-conditioning adoption causes heat-related mortality to decrease over time. But the climate-suicide link does not show the same pliability.

“We don’t see a mitigating effect of air-conditioner adoption on any of these effects,” Burke said. “Even once you control for income, you still don’t see air-conditioner use come through as a factor. Suicide is a fundamentally different animal than these other types of mortality, like cardiac mortality, that you see in the literature.”

“If there is going to be adaptation in the future, it has to be unprecedented. It has to be unlike anything we’ve seen over the past half century in the United States,” he added.

The new study doesn’t address the underlying mechanism beneath this climate-suicide relationship, though the authors report that certain neurotransmitters, including serotonin, are also used to regulate the body’s temperature. In an interesting coincidence-that-may-not-be-a-coincidence, unusually hot days seem to have nearly the same effect on interpersonal violence in the United States as they do on suicide rates.

Burke, who also co-authored the interpersonal-violence study, called the effects “startlingly similar” in size. “It’s a hypothesis, but there could be a similar biology that links your willingness to harm someone else with your willingness to harm yourself,” he said.

Until researchers understand that mechanism better, however, there is only one clear step that can blunt that future suicide-rate increase: We have to keep the planet’s temperature from rising.

“If we reduce future temperature increase, it doesn’t matter where you are, the U.S. or Mexico,” Burke said, “you are going to benefit from that reduction.”

Burke added that it was “clearly baloney” to claim that climate change is “the most important cause of suicide.”

“People are really experiencing this day to day, and we want to be careful around that,” he said. But from a societal perspective, “when we look at the data, we see a relationship that is very strong.”

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.