Road markings appear distorted as the asphalt starts to melt due to the high temperature in New Delhi on May 27, 2015. (Harish Tyagi/European Pressphoto Agency)

The temptation to paint a dire picture of climate change, at a time when the Trump administration seems bent on questioning a widely accepted body of climate science and withdrawing from international agreements, is clear. But the picture still has to be plausible and accurate, a number of scientists argued this week in response to a lengthy article in New York Magazine.

The article by David Wallace-Wells is entitled “The Uninhabitable Earth” and begins with the sentence, “It is, I promise, worse than you think.” It runs through a number of extreme climate possibilities — most prominently, the idea that some parts of the Earth will experience a combination of heat and humidity so intense that human beings won’t be able to survive outdoors in some regions.

The article gives the sense of being fairly definitive, saying it is “the result of dozens of interviews and exchanges with climatologists and researchers in related fields and reflects hundreds of scientific papers on the subject of climate change.” Wallace-Wells also warns that he’s describing worst-case scenarios that might not be realized but could occur if the world does not take action. “But those scenarios, and not the present climate, are the baseline,” he writes. “In fact, they are our schedule.”

Yet a number of scientific responses have been highly critical.

Pennsylvania State University’s Michael Mann, a climate researcher known for skewering skeptics of climate change, took the lead in debunking the Wallace-Wells story Monday, writing, “The article argues that climate change will render the Earth uninhabitable by the end of this century. Extraordinary claims require extraordinary evidence. The article fails to produce it.”

Mann objected in particular to Wallace-Wells’s statement regarding “satellite data showing the globe warming, since 1998, more than twice as fast as scientists had thought,” calling it “erroneous.” More broadly, he argued that climate change is bad enough without having to exaggerate the dangers:

The evidence that climate change is a serious problem that we must contend with now, is overwhelming on its own. There is no need to overstate the evidence, particularly when it feeds a paralyzing narrative of doom and hopelessness.

As Mann also notes, the story starts out by describing the massive amount of carbon that lies in Arctic permafrost, “all of it scheduled to be released at a date that keeps getting moved up,” and suggesting that much of it will be released as the hard-hitting greenhouse gas methane.

That’s very difficult to square with existing research: A major study published in 2015 by a large group of permafrost experts found that only about 5 to 15 percent of permafrost carbon was likely to be released during this century, and that most of it would be in the form of the slower-warming carbon dioxide, not methane. (Granted, that’s still a substantial addition to the world’s emissions.)

Another climate researcher, Bob Kopp of Rutgers University, commented, “Overall, the article highlights important effects that have been discussed in the literature, but in a manner that is often sloppy and hyperbolic. It would have been helpful had the reporter identified his sources, which makes it difficult to check what he intended in some points.”

Tweets by other scientists and climate experts give a sense of their worries about the article:

In response to questions from The Washington Post, Wallace-Wells sent a series of emails, saying that he was “extremely careful in working the science of climate change into this piece, mindful all along of the many problems that could arise in the translation. I was especially concerned because I have a tremendous, almost inexpressible, respect for the researchers and scientists who are doing this work, and did not want to misleadingly portray any of that research.”

He said he had not seen anything yet that required a correction, other than a minor error, and that he had run many sections by climate experts. (He defended his work further in a series of tweets here.)

He also defended his reliance on dire scenarios, saying that he had worked from the worst warming scenario presented by the United Nation’s Intergovernmental Panel on Climate Change — basically, a “do nothing” scenario for the world — and had considered a warming of 4, 6, and 8 degrees Celsius in the story. “My purpose in writing the story wasn’t to survey the median scenario, it was to survey the worst-case scenarios, because I believed — and still believe — that the public does not appreciate the unlikely-but-still-possible dangers of climate change,” he said.

The “likely” warming expectation, for the worst warming scenario presented by the IPCC in 2013, is 2.6 to 4.8 degrees Celsius for the 2081-2100 period. So in effect, the story is often going beyond what is deemed “likely” even in this worst scenario.

The story runs through threats to food supplies, risks of increased violence and fears of long-dormant viruses awakening from Arctic permafrost (another probably overblown fear), but its centerpiece focuses on the danger of extreme heat combined with high humidity. This is something scientists like Kopp have highlighted, and it’s a very real issue — indeed, a seminal study on the subject was published in 2010. So it’s been known for a while.

A key concept to understand here is the “wet bulb temperature,” which Kopp and his colleagues define as “essentially a measure of how well you can cool your skin by sweating, which is how humans stay alive in the worst heat.” But how frequently will these dangerous conditions occur?

Kopp has found that in the United States, by the century’s end under the IPCC’s worst warming scenario, “a third of the population would be expected to experience a day or more of extraordinarily dangerous conditions, with wet-bulb temperatures exceeding 33°C, or 91°F, in a typical summer. At wet-bulb temperatures exceeding 33°C, about an hour of vigorous, shaded activity leads to skin temperatures of 100°F and core body temperatures of 104°F — the threshold for heatstroke, which can be fatal.” Kopp fully acknowledges that hotter and more humid regions will be even more dangerous under this extreme scenario.

Wallace-Wells runs with the concept and assumes very high warming levels when writing about regions of the Earth becoming places where humans won’t be able to cool down:

At seven degrees of warming, that would become impossible for large portions of the planet’s equatorial band, and especially the tropics, where humidity adds to the problem; in the jungles of Costa Rica, for instance, where humidity routinely tops 90 percent, simply moving around outside when it’s over 105 degrees Fahrenheit would be lethal. And the effect would be fast: Within a few hours, a human body would be cooked to death from both inside and out.

Wallace-Wells doesn’t make clear whether he means Celsius or Fahrenheit in the story (an increase of one degree Celsius is an increase of 1.8 degrees Fahrenheit) which makes a big difference. But the study he seems to be relying on considered the possibility of seven degrees Celsius, as well as 11 or 12 degrees Celsius, and derived dire results. The point, therefore, is not that these things can’t happen at all, but rather that these are extremely high levels of warming we’re talking about.

But what are the odds? That’s the crucial question. In light of current energy trends and the Paris climate agreement, it seems more likely at present that human society will slowly bend its emissions curve downward, missing targets set by climate scientists (and blowing by 2 degrees Celsius of warming) but not hitting these worst-case scenarios, either. That doesn’t mean there won’t be major impacts — there will — but it does mean there is good reason for hope we can avoid the very worst outcomes.

“You raise the question of odds, and it’s a fair critique, I think,” said Wallace-Wells. “I could’ve included some estimates of probability and likelihood, and in some earlier drafts I did work in a few of those numbers.” But he said he felt such estimates are “quite speculative.” Even as he defended outlining worst-case scenarios, he also agreed by email that about 2.5 or 3 degrees Celsius of warming appears most likely this century.

It is difficult to tell the story of climate change. It never happens on any particular day; it never moves very rapidly; it does not directly cause individual weather events. But a number of scientists are saying that this particular story has gone too far.