In 1980, a nuclear power plant in Forsmark, Sweden, began pumping cooling effluent from its reactors into “Biotest Lake,” a manmade, 84-hectare enclosed body of water jutting out from the Baltic Sea. Separated from its aquatic environs and featuring a habitat that was consistently some six to 10 degrees Celsius warmer than the surrounding Baltic, the lake provided, as one group of researchers put it, “an unprecedented model to examine the long-term physiological responses of temperate fishes facing a severe climate-warming scenario under ecologically realistic conditions.”¹

Some native species quickly disappeared from Biotest Lake,² while others persisted. The European perch, in particular, proved remarkably resilient. But it became clear that it was not the same fish it used to be. Biotest perch had apparently undergone some degree of genetic selection and physiological adaptation: These fish were larger than control specimens, had smaller organ masses, and, crucially, had a lower resting heart rate than control fish placed in a warmed environment.

The popular imagination, when thinking about climate change impacts, often focuses on sea-level rise and how coastal dwellers will adapt. The poster for the Hollywood disaster film The Day After Tomorrow shows the torch of the Statue of Liberty rising from the sea. Kim Stanley Robinson’s novel New York 2140 imagines a submerged Gotham in which the rich live on safe upper floors of Manhattan skyscrapers.

But heat may be the bigger story. Climate scientists Steven Sherwood and Matthew Huber point out that “if warmings of 10 degrees Celsius were really to occur in next three centuries, the area of land likely rendered uninhabitable by heat stress would dwarf that affected by rising sea level.” Heat stress, scientists argue, “deserves more attention as a climate change impact.” There are already projections that some places in the Persian Gulf could be, without “significant mitigation,”³ virtually uninhabitable by the end of the century. A paper by Alistair Woodward, head of epidemiology and biostatistics at the University of Auckland, and his colleagues projects that by 2085 only a handful of cities outside western Europe could put on a “low risk” Summer Olympics. Already, the 2022 World Cup, hosted by Qatar, has been moved to winter, a cultural shift on the order of pushing Christmas to July.

Unlike the fish in the Biotest lake, we humans can adapt by changing our behaviors. And we are already. “If the present-day population had the same sensitivity to heat as that of the early 20th century,” one analysis suggests, “the weather observed during the period 1980–2009 would have resulted in more than four times the number of observed deaths (2,993 versus 689).”

But as the temperature continues to climb, changing our behavior may not be enough to protect us. Will we wither away, or, like the European perch, will our physiology begin to adapt?


Sweating, according to one theory, is a central part of the story of human evolution. Human ancestors transitioned to a savannah environment from cooler, forested environments precisely because of climate change: an increased aridification that may have been the result of global climate pulses, or more local tectonic uplifts that cast much of East Africa into a rain shadow. We were suddenly compelled to cover long distances in the heat to track down more sporadic food sources.

So, as one strand of argument goes, we lost our fur and switched almost entirely to thermal, or eccrine sweat glands (“the most copious water-producing gland possible”). Humans became the only naked, bipedal mammals whose bodies are primarily cooled by an eccrine glandular system. “In humans,” notes physiologist Hanns-Christian Gunga, “the maximum sweat production in relation to the body surface area is markedly higher than in any other organisms.”¹⁰ Our newly upright, more slender posture also helped: Our bodies moved farther away from the hot ground and became less exposed to overhead sun.

The changes allowed us to become persistence runners, able to track prey for hours without developing heatstroke,¹¹ even though running generates as much as 10 times more heat than walking.¹² (It’s no coincidence that one of the greatest volumes of sweat output ever tracked in a human was that of noted marathoner Alberto Salazar.¹³)

Heat, then, helped make us who we are today. Like all animals, humans are thermodynamic machines that need to maintain an energy balance. As our environment warms up, our balance changes.

Most of us have a remarkable ability to acclimatize to heat, a process that typically takes a few weeks but begins almost immediately. Our bodies adapt in a number of ingenious ways.¹⁴ The saline content of our sweat, initially as high as 60 milliequivalents of solute per liter (mEq/L), gradually declines, getting as low as 10 mEq/L as our body fights to retain valuable sodium. Electrolytes begin to lower the vapor pressure of our sweat itself, making evaporative cooling more efficient. Our metabolism shifts, plasma volume expands, proteins that protect against heat shock build up. Research subjects who can scarcely complete a walking test in the heat on day one are nailing it by day seven.

Perhaps not surprisingly, in light of our purported running past, fitter subjects are better able to handle increased temperatures, notes Jennifer Vanos, a researcher in biometeorology at University of California, San Diego. To adapt to heat, of course, you have to be in the heat. “You don’t want to be running on a treadmill in air conditioning every day” Vanos says. “It’s good to expose yourself to warmer temperatures.” And the adaptation only works up to a point. Boston Marathon winning times decline two minutes for every 10-degree increase in temperature.¹⁵

These are all acute, short-term changes. But there is evidence we can adjust ourselves to environmental conditions over longer time horizons. A 1968 study found that Bantu workers in South African gold mines — where heatstroke deaths were common — had a remarkable heat resiliency. The miners were placed in climate chambers with steadily increasing “wet bulb” temperatures (equal to the air temperature at 100 percent humidity and lower than the air temperature at lower humidity). While things did get iffy at two hours at 100 degrees Fahrenheit (“a number of men became aggressive, a few became hysterical, and a few maintained a stoical silence”), the researchers were struck by the off-the-charts performance of the Bantu subjects compared to benchmarks noted in other studies. It wasn’t better fitness, the researchers noted, but a “greater stability of the circulatory system.”¹⁶

Over even longer time horizons, we know that more transformative acclimatization can occur. For example, where a person lives — and particularly, where they spend the first few years of life — influences how they sweat. Arctic peoples like the Ainu, research has found, have fewer sweat glands than people dwelling in tropical climates. One 1974 study, titled “Regional Sweating in Eskimos Compared to Caucasians,”¹⁷ found that adult male Eskimo subjects had more sweat glands on their faces and comparatively fewer on areas of the body typically covered by winter clothing — as if they had very specifically adapted to the needs of their environment.


The Biotest perch exposed to warm nuclear effluent could adapt only up to a point, and then they hit a physiological ceiling. At a certain threshold — 4.6 degrees Celsius above their Biotest high temperatures — the perch were more not likely than their unadapted Baltic-dwelling neighbors to survive a temporary heat shock. The perch had adapted, but they were living on the edge of survivability, and this only became clear by hitting them even harder.

Humans will also have their own ceiling. Sherwood and Huber note that the second law of thermodynamics “does not allow an object to lose heat to an environment whose TW” — that’s the wet bulb temperature again — “exceeds the object’s temperature, no matter how wet or well ventilated.” In other words, once a certain threshold is reached, we can no longer cool ourselves by convection or evaporation. Our environment becomes, in effect, a steam room; sweating, our great adaptive tool, is no longer effective. Those savanna areas where evolutionary changes were said to have taken place — places like Kenya’s Turkana Basin¹⁸ — were hot, but not necessarily the wet-bulb hellscapes that some climate change models are projecting.

Studies have identified the “critical thermal maximum” as 35 degrees Celsius, which is the same as human skin temperature. (Our core temperature is slightly higher, at 37 degrees.) As Jeremy Pal and Elfatih Eltahir write in Nature Climate Change, “35 degrees Celsius is the threshold value of Tw beyond which any exposure for more than six hours would probably be intolerable even for the fittest of humans, resulting in hyperthermia,” or heatstroke. Vanos points out that the survivability metric is “just based off of pure physics, [and has] never really been tested.”

Sometime this century, however, it may be. In recent meteorological history, wet-bulb temperature has rarely maxed out at more than 31 degrees Celsius. And good thing: One South African mine study found that “90 percent of all heatstroke cases occurred at wet-bulb temperatures of 30.0 degrees Celsius.”¹⁹ But Pal and Eltahir project that the deadly 35-degree threshold will increasingly be reached by the end of the century, in regions like the Persian Gulf and the densely inhabited regions around the Indus and Ganges river basins.²⁰ They are talking about temporary heat waves, and, presumably, people living in those temperatures will, if they can, make behavioral adaptations: staying out of the sun, retreating into air conditioning, burrowing underground like desert tortoises.

But not everyone will be able to burrow and withdraw. And as temperatures continue to climb and indoor air-conditioning systems are strained, some adaptive responses will begin to kick in. In the very long-term, could we adapt to the new climate normal the way our savanna-dwelling forebears once did? Could our organs change, like the European perch’s did? Could our sweat become more better suited for aridity? Would we become taller and longer limbed so as to dissipate heat more reliably? Could our metabolisms undergo a permanent shift?

Humans are master adapters. The “variability selection” hypothesis, promoted by Rick Potts, director of the Smithsonian Institution’s Human Origins Program, suggests that we did not evolve to meet the needs of one environment but to the exigencies of novel and changing environments. “Key hominid adaptations, in fact, emerged during times of heightened variability,”²¹ Potts writes. We are adaptive machines, living in a remarkable variety of conditions across the planet. Witness those living at elevated environments, with much less oxygen than the rest of us require, presumably thanks to genetic variations passed on from ancestors.²² Similarly, archaeologist Patrick Roberts and anthropologist Brian Stewart argued this month in Nature Human Behavior that humans have carved out a unique ecological niche, which they call “the generalist specialist,” defined by an ecological plasticity.²³ “Not only did [our species] occupy and utilize a diversity of environments, but it also specialized in its adaptation to some of these environmental extremes.”

But it’s hard to pinpoint specific climate-related adaptations.

Arslan Zaidi, a researcher in genetics in the Department of Anthropology at Penn State, has tried to find some. He is the co-author of a paper²⁴ investigating the idea that variations in human nostril shape may be the result of adaptations to climate — the thought being that people in warmer climates have wider nostrils, which helps as a sort of internal air conditioning for the human body. The paper finds that nares width is only “weakly correlated” with temperature, and that other factors, like sexual selection, might have played equally important roles.

This speaks to the difficulty in trying to anticipate a human physiological response to rising temperatures. “Evolution,” Zaidi says, “is not as deterministic as one might think.” Randomness enters the picture, via genetic drift and mutation, so it “impossible to predict with confidence what’s going to happen in the future.” Natural selection, he adds, is “a weak evolutionary force in humans, especially when it comes to traits that are not lethal or absolutely essential for our survival at an early age.” Heat is already a threat, but we don’t go out of our way to subject our bodies to it.

Evolution also takes a long time. “There is a wait for beneficial mutations to arise — if they ever will — then reach a high enough frequency in the population to be ‘noticed’ by selection,” Zaidi says.“We don’t know how fast adaptation will or can occur,” says epidemiologist Woodward. “But it seems unlikely that biological change will be sufficient. The rate of change is the critical factor, and warming on this occasion is occurring orders of magnitude faster than ever before, as far as we can tell.”

It’s not clear whether climate change will become the latest episode in an evolutionary history full of successful adaptations, or whether the changes this time around will simply be too rapid. But we are warming up our very own Biotest Lake, and the clock is ticking.

Update: An earlier version of this story misstated the date that the Swedish plant went online. It was 1980. The story also misidentified Jennifer Vanos’ affiliation. She is at UCSD.