According to a recent study, parts of the Earth could start to become uninhabitable within a century. Surely it cannot be true?
IT IS the late 23rd century. Houston, Tel Aviv, Shanghai and many other once-bustling cities are ghost towns. No one lives in Louisiana or Florida anymore, and vast swathes of Africa, China, Brazil, India and Australia are no-go zones, too. That’s because in all of these places it gets hot and humid enough to kill anyone who cannot find an air-conditioned shelter.
This is the nightmare scenario outlined in a study published earlier this year. If we carry on as we are, it claimed, in as little as a century a few small areas might start to get so hot in summer that no one could survive without air conditioning. Three centuries from now, up to half of the land where people live today would regularly exceed this limit.
“I knew just from basic physics that there would be a point at which heat and humidity would become intolerable, and it didn’t seem that anyone had looked at that from a climate change perspective,” says Steven Sherwood, an atmospheric scientist at the University of New South Wales in Sydney, Australia. “When you look at the data, it becomes pretty clear that it wouldn’t take as much climate change as people seem to think to hit this.”
This is an astounding claim. Scientists have long warned that climate change will have serious consequences: big sea-level rises, floods, droughts, more extreme weather, extinctions and so on. But if Sherwood and co-author Matthew Huber of Purdue University in Indiana are right, huge parts of the planet could effectively become uninhabitable.
So New Scientist set out to discover if their claim really is plausible. What is the limit of human survival, and could global warming really lead to this limit being exceeded in some areas?
Even today, heatwaves can kill tens of thousands of people. In France alone, more than 14,800 people died of heat stroke in 2003. Hurricanes, tornadoes and floods might get all the headlines, but in the US heatwaves claim more lives each year than all of these phenomena combined.
The victims of heatwaves are usually the most vulnerable: the sick, the elderly and the very young. As heatwaves become more severe, though, the proportion of the population dying will rise. Even healthy adults acclimatised to heat will succumb if it stays too hot and too humid for too long. To function normally, we have to maintain a core body temperature of around 37 °C. If it rises above about 42 °C, we die.
Exactly why is still not understood. The body diverts blood to the skin to try to cool off, which cuts the blood supply to the gut. One theory is that bacterial toxins from damaged guts start leaking into the bloodstream, eventually causing multiple organ failure.
What is clear is that to prevent our core temperature rising too high, our skin temperature must not exceed 35 °C for more than a few hours. In dry climates sweating will cool the skin sufficiently even in temperatures of 45 °C or more. But in humid climates where the air is nearly saturated with moisture, sweating makes little difference.
So temperature alone is a very poor guide to what people can survive. A better indicator is the “wet-bulb temperature”. This is the temperature that a mercury thermometer wrapped in a wet cloth would record. It is a measure of both heat and humidity, and reflects the temperature you could lower your skin to by sweating.
Even fit and healthy people couldn’t survive sustained wet-bulb temperatures above 35 °C, say Sherwood and Huber. This heat-stress limit applies even to people sitting naked in the shade next to a fan. Without air conditioning or access to cooler or less humid places, they will die.
The claim that people cannot survive a wet-bulb temperature of 35 °C or more for long is reasonable, says Chris Byrne, an exercise physiologist who specialises in human thermoregulation at the University of Exeter, UK. “At any temperature above that, we switch from a state where we’re losing heat from the skin to the environment to one where the environment imposes a heat load through the skin,” he says. “There’s no doubt that if those conditions arise, you’re probably looking at a lethal situation for the vast majority of the population.”
To find out whether this heat-stress limit is ever exceeded today, Sherwood and Huber looked at patterns of temperature and humidity around the world during the past decade. In a few places, such as Death Valley in California, temperatures can hit 50 °C or more.
There is far less variability in wet-bulb temperatures, though, as the highest temperatures tend to occur in deserts with low humidity. At present, annual maximum wet-bulb temperatures almost never exceed 31 °C (see graphs).
There is a reason for this, a kind of natural thermostat: the hotter and more humid air becomes, the more likely it is to rise and be replaced by cooler air. That’s because humid air is less dense than dry air of the same temperature and pressure, as water molecules weigh less than those of oxygen and nitrogen. What’s more, when humid air starts to rise, the water vapour begins to condense and release latent heat, warming the air and making it rise even higher.
In weather jargon, hot and humid air is likely to be unstable, often leading to thunderstorms that cool things down. “That means that no matter where you go around the world, the wet bulbs top out at around 30 or 31 °C,” says Sherwood.
But instability is relative: air rises only if it is less dense than surrounding air. So if the entire tropics get warmer and more humid, air will have to be somewhat hotter and more humid before it starts to rise. Global warming, in other words, will crank up the thermostat.
To work out how wet-bulb temperatures will change as the world warms, Sherwood and Huber turned to a computer model. The take-home figures: for every 1 °C that the global average temperature rises, maximum wet-bulb temperatures will rise by about 0.75 °C (Proceedings of the National Academy of Sciences, vol 107, p 9552).
Other researchers see no problem with this prediction. “The climate modelling here is very solid,” says Peter Stott of the Met Office Hadley Centre in Exeter, UK, which researches climate change. “Having chatted to a couple of colleagues about this, we think what they are saying seems entirely reasonable and consistent with what we would expect.”
The upshot is that if global average temperatures rise by 7 °C, the maximum wet-bulb temperatures in a few places will start to exceed the 35 °C survival barrier for periods of hours or days. Of course, heat stress won’t suddenly start to be a problem only at this point. Rather, as heat and humidity slowly rise from today’s levels, heatwaves will kill more and more people. Their economic impact will also climb as physical labour outdoors or in buildings with poor air-conditioning becomes increasingly difficult.
With global warming of 7 °C, however, the heat and humidity would start to become intolerable in some places. “We’ll be seeing migrations out of hot and humid countries where people can’t survive the heatwaves,” says Sherwood.
If the global average temperature rises by 12 °C, half of the land inhabited today would become too hot to live in (see map). The uninhabitable regions are likely be those with the highest heat stress today, including the Amazon basin, India, parts of Africa, China, Australia and the south-eastern US.
That would have a nightmarish outcome. In theory, people in affected regions could adapt, living in underground shelters during the summer months, for instance, or not leaving air-conditioned houses or cars for long without cooling suits.
But the cost of running air conditioning constantly could be unaffordable for many individuals and businesses. Most livestock, for instance, would perish unless kept in air-conditioned barns at times. Even worse, a prolonged power cut could cause mass fatalities. “That’s a scary scenario,” says Sherwood. It seems likely that most people would move to cooler climes – leading to conflict over land and resources.
“The scenario they’ve laid out is pretty devastating. It’s a much more serious and catastrophic outcome than people have identified before in the context of heat-related mortality,” says Patrick Kinney, director of the climate and health programme at Columbia University’s Mailman School of Public Health in New York. “It seems to be based on sound reasoning, and good models and data. People have already thought about ill-health effects of climate change, but nobody that I know of has considered there being a threshold above which it basically becomes impossible for people to live.”
Kinney says we should be cautious about quantitative statements, such as “half the inhabited land”. “But the fact that any amount of the world would become uninhabitable, and we don’t know the exact proportion – that to me is alarming,” he says.
So there seems little reason to doubt that if the world gets warm enough, parts of it will start to exceed the heat-stress limit. The big question, then, is whether the planet will get warm enough. Could there really be a global temperature rise of 7 °C or more? The short answer is yes, in the long term.
How much the world will warm depends on two things: how much more carbon dioxide we pump into the atmosphere and how much warming that CO2 produces, also known as climate sensitivity. According to the Intergovernmental Panel on Climate Change (IPCC), every doubling of CO2 in the atmosphere will raise the temperature by between 1.9 and 4.5 °C, with 3 °C being the most likely value.
If climate sensitivity is as low as 1.9 °C, then it would take centuries for the planet to warm by 7 °C even if we continue pumping out lots of CO2. On the other hand, if climate sensitivity is as high as 4.5 °C, we could hit the 7 °C point within a century if we carry on as we are, although this could still be avoided as long as we slash emissions soon enough.
The catch is that the IPCC’s estimates of climate sensitivity are based only on fast feedback processes that are relatively easy to model on computers. They exclude slower processes such as the oceans switching from absorbing CO2 to emitting it, or the release of methane from thawing permafrost. Studies of Earth’s past suggest that actual sensitivity could be higher than computer models predict, with a few suggesting it could be as high as 7 °C (New Scientist, 30 June, p 38). If long-term sensitivity is high, it will be much harder to avoid big temperature rises over the next few centuries.
“Most of the discussion has been about the 21st century, but warming isn’t going to stop in 2100 unless our emissions have fallen almost to zero by then, and that would require heroic efforts,” says Sherwood. “If you consider that carbon releases might be a little higher than the most likely value and that the climate might be quite sensitive to inputs of energy, it’s not too hard to get up to 10, 12 or even 15 °C by the 23rd century.”
The danger is that if we pump too much CO2 into the atmosphere, large temperature rises might become inevitable even if all human greenhouse gas emissions cease. “We need to think about how to ensure that a large fraction of the fossil fuels are simply left in the ground,” says Sherwood. “That’s going to take a change in direction that many people are not yet seriously taking on board.”
Even fewer are taking on board the idea that parts of the planet could become too hot for humans to survive. If there is a fault in Sherwood and Huber’s reasoning, however, none of the researchers contacted by New Scientist could identify it.
“Scientists tend to be conservative – we stick to what we know and avoid speculating about things that are hard to pin down,” says Sherwood. “I think heat stress is an important impact of climate change that we’ve missed and there may be others, particularly with large, but possible, warmings.”
How did mammals survive?
IN A warmer world, will some areas really get too hot for humans to survive? One way to try to answer this question is to look at past warm periods, such as the Palaeocene-Eocene thermal maximum 56 million years ago. Before the PETM, Earth was already as much as 10 °C warmer than it is now. Then, over 20,000 years, the temperature shot up by a further 5 to 9 °C.
There were no humans around then, of course, but there were plenty of other mammals. Is there any evidence that heat stress was a major problem for them?
Some mammals have core body temperatures as high as 40 °C, so in theory they should be able to survive levels of heat and humidity that would kill humans. Even so, if calculations by Steven Sherwood at the University of New South Wales in Sydney, Australia, and Matthew Huber at Purdue University in Indiana are right (see main story), in some tropical regions during the PETM, heat stress should have become a problem even for mammals well adapted to hot, humid conditions.
Fossil records are very patchy, so it is unclear if heat stress drove any mammals from the hottest and most humid areas. It does appear, however, that many mammals became smaller during this time, with some halving in mass.
One proposed explanation for this is that foliage became less nutritious and digestible because of high CO2 levels. But a study published earlier this year concluded that other factors must have played a role, as some carnivores shrank too (Journal of Mammalian Evolution, DOI: 10.1007/s10914-010-9141-y).
One explanation is that heat stress drove the evolution of smaller mammals. Small animals have a higher surface area relative to their internal volume, so they can cool themselves more efficiently. “We think this provides a nice way of explaining what has long been known to palaeontologists, which is that mammals get smaller in warmer climates,” says Sherwood.
Article by Hazel Muir.