We live in the best of all possible worlds. This much-lampooned idea originated in the early 18th century, when the German polymath Gottfried Leibniz was pondering what sort of world a benevolent God would make. In the past couple of decades, however, it has acquired a new resonance as we have spied out worlds orbiting other stars.
The question is whether any of these, too, might support life. Underpinning each assessment is a surprisingly simple process that measures each of our potential planetary twins – or, more likely, near cousins – against a presumed earthly idyll. To host life, a world must, like Earth, be rocky, watery and orbiting in a star’s “Goldilocks zone”, where things are not too hot, not too cold, but just right.
But do we actually inhabit the best of all possible worlds? As we come to understand better how the properties of stars, planets and their atmospheres combine to produce life-friendly worlds, it seems Earth’s own habitability is more precarious than we had assumed. That has far-reaching consequences for the likelihood of life on other planets – and for the fate of life on Earth.
Our assumption of Earth’s perfection has largely rested on one fact: our planet is full of life. Life here is invariably constructed from carbon and reliant on liquid water, and there are good reasons beyond egotism to believe that, as it is on Earth, so it is in the heavens. Carbon and water are two of the most common substances in the universe. In tandem, they provide an extravagance of durable chemical products unmatched by any other obvious combination of elements.
The requirement for liquid water means any life-bearing planet must occupy a slim sliver of the space surrounding a star. Too close to the thermonuclear furnace, and water will boil off as steam. Too far away, and it will freeze to ice, consigning life to a frigid fate. Where exactly these boundaries are in a given planetary system depends on a star’s mass and age, which in turn determine how much heat and light it radiates. Earth seems to be snugly sandwiched in the sun’s sweet zone: the best of all possible worlds, at least in our solar system. The laws of physics as we understand them are the same throughout the universe, so presumably any other small, rocky planet in a similarly temperate orbit could also be a Goldilocks world.
If only it were that simple. Estimating where the Goldilocks zone lies depends on other assumptions about a potentially habitable planet’s nature besides the presence of liquid water. Based on its position in the solar system alone, Earth’s average surface temperature should be well below freezing. Its saviour is a heat-trapping atmosphere laced with the greenhouse gases carbon dioxide and water vapour. Such an atmosphere is thought to be a typical result of the way rocky planets form. If Earth’s comfort blanket were much thicker or thinner, however, or had a different chemical make-up, the planet could rapidly cease to be so amenable to life.
Our neighbour Venus illustrates the point. Venus seems to have started out habitable, with a relatively Earth-like ocean and atmosphere. Its proximity to the sun rapidly turned those blessings into a curse. Water began to boil off from the oceans into the atmosphere, where its heat-retaining qualities caused temperatures to rise still further. The result was a runaway greenhouse effect that sterilised the planet as all the CO2 was baked out of its crust and into its atmosphere. Under its stifling sky of almost pure CO2 today, Venus’s surface temperature is some 460 °C – above the melting points of tin, lead and zinc.
In 1993, geoscientist James Kasting of Pennsylvania State University in State College set out to pin down a lot more precisely where the Goldilocks boundaries lie. He and his colleagues examined how varying the intensities and wavelengths of sunlight falling on an idealised Earth affected its atmosphere and surface temperature. Increasing the incident sunlight by some 10 per cent – equivalent to moving Earth inwards from its present position of 1 astronomical unit (AU) from the sun to 0.95 AU – produced a temperature rise that sent water vapour soaring high up into the atmosphere, where it dissipated into outer space. Over tens of millions to hundreds of millions of years, such a “moist greenhouse” would entirely desiccate the Earth and eradicate all surface life (Icarus, vol 101, p 108).
When Kasting tried to place the habitable zone’s outer limit – the point where the fall in temperature is enough to cause irrecoverable global glaciations – he found it to be about 1.67 AU from the sun, slightly beyond the orbit of Mars. Already, these early calculations began to crack Earth’s Goldilocks facade. Earth is not slap bang in the centre of the Goldilocks zone, but significantly towards its inner edge.
Kasting’s climate model was in some respects rather basic, simulating a single, uniform strip of atmosphere essentially devoid of clouds and weather systems. In other ways it was quite elaborate, for example incorporating the sort of positive feedback effects by which increasing atmospheric water vapour leads to a runaway greenhouse effect. It could also replicate many of the climatic quirks of other solar system planets. Extrapolating its conclusions, Kasting was able to chalk in the inner and outer boundaries of the habitable zone around a wide variety of stars of different sizes and luminosities. These have been the gold standard for hunters of habitable planets ever since.
Until now. At the beginning of this year, working with Kasting and a few others, Penn State researcher Ravi Kopparapu updated the calculations for the first time in two decades. A lot has changed, Kopparapu points out. Above all, we now know that water vapour and CO2 are better at absorbing certain wavelengths of infrared light than we used to think. That affects the potency of each gas’s greenhouse heating. Rerunning the models produced a simple, unambiguous result: for a planet like Earth, the habitable zone around all types of stars lies slightly further out than we had assumed (arxiv.org/abs/1301.6674).
That has knock-on effects as we search for promising habitats around other stars. A few previously discovered extrasolar planets have been struck from the list, whereas others are beginning to look more promising. It also enhances the prospects for life around the small, cool “M-dwarf” stars that are the most prevalent in our immediate neighbourhood, shifting their habitable zones outwards towards the sort of distances where most small, rocky planets have so far been found. “It looks like nearly half of all M-dwarfs should have an approximately Earth-sized planet in their habitable zones,” says Kopparapu. As new planet hunts focusing on M-dwarfs start off over the next few years, we should expect to find at least three or four possible Goldilocks worlds right next door.
Teetering on the brink
For the original Goldilocks planet, however, the implications are much murkier. The inner edge of the solar system’s habitable zone moves outwards from 0.95 AU to 0.99 AU. In other words, were Earth just 1 per cent closer to the sun, its water could begin to steam off into space as a moist-greenhouse effect kicks in. Rather than being at a comfortable distance from the edge of the Goldilocks zone, we are teetering on the brink.
That portends an alarming future. As our sun ages, it is fusing hydrogen at higher and higher temperatures and becoming more luminous, pushing the inner edge of the Goldilocks zone outwards. “It suggests the end could come sooner than we thought,” says Kasting. Earth could technically begin to lose water “as early as tomorrow”, he says. More likely is that we can lop a few hundred million years off Earth’s generally accepted remaining habitable time of a billion years or so.
Raymond Pierrehumbert, a climate scientist at the University of Chicago, remains sanguine. Although he acknowledges that the calculations are useful for establishing reference points for habitable zones, he thinks the prospect of imminent desiccation is a mirage that reflects the model’s deficiencies. “Most of what it leaves out arguably makes a planet more habitable, not less,” he says.
Take the effects of clouds. In atmospheres with amounts of water vapour close to the moist or runaway greenhouse limits, clouds are more likely to form at lower altitudes, reflecting more sunlight back into space and cooling the surface beneath. In fact, it is hard to know exactly how an atmosphere that is perhaps 50 per cent water will behave. “A big rainstorm could make you lose half your atmosphere in one area,” says Pierrehumbert, and that might suck in more air from all around, changing the dynamics of the entire atmosphere.
Such oddities, paired with water vapour’s tendency to amplify other sources of warming, mean that some of the most resilient Goldilocks worlds may be those with only trace amounts of water. Astrophysicist Sara Seager at the Massachusetts Institute of Technology and her colleagues have been modelling “desert” planets with insufficient water to support global oceans or a steamy atmosphere. They conclude that a desert planet with just 1 per cent atmospheric water vapour could maintain habitable surface temperatures as close as 0.5 AU to a sunlike star, well inside the orbit of Venus (arxiv.org/abs/1304.3714).
Kasting, Pierrehumbert and their colleagues have recently secured NASA funding to jointly develop a fully three-dimensional climate model, one that will build in more robust and realistic cloud feedbacks and hydrological cycles for a wide variety of potentially habitable rocky planets. Other groups, notably those of François Forget at Pierre and Marie Curie University in Paris, France, and of Jochem Marotzke at the Max Planck Institute for Meteorology in Hamburg, Germany, are also on the case. “Understanding full atmospheric circulation and its effects on clouds and water vapour is the real frontier we’re all moving toward,” Pierrehumbert says. “This is what will ultimately have the largest effect on drawing the habitable zone’s boundaries.”
Yet at least in Earth’s case, the greatest uncertainty may now lie with us humans. Thanks to our burning of fossil fuels, the atmosphere’s CO2 content is currently crossing the 400 parts per million threshold, up from a pre-industrial average of 280 ppm. In its most recent consensus report, published in 2007, the United Nations Intergovernmental Panel on Climate Change pegged the most probable eventual warming from a doubling of CO2 – a level we are likely to reach in the second half of this century – at 3 °C.
Compared with the fate of Venus, that sounds innocuous. But this temperature rise is roughly equivalent to shifting Earth’s orbit inwards by 1 per cent, to 0.99 AU – precisely where the habitable zone’s inner boundary lies in the latest model. Might further greenhouse emissions push us over the edge, eventually to follow Venus’s destiny?
Various studies in recent decades have concluded that this outcome is highly unlikely. Even if we managed to burn most of the planet’s economically recoverable fossil fuel reserves, not merely doubling atmospheric CO2 but increasing it by a factor of 8 or 16, the worst outcome would be only a moderately moist greenhouse.
In work yet to be published, geochemist Colin Goldblatt at the University of Victoria in British Columbia, Canada, has revisited the question using the same updated information on greenhouse-gas absorption that was used to revise the habitable zone boundaries. His conclusion is that if Earth’s present-day atmosphere were to enter a very hot and moist greenhouse state, the extra water vapour would absorb more infrared sunlight than previously appreciated. That could reduce cloud and ice cover, further increasing heat absorption. In turn, that could trigger a runaway greenhouse – but only if we also chose to use the energy from all the Earth’s coal, oil, and gas for no other purpose than to cook up even more CO2 from limestone.
That reassurance does nothing to lessen the shorter-term impacts of climate change on human society in the coming decades and centuries. But palaeoclimate records suggest that, over scales of hundreds of millions of years, our planet is remarkably resilient to changes to its thermostat. “Despite high levels of atmospheric CO2 in the remote past, Earth hasn’t left the habitable zone,” says Goldblatt.
That’s probably down to Earth’s carbonate-silicate cycle. This feature appears to be unique, in the solar system at least, thanks to our planet’s singular combination of oceans and tectonic activity. As rising temperatures start to steam water from the oceans, increased rainfall washes more CO2 out of the atmosphere, ultimately sequestering it in rocks for many millions of years before it is eventually burped back out by volcanoes.
It is Earth’s natural temperature regulator – and perhaps its greatest claim to be the best of all possible worlds. It could also, ironically, be life’s ultimate undoing, perhaps even before the sun would otherwise become too bright for comfort. According to calculations performed by Kasting and his student Ken Caldeira in 1992, perhaps a billion years from now Earth’s interior will have cooled enough to substantially reduce volcanism, slowing the carbonate-silicate cycle and leaving so much carbon locked in rock that Earth’s atmosphere will cease to support most forms of photosynthesis (Nature, vol 360, p 721).
Till then, though, is life on Earth safe? Given our vast uncertainties about how climates work – and now even about Earth’s position in the Goldilocks zone – Goldblatt reckons it is probably unwise to take too much for granted. “It’s like playing tag on top of a cliff on a foggy day. No one’s fallen off yet, but you don’t know how close the edge is.”
(Article by Lee Billings, who is a freelance writer based in New York City)