Tweet
At least, that's what I take from this essay by Charles Stross.
Also, our odds of finding ET with SETI are depressingly small.
How habitable is the Earth?
(Pay attention at the back: this is a trick question.)
We H. Sapiens Sapiens appear to be an infestation on this planet. After the slow-burning evolution of hominins in Africa, our ancestral populations erupted out into Eurasia in a geological eye-blink, spread into the Americas by way of the Bering land bridge (sea levels being somewhat lower during the ice ages) and finally reaching even the remotest islands of oceania around twelve thousand years ago. Today we're ubiquitous. Even our pre-industrial ancestral cultures, from those resembling the inuit to the antecedents of the tuareg, occupied a slew of geographical environments that put cockroaches to shame.
So you'd think that, to a first approximation, the Earth is inhabitable by human beings. And this tends to colour our approach the prospects of finding extrasolar planets that might be hospitable to human life (if we could ever get there from here).
Actually, I think this is not quite the case. In fact, to a first approximation, from the perspective of prospective interstellar colonists, the Earth is uninhabitable. That we could imagine otherwise bespeaks a profound cognitive bias on our part (and a degree of relativism: because when all's said and done, the Earth is a lot less hostile than, say, the surface of Venus or the cloud base of Jupiter).
Why is the Earth uninhabitable?
Let's play a thought-experiment ...
I want you to imagine that, instead of being a perplexed mostly-hairless primate reading a blog, you're the guiding intelligence of an interstellar robot probe. You've been entrusted with the vital mission of determining whether a target planet is inhabitable by members of your creator species, who bear an eerie resemblance to H. Sapiens Sapiens. To gauge the suitability of the target world you've been given an incubator that can generate decorticated human clones — breathing meat-machines with nobody home up top. When you get to the destination you're going to transfer them to the surface and see how long they survive. If it can make it through 24 hours (or one diurnal period), congratulations! — you've found a potential colony world; one so hospitable that a naked and clueless human doesn't die on their first day out.
Your first destination planet is the cloud-whorled third planet out from an undistinguished G2 star, orbited by an airless, tidally-locked moon with roughly 1.3% of the planet's own mass. (Sound familiar? It should be.) You start sending down meat-machines to probe the surface at random. What conclusions do you draw about the inhabitability of Earth?
Let's start with Earth in its current configuration.
78% of the surface area is seawater. Drop a naked meat puppet there and it's going to go glug glug glub ... tritely, this is Not A Good Start.
Of the remaining 22%, about one third is either mountain ranges, deserts, or ice caps. It's reasonable to say that, in the absence of protective equipment, the meat probes are going to die of exposure in less than one diurnal period — possibly in as little as an hour if they're unlucky enough to land in the middle of the Antarctic winter.
We're down to about 15% of the planetary surface — 15% that isn't lethal without life support equipment such as boats, tents, and clothing. Our meat probes can breathe the air without their lungs freezing or dessicating. They aren't going to drown rapidly. And they aren't going to roll off a cliff. They might get a tad sunburned or hypothermic depending on the weather, and they might be eaten by a mountain lion or bitten by a rattlesnake, but they stand a reasonable chance of making it through 24 hours on the surface without dying.
Triumph! We have confirmed that a small part of this planet is inhabitable. Except ... I cheated. I pulled a fast one on you. Because I picked Earth in its current configuration — as it is today.
As a species, H. Sapiens has only been around for somewhere in the range 70,000-200,000 years. We are fine-tuned for survival on the Earth of this time frame. However, the Earth is a lot more than 200 kiloyears (Ky) old; the surface formed roughly 4.6 Gy ago (gigayears — 1Gy = 1,000,000,000 years). And we can expect the Earth to persist for about another 3-5 Gy, until the sun leaves the main sequence of the Hertzsprung-Russell diagram and becomes a red giant, presumably swallowing the Earth (or at any rate rendering it too crispy for comfort). So if we're being honest (and not cherry-picking our candidate stellar colony mission targets) we've got a 8-10Gy span to probe.
Back up far enough in the Earth's time-line (before about T minus 4.6 Gy) and we run into the formation of the solar system — and the proto-Earth, before its postulated oblique impact with Theia, a proto-planet roughly the size of Mars — debris from which collision condensed into our moon, Luna. Earth definitely wasn't a candidate for human inhabitation in those days — a largely airless blob of molten rock, under continuous heavy bombardment by planetary embryos and small planetesimals thrown out of the asteroid belt by Jupiter, as the two large gas giants (Jupiter and Saturn) churned the protoplanetary disk. Indeed, just about anywhere in the inner solar system at any time prior to the end of the Late Heavy Bombardment (which ended at T minus 3.8 Gy) is a poor candidate for a space colony — it was the LHB that resurfaced and cratered the moon, and presumably did a similar number on the Earth's then-fragile surface. On the other hand, this period left us geological evidence in the shape of minerals dated to the Hadean aeon. We know relatively little about the Hadean, although there's some recent evidence to support active plate tectonics and a surface temperature compatible with liquid water.
But don't get the idea that late Hadean Earth was a fun place to be. For one thing: that great big moon of ours didn't condense from a debris cloud at its current orbital distance. Tidal dragging is widening the lunar orbit by about 3.8 metres per century; it's now orbits roughly twice as far out as it did when it formed. Which in turn means that the young Earth spun on its axis far faster than it does today, and the tides the newborn molten-faced moon raised during the Hadean aeon would have been something to behold (preferably from a very high altitude). In fact, early Earth was a very alien world indeed. To hit on wikipedia (because I'm feeling too lazy to type):
Recent evidence suggests the oceans may have begun forming by 4.2 Ga.[22] At the start of the Archaean eon, the Earth was already covered with oceans. The new atmosphere probably contained ammonia, methane, water vapor, carbon dioxide, and nitrogen, as well as smaller amounts of other gases. Any free oxygen would have been bound by hydrogen or minerals on the surface. Volcanic activity was intense and, without an ozone layer to hinder its entry, ultraviolet radiation flooded the surface.
We don't know when life got started on Earth. However, we do know that the early history of life relied on anaerobic processes for a surprisingly long time. It wasn't until roughly 580 My ago (at the end of the Proterozoic era) that free oxygen came to dominate our planet's atmospheric chemistry. Back to wikipedia:
The most widely accepted chronology of the Great Oxygenation Event suggests that oxygen began to be produced by photosynthesis by organisms (prokaryotic, then eukaryotic) that emitted oxygen as a waste product. These organisms lived long before the GOE, perhaps as early as 3,500 million years ago. The oxygen they produced would have almost instantly been removed from the atmosphere by the weathering of reduced minerals, most notably iron. This 'mass rusting' led to the deposition of banded iron formations. Oxygen only began to persist in the atmosphere in small quantities shortly (~50 million years) before the start of the GOE. Without a draw-down, oxygen can accumulate very rapidly: at today's rates of photosynthesis (which are admittedly much greater than those in the plant-free Precambrian), modern atmospheric O2 levels could be produced in around 2,000 years.
There are elaborate and fascinating competing theories to explain why it took so long for oxygenation to take off: even the Moon gets blamed (the huge 50 metre tides that churned the anoxic early oceans and would have sucked premature photoautotrophs to their doom deep below: oxygenation had to wait for the moon to migrate out far enough for the tides to die down, permitting photosynthetic organisms to thrive near the surface). The point to take away from this is that, between T minus 4.6 Gy and T minus 0.56 Gy, the Earth's atmosphere was largely free of oxygen. Meat probe says: "aaargh, choke".
Even after the oxygen catastrophe, our space probe isn't going to find a terribly hospitable planet. The sun's luminosity is increasing by around 6% every billion years. The Proterozoic earth was a cold place by our standards; there were ice ages in which glaciation closed in on the equator. And there was continental drift. Continental drift leaves faint traces: we know that 250 My ago, our current continents were united in a single land mass (called Pangaea). There's some evidence that between 800 My and 550 My another single supercontinent (Pannotia or Vendia), and of another mass 1000-850 My ago (Rodina). And there seems to be a correlation between the occurence of "snowball Earth" events (those equator-reaching ice caps) and the unitary supercontinents. Back to wikipedia:
Most paleoclimatologists think the cold episodes had something to do with the formation of the supercontinent Rodinia. Because Rodinia was centered around the equator, rates of chemical weathering increased and carbon dioxide (CO2) was taken from the atmosphere. Because CO2 is an important greenhouse gas climates cooled globally.
Supercontinents too close to the equator: snowball earth. (Meat probe turns blue and shivers.)
So here's the upshot: of the 4.6 Gy of Earth's known history, there's only been enough oxygen in the atmosphere for us to survive for about 0.5 Gy. For roughly 90% of the Earth's history we couldn't even breathe the air. And about 10-25% of the time, there have been ice ages so savagely fierce that the glaciers reached the tropics: odds are good that any meat probe landing on solid ground during these periods would rapidly die of exposure. So historically, Earth has only been inhabitable about 8% of the time — assuming you are lucky enough to find some solid ground. Once you factor in the random surface distribution, we're down to about 2% survivability.
Now let's look at the future of the solar system.
We know the sun is steadily brightening by about 6% per Gy. It's postulated that within a couple of Gy, solar output is going to have some unpleasant effects. Ultraviolet radiation can split the covalent bonds that hold water molecules together, high in the atmosphere: and hydrogen ions (or, more likely, hydrogen molecules) can be blasted right out of the ionosphere by the same mechanism. The slow, steady loss of Earth's water is a one-way process, but exacerbated by warming (more water vapour in the upper atmosphere means more hydrogen is lost). As hydrogen loss proceeds, we end up with a carbon-dioxide dominated atmosphere and a runaway greenhouse effect like that of Venus.
There are other mechanisms that might render the Earth uninhabitable by our kind of life. Over geological time, the partial pressure of oxygen in the atmosphere has risen. With more solar energy inputs, it may be that oxygen levels continue to soar. Above about 28%, even waterlogged biomass will burn handily: and there are indications that atmospheric oxygen (currently down around 16%) has been well over 20% in the past. If oceanic photoautotrophs pump out too much of the stuff, the continents may well be burned back to bedrock by the resulting lightning-triggered fires.
All of this leaves aside the prospects for either an anthropogenic catastrophe, or the evolution (or creation) of new types of chemoautotrophs that have a drastic effect on the Earth's atmospheric chemistry. Or something else. Phase of the moon, perhaps. (What happens when the moon's tidal drag diminishes further, reducing long-term deep-ocean mixing? Anyone got a clue?)
The upshot is, we may well be most of the way through the Earth's inhabitable epoch. In which case, of the 4 Gy remaining, we may have 0.3 — 1 Gy to go with an oxygen-dominated atmosphere and water close to its triple point — the minimum necessary critera for human survival on a planetary surface.
So, back to the gedankenexperiment. Currently, a random meat probe dropped on the Earth's surface has something like a 15% chance of finding it survivable. But a random sampling over the historical epoch would return a survivability probability of around 1%. And over the future epoch, it's likely similar, unless we're erring massively on the side of pessimism about the prospects for our atmospheric composition remaining stable.
Ergo: to a space probe searching for somewhere that our kind of life can thrive, a truly random sampling of the Earth's surface (distributed over both time and area) would probably result in the conclusion that the planet is uninhabitable. See also: Fermi Paradox
Of course, a random sampling of Mars or Pluto would give an even lower probability of finding the planet to be inhabitable. But that's not the point. The point is this: we are finely tuned survival machines that have evolved to survive in a niche on one particular planet in one particular epoch. Even our own planet is unimaginably hostile to our kind of life for most of its history. And while survival outside that niche is possible with the assistance of a horrendously complex toolkit we call "civilization", we've yet to try it somewhere where we can't count on the basics (free oxygen and triple point water).
(Pay attention at the back: this is a trick question.)
We H. Sapiens Sapiens appear to be an infestation on this planet. After the slow-burning evolution of hominins in Africa, our ancestral populations erupted out into Eurasia in a geological eye-blink, spread into the Americas by way of the Bering land bridge (sea levels being somewhat lower during the ice ages) and finally reaching even the remotest islands of oceania around twelve thousand years ago. Today we're ubiquitous. Even our pre-industrial ancestral cultures, from those resembling the inuit to the antecedents of the tuareg, occupied a slew of geographical environments that put cockroaches to shame.
So you'd think that, to a first approximation, the Earth is inhabitable by human beings. And this tends to colour our approach the prospects of finding extrasolar planets that might be hospitable to human life (if we could ever get there from here).
Actually, I think this is not quite the case. In fact, to a first approximation, from the perspective of prospective interstellar colonists, the Earth is uninhabitable. That we could imagine otherwise bespeaks a profound cognitive bias on our part (and a degree of relativism: because when all's said and done, the Earth is a lot less hostile than, say, the surface of Venus or the cloud base of Jupiter).
Why is the Earth uninhabitable?
Let's play a thought-experiment ...
I want you to imagine that, instead of being a perplexed mostly-hairless primate reading a blog, you're the guiding intelligence of an interstellar robot probe. You've been entrusted with the vital mission of determining whether a target planet is inhabitable by members of your creator species, who bear an eerie resemblance to H. Sapiens Sapiens. To gauge the suitability of the target world you've been given an incubator that can generate decorticated human clones — breathing meat-machines with nobody home up top. When you get to the destination you're going to transfer them to the surface and see how long they survive. If it can make it through 24 hours (or one diurnal period), congratulations! — you've found a potential colony world; one so hospitable that a naked and clueless human doesn't die on their first day out.
Your first destination planet is the cloud-whorled third planet out from an undistinguished G2 star, orbited by an airless, tidally-locked moon with roughly 1.3% of the planet's own mass. (Sound familiar? It should be.) You start sending down meat-machines to probe the surface at random. What conclusions do you draw about the inhabitability of Earth?
Let's start with Earth in its current configuration.
78% of the surface area is seawater. Drop a naked meat puppet there and it's going to go glug glug glub ... tritely, this is Not A Good Start.
Of the remaining 22%, about one third is either mountain ranges, deserts, or ice caps. It's reasonable to say that, in the absence of protective equipment, the meat probes are going to die of exposure in less than one diurnal period — possibly in as little as an hour if they're unlucky enough to land in the middle of the Antarctic winter.
We're down to about 15% of the planetary surface — 15% that isn't lethal without life support equipment such as boats, tents, and clothing. Our meat probes can breathe the air without their lungs freezing or dessicating. They aren't going to drown rapidly. And they aren't going to roll off a cliff. They might get a tad sunburned or hypothermic depending on the weather, and they might be eaten by a mountain lion or bitten by a rattlesnake, but they stand a reasonable chance of making it through 24 hours on the surface without dying.
Triumph! We have confirmed that a small part of this planet is inhabitable. Except ... I cheated. I pulled a fast one on you. Because I picked Earth in its current configuration — as it is today.
As a species, H. Sapiens has only been around for somewhere in the range 70,000-200,000 years. We are fine-tuned for survival on the Earth of this time frame. However, the Earth is a lot more than 200 kiloyears (Ky) old; the surface formed roughly 4.6 Gy ago (gigayears — 1Gy = 1,000,000,000 years). And we can expect the Earth to persist for about another 3-5 Gy, until the sun leaves the main sequence of the Hertzsprung-Russell diagram and becomes a red giant, presumably swallowing the Earth (or at any rate rendering it too crispy for comfort). So if we're being honest (and not cherry-picking our candidate stellar colony mission targets) we've got a 8-10Gy span to probe.
Back up far enough in the Earth's time-line (before about T minus 4.6 Gy) and we run into the formation of the solar system — and the proto-Earth, before its postulated oblique impact with Theia, a proto-planet roughly the size of Mars — debris from which collision condensed into our moon, Luna. Earth definitely wasn't a candidate for human inhabitation in those days — a largely airless blob of molten rock, under continuous heavy bombardment by planetary embryos and small planetesimals thrown out of the asteroid belt by Jupiter, as the two large gas giants (Jupiter and Saturn) churned the protoplanetary disk. Indeed, just about anywhere in the inner solar system at any time prior to the end of the Late Heavy Bombardment (which ended at T minus 3.8 Gy) is a poor candidate for a space colony — it was the LHB that resurfaced and cratered the moon, and presumably did a similar number on the Earth's then-fragile surface. On the other hand, this period left us geological evidence in the shape of minerals dated to the Hadean aeon. We know relatively little about the Hadean, although there's some recent evidence to support active plate tectonics and a surface temperature compatible with liquid water.
But don't get the idea that late Hadean Earth was a fun place to be. For one thing: that great big moon of ours didn't condense from a debris cloud at its current orbital distance. Tidal dragging is widening the lunar orbit by about 3.8 metres per century; it's now orbits roughly twice as far out as it did when it formed. Which in turn means that the young Earth spun on its axis far faster than it does today, and the tides the newborn molten-faced moon raised during the Hadean aeon would have been something to behold (preferably from a very high altitude). In fact, early Earth was a very alien world indeed. To hit on wikipedia (because I'm feeling too lazy to type):
Recent evidence suggests the oceans may have begun forming by 4.2 Ga.[22] At the start of the Archaean eon, the Earth was already covered with oceans. The new atmosphere probably contained ammonia, methane, water vapor, carbon dioxide, and nitrogen, as well as smaller amounts of other gases. Any free oxygen would have been bound by hydrogen or minerals on the surface. Volcanic activity was intense and, without an ozone layer to hinder its entry, ultraviolet radiation flooded the surface.
We don't know when life got started on Earth. However, we do know that the early history of life relied on anaerobic processes for a surprisingly long time. It wasn't until roughly 580 My ago (at the end of the Proterozoic era) that free oxygen came to dominate our planet's atmospheric chemistry. Back to wikipedia:
The most widely accepted chronology of the Great Oxygenation Event suggests that oxygen began to be produced by photosynthesis by organisms (prokaryotic, then eukaryotic) that emitted oxygen as a waste product. These organisms lived long before the GOE, perhaps as early as 3,500 million years ago. The oxygen they produced would have almost instantly been removed from the atmosphere by the weathering of reduced minerals, most notably iron. This 'mass rusting' led to the deposition of banded iron formations. Oxygen only began to persist in the atmosphere in small quantities shortly (~50 million years) before the start of the GOE. Without a draw-down, oxygen can accumulate very rapidly: at today's rates of photosynthesis (which are admittedly much greater than those in the plant-free Precambrian), modern atmospheric O2 levels could be produced in around 2,000 years.
There are elaborate and fascinating competing theories to explain why it took so long for oxygenation to take off: even the Moon gets blamed (the huge 50 metre tides that churned the anoxic early oceans and would have sucked premature photoautotrophs to their doom deep below: oxygenation had to wait for the moon to migrate out far enough for the tides to die down, permitting photosynthetic organisms to thrive near the surface). The point to take away from this is that, between T minus 4.6 Gy and T minus 0.56 Gy, the Earth's atmosphere was largely free of oxygen. Meat probe says: "aaargh, choke".
Even after the oxygen catastrophe, our space probe isn't going to find a terribly hospitable planet. The sun's luminosity is increasing by around 6% every billion years. The Proterozoic earth was a cold place by our standards; there were ice ages in which glaciation closed in on the equator. And there was continental drift. Continental drift leaves faint traces: we know that 250 My ago, our current continents were united in a single land mass (called Pangaea). There's some evidence that between 800 My and 550 My another single supercontinent (Pannotia or Vendia), and of another mass 1000-850 My ago (Rodina). And there seems to be a correlation between the occurence of "snowball Earth" events (those equator-reaching ice caps) and the unitary supercontinents. Back to wikipedia:
Most paleoclimatologists think the cold episodes had something to do with the formation of the supercontinent Rodinia. Because Rodinia was centered around the equator, rates of chemical weathering increased and carbon dioxide (CO2) was taken from the atmosphere. Because CO2 is an important greenhouse gas climates cooled globally.
Supercontinents too close to the equator: snowball earth. (Meat probe turns blue and shivers.)
So here's the upshot: of the 4.6 Gy of Earth's known history, there's only been enough oxygen in the atmosphere for us to survive for about 0.5 Gy. For roughly 90% of the Earth's history we couldn't even breathe the air. And about 10-25% of the time, there have been ice ages so savagely fierce that the glaciers reached the tropics: odds are good that any meat probe landing on solid ground during these periods would rapidly die of exposure. So historically, Earth has only been inhabitable about 8% of the time — assuming you are lucky enough to find some solid ground. Once you factor in the random surface distribution, we're down to about 2% survivability.
Now let's look at the future of the solar system.
We know the sun is steadily brightening by about 6% per Gy. It's postulated that within a couple of Gy, solar output is going to have some unpleasant effects. Ultraviolet radiation can split the covalent bonds that hold water molecules together, high in the atmosphere: and hydrogen ions (or, more likely, hydrogen molecules) can be blasted right out of the ionosphere by the same mechanism. The slow, steady loss of Earth's water is a one-way process, but exacerbated by warming (more water vapour in the upper atmosphere means more hydrogen is lost). As hydrogen loss proceeds, we end up with a carbon-dioxide dominated atmosphere and a runaway greenhouse effect like that of Venus.
There are other mechanisms that might render the Earth uninhabitable by our kind of life. Over geological time, the partial pressure of oxygen in the atmosphere has risen. With more solar energy inputs, it may be that oxygen levels continue to soar. Above about 28%, even waterlogged biomass will burn handily: and there are indications that atmospheric oxygen (currently down around 16%) has been well over 20% in the past. If oceanic photoautotrophs pump out too much of the stuff, the continents may well be burned back to bedrock by the resulting lightning-triggered fires.
All of this leaves aside the prospects for either an anthropogenic catastrophe, or the evolution (or creation) of new types of chemoautotrophs that have a drastic effect on the Earth's atmospheric chemistry. Or something else. Phase of the moon, perhaps. (What happens when the moon's tidal drag diminishes further, reducing long-term deep-ocean mixing? Anyone got a clue?)
The upshot is, we may well be most of the way through the Earth's inhabitable epoch. In which case, of the 4 Gy remaining, we may have 0.3 — 1 Gy to go with an oxygen-dominated atmosphere and water close to its triple point — the minimum necessary critera for human survival on a planetary surface.
So, back to the gedankenexperiment. Currently, a random meat probe dropped on the Earth's surface has something like a 15% chance of finding it survivable. But a random sampling over the historical epoch would return a survivability probability of around 1%. And over the future epoch, it's likely similar, unless we're erring massively on the side of pessimism about the prospects for our atmospheric composition remaining stable.
Ergo: to a space probe searching for somewhere that our kind of life can thrive, a truly random sampling of the Earth's surface (distributed over both time and area) would probably result in the conclusion that the planet is uninhabitable. See also: Fermi Paradox
Of course, a random sampling of Mars or Pluto would give an even lower probability of finding the planet to be inhabitable. But that's not the point. The point is this: we are finely tuned survival machines that have evolved to survive in a niche on one particular planet in one particular epoch. Even our own planet is unimaginably hostile to our kind of life for most of its history. And while survival outside that niche is possible with the assistance of a horrendously complex toolkit we call "civilization", we've yet to try it somewhere where we can't count on the basics (free oxygen and triple point water).
Also, our odds of finding ET with SETI are depressingly small.
Comment