[I wrote this in 1998 for the late Chris Boyce’s Et-Presence website]
When we ask ourselves this question about other sophonts [“wise beings”] that might share our Universe perhaps we don’t fully appreciate just how important location might be for the fortunes of an intelligent race and our chances of meeting them. We’ve become used to the spanning of vast distances in mere moments thanks to “Star Wars” and “Star Trek”, while in the real world Einstein’s laws still refuse to budge. Virtually every star-system in the endless quest of the “Voyager” has an intelligent species, or at least an M class planet, but in the real Galaxy such might be few and far between. If this is the case then journeys between habitable worlds, let alone civilisations, might take centuries and successful colonisation might be difficult beyond imagining. How can we estimate the relevant factors? And what are the best estimates?
A habitable planet suitable for immediate colonisation by humans might be a chimaera, since such a world would possess an oxygen-rich atmosphere and consequentially an advanced biosphere that might be incompatible with our own. But inhabitants of such planets won’t have such difficulties, so how common are habitable worlds? We currently have no direct evidence for the existence of any planet smaller than Neptune around a Sun-like star, though a few smaller planets are known to orbit pulsars. A few years from now might see a change in our knowledge, as telescopic studies are now possible that can reveal planets via micro-lensing events. A space-based telescope could potentially survey hundreds of thousands of stars over several years and produce valuable statistics on the frequency and diversity of planetary systems. So perhaps estimates are too premature, but educated guesses can give us some sense of the problem before we look directly.
Our best theories of planetary formation involve the accretion of planets from an orbiting dust-disk around a newly forming star or stars. Recent studies have shown dust disks can form around binary star-systems, while telescopic surveys have revealed a wealth of dust-disks around embryonic stars and some remnant disks that form rings around a star-centred space now possibly occupied by planets. Planet formation does not seem to be difficult, but will those planets be suitable for life? The best current theories involve accretion of terrestrial planets from the dust disk over millions of years, while Jupiter scale planets potentially form rapidly via gravitational instabilities. Computer accretion modelling produces a multitude of asteroid-sized bodies colliding chaotically over tens of millions of years, finally forming a group of planets similar in size and number to the present terrestrial planets, though in more erratic orbits. Detailed modelling reveals that the final collision phases can also produce planet-moon systems like the Earth and the Moon fairly frequently.
So our best models replicate our solar system reasonably frequently. In spite of appearances the freakish “Hot Jupiters” that have been found around other stars are just that, freakish, involving 5 % of stars. They seem to form around stars richer in “metals” (C, N, O etc..) than our Sun. More well behaved systems are probably more common, just harder to detect. For example, Lalande 21185 possesses two Jovian planets in reasonable orbits, according to astronomer George Gatewood. But once we have found sets of planets around stars, how frequently will they prove to be habitable? Of four terrestrial planets in our own system one is habitable [Earth] and another marginal [Mars]. Both have the correct rotation rate for life, but Mars is too small to sustain the geochemical cycling that long-term biospheres need.
Let’s look at the statistics in our system:
Right Rotation: 2/4
Right Mass: 2/4
Possesses a large Moon: 1/4
Right distance from the Sun: 1/4
Point 3, a large Moon, is probably required to maintain stability of a planet’s axial inclination over geological time, else the planet will tip over chaotically and produce weather extremes incompatible with advanced life forms. The other points are not-controversial characteristics required for habitability. If we invoke the Maximum Likelihood Principle, that what we see is representative of what there is elsewhere, and use these solar frequency ratios as probabilities, then the chance of a habitable planet around a suitable star is ~ 0.015. The probability might be lower because a planet with an oxygen-rich/carbon dioxide-poor atmosphere is climatically unstable – closer to its star and it suffers a run-away greenhouse [in Earth’s case closer than 0.95 AU], further away and it suffers runaway glaciation [further away than 1.1 AU.] Large planets with dense carbon dioxide atmospheres can remain climatically stable over a broader range, and the range increases if methane is present. Neither gas can exist abundantly with high oxygen levels however.
Could advanced alien life exist in a non-oxygen atmosphere? Low oxygen environments with “advanced life” are known on Earth, but none are known that support vertebrates. And more importantly low oxygen levels are incompatible with the first tool of advanced technology – fire. While alien animal life will be interesting to study, we’re looking for neighbours not just life forms. I can’t dogmatically say intelligent life can’t exist in a non-oxygenated atmosphere, but I’ll bet that it probably doesn’t. And we certainly can’t exist in such an environment, so the restriction on habitable planets still applies – 0.95 AU – 1.1 AU, at Earth equivalent insolation levels. If terrestrial planets can be said to range between 0.4 – 5 AU, around a Solar-mass star, so the probability of an individual planet occupying the habitable zone will be ~ 0.03, rather than 0.25. In a system of four planets the probability of a planet occupying the habitable zone will be ~ 0.115. Overall the probability of a habitable planet around a suitable Sun-like star is ~ 0.72%.
The kind of star a planet orbits is also vitally important. Stellar luminosity is strongly tied to the star’s mass – a star twice as massive as the Sun burns perhaps 16 times faster, while a star half as massive burns 16 times slower, with corresponding effects on life-times of the stars. Related to this fact is the question of just how long a complex biosphere, capable of generating an oxygen-rich atmosphere and intelligent life, takes to evolve. In the Earth’s case oxygen levels reached current day levels after about 3.7 billion years, coinciding with the appearance of complex animal life. Prior to this time various chemical reactions involving dissolved metals and sulphur probably kept oxygen levels low, and various geological processes kept the continents mostly underwater. It seems a long period of geological evolution preceded animal life’s appearance, and once it appeared it took ~ 600 million years to evolve human-level intelligence. If a star lasts on the Main Sequence for less than ~ 4 – 5 billion years, then it probably won’t develop intelligent life. So stars have to be roughly less than 1.3 Solar masses.
Another restriction on the type of star involves tidal forces. Radiation received from a star follows an inverse square law, while its tidal forces follow an inverse cube law. Luminosity of stars follows a roughly fifth power law. So around smaller stars a habitable planet has to be really close to its star, probably so close that its rotation will eventually become tidally locked with its star, like our Moon is locked to the Earth. Stars of lower luminosity also tend to be cooler than brighter stars, and so their spectrums are shifted into lower frequencies making them redder. This may preclude high-energy photochemistry required for photosynthesis able to liberate oxygen from water. Combining these two factors suggest that habitable planets exist around stars of mass greater than 0.8 solar masses.
That’s quite a restrictive range, and it means only about 2% of stars are suitable, and combined with the rough estimates of 0.72% of a system’s terrestrial planets being suitable, it means that only ~ 1/7000 of stars will have habitable planets. Less if certain kinds of binaries are excluded from having planets, which may make the probability 1/14,000. How far apart would they be on average? In our part of the Galaxy stars are fairly thinly spread out – between one per 300 to 600 cubic light years – so habitable planets will be very roughly 128 – 161 light years apart. To further complicate matters not all stars are the same age. If Sun-like stars are randomly distributed in age in a Galaxy of 10 billion years [10 Gyr], then suitable stars will be at most 60% of the total, and probably more like 20% [probable lifetime of the oxygenated biosphere.] Then habitable planets will be 151 – 275 light years apart.
My estimates are rather rough, but I am using what we know at present, and I think that an estimate of 1/12,000 – 1/70,000 stars with habitable planets is reasonable. That’s still between 17 million and 3 million habitable planets in our Galaxy. But the Galaxy is a big place. What it means for our search for neighbours is that we’ll have to look further than “Star Trek” implies. Using present day technology, nuclear-pulse starships capable of a delta-vee of 0.1c are possible. If magnetic braking, as Robert Zubrin proposes, is used then that will be the cruise velocity. To reach another habitable planet could then take up to 2750 years, which is humanly impossible. More advanced star-drives, using beamed propulsion, could achieve cruising speeds of 0.3 – 0.9 c, trips about 920 to 305 years long, which are still rather beyond human possibility.
Life extension is becoming a medical possibility, and reversal of aging is also looking possible, but interstellar travel will still occupy a large portion of foreseeable lifetimes. Human brains probably saturate in information storage after ~ 1000 years, so a long interstellar journey would still be a great cost to even long-lived humans. Suspended animation may be possible, but no one yet knows how it might be done. Perhaps nanotechnology will allow human bodies to be locked up in some sort of crystal form to halt biology totally, just like brine shrimp manage to do for centuries at a time. In space there may be a limit on just how long suspension can last due to genetic damage from cosmic rays, but 1000 years might be possible. But there may be easier targets closer to home for colonisation.
In our own system we find half the inner planets are suitable for relatively easy terraforming. One, our own Earth, was “terraformed” by cyanobacteria gigayears ago. Mars today requires extensive transformation, but if it were larger and warmer terraforming would be easy. Non-oxygenated planets with liquid water are probably quite common. Even planets of redder stars might be suitable, which would increase our chances of finding worlds to transform. Stars massing between 0.5 solar masses and 1.5 solar masses might be suitable. One estimate suggests that 15% of stars will have suitable planets for terraforming, which implies about 30 – 100 billion suitable worlds in our Galaxy. By the time humans encounter a truly habitable planet they may have already colonised ~ 5,000 planets, many of which will have anoxic biospheres of their own. By that stage complex life might seem extremely precious, worth preserving at any cost. With 5,000 worlds worth of terraforming experience the human race can probably afford to by-pass the star systems of habitable planets entirely, leaving behind robotic observation platforms.
More aggressive colonisation programs might involve launching smaller robotic vehicles capable of carrying its colonists as DNA codes and computer software. Once a colony is established the probe will self-replicate and launch off copies towards new stars. Such a colonising program could achieve effective speeds approaching the speed-of-light, if the replication process is rapid, so the Galaxy might be colonised in ~ 135,000 years. Frank Tipler has used this argument against the existence of any other sophonts in our Galaxy, perhaps even the Universe, but such an attitude would be premature. A self-replicating coloniser will itself need to be a sophont, doubtless programmed with ethical concerns similar to its creators, and after exploring many thousands of stars it too may realise the rarity of complex life. The “fleet” of colonisers may communicate amongst themselves, so they may have a social existence and develop an ethic of their own – where animal life is found, observe.
Our solar system is a big and quite rich place. Resources freely available in space seem to get more abundant the further from the Sun we get. First we have the Near Earth Objects, then the Main Belt, the Jovian Trojan asteroids some 5 times more abundant than the Main Belt, and at the very fringes, the Kuiper Belt, which has tens of thousands of cometoids larger than 100 kilometres across. Compared with mere hundreds of Main Belt asteroids of that size, the Kuiper Belt is massive. This would provide plenty of room and material for alien starbases to utilise, and while sunlight would be dim, it would still provide useable energy. Kuiper Belts are doubtless more common than habitable planets, perhaps even planets themselves, and would be logical sites for most scientific operations in any star system. Our own might be home to hundreds of alien bases and we would be none the wiser.
Are they there, virtually on our doorsteps? If we are to believe the hundreds of reports of UFOs every year then we can be fairly certain They are, but Their behaviour is hardly friendly or intelligently covert. UFOs seem to behave in ways that enhance belief more than reason, gullibility more than scepticism. Or at least the UFOs that seem to be trying to contact us fit that category – they tantalise rather than illuminate. But some high-altitude UFOs which move rapidly through our skies, and that don’t seem to be mere fireballs, may be just passing through. Why pass by this little planet, since high-speed tours seem unlikely? Perhaps there are means of travelling faster-than-light that require gravitational anchoring to suitable planets, some sort of wormhole stargate. A habitable planet provides a relatively friendly place to crash-land in an emergency and a potential destination once proper contact is made. That may be, but I am sceptical of any beings that intrude so strangely into our lives, and most UFO reports seem to be of quite mundane phenomena anyway. I’ve seen enough strange lights in the sky and if they’re neighbours I wish they’d just stop and say hello.
Galactic colonisation takes a far shorter period of time than the lifetime of the Galaxy itself. Even more pessimistic estimates imply a colonisation time of about 100 million years, an expansion speed of 0.1% of c. In such scenarios colonisers travel out at 0.1c and then take ~ 900 years to send out their own colonisers, which may be pessimistic. Replicating robotic colonisers might spread at 0.9 c, some 900 times faster. Some writers have suggested that astrophysical processes might limit the expansion of civilisations. An extreme version of this argument was advanced by Duncan Lunan in1986. He proposed that we might be experiencing a brief period of high-energy cosmic radiation. Being so brief, older space-faring civilisations may have evolved in a more benign Galactic environment, and be totally unprepared for high-energy cosmic radiation. A wave of slow death might spread unforseen through the Galaxy, wiping out space-based civilisations and leaving the Galaxy relatively untouched by sophonts. A more recent argument by James Annis  suggests that gamma-ray bursts [GRBs] might effectively reset the evolution clock on inhabited planets, by killing off terrestrial life and damaging the marine biosphere. GRBs have the potential to destroy half the ozone layer of any planet within tens of thousands of light years, so this is plausible. Annis estimates that evolution of intelligence and galactic colonisation takes ~ 200 million years, and only “recently” in Galactic time have GRBs bursts decreased in frequency sufficiently to allow intelligence time to evolve. Sophonts through out the Galaxy might be about to launch forth. Annis uses a simple exponential decay for GRB frequency and assumes Galaxy-wide devastation for each burst, so his argument may be overly simplistic. Recent studies postulate that GRBs focus their energy in a beam, rather than putting it out in one big flash, which suggests the threat is much more localised and less damaging than first imagined.
My personal theory involves a combination of factors – Galactic chemical evolution, GRBs, stellar density and the intricacy of intelligence itself. The Galaxy gets denser the closer to the Galactic Core one is, and this has certain life-unfriendly consequences. More stars packed closer together means more frequent stellar fly-bys and disruption of comet-clouds. Even closer to the core, and Outer planets become unstable in their orbits. While that may cause disruption of Inner planets it also can disrupt a system’s Kuiper Belt, causing a shower of planet-killing cometoids. A Kuiper Belt cometoid 100 km across, impacting with the Earth, would boil the oceans and produce a greenhouse effect that would raise temperatures to 2000 K. This becomes very likely towards the Core. Another factor is the rise in metallicity of stars as the Core is approached. As recent finds of “Hot Jupiters” are suggesting, higher metallicity stars produce more massive planets and these interact violently, throwing planets out of the system or causing them to fall towards their star. Close-in planets may then cause super-flares that briefly bake the planets of the star system. Very unfriendly place to live. Metallicity decreases away from the Galactic Core, resulting in lighter preplanetary nebulae, and lighter planets. The effect of this is as yet unknown, but it may result in fewer Jovians. In our solar system’s history Jupiter acted as a confining influence on the Inner system – perhaps bigger terrestrials need a Jupiter to form? Combining these factors might mean that habitable planets are found in a ring of roughly the same age and distance from the Galactic Core. Sophonts may then have started evolving at roughly the same time. This would explain why no Galaxy spanning engineering projects have yet been seen. However in Intergalactic space the mysteriously dim Low Surface Brightness galaxies might be undergoing transformation, their stars steadily being wrapped in Dyson Shells.
If truly Earth-like worlds are rare I wonder if we will attempt the long journey between stars. Only a mature space-based civilisation will be likely to attempt journeys lasting hundreds of years, even by probes. After arriving in other star systems will they decide to terraform or remain space-based? A mixture of both strategies would be reasonable, since planets have much longer life-times than built structures, but the motivations of such a civilisation is hard to imagine. Stability might be more important than the urge to expand. To fill the Galaxy with one’s species seems more like a Crusade than a logical choice. Curiosity may be the main motivation which might cause sophonts to opt for the minimal self-replicating robot approach, touching lightly on many star-systems, and setting up a Galactic communications web.
In all this the only way to really know is to go and look for ourselves. We might find space-junk that has drifted into our star system – indeed some anomalous and very ancient “artefacts” might be just that. Also some meteor plasma trails have shown colouration consistent with complex alloys unlikely to occur in nature. We might find mining scars on asteroids, or abandoned scientific bases. We might find colonies in the Kuiper Belt, or floating in the atmospheres of the Jovians. Or we might scour the Galaxy to find Them, struggling on their homeworlds to become civilisations capable of Galactic expansion. We are still at that stage, and I wonder what They’re thinking if They’re here.
Would we recognise their presence? Our bodies have evolved according to the blind guide of natural selection and aren’t designed for travelling the immensity of the Galaxy. I’ve mentioned nanotechnology as a means of suspended animation, but it could provide something more – total independence from undesigned biology. If the transfer of biological minds from the wetware of neurones to some more suitable hardware is achieved, then smaller physical forms might be possible or even existence as virtual entities in a computer. The absolute limit to miniaturising human-level intelligence is not yet known, but it might make very small starships possible – kilograms rather than kilotons. In 1991 the Earth was “buzzed” by a small and unusual asteroid, 1991 VG. According to astronomer Duncan Steele its light-curve showed all signs of it being metallic, even manufactured, but its orbit matches no known Earth vehicle of the estimated size ~ 10 – 20 metres. Perhaps it is a starship observing us, manned by “compressed” sophonts. NASA should send a probe – to intercept or at least fly-by – from the Deep Space series. DS 9, maybe?