The Space Reviewin association with SpaceNews

Soace colony illustration
Future human exploration beyond the solar system will depend on the ability to establish space colonies within our solar system. (credit: NASA/Ames)

What future for intelligent life in space?

Bookmark and Share

Thirty-five years ago, Gerard K. O’Neill wrote: “We are so used to living on a planetary surface that it is a wrench for us even to consider continuing our normal human activities in another location” (The High Frontier, p.25). He concluded that the best place for a growing industrial society is not on the Earth, or the Moon or Mars, but “somewhere else entirely”: an array of artificially constructed space colonies.

While a planet is a good place for life to get started, once it has reached the stage of industrial development, its further growth depends on the use of technology to construct artificial space colonies.

Judging from current attitudes, his fundamental insight is still enough of “a wrench” that it has yet to gain wide acceptance. For example, the material published so far by the DARPA-NASA Ames 100-Year Starship Study ignores colonies in space, despite their obvious relevance, as does Lou Friedman’s report on their recent meeting (see “Fly me to the stars”, The Space Review, January 24, 2011). Joy Shaffer’s 2004 essay “Better Dreams” at, enthusiastically referenced by one Space Review commenter, explicitly excludes colonies in space: “there is no need to massively industrialize any place in the solar system beyond the elevator terminals and power stations at geosynchronous orbit”. Even the Tau Zero Foundation focuses on “the ultimate goal of reaching other habitable worlds”.

In this essay I want to revisit some of the fundamentals of O’Neill’s concept and demonstrate their continued relevance. In order to live and function, multicellular creatures such as ourselves need land area with gravity and an atmosphere. A planet represents a highly inefficient means of providing these basic essential services. Mars, for example, has a mass of 6.42x1020 tonnes and a surface area of 1.44x108 square kilometers, thus 4.45 trillion tonnes of matter are required for each square jilometers of habitable area.

For comparison, the space colony design Island One has an estimated total mass of about 3.5 million tonnes (including comprehensive radiation shielding) and offers a pressurised surface area of 1.09 square kilometers for a population of 10,000 people (O’Neill, p.57–59). The space colony therefore requires 3.2 million tonnes of structure for each habitable square kilometer, a mass less than that of Mars by a factor of about a million.

The reason is clearly that in the case of the Earth, Mars, or any other planetary sized body, almost all the mass of the planet serves no function, from the point of view of its surface life, other than to provide surface gravity, and in some cases also a magnetic field and geothermal heat, though smaller bodies such as the Moon provide little or none of these. Planets are low-tech, compressive structures, as indeed they have to be.

An artificial space colony that provides gravity by rotation, on the other hand, must be maintained in tension, requiring conscious design and high-tech materials. In exchange for the investment of engineering effort, the efficiency in the use of raw materials to provide habitable surface area is increased by about a factor of a million (a factor that varies within an order of magnitude each way depending on the particular sizes of space colony and planet being compared).

The conclusion has to be drawn that, while a planet is a good place for life to get started using unconscious means that can evolve spontaneously from the chemical substrate, once life has reached the stage of industrial development, its further growth depends on the use of technology to construct artificial space colonies, which use the material resources of planetary systems at a much higher level of efficiency.

Since, in the O’Neill colony design, about 86 percent of the mass consists of passive radiation shielding, this conclusion is not strongly dependent on the precise proportions of rocks, metals, and volatiles available. Clearly, if adequate protection from cosmic rays can one day be achieved by the use of magnetic fields, the balance swings even further in favor of designed habitats as opposed to natural ones.

The project of sending humans to the stars is absolutely dependent upon prior large-scale space colonization.

In this way, very large future human populations are conceivable. For example, John S. Lewis has reckoned that the material resources of the main asteroid belt, together with large-scale use of solar power, would allow at least 10 million billion people to support themselves (Mining the Sky, p.196). When one adds in the Jupiter trojans and the opportunities presented by the outer solar system, even larger populations become possible. Frank Drake and Dava Sobel have put the overall carrying capacity of our system at “more than a hundred billion billion human beings” (Is Anyone Out There?, p.128), while Marshall Savage suggests an even larger figure (The Millennial Project, p.303). We do not need to quibble over orders of magnitude in order to make the point that the propensity for economic and population growth characteristic of industrial civilization is well matched with the opportunities offered by its local environment, provided that the 21st century sees a shift of the focus of industrial and population growth away from Earth and onto large-scale development of the natural resources of near-Earth space.

Clearly the Moon and Mars will remain the main initial targets for astronaut exploration. But the necessity to protect travelers to and from these worlds from solar storm radiation, as well as the advisability of providing them with a safe haven en route, forces a transport architecture based on well-shielded and well-provisioned—but consequently heavy—cycler stations. In the case of Mars in particular this drives the designer’s attention towards use of materials from near-Earth asteroids, some of which may be volatile-rich, found in orbits that mimic cycler trajectories.

From there it is a short step towards using asteroidal materials to create and supply space habitats which are independent of the Earth-Mars traffic, leading to a gradual expansion of habitable volume spreading out to the main asteroid belt. The resource limits to growth of this kind will not appear for centuries or millennia to come, depending on growth rates over time.

Other advantages mentioned by O’Neill include full-time, full-strength access to solar power, and the option to choose different or zero gravity levels for different purposes. While transport within a single colony or between two adjacent ones would also be easier than on Earth, colonies in widely differing orbits would not necessarily benefit from cheaper transport, though they would still be mutually accessible without having to climb up out of one steep gravitational well and down into another.

With this picture of space colonization based on the material resources of the solar system’s smallest bodies firmly in mind, a number of consequences follow.

First, the project of sending humans to the stars is absolutely dependent upon prior large-scale space colonization. To begin with, the passengers on any interstellar mission will be devoting the rest of their lives to the voyage and the explorations at their destination: a return journey within a human lifetime is hardly conceivable (barring some magical new propulsion technology, and even that is hardly likely to come cheap).

This means that no crewed starship will be dispatched until the viability of a space habitat has been demonstrated for at least one complete human lifetime (including one or more reproductive cycles, unless the starship is conceived as a suicide mission). With space colonization in progress, spurred by general economic and population growth, such a demonstration will be a matter of course, and will be funded by the broader economy. Without it, the demonstration will be an expensive one-off project, and volunteers (together with their yet unborn offspring) will have to renounce all claim to a normal life.

If space agencies want to promote progress in space development, they will need to think more carefully about how they coordinate their exploration with commercial ventures that hope to capitalize on those exploratory activities.

Again, the practicality of interstellar spaceflight varies enormously between robotic and manned programs. Travelling to the Moon, the Surveyor robots weighed about a tonne (including descent propellants), as against the 45 tonnes of the complete Apollo spacecraft. But with sights set on Alpha Centauri, a robotic probe could be very much lighter than those of the 1960s (Lou Friedman talks of “masses less than 10 kilograms”), whereas the irreducible need on a manned vehicle for long-term life-support systems and all the equipment, ancillary vehicles and reserves the crew are likely to need for the rest of their working lives, plus an exotic high-energy propulsion system, will surely push the vehicle mass up towards the order of 10,000 tonnes (25 times the mass of the ISS, or the size of a small oceangoing ship).

Thus the step up in scale from a robotic starship to a manned one looks to be on the order of a factor of a million in terms of mass and hence of energy cost. The absolute minimum energy cost for a 10 kilogram probe with a cruising speed of 0.05c which takes about 80 years to cross to Alpha Centauri is twice its kinetic energy at that speed, or 143 megawatt-years (assuming a perfectly efficient propulsion system!) For even a minimal manned starship weighing say 5,000 tonnes, that energy cost jumps to 71 terawatt-years, equivalent to total current global industrial energy consumption for about five years (and in practice a multiple of these energies will be required when the inevitable inefficiencies of the propulsion system are factored in).

It is therefore effectively inconceivable that a merely planetary economy would ever be rich enough to afford to fuel a single manned starship, unless it went for a slow, multi-generational approach: travelling at 0.005c, our 5,000-tonne starship would take 800 years to make the crossing, with an energy cost about a twentieth of current annual global consumption (plus inefficiencies). But with the millionfold growth in population, and thus in the overall solar system economy, made possible by the transformation of our species from a planet-bound to a space-based one, the energy cost of even a fast manned starship program sinks to a fraction of civilization’s annual energy budget.

In the context of interstellar exploration such as that now being contemplated by the 100-Year Starship Study, a consequence of space colonization within our own solar system is that all main-sequence stars can become targets for colonization by industrial civilizations, from M dwarfs such as Proxima Centauri to A stars such as Sirius and Altair. That assumes that those stars are accompanied by asteroidal rubble similar to our own system’s asteroid and Kuiper belts, which, according to a well-known astronomer of my acquaintance, is indeed very likely.

This acts as a great enabler to manned interstellar spaceflight. Were human star-farers to turn up their noses at any star lacking an Earth analogue planet in a closely Earth-like orbit, they might well have to travel tens or hundreds of light-years in order to reach the nearest acceptable destination. (Obviously we cannot yet rule out Earth-like worlds orbiting the very nearest stars, but given the unexpected variety of the extrasolar planets that have been discovered so far, does anyone really believe that new Earths are likely to be common?)

This means that the quest to find earthlike planets becomes a purely scientific one. Should we identify an extrasolar Earth analogue, complete with indigenous life of its own, its value to us as a target of non-invasive scientific study will be far greater than its value for colonization or resource extraction. The immigrants will be accustomed to living permanently in artificial space structures, and will have no reason to change their lifestyle after arrival, though visits to planetary surfaces for science and recreation will no doubt take place. The scenario of James Cameron’s movie Avatar, which posits human star-travelers coming into conflict with indigenous inhabitants of such a planet, is therefore not a realistic one.

A final consequence is that if space agencies want to promote progress in space development, whether for economic or scientific reasons, or both, they will need to think more carefully about how they coordinate their exploration with commercial ventures that hope to capitalize on those exploratory activities. While it is excellent news that Space Adventures has announced the purchase of more Soyuz seats for fare-paying passengers on trips to the ISS and even around the Moon, this is still happening more in spite of space agency planning than as a result of any coherent strategy to make hardware and technologies purchased at considerable expense with taxpayers’ money useful and profitable in wider society.

Space agencies alone cannot create the complex, dynamic space economy implied by the vision of large, permanent extraterrestrial human populations, which is why NASA and ESA in particular need to pay closer attention to fostering commercial growth in the difficult space environment—starting, obviously, with those companies focused on getting the cost per seat to orbit down, and the traffic level up.

To the best of current knowledge, almost all the matter and energy in the galaxy that could, in principle, be used to support life and civilization is found in planetary systems that likely have either no indigenous life at all or only microbiotic life. Given that space colonization is undoubtedly technically possible (the unhappy experience of the mismanaged Biosphere 2 project notwithstanding), and that interstellar flight is a logical outlet for the energies of an interplanetary-scale civilization, it follows that life in the galaxy will tend over time to become dominated by expansionary, space-based civilizations.

Our own species therefore stands at a unique crossroads in its career: whether to perfect these technologies and move outwards from Earth, effectively making its heritage immortal through a variety of far-flung successor civilizations, each one orders of magnitude larger than that possible on Earth alone, or whether to retreat from a high-tech future and face ultimate decline and extinction on Earth.

Gerard O’Neill’s strategy of economically self-sustaining space colonisation, first published in 1976, is still the touchstone by which future plans for space exploration and development must be judged.