I grew up on generation ship stories. I’m thinking of tales like Heinlein’s Orphans of the Sky, Aldiss’ Non-Stop, and (from my collection of old science fiction magazines) Don Wilcox’s “The Voyage that Lasted 600 Years.” The latter, from 1940, may have been the first generation ship story ever written. The idea grows out of the realization that travel to the stars may take centuries, even millennia, and that one way to envision it is to take large crews who live out their lives on the journey, their descendants becoming the ones who will walk on a new world. The problems are immense, and as Alex Tolley reminds us in today’s essay, we may not be fully considering some of the most obvious issues, especially closed life support systems. Project Hyperion is a game attempt to design a generation ship, zestfully tangling with the issues involved. The Initiative for Interstellar Studies is pushing the limits with this one. Read on.

by Alex Tolley

“Space,” it says, “is big. Really big. You just won’t believe how vastly, hugely, mindbogglingly big it is. I mean, you may think it’s a long way down the road to the chemist, but that’s just peanuts to space..” The Hitchhikers Guide to the Galaxy – Douglas Adams.

Detail of book cover of Tau Zero. Artist: Manchu.

Introduction

Science fiction stories, and most notably TV and movies, avoid the fact of the physical universe being so vast by various methods to reduce travel time. Hyperspace (e.g. Asimov’s galaxy novels), warp speed (Star Trek), and wormholes compress travel time to be like that on a planet in the modern era. Relativity will compress time for people at the cost of external time (e.g. Poul Anderson’s Tau Zero). But the energy cost of high-speed travel has a preferred slow-speed method – the space ark. Giancarlo Genta classifies this method of human crewed travel: H1 [1]. a space-ark type, where travel to the stars will take centuries. Either the crew will be preserved with cryosleep on the journey (e.g. Star Trek TOS S01E22, “Space Seed”) or generations will live and die in their ship (e.g. Aldiss Non-Stop) [2].

This last is now the focus of the Project Hyperion competition by the Initiative for Interstellar Studies (i4is) where teams are invited to design a generation ship within various constraints. This is similar to the Mars Society’s Design a City State competition for a self-supporting Mars city for 1 million people, with prizes awarded to the winners at the society’s conference.

Prior design work for an interstellar ship was carried out between 1973 and 1978, by the British Interplanetary Society (BIS). It was for an uncrewed probe to fly by Barnard’s star 5.9 lightyears distant within 50 years – their Project Daedalus.[3] Their more recent attempt at a redesign, the ironically named Project Icarus, failed to achieve a complete design, although there was progress on some technologies [4] Project Hyperion is far more ambitious based on the project constraints including human crews and a greater flight duration .[5].

So what are the constraints or boundary conditions for the competition design? Seven are given:

1. The mission duration is 250 years. In a generation ship that means about 10 generations. [Modern people can barely understand what it was like to live a quarter of a millennium ago, yet the ship must maintain an educated crew that will maintain the ship in working order over that time – AT].

2. The destination is a rocky planet that will have been prepared for colonization by an earlier [robotic? – AT] ship or directed panspermia. Conditions will not require any biological modifications of the crew.

3. The habitat section will provide 1g by rotation.

4. The atmosphere must be similar to Earth’s. [Presumably, gas ratios and partial pressures must be similar too. There does not appear to be any restriction on altitude, so presumably, the atmospheric pressure on the Tibetan plateau is acceptable. – AT]

5. The ship must provide radiation protection from galactic cosmic rays (GCR).

6. The ship must provide protection from impacts.

7. The crew or passengers will number 1000 +/- 500.

The entering team must have at least:

  • One architectural designer
  • One engineer
  • One social scientist (sociologist, anthropologist, etc.)

Designing such a ship is not trivial, especially as unlike a Lunar or Martian city, there is no rescue possible for a lone interstellar vehicle traveling for a quarter of a millennium at a speed of at least 1.5% of lightspeed to Proxima, faster for more distant stars. If the internal life support system fails and cannot be repaired, it is curtains for the crew. As the requirements are that at least 500 people arrive to live on the destination planet, any fewer survivors, perhaps indulging in cannibalism (c.f. Neal Stephenson’s Seveneves), means this design would be a failure.

Figure 1. Technology Readiness Levels. Source NASA.

The design competition constraints allow for technologies at the technology readiness level of 2 (TRL 2) which is in the early R&D stage. In other words, the technologies are unproven and may not prove workable. Therefore the designers can flex their imaginations somewhat, from propulsion (onboard or external) to shielding (mass and/or magnetic), to the all-important life support system.

Obviously, the life support system is critical. After Gerry O’Neil designed his space colonies with interiors that looked rather like semi-rural places on Earth, it was apparent to some that if these environments were stable, like managed, miniature biospheres, then they could become generation ships with added propulsion. [6]. The Hyperion project’s 500-1500 people constraint is a scaled-down version of such a ship. But given that life support is so critical, why are the teams not required to have members with biological expertise in the design? Moreover, the design evaluators also have no such member. Why would that be?

Firstly, perhaps such an Environmental Control and. Life Support System (ECLSS) is not anticipated to be a biosphere:

The requirements matrix states:

ECLSS: The habitat shall provide environmental control and life support: How are essential physical needs of the population provided? Food, water, air, waste recycling. How far is closure ensured?

Ecosystem: The ecosystem in which humans are living shall be defined at different levels: animals, plants, microbiomes.

This implies that there is no requirement for a fully natural, self-regulating, stable, biosphere, but rather an engineered solution that mixes technology with biology. The image gallery on the Hyperion website [5] seems to suggest that the ECLSS will be more technological in nature, with plants as decorative elements like that of a business lobby or shopping mall, providing the anticipated need for reminders of Earth and possibly an agricultural facility.

Another possibility is that prior work is now considered sufficient to allow it to be grafted into the design. It seems the the problem of ECLSS for interstellar worldships is almost taken for granted despite the very great number of unsolved problems.

As Soilleux [7] states

“To date, most of the thinking about interstellar worldships has focused, not unreasonably, on the still unsolved problems of developing suitably large and powerful propulsion systems. In contrast, far less effort has been given to the equally essential life support systems which are often assumed to be relatively straightforward to construct. In reality, the seductive assumption that it is possible to build an artificial ecosystem capable of sustaining a significant population for generations is beset with formidable obstacles.”

It should be noted that no actual ECLSS has proven to work for any length of time even for interplanetary flight, let alone centuries for interstellar flight. Gerry O’Neill did not make much effort beyond handwaving for the stability of the more biospheric ECLSS of his 1970s-era space colony. Prior work from Bios 3, Biosphere II, MeLISSA, and other experiments, has demonstrated very short-term recycling and sustainability, geared more towards longer duration interplanetary spaceflight and bases. Multi-year biospheres inevitably lose material due to imperfect recycling and life support imbalances that must be corrected via venting. The well-known Biosphere II project couldn’t maintain a stable biosphere to support an 8-person crew for 2 years, accompanied by a 10% per year atmosphere loss.

On a paper design basis, the British Interplanetary Society (BIS) recently worked on a more detailed design for a post-O’Neill space colony called Avalon. It supported a living interior for 10,000 people, the same as the O’Neill Island One design, providing a 1g level and separate food growing, carbon fixing, and oxygen (O2) generating areas. However, the authors did not suggest it would be suitable for interstellar flight as it did require some inputs and technological support from Earth. Therefore it remains an idea, and as with the hardware experiments to date, such ECLSSs remain at low TRL values. While the BIS articles are behind paywalls, there is a very nice review of the project in the i4is Principium magazine [7].

Table 1. Inputs to support a person in Space. Source: Second International Workshop on Closed Ecological Systems, Krasnoyarsk, Siberia, 1989 [8]

Baseline data indicates that a person requires 2 metric tonnes [MT] of consumables per year, of which food comprises 0.2 MT and oxygen (O2) 0.3 MT. For 1000 people over 250 years that is 500,000 MT. To put this mass in context, it is about the greatest mass of cargo that can be carried by ships today. Hence recycling to both save mass and to manage potentially unlimited time periods makes sense. Only very short-duration trips away from Earth allow for the consumables to be brought along and discarded after use, like the Apollo missions to the Moon. Orbiting space stations are regularly resupplied with consumables, although the ISS is using water recycling to save on the resupply of these consumables for drinking, washing, and bathing.

If an ECLSS could manage 100% recycling, then if the annual amount is the buffer to allow for annual crop growth, the ship could support the crew over the time period with just over 2000 MT, but of course with the added requirement for power and infrastructure and replacement parts to maintain the ECLSS.

Figure 2. Conceptual ECLSS process flow diagram for the BIS Avalon project.

A review of the Avalon project’s ECLSS [7] stated:

“A fully closed ECLSS is probably impossible. At a minimum, hydrogen, oxygen, nitrogen, and carbon lost from airlocks must be replaced as well as any nutrients lost to the ecosystem as products (eg soil, wood, fibres, chemicals). This presents a resource-management challenge for any fully autonomous spacecraft and an economic challenge even if only partially autonomous…Recycling does not provide a primary source of fixed nitrogen for new plant growth and significant quantities are lost to the air as N2 because of uncontrollable denitrification processes. The total loss per day for Avalon has been estimated at 43 kg to be replaced from the atmosphere.”

[That last loss figure would translate to 390 MT for the 1/10 scaled crewed Hyperion starship over 250 years.]

Hempsell evaluated the ECLSS technologies for star flight based on the Avalon project and found them unsuited to a long-duration voyage [9]. It was noted that the Avalon design still required inputs from Earth to be maintained as it was not fully recycled:

“So effective is this transport link that it is unlikely Earth orbiting colonies will have any need for the self-sufficiency assumed by the original thinking by O’Neill. It is argued that a colony’s ability to operate without support is likely to be comparable to the transport infrastructure’s journey times, which is three to four months. This is three orders of magnitude less than the requirements for a world ship. It is therefore concluded that in many critical respects the gap between colonies and world ships is much larger than previous work assumed.”

We can argue whether this will be solved eventually, but clearly, any failure is lethal for the crew.

One issue is not mentioned, despite the journey duration. Over the quarter millennium voyage, there will be evolution as the organisms adapt to the ship’s environment. Data from the ISS has shown that bacteria may mutate into more virulent pathogens. A population living in close quarters will encourage pandemics. Ionizing radiation from the sun and secondaries from the hull of a structure damages cells including their DNA. 250 years of exposure to residual GCR and secondaries will damage DNA of all life on the starship.

However, even without this direct effect on DNA, the conditions will result in organisms evolving as they adapt to the conditions on the starship, especially the small populations, increasing genetic drift. This evolution, even of complex life, can be quite fast, as the continuing monitoring of the Galápagos island finches observed by Darwin attests. Of particular concern is the creation of pathogens that will impact both humans and the food supply.

In the 1970s, the concept of a microbiome in humans, animals, and some plants was unknown, although bacteria were part of nutrient cycling. Now we know much more about the need for humans to maintain a microbiome, as well as some food crops. This could become a source of pathogens. While a space habitat can just flush out the agricultural infrastructure and replace it, no such possibility exists for the starship. Crops would need to be kept in isolated compartments to prevent a disease outbreak from destroying all the crops in the ECLSS.

If all this wasn’t difficult enough, the competition asks that the target generation population find a ready-made terrestrial habitat/terraformed environment to slip into on arrival. This presumably was prebuilt by a robotic system that arrived ahead of the crewed starship to build the infrastructure and create the environment ready for the human crew. It is the Mars agricultural problem writ large [10], with no supervision from humans to correct mistakes. If robots could do this on an exoplanet, couldn’t they make terrestrial habitats throughout the solar system?

Heppenheimer assumed that generation ships based on O’Neill habitats would not colonize planets, but rather use the resources of distant stars to build more space habitats, the environment the populations had adapted to [6]. This would maintain the logic of the idea of space habitats being the best place for humanity as well as avoiding all the problems of trying to adapt to living on a living, alien planet. Rather than building space habitats in other star systems, the Hyperion Project relies on the older idea of settling planets, and ensuring at least part of the planet is habitable by the human crew on arrival. This requires some other means of constructing an inhabitable home, whether a limited surface city or terraforming the planet.

Perhaps the main reason why both the teams and evaluators have put more emphasis on the design of the starship as a human social environment is that little work has looked at the problems of maintaining a small population in an artificial environment over 250 years. SciFi novels of generation ships tend to be dystopias. The population fails to maintain its technical skills and falls back to a simpler, almost pretechnological, lifestyle. However, this cannot be allowed in a ship that is not fully self-repairing as hardware will fail. Apart from simple technologies, no Industrial Revolution technology that continues to work has operated without the repair of parts. Our largely solid-state space probes like the Voyagers, now over 45 years old and in interstellar space, would not work reliably for 250 years, even if their failing RTGs were replaced. Mechanical moving parts fail even faster.

While thinking about how we might manage to send crewed colony ships to the stars with the envisaged projections of technologies that may be developed, it seems to me that it is rather premature given the likely start date of such a mission. Our technology will be very different even within decades, obsoleting ideas used in the design. The crews may not be human 1.0, requiring very different resources for the journey. At best, the design ideas for the technology may be like Leonardo Da Vinci’s ideas for flying machines. Da Vinci could not conceive of the technologies we use to avoid moving our meat bodies through space just to communicate with others around the planet.

Why is the population constraint 1000 +/- 500?

Gerry O’Neill’s Island One space colony was designed for 10,000 people, a small town. The BIS Avalon was designed for the same number, using a different configuration that provided the full 1g with lower Coriolis forces. The Hyperion starship constrains the population to 1/10th that number. What is the reason? It is based on the paper by Smith [14] on the genetics and estimated safe population size for a 5-generation starship [125 years at 25 years per generation?]. This was calculated on the expected lower limit of a small population’s genetic diversity with current social structures, potential diseases, etc. This is well below the number generally considered for long-term viable animal populations. However, the calculated size of the genetic bottleneck in our human past is about that size, suggesting that in extremis such a small population can recover, just as we escaped extinction. Therefore it should be sufficient for a multi-generation interstellar journey.

From Hein[14]:

While the smallest figures may work biologically, they are rather precarious for some generations before the population has been allowed to grow. We therefore currently suggest figures with Earth-departing (D1) figures on the order of 1,000 persons.

But do we need this population size? Without resorting to a full seed ship approach with the attendant issues of machine carers raising “tank babies”, can we postulate an approach without assuming future TRL9 ECLSS technology?

As James T Kirk might say: “I changed the conditions of the test competition!”

Suppose we only need a single woman who will bear a child and then age until death, perhaps before 100. The population then would be a newborn baby, the 25-year-old mother, the 50-year-old grandmother, the 75-year-old great-grandmother, and the deceased great-great-grandmother. 3 adults and a newborn. At 20, the child will count as an adult, with a 45-year-old mother, a 70-year-old grandmother, and possibly a 95-year-old great-grandmother, 4 adults. Genetic diversity would be solved from a frozen egg, sperm, and embryo bank of perhaps millions of individual selected genomes. The mother would give birth only to preselected, compatible females using implantation (or only female sperm) to gain pregnancy, to maintain the line of women.

This can easily be done today with minimal genetic engineering and a cell-sorting machine [11]. At the destination, population renewal can begin. The youngest mature woman could bear 10 infants, each separated by 2 years. The fertile female population would rapidly increase 10-fold each generation, creating a million women in 6 generations, about 150 years. Each child would be selected randomly from the stored genetic bank of female sperm or embryos. At some point, males can be phased in to restore the 50:50 ratio. Because we are only supporting 4 women, almost all the recycling for food and air can be discarded, just recycling water. Water recycling for urine and sweat has reached 98% on the ISS, so we can assume that this can be extended to the total water consumption. Using a conservative 98% recycle rate for water (sanitary, drinking, and metabolic), O2 replenishment from water by electrolysis (off-the-shelf kit) and an 80% recycle rate of metabolic carbon dioxide (CO2) to O2 and methane (CH4 by a Sabatier process), implies that the total water storage needs only be 5x the annual consumption. The water would be part of the hull as described by McConnell’s “Spacecoach” concept.

Based on the data in Table 1 above, the total life support would be about 125 MT. The methane could be vented or stored for the lander propulsion fuel. For safety through redundancy, the total number of women population might be raised 10x to 30-40, requiring only 1250 MT total consumables. All this does not require any new ECLSS technology, just enough self-repair, repairable, and recyclable machinery, to maintain the ship’s integrity and functioning. The food would be stored both as frozen or freeze-dried, with the low ambient temperature maintained by radiators exposed to interstellar space. This low mass requires no special nutrient recycling, just water recycling at the efficiency of current technology, O2 recycling using existing technology, and sufficient technology support to ensure any repairs and medical requirements can be handled by the small crew.

Neal Stephenson must have had something like this in mind for the early period of saving and rebuilding the human population in his novel Seveneves when the surviving female population in space was reduced to 7. Having said this, such a low population size even with carefully managed genetic diversity through sperm/ar embryo banks, genetic failure over the 10 generation flight may still occur. Once the population has the resources to increase, the genetic issues may be overcome by the proposed approach.

Table 2. ETHNOPOP simulation showing years to demographic extinction for closed human populations. Bands of people survive longer with larger starting sizes, but these closed populations all eventually become extinct due to demographic (age & sex structure) deficiencies. Source Hein [14]

From table 2 above, even a tiny, natural,human population may survive hundreds of years before extinction. Would the carefully managed starship population fare better, or worse?

Despite this maneuver, there is still a lot of handwavium about the creation of a terrestrial habitable environment at the destination. All the problems of building an ECLSS or biosphere in a zone on the planet or its entirety by terraforming, entirely by a non-human approach, are left to the imagination. As with humans, animals such as birds and mammals require parental nurturing, making any robotic seedship only partially viable for terraforming. If panspermia is the expected technology, this crewed journey may not start for many thousands of years.

But what about social stability?

I do applaud the Hyperion team for requiring a plan for maintaining a stable population for 250 years. The issue of damage to a space habitat in the O’Neill era was somewhat addressed by proponents, mainly through the scale of the habitat and repair times. Major malicious damage by groups was not addressed. There is no period in history that we know of that lasted this long without major wars. The US alone had a devastating Civil War just 164 years ago. The post-WWII peace in Europe guaranteed by the Pax Americana is perhaps the longest period of peace in Europe, albeit its guarantor nation was involved in serious wars in SE Asia, and the Middle East during this period. Humans seem disposed to violence between groups, even within families. I couldn’t say whether even a family of women would refrain from existential violence. Perhaps genetic modification or chemical inhibition in the water supply may be solutions to outbreaks of violence. This is speculation in a subject outside my field of expertise. I hope that the competition discovers a viable solution.

In summary, while it is worthwhile considering how we might tackle human star flight of such duration, there is no evidence that we are even close to solving the issues inherent in maintaining any sort of human star flight of such duration with the earlier-stage technologies and the constraints imposed on the designs by the competition. By the time we can contemplate such flights, technological change will likely have made all the constraints superfluous.

References and Further Reading

Genta, G Genta, “Interstellar Exploration: From Science Fiction to Actual Technology,”” Acta Astronautica Vol. 222 (September 2024), pp. 655-660 https://www.sciencedirect.com/science/article/pii/S0094576524003655?via%3Dihub

Gilster P, (2024) “Science Fiction and the Interstellar Imagination,” Web https://www.centauri-dreams.org/2024/07/17/science-fiction-and-the-interstellar-imagination/

BIS “Project Daedalus” https://en.wikipedia.org/wiki/Project_Daedalus

BIS “Project Icarus” https://en.wikipedia.org/wiki/Project_Icarus_(interstellar)

i4is “Project Hyperion Design Competition” https://www.projecthyperion.org/

Heppenheimer, T. A. (1977). Colonies in space: A Comprehensive and Factual Account of the Prospects for Human Colonization of Space. Harrisburg, PA: Stackpole Books, 1977

Soilleux, R. (2020) “Implications for an Interstellar World-Ship in findings from the BIS SPACE Project” – Principium #30 pp5-15 https://i4is.org/wp-content/uploads/2020/08/Principium30-print-2008311001opt.pdf

Nelson, M., Pechurkin, N. S., Allen, J. P., Somova, L. A., & Gitelson, J. I. (2010). “Closed ecological systems, space life support and biospherics.” In Humana Press eBooks (pp. 517–565). https://doi.org/10.1007/978-1-60327-140-0_11

Hempsell, M (2020), “Colonies and World Ships” JBIS, 73, pp.28-36 (abstract) https://www.researchgate.net/publication/377407466_Colonies_and_World_Ships

Tolley, A. (2023) “MaRMIE: The Martian Regolith Microbiome Inoculation Experiment.” Web https://www.centauri-dreams.org/?s=Marmie

Bio-Rad “S3e Cell Sorter” https://www.bio-rad.com/en-uk/category/cell-sorters?ID=OK1GR1KSY

Allen, J., Nelson, M., & Alling, A. (2003). “The legacy of biosphere 2 for the study of biospherics and closed ecological systems.” Advances in Space Research, 31(7), 1629–1639. https://doi.org/10.1016/s0273-1177(03)00103-0

Smith, C. M. (2014). “Estimation of a genetically viable population for multigenerational interstellar voyaging: Review and data for project Hyperion.” Acta Astronautica, 97, 16–29. https://www.sciencedirect.com/science/article/abs/pii/S0094576513004669

Hein A.M,, et al (2020). “World ships: Feasibility and Rationale” Acta Futura, 12, 75-104, 2020 Web https://arxiv.org/abs/2005.04100