Loading
Old version — revision 1
This is a fixed snapshot of Closed-loop life support, saved by Import as part of the initial corpus import. It is not edited and it is not updated; the article may have changed since.
Edit summary: Initial import of content/closed-loop-life-support.md — the filesystem corpus, unchanged. Not an edit.
Engineering that regenerates a crew's air, water, and food from their own waste, reducing the resupply mass that otherwise makes long-duration spaceflight impossible.
Closed-loop life support is the engineering of a habitat that regenerates the consumables its crew uses — oxygen, water, and in the fully closed case food — from metabolic and system waste, rather than carrying or resupplying them. It is a mass problem before it is a biology problem. An unrecycled crew consumes roughly five kilograms of oxygen, water and food per person per day, so a three-year Mars mission for four people implies a resupply mass no launch architecture can deliver. Every fraction of a loop that closes converts consumable mass into hardware mass and power, and the design question is where that trade turns favourable.
Four loops are usually treated separately.
The atmosphere loop removes carbon dioxide, supplies oxygen, controls trace contaminants and manages pressure and humidity. On the International Space Station, carbon dioxide is captured on regenerable zeolite sorbent beds and oxygen is produced by electrolysing water. A Sabatier reactor closes part of the carbon loop by combining captured carbon dioxide with hydrogen from electrolysis to make methane and water; the methane is vented, so hydrogen, and therefore water, leaves the system permanently. Closing that last step requires methane pyrolysis to recover the hydrogen, which has been developed but not operationally flown at scale.
The water loop is the most closed in practice. Urine is processed by vacuum distillation, condensate and hygiene water are collected, and everything passes through multifiltration and catalytic oxidation. The addition of a brine processor, which extracts water from the residual concentrate the distillation leaves behind, raised station recovery to roughly 98 per cent.
The food loop is not closed at all. Every calorie consumed in orbit since 1961 has been launched from Earth. Small plant growth facilities aboard the station produce leafy greens, radishes and peppers in quantities that are nutritionally and psychologically welcome and metabolically trivial.
The waste loop is largely open. Solid waste is compacted and disposed of; the carbon, nitrogen and phosphorus it contains are not recovered.
ISS life support is a hybrid physicochemical system with essentially no biology in the loop, and it works. Water recovery near 98 per cent and oxygen generation from recovered water mean the station's consumable resupply is dominated by food. That is a satisfactory position for a vehicle in low Earth orbit with regular cargo flights, and an inadequate one for Mars.
Reliability rather than closure is the operational pain point. The oxygen generation and urine processing assemblies have both required repeated maintenance and part replacement, and the spares logistics that supports them is itself a resupply stream. A system that fails every few months is acceptable when replacement parts are ninety minutes away, and not otherwise. Assessments of readiness in this domain frequently note that the relevant Technology readiness level for Mars is not about whether a component works but whether it works unattended for a thousand days.
In-situ resource use changes the arithmetic where it is available. The Martian atmosphere is over ninety-five per cent carbon dioxide, and MOXIE demonstrated oxygen production from it at gram scale on the surface.1 Subsurface water ice would supply hydrogen. A surface base that can mine its own oxygen and water does not need the same degree of loop closure as a transit vehicle, which is one reason mission architectures front-load risk into the cruise phase — the same phase in which radiation dose and deconditioning also peak, and the phase that proposals for crew torpor target.
Closing the food loop requires organisms. Two long-running programmes define the field.
BIOS-3 in Krasnoyarsk demonstrated in the 1970s and 1980s that a sealed facility with algal cultivators and hydroponic wheat could regenerate air and water essentially completely and supply a substantial share of the crew's calories, with small crews and runs of up to six months.2 Its results remain among the most credible in the literature precisely because the system was small, instrumented, and honest about what it did not close.
ESA's MELiSSA takes a compartmentalised microbial approach: anaerobic liquefaction of waste, photoheterotrophic and nitrifying stages that convert the products into nitrate, a photoautotrophic compartment growing cyanobacteria, and a higher-plant compartment producing food and oxygen for the crew. A pilot plant in Barcelona has operated compartments in series with animal crew analogues. Progress has been steady and slow, which is what a system with this many coupled biological stages should be expected to produce.
Crop area is the binding constraint on full food closure. Estimates cluster in the range of tens of square metres of continuously cropped area per person, and the power demand for artificial lighting typically dominates the entire habitat's energy budget. Cultured animal cells have been proposed as a compact protein source using the methods of Tissue engineering, and engineered microorganisms — including strains built with the tools described in Synthetic genomes and biocontained by the synthetic auxotrophy of Genetic code expansion and recoding — appear in most forward-looking architectures. None has flown.
Small closed ecosystems are unstable in ways large ones are not. The buffering capacity of the Earth's atmosphere and oceans is enormous relative to any perturbation a population of eight can generate; the buffering capacity of a sealed structure is not. Biosphere 2 illustrated this comprehensively. Atmospheric oxygen fell from the normal 20.9 per cent to about 14.2 per cent over sixteen months and had to be injected. The cause was traced to microbial respiration in organic-rich soils generating carbon dioxide that was then absorbed by uncured concrete, so oxygen disappeared without a corresponding rise in carbon dioxide to signal where it had gone.3 Food production fell short, the crew lost weight, most vertebrate and pollinator species died, and ants and cockroaches came to dominate the fauna.
The lesson usually drawn is that Biosphere 2 failed. The more useful lesson is that it failed in a way nobody predicted, through a concrete-chemistry interaction with a soil microbial community, and that closed-system engineering is dominated by such couplings.
Other persistent problems are more mundane. Volatile organic compounds accumulate and must be catalytically oxidised. Biofilms form in water lines and have caused recurring maintenance on the station. Crop pathogens in a monoculture with no natural enemies can propagate through a hydroponic system rapidly. Menu monotony degrades crew intake in ways that show up as weight loss in every long-duration analogue.
Closure percentages are not comparableA system described as "98 per cent closed" may be quoting water recovery, total mass balance, or oxygen regeneration, and may or may not count food. Because food dominates the unclosed mass in every flown system, headline closure figures that exclude it can differ by an order of magnitude from figures that include it. Comparisons between facilities generally require reading the definitions.
The dominant risk is coupled failure. Physicochemical and biological subsystems share water, air and power, so a fault in one propagates. Redundancy is expensive in mass and does not help against a common-mode failure such as a contaminant that poisons both the crop and the microbial compartments.
Introducing engineered organisms adds a containment question. A bioregenerative loop is a deliberately constructed ecosystem, and organisms designed for it would eventually be released, intentionally or otherwise, into a planetary environment — a scenario that has driven planetary protection policy for decades and that shares structure with the concerns raised in Mirror life. How much weight the Precautionary principle should carry when the alternative is a crew running out of food is not a question the literature has resolved.
The near-term trajectory is incremental: better reliability, methane pyrolysis to close the hydrogen loop, higher-yield crop chambers, and improved trace contaminant control. None of that produces a closed system, and no agency's current Mars architecture assumes one; the baseline is high water and oxygen recovery with all food carried.
Full closure only becomes mandatory when resupply is unavailable: a permanent settlement, or the multi-generation voyages considered under Generation ship biology. At that point the problem stops being spacecraft engineering and becomes ecology, with all the stability questions that implies and none of the redundancy the Earth provides. It is not obvious that a system small enough to launch can be stable for centuries, and the two-year Biosphere 2 record remains the longest human test of the proposition. Whether the answer is better engineering, larger systems, or adapting the crew instead — the argument made in Pantropy — is unsettled, and bears directly on any serious account of the Future of humanity beyond Earth.
paperHoffman, J.A. et al. "Mars Oxygen ISRU Experiment (MOXIE) — Preparing for human Mars exploration." Science Advances, 2022. ↩
paperSalisbury, F.B., Gitelson, J.I., Lisovsky, G.M. "Bios-3: Siberian Experiments in Bioregenerative Life Support." BioScience, 1997. ↩
paperSeveringhaus, J.P., Broecker, W.S., Dempster, W.F., MacCallum, T., Wahlen, M. "Oxygen loss in Biosphere 2." Eos, Transactions American Geophysical Union, 1994.↩The concrete sink was identified after the fact; oxygen fell without the matching rise in carbon dioxide that would have shown where it went.