Living in a Moon Habitat

Status. Independent working paper, not peer-reviewed. Every number on this page that can be computed is computed by wp136-verify.py, which was run before the sentences were written. Every figure that comes from someone else is labelled quoted and was read once by fetch on 2026-10-21 from the page named, not from the primary document. Nothing here measures a person. The design ideas are hypotheses, and the page says so each time.

§1 · The brochure and the Dragon bathroom

The phrase "months in a portapotty" is the author's, and it is a joke with a point. It is also not a description of any product on sale. What is sold to private customers today runs from minutes to about two weeks, according to one aggregator page (quoted, not checked against the operators): New Shepard, about 11 minutes, with an estimated price of $150,000 to $450,000 and flights paused since January 2026; Axiom, 10 to 14 days at about $55 million. Per unit time, the script gets $13.6k to $40.9k per minute for the first and $3.93M to $5.5M per day for the second. Those are arithmetic on a secondary source's estimates, not prices an operator has confirmed.

Months is a different case: a lunar stay, which is a projection and not a product. The nearest thing to a commercial habitat is Vast's Haven-1, and the Wikipedia page on it (quoted) says it is planned for no earlier than Q1 2027 and still in development. It lists 45 m³ of habitable volume and four crew. That is 11.25 m³ per person, a cube 2.24 m on a side. The page says there will be no on-board bathroom and that crew will use the one on the Crew Dragon they arrived in. The same page gives the crew stay three ways: thirty days, 10 to 30 days, and typically two weeks. This script does not resolve that, and neither does this paper.

§2 · The Moon gives no day

A lunar day is one synodic month, 29.53 Earth days: about 14.77 days of light and 14.77 of dark, 708.7 hours, or 29.53 cycles of 24 hours. So any 24-hour cycle inside a lunar habitat is supplied by the habitat. A sunlit polar ridge shortens the dark but does not give a day: a DLR report on the Connecting Ridge, read earlier this month and not re-checked for this page, gives 3 to 5 days of darkness there. Light is only one of the cues that keep a body clock on 24 hours, along with meal times, temperature, activity and social routine, and this paper treats the whole set as the thing to design.

What this section does not say. Circadian alignment and allostatic load are related but not the same thing. Allostatic load is the cumulative wear from chronic stress, and circadian disruption is one contributor next to isolation, confinement, noise, workload, partial gravity and radiation. A habitat that keeps a 24-hour cycle has not thereby lowered allostatic load. That is a measurement, and no one has made it here.

§3 · Sound

Outside there is no sound, because there is no atmosphere. Inside there is the life-support hardware. The NASA acoustics technical brief (quoted, via a summary; the standard cited is NASA-STD-3001 Vol. 2) sets the following limits for long missions: continuous noise in work areas at NC-50, crew quarters and sleep areas at NC-40 on missions over 30 days, tones and narrow-band components at least 10 dB below the broadband level of the octave band containing them, alarms under 95 dBA at the receiver, and, during sleep periods, impulse and intermittent noise within 10 dB of background. The brief describes the crew of the Space Shuttle mission STS-40 as "very irritated" during operations and sleep, and has no behavioural-health section. So the agency already treats tones and intermittent sounds as special, which is the substance of "buzzing drives people crazy", and it treats them as something to limit, not to design for.

A factor of 10 dB is a factor of 10 in power and 3.16 in pressure amplitude (computed). A tone 10 dB below its octave band is therefore one tenth of that band's power.

A hard-walled room has resonances. The axial modes are $f_n = n\,c/2L$. With $c = 343.2$ m/s (air at 20 °C, an assumption of this page):

The 10 m mode at 17.2 Hz is below the nominal 20 Hz lower edge of hearing. The table says where the resonances sit, and nothing about how any of them is perceived or whether any affects a person. That is the line between archaeoacoustics as measurement and as claim, and it is the same line the Hypogeum correction on the Soundworks pages draws.

Two facts about gas, under the ideal-gas model. The speed of sound $c=\sqrt{\gamma R T/M}$ contains no pressure, so a habitat kept at reduced pressure has the same mode frequencies as one at sea level at the same temperature and gas. The characteristic impedance $\rho c$ does scale with pressure: at 70.3 kPa it is 69% of the sea-level value and at 55.2 kPa 54% (the two reduced pressures are illustrative, not a specification). A helium-oxygen mix of 21/79 has $c$ = 1.82 times air, which moves the 10 m first mode from 17.2 Hz to 31.3 Hz (an earlier guess of more than 2.5 times was wrong, and the run said so).

§4 · Light and smell

Light. The requirement we propose is eight hours of full dark in the sleep quarters: no indicator LEDs, screens or door-edge glow visible from the bunk, and a way to block any that remains. For night movement, a very dim warm light that the sleeper switches on beats a standing glow. For the watch room, dim red or dark. All of this is a design hypothesis. Red light is generally regarded as less disruptive to the body clock and to dark adaptation than blue-white light; no source is cited here, and the size of the difference is not claimed. The NASA acoustics brief covers noise and says nothing on light.

Smell. On a closed air loop, anything added circulates for months. The hypothesis is to add no fragrance, to remove bad smells at the source, and to give the galley its own air return so cooking does not reach the sleep quarters or the watch room. It is also commonly said that sleeping people do not reliably wake to the smell of smoke. That is from memory, unverified here, and it matters because it would mean smell cannot be the safety channel and detection has to be by sensor. Mint is the example of a plant that serves several roles: a growing, touchable, aromatic plant for tea in the galley, with almost no calories, so its value is flavour and ritual, and with a medical claim limited to what we have not checked. I recall some evidence for peppermint oil for irritable-bowel symptoms; it is not verified here and I would not state a strength.

§5 · Sleep, and the watch

Fire, loss of pressure and toxic air still have to wake the crew. The design question is how to make the emergency alarm the only thing that can. Routine alerts go to a log or the day crew. A graded system, an emergency wake signal that cannot be confused with an everyday sound, and a vibrating wristband or bunk pad are options. None is known here to have been flight-tested.

Someone has to hold a night watch, or the habitat has to run itself with an on-call duty person who sleeps normally and is woken only by a real alarm. If a person is awake on watch, they are off the shared rhythm. With the author's rule of at most three consecutive nights per person, and one person on watch every night, the script computes the share of nights each person spends on watch: 33.3% for a crew of 3, 25% for 4, 20% for 5, 16.7% for 6. The control, a fixed night watch, puts 100% on one person and 0% on the rest. Rotation was chosen on purpose: a permanent night watch would cut that person off from the crew, and the author's position is that people need to be included more, not less. The cost is that each person in a crew of four spends a quarter of their nights off the shared rhythm. The three-night cap is the author's hypothesis; its basis is not claimed here. A person who prefers nights may choose longer blocks, by choice and never by assignment, and the rotation should come with protected shared time: at least one meal or gathering a day for everyone awake, and an overlap at handover.

People woken from deep sleep are generally impaired for a short time afterwards (sleep inertia, from general sleep science, not sourced here), so emergency procedures should be short and practised, with no hard decisions in the first minutes.

§6 · Touch, animals, plants

Touch. Touch plausibly helps regulate stress, and going without it for months may be hard. That is a hypothesis; I recall mixed findings on touch and stress hormones and have not checked them. Deep-pressure items (weighted blankets, compression vests) and robot huggers exist; the evidence for both is unchecked here, and a machine should be optional and never a reason to reduce human contact. The author's position, added in conversation: the best source of touch is not necessarily another crew member, so the design should not assume it is.

Animals. A systematic review of dog-based interventions in prisons exists (cited below; its page was blocked to my reader, so its results are unread here). A puppy would collide with the habitat's own rules: barking and whining are intermittent sound, odour circulates on the loop, dander is an allergen in a sealed volume, food, water and waste count against the mass budget, and the animal's welfare in launch, confinement and one-sixth gravity is untested. A calmer adult animal, a small animal or a robotic pet are alternatives to test.

Plants. Plants do respond to touch: some fold their leaves, many grow differently under repeated touch, and aromatic herbs release scent when brushed (general botany, not sourced here). The plant responds to the touch; whether the person benefits is the hypothesis. I do not know of a common food plant with a strong touch response. A pettable, edible, soft plant is more plausible than a huggable tree. A daily care task, tending and harvesting for tea, may supply part of what the prison programs supply: a living thing that depends on you. That, too, is a hypothesis. NASA's VEG-04 abstract (cited) says crews filled in behavioural-health surveys on time spent, enjoyment, engagement and sensory stimulation, and reports no results; it frames plant tending as a possible countermeasure, not a demonstrated one.

§7 · The small room, the practice, and the load

Films make it look easy: the crew drifts through a bright module, rested, fit and calm, and nobody is short of sleep or privacy. Sections 3 to 6 argued the sensory side of that. This section is about what a person does when the walls, the sound and the faces are all still there.

A room with a lock. The author tells of a person who said that on Earth she locked herself in the bathroom to cry, and in space she would go out on a spacewalk to stretch her legs. Taken as a design requirement: a small room with its own quiet, dim light and a lock, where a person can cry or just stop performing for a while. It has a price. At the lunar-surface volume factor of 133.1 kg/m³ quoted from the equivalent-system-mass paper ICES-2026-160 (not re-read for this page), 1 m³ of pressurised room is 133 kg and 2 m³ is 266 kg, volume only. Whether it improves anything is a hypothesis.

A room that comes with you. The author's position is that a trained contemplative practice (oneness, advaita, looking at the cosmos and thinking I am that) gives a person a retreat they carry, and that with training the smell, the sound and the light recede. What I can say: a 2014 review of meditation programmes (Goyal et al., read as an abstract only) rated the evidence for sleep as low or insufficient and found no advantage over active treatments, and mindfulness has been proposed for isolated, confined crews (the MINDFUL-ICE II work and Pagnini et al. 2019, abstracts only). I know of no evidence that a practice removes a stimulus. It may change how a person responds, and exposure limits must stay in force whatever any crew member says they notice. The author also describes reports of time stretching during deep practice. Those are self-reports; this page does not treat them as a physical effect, and nothing here tests them.

The view. Seeing the cosmos or the Earth may help. That is from memory (the astronauts’ “overview effect”), unchecked, and a window on a lunar surface habitat is a structural and radiation problem as well as a comfort. The retreat room can have a screen if it cannot have a window.

The load. Sitting still for months is bad for bone and muscle. The proven countermeasure is resistance exercise on heavy hardware; in 42 ISS astronauts with adequate energy and vitamin D, the ARED device preserved bone density where the earlier iRED did not (from an abstract read in an earlier session). Food and vitamin D come first; no exercise makes up for their absence. The author’s ladder, from least to most hardware: elastic bands (which can be stacked to heavy loads; the author’s point, plausible, unchecked here, and a band’s force rises as it stretches, unlike a weight), slow paced loading of a muscle against its opposing muscle (the “dynamic tension” tradition), movement practice such as capoeira, and the device. The evidence against the middle rungs is stated plainly: bone responds to dynamic strain, a static hold acts like no load, and I know of no trial showing that co-contraction or bands alone maintain bone density. The nearest prior art is a single ISS case with an electrically stimulated antagonist (one subject, no statistics). The author’s answer is that the combination is untested and worth testing, which is fair, and is the test in §8.

Capoeira on the Moon. For the same takeoff speed, jump height and flight time scale as 1/g, about 6.05 times at g = 1.62 against 9.81 m/s². An illustrative 3 m/s takeoff, chosen for this page, gives 0.46 m on Earth and 2.78 m on the Moon, 0.61 s against 3.70 s in the air. A habitat ceiling would cap it, and so would landing in a pressurised module; the roda on the Moon is slow, floaty and low-ceilinged. Capoeira also supplies a pacing cue, which a paced practice needs: it is played to music. The series’ own Berimbau Machine has a bass at D2 = 73.42 Hz and a 12-step atabaque cycle with palmas (clapping) at 104 BPM, a step of 0.2885 s and a cycle of 3.46 s. The tempo is a parameter, not a prescription, and the claps follow a simple pattern chosen for the machine, not a traditional toque. I have not heard it; I only read its code and rendered its plate. In a habitat, sound from it falls under the acoustic limits of §3, so headphones.

Where the dose comes from. The author, as a young banker at Banco do Brasil, trained and then taught colleagues stretches at the workplace: 15 minutes in the morning and 15 in the afternoon, run in the hours when the branch was closed to the public (so it did not compete with customers). Management was pleased, and the author reports that sick leave fell about 30% and that the practice later spread through Brazilian public offices as ginástica laboral. None of that is sourced here: I have no record of the 30%, and I could not find the Brasília law the author recalls from about 2003–2005 (one search; absence is not established). What I found is a bill, PL 4.859/2026, filed 3 August 2026 (a news report), that would let each level of government run programmes of up to 15 minutes a day, optional, with workers free to opt out. So 2 x 15 minutes is a reference starting dose for a trial, not an established one. At 6 days in 7 for 180 days it is 77.1 hours per person. Scheduled for everyone but voluntary to join, self-paced and logged, practised by repetition until it is automatic, it is a plausible design; and because back pain is commonly reported in flight (from memory, unchecked), pain may be what keeps people at it. The skill from repetition is not the bone stimulus: gymnasts’ dense bones are usually credited to impacts, not to slow holds (from memory, unchecked).

§8 · What would test it

The test is the same for every idea above: an analog habitat study comparing a crew with the sleep-protection, sensory and routine package against one without, on sleep timing and duration, mood, and stress markers. For the practice in §7 the arms would be none, bands only, the practice at a reference dose, and the practice at another dose, with minutes logged, back-pain scores and bone, muscle and spine measures, and food and vitamin D the same in every arm. Candidate markers are cortisol rhythm, heart-rate variability and blood pressure; I have not checked which markers spaceflight studies use. As this series requires for empirical rows (R30), the hypotheses should be written down and dated before any data is read, and the test should be shown able to detect an effect before a null result counts against anything. No such study exists on this page.

Power for the room, from a NASA release

On 8 October 2026 NASA and the Department of Energy signed an agreement on space nuclear power and propulsion, effective 1 November 2026 (NASA release 26-084, 8 Oct 2026, as pasted by the author). As described there, a launch-ready lunar surface reactor is due by 2030 under a December 2025 executive order, and Lunar Reactor-1 is the fission power system meant to carry the future Moon Base through lunar night. For this page the point is narrow: every sleep, light and sound idea above assumes the crew has power through the dark, and the release names the system intended to supply it. It gives no electrical output, no reactor location relative to quarters, and no noise or vibration figures, so nothing in §3 or §5 changes. Reactor hum near sleeping quarters is a question this page cannot answer.

§9 · What each statement is

§10 · The pageant problem

Sooner or later someone asks what the crew will look like. The honest answer is that nobody has measured it. Less stressed people tend to look less stressed, and we offer that as an untested hope, held lightly, because the Moon has no good mirrors: the mirror is a camera feed, and the lighting is whatever the habitat supplies at that hour.

A pageant needs a stage, a panel and an audience. The habitat has 11.25 m³ per person and, at the first commercial station, no bathroom (see §1). The swimsuit round would require a pressure suit. The talent round is mostly making tea. In the interview round, asked what she wants most for the world, the contestant says: quiet, dim light and a door that locks. The sash reads Mind Wide Shut: eyes closed, score open.

The author reports having heard that a lunar pageant is in the works. One search on 2026-10-21 found no announcement, and one search does not establish absence (R15). No statement by anyone is quoted or attributed on this page. If a pageant is announced, this section will be revised. Until then the judging criteria are the ones we can measure: sleep, stress markers, and how quickly a person comes back on a signal.

Sources

NASA acoustics technical brief · Haven-1 (Wikipedia) · Space tourism price index (aggregator) · NASA VEG-04 abstract · Dog-based interventions in prison populations: a systematic review (not read) · PL 4.859/2026, news report (Correio da Manhã)

Also cited without a link I have checked: Goyal et al. 2014 (meditation programmes); MINDFUL-ICE II; Pagnini et al. 2019; Smith et al. 2012 (ARED); Meakin et al. 2014 (mechanostat); ICES-2026-160 (volume factor). Full citations are to be verified before this page is relied on.

Companion script: book6/wp136-verify.py. Related: WP71, the Hypogeum machine and its correction.