Case Study: Designing Life Support for a Crewed Mars Transit

"On the way to Mars, the spacecraft is not carrying the crew — the crew is carrying the spacecraft's ability to keep them alive, and every kilogram of it fights the rocket equation."

Executive Summary

In the first case study we audited a finished system in the friendliest possible environment — low Earth orbit, hours from resupply. Here we do the hard thing: design the human-systems architecture for a crewed Mars mission, where resupply is impossible and the environment is at its most lethal. Given a crew and a mission length, we will show that open-loop life support is a non-starter, size a closed-loop system that actually closes the mass, budget the radiation dose and carve a storm shelter out of consumables the mission already carries, and make the artificial-gravity decision that every Mars study wrestles with. The result is a human-systems mass budget — the exact package that feeds the Mars mission of Chapter 34 and the capstone MDR. Numbers are rounded planning figures (Tier 2), with illustrative hardware masses flagged (Tier 3).

Skills applied

• Showing why long missions must recycle, via the open- vs closed-loop mass comparison (§28.1–28.2).
• Sizing a closed-loop system and identifying the commodity that now dominates (§28.2).
• Budgeting radiation dose against a career limit and sizing a storm shelter from consumables (§28.3).
• Making the artificial-gravity decision with $a = \omega^2 r$ and the comfort tradeoff (§28.5).

Background

The requirement

• Crew: $4$.
• Duration: a conjunction-class round trip — roughly $180\ \text{days}$ out, a $\sim 500$-day surface stay, and $180\ \text{days}$ back — about $900$ days the life-support system must sustain.
• Resupply: none. Once trans-Mars injection fires, the crew has what they launched with (plus anything pre-positioned on Mars — outside our scope).

The one hard truth

Distance changes everything. In LEO the ISS could ship up the hard commodities; here there is no ship. Every deficiency in loop closure, every gram of shielding, every design margin must be launched at the start and dragged the whole way — and, per the rocket equation of Chapter 3, accelerated past escape velocity. This is the theme mass is the enemy at its most unforgiving.

Phase 1: Open-loop is dead on arrival

Start with the counterfactual, to feel the wall. Open-loop consumables for the mission, at the chapter's supplied rate of $5.14\ \text{kg}$ per crew-day:

$$ M_{\text{open}} = 4\ \text{crew} \times 900\ \text{days} \times 5.14\ \frac{\text{kg}}{\text{crew-day}} = 3{,}600 \times 5.14 = 18{,}504\ \text{kg} \approx 18.5\ \text{t}. $$

And that counts only drinking water. Include hygiene — say a realistic total water demand near $6\ \text{kg}$ per crew-day rather than $2.5$ — and the supplied rate climbs to $8.64\ \text{kg}$ per crew-day, giving

$$ M_{\text{open,realistic}} = 3{,}600 \times 8.64 \approx 31{,}100\ \text{kg} \approx 31\ \text{t}. $$

Between eighteen and thirty-one tonnes of consumables — as much as an entire large payload — thrown toward Mars and consumed en route. No architecture survives that. Recycling is not an optimization here; it is the price of admission.

Phase 2: Size the closed-loop system

Now close the loops. Food cannot be recycled, so it is launched in full. Water is recovered at $90\%$ and oxygen at $50\%$, exactly as in the ISS audit, leaving only makeup streams. To those we add an illustrative recycler hardware mass of $M_{\text{hw}} = 2{,}500\ \text{kg}$ — heavier than the ISS-audit figure because a Mars system must be more robust and self-maintaining (Tier 3).

$$ M_{\text{closed}} \approx 11.4\ \text{t}, \qquad \text{saving vs lean open-loop} = 18.5 - 11.4 = 7.1\ \text{t}. $$

Against the realistic (with-hygiene) open-loop of $31\ \text{t}$, the saving is far larger — closing the water loop nearly erases the hygiene-water penalty — so recycling saves on the order of $18$–$20\ \text{t}$.

🔗 Connection: Notice what has happened to the mass. In the open-loop world water dominated; here, with water recovered, food is now the single biggest human-systems mass ($6.48\ \text{t}$, more than half the closed-loop total). The unclosed loop always becomes the driver. This is precisely why bioregenerative food research (growing crops in transit) and pre-positioning food on Mars are studied so hard: food is the last, hardest loop, and on a Mars mission it is the one left holding the mass.

def closed_loop_mass(crew, days, hw=2500, w_rec=0.90, o2_rec=0.50):
    food     = crew * days * 1.8
    water_mk = crew * days * 2.5 * (1 - w_rec)
    o2_mk    = crew * days * 0.84 * (1 - o2_rec)
    return food, water_mk, o2_mk, food + water_mk + o2_mk + hw

f, w, o, tot = closed_loop_mass(4, 900)
print(f"food {f:.0f} | water {w:.0f} | o2 {o:.0f} | total {tot:.0f} kg")
open_lean = 4 * 900 * 5.14
print(f"open-loop (lean): {open_lean:.0f} kg")
# Expected output:
# food 6480 | water 900 | o2 1512 | total 11392 kg
# open-loop (lean): 18504 kg

Phase 3: Radiation — budget the dose, then build the shelter

Life support keeps the crew breathing; radiation is what quietly ages them. Budget the mission dose from §28.3's rates (Tier 2). The two $180$-day cruises spend $360$ days in deep space at $\sim 1.8\ \text{mSv}$ per day; the $\sim 500$-day surface stay, partly shielded by Mars's thin atmosphere and the ground beneath, runs at roughly $0.7\ \text{mSv}$ per day:

$$ D \approx (360 \times 1.8) + (500 \times 0.7) = 648 + 350 = 998\ \text{mSv} \approx 1.0\ \text{Sv}. $$

Against NASA's current career limit of about $0.6\ \text{Sv}$, a single mission overspends a career's allowance — before any solar particle event. This is a genuine, unsolved constraint: the honest options are a faster transit (less cruise time), better shielding, accepting elevated lifetime risk, or all three.

The acute threat — a large SPE mid-cruise — we defend against with a storm shelter, and here the mission's own consumables become the shield. Suppose we want a shelter wall of areal density $20\ \text{g/cm}^2$ (enough to blunt an SPE) around a compact refuge of about $10\ \text{m}^2$ of wall area. Areal density converts to a surface mass:

$$ 20\ \frac{\text{g}}{\text{cm}^2} = 200\ \frac{\text{kg}}{\text{m}^2}, \qquad M_{\text{shelter}} = 200\ \frac{\text{kg}}{\text{m}^2} \times 10\ \text{m}^2 = 2{,}000\ \text{kg}. $$

The mission already carries $6{,}480\ \text{kg}$ of food and a large water inventory — several times the $2{,}000\ \text{kg}$ needed. So the shelter's shielding is nearly free in launch mass: arrange the food lockers and water tanks as the walls of the sleep refuge, and the crew rides out the storm behind commodities they were carrying anyway.

🔧 Engineering Reality: There is a subtlety the areal-density sum hides: food gets eaten. Early in the mission the food wall is thick; by the return leg much of it is gone, thinning exactly the shielding the crew may still need. Good designs lean on the water inventory (topped up by the recycler and slower to deplete) and on stored waste, and they place the shelter where the vehicle's tanks and structure add their mass for free. Shielding that changes thickness over the mission is the kind of second-order trap that separates a sketch from a flyable design.

Phase 4: The artificial-gravity decision

Nine hundred days is long enough that the microgravity losses of §28.4 — bone, muscle, and especially the possibly-irreversible vision syndrome (SANS) — become mission risks, not nuisances. Do we spin?

A rigid $1\,g$ ring at a comfortable $2\ \text{rpm}$ would need a $224\ \text{m}$ radius (§28.5) — far too massive to launch. But we do not need a rigid ring, and we may not need a full $g$. Tether the habitat to the spent trans-Mars-injection stage with a $\sim 100\ \text{m}$ cable and spin the pair; the crew module sits at a radius of roughly $r = 50\ \text{m}$ from the spin center. Aim not for $1\,g$ but for Mars-level partial gravity, $a = 0.38\,g = 3.73\ \text{m/s}^2$:

$$ \omega = \sqrt{\frac{a}{r}} = \sqrt{\frac{3.73}{50}} = \sqrt{0.0746} = 0.273\ \text{rad/s}, \qquad \text{rpm} = 0.273 \times \frac{60}{2\pi} = 2.6\ \text{rpm}. $$

A gentle $2.6\ \text{rpm}$ on a cheap tether buys Mars-level gravity — enough, plausibly, to blunt the worst of the deconditioning and to pre-adapt the crew to the gravity they will land in.

import math
def ag_rpm(r, a):
    omega = math.sqrt(a / r)
    return omega * 60 / (2 * math.pi)

print(f"0.38 g at 50 m: {ag_rpm(50, 0.38*9.81):.1f} rpm")
print(f"1.00 g at 50 m: {ag_rpm(50, 9.81):.1f} rpm")
# Expected output:
# 0.38 g at 50 m: 2.6 rpm
# 1.00 g at 50 m: 4.2 rpm

The counterargument is real: a tethered spin adds the complexity of deploying, despinning to maneuver or communicate with a fixed antenna, and the risk of a tether failure — against a proven, if imperfect, alternative (exercise). Every Mars study weighs this trade, and none has yet judged artificial gravity worth its cost over hard exercise. Our design reserves the tether option and books the mass for a robust exercise suite; which one flies is a decision we leave open, exactly as the real programs do.

Phase 5: The human-systems mass budget

Roll it up into the number the mission architect actually needs — the human-systems mass this crew adds to the vehicle:

That $\sim 11.4\ \text{t}$ is the figure that flows into the Mars mission's overall mass statement in Chapter 34 and, through the rocket equation, into the trans-Mars-injection propellant. Sanity check: it is comfortably less than the open-loop $18.5\ \text{t}$ (and far below the realistic $31\ \text{t}$), it is food-dominated as expected once the loops are closed, and it is of the same order as the pressurized cargo of a single large logistics vehicle — a plausible slice of a Mars transit habitat. The design closes.

Discussion Questions

1. In the open-loop world water dominated the mass; in the closed-loop design food does. Explain the handoff, and why it points designers toward bioregenerative food and Mars pre-positioning.
2. The mission dose ($\sim 1\ \text{Sv}$) exceeds the career limit ($\sim 0.6\ \text{Sv}$). Of the honest options (faster transit, more shielding, accept risk), which does the rocket equation make expensive, and why?
3. Why is the storm shelter "nearly free" in launch mass while shielding the whole habitat to the same level would be unaffordable? Tie your answer to how shielding mass scales.
4. The artificial-gravity tether gives Mars-level gravity at a gentle $2.6\ \text{rpm}$. What operational complications does spinning the stack introduce that a non-spinning vehicle avoids?

Your Turn: Extensions

• Option A (design). Re-size the closed-loop system for a crew of $6$ on the same $900$-day mission. Which grows faster — the fixed hardware or the consumables — and what does that say about crew size?
• Option B (computation). Add a mission_dose(cruise_days, surface_days) function and a shelter_mass(areal_density_gcm2, area_m2) function to the code above; reproduce the $\sim 1\ \text{Sv}$ and $2{,}000\ \text{kg}$ results. (Do not run them — hand-trace and add # Expected output:.)
• Option C (your mission). If you ever imagine a crewed variant of your project track, carry this design's method — consumables, loop closure, shelter, artificial-gravity decision — into your MDR's crew note, and record the human-systems mass it would add.

Key Takeaways

1. Distance forces closure. Open-loop is $18.5$–$31\ \text{t}$ of consumables for a crew of four to Mars — impossible; closed-loop brings it to about $11.4\ \text{t}$, and beyond a few months recycling is the only architecture that closes the mass.
2. Food becomes the driver once water and oxygen are recycled — the last, hardest loop holds the mass.
3. The storm shelter is nearly free because it reuses consumables (water, food) the mission already carries as shielding — but watch the food wall thin as it is eaten.
4. Artificial gravity is a genuine open trade: a tethered spin buys Mars-level gravity at a gentle $2.6\ \text{rpm}$, but complexity and risk have so far kept every real Mars study on exercise instead.
5. The output is a human-systems mass budget (~$11.4\ \text{t}$) that feeds straight into the Mars mission architecture of Chapter 34 and the capstone MDR.