Skip to content

Constraint 17: Water Recycling Efficiency

Physical Motivation

A space habitat has no connection to Earth's hydrological cycle. Every litre of water that escapes the system — through evaporation venting, chemical reactions, equipment losses, or biological output that is not recaptured — must be replaced by water launched from Earth or extracted from local resources (lunar ice, asteroids). At current launch costs, water is the most mass-expensive consumable to resupply. The habitat's water recycling loop must therefore approach thermodynamic closure.

The Closed-Loop Water Balance

Let \(N\) be the colony population, \(D_{pp}\) the daily domestic water demand per person (L/day), and \(\eta\) the recycling efficiency (fraction recovered per cycle). The total daily water throughput is:

\[D_{total} = D_{pp} \cdot N \quad [\text{L/day}]\]

Water that is not recovered constitutes a net daily loss:

\[L_{day} = D_{total} \cdot (1 - \eta) \quad [\text{L/day}]\]

Since the density of water is approximately 1 kg/L, the annual mass loss is:

\[L_{year} = L_{day} \cdot 365 \quad [\text{kg/year}]\]

For long-term self-sufficiency without routine resupply, the recycling efficiency must meet a minimum threshold \(\eta_{min}\):

\[\eta \geq \eta_{min}\]

This is the feasibility condition.

What This Model Covers

This constraint models domestic water only: drinking, food preparation, hygiene, and laundry. Agricultural water is a separate, largely self-contained subsystem — evapotranspiration from crops is recovered via greenhouse condensation at efficiencies >97%, making it a tighter loop than the domestic system (Hendrickx et al. 2019). Modelling the two separately reflects how they are engineered in practice (ECLSS handles domestic; the greenhouse handles agricultural).

Numerical Reference

NASA's Baseline Values and Assumptions Document (BVAD) specifies domestic water demand for a space habitat at approximately 22 L/person/day (Hanford 2004):

Use Demand (L/person/day)
Potable (drinking/cooking) 2.0
Oral hygiene 0.4
Hand/face wash 4.1
Shower 2.7
Clothes washing 12.5
Urinal flush 0.5
Total ~22

The model default of 20 L/person/day is close to this figure.

ISS ECLSS Baseline

The International Space Station Environmental Control and Life Support System (ECLSS) has progressed through two distinct performance eras:

Pre-BPA era (~2009–2021): ~93% recovery

Original ECLSS achieved approximately 93% water recovery through two subsystems (Carter et al. 2009):

  • Urine Processor Assembly (UPA): Distils urine to brine; ~85% recovery
  • Water Recovery System (WRS): Processes condensate + UPA distillate; combined recovery ~93%

Current era (with BPA, 2024): 98% recovery

NASA's Brine Processor Assembly (BPA), activated in 2023–2024, recovers approximately 95–98% of the residual water from the UPA brine output. Combined with an improved UPA achieving ~87–98% urine recovery, the total system now achieves 98% water recovery — the Mars mission threshold (Gatens et al. 2024). This was described as a milestone for long-duration exploration.

ISS net consumption is approximately 3.6 L/person/day — a result of severe rationing, not of high efficiency. A comfortable long-duration colony running full sanitation at 20 L/person/day at the historical 93% efficiency would lose 1.4 L/person/day per person.

At 8,000 people, that historical 93% loss rate is 11,200 L/day = 4,088 tonnes/year — roughly 1,360 Falcon 9 payloads annually. Even at the current ISS level of 98%, loss is 818 t/year. For a fully isolated habitat, any non-zero loss eventually depletes the supply; 98% is viable only if loss is covered by local resource extraction (ISRU) or very infrequent resupply.

One analysis of the 98% regime noted the operational margin is "too small for comfort" when accounting for disposal paths — hygiene towels, wipes, and contamination losses — that bypass the recovery system entirely. Future designs targeting 99%+ are an active NASA research direction.

Required Efficiency for Self-Sufficiency

For a colony of \(N\) people at demand \(D_{pp}\) and a target maximum annual loss of \(L_{max}\) kg/year, the required efficiency is:

\[\eta_{min} = 1 - \frac{L_{max}}{D_{pp} \cdot N \cdot 365}\]

For \(L_{max} = 0\) (fully closed loop): \(\eta_{min} = 1.0\), which is physically unachievable. In practice the goal is to reduce losses to a level manageable by local resource extraction (lunar ice mining, comet ice) or very infrequent resupply.

The model default \(\eta_{min} = 0.98\) represents the minimum efficiency for a colony where annual water loss is small enough to be covered by in-situ resource utilisation (ISRU).

Consequences of Failing the Constraint

\(\eta\) Annual loss (8,000 people, 20 L/day) Equivalent launches
0.90 (ISS level) 4,088 t/year ~1,360/year
0.95 2,044 t/year ~680/year
0.98 (threshold) 818 t/year ~273/year
0.99 409 t/year ~136/year
0.999 41 t/year ~14/year

The Falcon 9 equivalent is based on ~3 t useful payload to orbit.

Model Inputs

Symbol Parameter Default Source
\(D_{pp}\) water_per_person_day_liters 20 L/day Hanford 2004
\(\eta\) water_recycling_efficiency 0.98 Design target
\(\eta_{min}\) min_water_recycling_efficiency 0.98 Design target

Key Insight

The default (0.98) represents NASA's stated minimum for Mars/permanent missions and has now been demonstrated on the ISS (2024) with the BPA. It is well-supported by the literature.

One important caveat: 98% is the thermodynamic recovery from the active ECLSS loop. Actual habitat water loss also includes passive disposal paths (wipes, hygiene towels, contaminated water) not captured by the system. The effective whole-habitat efficiency may be lower. For a permanent colony, the "system efficiency" the model checks is the ECLSS fraction; the designer must separately minimise non-loop waste paths.

Users can slide water_recycling_efficiency below 0.98 to see what pre-BPA ISS-level loss rates imply at colony scale.

References

  • Carter, Layne, et al. "Water Recovery System (WRS) and Urine Processor Assembly (UPA) Status." 38th International Conference on Environmental Systems. 2009. (Carter et al. 2009) — Documents pre-BPA 93% performance.
  • Gatens, Robyn, et al. "Status of ISS Water Management and Recovery." 54th International Conference on Environmental Systems. NTRS 20240005472. 2024. (Gatens et al. 2024) — Documents 98% BPA achievement and Mars mission requirement.
  • Hanford, Anthony J. Advanced Life Support Baseline Values and Assumptions Document. NASA/CR-2004-208941. NASA, 2004. (Hanford 2004)
  • Hendrickx, Lieve, et al. "Microbial ecology of the closed artificial ecosystem MELiSSA." Advances in Space Research 23.12 (2019): 1–15. (Hendrickx et al. 2019)
  • Wieland, Paul O. Designing for Human Presence in Space: An Introduction to Environmental Control and Life Support Systems. NASA Reference Publication
  • NASA, 1994. (Wieland 1994)