Systems Engineering Overview for Regenerative Life-Support Systems Applicable to Space Habitats


The Life-Support Systems Group of the 1977 NASA Ames Summer Study conducted analyses initially to identify areas where partial or complete closure of the life-support system, for various types of space habitats, should be considered, and to develop a sound methodology for identifying, screening, and evaluating alternative closure schemes. Results of these analyses, including the systems engineering considerations that affect technology development planning and management for regenerative life-support systems, are discussed. A recommended methodology is also presented as a basis for future technology-development activities in this area.


The essentials for human survival are commonly referred to as the life-support system (LSS). Requirements for life support on manned space missions include those provisions and accommodations that are necessary to maintain human occupants of the vehicle or space station in a functional state. A functional state is defined as the willingness and physical ability to perform assignments effectively and to accept responsibilities for the duration of the mission. Thus, in general, the necessary provisions and accommodations consist of a primary set (adequate food, potable water, breathable air, and safety from identifiable hazards) and a secondary set (waste disposal, personal hygiene, medical care, and reasonable comfort). It is recognized that these factors affect the psychological, as well as the physiological well-being; however, the focus of this paper is on physiological considerations.

This report was prepared by the Life-Support Systems Group of the 1977 NASA Ames Summer Study on "Space Settlements and Industrialization Using Nonterrestrial Materials." This report summarizes the group's analyses and recommendations concerning several issues that were derived from the objectives of the group's study activity. These objectives included:

The 1977 NASA Ames Summer Study was concerned principally with industrial production in space using nonterrestrial materials, housing, and life support for an industrial work force in space, and space settlements per se. The study planners reasoned that the earliest industrial activities in space probably will depend on "open loop" life-support provisions, including resupply from Earth for food and perhaps some partial regeneration of atmosphere components and water. However, beyond this early transitional period, it is believed that long-term space settlements for space industrialization will require essentially completely regenerative, or "closed," life-support systems. Therefore, the charter of the Life-Support Systems Group for the 1977 Summer Study was to determine the requirements for achieving this closure, including particularly the research that will be necessary over a period of many years.

The group's findings during the summer study are presented in this paper and its companion, subsequent paper entitled "Research Planning Criteria for Regenerative Life-Support Systems Applicable to Space Habitats." The information presented in these two papers represents interim results on systems analysis, the identification of research options, and the formulation of guidelines for the eventual assignment of research priorities. However, follow up studies are anticipated and they will be reported later with more detailed assessments of research and development options and recommended priorities for resource allocations.


With minor exceptions, life-support functions on manned space missions to date have been accomplished by (1) the on-board storage at launch, of food, water, and atmosphere components; (2) removal (but not conversion) of trace contaminants and excess carbon dioxide; and (3) storage and return to Earth of waste residues. Provisions for personal hygiene and medical care were very limited until the Skylab missions. These life-support procedures to date have at least been inherently reliable and safe if not a source of comfort, convenience, and pleasure for crew members.

The principal focus of prior NASA-sponsored research and development on life-support systems has centered on partial closure of the mass exchange through the use of physicochemical processes for atmosphere revitalization, wash water regeneration (non potable use), and waste conversion to veritable gases, reusable (potable) water, and a storable, compact ash. Food production and regeneration have received very little attention for space-mission applications.

Several trade-off determinations combined to provide the rationale for this orientation of LSS design and technology development until the present. These include the following:

Funding priorities for life-support systems research and development are assigned according to projected time frames of need and readiness for mission applications. Therefore, in view of the limited resources that were available for advanced technology development in these areas, research on completely closed systems and, in particular, biological processing components, received little or no support, except for minor exploratory efforts that decreased to virtually nothing several years ago.

Current interest in self-sustaining,. regenerative, life-support systems derives principally from the possibility that habitats will be needed in the coming decades to support missions that could well involve a large number of people working in space for long periods of time and, in some cases, at sites that are quite remote from the Earth.

The inputs required to support a person in space are shown in table 1. Over the course of a year, the average person consumes 3 times his body weight in food, 4 times his weight in oxygen, and 8 times his weight in drinking water. Over the course of a lifetime, these items amount to over a 1000 times an adult's weight.

Type of input Amounts,
kg per person per:
Day Year Lifetime
Food (dry) 0.6 219 15,300
Oxygen 0.9 329 23,000
Drinking water 1.8 657 46,000
Sanitary water 2.3 840 58,800
Subtotal 5.6 2,045 143,100
Domestic water 16.8 6,132 429,240
Total 22.4 8,177 572,340

As discussed above, for long-term space missions, it will probably be necessary to replenish or recycle supplies on board. The quantities of supplies required increase with the duration of the flight and the cost of supplies required for the journey is proportional to their weight at launch. At some break-even point it will probably be less costly to send up the equipment to recycle the supplies from wastes than to use up the provisions on a once-through basis.

Consider, for example, wash water which is relatively easy to recycle because the contaminant levels are low (about 350 ppm; ref.1). A number of processes have been investigated for wash water recycle, including systems based on multifiltration, ultrafiltration, reverse osmosis, and vapor compression distillation (ref. 2). A schematic (ref. 3) of a system based on reverse osmosis (RO), with vapor compression distillation to concentrate the rejected wastes, is shown in figure 1. Waste water from a storage tank is passed through progressively finer cartridge filters to remove particulates that can foul the reverse osmosis membranes. The filtered water enters a recirculation loop of RO modules, where the water that permeates the membranes is collected and stored at pasteurization temperature. The concentrated "brine" is sent to a vapor compression distillation unit which reduces the contaminants to a concentrated slurry, either to be stored on-board or processed further.

Even a recycling system such as this requires that replacement filters, membranes, and chemical additives be brought from Earth, with their quantity depending on the length of the flight. Thus the total weight at launch of the system increases with flight duration.

For a six-person crew, launch weight as a function of flight duration is shown in figure 2, for state of the art as well as projected technology. Also shown is the launch weight of an alternative system, multifiltration, based on sequential filtration with cartridge filters, activated carbon, and ion-exchange resins. Based on current knowledge, an RO-type of system for wash water recycle would probably be preferred for a long-duration mission.

Human wastes and other organic wastes (food wastes and trash) can be turned into carbon dioxide, water, and ash by presently available methods. Wet and dry oxidation systems have been built and tested on a demonstration scale. In wet oxidation, an aqueous slurry is treated at moderate temperatures (150o- 250o C) under high pressures (8.3X106 to 1.5X107 N/m2 (ref.4). In dry oxidation, toilet wastes and aqueous slurries of pulverized trash are first concentrated by evaporation of water, and the concentrate is then incinerated (ref.5). Water vapor and off-gases from the incinerator are passed through catalytic oxidizer beds to completely convert organic contaminants to CO2 and H2O.

A number of physicochemical techniques have been suggested for recycling oxygen from CO2 (refs. 6, 7). These methods involve separation of CO2 from air followed by reaction of CO2 to produce O2 and carbon by-products (methane in the Sabatier process, or solid carbon in the Bosch process). However, processes of this type are not suitable for a closed life-support system for flights of long duration. Although oxygen is generated for reuse, the carbon present in the food that was consumed is ultimately converted to a solid form and must be stored. To close the system, the carbon must be returned to the food cycle from which it originated.

All of the carbon in human food originates either directly or indirectly (i.e., through animal consumption) from vegetation. Thus, to close the system, all of the carbon in the outputs from living components of the system (e.g., humans, plants, and animals) should be converted back to carbon dioxide so that it can be photosynthesized back into vegetation. In this manner, it is possible to satisfy simultaneously the recycle requirements for oxygen and hydrogen. For example, if the average carbon, hydrogen, and oxygen composition of vegetation is represented by the formula CHm On, then the stoichiometry of net plant growth (photosynthesis less respiration) is:

A portion of the edible plant output is oxidize d back to CO2 and H2O directly by human and animal metabolism. The remainder of the edible portion is partially oxidized to products which appear in urine, feces, perspiration, and exhaled breath. If these partially Oxidized products, together with the nonedible portion of the plants, are subsequently oxidized to CO2 and H2O, then the net effect of human and animal metabolism plus waste oxidation is the reverse of the net photosynthesis stoichiometry. Thus, incorporation of a process like wet oxidation or incineration for complete conversion of organic wastes to CO2 and H2O simultaneously closes the carbon, oxygen, and hydrogen recycle loop.

The logistical basis for the recycling of the wash water, as discussed above and characterized in figure 2, also is applicable to the regeneration of the food supply. Figure 3 illustrates that economic break-even or cross-over points occur when comparing the alternatives of either regenerating food at the space settlement (with some nominal resupply from Earth) or continually resupplying all food from Earth (with no regeneration). The ranges of cumulative food-weight transportation requirements per inhabitant, shown as functions of mission duration (or continuous time in space) in figure 3, are based on best available information. The state of the art for food storage was taken as 650 kg per person per year, which is approximately that for present Space Shuttle food-supply planning. This includes thermo-stabilized and intermediate moisture foods to achieve a level of palatability and general acceptability better than that of the food used on Apollo missions (about 400 kg/person/yr). The optimistic estimate for stored foods is about 475 kg per person per year. This is based on the analysis summarized in table 2, which characterizes a reasonable diet and assumes, optimistically, that technology developments in the future will be able to accommodate the freeze-drying and packaging of these dietary components in an acceptable palatable form within the per-person annual resupply weight value of 475 kg.

Food item Food weights,
grams per person per day
Natural Dry Freeze-dried
and packaged
Cereals 175 154 180
Potatoes 142 114 110
Sugar 75 75 80
Beans and nuts 36 33 38
Vegetables and fruit 608 535 500
Meat 218 88 100
Eggs 76 18 22
Fish 18 6 7
Milk 785 103 110
Fat 55 ~55 60
Drinks (dry) 12 12 15
Alcohol 30 30 70
Salts 15 15 16
Total 2,245 1,238 1,308

It should be noted, however, that foods could be chosen differently to minimize the total weight and perhaps reduce this optimistic weight-requirement value by a factor of 2 (i.e., to about 220 kg/person/yr). This weight-conserving food supply does not represent conventional terrestrial diets, however, and its acceptability may be limited. For this reason it was not used in this analysis as the optimistic possibility for food storage.

The optimistic value for food-regeneration requirements shown in figure 3 was taken as 1500 kg/person (total biomass holdup for the self-sustaining, steady-state condition). Therefore, that amount of total food-producing material must be transported from Earth before the food regeneration process can reach this condition. The slope of this line represents an optimistic assumption that only 100 kg/person of resupplied biomass or food must be transported annually from Earth to compensate for supplemental requirements, attrition of material, etc. The initial supply value of 1500 kg/person agrees well with the value presented in the Final Report of the 1975 NASA Ames Summer Study (ref. 8), although this figure was derived independently in a more rigorous manner by members of the Life Support Systems Group of the 1977 NASA Ames Summer Study.
The pessimistic value for food-regeneration biomass requirements was estimated to be 4000 kg/person. This amounts to between 2 and 3 times the optimistic value, and was selected only to obtain some estimate of the sensitivity of the crossover point to the magnitude of the value assigned to food-regeneration biomass requirements. Similarly, the slope of the pessimistic regeneration line was chosen to be significantly greater than that for the optimistic regeneration line, to portray sensitivity. This slope represents a pessimistic assumption of an annual resupply requirement of 350 kg/person.

Figure 3 illustrates several points:

At this point it should be emphasized that the food storage versus food-regeneration comparisons have been based on the required weight to be transported to the space habitat for each case. Assuming transportation cost to be the only significant economic factor, this is a reasonable basis for direct comparisons. However, in reality, the cost of research and development necessary to achieve stable regeneration of food, compared with, for example, the amount of importation of supplies characterized by the optimistic regeneration line in figure 3, might be high enough to greatly change the crossover point to a much longer mission duration in a total-cost comparison of state-of-the-art food storage with resupply. Nevertheless, since the weight (and cost) of subsystems needed to provide atmosphere regeneration and water reclamation are not included for the stored-food cases in figure 3, these comparisons are quite conservative.

Another very significant factor would be the cost of transporting material into space. A greater amount for this cost might significantly offset extensive research and development costs to achieve the optimistic food regeneration capability. Furthermore, consideration must also be given to the extent of transportation cost that might be required to provide equipment for cultivating, harvesting, processing, preparing, and delivering the food in a regenerative system, and for the additional waste processing associated with such a system.

Therefore, the cost-benefit analysis for the comparison of resupply versus regeneration of food is very complex and cannot be accomplished effectively until reliable capability and cost estimates can be made. Even more complex and information-dependent is the selection and design of an integrated regenerative life-support system for a space habitat. Therefore, the information requirements must be fully understood and then satisfied through careful planning.


Integrated systems design refers to the process of defining the requirements that must be satisfied by a regenerative life-support system, identifying component and subsystem options, screening and selecting among these options for integration into an effective system that satisfies all requirements, and then integrating this life-support system design into the overall habitat design. By far the best approach is to accomplish all of this by way of a total system analysis and design procedure that considers all component and subsystem interactions at the same time. The result has been demonstrated extensively to be superior to the outcome of the alternative procedure of fixing or optimizing one aspect of the design and then trying to retrofit other subsystems into this fixed arrangement such that typically, the latter subsystems cannot function optimally. The best approach requires careful attention to all identifiable options and the eventual selection of a totally optimum combination of components and subsystems that will meet all performance requirements in a stable, safe, and reliable manner.

This is especially vital for completely regenerative life-support systems which will very likely include biological as well as physicochemical components. The balancing of the input and output streams among the components to maintain stable operation can be expected to be extremely sensitive. This requires detailed consideration of interface constraints inside and outside the system. The task would probably be much easier if only terrestrial conditions needed to be considered. A great deal more is known about the performance of humans, animals, and plants in terrestrial gravity and in an air atmosphere. Almost nothing is known about the effects of long-term zero-gravity or fractional-gravity exposure on plants and animals. In general, this means that enough data will be needed, at a minimum, to support extensive sensitivity analyses for the various systems design scenarios that should be considered. This must include characterization of the effects that choices of external factors (gravity, atmosphere, adjacent activities, etc.) can have on the closed life-support system option, as well as the effects that each life-support system option can have on other elements of the habitat design.

Overall, this requires that mission planners and conceptual designers remain aware of and give adequate consideration to the sensitivity of the regenerative life-support system's selection, design, and performance, and its potential effect on total habitat design. Otherwise, these will be neglected and the life-support system will be developed in a retrofit mode. This would very likely result in a low probability of optimality in the selection or operation of the life-support system. Furthermore, the consequences could well be disastrous.

As was discussed earlier in this paper, there are various approaches to and degrees of closure. At the present time not enough is known to support the identification of optimum choices. However, a particularly effective procedure for the identification and characterization of options involves the steps shown in figure 4. The procedure begins by selecting a diet scenario which is disaggregated into the types and quantities of food that will be specified to meet the dietary requirements. From this list of required food the resources necessary for producing the vegetables, fruits, and grains on the list are determined. From the agricultural characteristics of these crops the amount of waste that will be generated (i.e., material not consumed by the humans) is estimated. Then the requirements for feed and resources to produce the animal-derived food components of the diet scenario are determined, based on the utilization of the byproducts from the crops. The waste generated from the production and processing of the animal-derived food can be estimated at this point.

Next, it is necessary to determine the waste-processing scheme needed to regenerate the required inputs to the plants and animals (i.e., nutrients, water, carbon dioxide, etc.). Processing requirements for food production and preparation can be estimated at this point. These include equipment, facilities, and resources (e.g., water and energy) for milking cows, processing the milk and its derivative products, baking bread, processing meat and vegetables, refrigeration, packing, and storing, to name a few. Finally, it is necessary to determine the stability characteristics of the proposed system that has been put together conceptually by this procedure, and identify requirements for monitoring and control.

This latter step provides the basis for the analysis of the system's sensitivity to perturbations at its internal and external interfaces.

The major internal features of the life-support system that can be varied to generate different scenario options are the diet and the method of closure. The method of closure includes both the functions in the system that will be closed and the component biological and physicochemical species that will be employed to accomplish these functions in a closed mode. Although the options that are possible for the variation of these features are many, it would be reasonable to start by selecting a diet that closely represents conventional terrestrial choices of food and then determining the component requirements for full closure of the life-support system. From a baseline case of this type, subsequent analyses can progress to other diet and closure options. For each scenario option, the trade-offs associated with it should be examined carefully in comparison with other proposed options. Analyses of this type are extremely valuable in identifying research required to improve the data base such that the evaluational comparisons will be more reliable.

Several scenarios have been considered and reported to date. The diet associated with the scenarios considered by the 1975 Ames Summer Study team (ref. 8), Henson and Henson (ref. 9), and Soviet researchers (ref. 10) did not closely represent conventional terrestrial choices of food. A scenario based on a more conventional diet is currently being formulated and evaluated by the Bioenvironmental Systems Study Group on NASA Contract No. NASw-2981. The foundation for this diet is presented in the subsequent companion paper prepared by the Life Support Systems Group of the 1977 NASA Ames Summer Study (entitled "Research Planning Criteria for Regenerative Life-Support Systems Applicable to Space Habitats").

It was mentioned earlier in this report that the life-support system also has external "trade-off interfaces." These are defined as environmental or operational conditions that the life-support systems share with other components or systems of the habitat. For example, the gravitational force designed into the habitat is an environmental condition that is shared by all systems; any connections or interlocks between the section of the habitat in which the regenerative life-support system functions and, perhaps, a manufacturing section, would represent shared operational interfaces. Good systems engineering practices require that these interfaces be selected with careful attention to, and adequate knowledge of, the effect of the selected conditions on all systems that share these interfaces. The choice is then optimized with respect to all these systems through trade-off analyses that are based on a sound rationale.

Typical external interfaces for a regenerative life-support system in a space habitat are discussed below.

Atmosphere Environment

The parameters of atmosphere selection include total pressure, composition, and the individual partial pressures of the principal constituents (oxygen, carbon dioxide, and nitrogen, or other inert diluent). The factors involved in the selection of any atmosphere other than terrestrial air are very complex. Nevertheless, it is very clear that hasty decisions must not be made simply for the convenience of engineering design or other logistical or operational considerations. The health and safety of the habitat's occupants must come first, and this includes the need for a stabilized and optimally performing life-support system. As stated earlier, there is a serious lack of information about the effects of exposure to nonterrestrial atmospheres on humans, plants, and animals. For this reason, pure oxygen or oxygen-rich atmospheres must be avoided until an adequate understanding of the effects of multigeneration exposure of biological species (including microbiological candidates) to such atmospheres has been achieved. Obviously, the length of the time delay and cost incurred in deriving this level of understanding could far outweigh the anticipated benefits from the selection of such an atmosphere, and must be considered in any trade-off analyses. Certainly, from Apollo Project experience, it must be realized that use of such atmospheres involves a significant increase in the risk of fire and imposes severe restrictions on the selection of materials that can be used in the habitat (refs. 11-13).

With respect to the carbon dioxide composition, the current maximum limit for manned spacecraft environ- ments is 7.6 mm of Hg partial pressure with a nominal level of 5 mm of Hg Research to date by plant researchers in the controlled environments of phytotron chambers indicates that this carbon dioxide level is greater than the maximum level of enrichment of carbon dioxide in terrestrial air for which plant-growth enhancement has been achieved (refs. 14 and 15). Therefore, it does not appear that different atmospheric compositions will be required to satisfy the general needs of humans, plants, and animals. Nevertheless, the optimal level of CO2 still remains to be determined.

The emphasis expressed above on the advantages of using terrestrial air does not entirely limit considerations to "one standard Earth atmosphere." Stable ecologies exist on Earth at high-altitude locations where the total pressure of the atmosphere is reduced to about 2/3 the standard Earth atmosphere. However, the ratio of oxygen to nitrogen is still about 20/80 (ref. 16), not oxygen enriched. Therefore, these types of atmospheres also qualify for reasonable consideration. The extent of the restrictions, if any, imposed on agriculture (e.g., variety and productivity), on effects on human health and epidemiology, etc., should be investigated thoroughly.

Gravity Environment

The effects of zero-g and low-g exposure on higher plants and food-producing animals are not known. Effects on humans have been documented but not fully explained. This suggests that, at a minimum, the serious consideration of a gravitational environment for the habitat which is other than that on Earth will add significantly to the needed research and development time. Until adequate information can be obtained that ensures the health and safety of the habitat's human occupants in such an environment, and the stability and optimal performance of its life-support system, only terrestrial gravity should be considered for space habitats.


The general data base is probably adequate for the specification of illumination requirements for humans, plants, and animals. However, these specifications need to be formulated to provide guidelines for the design of adequate filters or reflectors for the alteration of sunlight, to allow the use of solar illumination in space, or completely artificial illumination using solar energy as a power source. If the habitat must be completely shielded with opaque material to eliminate the cosmic radiation hazard, artificial illumination would be necessary. The results of extensive research show that plants grow very well in artificial illumination (refs. 14, 15). Similar information on illumination requirements (e.g., periodicity, spectral characteristics, and intensity) that are optimum for humans and food-producing animals needs to be compiled.

Special Constraints

These include area, volume and configurational (or shape) limits within which the life-support system must function. Several estimates of area requirements for habitat occupants and their regenerative life-support system have been published (refs. 8, 9), based on different scenarios for regeneration of food. These demonstrate that once the scenario has been quantitatively modeled, various configurations for the arrangement of components of the life-support system can be studied and area and volume requirements for each can be determined. The principal focus of such analyses should be the "packing intensity" potential for various components of each candidate design scenario. In this manner, area, volume, and shape considerations become combined; otherwise the specification of an area requirement alone is quite meaningless since an effective use of an available volume, through higher packing density, can reduce this area requirement. Trade-offs arise, of course, in most layout planning studies. For example, if plants are to be grown with a higher packing density, perhaps involving vertical stacking, provisions for adequate illumination become more complex.

Nevertheless, these problems probably can be resolved from state-of-the-art knowledge and design methodology, and do not require extensive research. Human social and psychological aspects of spatial planning are another matter, and these must be considered carefully in habitat design. However, for the purposes of this report these factors are not considered to be within the domain of the regenerative life-support system, per se. In a sense, social and psychological factors represent components of a separate system in the habitat which share certain interfaces with the regenerative life support system.

Radiation Environment

The results of studies on long- and short-term radiation exposure effects show quite conclusively that living systems, particularly humans and animals, must be provided with protection from ionizing radiation that is present in the space environment. Effects of sublethal overexposure include genetic and developmental effects, the production of tumors and leukemia in humans and animals, and generalized life-shortening changes in the vascular and central nervous systems. Therefore, any attempts to reduce the structural weight of a habitat design must not be allowed to compromise the health and safety of human occupants or the stable and optimal operation of the regenerative life-support system by decreasing the radiation shield to even a marginally effective thickness. Further studies will be required to provide the basis for specifying adequate shielding requirements for space habitats. In general, however, life-support systems studies should assume that the radiation environment will not be different from terrestrial conditions.

After several reasonable scenarios for partially or completely regenerative life-support systems have been formulated - based on the selection of internal interfaces, and then analyzed for sensitivity to variations in the external interfaces, perturbations at the internal interfaces (stability characteristics), and the reliability of assumptions - trade-off analyses can be performed to select the most promising options for further consideration. Initially, the trade-off studies can be used to identify auspicious areas of research to improve the data base on performance and stability characteristics. However, as the database improves as a result of technological developments, the trade-off comparisons should be updated regularly. The results can provide a basis for (1) determining cost effectiveness of the research investment to date; (2) redirecting future efforts; (3) predicting and comparing the time of readiness for use (or "technological maturity") for the various options; and (4) identifying new system-scenario options based on combinations of particularly attractive components or subsystems to form hybrid groupings.

This iterative procedure is strongly recommended as a tool to provide effective management of technology-development programs in the multidisciplinary fields that will contribute to the development of a successful regenerative life-support system for space habitats. Methodology already developed specifically for trade-off analyses for life-support systems components (refs. 17, 18) can readily be adapted for the procedure described above.


This report has summarized criteria for the closure of the life-support system for space habitats and factors to be considered when assessing the relative benefits and costs of closure in comparison with nonregenerative options. In addition, a methodology was specified for identifying, screening, and evaluating candidate processing components for a regenerative life-support system through an iterative procedure that involves scenario formulation and modeling together with trade-off analyses. The procedure also includes guidelines for effectively managing the associated technology-development program. For all of the above topics of this report, it has been stressed that the limits of the present data base for the relevant areas of technology preclude a priori conclusions, at this time, regarding the ultimate role that regenerative life-support systems will have as a function of time in space habitats or the optimum choices to be made among potential processing components to formulate an "ideal" system scenario. Therefore, the highest priority short-term goal of future studies must be the identification of key data-base needs. From there the longer-term research required to satisfy these needs can be specified and prioritized.


Table of Contents