This is an excellent paper for providing a high level overview of how to design a life support system. It of course retains the standard NASA bias against fully bioregenerative systems on the basis of heritage, weight, and complexity. Below are the highlights of the paper with my detailed comments for each section.

LIFE SUPPORT SYSTEM DESIGN

1. Life support subsystems are defined by function - oxygen generation, carbon dioxide removal, wastewater processing, etc. 2. The basic architecture of life support systems is stable. 3. Life support systems will probably evolve one subsystem or component at a time. 

I absolutely agree with this, with the caveat that biology tends to be multifunctional, so that when replacing for instance static food supplies with agriculture, the need for mechanical CO2 removal becomes negligible. 

HUMAN METABOLIC NEEDS 
1. A crewmember requires about 5 kg (11 Ibs) of drinking water, hydrated food, and oxygen per day. 2. Water for drinking and food preparation dominates. The 5 kg is roughly 1/2 water, 1/3 hydrated food, and 1/6 oxygen. 3. Dehydrating food can reduce food resupply mass by 2/3. 4. Food solids provide about 5 Calories per gram. 5. Respiration produces about 3.4 Calories per gram of oxygen consumed. 

Presupplied dehydrated food is not sustainable in a closed system, except as a method of storing food grown in the system. 

 

HYGIENE WATER REQUIREMENTS 
1. The hygiene (washing) water is typically 25 kg/day, 5 times the 5 kg/day mass of oxygen, hydrated food, and food preparation and drinking water. 2. Including hygiene water, a crewmember requires about 30 kg of water, food, and oxygen per day. 3. The hygiene (washing) water is typically 10 times the mass of the food preparation and drinking water. 4. The minimum consumed (food preparation and drinking) water is roughly 1/3 of nominal. 5. The minimum hygiene (washing) water is roughly 1/3 of nominal. 6. Two-thirds reduction in total water use, to 10 kg/day, is possible. 

Lesson being that lots of water is needed, and not just for human metabolism and hygeine. NASA missions generally try to minimize water due to weight considerations. Water, however, can be used for a lot of things in a system including mass transfer of micronutrients, radiation shielding, thermal mass and heat transfer. At low temperatures, ice can even be used structurally (e.g. pykrete). A large water buffer provides system stability, and should be integral to the system design.

ATMOSPHERE LOSSES 
1. Nitrogen resupply is needed to make up atmosphere leakage and airlock loss. 2. The daily atmosphere loss due to leakage and airlock operation can equal 2-3 kg/day. 3. Leakage and airlock loss of oxygen can equal 10% of the metabolic use of oxygen by the crew. 

I have never understood why folks insist on using the day as a measurement of time. A "day" is ambiguous depending on planet and context: it could be a full cycle of day/night, just the diurnal period, a mission day, a workday, or a period of 24hrs. Makes more sense to just reference the number of hours, in this case ~100g/hr, although it is probably heavily influenced by use of airlocks, so that should be a separate measurement (e.g. 500g/use).  

Clearly, there is opportunity for improvement in airlock leakage, which I think is largely addressed in the latest Z-series suit designs. 

 

CREW SIZE 
1. The mass of oxygen, water, and food consumed increases directly with crew size. 2. The size and cost of the life support system increase with crew size, but less than directly. Some economies of scale are possible. 

Buffer, buffer, buffer.

 

DURATION 
1. The mass of oxygen, water, and food consumed increases directly with duration. 2. The processing capacity of the life support system hardware does not increase with duration. 3. Longer duration missions require hardware with higher reliability, maintainability, and repairability, more spares, and longer life. 4. The optimum amount of recycling (of oxygen, water, and possibly food) increases as mission duration increases. 5. For duration beyond a few weeks, regenerable technologies should be used to recycle water. 
6. For duration beyond a few weeks, regenerable technologies rather than lithium hydroxide should be used to remove carbon dioxide. 7. For duration beyond a few weeks, oxygen regeneration should be used. 8. For duration of several years or permanent bases, food production might be considered. 9. Longer duration missions require more waste processing for stabilization, storage, and sanitation. 

It should be apparent that for long term habitation of space, bioregeneration is essential. Why this is not a higher priority for NASA, I don't know (and is why I created the website)!  Hardware sparing is another topic. For self sufficiency, it implies some basic manufacturing capabilities.

DESTINATION 
1. Planetary missions should be designed for significantly reduced mass. 2. Earth orbit missions should not be designed for minimum mass. 3. Long travel time requires closed loop life support in microgravity. 4. Longer resupply delay requires increased storage and spares, and higher system maintainability and reliability at a remote base. 5. Life support systems must operate in microgravity and possibly in planetary gravity. 6. Planetary dust and atmosphere and soil chemistry must be considered. 7.In situ resources can be used to reduce resupply and recycling needs. 8. Design cost analysis should consider the total mission cost, not only development or launch or operations cost. 9. Higher resupply cost justifies tighter atmosphere leakage and loss specifications. 

I'm not sure I agree with the first statement, particularly if the interplanetary transit reuses a vehicle/habitat. Planetary surface operations should leverage ISRU.

There is little reason to design a habitat for sustained microgravity. Earth biology is evolved for gravity; we should be designing our systems with centrifugal simulation for the foreseeable future. Biology may eventually adapt to microgravity, but it will be gradual, and for specialized use. 

Soil chemistry optimization will need to be tailored constantly based on local conditions. 

PLANNED OPERATIONS 
1. EVA suits typically use nonregenerable life support. 2. One EVA in an open loop suit requires about 3 kg of LiOH and 5 kg of water. 

EVA suits should return wastes to habitat for recycle, rather than venting them.

SYSTEMS INTEGRATION AND RECYCLING 
1. Full closure is impossible. 2. Some resupply is always necessary. Some regeneration and recycling is usually economic. Life support commodities should be provided by a cost-effective combination of resupply and recycling. 3. Waste should be recycled only if we need the recovered resource. Otherwise, we should stabilize and store or dump the waste. 4. The cost-effective amount of water recovery depends on the water balance. 5. Fuel cells or a hydrated food supply can provide significant water. 6. We can generate oxygen by using electrolysis rather than by reducing carbon dioxide if excess water is available. 7. We should reduce carbon dioxide to recover oxygen if the system must recycle most of the water.  8. Only 80-90% of the crew oxygen can be recovered from the carbon dioxide. 9. Metabolism of food produces about 1/3 kg/day/crewmember more water than consumed. 10. The missing 15% of the crew oxygen can be recovered from 0.1 4 kg of water. We still have about 0.2 kg/day/crewmember of excess water. 11. If dehvdrated food is supplied, the water recycling system can loose only 0.7% without resupply 12. If hvdrated food is supplied, the water recycling system can loose 5% without resupply. 13. The harvest index of plants grown for food is roughly 50%. 14. If we grow roughly half the food, plant growth will supply all the oxygen required by the crew and remove all the carbon dioxide generated by the crew.  15. The fraction of the food that must be grown to provide all crew oxygen and absorb all crew carbon dioxide equals the harvest index.  16. We need to oxidize solid waste to produce carbon dioxide if the plants produce more than roughly half (the harvest index) of the crew food.  17. If the plants produce all the crew food, we need to oxidize all the waste.

The 2nd Law of Thermodynamics makes the first statement true on a theoretical basis. Indeed, the Earth itself is not completely closed system, as it it losing hydrogen and other gases, but on the practical level it is closed. As I've asserted before, system stability must be measured in terms of time. The Earth's ecosystem is stable on the order of millions of years.  A viable CELSS habitat must be ulitmately stable on the order of multiple human generations.

Wastes should nearly always be recovered, although when a habitat is being generated from local resources, the proportion of elements may not be ideal for the habitat, so there will necessarily be waste from these (and any ISRU) operations. 

Regarding 14-17, I believe this imbalance can be addressed by having other animals in the food chain (rabbits, quail, fish, crustaceans, worms, BSF larvae, etc), as they can provide the additional CO2 needed by the plants and help to oxidize the fibrous plant parts indigestible by humans. 

HARDWARE AND TECHNOLOGY 
1. It is likely that the life support systems for the next human mission will be similar in concept and technology to those developed for the International Space Station (ISS). 2. They will be physico-chemical rather than bioregenerative. 3. For planetary missions, ISS-like physico-chemical subsystems must be reengineered for minimum mass. 

Only because that is where the dollars and effort are going.