Crop Selection in Controlled Ecological Life Support Systems

  • Keith D. CrookerEmail author
Living reference work entry


Long-term human habitation beyond the confines of Earth will require closed-loop agricultural habitats that provide oxygen replenishment and edible biomass for future space explorers. These habitats must miniaturize terrestrial ecosystems to replicate the life-supporting biome of our planet, and allow cultivation of species-diverse plant communities for optimum human habitability. By utilizing Controlled Ecological Life Support Systems (CELSS), habitats can balance carbon dioxide acquisition, water usage, light utilization, and crop yield to increase the robustness and survivability of a habitat’s plant species mix in a space environment. Two different plant photosynthetic pathways, called C3 and C4 photosynthesis, can each metabolize and store carbon at different rates that affect their CO2 acquisition and crop yield. Previous experiments in CELSS have focused on C3 food crop species despite C4 species showing a more efficient carbon utilization level under wider environmental tolerances. This report reviews the two different photosynthetic pathways and compares the carbon biomass produced by three historical CELSS C3 crop experiments to terrestrial C4 crop experiments. While the results did not show a significant difference, they did suggest enough divergence to demonstrate possible benefits of a mixed C3/C4 crop species diversity in future CELSS habitats.


Biomass production C3 photosynthesis C4 photosynthesis Calvin-Benson cycle Carbon dioxide uptake Carbon mass balance Controlled Ecological Life Support Systems Closed-loop life support Crop diversity Spaceflight crew nutrition Space habitat design 


Life support systems for long-term manned space missions require the establishment of crew habitat enclosures that ensure basic health and safety and establish an efficient work environment for mission operations (Aydogan-Cremaschi et al. 2009). These systems have been proposed for construction of extra-planetary human habitats on the moon (Blum et al. 1994), Mars (Nelson et al. 2008a), as well as in microgravity conditions (Kitaya et al. 2000). Controlled Ecological Life Support Systems (CELSS) are self-contained habitats that will rely heavily on a mass-balance approach between the atmospheric, terrestrial, biological, and waste system components operating in a closed-loop, steady-state system (Blum et al. 1995).

Primary to CELSS crew habitats are the ability to convert carbon dioxide and water to food and oxygen, usually requiring the use of photosynthetic organisms as the mechanism (Lehto et al. 2006). Crop selection that optimizes both the radiation (light) use efficiency of the plants and their nutritional content is vital to increasing the overall success of the habitat (Bartsev et al. 2003). Numerous experiments have been conducted over the past several decades to study the different aspects of CELSS, many of them to determine which plant species are best suited for space agriculture (Nelson et al. 2008b).

In molecular plant biology, plant species differ in their methods of carbon dioxide acquisition for biochemical processing. Two of these methods, C3 and C4 photosynthesis, utilize either a 3-carbon molecule (phosphoglyceric acid) or a 4-carbon molecule (oxaloacetate) to metabolize carbon dioxide (Killops and Killops 2005; Bugbee et al. 1994). Due to a higher light-use efficiency in C4 photosynthesis, it is theorized that C4 photosynthesis is an evolutionary adaptation of C3 photosynthetic plants in response to reduced atmospheric carbon dioxide levels during the Cenezoic Era (Ehleringer et al. 1997). These adaptations in C4 plants have resulted in measurable differences in photosynthesis efficiency from C3 plants under specific sets of environmental conditions (Ehleringer et al. 1997; Bugbee et al. 1994). These differences may impact the overall efficiency of CELSS habitats if they rely on C3 plants alone.

While C4 plants have distinct advantages over C3 plants in conditions such as variable light, temperature, and CO2 concentrations, little attention is given to the selection of crop species in terrestrial agricultural systems for differences in photosynthetic chemical pathways. In a review of select experiments from published scientific literature, this same lack of discrimination between C3 and C4 plant selection has migrated to CELSS experiments despite the possible advantages that C4 plants have under variable environmental conditions prevalent in CELSS habitats. The problem examined in this study is the reliance on primarily C3 plants in CELSS despite the possible advantages that C4 plants may have due to a higher expected photosynthetic efficiency. It is hypothesized that a significant difference exists between the photosynthesis efficiencies of C4 plants and C3 plants utilized within CELSS habitats.

Background and Review of C3 Versus C4 Photosynthesis

Photosynthesis is the chemical process by which plants convert sunlight, carbon dioxide, and water into oxygen and sugar. In higher vascular plants, this occurs in cells organelles called chloroplasts, which utilize the plant-specific chemical, chlorophyll. Chloroplasts perform photosynthesis via two different biochemical pathways called photophosphorylation and the Calvin-Benson cycle (Purves et al. 1998; Fig. 1).
Fig. 1

A simplified flow diagram of photosynthesis, showing the conversion of sunlight and water into oxygen, ATP, and NADPH+H, which is utilized in the Calvin-Benson cycle to convert carbon dioxide to the sugar glucose

Photophosphorylation, known as the “light reactions,” is the pathway of photosynthesis that produces oxygen by transferring electrons from water to the chlorophyll molecule via absorption of specific wavelengths of sunlight, producing diatomic oxygen and hydrogen ions. Chlorophyll, after becoming charged with electrons from the light reactions, undergoes a series of chemical reactions to create two different compounds called adenosine triphosphate (ATP) and nicotinamide adenine dinucleotide phosphate (NADP), the latter transformed to a reducing agent by combining with the hydrogen ions to form NADPH+H (Purves et al. 1998). Acting as an electron transport to supply energy generated from photophosphorylation, ATP and NADPH+H are the two compounds required for the second biochemical pathway of photosynthesis, the Calvin-Benson cycle (Crafts-Brandner and Salvucci 2004). Also known as the “dark reactions” (named such as it is the portion of photosynthesis that does not directly require sunlight), the Calvin-Benson cycle is the photosynthetic pathway responsible for producing sugars (glucose) from carbon dioxide (Purves et al. 1998).

In most plants, there is one specific type of cell that performs photosynthesis, called the mesophyll cell. In these cells, carbon dioxide absorbed into the plant leaf is converted to the sugar glucose, which is required for plant metabolism and growth. The Calvin-Benson cycle is the second biochemical pathway of photosynthesis and starts by combining carbon dioxide with a five-carbon sugar called ribulose bisphosphate (RuBP). Using an enzyme called rubisco, the Calvin-Benson cycle combines CO2 and RuBP to form 3-phosphoglyeric acid, or 3PG. This process is known as carbon fixation. In order for the mesophyll cell to produce the sugar glucose, the Calvin-Benson cycle utilizes ATP and NADPH+H obtained from photophosphorylation to reduce 3PG to glyceraldehyde 3-phosphate, or G3P, which is then converted to glucose as the product of the entire reaction (Purves et al. 1998). This entire process, which extensively utilizes the 3-carbon molecule 3PG to produce the sugar glucose, describes the primary photosynthetic mechanism that a majority of vascular plants utilize as part of their metabolism and is given the moniker, C3 photosynthesis (Ehleringer and Bjorkman 1977).

As mentioned previously, most vascular plants utilize the 3-carbon molecule 3PG to convert carbon dioxide to glucose in the carbon fixation process. Most plant species around the globe utilize this one pathway, called C3 photosynthesis, thus characterizing them as C3 plants (Ehleringer et al. 1997). However, in 1966, another method by which plants undergo carbon fixation was discovered (Ehleringer and Bjorkman 1977), now known as C4 photosynthesis. These plants are similarly characterized as C4 plants.

Unlike C3 plants, C4 plants have more than one type of cell that can undergo photosynthesis. In C4 plants, not only does photosynthesis occur in mesophyll cells (as in C3 plants), but another type of cell, called the bundle-sheath cell, also contributes to plant photosynthesis (Fig. 2). In the mesophyll cells of C4 plants, carbon dioxide is combined with a three-carbon acid called phosphoenolpyruvate (PEP) to form the four-carbon compound, oxaloacetate, using the enzyme phosphoenolpyruvate carboxylase (PEP carboxylase). This is different from C3 plants in which carbon dioxide is combined with RuBP using the enzyme rubisco to form 3PG, the three-carbon compound for which C3 photosynthesis is based. Furthermore, the oxaloacetate produced in C4 mesophyll cells diffuses into the bundle sheath cells of C4 plants, where it is split back into CO2 in a process called decarboxylation, allowing the CO2 to become incorporated into the Calvin-Benson cycle where the normal cycle of glucose production occurs (Purves et al. 1998).
Fig. 2

A simplified flow diagram of the pathways of carbon dioxide to glucose formation within the cells of C3 plants and C4 plants. Note that the bundle sheath cells in C4 photosynthesis are more internal to the leaf than the mesophyll cell

Photosynthesis efficiency between C3 and C4 plants differ with respect to temperature and carbon dioxide concentration and, in turn, will affect the rate of carbon dioxide fixation of each plant type. Carbon fixation within a C3 leaf relies on the ability of the enzyme rubisco to convert CO2 and RuBP into 3PG, which occurs at optimum efficiency within a specific temperature range and CO2 level. Carbon fixation within a C4 leaf relies on the enzyme PEP carboxylase to convert CO2 and PEP into oxaloacetate, which occurs at optimum efficiency over a higher range of temperatures than C3 leaves. Furthermore, the affinity of PEP carboxylase (C4 carbon-fixing enzyme) for CO2 is higher than rubisco (C3 carbon-fixing enzyme), and so, C4 plants have the ability to build up higher concentrations of CO2 in their leaf tissues. This allows the C4 plant to continue producing glucose from CO2 in the Calvin-Benson cycle even after the plant is forced to stop absorbing other resources, such as water, due to drought conditions (Purves et al. 1998). All of these factors combined result in the adaptation of C4 plants to wider environmental extremes such as drought, and fluctuations in carbon dioxide levels and temperatures (Ehleringer et al. 1997).

Numerous studies exist that demonstrate the differences in photosynthesis efficiency between C3 and C4 plants. For instance, Crafts-Brandner and Salvucci (2004) showed that high temperatures and elevated CO2 levels limit the ability of the enzyme rubisco to produce 3PG from CO2 in the Calvin-Benson cycle within the leaves of cotton plants, a C3 crop (Crafts-Brandner and Salvucci 2004). Additionally, Ehleringer and Bjorkman (1977) showed that California brittlebush (a C3 plant) had a significant drop in photosynthesis efficiency over a temperature range of 14–38 °C, while tumbling saltbush (a C4 plant) remained relatively constant over a similar temperature range (Ehleringer and Bjorkman 1977). Finally, Vu and Allen (2009) found that C4 sugarcane plants have a higher tolerance to drought stress at elevated CO2 levels due to the water-use efficiency of the C4 photosynthetic pathway (Vu and Allen 2009).

History of Crop Selection in CELSS

Selection of crop species based on the edible biomass production of agricultural plants is vital for crew nutritional requirements, efficiency of carbon dioxide uptake, and generational seed propagation for optimum performance of a long-duration CELSS habitat. The primary literature is replete with plant studies specific to CELSS. However, there are few studies that address the differences in biomass growth and light utilization between C3 and C4 plants with respect to crop selection in a closed habitat. Bugbee et al. (1994) studied the effects of elevated CO2 levels in wheat and rice in growth chambers, citing that C3 photosynthesis is usually expected by plant biologists to be more efficient than C4 photosynthesis, suggesting that CELSS crop selection should focus primarily on C3 crops. However, their results indicated that there was a limit to that efficiency, demonstrating that past a certain CO2 and temperature level, C3 photosynthesis becomes less efficient (Bugbee et al. 1994). While CELSS crop studies are common, the continuing focus on only C3 plants may be overlooking inherent advantages in C4 crops such as a higher efficiency of CO2 uptake (Ehleringer et al. 1997), water use efficiency (Vu and Allen 2009), and temperature (Ehleringer and Bjorkman 1977). A search of the literature has found several CELSS experiments that studied C3 crop yield within closed growth chambers using carbon dioxide and light as control variables.

As mentioned previously, CELSS habitats are expected to rely heavily on a mass-balance approach between the biotic and abiotic components of a closed-loop, steady-state system, exploiting plants and their ability to convert carbon dioxide and water to food and oxygen (Aydogan-Cremaschi et al. 2009). Most CELSS studies have focused heavily on C3 plant selection without considering the benefits of C4 photosynthesis, which can have distinct advantages with respect to limited water and sunlight, as well as fluctuating environmental parameters such as temperature and carbon dioxide levels. Such limited resources and variables could prove vital to the success of any future operational CELSS habitat. Three key historical CELSS studies provided sufficient information to determine the photosynthetic efficiency of the C3 plant species used in each experiment. These studies include the NASA Biomass Production Chamber (Wheeler et al. 2008), the BIOS-3 experiment (Manukovsky et al. 2005), and the Biosphere 2/Laboratory Biosphere experiments (Nelson et al. 2008a, b).

The NASA Biomass Production Chamber at Kennedy Space Center, Florida, demonstrated the ability to operate a long-term CELSS plant production facility for over 10 years utilizing a hydroponic system with a nutrient film delivery system. With a growing space of 20 m2 and an inside closed volume of 113 m3, candidate crops selected for the experiment included wheat, soybean, potato, tomato, and lettuce, producing an average of 11.3 g/m2 per day of edible biomass and sequestered an average of 36.0 g/m2 per day of carbon dioxide. The project was designed only to study the bio-regenerative processes of crop horticulture in a closed system and did not include a human crew for habitation studies. It is the longest running successful CELSS experiment to date (Wheeler et al. 2008).

The Russian BIOS-3 experiment was a CELSS experiment conducted in the 1970s and 1980s that performed several different closure experiments complete with crew and habitats capable of supporting over 300 m2 of agricultural growth. The BIOS-3 experiment utilized a “conveyor belt” approach, whereby the cultivations of the crops were performed on a rotating basis, ensuring a constant plant biomass during the early growth and harvest stages of the plant lifecycle. Growing experiments utilized crops such as wheat, onion, carrot, radish, beet, chufa, dill, and cabbage, yielding 19.35 g/m2 per day of edible biomass. BIOS-3 utilized a soil-like substrate using composted or incinerated (oxidized) material from inedible plant detritus and other wastes, sequestering an estimated 8.6 g/m2 per day of CO2 and producing 19.35 g/m2 per day of edible biomass. The results of the BIOS-3 experiment showed that CELSS agriculture could provide for over two-thirds of the needs of the crew over the course of the project (Manukovsky et al. 2005).

While the original Biosphere 2 experiment in Oracle, AZ utilized almost a dozen crop species prior to its closure in the 1990s, researchers from that program embarked on a new, smaller-scale research program called Laboratory Biosphere. The Laboratory Biosphere experiment utilized cowpeas and pinto beans to study crop yields in a closed habitation environment, which produced 5.2 g/m2 per day of edible biomass for 80 days, sequestering CO2 with a daily rate of 37 ppm, or approximately 1.425 L of CO2 per day based on a total enclosure volume of 38.5 m3. The short duration of the study focused only on one generation of crops, extrapolating potential seed yield based on species for a speculated second-generation cultivation (Nelson et al. 2008b).

The historical CELSS studies outlined above represent only a small portion of the scientific literature for cultivation of crop species in closed habitat systems. These studies utilized crop yield as a function of CO2 along with additional parameters such as light, water availability, and temperature. All of these variables, either separately or together, could potentially be used to determine the photosynthesis efficiency of the CELSS studies.

Comparison Between CELSS Crop Selection and C4 Crop Experiments

To compare crop selection in CELSS with crop experiments that focused on C4 plants, photosynthesis efficiency was measured as a ratio of total dry weight of biomass to total carbon dioxide assimilated over an equal amount of time and/or growing area. This is a crude measure of photosynthesis efficiency, as many carbon mass-balance studies take great lengths to measure carbon dioxide as a function of light absorbed (PAR), converting it to carbon net assimilation through a mass balance equation of photosynthesis and respiration at the leaf surface. In this manner, CO2 assimilation is inferred rather than directly measured. However, for this review, the actual mass of CO2 was provided in the closed-chamber studies in all the CELSS experiments (BIOS-3, the Biomass Production Chamber, and Laboratory Biosphere), thereby surpassing the need for calculating CO2 acquisition from a complex set of photosynthesis equations.

As detailed previously, the CO2 is incorporated into the Calvin-Benson cycle to produce glucose, which is then converted to other biochemical compounds that are incorporated into plant tissues. Therefore, the mass of the overall plant matter is directly proportional to the carbon dioxide absorbed (Desjardins 1985). For simplicity purposes, the data collected from the CELSS studies were restricted to the mass of carbon dioxide assimilated and dry weight of biomass. For BIOS-3, data from a 1977 closure event involving human habitation were utilized. Plant biomass was calculated as a sum of the edible biomass (2141 g/day) and plant wastes (3031.4 g/day) and using a multiplication factor of 0.43 to account for water as discussed in the study. Carbon dioxide was calculated as both input from the closed atmosphere (2693 g/day) and input from soil respiration (3193.3 g/day). The NASA Biomass Production Chamber provided total mass amounts of dry matter biomass and CO2 over the course of the entire experiment for each species of crop used. These were calculated into mass per day for consistency before comparing them to the data from other CELSS experiments. The total dry biomass produced by each C3 species in the experiment was 854.2 kg over a total time of 1899 days. The total CO2 fixed was 1304 kg over the same period. These results are reported in Table 1 in units of kg per day.
Table 1

Total carbon dioxide and biomass data from CELSS experiments


Mass of CO2 acquired

Dry weight biomass

Biomass/CO2 ratio


2.45 kg/day

1.30 kg/day


Biomass Production Chamber

0.45 kg/day

0.69 kg/day


Laboratory Biosphere

6.49 kg

3.6 kg


Laboratory Biosphere provided similar information as the Biomass Production Chamber, segregating data for the two plant species into total biomass and total CO2 fixed from planting to harvest. These were similarly calculated into mass per day for consistency. Carbon dioxide and biomass accounting were published a year after the experiment in a separate paper from the original Nelson et al. (2008a), and which yielded a net total of 147.4 moles of CO2 from planting to harvest, which converts to 6.49 kg using the molecular weight for CO2 of 44 g/mol (Dempster et al. 2009). The same paper recorded a total aboveground biomass of 120 moles, which converts to 3.6 kg using the molecular weight for glucose of 30 g/mol (Dempster et al. 2009). Results are recorded in Table 1. Daily biomass and CO2 acquisition rates were not calculated since units effectively cancel out when converting to the biomass/CO2 ratio.

For each CELSS experiment, it was determined that the plants utilized were all of the C3 photosynthetic variety. BIOS-3 utilized wheat, onion, carrot, radish, beet, chufa, dill, and cabbage. The NASA Biomass Production Chamber utilized wheat, soybean, lettuce, potato, and tomato. Laboratory Biosphere utilized only two crops, cowpeas and pinto beans. Due to these crop usages, data from non-CELSS experiments had to be obtained in order to compare the photosynthesis efficiencies of the CELSS C3 plants with C4 plants.

Obtaining control data for C4 plants outside of CELSS proved difficult because many of the studies were of crops in open systems, such as agricultural fields, instead of closed chambers as used in the CELSS studies. Closed chamber experiments existed but did not provide sufficient information to compare with the CELSS studies, most notably, carbon dioxide. While most of these studies provided dry biomass information on the C4 crops, carbon dioxide was not reported on a mass-basis, but rather, as instantaneous measurements of the ambient air. This is logical for determining plant growth and photosynthesis efficiency in open-air systems with respect to volumetric concentration of CO2, but it does not allow for a mass comparison with the CELSS experiments. Therefore, it was necessary to perform an additional search of the scientific literature to find historical experiments that analyzed C4 crops with respect to both biomass and mass of carbon dioxide acquired.

Three experiments within two studies were identified as suitable control experiments for comparing photosynthesis efficiency between a C4 crop and the C3 crops analyzed within the CELSS experiments. Jans et al. (2010) performed an analysis of a 4.6 hectare field of maize (a C4 crop) in the Netherlands to quantify the carbon exchange with a focus on plant phenology (differences between roots, stem, leaves, etc.). Loomis and Lafitte (1987) also performed a study on maize but utilized elemental analysis (a type of chemical detection instrument for detecting organic elements such as carbon, oxygen, nitrogen) to directly determine carbon acquisition in relation to water stress. Each of these studies provided mass of carbon acquired through photosynthesis, either as a mass measurement of carbon dioxide or as total mass of carbon, which could be converted to mass of carbon dioxide.

Jans et al. (2010) provided their study results as grams per square meter (g/m2) for dry weight biomass and tons of carbon per hectare (tC/ha) for carbon dioxide acquired through photosynthesis. Because the size of the agricultural field was reported, the area could be calculated out of the reported data, and the mass converted to kilograms. Jans et al. (2010) reported a total dry mass of 1750 g/m2 over the course of the study (282 days) and divided the mass among leaves, stems, roots, and corncobs. In order to ensure that the data were congruent to the CELSS studies, two of which accounted for only aboveground biomass thus counting the root biomass as soil organic matter, the root biomass was also removed from the Jans et al. (2010) study, leaving a total dry biomass of 1630, which, accounting for the 4.6 hectare plot (converted to 46,000 m2), equates to 74,980 kg of aboveground dry weight biomass. Carbon dioxide data were recorded in tons of carbon per hectare (tC/ha) and reported as maximum gross assimilation (i.e., gross primary production, or GPP) as a function of photosynthesis, accounting for respiration back into the environment to arrive at the net ecosystem exchange (NEE), which is the total amount of carbon absorbed by the system as CO2, and congruent to what was measured in the CELSS experiments. The net ecosystem exchange at the time of harvest (day 282) was −5.97 tC/ha, or 5.97 if considered as a positive uptake value. For the purposes of this review, total carbon was converted to total carbon dioxide by using the ratio of molecular weight of carbon dioxide to the atomic weight of carbon, which is 44/12, or 3.67, and when applied to the NEE, equates to 21.89 tCO2/ha, or 100,694 kg of CO2 accounting for 4.6 hectares and converting metric tons to kilograms. Results of this data re-analysis are recorded in Table 2 above.
Table 2

Total carbon dioxide and biomass data from C4 control experiments


Mass of CO2 acquired

Dry weight biomass

Biomass/CO2 ratio

Jans et al. 2010

100,694 kg

74,980 kg


Loomis and Lafitte 1987 (1)

13,273 kg/ha

8590 kg/ha


Loomis and Lafitte 1987 (2)

29,847 kg/ha

18,700 kg/ha


Loomis and Lafitte (1987) provided their data as kilograms per hectare (kg/ha) for dry weight biomass and kilograms of carbon per hectare (kg C/ha) for photosynthesis and respiration. A large set of data were obtained by this experiment, which studied two harvests of maize crop grown in open fumigation chambers, subjected to four different levels of CO2, as well as separate groups that were well-watered and subjected to water stress. The control studies of two separate harvests were utilized for this review, which included maize crops that were under ambient CO2 and well watered. Harvest 1 was reported to have a dry biomass of 8590 kg/ha and a net photosynthesis (accounting for respiration) of 3620 kg C/ha, which converts to 13,273 kg CO2/ha using the 44/12 conversion ratio. Similarly, harvest 2 was reported to have a dry biomass of 18,700 kg/ha and a net photosynthesis of 8140 kg C/ha, which converts to 29,847 kg CO2/ha. Results of this data re-analysis are recorded in Table 2 below.

Analysis and Discussion

Data collected on the photosynthesis efficiency of C3 plants utilized in three historical CELSS experiments were compared to similar data collected on the photosynthesis efficiency of C4 plants in terrestrial crop experiments. These data encompassed the overall mass of plant material formed during a defined growth period (biomass) as well as mass of carbon-dioxide absorbed over the same amount of time. This formed a simplistic representation of photosynthesis efficiency in a mass-to-mass ratio of biomass over CO2 uptake.

The biomass/CO2 ratios could only be calculated from experiments that provided data on both biomass and carbon dioxide on a mass-basis. The three historical CELSS experiments provided these data directly, with little need for recalculation or re-interpretation for consistency between data, save that of conversion between units (i.e., grams to kilograms). Terrestrial C4 control data were less forthcoming, and while dry biomass was regularly reported in a majority of the published studies, difficulty was encountered with direct reporting of the mass of CO2 absorbed by the plants in the experiments. Only the Jans et al. (2010) and Loomis and Lafitte (1987) papers provided sufficient information on CO2 absorption that could be utilized for a comparative analysis with CELSS. In both papers, the mass CO2 absorbed were and indirect measurements derived from growth respiration of the plants, taking into account variables such as light absorbed, maintenance respiration (biological processes within plants that maintain metabolism), and phenology (reduced photosynthesis during the juvenile shoot stage of the plant). These results formed the mass data required to calculate the biomass/CO2 ratios for the CELSS studies and C4 control studies.

Overall, the data reported for each study group possessed a small sample size (n = 3 for 2 comparative groups) but were sufficient enough for an analysis of variance between biomass/CO2 ratios. While there were three separate CELSS studies analyzed (BIOS-3, NASA Biomass Production Chamber, Laboratory Biosphere), there were only two papers consulted for the C4 Control studies (Jans et al. 2010; Loomis and Lafitte 1987). However, Loomis and Lafitte (1987) provided a large enough data pool to extract two separate crop plots that were treated as separate C4 Control studies. While two of the three CELSS studies provided data for separate C3 crops (NASA Biomass Production Chamber and Laboratory Biosphere), it was determined that grouping the plants together into total biomass and total CO2 consumed would provide the most consistent data within the CELSS study group, as the BIOS-3 experiment lumped all C3 plants together into one input value. Although other C4 species were considered for inclusion in the control data, only maize (corn) was considered (USDA 2011). This was done for several reasons: (1) To ensure consistency in the control data (additional species would have increased the standard deviation thus adding to the lack of precision), (2) corn is the most widespread agricultural crop of the C4 crop species, and (3) the mass-based CO2 data limited the number of studies that could be utilized within the scope of this report.

The biomass/CO2 ratio of the CELSS experiments and C4 control experiments were grouped separately, and a one-way, single-factor analysis of variance (ANOVA) was performed. The CELSS average was 0.58 with a standard deviation of 0.06 (n = 3). The C4 Control average was 0.67 with a standard deviation of 0.05 (n = 3). A chart of these results is shown in Fig. 3 below. Using a 95% confidence interval, the ANOVA f-test comparing the two sets of data revealed a p-value of 0.16.
Fig. 3

Results of the biomass/CO2 comparison between the CELSS data (s.d. = 0.06) and the C4 Control data (s.d. = 0.05)

During the analysis, it was observed that one of the CELSS biomass/CO2 ratios was higher than the others by over 0.1, which happened to be the NASA Biomass Production Chamber experiment. This indicated that the data point may have been skewing the CELSS data during the ANOVA, and so, a separate ANOVA was made removing that data point. Once that point was removed, the p-value between the two sets of data dropped to 0.06, and the standard deviation for the CELSS data set dropped to 0.02 (n = 2). While this still does not bring the p-value to the 0.05 needed to fall within the 95% confidence interval, it does bring it within the 90% confidence interval, suggesting some significant difference. The reason why the NASA Biomass Production Chamber may have had a higher biomass/CO2 ratio than the BIOS-3 or Laboratory Biosphere experiments is because NASA accounted for the entire plant when measuring dry biomass in their experiments, while the other two excluded roots.

The biomass/CO2 ratio used for analysis is a rudimentary measure of photosynthetic efficiency in plant organisms and does not account for factors utilized by more robust photosynthesis studies. While this metric was considered accurate enough for the purposes of this review, there are several other methods that could have increased accuracy and precision. For instance, Luo et al. (2000) and Ehleringer and Bjorkman (1977) utilized light response data to calculate quantum yield, which measures the efficiency of light utilization in photosynthesis. Additionally, Crafts-Brandner and Salvucci (2004) used biochemical methods to measure rates photosynthesis through direct analysis of the carbon-dioxide fixing enzyme rubisco within the leaf tissues of plants. Zhu et al. (2008) went so far as to define photosynthesis efficiency as the energetic cost of photorespiration, using net ATP use in the plant as a measure of photosynthesis. None of these methods were used because they would have been beyond the scope of this review and requiring a higher level of data robustness.

Conclusions and Recommendations

Using the ratio of total plant biomass to carbon dioxide uptake, photosynthesis efficiency of C3 plants used in three historical CELSS experiments was compared to the photosynthesis efficiency of C4 plants from two terrestrial crop studies as a control. Based on the results of the ANOVA between the two sets of biomass/CO2 ratios for the CELSS studies and the C4 Control studies, the ANOVA f-test p-value of 0.16 indicates that there was no significant difference between the photosynthesis efficiency of C3 plants in CELSS and the C4 plants in the controls. This conclusion suggests that the sole use of C3 crops in CELSS habitats would be appropriate if the efficiency of any specific system is focusing strictly on plant biomass. While the results did not statistically support the original hypothesis, there was enough information to suggest that C4 crops may offer an advantage to CELSS habitats if more precise methods to measure photosynthesis efficiency were employed, and additional environmental factors were investigated such as the effect of variable water use, temperature, and light availability. Ideally, C3 and C4 crops should be investigated both separately and together to determine if optimal CELSS plant growth occurs as a heterogeneous mix of diverse crops species.

Overall, the purpose of this study was not to suggest that C4 crops should replace C3 crops in CELSS. Rather, future closed-loop space habitats that utilize crop agriculture and photosynthesis to maintain a human population should utilize a diverse set of both types of crops, and possibly even noncrop species that potentially have higher photosynthesis efficiencies for the sole purpose of carbon cycling and CO2 sequestration within CELSS. As with many biological systems on Earth, species diversity adds to the robustness and adaptability of an ecosystem, allowing it to adapt to environmental variability (Cox 1997). Due to the inhospitable extremes that will undoubtedly be encountered by CELSS in space, this concept of crop diversity should be adhered to in any plant selection for study in CELSS.



  1. Aydogan-Cremaschi S, Orcun S, Blau G, Pekny J, Reklaitis G (2009) A novel approach for life-support-system design for manned space missions. Acta Astronaut 65:330–346CrossRefGoogle Scholar
  2. Bartsev SI, Mezhevikin VV, Okhonin VA (2003) Systematic approach to life support system analyses and integration. Adv Space Res 31(7):1823–1832CrossRefGoogle Scholar
  3. Blum V, Gitelson JI, Horneck G, Kreuzberg K (1994) Opportunities and constraints of closed man-made ecological systems on the moon. Adv Space Res 14(6):271–280CrossRefGoogle Scholar
  4. Blum V, Andriske M, Eichhom H, Kreuzberg K, Schreibman MP (1995) A controlled aquatic ecological life support system (CAELSS) for combined production of fish and higher plant biomass suitable for integration into a lunar or planetary base. Acta Astronaut 37:361–371CrossRefGoogle Scholar
  5. Bugbee B, Spanarkel B, Johnson S, Monje O, Koerner G (1994) CO2 crop growth enhancement and toxicity in wheat and rice. Adv Space Res 14(11):257–267CrossRefGoogle Scholar
  6. Cox GW (1997) Conserving genetic diversity. In: Conservation biology, 2nd edn. Wm. C. Brown Publishers, Dubuque, pp 258–267Google Scholar
  7. Crafts-Brandner SJ, Salvucci ME (2004) Analyzing the impact of high temperature and CO2 on net photosynthesis: biochemical mechanisms, models and genomics. Field Crop Res 90:75–85CrossRefGoogle Scholar
  8. Dempster WF, Nelson M, Silverstone S, Allen JP (2009) Carbon dioxide dynamics of combined crops of wheat, cowpea, pinto beans in the Laboratory Biosphere closed ecological system. Adv Space Res 43:1229–1235CrossRefGoogle Scholar
  9. Desjardins RL (1985) Carbon dioxide budget of maize. Agric For Meteorol 36:29–41CrossRefGoogle Scholar
  10. Ehleringer J, Bjorkman O (1977) Quantum yields for CO2 uptake in C3 and C4 plants. Plant Physiol 59:86–90Google Scholar
  11. Ehleringer JR, Cerling TE, Helliker BR (1997) C4 photosynthesis, atmospheric CO2, and climate. Oecologia 112:285–299CrossRefGoogle Scholar
  12. Jans W, Jacobs C, Kruijt B, Elbers J, Barendse S, Moors E (2010) Carbon exchange of a maize (Zea mays L.) crop: influence of phenology. Agric Ecosyst Environ 139:316–324CrossRefGoogle Scholar
  13. Killops SD, Killops VJ (2005) An introduction to organic geochemistry., 2nd edn. Wiley-Blackwell, Hoboken, p 408Google Scholar
  14. Kitaya Y, Tani A, Goto E, Saito T, Takahashi H (2000) Development of a plant growth unit for growing plants over a long-term life cycle under microgravity conditions. Adv Space Res 26(2):50–62CrossRefGoogle Scholar
  15. Lehto KM, Lehto HJ, Kanervo EA (2006) Suitability of different photosynthetic organisms for an extraterrestrial biological life support system. Res Microbiol 157:69–76CrossRefGoogle Scholar
  16. Loomis R, Lafitte H (1987) The carbon economy of a maize crop exposed to elevated CO2 concentrations and water stress, as determined from elemental analyses. Field Crop Res 17:63–74CrossRefGoogle Scholar
  17. Luo Y, Hui D, Cheng W, Coleman JS, Johnson DW, Sims DA (2000) Canopy quantum yield in a mesocosm study. Agric For Meteorol 100:35–48CrossRefGoogle Scholar
  18. Manukovsky NS, Kovalev VS, Somova LA, Gurevich YL, Sadovsky MG (2005) Material balance and diet in bioregenerative life support systems: connection with coefficient of closure. Adv Space Res 35(9):1563–1569CrossRefGoogle Scholar
  19. Nelson M, Dempster WF, Allen JP (2008a) Integration of lessons from recent research for “Earth to Mars” life support systems. Adv Space Res 41:675–683CrossRefGoogle Scholar
  20. Nelson M, Dempster WF, Allen JP, Silverstone S, Alling A, Van Thillo M (2008b) Cowpeas and pinto beans: performance and yields of candidate space crops in the laboratory biosphere closed ecological system. Adv Space Res 41:748–753CrossRefGoogle Scholar
  21. Purves WK, Orians GH, Heller HC, Sadava D (1998) Photosynthesis: energy from the Sun. In: Associates S (ed) Life: the science of biology, 5th edn. W. H. Freeman and Company, Sunderland, pp 165–190Google Scholar
  22. USDA, NRCS (2011) The PLANTS database. National Plant Data Center. Retrieved 18 Mar 2011
  23. Vu JC, Allen LH Jr (2009) Growth at elevated CO2 delays the adverse effects of drought stress on leaf photosynthesis of the C4 sugarcane. J Plant Physiol 166:107–116CrossRefGoogle Scholar
  24. Wheeler RM, Mackowiak CL, Stutte GW, Yorio NC, Ruffe LM, Sager JC, Prince RP, Knott WM (2008) Crop productivities and radiation use efficiencies for bioregenerative life support. Adv Space Res 41:706–713CrossRefGoogle Scholar
  25. Zhu X, Long SP, Ort DR (2008) What is the maximum efficiency with which photosynthesis can convert solar energy into biomass? Curr Opin Biotechnol 19:153–159CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2019

Authors and Affiliations

  1. 1.Department of Math, Physical & Life Sciences, WorldwideEmbry Riddle Aeronautical UniversityDaytona BeachUSA

Section editors and affiliations

  • Roberto Furfaro
    • 1
  1. 1.Department of Systems and Industrial EngineeringThe University of ArizonaTucsonUSA

Personalised recommendations