Content uploaded by Abigail Alling
Author content
All content in this area was uploaded by Abigail Alling
Content may be subject to copyright.
250250
Lessons Learned from Biosphere 2 and Laboratory Biosphere Closed Systems
Experiments for the Mars On Earth
®
Project
Abigail Alling
1
, Mark Van Thillo
1
, William Dempster
2
, Mark Nelson
3
, Sally Silverstone
1
, John Allen
2
1
Biosphere Foundation, P.O. Box 201 Pacic Palisades, CA 90272 USA
2
Biospheric Design (a division of Global Ecotechnics) 1 Bluebird Court, Santa Fe, NM 87508 USA
3
Institute of Ecotechnics, 24 Old Gloucester St., London WC1 U.K.
Abstract
Mars On Earth® (MOE) is a demonstration/research project that will develop systems
for maintaining 4 people in a sustainable (bioregenerative) life support system on Mars. The overall
design will address not only the functional requirements for maintaining long term human habitation
in a sustainable articial environment, but the aesthetic need for beauty and nutritional/psychological
importance of a diversity of foods which has been noticeably lacking in most space settlement designs.
Key features selected for the Mars On Earth® life support system build on the experience of operating
Biosphere 2 as a closed ecological system facility from 1991-1994, its smaller 400 cubic meter test module
and Laboratory Biosphere, a cylindrical steel chamber with horizontal axis 3.68 meters long and 3.65
meters in diameter.
Future Mars On Earth® agriculture/atmospheric research will include: determining optimal light
levels for growth of a variety of crops, energy trade-offs for agriculture (e.g. light intensity vs. required
area), optimal design of soil-based agriculture/horticulture systems, strategies for safe re-use of human
waste products, and maintaining atmospheric balance between people, plants and soils.
Key words
: agriculture, air, closed system, sustainable
1. Introduction: The Emergence of Closed
Ecological Systems
New kinds of laboratory and experimental facilities
open up new scientific possibilities. Closed ecological
systems of varying size and complexity have been
developed over the past few decades: from laboratory-
sized microbial ecosystems to larger systems capable of
human life support that contain considerable biodiversity
and a variety of ecosystem types. Just as the concept
of closure was crucial to the development of classical
thermodynamics and physics, the development of closed
ecological systems has opened the study of ecosystems
and biospheric systems to a truly experimental approach
(Morowitz
et al
, 2005). Figure 1 shows not only the
classical approach to closure, but new possibilities due
to the addition of information that is gained through the
process of creating closed ecological systems.
To date, most closed system research for human life
support has focused on food production and the recycling
of air and water using ecological systems that are reliable
and sustainable. Some of the emphasis in bioregenerative
technology development has been driven by the need to
reduce consumables (such as food which would have
to be resupplied at high cost), while at the same time
recognizing that sustainable life support systems must
be designed to maximize efciency and productivity in
order to keep areas, volumes and energetic requirements
at a minimum. For example, at the heart of efforts to
replace food consumables for life support, scientists
have focused their research on two approaches: those
that involve soil-based/traditional agriculture systems
as compared with hydroponic/genetically modied crop
cultivars.
Closed systems have also been used to study
ecolog i c al sy stems with the aim of su s tainabl e
development and environmental protection on Earth.
An increased awareness of the ecological challenges
Fig.1.
Schematics of differing types of closed systems. a)
Complete closure – system closed to the flow of energy and
matter (adiabatic walls). b) System open to the ow of energy,
but closed materially (diathermal walls). c) System open to the
ow of energy and matter.
T
represents energy, and the symbol
m
(chemical potential) indicates open to the flow of matter. d)
closed ecological systems: open to information (
I
) and energy (
T
)
but essentially closed to material exchange. Examples include
the Earth’s biosphere, Biosphere 2, CEEF.
µ
T+I
T
T
© 2005 Jpn. Soc. Biol. Sci. Space
Biological Sciences in Space, Vol.19 No.4 (2005): 250-260
Article ID: 051904035
Presented at International Conference on Research High-
lights and Vanguard Technology on Environmental Engi-
neering in Agriculture Systems, Kanazawa, Japan, Septem-
ber, 2005
a)
b)
c)
d)
251
Alling, A.
et al
.
251
facing humanity – global warming, industrial pollution,
loss and degradation of natural ecosystems, species
extinctions and reliance on nite resources such as non-
renewable fossil fuels - has led to dramatically changed
perceptions with regards to our global biosphere. These
perspectives on how to live sustainably with the Earth’
s biosphere have many similarities with the objectives of
creating man-made biospheres and bioregenerative life
support systems for space. Bioregenerative life support
systems developed for small, partially or fully materially
closed ecosystems will provide new data and incentives
for Earth-based applications that will encourage a shift
from the destructive mindset of “unlimited resources” to
that of conserving, recycling and sustainably managing
limited resources. The concept of “zero-emission” as a
goal for industry reects this change of thinking.
Previously, pollution has been viewed as an inevitable
by-product of the human economy with negative impacts
on local ecosystems and the global biosphere. Now, just
as in a man-made closed ecological system, the challenge
of industrial waste is to maximize recycling between and
within processes, to reduce or replace non-recyclable
or toxic compounds, and to develop bioremediation
to safeguard the environment. In any closed system of
whatever scale, there is no “away.” Closed artificial
systems offer an extremely useful laboratory environment
for the study of complex systems. Small scale man-made
closed ecological systems, combined with their differing
ratios of soil, water and atmosphere (compared with our
global biosphere) accelerate the rate of elemental cycles.
This provides a laboratory that puts their relationships,
so to speak, under a time microscope. Table 1 presents
such a comparison between the Earth (Biosphere 1),
Biosphere 2, and Laboratory Biosphere closed ecological
systems.
Depending on the system’s size and its ratio of
components, each closed ecological system has unique
“metabolic” characteristics (Table 2). This presents both
a challenge on how to properly manage and balance
biogeochemical cycles inside closed ecological systems,
as well as an opportunity because they offer a superb
experimental tool for investigation of such cycles. They
open up the potential for “comparative biospherics” as
a way of gaining insight into Earth’s biosphere; just as
Earth Biosphere 2 Laboratory Biosphere
Soil Area: m
2
1.5 x 10
14
6300 5.37
Atmospheric Volume: m
3
4.3 x 10
18
(*1)
Average
180000
Average
40
Water Surface/ m
2
Land
2.4 0.2 0.1 – 0.6
Water Volume/ m
2
Land
8300
(*2)
0.9 0.07
Atmospheric Volume/ m
2
Land
~29,000
(*1)
29 6.8
Leak Rate
Negligible input or loss < 10%/year 1%/day
Table 1
Ratio comparisons of land (soil) surface to water surface, water volume and atmospheric volume.
Earth Biosphere 2 Laboratory Biosphere
Ratio of Biomass C:
Atmospheric C
1:1
(at 350 ppm CO
2
)
100:1
(at 1500 ppm CO
2
)
96-280:1
(mature crop to atmosphere at
1500 ppm CO
2
)
Ratio of Soil C:
Atmospheric C
2:1 5000:1
1650:1
(atmosphere at 1500 ppm CO
2
)
Estimated C
Residence Time in
Atmosphere
3 years 1 – 4 days 0.5-2 days
Table 2
Estimates of carbon ratios in biomass, soil and atmosphere in the Earth’s biosphere, Biosphere 2 and the Laboratory Biosphere
facility and an estimate of carbon cycling time as a consequence (data from Bolin and Cook, 1983; Dempster et al, 2004; Nelson et al,
2003a; Schlesinger, 1991).
Note: Biosphere 2 biomass to atmosphere carbon ratio is 100:1 if dry weight of biomass is 27 metric tons. Laboratory Biosphere
atmospheric carbon is about 20 grams at 1500 ppm. Ratios 96 - 280 would be for biomass of 4.8 kg to 14 kg dry weight. For
Laboratory Biosphere, the ratio soil C to atmosphere C calculated using January 2003 analysis with 5% organic matter in soil.
*1 = Equivalent at standard pressure, 15°C; *2 = Assuming average ocean depth of 3400 m
252252
comparative planetology has provided insight into the
dynamics of our home planet.
2. Biosphere 2
The Biosphere Consortium (Biosphere Foundation,
Biospheric Design and Institute of Ecotechnics) has two
decades of experience with the design and research of
closed ecological systems using the Biosphere 2 Test
Module (1986-1989), Biosphere 2 (1990-1994), and the
Laboratory Biosphere (2002-present). This know-how
is now being applied to the design for a future project
called Mars On Earth® (MOE).
2.1 Biosphere 2 Test Module
The Biosphere 2 Test Module is a 400 m
3
materially-
closed ecological facility, the largest such facility built
prior to the completion of Biosphere 2. It was designed
to test both the engineering and structure planned for the
much larger Biosphere 2, and life system interactions in
conditions of a closed ecological system. In operating
the testbed, there were progressive approximations made
towards creating and operating a successful integrated
life support system. For example, initial experiments
included the testing of two sealing methods, several
generations of analytic/sensor systems and the first
application of the variable volume chamber concept. The
Biosphere 2 Test Module was sealed underground with
a steel liner and was connected via an air duct with a
variable volume chamber (lung). This “lung” allowed the
atmosphere to expand and contract without exerting force
on the facility’s structure. The Biosphere 2 Test Module
achieved a leak rate of 24% per year (Nelson
et al
, 1992)
and it was the first bioregenerative facility to achieve
air purification through biological means (vs. catalytic
burners), water cycling, and human waste and domestic
gray water waste recycling. Food was grown to supply
nutrition during the short-term human closures, but the
limited growing area was not adequate to support long
periods of human habitation. Over 60 person-days were
logged in experiments, including a three-week closure in
November 1989 (Alling
et al
, 1993, Alling
et al
, 1990;
Nelson
et al
, 1992).
2.2 Biosphere 2 Overview
Biosphere 2, a 1.27 hectare system, was the first
system designed with a diversity of biomes/ecosystem
types (Fig. 2), thus making it a laboratory for study
of the Earth’s biosphere as well as a facility for the
development of bioregenerative technologies for human
life support (Allen
et al
, 2003; Allen, 2000; Allen and
Nelson, 1999). The rst Biosphere 2 experiment, 1991
– 1993, was an excellent prototype for a manned space
mission: eight people lived for two years in a man-made
biosphere. All of the water and wastewater was recycled,
air was recycled and puried (there was an addition of
oxygen which will be discussed below), and 81% of all
the food for the two-year mission was grown inside (the
remaining 19% was supplied with seed stock and food
grown inside the system prior to closure). Observations
of the group dynamics and human health during the
two years also provided invaluable information with
regards to designing biospheric systems for long-term
space missions (Allen, 2002; Alling
et al
, 2002). This
was a major factor in the Biosphere 2 design – systems
were designed to be safe, reliable and efcient as well as
satisfying and aesthetically pleasing for its inhabitants.
While crews have survived in submarines, cramped
space stations and underground isolation chambers
for periods of months to years, these types of sterile
and mechanical environments are hardly conceivable
as permanent habitations for people. There are also
signicant concerns about the long-term reliability and
stability of such systems if they are to be used in space,
outside the safety net of the Earth’s biosphere.
Biosphere 2 was also a laboratory for the study of
global ecology where the dynamics and operational
characteristics of large and complex ecological systems
could be studied from microbial diversity to the scale of
biogeochemical cycles and global changes, from genetics
to population dynamics. It was an unprecedented
attempt to both further the study of the biosphere and
develop sustainable systems for Earth. Recognizing
that Earth (Biosphere 1) is an amazingly successful
closed ecological system, the builders of Biosphere 2
incorporated many aspects of Earth’s biosphere into, by
comparison, the tiny Biosphere 2. Biosphere 2’s design
included analogues of major natural biomes such as
rainforest, savannah, ocean, desert and marsh, plus two
anthropogenic biomes, agriculture and human habitat
(Fig. 2). Overall there were some 3800 species of plants
and animals.
14
15
16
17
18
19
20
21
0 100
200
500
OXYGEN CONCENTRATION, PERCENT
DAY OF CLOSURE
300
400
Fig. 2.
Schematic layout of Biosphere 2 showing the human
habitat, intensive agriculture, natural biome analogues, variable
pressure facilities (lungs) at the facility, Oracle, Arizona. The
airtight footprint of Biosphere 2 was 1.27 hectares and overall
volume around 200,000 m
3
. It was operated as a closed
ecological system from 1991-1994.
253
Alling, A.
et al
.
253
2.3 Oxygen Dynamics in Biosphere 2
A central feature of Biosphere 2, which makes many
of the observed data particularly meaningful, was its high
degree of closure. Degree of closure must be such that
its leak rate is slow compared to internal processes so
that the leakage has negligible effects on those internal
processes (Dempster, 1994). Biosphere 2 achieved a
leak rate of less than 10 % per year. The significance
of this is dramatically illustrated by changes of oxygen
concentration over a period of 475 days (Dempster, in
press).
Fi g u r e 3 sh o w s o x y g en co nc en t r a t i on fr om
September 26, 1991 to January 13, 1993, a period of 1.3
years. Although the declining concentration of oxygen
is very slow, averaging only 140 ppm d
-1
, the fact that
loss can accumulate and reach a large difference from
normal oxygen concentration of 20.9 percent allows
the process to be accurately measured and studied. In
Fig. 4, compare how the decline would have appeared
for different hypothetical leak rates of Biosphere 2 as
determined by computer simulation. For perfect closure
(0 % per year), the curve is only slightly different, but
at higher leak rates, as are common in many CES test
facilities (such as the space shuttle which leaks 1.9
%/day), it may be difcult to become aware that a loss
of oxygen is occurring. A substantial cumulative loss
of oxygen in a remote life support system will be life
threatening. It is essential that systems developed and
tested for future remote missions identify cumulative
atmospheric changes that can only be observed when
leak rates are very low. Thus for investigation of subtle
changes in biogeochemical cycles and atmospheric
composition, it is crucial to achieve a high degree of
closure or the differences will be undetectable.
2.4 Carbon Dioxide Dynamics in Biosphere 2
Fluxes of biogeochemical elements can be rapid in
small, closed ecological systems because of the high
concentrations of biotic elements, and comparatively
small atmosphere. Even with Biosphere 2’s volume
of some 200,000 m
3
, 83 kg of carbon will raise the
concentration of CO
2
from 350 to 1500 ppm in the
facility’s atmosphere. Carbon dioxide showed daily
fluctuations of 500-600 ppm on a bright sunny day
(Fig. 5). This large swing is due to the two orders of
magnitude higher ratio of carbon in plant biomass to
atmospheric carbon in Biosphere 2 compared to Earth
(Table 2) and the active photosynthesis of the vegetation
during daylight hours and dominance of plant and soil
respiration during dark night time. The large seasonal
light variation at the project site in southern Arizona (from
9.5 hours of light in winter to 14.5 hours of sunlight in
Fig. 3.
Oxygen concentration in Biosphere 2 September 26,
1991 - January 13, 1993.
14
15
16
17
18
19
20
21
OXYGEN CONCENTRATION, PERCENT
DAY OF CLOSURE
A
B
C
D
E
F
G
G - 10% / DAY
F - 1.9% / DAY
E - 400% / YEAR
D - 100% / YEAR
C - 50% / YEAR
B - BIOSPHERE 2 ACTUAL, 10% / YEAR
A - 0 % / YEAR (PERFECT CLOSURE)
0 100 200 300 400 500
Fig. 4.
Oxygen concentration in Biosphere 2, actual and
simulated for six other leak rates.
CO
2
CONCENTRATION IN BIOSPHERE 2
6000
5000
4000
3000
2000
1000
0
0 100 200 300 400 500 600 700
DAY SINCE CLOSURE
SEPT. 26, 1991 -SEPT. 26, 1993
CO
2
CONCENTRATION, PPM
Fig. 5.
Carbon dioxide average daily concentration in the
Biosphere 2 atmosphere during the two year closure experiment,
1991-1993.
254254
summer) is also reected in Biosphere 2 carbon dioxide
dynamics (Figs. 5 and 6).
Some of the strategies employed for managing
CO
2
included lowering night time temperatures (which
reduces phytorespiration), suspending composting and
minimizing soil disturbance during low-light seasons,
pruning to stimulate regrowth, and drying of biomass for
long-term storage. A physico-chemical precipitator was
also installed inside Biosphere 2 and it was used to lower
carbon dioxide concentrations by 100 ppm per day when
necessary in the winter months (Nelson
et al
, 1994).
2.5 Food Production in Biosphere 2
During the first Biosphere 2 closure experiment
from 1991–1993, approximately 81% of food for the
8-person crew was grown on the 2000 m
2
agriculture
system; the remaining 19% came from seed stock and
food previously grown in Biosphere 2 that was replaced
over the years through successive harvests. (On year
three, 100% of the food was grown inside.) Productivity
was directly related to ambient light levels. Of critical
importance was the fact that 50-55% of the natural
Arizona light was removed by the glass space frame
(45-50% reached the Biosphere agriculture system)
and the 1991-1993 experiment occurred during two el-
Nino years which caused abnormally low light levels.
For example, a wheat crop grown with total light of
679 mol·m
–2
PPF
yielded only 40 g·m
-2
, while a crop
grown with 2022 mol·m
–2
PPF
(nearly three times as
much light) yielded 240 g·m
–2
or six times as much
grain (Silverstone and Nelson, 1994). Experience gained
from Biosphere 2’s 1991-1993 closure experiment led
to improvements that included the addition of some
articial light in the agricultural system that helped make
possible the increased yields and 100% food sufciency
achieved in the second March to September 1994 human
closure experiment (Marino
et al
, 1999).
Over 86 crops (including herbs) were included
in the first closure experiment. The diet for the eight
crew members of Biosphere 2 included milk (from
African pygmy goats) eggs (from the system’s domestic
chickens), meat (from the goats, chickens and Ossabaw
feral pygmy pigs), and sh (from Tilapia grown as in the
rice/azolla water fern paddies). A computer program kept
track of nutrient intake and helped plan forward planting
of crops to ensure a balanced diet. The agricultural area
also produced the fodder necessary for the domestic
animals as well as direct human food crops. The reliance
on ambient sunlight, reduced by 50-55% in passing
through the glazed envelope, limited crop productivity
and might differ in space application and other closed
ecological systems where advantage may be taken of
enhanced artificial light techniques to boost yields and
reduce area required for agriculture.
Other variables, apart from reduced light, may
have also affected crop productivity in Biosphere 2.
These include elevated carbon dioxide concentrations,
decreased oxygen during much of the first closure
mission, the ratio of direct to diffuse light above
the canopy, and the absence of strong winds which
impacts plant strength and pollination. These and other
differences can be further investigated using facilities
like CEEF, Laboratory Biosphere and Mars On Earth®
facilities.
Bi o sp h er e 2 w a s th e fi rs t clo s ed ec ol o gi cal
system to use soil as the plant growth media rather
than hydroponics. There are a number of reasons for
using soil. Since Biosphere 2 was a laboratory for the
study of biospherics, the use of soil made systems far
more comparable to natural ecosystems on Earth. In
addition, buildup of trace gases and pollutants are a
major health and operational concern, the intense and
diverse microbiological biota within soils purifies air
and is critical component of biogeochemical cycles
(Alling
et al
, 1990: Alling
et al
, 1993, Frye and Hodges,
1990). Soil also simplies the waste recycling systems
for animal and human wastes and inedible portions of
crops. Composting and constructed wetlands for biomass
reuse and wastewater treatment are far less energy-
consumptive than alternatives like wet oxidation or
incineration. Agricultural soils, constructed wetlands
and compost systems operate by time-tested biological
mechanisms. The biomass produced in constructed
wetlands can be used for fodder for domestic animals as
was done in Biosphere 2, or returned to the soil through
composting, thus replenishing the nutrients that crops
remove from the soil. Hydroponics, on the other hand,
requires a steady supply of chemical nutrients, fertilizers
and pesticides that cannot be produced from within the
system (Glenn and Frye, 1990, Nelson
et al
, 1994).
Hence, a critical aim for the design of closed ecological
systems for life support is not only supply of a complete,
nutritionally balanced diet, but the maintenance of
sustaining a fertile system.
3500
3000
2500
2000
1500
1000
500
June
December
December 1991
June 1992
Time (day of the month)
CO
2
(ppm)
70
50
30
10
PPF (moles)
m
2
/day
0
7
14
21
28
Fig. 6.
Day-night oscillation of CO
2
, about 600 ppm, is shown
in this graph of a winter (top) and a summer month (lower). The
histogram at bottom also shows total daily light for the winter
month (shaded bar) and for the summer month (clear bar).
255
Alling, A.
et al
.
255
An agricultural system in a materially closed system
must be virtually pollution-free since air, water and
soil buffers are so small and cycling times so rapid.
Introducing pesticides and herbicides risks
contamination and serious health hazards.
With no process to disperse or dilute air and
water pollution; it will just appear as a problem
“somewhere else”. The truth that the “world is
our backyard” is immediately obvious in a small
closed ecological system, but is only beginning to
be understood on a planetary scale. Therefore in
place of toxic chemicals, a variety of biological
and cultural methods of pest and disease control
(known as Integrated Pest Management) must be
utilized for the agricultural crops. IPM techniques
include: selection of resistant crops, small plots
with frequent replantings, switching between
several cultivars (varieties) of the major crops,
maintenance of “beneficial insect” populations
(ladybugs, praying mantis, parasitic wasps etc.)
to control pest insects, intercropping and manual
control when necessary. In addition “safe sprays”
such as soap, light oil or
Bacillus thuringensis
can
be used and were used in Biosphere 2 (Silverstone
and Nelson, 1996).
3.0 Laboratory Biosphere
Since 2002, a series of closed ecological
system experiments has been conducted in the
Laboratory Biosphere, a 34-43 m
3
closed system
(Fig. 7) in Santa Fe, New Mexico, USA. The
Laboratory Biosphere was created as a testbed
to continue experiments using a sustainable soil
based agriculture system, artificial lights and
the small volumes of components more suitable
for space life support (Nelson
et al
, 2003b,
Dempster
et al
, 2004). Table 3 summarizes
the mass and volume of the components in the
Laboratory Biosphere closed ecological facility. Initial
experiments, 2002-2004, used crops of soybeans, dwarf
wheat, compact sweet potatoes, pinto bean and cowpea
to study biogeochemical cycles, atmospheric dynamics,
crop productivity, nutrient recycling, soil fertility, light
efciency and observations of Mir 3
rd
generation wheat
seed.
3.1 Laboratory Biosphere Food Production
Interesting observations include the experimentation
with sweet potato (TU-82-155) that had a 50% higher
productivity using the soils system than trials using
hydroponic methods with the same variety of sweet
potato at NASA Johnson Space Center (Barta, 2001;
Nelson
et al
, 2005). The projected yields of sweet
potato for the Mars On Earth facility® were initially
extrapolated at twice Biosphere 2’s best yields of 16g
m
-2
d
-1
(wet weight at 25 mol m
-2
d
-1
light) or 32 g m
-2
d
-1
(wet) with the 50 mol m
-2
d
-1
of light planned for
the facility (Silverstone
et al
, 2003). The experiment in
the Laboratory Biosphere at around 44 mol m
-2
d
-1
light
averaged 58.7 g m
-2
d
-1
(wet) which was therefore 83%
higher than the yield data projected for the MOE facility
Fig. 7. Laboratory Biosphere Facility, Santa Fe, New Mexico,
USA.
Laboratory Biosphere Data Wheat Sweet Potato
Planting density: plants/m
2
,
East side / West side
400/800 16 / 16
Light regime hours
on / hours off
16/8 18/6
Light intensity
mmol m
-2
s
-1
840 699
Daily light
mol m
-2
d
-1
(typical)
48.4 45.3
Days from planting to harvest 87 126
Total light
mol m
-2
(actual)
3981 5568
Total dry biomass, g,
East side / West side *
3104/4387 6800/ 4907
Average dry biomass: g m
-2
East side / West side
1156/1634 2533/ 1828
Average dry biomass rate, g m
-2
d
-1
East side / West side
13.3/18.8 20.1/ 14.5
Average dry edible, g m
-2
East side / West side
566.5/812.3 1657/ 1202
Average dry edible rate, g m
-2
d
-1
East side / West side
6.5/9.3 13.1/ 9.5
Overall Harvest index 0.49 0.66
Light efciency-biomass, g/mol
East side / West side
0.29/0.41 0.45/ 0.33
Light efciency, dry edible, g/mol,
East side / West side
0.14/0.2 0.30/ 0.22
Table 3
Summary of light input and results from the wheat and sweet
potato experiments in the Laboratory Biosphere.
*Dry biomass for sweet potato tubers was estimated by drying samples to
determine a dry weight to wet weight ratio and multiplying wet weights by
the ratio so obtained which was 0.194.
256256
(Table 6). This demonstrates the importance of testing
crop cultivars and growing conditions in a closed system
that is the size, scale and using the proposed engineering
systems for a Mars base.
Compared with Biosphere 2 production of wheat
at relatively low light (6.4 - 16.4 mol m
-2
d
-1
), with
average wheat production of 120 g seed m
-2
, the average
production from this study is over 5 times higher, and
the best area over 7 times higher. At the highest light in
Biosphere 2, 16.4 mol m
-2
d
-1
, yield was 240 g m
-2
about
1/3 that of the current study; total light was 2022 mol m
-2
(about half of this study) and light efciency was 0.12
g seed/mole, about 30% less efficient than the wheat
crop in the Laboratory Biosphere (Nelson
et al
, in press:
Silverstone and Nelson, 1996).
The projected wheat yields for the MOE facility (Table
5), derived using three times the average Biosphere 2
wheat yields to reflect the increased light levels of 50
mol m
-2
d
-1
planned for MOE were estimated at 7.3 g m
-2
d
-1
(Salisbury
et al
, 2002; Silverstone
et al
, 2003). The
recent yields in the Laboratory Biosphere (7.9 g m
-2
d
-1
)
were thus 8% better than those projections.
Laboratory Biosphere crop results (Table 3) suggest
that soil-based systems are comparable with and
sometimes out produce the yield of hydroponic systems
which have predominated previous bioregenerative space
life support systems. This is signicant since soils will
reduce dependence on highly technical systems which
rely on consumables like hydroponic chemicals as well
as provide soil microbiota that will assist in completion
of biogeochemical cycles and reduction of problematic
trace gases (Alling
et al
, 1990; Frye and Hodges, 1990;
Bohn, 1972). In addition, both the sweet potato and
wheat results show that the increase in light provided in
the Laboratory Biosphere as compared with Biosphere
2 results in greater yields. This is turn will enable a
reduction in space requirements for food production,
and support the strategy of increasing light supplied to
optimize plant efciency and productivity (Salisbury
et
al
, 2002).
3.2 Laboratory Biosphere Atmospheric Dynamics
On planet Earth humans have the great luxury of
nearly invariant atmospheric composition. The opposing
processes of photosynthesis by plant life and respiration
by animal life (especially soil respiration), as moderated
by the biogeosphere as a whole, are so nely balanced
with each other that it has required the development of
advanced technology within the last century to detect
any variations at all. The Biosphere 2 experiment
conclusively demonstrated that in small closed systems,
the atmosphere was the most dynamic component of the
biosphere. And further, that the atmosphere was far more
interactive with life in the soil and water then previously
understood. Now we contemplate constructing small
biospheres that can sustainably support a few humans in
food, water and air. To do that, a sufcient approximation
to balance photosynthesis and respiration for human life
support must be achieved. A balanced system cannot
be assembled without a detailed understanding of the
behavior of its most important components including
food crops as they progress from seedlings to harvest.
Recent successive experiments in Laboratory
Biosphere with first wheat and then sweet potato
crops provided data on total biomass and edible food
production, water evapotranspiration rates and the
fixation rate of carbon dioxide at every stage of crop
growth. Signicant differences between these two crops
were observed (Dempster
et al
, 2005). Figure 8 shows
the CO
2
concentration during each experiment.
While these CO
2
concentration curves may seem very
similar, important differences are revealed by a study
of rates of net fixation and respiration determined by
the first time derivatives taken at hundreds of selected
segments of each curve. The net fixation rates are
presented in Figs. 9 and 10. Fixation is represented as
a negative number and respiration as positive which
enables both processes to be shown on the same graph
without overlap of these opposites.
This analysis reveals very noticeable differences
between the two crops. Wheat rapidly increases net
xation rates to a maximum circa days 25 – 30 and then
net xation decreases almost as rapidly to a denite end
point just before harvest circa day 80. This corresponded
with the qualitative observation that the plants became
brow n and dry wi th full y develo p e d seed while
vegetative growth had ended at about the same time. In
contrast, net fixation by sweet potato increased, not as
rapidly, up to about day 40 and then with some variations
continued at roughly the same rate until the arbitrarily
exercised harvest at day 126. Vegetative growth was
0
2000
4000
6000
8000
0 20 40 60 80
CO2, PPM
DAYS AFTER PLANTING
LAST CO2
INJECTION
FIRST CO2
INJECTION
PERSONS
BREATHING
24-HR. RESPIRATON
EXCEEDS FIXATION
24-HR. FIXATION MATCHES
RESPIRATION HERE
24-HR. FIXATION
EXCEEDS RESPIRATION
WHEAT
0
2000
4000
6000
8000
10000
12000
0 20 40 60 80 100 120
CO2, PPM
DAYS AFTER PLANTING
24-HR. FIXATION MATCHES
RESPIRATION HERE
24-HR. RESPIRATION
EXCEEDS FIXATION
24-HR. FIXATION
EXCEEDS RESPIRATION
FIRST CO2
INJECTION
LAST CO2 INJECTION
PERSONS
BREATHING
SWEET
POTATO
Fig. 8.
CO
2
in Laboratory Biosphere during wheat and sweet
potato experiments. CO
2
initially rises while the new plants
are very small then falls rapidly as they grow larger and
photosynthesis increases. 58 injections maintain CO
2
levels for
wheat (days 15 – 69) and 91 injections for sweet potato (days 20
– 124).
257
Alling, A.
et al
.
257
evidently still unabated and may well have continued for
much longer.
For both crops, many net fixation and respiration
rates were determined at differing CO
2
concentrations.
Points are plotted for both 1200 ppm (squares) and 2000
ppm (dots) in these two graphs (Figs. 9 and 10). While
there may be some isolated exceptions, it is generally
seen that net xation rates are greater at the higher CO
2
concentration. This suggests that vegetative growth is
enhanced with increasing CO
2
, at least up to 2000 ppm.
These represent two of the many required building blocks
of data that can contribute to assembling a complex
system such as Mars On Earth®.
4. Mars On Earth® Project
Biosphere Foundation’s Mars On Earth® (MOE)
project will simulate a 4 person sustainable life support
system designed for Mars. This closed system, which
has a footprint of approximately 800 m
2
, will be used
to develop space-based life support systems, such as
water and wastewater recycling, food production, air
purification, etc. and to develop space engineering
and technology for the Mars Base. The test bed will
be atmospherically closed to examine biogeochemical
processes, but open to information, energy and certain
material exchanges. As a modular system, the Mars Base
will have the exibility to add on chambers – hence the
number humans and their length of occupation inside
the system will vary depending upon the experiment
protocol. In particular, the overall design will address
not only the functional requirements for maintaining
long term human habitation in a sustainable artificial
environment, but the aesthetic need for beauty and
diversity of foods which has been noticeably lacking in
most space settlement designs.
Once the facility has demonstrated that a biospheric
life support system is feasible and is desirable for
humans to inhabit, human future in space will become
a real possibility. People will realize that it is a very
exciting frontier to be explored. This investigation into
life support systems will not only yield data for space
exploration but also information that can be used to
understand and preserve the ecological health of our
own planet. The goal is to produce a complete diet and
to recycle all waste products including human waste
from the crew (Allen and Alling, 2002; Silverstone
et al
.
2003).
The proposed opaque structure, designed to have 4
modular units each composed of approximately 5000
m
3
(total of 20,000 m
3
), will be pressurized at about 1-2
psi above ambient. Two of the units will be used for
eld agriculture, each with two levels totaling 245 m
2
,
and another with 125 m
2
as a horticulture garden and
utility water tank/pond storage. The fourth unit will be
a 250 m
2
habitat with two levels for the kitchen, ofce/
communication center, food processing, living quarters
and Wastewater Garden for processing human waste. The
agriculture/horticulture design is based on 50-70 mol m
-2
d
-1
of articial light on a schedule of 14 – 18 hours/day
depending on crop requirements. The electrical load
required to operate the lights, both for plant growth
and human needs, pumps, cooling, and other technical
-1 50
-1 00
-50
0
50
0 20 40 60 80
FIXATION & RESPIRATON RATES mmol h
-1
m
-2
D A Y S A F T E R P L A N TIN G
D A R K R E S P IR AT IO N
LIG H T
FIX A T IO N
12 00 P P M (SQ U A R E S )
20 00 P P M (D OT S )
W H E AT
Fig. 9.
Fixation and respiration rates for wheat at 1200 ppm
(squares) and 2000 ppm (dots).
-60
-40
-20
0
20
40
0 20 40 60 80 100 120
FIXATION & RESPIRATON RATES, mmol h
-1
m
-2
DAYS AFTER PLANTING
DARK RESPIRATION
LIGHT
FIXATION
AT 1200 PPM (SQUARES)
AT 2000 PPM (DOTS)
SWEET
POTATO
Fig. 10.
Fixation and respiration rates for sweet potato at 1200
ppm (squares) and 2000 ppm (dots).
258258
equipment is estimated to be around 1.2 – 1.5 MW.
Potable water will be generated using methods of
condensation with back up UV sterilizers for further
treatment. The irrigation water will be a mix of subsoil
drainage, potable water, utility water and treated
wastewater. A mixing system will be available to mix
the water prior to irrigation. A 28,000 liter (7,000
gallon) utility water tank will be included as the main
water reservoir given that at any one time the amount
of water that can be held in the soil may vary due to
differing harvest strategies. This leaves a reserve that
will be held in an open tank simulating a natural pond
environment located in the horticulture chamber. The
pond will simulate a rocky pool with wild aquatic
ecology and will provide the drew with the aesthetic
beauty of a wilderness pool. The waste recycling system
will combine an anaerobic septic tank and a subsurface
flow wetland plant treatment facility, technology we
developed called Wastewater Gardens® (Nelson
et al
,
2002). The system will be able to handle 40 liters (10
gallons) per person per day with a four-
day residence period at an overall area
of 9.3 m
2
(100 sq ft). The depth of the
wetland subsurface marsh tank will be 0.6
m (2 ft). Plants will be selected for edible
use such as banana and papaya trees,
along with other owering shrubs, such as
canna lily and arrowheads.
The MOE diet, 3000 kcal, 79 g protein
and 35 g of fat per person/day, has been
designed using the data and experience
from Biosphere 2 and the Laboratory
Bi o s p he re (N e l s on
e t a l
, i n pr e s s;
Silverstone
et al
, in press; Silverstone
et
al
2003; Walford 1992). While we are
presently using the Laboratory Biosphere
to test other MOE crop candidates, the
system is designed to provide a diet that
will utilize ten crops (Table 4) selected
bec a use of th eir p r ove d succ e ss in
Biosphere 2: wheat, rice, sweet
potato, peanut, soybean, pinto,
beetroot, winter squash, banana
and papaya supplemented with
vegetables, herbs, and spices.
T h e s e c r o p s a r e p r o d u c t i v e ,
hardy, dependable and relatively
easy to harvest and process with
a minimum of equipment. The
ease of processing is an important
consideration given the desire to
minimize crew time necessary for
food production.
Table 5 shows some estimated
yield figures and necessary crop
growing area given light input of
50 mol m
-2
d
-1
. The yield figures
taken for the basis of these calculations are based on
extrapolations from the best producing crop of each
type grown in Biosphere 2 during the first two-year
closure experiment. Apart from the wheat, these crops
were grown during the second summer with an average
light level of 25 mol m
-2
d
-1
. For all the crops except
wheat, papaya and banana, a doubling of yield was
extrapolated with light increased to 50 mol m
-2
d
-1
. These
yield gures have yet to be proven for the crops listed;
however work carried out at Utah State University
(Bugbee
et al
1988; Salisbury
et al
, 1987) on the limits
of crop productivity would indicate that these yields are
reasonably conservative using a light level of 50 mol m
-2
d
-1
. After further research, design scenarios may call for
less growing area to supply a Mars exploration crew of
4-5. The yield projected for banana and papaya are based
on the production rates obtained in Biosphere 2. The
wheat crop with the best yield during the rst Biosphere
2 mission was grown in the winter with an average light
level of 16 mol m
-2
d
-1
.
Crop
kcal/person
per day
Grams of crop
per day
Protein from
crop per day
Fat from crop/
day
Rice
450 128.57 16.71 1.29
Wheat
300 90.91 11.82 1.82
Sweet Potato
750 707.55 7.08 1.98
Peanut
150 25.68 6.68 12.33
Soybean
150 37.31 2.99 6.72
Pinto bean
300 87.72 21.05 0.75
Winter squash
225 354.89 3.55 0.35
Beet root
225 505.62 5.06 0.1
Banana
300 500 3 10
Papaya
150 576.92 1.73 0.4
Total 3000 3015.18 79.66 35.74
Table 4
Calculation of protein & fat from ten crops in the MOE facility based on 3000 kcal/
person/day.
Crop Daily kcal
Yield:
50 mol/ m
2
/d
Yield:
kcal/ m
2
/d
Area required
(m
2
)
Rice
1800 0.011 40.55 44.39
Wheat
1200 0.007 24.38 49.22
Sweet Potato
3000 0.032 33.69 88.51
Peanut
600 0.003 16.32 36.77
Soybean
600 0.003 10.64 56.41
Pinto bean
1200 0.007 25.36 27.32
Winter squash
900 0.085 54.32 16.57
Beet root
900 0.046 20.45 44.01
Banana
1200 0.050 29.64 40.48
Papaya
600 0.108 28.68 20.92
Salad &
Greens
33
Totals 12000 478
Table 5
Estimated area needed for a complete nutritional diet for the MOE 4 person
crew using ten major crops. The estimates are based on Biosphere 2 yields and 50
mol/m
2
/d.
259
Alling, A.
et al
.
259
Results from the MOE facility will be important in
demonstrating what crops can produce reliable yields
and which ones should be selected for a Mars mission
to produce a complete nutritional diet. The facility will
also be used to determine the most efficient types of
lights, optimal light levels for growth of a variety of
crops, energy trade-offs for agriculture (e.g. energy
required for given light intensity vs. required cropping
area), capabilities of Martian soils and their need for
enrichment and elimination of oxides, strategies for use
of human waste products, and maintaining atmospheric
balance between people, plants and soils.
References
Allen, J. (2002) People challenges in biospheric systems for
long-term habitation in remote areas, space stations, moon,
and Mars expeditions.
Life Support Biosph. Sci.
,
8
, 67 -70.
Allen, J. (2000) Artificial Biospheres as a model for global
ecology on planet earth.
Life Support Biosph. Sci.
,
7
, 1-10.
Allen, J., Alling, A. (2002) The design approach for Mars On
Earth®, a biospheric closed system testing facility for long-
term space habitation.
American Institute of Aeronautics and
Astronautics. IAC-02-IAA.8.2.02.
Allen, J., Nelson, M. (1999) Biospherics and biosphere 2
mission one (1991-1993).
J. Ecol. Eng.
,
13
, 15-29.
Allen, J.P., Nelson, M., Alling, A. (2003) The legacy of
Biosphere 2 for the study of biospherics and closed
ecological systems.
Adv. Space Res.
,
31
(7), 1629-1640.
Alling, A, Nelson, M., Silverstone, S., Van Thillo, M. (2002)
Human factor observations of the Biosphere 2, 1991 – 1993,
closed life support human experiment and its application to
long-term manned mission to Mars.
Life Support Biosph.
Sci.
,
8
, 71 -82.
Alling, A., Leigh, L., MacCallum, T., Alvarez-Romo, N. (1990)
Biosphere 2 Test Module Experimentation Program. In
Biological Life Support Systems
(eds by Nelson, M. and
Soffen, G.A.) Synergetic Press, Tucson, 23-32.
Alling, A., Nelson, M., Leigh, L., Frye, R. MacCallum, T.,
Alvarez-Romo, N. (1993) A summary of the Biosphere
2 Test Module Research. In
Microcosms and Mesocosms
in Scientific Research
(eds. Beyers, R. and Odum, H.T.)
Springer-Verlag, New York, 1993.
Barta, D.J., Henderson, .K., Desmond, M., Henninger, D.L.
(2001) Gas exchange, transpiration and yield of sweet
potato grown in a controlled environment.
Life Support
Biosph. Sci.
,
7
(1), 46.
[Abstract only]
Bol i n , B. and Co ok, R. B. (1 9 83) (E ds.).
The majo r
biogeochemical cycles and their interactions
, John Wiley &
Sons, NY, 1983.
Bohn, H.L. (1972) Soil adsorption of air pollutants.
J. Environ.
Qual.
,
1
, 372-377.
Bugbee, B., Salisbury, F.B. (1988) Exploring the limits of crop
productivity. 1. Photosynthetic efciency of wheat in high
irradiance environments.
Plant Physiol.
,
88
, 869-878.
Dempster, W.F. Tightly closed ecological systems reveal
atmospheric subtleties - experience from Biosphere 2.
Presented at COSPAR, Paris, July 2004.
Adv. Space Res
. [in
press]
Dempster, W.F. (1994) Methods for Measurement and
Control of Leakage in CELSS and their Application and
Performance in the Biosphere 2 Facility.
Adv. Space Res.
,
14
(11), 331-335.
Dempster, W.F., Allen, J.P., Alling, A., Silverstone, S., Van
Thillo, M. (2005) Atmospheric dynamics in the “Laboratory
Biosphere” with wheat and sweet potato crops.
Adv. Space
Res.,
35
(9), 1552-1556.
Dempster, W.F., Van Thillo, M., Alling, A., Allen, J.P.,
Silverstone, S., Nelson, M. (2004) Technical review of the
Laboratory Biosphere closed ecological system facility.
Adv.
Space Res.
,
34
, 1477-1482.
Frye, R., Hodges, C. (1990) Soil Bed Reactor Work of the
Environmental Research Laboratory of the University of
Arizona in Support of the Research and Development of
Biosphere 2, In
Biological Life Support Technologies
,
(eds Nelson, M. and Soffen, G.A.), NASA CP-3094 and
Synergetic Press, Oracle, AZ: pp. 33-40.
Glenn, E.P., Frye, R. (1990) Soil Bed Reactors as Endogenous
Control Systems of CELSS, In
Workshop on Artificial
Ecological Systems
, Proceedings of Meeting in Marseilles,
France, October, sponsored by DARA and CNES, 41-58.
Marino, B., Mahato, T.R., Druitt, J.W., Leigh, L., Lin, G.,
Russell, R.M., Tubiello, F.N. (1999) The agricultural biome
of Biosphere 2: Structure, composition and function.
Ecol.
Eng.
,
13
, 199-234.
Morowitz, H., Allen, J.P., Nelson, M., Alling, A. (2005) Closure
as a Scientific Concept and its Application to Ecosystem
Ecology and the Science of the Biosphere.
Adv. Space Res.
,
36
(7), 1305-1311.
Nelson, M., Dempster, W.F., Alvarez-Romo, N., MacCallum,
T. (1994) Atmospheric Dynamics and Bioregenerative
Technologies in a Soil-Based Ecological Life Support
System: Initial Results from Biosphere 2.
Adv. Space Res.
,
14
(11), 417-426.
Nelson, M., Dempster, W.F., Silverstone, S., Alling, A., Allen,
J.P., Van Thillo, M. (2005) Crop Yield and Light/Energy
Efciency in a Closed Ecological System: Two Laboratory
Biosphere Experiments.
Adv. Space Res.
,
35
(9), 1539-1543.
Nelson, M., Allen, J., Alling, A., Dempster, W.F., Silverstone,
S. (2003a) Earth applications of closed ecological systems:
relevance to the development of sustainability in our global
biosphere.
Adv. Space Res.
,
31
(7), 1649-1656.
Nelson, M., Alling, A, Dempster, W.F., Van Thillo, M., Allen, J.
(2002) Potential integration of wetland wastewater treatment
with space life support systems.
Life Support Biosph. Sci.
,
8
(3-4), 149-154.
Nel s o n , M. , De m p s ter, W. F., A l ling, A., A llen, J.P.,
Rasmussen, R., Silverstone, S., Van Thillo, M. (2003b)
Initial experimental results from the Laboratory Biosphere
closed ecological system facility.
Adv. Space Res.
,
31
(7),
1721-1730.
Nelson, M., Leigh, L., Alling, A.
et al
. (1992) Biosphere 2 Test
Module: A Ground-Based Sunlight-Driven Prototype of a
Closed Ecological System.
Adv. Space Res.
,
12
(5) 151-156.
Salisbury, F.B, Dempster, W.F., Allen, J.P., Alling, A,
Bubenheim, D., Nelson, M., Silverstone, S. (2002) Light,
260260
plants, and power for life support on Mars.
Life Support
Biosph. Sci.
,
8
(3-4), 161-172.
Salisbury, F.B., Bugbee, B., Bubenheim, D. (1987) Wheat
production in controlled environments,
Adv. Space Res.
,
7
(4), 123-132.
Schlesinger, W.H. (1991)
Biogeochemistry: An Analysis of
Global Change
, Academic Press, NY, 1991.
Silverstone, S., Nelson, M. (1996) Food Production and
nutrition in Biosphere 2: results from first mission,
September 1991-September 1993.
Adv. Space Res.
,
18
(4-5),
49-61.
Silverstone, S., Nelson, M. Alling, A., Allen, J.P. (2005) Soil
and crop management experiments in the Laboratory
Biosphere: an analogue system for the Mars on Earth®
facility.
Adv. Space Res.
,
35
(9), 1544-1551.
Silverstone, S., Nelson, M. Alling, A., Allen, J. (2003)
Development and research program for a soil-based
bioregenerative agriculture system to feed a four person
crew at a Mars base.
Adv. Space Res.
,
31
(1), 69-75.
Walford, R.L., Harris, S.B., Gunion, M.W. (1992) The
calorically restricted low-far nutrient-dense diet in Biosphere
2 signicantly lowers blood glucose, total leukocyte count,
cholesterol, and blood pressure in humans.
Proc. Natl. Acad.
Sci.
,
89
(11), 533-537.
