CControl of habitat’s carbon dioxide level by biomass burning
Pekka Janhunen a, ∗ a Finnish Meteorological Institute, Helsinki, Finland
Abstract
Consider a free-space settlement with a closed ecosystem. Controlling the habitat’s carbon dioxide level is a nontrivialproblem because the atmospheric carbon bu ff er per biosphere area is smaller than on Earth. Here we show that theproblem can be solved by burning agricultural waste. Waste biomass is stored and dried, and burned whenever plantgrowth has lowered the atmospheric carbon dioxide level so that replenishment is needed. The method is robust, low-tech and scalable. The method also leaves the partial pressure of oxygen unchanged. In the initial growth phase of thebiosphere, one can obtain the carbon dioxide by burning sugar or carbon, which can be sourced from carbonaceousasteroid materials. This makes it possible to bootstrap the biosphere without massive biomass imports from Earth. Keywords: space settlement, closed ecosystem, carbon cycle
1. Introduction
Space settlements need a nearly closed ecosystem forfood production. One of the fundamental parts of aclosed ecosystem is the carbon cycle. In the carboncycle (Fig. 1), plants fix carbon from atmospheric CO by photosynthesis, producing biomass [approximatelysugar, net formula n (CH O)] and liberating oxygen,CO + H O + light → CH O + O . (1)The biomass is consumed and metabolised by decom-posers, animals and people. Metabolism is the reversereaction of photosynthesis,CH O + O → CO + H O + energy . (2)On Earth, the atmospheric CO and the biosphericCH O contain comparable amounts of carbon. This isso because the amount of carbon in the atmosphericCO is 1.66 kgC / m , while the world average biosphericcarbon is 1.08 kgC / m [1] . Because the atmosphericcarbon bu ff er is large, on Earth the atmospheric CO level is not sensitive to fluctuations in the primary pro-duction of the biosphere. The Earth’s atmosphere ismassive (10 tonnes per square metre), while most of ∗ Corresponding author
Email address: [email protected] (Pekka Janhunen)
URL: (Pekka Janhunen)
550 billion tonnes of carbon[1, Table 1] is 1.08 kgC / m . CO O Photo-synthesis CO O Metabolism,decompositionEarth’s atmosphereLarge pool of N , O and CO Primary producers(autotrophs)–Natural plants–Agricultural plants Consumers(heterotrophs)–Herbivores,predators,decomposers–Bacteria, fungi,animals, peopleCH OBiomass(eating)
Figure 1: Carbon cycle on Earth.
Earth’s surface area is open ocean, desert or glacier sothat the globally averaged biomass areal density is onlymoderate. For example in average African tropical rain-forest, the carbon stock is 18.3 kgC / m i.e. 183 Mg / ha[2, Table 2], which is as much as 17 times larger thanthe global average.In a space settlement, the atmosphere mass is likelyto be much less than 10 tonnes / m . In O’Neill’s originallarge habitat concepts [3, 4], the atmosphere had sev-eral kilometres depth. However, a massive atmosphereincludes a lot of nitrogen. Nitrogen is not too abun-dant on asteroids, and would only be widely availablein the outer solar system. One way to avoid the nitro- NSS Space Settlement Journal August 13, 2019 a r X i v : . [ phy s i c s . pop - ph ] A ug FEASIBILITY OF A CLOSED ECOSYSTEM gen supply problem would be to use a reduced pressurepure oxygen atmosphere, but then the risk of fire wouldbe increased since the flame is not cooled by inert gas.Also birds and insects (needed for pollination) wouldhave di ffi culty in flying in a pure oxygen atmosphere,because its mass density would be several times lessthan on Earth. Hence it is likely that most settlementswould prefer to use a shallower N / O atmosphere ofe.g. ∼
50 m depth [5]. A 50 m height allows forestswith maximum tree height of ∼
30 m plus some roomfor horizontal winds to mix gases above the treetops.The nitrogen (47 kg / m ) can be obtained from the as-teroids, as a byproduct of the mining that produces thecombined structures and radiation shielding of the set-tlement (10 kg / m ).Carbon dioxide is necessary for plants to grow. Tomaintain good growth, the concentration should be atleast ∼
300 ppmv (parts per million by volume). Thepre-industrial level on Earth was 280 ppmv, which, aswe know, already allowed plants to grow reasonablywell. On the other hand, for human safety the amountshould not exceed ∼ bybreathing.A shallow atmosphere is unable to absorb fluctua-tions in the biomass carbon pool while keeping the CO level within safe bounds. The timescales can be ratherfast. A tropical rainforest can bind 2.0 kgC / m / year[6, 7], so in a shallow 50 m atmosphere, maximal plantgrowth could reduce the concentration of CO by 1000ppmv in as short time as 4.5 days. In temperate forestthe rate of biomass production is somewhat less (1.25kgC / m / year) and in cultivated areas even less (0.65kgC / m / year)[6, 7], but the timescales are still onlyweeks. Hence the atmospheric CO must be controlledby technical means, which is the topic of this paper.
2. Feasibility of a closed ecosystem
There are many examples of nearly semi-closed smallecosystems that interact with the rest of Earth’s bio-sphere mainly via air only: a potted flower, a vivarium, afenced garden, a small island, etc. To turn a semi-closedsystem into a fully closed one, one only needs to worryabout a few gases. This is an engineering task, wherethe complexity of biology has been factored out. Morespecifically, there are five parameters to consider: 1. O partial pressure. Oxygen is needed for hu-mans and animals to breath, and the partial pres-sure should be about 0.21 bar.2. N partial pressure. Nitrogen is needed for firesafety and for birds and insects to fly, and the par-tial pressure should be about 0.79 bar.
3. CO concentration. Carbon dioxide is needed byplants to grow, but too high a value is unsafe topeople. The allowed range is 300–2000 ppmv.4. CH concentration. Methane is not needed so thelower limit is zero, but if generated by the bio-sphere, it is tolerable up to 30 mbar, which is wellbelow the ignition limit of 44 mbar. Methane’sonly health e ff ect is oxygen displacement, whichis however negligible at 0.03 bar.5. Other gases should remain at low concentration.Considering oxygen, a biosphere does not fix itfrom the atmosphere. The oxygen atoms that biomassCH O contains originate from the water that entersphotosynthesis. When organisms do metabolism andbreathe (Eq. 2), they transform O molecules into CO molecules, but the process involves no net transfer of Oatoms from the atmosphere into the body. Hence onedoes not need to do anything special to maintain theright O partial pressure.Considering N , a biosphere fixes some of it since ni-trogen is a key nutrient, present in proteins and DNA.The C:N ratio of cropland soil is 13.2 and for otherbiomes it varies between 10.1 and 30 [8, Table 1]. Forleaves, wood and roots the C:N ratio is higher [8]. Toget an upper limit, the carbon stock of average Africanrainforest is 18.3 kgC / m [2]. With the minimal soilC:N ratio across biomes of 10.1, this corresponds to1.81 kgN / m of fixed nitrogen. But the mass of nitro-gen in a 50 m high atmosphere is 46 kgN / m , so clearlythe biosphere can assimilate only a small fraction of at-mospheric N . Hence one does not need to do anythingspecial with N , either. Its partial pressure will remainsu ffi ciently close to the initial value. Circulation of ni-trogen from the point of view of nutrient supply is arelated topic [9], which is however outside the scope ofthis paper.Thus, since N and O are not changed too muchby the biosphere, the task of maintaining a good atmo-sphere is reduced to three issues:1. Maintaining CO within the 300–2000 ppmvbounds. This is treated in the next section. We do not consider argon and other noble gases because they areeven less abundant on asteroids than N . Also, at high concentrationssome noble gases have narcotic e ff ects. NSS Space Settlement Journal
BIOMASS BURNING
2. Ensuring that if net methane is emitted by the bio-sphere, its concentration does not increase beyond ∼
3. Ensuring that the concentration of other gases staylow. This may possibly happen automatically, be-cause plants are known to remove impurities fromair [10]. We shall say a bit more on this in the Dis-cussion section below.
3. Biomass burning
Above we described the carbon cycle problem of theorbital space settlement. The problem is that the settle-ment’s atmosphere is much shallower than on Earth, andhence the atmospheric carbon bu ff er is much smallerthan the biospheric carbon stock. Fluctuations in theamount of biospheric carbon can occur for many rea-sons, and the fluctuations would cause the atmosphericCO concentration to go o ff bounds.A way to solve the problem is to store some biomassand to burn it when the atmosphere needs more CO (Fig. 2). Agricultural waste is a necessary byproductof food production. One stores the waste biomass insuch a way that it does not decompose and then burns itat a controlled rate. Methods to store biomass includedrying, freezing and freeze-drying. Drying is feasible atleast if the relative humidity is not too high.It is su ffi cient for only part of the biomass to gothrough the storing and burning pathway. The higherthe burned fraction is, the larger is the CO control au-thority of the scheme. The control authority is su ffi cientif the total amount of carbon in the settlement exceedsthe maximum mass of carbon that can be fixed in liv-ing organisms at any one time. When the atmosphericCO drops below a target value, one burns some storedbiomass. If there is too much of CO in the atmosphere,one ceases the burning activity for a while. After somedelay plant growth will take down the CO concentra-tion.Burning consumes oxygen, but the same amount ofoxygen is liberated into the atmosphere when the CO is used by photosynthesis (Eqs. 1 and 2). Thus the O concentration stays constant, apart from an insignificant On Earth the methane concentration is 1.8 ppmv, which is re-sponsible for part of the terrestrial greenhouse e ff ect. For atmosphericheight of 50 m, a similar greenhouse e ff ect arises at 200 times higherconcentration, i.e. at 360 ppmv. Thus a 3 % (30,000 ppmv) methaneconcentration would cause a significant greenhouse e ff ect for a 50 matmosphere, which should be taken into account in the settlement’sheat budget. Greenhouse e ff ects are nonlinear so quantitative predic-tion would need modelling. part that exists temporarily as CO . This is especiallyadvantageous in the build-up phase of the biosphere. Inthe build-up phase, one needs to add carbon constantlyto the atmosphere, as trees and other plants are grow-ing. Depending on the type of ecosystem we are build-ing, the growth phase might last up to tens or even hun-dreds of years as trees grow and the soil builds up. Itis not necessary to wait for the growth phase to finishuntil people can move in, but while the growth phaseis ongoing, one must be prepared to put in new car-bon as needed to avoid CO starvation. If this carbonwould be added in the form of new CO from an ex-ternal tank, for example, the level of atmospheric oxy-gen would build up. However, if one adds the carbonby burning biomass, sugar or carbon, the O level staysconstant. Carbon can be sourced from carbonaceousasteroids. Possibly sugar [net formula n (CH O)] can besynthesised from C-type asteroids as well. Thus the bio-sphere can be bootstrapped without massive importingof biomass from Earth.When burning biomass, the rate must be controllableand fire safety must be maintained. One also wants tominimize smoke production (particulate emission), be-cause otherwise the settlement’s sunlight-passing win-dows would need frequent washing and because wewant to avoid atmospheric pollution [11]. One wayto facilitate clean burning is to mechanically processthe biomass (or part of it which is used in the ignitionphase) into some standardised form such as pellets [12]or wood chips. It is also possible to use a bioreactorto turn the biomass into biogas (methane) which burnswithout smoke. To further reduce smoke, one might addan electrostatic smoke precipitator in the smokestack. Acombination of approaches is also possible. One can ig-nite the flame using easy fuel and then continue withmore unprocessed material. The burning activity couldbe continuous, but in a 50 m high atmosphere, enoughconstant CO is reached by a daily burning session.Atmospheric pollution should be avoided, so smokeproduction should be minimised. However, plants andsoil are known to clean up the atmosphere rather well[10]. Hopefully, if the above measures to promote cleanburning are used, the plants can accomplish the rest sothat the atmosphere remains clean. To investigate thequestion experimentally, one could burn biomass insidea greenhouse by di ff erent methods, while using standardair quality monitoring equipment for measuring the at-mosphere. Burning hydrocarbons ( ∼ CH ) in the buildup phase is not rec-ommended, because then net consumption of O would take place asoxygen would be bound with hydrogen to make water. NSS Space Settlement Journal
DISCUSSION CO O Photo-synthesis CO O Metabolism,decomposition CO O ControlledburningShallow atmosphere (N , O ), small pool of CO Primaryproducers(plants) Consumers,incl. people Stored drybiomassbankBiomass(Eating) Agricultural waste
Figure 2: Carbon cycle in the settlement.
In a rainforest, the maximum carbon fixation rateis 2 kgC / m / year and in a cultivated area it is 0.65kgC / m / year (see last paragraph of Introduction). Ifthe average is ∼ / m / year and if 50 % of it isburned while the remaining part is decomposed nat-urally or eaten as crop, then the burned amount is0.5 kgC / m / year, which corresponds to 34 kg of drybiomass per hectare per day. When wood is burned,the mass fraction of ash varies between 0.43 and 1.8per cent [13, Table 1], so that the ash produced is a fewhundred grams per day per hectare. The ash must be dis-tributed evenly back into the environment. The amountof ash is modest enough that the settlers could even dothe spreading manually if they wish. The heat producedby the burning is of the order of 0.8 W / m as a temporalaverage, which is two orders of magnitude less than theheat dissipation of sunlight, or artificial light if that isemployed.In reality, a smaller burning rate than this calcula-tion would probably su ffi ce. It is only necessary to burnenough biomass to maintain su ffi cient control authorityof the CO level. Burning as much 50 % of the growthis likely to be overkill, but we assume it to arrive at aconservative estimate.Animal and human wastes are not burned, but com-posted to make leaf mold which is spread onto the fields.Our recommendation is to primarily burn agriculturalplant waste which is poor in non-CHO elements, com-prising substances such as cellulose, lignin and starch.In this way we avoid unnecessarily releasing fixed ni-trogen and other valuable nutrients into the atmosphere,where they would also be pollutants.
4. Backup techniques
As was pointed out above, typically the biosphere isnot able to fix so much oxygen or nitrogen that it wouldchange the atmospheric concentrations of these gasestoo much. However, to facilitate dealing with accidentscenarios like air leakages or atmospheric poisonings,having compressed or liquefied O and N availablecould be desirable . If so, it may make sense to alsohave a mechanism available for moving O and N se-lectively from the habitat into the tanks by e.g. cryo-genic distillation of air [11]. If such process is imple-mented, then CO is also separable. For managing CO ,such process would be energetically ine ffi cient becausee.g. to reduce the CO concentration into one half, onehas to process 50 % of the air by liquefaction, separat-ing out the CO and returning the O and N back intothe settlement. However, if energy is available, energye ffi ciency is not a requirement for backup strategies.Chemical scrubbing of CO into amines or hydroxidesis another possible backup strategy for emergency re-moval of CO . Table 1 lists these alternatives and theirpotential issues.
5. Discussion
As described in Section 2, gardens, vivariums andother widespread examples of semi-closed (i.e., onlygases exchanged) ecosystems show that closed bio-spheres are feasible. The only issue is to maintain theright atmospheric composition, but this is only a tech-nical problem to which there are many solutions. The In addition, one probably wants to divide the settlement into sep-arately pressurisable sectors [5] so that people can be evacuated froma sector that su ff ered an accident. NSS Space Settlement Journal
DISCUSSION
Table 1: Some alternatives of habitat CO control. Method Potential issues
Biomass burning –Smoke–Need to handle fireCryo-distillation –Power-intensiveor air –Reliability concern / moving parts–Mass overhead of CO tanksScrubbing –Reliability concern / moving partsinto amines –Safety concern due to chemicalsor hydroxidesbiomass burning is one of them. The complexity of bi-ology cannot spoil the feasibility of closed biospheres.If it could, it would already have been seen in gardensand vivariums. The complexity of biology is factoredout of the feasibility equation.The biomass burning method works, as such, onlyin a tropical climate with no dark season. During darkseason photosynthesis is stopped and the level of CO2would probably build up too high in the atmosphere.Therefore, if seasons are wanted, one has to use sector-ing such as discussed in Janhunen [5]. Di ff erent sectorsmust then be phased in di ff erent seasons and air must beexchanged between sectors.Biomass burning seems to be a straightforward, scal-able, low-tech and reliable solution. A possible draw-back is the production of smoke. As on Earth, plantsand soil are absorbers of air pollution, but productionof smoke should nevertheless be minimised to preventhealth issues. In addition, smoke in a settlement envi-ronment is more harmful than on Earth, because the set-tlement has windows through which sunlight enters, orif it is artificially lighted, the lamps have cover glasses.The production of smoke can be minimised by technicalmeans such as igniting the fire by a biogas flame or bymechanically making the biomass into pellets or othergranular form.Biomass burning involves fire, and fire is in principlea risk because conflagration in a space settlement wouldbe very dangerous. Concerning fire risk in general, it isnot feasible to eliminate it entirely by removing all pos-sible ignition sources, e.g. because electric equipmentis necessary and malfunctioning electric equipment isa potential ignition source. The risk of wildfire can belowered by having frequent artificial rain so that the en-vironment is fresh and green. Lush nature also boostsagricultural output and is good for aesthetic reasons.However, not everything can be humid since the storedbiomass must be dry in order to burn cleanly. Thus therelative humidity should be less than 100 %, which is also convenient for people. To reduce the fire risk fur-ther, an easy way is to store the dry biomass far from thelocations where it is burned. Artificial rain or sprinklersystem must be possible to turn on quickly in case a firebreaks out.Also other approaches for reducing the fire riskare possible. For example, one can freeze-dry thebiomass and store it in a refrigerated space. Storage un-der nitrogen-enriched atmosphere is another possibility,which eliminates the fire risk during storage. Nitrogen-enriched gas can be made e.g. by filtering air throughcertain polymeric membranes.The methods discussed in this paper do not involvemoving materials through airlocks. Thus there is no is-sue of losing atmospheric gases into space.After O , N and CO are controlled, the remainingissue is how to keep the level of other volatile com-pounds low. Plants remove harmful impurities [10],but they also produce some volatile organic compounds(VOCs) of their own, such as isoprene and terpenes.This smell of plants can be experienced e.g. in green-houses and it is generally considered pleasant. How-ever, too much of a good thing is potentially a bad thing,so let us briefly discuss loss mechanisms of VOCs. It isthought that the hydroxyl radical OH is an important“detergent” of the troposphere that oxidises VOCs [14].On Earth, the primary formation of OH is by solar UVand is highest in the tropics where the solar zenith an-gle is smallest, the stratospheric ozone layer is thinnestand the humidity is highest [14]. Thus, in the settle-ment it might be a good idea not to filter out the solarUV entirely, but let a small part of it enter so that theUV radiation level mimics the conditions in Earth’s tro-posphere, thus maintaining some OH to remove VOCsand also methane by oxidation.One of the referees pointed out that the carbon stockof soil might potentially grow in time due to incom-plete decomposing. While certain biomes like some wetpeatlands exhibit slow continuous carbon accumulation,typical biomes such as forests have moderate carbonstocks [8] that presumably have not essentially growneven in millions of years. Earth’s significant fossil coaldeposits are thought to have been accumulated beforelignin-degrading organisms developed around the endof the Carboniferous period [15]. On modern Earth, ter-mites are good lignin decomposers [16] so their pres-ence in the habitat ecosystem could be beneficial for ef-ficient carbon circulation.5 NSS Space Settlement Journal
EFERENCES
6. Summary and conclusions
Controlling a habitat’s carbon dioxide level is a non-trivial problem because the atmospheric volume per bio-sphere area is typically much smaller than on Earth.The problem is important because too low CO ( (cid:47) (cid:39) level in the settlement’satmosphere drops too low. The method is straightfor-ward, robust and low-tech. It ensures large control au-thority of the CO while keeping the O partial pressureunchanged. The method scales to habitats of all sizes.In the initial growth phase of the biosphere, onecan obtain the CO by burning sugar or carbon. Theycan be sourced from carbonaceous asteroid materials sothat bootstrapping the biosphere does not require liftinglarge masses from Earth.Closed ecosystems in habitats are feasible. We knowthis because there are many examples of semi-closedecosystems such as gardens – and because it has beendone e.g. in Biosphere-II and BIOS-1, 2 and 3[17].Maintaining the atmosphere is an engineering problemthat can be solved. For gases other than CO , the prob-lem is in fact solved automatically. For the control ofCO , the biomass burning method seems simple and ef-fective.
7. Acknowledgement
The results presented have been achieved under theframework of the Finnish Centre of Excellence in Re-search of Sustainable Space (Academy of Finland grantnumber 312356). I am grateful to journalist HannaNikkanen for providing her compilation of papers onthe topic.
References [1] Bar-On, Y.M., R. Phillips and R. Milo, The biomass dis-tribution on Earth, Proc. Nat. Acad. Sci. 115 (25) (2018)6506–6511. [2] Sullivan, M.J.P., J. Talbot, S.L. Lewis et al., Diversityand carbon storage across the tropical forest biome, Scien-tific Reports 7 (2017) 39102. [3] O’Neill, G.K., The colonization of space, Physics Today 27 (9)(1974) 32–40.[4] O’Neill, G.K., The high frontier, New York (1977) 288. [5] Janhunen, P., Natural illumination solution for rotatingspace settlements, NSS Space Settlement J. (2018). https://space.nss.org/media/NSS-JOURNAL-Natural-Illumination-for-Rotating-Space-Settlements.pdf [6] Wikipedia article on biomass (ecology), https://en.wikipedia.org/wiki/Biomass_(ecology) [7] Ricklefs, R.E. and G.L. Miller, Ecology (4th ed.), Macmillan(2000), 192.[8] Wang, Y.P., R.M. Law and B. Pak, A global model of carbon, ni-trogen and phosphorus cycles for the terrestrial biosphere, Bio-geosci. 7 (2010) 2261–2282.[9] Jewell, W.J. and L.S. Valentine, Biological processes inclosed ecosystems, Space Manufacturing 14, Space StudiesInstitute, Oct. 29–31 (2010). http://ssi.org/2010/SM14-proceedings/Biological-Processes-in-Closed-Ecosystems-Jewell-Valentine.pdf [10] Wolverton, B.C., A. Johnson and K. Bounds, Interior land-scape plants for indoor air pollution abatement, Final re-port, NASA-TM-101766 (1989). https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19930073077.pdf [11] Soilleux, R.J. and S.D. Gunn, Environmental control and lifesupport (ECLSS) for large orbital habitats: ventilation for heatand water transport and management, NSS Space SettlementJ.(2018). https://space.nss.org/media/NSS-JOURNAL-ECLSS-for-Large-Orbital-Habitats-Ventilation-and-Heat-Transport.pdf [12] Thomson, H. and C. Liddell, The suitability of wood pellet heat-ing for domestic households: a review of literature, Renewableand Sustainable Energy Rev. 42 (2015) 1362–1369.[13] Misra, M.K., K.W. Ragland and A.J. Baker, Wood ash compo-sition as a function of furnace temperature, Biomass and Bioen-ergy 4, 2 (1993) 103–116.[14] Lelieveld, J., F.J. Dentener, W. Peters and M.C. Krol,On the role of hydroxyl radicals in the self-cleansingcapacity of the troposphere, Atmos. Chem. Phys. 4(2004) 2337–2344. [15] Floudas, D., et al., The Paleozoic origin of enzymatic lignindecomposition reconstructed from 31 fungal genomes, Science336 (2012) 1715–1719.[16] Butler, J.H.A. and J.C. Buckerfield, Digestion of lignin by ter-mites, Soil Biology and Biochemistry 11 (1979) 507–513.[17] Salisbury, F.B., J.I. Gitelson and G.M. Lisovsky, Bios-3:Siberian experiments in bioregenerative lif support, BioScience47 (1997) 575–585.6