The entry says that the institute published its result as six different percentages and that all of them are true. This piece sets the tables out so that the reader can do the arithmetic. Everything here is from the Krasnoyarsk group’s own papers: the 1989 paper on the last run, printed by NASA; the 1997 paper written with Frank Salisbury; and the 2008 retrospective by Gitelson and Lisovsky. Where two versions of the same table differ, both are given.

Table one: what one person needs a day

The 1989 paper’s Table 1 lists the daily requirement of one crew member, in grams, without recycling and with recycling as in Bios-3. The 1997 paper reprints it as its Table 3.

Without recycling: dry food, 924; oxygen, 1,220 (the 1997 version says 1,283); drinking water, 5,133; sanitary water, 5,696; sanitary materials, 9.5; cooking salt, 28. Total, 13,010.5 (1997: 13,073.5).

With recycling in Bios-3: dry food, 208; chemical compounds for the nutrient medium, 350; sanitary materials, 9.5; absorbents for water purification, 2.7; cooking salt, 28. Total, 598.2. Oxygen and water do not appear, because they were made inside; the 1997 version footnotes that “oxygen and water were recycled 100%; hence, they do not appear in this column.”

The note under the 1989 table: “The sampling of a substance for investigation outside the system isn’t presented in this table.” The 1997 version adds that with recycling “the requirements of the crew cannot be separated from the requirements of the ecosystem as a whole”, which is why the 350 grams of plant nutrients sit in a person’s column.

Divide 598.2 by 13,010.5 and you get 4.6 percent. That is the 95 percent: the crew’s needs were “met 95% (in weight) with the resources inside the system”. Notice what dominates the top line. Water is 10,829 of the 13,010.5 grams, oxygen 1,220, food 924. The plants made all the water and all the oxygen, and if they had made no food at all the figure would have been 13,010.5 less 10,829 less 1,220, which is 961.5 grams needed from outside, or 92.6 percent closure. The 95 percent is a sentence about water.

Table two: the food

The 1989 paper’s Table 2, “System recycling of food consumption by humans”, covers one month of the 1983 to 1984 run with two men. Consumed by the crew in the month: 27,716 grams of absolutely dry substance; 4,369 grams of protein; 4,679 of lipids; 16,030 of carbohydrates; 119,699 kilocalories. Edible biomass provided by the plants grown in the system: 21,484 grams dry; 2,851 of protein; 1,652 of lipids; 15,290 of carbohydrates; 83,609 kilocalories. Portion of crew consumption being recycled: 77.5 percent of dry substance, 65.2 of protein, 35.3 of lipids, 95.4 of carbohydrates, 69.8 of calories.

So the plants made nearly all the starch, two thirds of the protein and a third of the fat. The rest was the lyophilised meat through the airlock, which the 1997 paper says was about a quarter of the food by mass. These are the 77.5 and the 70.

The design point: 13 square metres

The 1989 paper’s finding from the whole series is that “13 m² area occupied by plants is sufficient for a continuous supply of one man’s oxygen, water (at the expense of transpirational moisture condensate) and for 40-45% of his food”, under round-the-clock light of 150 to 160 watts per square metre of photosynthetically active radiation, which took about 200 watts of lamp power per square metre. That is the 40 to 45. The paper notes that this is the flux the sun delivers to every 3.5 square metres in Earth orbit and to 8 square metres at Mars, so a system in space would need light-collecting surfaces several times the area of its crops.

The 1997 paper’s Table 2 gives the trade-off as a line. For 85 percent closure of the whole system: 13 square metres of plants per person, two kilowatts of photosynthetically active radiation, relative size 1. For 95 percent: 30 square metres, 4.6 kilowatts, relative size 2.3. For 99 percent: 56 square metres, 8.5 kilowatts, relative size 4. The food line of that table runs 35, 50 and 100 percent respectively, oxygen and water 100 throughout. The last four points of closure cost nearly twice the plants and the power of the first ten.

The 91 and the 93

The 91 is the 1989 abstract’s “total closure of material turnover”, for the 1983 to 1984 run, and the paper does not show its working. It is lower than the 95 because it counts flows the Table 1 does not: the samples out through the airlock, the dried faeces and kitchen waste, the ash, the water in the meat. It is the honest headline number and it is the one the entry takes its title from.

The 93 is from the 2008 paper: “The atmosphere and water were regenerated completely while food regeneration amounted to 93% (Table 1).” The table it points to lists the daily yield of eleven crops on 39.40 square metres: wheat on 17.53 square metres giving 186.7 grams a day of edible dry mass; chufa on 13.27 giving 345.6; carrot, beet, radish, onion, dill, kale, cucumber, potato and sorrel on the rest; 681.1 grams of edible dry mass a day in all, and 1,442.2 of total dry mass, so that just under half of what grew could be eaten. A footnote says the fresh-mass total “comprises only vegetables”. The 39.4 square metres are two phytotrons’ worth, not three, and the chufa share is far higher than in the third run’s crop list, so this table is from a different run than Table 2 above, most likely the first. The 93 percent is of the plant part of the diet, with the meat outside the account, which is what the group’s 1981 and 1985 conference papers meant by “a full regeneration of atmosphere, water and ration vegetable part”.

The 50, the 70 and the 80

These are summaries, not measurements. The institute’s Bios-3 web page says that by 1968, in Bios-2, “food and nutrient recycling” reached about 50 percent; its history page and the Siberian Branch’s tribute of 2003 say the Bios-3 crews had “up to 80%” of their food from inside; the NASA plant physiologist Raymond Wheeler, with information from Gitelson, says “up to 70% of the food”; a Krasnoyarsk newspaper in 2005 said the cycle was about 60 percent closed and the push to close the last 40 had been shelved for want of money. None of these appears in the papers with a table under it.

The minerals: the balance that did not close

The 1997 paper followed twenty elements, seven major (nitrogen, sulphur, potassium, sodium, calcium, magnesium and phosphorus) and thirteen trace (iron, manganese, copper, zinc, tin, lead, aluminium, titanium, boron, nickel, chromium, vanadium and cobalt), in every liquid and solid that entered or left, analysed once or twice a month by chemical, photometric, atomic-absorption and flame-photometric methods.

Inputs: the nutrient salts, plus impurities in the salts and in soap and toothpaste (nickel, chromium, aluminium, lead, tin, titanium), plus what the building gave up. Outputs: dried faeces, kitchen waste, the ash of burned straw, and the samples. The result in the third run: mineral closure “only approximately 20% on average”, nitrogen about 40 percent; for the major elements, “input and output were nearly balanced, within the limits of error of analysis techniques (i.e., 10-15%)”; for the trace elements, “output exceeded input”. A 1997 paper by Gribovskaya, Kudenko and Gitelson gives the mineral closure as 24 percent and names the structural steel and titanium, the expanded-clay substrate and the catalysts as sources of the elements that built up.

In the second run the imbalance went the other way for some elements: nickel, aluminium, chromium and lead were ten to twenty times higher in the plants and solutions at the end than at the beginning, tin, titanium and zinc two to four times. The paper names the sources: solder, “especially the solder applied to the steel net used to collect the sewage water”, gave lead and tin; new untreated steel gave nickel and chromium; a porous water filter gave aluminium, lead, nickel, chromium, titanium, zinc, copper and vanadium; the catalysts in the thermocatalytic converter gave zinc, chromium, vanadium, titanium, iron and aluminium. In the third run, with a modified converter, different nets and solders and no new steel, “by the end of that experiment, there was no accumulation of elements in the plants.”

The 2008 paper’s judgement on all of this is a sentence: “The migration of trace elements in the experimental system has turned out to be a hardly controllable process.”

The gases: the balance that was burned

Two charts carry the gas balance. The first is figure 5 of the 1997 paper, carbon dioxide and oxygen through the second run, with the caption: “Sharp changes during the second half of the run were caused by the burning of inedible plant wastes. Note the mirror-image responses of carbon dioxide and oxygen.” Carbon dioxide ran between about 0.5 percent of the air, when the crops were doing especially well, and a little over 2 percent at the start of the run; the Krasnoyarsk view, from Lisovsky’s 1979 monograph, was that a dense canopy grew best above 0.3 percent, where American work at Utah State later put the wheat optimum at 0.12.

The second is figure 6 of the 2008 paper, ten lines over the 180 days of the first run: ammonia, acetic acid, acrolein, aldehydes, alcohols, hydrogen sulphide, mercaptans, water-soluble and water-insoluble organic substances, and carbon monoxide. The text says the concentrations “varied around the average values, remaining, on the average, parallel to the time axis, which confirmed that the atmosphere was in a state of dynamic equilibrium.” What held them there was not a balance between people and plants but the thermocatalytic filter at 600 to 650 degrees, which oxidised the lot to carbon dioxide and water. The 1996 paper by Gitelson and Okladnikov describes the one true balance in the gases: oxygen stays level only if the crew’s respiratory quotient, carbon dioxide out over oxygen in, equals the crops’ assimilation quotient, oxygen out over carbon dioxide in, and the group learned to tune the second with oil crops, chufa in the third run bringing it to 0.95, and the first with the diet.

What the ledger cannot contain

A balance sheet records the flows its keepers chose to follow, to the precision of their instruments, for the period of the run. This one followed air and water to completion, food to the gram, twenty elements to within ten or fifteen percent, and ten trace gases to a line that stayed level. It did not follow the volatile organics that the filter burned, because they were burned. It did not follow the crew’s metabolism as a variable in the gas balance, although the blood chemistry that a 2025 preprint reads that way was in the 1975 and 1976 reports all along. And it says nothing past April 1984. Every number above is a number for a run of four, five or six months, with the people inside adjusting the system every day. The ledger is exact about that, and so should anyone be who quotes it.

Sources

• I. I. Gitelson and colleagues, “Long-term experiments on man’s stay in biological life-support system”, NASA Conference Publication 10040, 1989, pages 61–66; also Advances in Space Research 9 (8), 65–71. Tables 1 and 2, the 13 square metres, the 91 and 95 percent. ntrs.nasa.gov
• Frank B. Salisbury, Josef I. Gitelson and Genry M. Lisovsky, “Bios-3: Siberian experiments in bioregenerative life support”, BioScience 47 (9), 575–585, 1997. Tables 2 and 3, figure 5, the balance-sheet section and the elements. simoc.space
• Iosef I. Gitelson and Genry M. Lisovsky, “Creation of closed ecological life support systems”, Journal of Siberian Federal University, Biology 1 (1), 19–39, 2008. Table 1, figure 6 and the 93 percent. elib.sfu-kras.ru
• I. V. Gribovskaya, Yu. A. Kudenko and J. I. Gitelson, “Element exchange in a water- and gas-closed biological life support system”, Advances in Space Research 20 (10), 2045–2048, 1997, abstract; J. I. Gitelson and Yu. N. Okladnikov, “Consistency of gas exchange of man and plants in a closed ecological system”, Advances in Space Research 18 (1–2), 205–210, 1996, abstract. pubmed 11542588, pubmed 11538965
• Institute of Biophysics, “BIOS-3” and “History” pages; “Zhizn’ v nauke”, Nauka v Sibiri 26, July 2003; Raymond M. Wheeler, Open Agriculture 2, 2017, figure 1 caption; Zolik Milman, Rossiyskaya Gazeta, 1 July 2005. The 50, 70, 80 and 60 percent. ibp.ru, prometeus.nsc.ru, sciences.ucf.edu, rg.ru
• D. A. Semyonov and A. A. Semyonova, arXiv 2504.18551, 2025, for the reading of the blood chemistry. arxiv.org