Fish, Salt, or Dirt

I watched a video last week of a fish tank feeding a garden.

The setup was simple enough that I could have built it in a weekend. A tank of fish on the bottom. A pump lifting water up to a row of vertical towers, plants set into the sides of the towers in little cups, water trickling down through the roots and back to the tank. The fish eat, the fish do what fish do in the water, the water goes up through the plants, the plants take what they need out of it, and clean water comes back down to the fish. Around and around. Nobody adds fertilizer. Nobody changes the water.

That is called aquaponics, and it is not a new idea. The Oklahoma State extension office traces the practice back to about 1400 A.D. ¹ What is new is that people are building it in garages and on rooftops with pond pumps and PVC pipe, and it works.

I am a practical engineer, not a farmer, and what caught me about that video was not the plants. It was the loop. That is a closed system with two products coming out of it, greens and protein, and one input going in, fish food. Anybody planning to feed people a long way from a grocery store ought to look hard at that.

So this article is about the loop, and about the two or three other loops you could build instead. I am not going to tell you which one to bet on. I have a leaning, and I will say what it is, but the honest answer is that nobody has run any of these long enough, far enough from help, to know. That is the question at the end.

What the fish are actually doing

Here is the part of the video they did not explain, and it is the part that matters.

Fish put out ammonia. It comes off the gills and out of the waste, and in a closed tank it will kill the fish in short order. In an aquaponics system the ammonia never gets the chance, because two families of bacteria living in the gravel and on the roots eat it. The first group turns ammonia into nitrite. The second group turns nitrite into nitrate. Both steps burn oxygen and both steps make acid, which pulls the pH down. Nitrate is plant food. The plants take it up, the water goes back to the tank clean, and the cycle repeats. ²

So it is not really a two-part system, fish and plants. It is a three-part system, fish and bacteria and plants, and the bacteria are the part you cannot see. You cannot buy them at the store and pour them in on day one, either. You have to grow them, and the extension people say to figure at least four weeks of cycling before you plant anything, with ammonia and nitrite both under about half a part per million before you call the system ready. ^{1 2}

That is a commissioning schedule. Anybody who has started up a plant recognizes it. You do not turn the process on and walk away. You run it empty, you watch the numbers, you wait for the biology to catch up with the hardware.

And here is the catch, the one I keep coming back to. Three living systems sharing one water loop means three things that can fail, and any one of them takes the other two with it. If the pump stops, the bacteria go anoxic and the ammonia climbs. If the fish die, the nitrate supply dies with them and the plants starve. If the plants crash, the nitrate has nowhere to go and it builds up in the tank. The pH that the bacteria want, around eight, is not the pH the plants want, closer to six, so everybody lives at a compromise around seven that none of them would pick. ^{1 3}

Coupled systems are efficient and coupled systems are fragile. That is the trade, and it is the same trade you make when you tie two process units together on the ground.

Option two: salt from a bag

The plainer version is hydroponics, and it is the one that has actually been run in the hardest place on Earth for twenty years.

The South Pole Food Growth Chamber was designed by the University of Arizona's Controlled Environment Agriculture Center with Phil Sadler's machine shop in Tempe, and delivered to Amundsen-Scott South Pole Station in February 2004. ⁴ It is a room, not a greenhouse. About 392 square feet of growing space, climate controlled, with automated carbon dioxide enrichment, a nutrient delivery system, and acid injection to hold the pH. ⁵ The station crew is isolated for more than eight months a year, living on frozen stores flown in during the summer, and the chamber is what puts a salad on the table in the middle of the dark. ⁴

There are no fish. The nitrate comes out of a bag of dry salts, dissolved into a reservoir and dosed into the water. The operator checks the solution with a hand meter and adjusts it with a graduated cylinder. ⁵ If the numbers drift, you add a measured amount and the numbers come back. Nothing has to die for that to happen and nothing has to grow for it to happen. It is a chemical system, not a biological one, and a chemical system is a lot easier to start, stop, and restart.

A team from Arizona built a full mass balance model of the chamber during the 2006 winter, monitoring all the carbon dioxide and water going in and the biomass coming out, remotely, over the internet from Tucson. ⁶ I have seen numbers from that work put the chamber's draw at a few hundred kilowatt-hours a day, which is a steady load in the neighborhood of twelve kilowatts. I cannot get behind the paywall to confirm the figure, so take that as a rough size and not a spec. But it tells you the shape of the thing: a small hydroponic room pulls about what a good-sized diesel generator puts out, all day, every day, mostly for the lights.

The cost of hydroponics is that bag of salt. It has to be shipped in. On the South Pole it comes in on a plane. On the Moon it comes in on a rocket, and every pound of it is a pound you did not spend on something else. Aquaponics makes its own nitrate out of fish food, and fish food is a lot lighter than the fish and the greens it turns into. That is the whole argument for the fish, right there.

Option three: no water at all, almost

The third version is the one the Germans ran at the other end of Antarctica, and it is the closest thing to a Moon-base food system anyone has built.

EDEN ISS was two twenty-foot shipping containers, bolted together and set down 400 meters from Germany's Neumayer III station on the ice shelf. ⁷ The German Aerospace Center, DLR, built it with a consortium of European partners and later NASA, and one man, Paul Zabel, ran it through the 2018 winter as the only operator. He walked out to it once a day through a cold porch and a service section, so the growing room itself never touched the outside air. ^{7 8}

The growing method was aeroponics. The roots hang in the air inside a closed trough and get misted with nutrient solution on a timer. No gravel, no medium, no standing water. That saves weight and it saves water, and it lets you get at the roots to look at them, which matters when you are trying to find out why something is dying.

The numbers from that first season are the best real data I know of. On twelve and a half square meters of growing area, about 135 square feet, the greenhouse produced more than 268 kilograms of edible food between February and November 2018. Cucumbers, lettuce, leafy greens, tomatoes, herbs, radish, kohlrabi. That works out to 27.4 kilograms per square meter per year, or 75 grams per square meter per day. ⁸ Enough that the station's ten-person overwinter crew was getting fresh salad instead of freeze-dried, in a place where the next plane was months away. ⁹

And two crops failed. The peppers and the strawberries never set fruit, most likely because there was nothing in that sealed room to pollinate them. ¹⁰ Read that twice. The lights worked, the pumps worked, the nutrient dosing worked, the plants grew, and they made no food, because a system that closes out the weather also closes out the bees. Somebody had to go in there with a paintbrush, or accept the loss. That is the kind of thing you only learn by running it.

I keep coming back to how familiar all of this is. Two containers on the ice, a generator, one operator walking out once a day to check on it, the nearest spare part a season away. Swap the ice for tundra and that is a North Slope camp. I have lived in that camp. The failures are never the ones on the drawing. They are the pump that cavitates because the reservoir ran low overnight, the timer that drifts, the heat trace that trips and lets a line freeze in the porch. Every one of those took out food in Antarctica for a day or a week, and every one of them will take out food on the Moon, and the Moon is a lot farther from the parts truck.

Option four: dirt

There is a fourth way, and it is the oldest one. You bring soil, or you make it, and you grow in it the way people have for ten thousand years. I wrote about that in the worms piece and I am not going to repeat it here, except to say what it costs and what it buys.

What it costs is mass. Soil is heavy, and soil that grows food has to be a foot deep at least. Nobody is launching that. You would have to make it in place, out of crushed rock and composted waste and the worms to work it, and that is a years-long project before you get the first tomato. What it buys is forgiveness. Soil is a buffer. It holds water when the pump quits, it holds nutrient when the dosing is off, it holds the bacteria and the fungi that make the whole thing run without anybody watching a meter. A soil bed that loses power for a day is fine. An aeroponics trough that loses power for a day is a rack of dead plants with dry roots.

Biosphere 2 tried to do it that way, three acres of soil under glass, and what happened there is its own story, which is coming. The short version is that the soil did a great many things the designers did not ask it to do. That is what soil is.

Can a fish ride a rocket

Before anybody dismisses the fish on the grounds that you cannot launch them, it has been done, more than once, and the fish were fine.

Japanese medaka, a little freshwater fish about an inch long, flew on the shuttle in 1994. Adults mated in orbit during a fifteen-day flight, the eggs developed normally, and the fry hatched in space. ¹¹ In 2012, thirty-two medaka rode up to the space station on a Soyuz with three crew members, and lived two months in a Japanese-built habitat called the Aquatic Habitat. ^{12 13} The habitat had its own recirculating water loop, and the loop cleaned the water with nitrifying bacteria. ¹⁴ Read that one twice, too. There was a working aquaponics biofilter on the space station in 2012. It just did not have any plants on the other end of it.

Fish take launch loads better than you would think, and the reason is that they are already floating. A fish does not hang its organs off a skeleton the way you do. It circulates a small volume of blood at low pressure in water that carries it, and the researchers who flew the medaka picked them partly because fish are less dependent on gravity than mammals in the first place. ¹³ Under a launch load the water around the fish loads with the fish, so there is very little for the g's to tear.

But I would not launch the fish anyway. I would launch the eggs. This is my own reasoning and not anybody's test data, so take it that way: fish eggs are tiny, they need no feeding on the trip, they tolerate cool storage, and a few thousand of them weigh less than a single adult tilapia. You ship the genetics, not the livestock, and you hatch them at the far end into a tank you have already cycled and proven with nothing living in it. That is the same argument I am going to make about seeds in another piece. It applies to anything with an egg.

Where I lean

I said I have a leaning. Here it is.

For a first habitat, the one where the nearest help is three days or eight months away, I would build hydroponics, the South Pole way, salt from a bag, and I would build it twice, two independent rooms, so that one can go down while the other feeds people. I would not couple the fish to the plants at the start, not because the fish idea is wrong, but because I would want to commission each loop by itself before I tied them together. That is how you start up a plant. You do not bring the whole thing up at once and see what happens.

Then, once the hydroponics is running and the crew is not depending on it to eat, I would start the fish in their own tank on their own filter, cycle it, prove it, and only then cross-connect the water. At that point you have aquaponics, and you got there without ever betting a meal on it.

And I would start making dirt on day one, in a corner, with the worms, knowing it would be years before it fed anybody, because the day it does, it is the only one of the four that keeps working when the power is out.

That is my order. Salt first, fish second, dirt last but started early.

The question

But I am a construction guy reading extension bulletins, and the people who actually ran these rooms, at the Pole, at Neumayer, in the garages with the fish tanks, know things I do not.

So here is the question, and it is for you. If you had to feed eight people for two years with no resupply, which loop would you bet on, and why? What did your system do that the plans did not say it would? What killed your fish? What broke at three in the morning?

Send it to hello@starboundmagazine.com. If enough of it comes in, the next piece on this will be yours, not mine.

Ideas in Star Bound are published open — take them, build them better. Our open proposal.

Comments

Star Bound is not a forum, so comments run the old way: if you have comments, please write to us at hello@starboundmagazine.com and they will be published here in the next update, with your name on them. Say if you would rather stay anonymous.

No comments posted on this one yet. Yours could be the first.

Share this article

Copy link Facebook X Email

A note on where this comes from. I am not a metallurgist, a spacecraft engineer, or a degreed engineer of any kind. But I am a practical one, with more than forty years in heavy industrial construction: welding, millwright work, hydraulics, mining, and thirty-three years of electrical construction and commissioning on the North Slope of Alaska, in a place that punishes anything built wrong. Much of that time I have worked as a field engineer, which means I am the one standing in front of the equipment when the drawing turns out to be wrong, and the fix has to be made on the spot, with what is on hand, and has to work the first time. Anyone who has done that long enough learns that a lot of engineering gets corrected in the field by people with more experience than the people who drew it. That is the eye I bring to these subjects: not what should work on paper, but what actually gets built, and what breaks. The facts I take from other people are footnoted so you can check them and read further. The ideas built on top of those facts are my own, and I would rather see them knocked down by somebody who knows better than left standing because nobody looked.

Sources

1. Oklahoma State University Extension, "Nitrification and Maintenance in Media Bed Aquaponics," fact sheet. ↩
2. New Mexico State University, "Important Water Quality Parameters in Aquaponics Systems," Circular 680. ↩
3. University of Hawaii, "Nitrogen Transformations in Aquaponic Systems." ↩
4. University of Arizona Controlled Environment Agriculture Center, "South Pole Growing Chamber." ↩
5. Aquaponics Journal, "Antarctic Hydroponics," as reproduced at arc306.blogspot.com, October 2009. ↩
6. Patterson, Giacomelli, Sadler, "Resource and Production Model for the South Pole Food Growth Chamber," SAE Technical Paper 2008-01-2011. ↩
7. "Microbial Monitoring in the EDEN ISS Greenhouse, a Mobile Test Facility in Antarctica," Frontiers in Microbiology, 2020. ↩
8. Zabel et al., "Biomass Production of the EDEN ISS Space Greenhouse in Antarctica During the 2018 Experiment Phase," Frontiers in Plant Science, 2020. ↩
9. European Space Agency ACT, "EDEN ISS – A space greenhouse analogue in Antarctica," June 2022. ↩
10. Space.com, "Antarctica Greenhouse Produces Cucumbers, Tomatoes and More in Mars-Like Test," September 2018. ↩
11. Ijiri et al., "Life-Cycle Experiments of Medaka Fish Aboard the International Space Station," Advances in Space Biology and Medicine, 2003. ↩
12. Space.com, "Next Space Station Crew to Try 'Fishy' Science," July 2012. ↩
13. Murata et al., "Histological and Transcriptomic Analysis of Adult Japanese Medaka Sampled Onboard the International Space Station," PLOS ONE, 2015. ↩
14. "NASDA next generation aquatic habitat for space shuttle and ISS," Advances in Space Research, 2004. ↩

The startup order in "Where I lean" and the egg-shipping argument are my own reasoning, not anything from the sources above. The South Pole power figure is secondhand and unconfirmed.