Starting the Mars Base

Part of my series on countering common misconceptions in space journalism.

This blog grew out of several fascinating discussions with Dr Margarita Marinova, a former SpaceX Senior Mars and Vehicle Systems Development Engineer who has also published several fascinating papers on terraforming. As all good conversations do, it began with a question:

How to start out the city on Mars?

I’ve written at length about working on Mars, Mars architecture, and Mars exploration systems. I’ve even written a book about industrializing Mars, which focuses on sustaining exponential growth of the city. But it’s still necessary to transition from Mars’ current state of robot habitation to human habitation. How does that look? What are the first few years like?

How to Feed a Mars Colony of 1 Million People | Space

I don’t know all the answers. Indeed, our media and art is usually pretty sparse in this regard. Beyond vague notions of arbitrarily capable robots or indentured servitude at Elon City, we lack a common vision for starting from scratch, with a will to overcome. This may be because we lack suitable analogies for an endeavor this ambitious. In my post on the Starship, I pointed out that SpaceX wants to lift the cargo constraint on city building, enabling logistics comparable to D-Day or the Berlin airlift. Certainly hundreds of tons of cargo lessens the difficulty of the operation, but what is the operation?

In order to understand how the operation works, we need to understand what it does. What mission does it serve? For want of a catchier phrase, the mission is to survive and then thrive.

Remember, Mars is a horrible place. No-one should want to go there. Humans need comprehensive life support and advanced technology to even survive the landing. Upon launch, the mission is to survive a six month voyage in space. Upon entering the atmosphere, the mission is to survive the next seven minutes. After executing a survivable landing, the mission is to survive for the first day, then week, then month, then year.

This constant expansion of the “time until certain death” is a microcosm of the later process of industrialization. At any point until complete industrialization, humans on Mars are vulnerable to certain disruptions, particularly of needed supplies. Later on, this margin may be measured in years but during the first few months, many other sources of risk need to be understood and retired.

After the first 18 months on the surface, the crew will face an important decision. Remain on Mars to keep building and be joined by more immigrants? Or bail out and take a Starship back to Earth? This is not an easy decision since neither option is without risk. In practice, this go/no go decision will be determined by whether enough baseline capacity has been built. Later, the Martians will need to build more space for future arrivals but in the first year, they need to first accommodate themselves!

To dive slightly deeper into technical details, the first year on Mars will be occupied by bringing basic needs online. There is plenty of discussion on the particulars, but there is agreement on the following:

  • Build and operate a giant solar farm to generate electricity.
  • Commission air miners to extract CO2 and nitrogen from the Martian atmosphere.
  • Set up a water mine, most likely some sort of ice well. Liquid water would be even better if available.
  • Set up chemical plant to produce methane, oxygen, ethanol and plastics.
  • Demonstrate agriculture with natural light and soils, or with containerized hydroponic systems.
  • Commission (tele)robotic systems of all kinds. Heavy construction, life support systems, etc. Anything that can be operated by computers or remote techs on Earth will be.
  • Demonstrate construction of pressurized habs on the surface, even if people are mostly living in the Starships or prefabricated structures.
  • Commission a protofab facility to produce one off parts or perform vital repairs.

In general it is impossible to do any kind of complex construction without accruing technical debt. Technical debt is where cutting corners saves time now but requires more time to fix in future. Under ideal circumstances, the Mars city will have a steadily diminishing shortage of available labor, so it makes sense to accrue some technical debt. As always, it’s best to build systems that maximize the decision and learning rate. Most decisions are relatively easily reversed, and do not require enormous deliberation. The same goes for technical debt. The trick is to identify the interfaces which will cause path dependency and real headaches down the road, and get them right the first time.

Finally, let’s consider the humble step function. There are several unprecedented achievements in building a Mars city which involve stepwise increments in difficulty. Getting humans to Mars. Operating a base on the surface. Getting humans back to Earth. Getting steadily more sophisticated industrial processes working. Even in 2020, the incremental difficulty of deep space human spaceflight is so high that it’s effectively a blocker, which is why no-one has left low Earth orbit for nearly 50 years. It is vital that actions and decisions be measured by the extent to which they reduce the height of these step functions.

There is no point in cutting a corner to get humans to Mars if doing so leaves the program without any path to sustainable growth. The Apollo program is an excellent example of trading one step function for another – and canceling the Saturn V was inevitable. Not only was it too expensive to operate indefinitely, the entire architecture had no path to getting cheap enough to run on an ongoing basis.

There are several different ways we can think about starting the Mars city. Reducing step functions. Aiming to survive and then thrive. Accumulating technical debt responsibly. Or just assembling Ikea furniture all day, every day, for years. The path should be made easier for those who follow.

26 thoughts on “Starting the Mars Base

  1. I wish more thought was put into this topic. Every futurist or documentary I have seen jumps to 25, 50, or 100+ years beyond the 1st mission to a stage where the pre-colony is already established. I want to know what supplies will be sent in advance. What will be the various missions and priorities in the first few days? weeks? and years. How much work can we get done in a single 12-18 month mission?

    Liked by 1 person

  2. The biggest step is to get there safely with enough stuff to survive until the next return window and return if things do not work out. SpaceX is rightly focusing on logistical issues — getting there and producing the fuel to get back. They will set up solar power, ice mining, and CO2 extraction, and perhaps hardened landing zones.

    Others — I am thinking of the participants in the NASA challenges and the space agencies — need to do the rest.

    I agree: habitats, food production and the manufacture of spare parts are important, and a geologist to do some science.

    Liked by 1 person

  3. I agree that the vast majority of the work will be done by remotely operated semi-autonomous machines, carrying out processes designed to tolerate being put on standby for the duration of the light speed lag from Earth. Tele-robotic, as you say. Rarely, they’ll page a local human response, to figure out something that can’t wait for cheaper Earth labor.

    The first task will be to make the high-mass low-precision components of more such machines. I think glass fiber and a few kinds of plastic will be good enough for housings, wheels, belts, pressure vessels, and so on.
    Even low-load nuts and bolts.

    I don’t think that buildings will be an early focus. I think the machines will work in inflatable structures filled with CO2, but no CO or perchlorate or whatever. The structures will be pressurized and insulated to varying degrees for different processes.
    My guess is that mostly it will be enough to allow liquid water, but not much more: maybe a tenth of an atmosphere and 4 degrees C.

    I don’t believe in agriculture. I believe in food manufacturing. I think the Martians will be able to produce food from CO2, H2O, N2, and trace elements, all just with chemistry, in equipment that has less imported mass than two years’ food supply. If not, they’ll be able to feed methane and O2 to microbes and eat the microbes.

    To get water, I imagine them covering icy regolith with basically a tarp, and pointing large mirrors at it so that the ice sublimates from the warm ground and collects as frost on a non-heated area of the covering.

    I think metals will come after plastics and glass fiber. Metals are difficult when you don’t have layers of stuff that precipitated out of the ocean after cyanobacteria switched the atmosphere to oxidizing, or rock that has been sitting in a hot rainy climate until everything more soluble than alumina has leached away, or hydrothermal deposits that have been exposed when a subduction-zone volcano eroded away. My current guess for large-scale metal production is the same fracking-for-minerals setup that I described in the comments on Underground Lair. Small-scale metal production probably won’t be worth doing, compared to bringing enough metal along to get by until the large-scale system gets going.

    Liked by 1 person

      1. SpaceX has been checking out landing sites in Arcadia Planitia which offer a number of advantages:

        – Ice is thought to be near the surface
        – They are near the equator
        – They are at low altitude, meaning a thicker atmosphere hence better protection from cosmic radiation, easier CO2 extraction, and easier landing
        – Some sites are near highlands offering more varied geology

        I picked this article because of the map.

        Liked by 2 people

      2. Good point about meteorite metal. I don’t know how much is there, but it’s a resource that Mars has more of than Earth because of the same things that cause it not to have the set of mineral resources we have here.

        As for water, yes, I’m thinking big. There are so-called “rock glaciers” on Mars, glaciers where enough ice has sublimated away that the surface is just rock, but it’s still shaped like a glacier and there’s probably a lot of ice still there on the inside. When I say “large” mirrors, I’m thinking lots of square km in total, aiming megawatts of heat at the icy ground, to extract water without having to move solids.

        Liked by 1 person

      3. The largest gold reserves on Earth were on the rim of the Vredefort crater in South Africa, the largest we know of. The Witwatersrand mines to the north and northwest, the north Free State mines to the southwest, and smaller mines along the Swaziland border to the east. So crater rims might be a good place to start looking.

        Liked by 1 person

      4. On Earth, we tap into one part of the water cycle for our agriculture, and let Sun-powered thermals (and ridge lift) condense the water back out of the atmosphere uphill of us.

        On Mars, we’ll need to explicitly remove agricultural transpiration water from the air – at which point we can use it again. No more gallon-per-almond requirements for newly mined water. The power budget for dehumidifying may be intense, though.

        Industry on Earth also uses a lot of water once-through. It might be tempting to mine lots of water on Mars for similar usage – but once you’ve used it, where do you put it? Putting massive piles of frozen low-grade toxic waste around your colony isn’t practical. So we’ll probably have to recycle industrial water as well as agricultural.

        If industrial and agricultural water is recycled, the need for new water decreases by orders of magnitude.

        Liked by 1 person

  4. We won’t want to invest much in any one base until we’ve done an extensive geological resource survey and seen which types of pilot ISRU plants can operate on Mars. Before humans land we could have scores of robotic drill sites looking for extractable water and useful ores.

    Liked by 1 person

  5. I looked up the shipping weight of a roll of aluminized mylar on Amazon. If I did the arithmetic right, it comes to about one and a half Starship loads per square km. So anything that needs lots of square km will have to wait until the stuff is being made locally.

    The solar constant at Mars is about half a kW per square meter, or about half a GW per square km.

    The heat of vaporization of water is about 2 kJ/g. So the sunlight per square km is enough to vaporize a quarter of a ton of water per second (again, if I did the arithmetic right).

    Each Starship load of fuel contains about 3k tons of water, as the H in CH4 and the share of the O2 that goes with it. So vaporizing that much water under standard conditions takes the heat in a square km of sunshine for 12k seconds, aka 3hr 20 min. The heat of vaporization under low pressure is lower, but they will be heating rock as well as vaporizing ice.

    Still, it sounds as though the Martians won’t need lots of square km of mirrors just to extract water, unless the percentage of ice in a rock glacier is lower than I imagine.

    Liked by 1 person

      1. I would certainly expect so. I think the shipping weight includes a cardboard tube that it’s wound around, too. But I want to err on the side of difficulty, when I’m thinking about stuff that I’m going to say that the Martians could do. I know that the standard thickness is practical for ordinary purposes, and if I said half the thickness or whatever, I wouldn’t know whether I was making it too thin.

        Liked by 1 person

  6. I’m guessing that the city will have to be located where the water is available, the topography is acceptable, and the latitude is in the right range, and that won’t leave a whole lot of options for having it be at a major deposit of any other resource. On Earth, transport is dirt cheap as long as it’s from one seacoast location to another. On Mars, eventually they will build a rail network connecting different mineral resources. But when it’s getting started there will be no long-distance transport other than by Starship. So the meteorite metal that I imagine is in small amounts: whenever they move regolith for other reasons, they’ll run it past a magnet and get whatever metallic meteorites are there. Some samples will be worth exporting back to Earth for research, but the bulk will be for local use.

    Liked by 1 person

    1. When it comes to ores on Mars we just don’t know. The surface has been much less chewed than on Earth but there’s probably still plenty of metamorphism, hydrological deposits, meteorite falls, etc.

      Liked by 1 person

      1. True. But we do know that they’ll have ordinary rock, at least. So they can do fracking for minerals, or they can crush it and separate by density and proceed from there, or they can smelt it with carbon un-burned from the air, and so on. If it turns out that they can do better by finding a site with both abundant water and some higher-grade mineral resources, even cooler.

        Liked by 1 person

      2. I just looked up how aluminum is refined, at the glance-at-Wikipedia level of detail. My guess now is that roughly the same process could start from whole rock. On a planet with oceans and bauxite, starting from whole rock, just because it’s conveniently located, is doing it the hard way. On Mars, I think it’s going to be the easy way.

        I still think plastics and glass fiber are likely to be easier, and fracking for minerals is likely to be more efficient at scale. But now I imagine that metal production will start at a fairly modest scale.

        Liked by 1 person

  7. Another thing that the Martians will want to produce locally, fairly early on, is refractory materials. When you melt lots of rock to make glass fiber and tempered glass bricks and beams, you need something to melt it in. When you smelt metal, you need something to smelt it in. Those crucibles are relatively low tech and high mass. A few will be brought along, to get things started, but as production ramps up, it will be worth making them. Materials include silicon carbide and metal oxides (aka ceramics). They’re readily available from rock and air, so they don’t affect site selection.

    Liked by 1 person

  8. Somehow the string of well thought out comments disappeared (!) that related to using a tarp on the ground, warmed by the sun, to extract water.

    I forwarded the comments to Kris Zacny, our leader at Honeybee Robotics Exploration, who replied that Greg Mungas from JPL proposed this idea over a decade ago. It has not been demonstrated to work by anyone yet. The problem: ice will not sublimate, it will melt, which then results in mud, not water vapor that can be condensed.

    At Honeybee, we are developing several systems to drill into underground glaciers and extract water to the surface. The drilling approaches include devices not so different from how oil wells are created, to devices that can drill small holes deeply and quickly. Both approaches seem to be effective and scalable, and are being developed for near term (2-5 year) deployments to other planetary surfaces.

    Liked by 1 person

      1. Makes sense, in the context of barely-icy regolith. I was thinking of a glacier that’s mostly still there, but covered by wind-borne dust and by the ice-borne debris from the ablated-away ice that was superficial enough to be affected by circadian temperature cycles. If barely-icy regolith could provide enough water, that would be really cool. It would mean more flexibility for the city to be sited for access to other resources.

        Liked by 1 person

  9. Or rather, if you apply the heat so fast that the vapor can’t escape fast enough to keep up, the water will melt temporarily. But it will boil as the vapor escapes. And if the water resource is large enough to support the city in the first place, there’s plenty of room to apply the heat at a lower rate over a larger area.

    Liked by 1 person

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s