"1M tons" to the surface - mass breakdown

[meekGee]

So just now Musk repeated his goal:

Elon Musk

@elonmusk
I am confident that Starship will land humans on Mars. That path is clear.

But what really matters is securing the future of consciousness, not just getting a small number of people to Mars.

That probably requires getting over 100,000 people and 1M tons of cargo to Mars.

The critical threshold to pass is making it such that Mars can grow even if supply ships from Earth stop coming for any reason.
7:59 AM · Oct 15, 2025
·

https://twitter.com/elonmusk/status/1978475722273620121

Which matches an average of 10 people and 100 tons per ship, 1000 ships per synode, 10 synodes.  Clearly an OOM estimate.

The 100k number feels right. A midsize town.

I'm a bit fuzzier on the mass - how much does a civilization seed weigh?

I want to come up with even a methodology to estimate that.

Things around me, in general, fall broadly into "commodities" and "products", where at a small level products are made from commodities.

Some products (electric wires for example) are made in such large quantities that they become commoditized. But the machines that make the wires, they're definitely not commoditized.

(I don't know if "products" is the right term here)

Looking around me, by far most mass is in commodities.

The process of setting up a transport-independenr colony is the process of chasing fabrication from the commodities end.

The stuff at the very top of the list would only be shipped in for a couple of synodes, until it's made on Mars and then it drops off the manifest.  But this requires shipping production equipment, until at some (much) later time even the equipment can be made on Mars.  Until that time, it counts towards the 1M ton.

Some stuff can be hybridized. When steel is produced, 90% of the mass of a bulldozer can be locally made, even if the finder parts still have to be imported.

Trying to make a comprehensive list that is actually justified is very difficult and honestly has too many guesses. (want to try?)

So I'm trying to come up with an alternative to just open-loop estimating it.

[meekGee]

One starting point is the mass/population ratio.

1M ton / 100,000 people is 10 tons per person.

This seems at first blush low, and definitely can't include ANY regularly consumed commodities, since that will break the budget immediately.

But this ratio is only for the 20 year view.  The initial ratio will be much much higher, but only for a small crew, and 2-3 synodes.

Here's a list of commodities and commoditized products, kind of in order, and since we don't allow it in the 1M ton budget, the equipment for making it must be

- power
- water
- oxygen
- methane

- most food supply
- methane-derived plastics
- locally sourced elements and minerals
- cement/concrete
- polymers that include locally sourced minerals
- glass
- metal

- extruded/cast plastic products
- wires, fasteners, extruded metal
- cement-based structures

- Hybrid PV panels (thin film from Earth  laminated to locally sourced backing plates and framing)
- Hybrid construction equipment using local metal castings.

---

Let's say being able to make these products in place, while very far from self sufficiency, is a requirement for increasing the population, since otherwise you'll need to supply all that from Earth.

This is a much more modest goal. How many synodes (and people) are required for only that?

I'd say more than 100, less than 1000.

(Thinking about what was achieved in companies of that range I've worked for, and how many of those people aere actually productive....)

So let's call this goal10, and say 30 people  at 100 tons/person.

And maybe that's the key to pushing on the 1M ton number. As follows (next post)

[Robotbeat]

The more population you have, the less that needs to be launched to LEO. At first, everything is one way, and if you want enough repellent to get all the way back to earth, that increases your mass per person by a massive amount.

Eventually, you can even use Martian propellant to do trans Mars insertion. You send it to Mars beforehand and it’s waiting to fuel up Mars bound starships in Leo.

[meekGee]

The more population you have, the less that needs to be launched to LEO. At first, everything is one way, and if you want enough repellent to get all the way back to earth, that increases your mass per person by a massive amount.

Eventually, you can even use Martian propellant to do trans Mars insertion. You send it to Mars beforehand and it’s waiting to fuel up Mars bound starships in Leo.

That's a lot of repellent! 

I think back traffic, as long as the colony is growing, will be a tiny amount of fore traffic.  Even if everyone wanted to eventually come back. But I think most won't.

I just can't imagine a Mars frontier having shades of suburbia.  I think SciFi that depicts them as somewhat spartan has got it right.  Some combination of both.

Have you read "The Forever War"?  You can never go home again.

[meekGee]

So maybe the key to grappling withh the 1M ton in 20 years goal is to break it into imaginable steps. Maybe allow the 20 year number to drift too.

Step 1: 10 people, 100 tons/person. (~1 person/ship, 1000 tons total)   Goal: First 4 items: power water oxygen methane.

Step 2: 100 people, 50 tons per person (~2 ppl/ship, 5000 tons total)  Goal: Next 7 items

Step 3: 1000 people, 33 tons per person (~3 ppl/ship ships, 33,000 tons total)  Goal: Next 3 items

Step 4: 10,000 people, 25 tons per person (~4 ppl/ship, 250,000 tons total) goals: finish aforementioned list.

I think once there, the required tons/person will drop by a lot, and so many more people can be brought in.

Each step is 1-2 synodes, depending. The list made a dent in the mass budget (30%) and a smaller one (11%) in the population target.

After this, if the goals have been met, crank it up.  10 ppl/ship, 10 tons/person, 1000 ships/synode.

With the influx of people, the goal will shift from reducing the tonnage of products shipped from earth (per person) to reducing the variety of products needed.

For example, the thin film PV cells themselves.  They may not account for much in terms of mass, but they are super critical to self-reliance.

Full semiconductor ability will be much further down the road, after things like bunny suits and clean room mats and e-stop buttons can be produced with impunity....

This means we have these distinct phases:

1: Reduce imported tonnage per person, allowing for more people to come in
2: Eliminate imported critical items, working towards self sufficiency
3: Focus on stupid shit because it's fashionable rn, and how quickly did we forget.

[Coastal Ron]

For assumptions it would help to understand if the colonists are living in one large building or lots of smaller ones?

It is unlikely that near-term colonists will be living in single-person dwellings, and that would be very resource intensive, but maybe you don't want or need to have all 100,000 colonists living in one place and one building, even though that may be the most efficient use of resources.

So maybe coming up with an assumption of what the "clumps" will be of colonists will be able to help understand the resources need for each "clump" of colonists?

For instance, is it assumed that all 100,000 colonists will be in one "city", or will they be spread out into multiple "towns"? One city would reduce the amount of infrastructure needed, but maybe that doesn't make sense from a colonization standpoint?

[Robotbeat]

Only reason to have more than one city is mineral resources or some site with particular scientific interest.

I expect one big city. Perhaps one or a few very large inflatable pressure tents with smaller pressurized buildings inside (layered safety, plus more comfort inside). Square kilometers of pressure tents. Casey Handmer has blogged about this
Inside the tents, I think you’d have individual pressurized buildings. At first, brought in pre-built but later built on-site using plastic and rolls of steel from Earth and later, after the steel mills are set up, Martian steel and plastic.

You would important all major structural materials at first. Your first square kilometer tents, etc, built in sections, would come from Earth. It sounds like a lot, but even at $1000/kg, you could import a pressure tent of a square kilometer (with a mass of 3000-6000tons) for about $3-6B. A great start for a city. Interior buildings can be built in shirt sleeve or with a low-encumbrance unpressurized suit. Early workers may camp light, like backpackers. 100kg of backpacking supplies is plenty for sparse accommodations per person while setting up factories for housing material and agriculture.


Mining outposts probably would use pre-built modules from Earth.

[Vultur]

One question - is that 1M tons of payload, or 1M total tons (including the structure of landed Ships which can be repurposed)?

Here's a list of commodities and commoditized products, kind of in order, and since we don't allow it in the 1M ton budget, the equipment for making it must be

- power
- water
- oxygen
- methane

These are going to be the things that would be made from the first synod; hardware delivered even before humans, though presumably set up by the first humans.

- most food supply
- methane-derived plastics

These are probably next, though I'm not sure methane would be the source of plastics. You could do that through oxidative coupling of methane, but syngas made by CO from CO2 -> CO + O2 and H2 from electrolysis of ice might be easier. That's a trade that would need to be done (and might deserve its own thread.)

- locally sourced elements and minerals
- cement/concrete
- polymers that include locally sourced minerals
- glass
- metal

Yeah, these are a step up in difficulty.

I am not sure if you'd use cement/concrete on Mars. The usual kind on Earth (Portland cement) is made from limestone. And if you're starting with a plastics industry before you start mining, you might make structures out of carbon fiber or something. The relative cost of composites and concrete might be very different from on Earth, due to the abundance of the starting materials. Even structural metals might be much less used than on Earth (though of course you'd still need metals for many other uses).

A lot of structure might also be tunneled rather than 'built'.

Only reason to have more than one city is mineral resources or some site with particular scientific interest

Or cultural reasons.

I don't think we should think of Mars settlement as driven by conventional economics. It's fundamentally an ideological project.

The closest historical analogy might be something like the early settlement of Utah. Or perhaps something like Post, Texas (designed as an ideal community in an area then very much frontier), since the motivation isn't religious, though arguably culturally/sociologically it fills a similar role.

I think you might have multiple sites to avoid an "all eggs in one basket" situation as well as provide a more 'spacious' settlement model.

Perhaps one or a few very large inflatable pressure tents with smaller pressurized buildings inside (layered safety, plus more comfort inside). Square kilometers of pressure tents.

I'd expect more underground construction (tunnel boring, etc)

Musk has talked about geodesic domes, but I think these would more fill the role of "city parks" rather than be the site of the cities themselves - they'd be green spaces/recreational/view the stars areas, not containing residences.

[meekGee]

One question - is that 1M tons of payload, or 1M total tons (including the structure of landed Ships which can be repurposed)?

small print....

I'd say the ships are extra, but overall that's just a factor of 2 and we're talking orders of magnitude.
So 1Mton payload, 1Mt ships.

[Vultur]

One question - is that 1M tons of payload, or 1M total tons (including the structure of landed Ships which can be repurposed)?

small print....

I'd say the ships are extra, but overall that's just a factor of 2 and we're talking orders of magnitude.
So 1Mton payload, 1Mt ships.

Good point, it's not that large a difference.

The Ships as a resource will probably strongly shape how the first "generation" of Mars industry develops though.

[Robotbeat]

Ships as a resource is for the $1000/kg epoch. Once costs get to $100/kg, you’ll necessarily have had to minimize ship mass expended per mass of payload, and the ships will be worth sending back as well.

[Robotbeat]

Underground volume is very expensive. This is even true on Earth. Which is why tunnels are expensive.

Only if we find caves will it be cheaper than the surface.

Radiation and micrometeorites are basically not a concern at low altitudes. This is particularly true after partial terraforming, which I think will start almost immediately.

And you wouldn’t use domes. You’d use cable-restrained air mattress type structures, like Casey Handmer blogged about.

https://caseyhandmer.wordpress.com/2019/11/28/domes-are-very-over-rated/

[InterestedEngineer]

some of that 1M tons better be Rare Earth Elements (REEs), because they don't exist in significant quantities on Mars.

motors, batteries, electronics all use REEs.

[DanClemmensen]

some of that 1M tons better be Rare Earth Elements (REEs), because they don't exist in significant quantities on Mars.

motors, batteries, electronics all use REEs.

REEs are used in the most cost-effective motors, batteries, and electronics here on Earth, but that's because they were designed for the current Earth economy. There are alternative designs that do not use REEs. A trivial example is a motor that uses electromagnets instead of permanent magnets. If you are trying for Martian self-sufficiency, you may have to use these alternatives. I suspect that in almost every case, these alternatives will not be cost-effective here on Earth.

[Robotbeat]

As an example, my 2013 Model S uses no rare earth magnets. the induction motor works fine.

[Bob Shaw]

Nickel-Iron meteorites are common on the surface of Mars to the extent that NASA rovers have discovered them in passing. There may be polar areas where they are eroded out of subliming ice and preferentially deposited (as in Antarctica).

[DanClemmensen]

Nickel-Iron meteorites are common on the surface of Mars to the extent that NASA rovers have discovered them in passing. There may be polar areas where they are eroded out of subliming ice and preferentially deposited (as in Antarctica).

If the Martian economy will be based on electricity, they will need to find copper. Gold or silver would also work, but not likely to be available in industrial quantities. It's possible to use aluminum, but that would be a major constraint, I think.

[Vultur]

some of that 1M tons better be Rare Earth Elements (REEs), because they don't exist in significant quantities on Mars.

motors, batteries, electronics all use REEs.

REEs are used in the most cost-effective motors, batteries, and electronics here on Earth, but that's because they were designed for the current Earth economy. There are alternative designs that do not use REEs. A trivial example is a motor that uses electromagnets instead of permanent magnets. If you are trying for Martian self-sufficiency, you may have to use these alternatives. I suspect that in almost every case, these alternatives will not be cost-effective here on Earth.

I think industry on Mars will be full of all kinds of things that don't make economic sense on Earth but have been technically doable for a long time.

 

Underground volume is very expensive. This is even true on Earth. Which is why tunnels are expensive.

It's certainly true on Earth, but Earth has breathable surface pressure. I'm not so sure it's a bad solution on Mars, for very large structures.

(I mean large by Earth structure standards, not just large by current space hardware standards.)

It's still probably not the cheapest option, but it may be much more robust, thus arguably cheaper over time at least when risk is figured in.

Radiation and micrometeorites are basically not a concern at low altitudes. This is particularly true after partial terraforming, which I think will start almost immediately.

Micrometeorites are definitely not an issue, but I don't personally think partial terraforming will start quickly, and it may not happen at all (it may not be needed/cost-effective by the time it's possible).

And you wouldn’t use domes.

Not as primary settlement structures, no. But Elon Musk has definitely talked about them, so I think they have some place in the plan.

IMO it's probably as relatively small green space/star viewing areas (not the giant city domes seen in much SF) for a primarily underground settlement.

[DanClemmensen]

Underground volume is very expensive. This is even true on Earth. Which is why tunnels are expensive.

Only if we find caves will it be cheaper than the surface.

Sometimes it's cheaper to dig than to build on the surface, for example in loess soil, which is extensive on Mars. The majority of the population of some areas in China lived in Yaodongs, and in the year 2000, 40 million Chinese were estimated to live in them, although this millennium-long traditional way of life is now on the decline.
    https://en.wikipedia.org/wiki/Yaodong