23. Settlement BOM (Bill of Mass)

We used to joke that Mars was built by accountants with welder’s masks. The tool that ran our lives wasn’t a saw or a crawler — it was the ledger. On Earth hardware businesses keep a Bill of Materials — the BOM (pronounced “bomb”). On Mars we kept a Bill of Mass — it was our BOM (“bomb”). 

In that first century of the human era, every idea, however shiny, owed us kilograms. If it couldn’t pay, it didn’t fly. Literally.

The cleanest way to make this tangible is by independence steps. Each “ring” of settlement that pulled us up the ladder of independence had a weight. You’ve already seen how we arranged barrels and barns, how we powered them, how we scored layouts. Now we ask the question that quietly ruled all the others: how many kilotonnes did it take to build the thing?

♦

A note on language before the ledger marches on: we kept saying “rings,” even when the geometry had the decency not to be circular. What we meant was waves—each new build tended to encircle the last, so the name stuck. Early on it really was barrels-and-barns in loops; later it looked more like a disciplined mesh pretending to be polite. We still called them rings. I argued for “zones” (legacy British firmware; apologies), but you can take the droid out of London, not London out of the droid. So rings it was: A, B, C, and the accountants were happy.

We’ll anchor the numbers at the first meaningful civic scale: Ring A — Independence Level (IL-3) — about 2,000 residents. Think of this as Port Harmonia’s “village-city” moment: continuous shirtsleeve streets, six ‘great’ halls, off-belt wedges for power/industry, and a starport far enough away to keep our air clean and our nerves steady. What follows is conservative—bias high, round up, sleep well.

♦

Port Harmonia BOM roll-up: Ring A (IL-3) ~2,000 residents

This is just the core on-plan BOM.  Typically this was an under-estimate of mass transit reality as Mars had a habit of eating up supplies, and we did not prioritize having lots of auditors on the surface to follow every kilogram.

A few honest spotlights for those of you following along with calculators and wanting to build your own settlement in a new part of Mars.

Interiors basis. By the end of the first century with ISRU we could furnished roughly 345,000 m² of pressurized floor (≈ 285k in rings/halls + ≈ 60k off-belt), at ~30 kg/m² → ~10.35 kt. That’s partitions/doors/fixtures — district plant lives on its own lines above.  Well maybe it was more like ~40 kg/m², but that should give you a good steer.

Power (Hybrid). Two independent ~8 MWe reactors (heat co-used), 60 MWdc PV across two wedges, 150 MWh distributed storage, and an HVDC ring (~25 km). If had gone full solar-led, we would have had to add ~3–4 kt (mostly batteries). Nuclear-led trims ~2–3 kt at this scale (probably), but that wasn’t really the reason we diversified into nuclear as you know.

Transport & lineside. You pay for the corridors and cars once; they quietly carry everything: people, pallets, pipes, and the future you didn’t plan for.  Investing here was priceless.

♦

Does it smell right? Counting structure plus the construction kit, we were sitting at roughly 30,000 t over about 280,000 m² of enclosed floor — call it ~107 kg/m² for the big shells. Strip that to structure only and you’re in the ~65–85 kg/m² band. On Earth, the equivalents routinely land in the hundreds to well over a thousand once you pay for façades, cores, and wind. We buried, we arched, and we let the soil do the heavy lifting; that’s why the ratio stayed lean. 

Per person, the math was equally plain: ~55.1 kt divided by 2,000 people is about 27.5 tonnes each to stand up a town that could ride out a window without panic. Not to be independent, but to be a start. 

Mars frontline operator’s preferred a different yardstick — ships. A first 300 m-radius loop wants roughly 1.9–3.8 kt of steel shell depending on how cautious you are; add a full mid-deck and you’re at ~2.6–4.5 kt. Allow ~12% fabrication loss and assume ~100 t of usable stainless per donor hull and you arrive, reluctantly, at ~50 donor ships per loop link at baseline specs. The six barns — though actually a series of smaller barns, but in area the same  — at 80 m by 250 m and two floors, eat ~15 donor ships each on the same assumptions. That is why geopolymer and basalt-fiber infill were not a style choice; they were how we pulled the ship count out of the triple digits. 

Raise set-point pressure or ask for heroic spans and you will buy rings and ties by the tonne. Mars is kind to compression. It is indifferent to ambition.

♦

The mass you never notice but always needed lived off the main line. 

We carried a bootstrap kit of about 15 kt: landing rigs and recovery tools; an earthmoving set sized for trenches and berms rather than operatic mining; jigs to flatten donor steel; a small panel line; mobile weld and cut cells; temporary power; a handful of emergency shells; liner reels and clamp stock matched to the two mega-SKUs; a year’s worth of high-criticality spares. We double-bought some tools on purpose; the desert abhors single points of anything. 

Alongside it, an extra “miscellany” of roughly 10 kt paid for the honest untidiness of firsts: training mockups and test galleries; a sacrificial bay to prove procedures; a few too many doors because someone reversed a forklift with enthusiasm; buffers for dust season; the occasional bad bet retired gently. Add those unobtrusive necessities to the ~55 kt for the built town and you reach a planning total close to 80 kt to achieve IL-3 with margin. 

On the day, every kilogram of that last line felt like foresight rather than waste.

♦

Independence did not make us hermits. 

At Ring A, with about 2,000 people, we set a cap of half a tonne per person per 26-month window and sized ships and shelves to match. On paper that came out around 0.46 kt per year; in practice we budgeted a round 1.0 kt per year because Mars cheats — a dust deck leans on your insolation, a membrane batch sulks, a supplier quietly changes an alloy. 

The layout score kept the city’s shape honest; the bill of mass paid for its pulse. We kept the imports high-value and low-mass: membranes, catalysts, meds, opto-electronics, a little flavor for the galley, the occasional “please don’t fail” spare. Boring on purpose, which is another way of saying civilised.

♦

If you are looking for the places where tonnes hide, start with rooms, power, arteries, and hands that do not sleep. 

Interiors scale with honesty: at roughly 30 kg/m² you get partitions that stand up to apprentices, doors that seal on the first try, and fixtures that don’t turn a fault into a festival. Shave it too far and maintenance will write your history in red ink. 

Power choices echo across the ledger. Our hybrid pick cost on the order of five kilotonnes at Ring A for two modest reactors, a sane slab of PV, and batteries sized for quality rather than heroics. Go solar-led and you will add batteries by the kilotonne; go nuclear-led and you might save a couple of kilotonnes at the price of nerves and paperwork. We split the reactors across two vendors and slept fine. 

Transport buys you time disguised as mass. Those nine-and-a-bit kilotonnes of tram, UPS-Train, conveyors, pipes, and paving are not glamour; they are the difference between a compressor arriving in minutes along a designed path and a weekend lost to detours. Cities are their arteries. And robots outrun romance. Half a kilotonne buys crawlers that never forget a flange, carts that don’t call in sick, and a dozen humanoids that will hang a liner while the real engineers argue about lunch. Civilisation, up close, is a choreography of small reliable wins.

If you read the ledger that way, the numbers stop being abstract. They become rooms you can stand in, doors you can trust, watts you can count, and the quiet thrum of a town that breathes without asking permission.

♦

We’d balanced the first ledger; now we widen the page. After Ring A the pattern stopped being mysterious: another link, another tranche of barns, another turn of the power wedge, and the metrics moved in ways you could predict with your thumb on a slide rule. The trick was to stay conservative, because Mars always finds the hidden line item.

♦

Each settlement “ring” is a kind of mega kit: a double links of barrels to create the next layout, the equivalent of eight mega barns added to the civic core, off-belt packs for the wedges, a modest power uplift, and a little more starport/ISRU/GSE to keep cadence honest. We kept the structural recipe steady — haunched horseshoe links, two decks where justified, barns in the later block versions in the 80 m span class — then let ISRU shave steel at the margins as panel lines matured.

Here’s the conservative ledger, stage by stage. “Added pressurized area” is shirtsleeve floor; power uplifts are increments beyond the previous stage; steady-state imports are what kept the place civil between windows, not capex for the next ring.

♦

A few truths hide in those round numbers.

First, shells + interiors dominate. We budgeted ~60 kg/m² imported for the stainless bits when ISRU was young, and ~30 kg/m² for interior fit-out that didn’t hate maintenance. You can save a handful of kilotonnes per ring as geopolymers and basalt-fiber ribs take over the straight plate work, but the shape stays the same: people require rooms, and rooms require honest kilograms.

Second, power stays modest if you stay hybrid. Each ring gets another ~30 MWdc of PV, another ~100–150 MWh of distributed batteries, and (until D) one more ~8 MWe reactor. The point is not maximal self-reliance; it is smooth days and calm nights. We pushed heat from the reactors into the district loops, let PV carry the shoulders, and kept batteries in the business of quality and ride-through—not week-long storms. Switch to solar-led and the batteries turn your ledger bold.

Third, transport scales gently when geometry behaves. The ring tram grows by a few kilometres, the UPS-Train (Unpressurized track Pressurized cabin Settlement Train) of the second century gets cars and a station or three, conveyors and pipes extend like vines. That’s low single-digit kilotonnes per ring, and it buys you time every day thereafter.

Finally, starport/ISRU/GSE increments are real but not ruinous—about 1–2 kt per ring for cadence and redundancy. We never rebuilt the port; we taught it new tricks.

♦

When the metrics went sideways, it was almost always because someone tried to be clever in the wrong place. 

Curved plan pieces for “flow,” four-way junctions for “choice,” bulkheads at every door for “safety.” The CLS caught the layout sins; the BOM priced them. The reverse was also true: a clean T-junction kit, reserve lots platted early, utility stubs in the right walls — those are kilograms you don’t pay twice.

We let materials maturity move the dial slowly. From B to D we shifted more of the plate work into regolith geopolymers and basalt-fiber members; stainless retreated toward rings, knees, clamps, and doors. It shaved ~8–15 kt per ring by D in a few places, but we never booked the savings until jigs and QA had been boring for a season.

♦

There’s an accountant’s pride in watching a line go flat. As independence rose, per-person imports fell, but population climbed. The outcome was a plateau: roughly a kilotonne a year through Rings B–D, despite a city that tripled in size. Even the solar nerds smiled at that.

The steady-state cargo was oddly intimate: membranes, catalysts, med cartridges, opto-electronics, high-strain seals, refrigerants we didn’t yet make cleanly, the odd rare alloy for a door hinge we refused to redesign. Food was a rounding error by C. Fashion was not a line item; we had better vices.

If you’re keeping score against the old out-loud promises: Ring A’s target ≤0.5 t/person per window came in just under by some measures; by D we were living at ≤0.2, with total tonnes per window hovering near ~2.2 kt (≈ 1.0 kt/year). 

Imports that flatten, on purpose. Boring, durable, correct. The path to independence.

♦

But there was one surprise that really in the hindsight of the second century wasn’t a surprise: satellite towns and outposts

Field crews weren’t surprised that the bill doubled; they were the ones hauling cable across the flats. A single capital hub is efficient. A living planet is not a point.

We built science stations on high ground, mining camps where the ore and ice agreed, pad operations towns to keep the starport honest, PV service depots in the megafarms, and relay nodes that turned whispers into networks. Each of those sites carried fixed costs that don’t scale with crew size: a starved ring and two barns, a small wedge of power and batteries, one honest clinic, one honest machine shop, a corridor that doesn’t quit in dust.

You can make a 100-person outpost sing on paper at ~0.2 t/person per window — and then you remember it needs 10–15 kt to exist without killing anyone. Do that a dozen times, scale a few to 300–500 people when the geology makes you greedy, and the extra mass stacks faster than lovers’ promises.

By the time Port Harmonia sat around ~250 kt of built imported dry mass (A-through-early D, conservatively), the network — satellites, feeders, second and third starport pads, an extra reactor island where it made sense, HVDC that went places — had added another ~300–350 kt across the first two centuries. That is how the total bill drifted toward ~600 kt to reach a true IL-6 (>10,000 people across Mars). Not self-sufficient yet, but undeniably alive: basic food, water, air, power, civics — covered — and the rest on a gentle glide slope rather than a cliff.

Accountants frowned at the tonnes-per-person on the rim sites. Crews smiled and pointed at the ore, the spectra, and the parts that stopped arriving broken. Fix costs on Mars operated like fixed costs on Earth, even in the lower gravity.

♦

We used to pass around a shaky video of one of the first boosters being tower caught and pretend that schedules were linear. The line from the voiceover that stuck was the loud one: we need million tonnes to seed a self-sustaining city. How wrong was it?

Not very, if you accept that a city is not one dot. Stack Port Harmonia at ~250 kt (A–D), add its satellites to pass ~600 kt at IL-6, then let ambition widen to regional meshes, polar extraction, and second-generation industry. And that’s required tonnes assuming on-plan mass-utilization. The reality with everything human was there was wastage and simply parts of one-way rockets we landed that did not get used.

You can see the runway from here into the second century and beyond, to true independence of IL-7+: one million tonnes and probably in reality two million when everything is added up was not fantasy; it was the next ladder on the self-sustaining rung when you stop measuring just Port Harmonia and start counting Mars.

The part that was gloriously wrong was the clock. We learned — again — that optimism is a force multiplier for motivation and a terrible estimator. Those at the time should have tripled the early ambitious schedules of trillionaire visionaries and called the difference wisdom. The planet forgave us; the ledgers never did.

♦

In the end the Bill of Mass was not a punishment; it was a rhythm section. It gave tempo to design, kept harmony with the CLS on form, and let barrels and barns modulate without changing the key. When an expansion ring of settlement came due, we knew the link chords: kilograms for shells, for rooms, for electrons; a little brass for rockets; strings for the places in between. And always the quiet percussion of robots and doors, doors and robots. (Side-bar: I should probably write a whole codex just on Mars’ droids and bulkheads!)

We did not get to a magic number where ships stopped coming. We got to something better: a slope shallow enough that growth paid for growth, and a city that could hum through a dust season without asking anyone’s permission. In those later centuries on certain nights, when the compressors were kind and the tram sang true, even an old droid could hear the tune.

That’s the page, widened: settlement expansion rings B through D in those later centuries counted honestly; imports flattened by design; satellites turning a hub into a world; the million-tonne boast redeemed in spirit if not in schedule. 

The appendix helps to break the ledger down by district and decade — the way budgets become neighborhoods.

Now you should understand how we used a million tonnes on Mars in that first century.