Nitrogen supply is the limiting constraint on the construction of pressurized habitats, and therefore on the rate at which gyrealms can be commissioned. The autopoietic industry has closed its supply loop on every other class of input: structural metals, silicates, water, carbon, and oxygen are all obtainable in quantity from bodies already in high orbit, and the energy to process them is free at the point of collection. Molecular nitrogen is the single exception. It cannot be made from rock in any useful quantity, it must be lifted out of a planetary gravity well, and it does not come back. Habitat construction is consequently paced not by the industry’s capacity to build — which is very large — but by the rate at which nitrogen can be delivered to the construction sites.
Why nitrogen#
A habitable atmosphere requires an oxygen partial pressure near 21 kPa and a diluent to make up the rest. The diluent is not optional. In a pure-oxygen atmosphere at the same partial pressure, ordinary materials ignite readily and burn with a flame speed that no practical fire-suppression system can outrun; flammability tracks the oxygen mole fraction rather than its partial pressure, so an atmosphere that is safe to breathe and safe to live in must be mostly something else.
Nitrogen is the diluent every terrestrial organism is adapted to. It is chemically inert at habitat conditions, physiologically neutral at one bar, and — being close in molar mass to oxygen — gives an atmosphere whose density, speed of sound, viscosity, and heat capacity reproduce the conditions under which Earth’s weather, flight, hearing, and respiration all evolved. A habitat filled with any other diluent is a habitat whose every fluid behaviour must be re-derived.
The difficulty is cosmochemical. Oxygen is bound in the silicates and oxides that make up most of the accessible mass of the inner Solar System and is recovered as a by-product of ordinary refining; carbon and hydrogen are abundant in carbonaceous and icy bodies. Nitrogen is volatile, was largely lost from the inner system during planetary formation, and survives in rocky bodies only in trace concentrations. It is present in lunar regolith at some tens of parts per million, implanted by the solar wind, and in the most nitrogen-rich carbonaceous chondrites at one to three thousand parts per million, bound into refractory organics and ammonium salts that must be liberated by bulk pyrolysis. Recovering a tonne of nitrogen from such material means heating and processing several hundred to many thousand tonnes of rock.
Earth’s atmosphere, by contrast, is 78% nitrogen by volume — an ore grade of roughly 755,000 parts per million, already in the gas phase, requiring nothing more exotic than compression and fractional liquefaction. No other accessible reservoir in the inner system is within two orders of magnitude of it.
The atmospheric inventory of a habitat#
The quantity required is large, and larger than the naive estimate. On a planet, atmospheric pressure falls off exponentially with altitude and the air is effectively over within a few scale heights. Inside a rotating habitat it is not: the apparent gravity falls linearly from full value at the floor to zero on the spin axis, so the pressure gradient vanishes as the axis is approached and the air column terminates at a finite pressure rather than tapering into vacuum.
For an isothermal atmosphere in a cylinder of radius R rotating to give floor acceleration g, the pressure at radius r is
P(r) = P₀ · exp[ −(R² − r²) / (2RH) ]
where H = RₛT/g is the scale height at floor conditions, about 8.4 km for Earth-normal air. Two results follow. Pressure on the axis is P₀·exp(−R/2H): for a 10 km radius, 56 kPa, comparable to the pressure at 4,800 m of terrestrial altitude — thin, but breathable, and present all the way up. And the air mass standing over each square meter of floor is
σ = (P₀/g) · [ 1 − exp(−R/2H) ]
in which P₀/g is exactly Earth’s own column, 10,330 kg·m⁻². The bracketed factor is 0.45 at a 10 km radius. A gyrealm of Coriopolis’s diameter therefore carries about 4,600 kg of air over every square meter of ground, of which some 3,500 kg is nitrogen — less than half of what Earth carries, at identical floor pressure, because the column both converges toward the axis and weighs less as it goes.
For an interior open to the axis, a cylinder 20 km in diameter and 80 km long holds on the order of 2.3 × 10¹³ kg of gas, of which roughly 1.75 × 10¹³ kg — seventeen and a half billion tonnes — is nitrogen. Interior structure cuts that substantially: an inner shell at 8 km radius truncates the column at about 2,000 kg·m⁻², under half the open figure, and takes the habitat’s nitrogen bill down with it.
Three properties of that inventory shape everything downstream.
It is a volume cost, not an area cost. The nitrogen bill is set by enclosed volume, and the open landscape of the main level is almost the whole of it. A decked interior three meters between floor and ceiling holds 3.7 kg of air per square meter of usable ground — roughly one twelve-hundredth of the open column. Habitable area can therefore be multiplied almost for free by decking, while every square meter of open sky is paid for at full price. Sky, not floor, is the expensive commodity in a gyrealm.
It scales with radius. Because the bracketed factor rises with R, a wide habitat costs more air per square meter of ground than a narrow one: halving the radius to 5 km drops the column from 4,600 to 2,650 kg·m⁻². Diameter is chosen against the horizon, the sky depth, and the Coriolis gradients that a narrow cylinder makes harsher — but it is chosen knowing that width is paid for in nitrogen.
It is not consumed. Nitrogen in a sealed habitat is inert and permanent. Biological fixation locks up a negligible fraction — the standing nitrogen in the soil and biomass of a fully established biome is of order 10⁹ kg against an atmospheric inventory of 10¹³, a few hundredths of one percent — and denitrification returns most of it. Demand for nitrogen is therefore proportional to construction, not to population. An existing habitat consumes none. This is the single most consequential fact in the political argument: halting construction would arrest the terrestrial drawdown immediately and cost no living resident anything at all.
The cost of the lift#
What makes nitrogen expensive is not the gas but the gravity well. Moving asteroidal material between orbits costs of order 0.5 MJ·kg⁻¹. Raising mass from Earth’s surface to low orbit costs at minimum about 34 MJ·kg⁻¹ — some 9.4 MWh per tonne — before any inefficiency. Nitrogen is thus dearer than every other input the industry touches by a factor of thirty to a hundred, and no advance in engineering can reduce the figure, because it is set by the depth of the well rather than by the method of climbing it. The atmosphere of a single large habitat represents on the order of 8 × 10²⁰ J of lift energy: several years of early-21st-century terrestrial electrical generation, expended on air.
Collection from the upper atmosphere by orbiting scoops avoids building a launcher and avoids atmospheric flight, but does not avoid the bill. A collector sweeping gas at orbital velocity must be thrust continuously against the drag its own intake produces, and the capture is inelastic: only half the work done goes into the gas’s orbital kinetic energy, the remainder appearing as heat that must be radiated. The thrust work is accordingly around 61 MJ·kg⁻¹, worse per tonne than an ideal lift and better in capital. Both routes are powered, in the end, from orbital solar collection, and the choice between them is one of infrastructure rather than physics.
Alternative reservoirs#
The outer Solar System is not nitrogen-poor. Titan alone carries an atmosphere of about 9 × 10¹⁸ kg at 147 kPa surface pressure, some 95% nitrogen by volume — more than twice the nitrogen in Earth’s entire atmosphere, standing on a body whose escape velocity is 2.6 km·s⁻¹ against Earth’s 11.2. Ammonia ices and the atmospheres of the gas giants hold vastly more again. On the narrow question of well depth, outer-system nitrogen is cheaper than Earth’s.
What separates it from the construction sites is not energy but time. Reaching the inner system from Saturn’s orbit requires shedding several kilometers per second of heliocentric velocity, recoverable in part by gravity assist and by aerocapture on arrival; more importantly, a one-way transit is measured in years, and establishing production at 9.5 AU means committing capital against a return that cannot begin to arrive for a decade or more. Earth’s atmosphere is not the cheapest nitrogen in the Solar System. It is the nearest in time, which under a constraint that binds construction schedules is the property that decides. The permanence of the constraint therefore rests on a logistics judgement rather than a physical one — which is precisely why it is arguable, and why it is argued.
Consequences for deployment#
Because nitrogen is the only supply-limited input, it is effectively the only priced one, and the marginal cost of a habitat approximates the cost of its atmosphere. Structure, machinery, and labour — the dominant costs of every previous era of construction — are supplied by an industry that reproduces itself and are close to free at the margin. Cost therefore scales with enclosed volume rather than with mass, complexity, or fit-out, which inverts the entire inherited economics of building.
The observable consequence is an industry idling below its own capacity. The Works can double their productive base on a timescale of days where feedstock and energy are abundant; habitat commissioning proceeds at a fraction of that rate, queued behind nitrogen delivery. Surplus capacity is absorbed by unpressurized products — power collection, splinky guideway, tugs, mirrors, refining and fabrication plant — which are unconstrained, and which is why the volume of unpressurized infrastructure in the inhabited orbits so greatly exceeds the volume of habitat.
Design practice responds along three lines:
- Reduced total pressure. Holding oxygen partial pressure at the physiological requirement and cutting the diluent lowers the nitrogen bill in direct proportion — an atmosphere at 0.7 bar with 30% oxygen uses roughly 40% less nitrogen than one at Earth-normal. It is bought against flammability, which follows mole fraction, and against a raft of secondary changes: water boils near 90 °C, wings and rotors need more speed, and the acoustic and convective behaviour of the interior shifts. Reduced-pressure atmospheres are standard in industrial and service volumes and are contentious in inhabited biomes.
- Substitute diluents. Argon, neon, and helium have all been examined and none solves the problem. Argon is a mere 0.9% of Earth’s atmosphere, and the radiogenic argon retained in rock occurs at the same parts-per-million grades that make rock a poor nitrogen ore; neon is rarer still; helium diffuses through structural materials and is lost from any habitat that holds it. The substitution problem is the nitrogen problem restated.
- Conservation. An inventory of 10¹³ kg makes even minute fractional losses expensive: a habitat losing one part in ten thousand per year sheds over two million tonnes annually. Airlocks are pumped down and their contents recovered rather than vented — a thousand cubic meters vented at full pressure is 1.2 tonnes gone, and a busy habitat cycles locks thousands of times a day. Hull leak budgets are specified in parts per million per year, punctures are treated as inventory losses as much as safety events, and a decommissioned habitat is drained before it is dismantled.
Terrestrial effects#
The export is measurable at the source. Removing one percent of Earth’s atmospheric nitrogen — roughly 3.9 × 10¹⁶ kg, the inventory of some two thousand large habitats — lowers mean sea-level pressure by about 7.5 hPa, or 0.74%, which is equivalent in barometric terms to raising the whole surface of the planet some 60 m in elevation. Because the extraction is selective, oxygen partial pressure is unchanged and there is no physiological effect; the oxygen fraction rises slightly, from 20.95% toward 21.1%, with a correspondingly slight increase in the flammability of everything. Reduced total pressure marginally weakens the pressure-broadening of the infrared absorption lines of carbon dioxide and water vapour, so the drawdown is, incidentally, a very small cooling intervention. Air density falls in proportion, with proportionate effects on aviation, wind loading, and the thermal inertia of the atmosphere.
At the fractions removed to date none of these is a crisis, and this is the crux of the dispute rather than an answer to it. The quantity that matters is not the present depletion but its integral: the drawdown is monotonic, the return flow is zero, and demand grows with the built inventory of habitats rather than declining with it. What is contested is the trajectory — the built area still to come, the date at which an outer-system reservoir could begin to substitute, and the discount properly applied to a loss distributed across every future inhabitant of the planet.
The argument is made harder, not easier, by the fact that emigration has been good for Earth in nearly every other respect. Living in a gyrealm is cheaper than living on Earth, so the outflow is not an elite one; terrestrial pollution, land pressure, and habitat destruction have all fallen sharply as population has moved upward. The gain is immediate, personal, and distributed among people who can name it. The loss is slow, planetary, and belongs to no one in particular. That asymmetry, and not any dispute about the underlying physics, is what has made the nitrogen question intractable.