Leaving Earth

Foundations

The exponential at the bottom of everything

Every destination in space has a price, and the price is delta-v: how much you need to change your velocity to get there.

The reason this is brutal is the rocket equation. To go faster you carry propellant. Propellant has mass. Carrying it means you need more propellant to accelerate the propellant. The required mass ratio grows exponentially with delta-v divided by exhaust velocity.

The practical consequence: reaching low Earth orbit means roughly ninety-four per cent of a rocket’s liftoff mass is propellant. Payload is a few per cent. That is not a failure of engineering, it is arithmetic, and better engines shift it modestly rather than removing it.

Getting to orbit is, famously, halfway to anywhere in the Solar System in delta-v terms. But it is still an exponential, and everything below is downstream of it.

The thing radiation does that people get wrong

On Earth you are shielded twice: by the magnetic field, which deflects charged particles, and by ten tonnes of atmosphere above every square metre.

Leave, and you meet galactic cosmic rays — high-energy charged particles from outside the Solar System, including heavy nuclei like iron.

A dosimeter rode inside the capsule that carried the Curiosity rover to Mars, and measured what an astronaut would receive on that journey: roughly 1.8 millisieverts per day, behind real spacecraft shielding. A round-trip cruise adds up to several hundred millisieverts, before a single day on the surface.

Now the counterintuitive part. You cannot solve this by adding shielding, at least not straightforwardly. When a high-energy heavy nucleus hits a shield, it does not simply stop. It fragments, producing a shower of secondary particles. Thin or moderate shielding can leave the dose worse than no shielding at all. Effective shielding means a lot of mass — and mass is exactly what the rocket equation charges you exponentially for.

The biological risk estimates carry large uncertainties, dominated by our poor understanding of what heavy ions do to tissue, since there is no population that has ever been exposed to them.

What a year in space does to a person

The most careful measurement available compared an astronaut who spent nearly a year in orbit with his identical twin on the ground — the same genome, different environments.

The findings were extensive: changes in gene expression, immune function, the microbiome, cognitive test performance, the structure of the eye, and telomere length behaving unexpectedly.

Most of it returned toward baseline within six months of landing. Some of it had not.

That is roughly the right shape of answer: the human body is more adaptable than pessimists expected and less than optimists hoped, and the effects are real and partly persistent.

And there is a gap in the data worth naming. Nobody has ever spent an extended period at partial gravity. We have data at 1 g and at approximately 0 g. Mars is 0.38 g. Whether that is enough to prevent bone and muscle loss is entirely unknown, and it is the single largest physiological unknown in any long-duration plan.

Mars, without the romance

Surface pressure is under one per cent of Earth’s. That is far below the altitude at which body fluids boil at body temperature. Every moment outdoors requires a pressure suit, not just an oxygen mask.

The atmosphere is mostly carbon dioxide, so it provides no radiation shielding worth the name and nothing to breathe.

There is no global magnetic field.

The soil contains perchlorates, which are toxic and would need removing before anything is grown in it.

Water exists, as ice, and that genuinely matters — it is the reason the poles and the shadowed polar craters are interesting.

The paper that ended the terraforming argument

The standard reply to all of the above is terraforming: release the frozen carbon dioxide, thicken the atmosphere, warm the planet, and let it become habitable over time.

In 2018 two researchers did the obvious thing and counted.

They inventoried every accessible carbon dioxide reservoir on Mars: the polar caps, gas adsorbed onto soil grains, carbonate minerals in the crust, everything the observations constrain.

Release all of it, and atmospheric pressure rises to a small fraction of what habitability requires. Not close. Not close enough for liquid water to be stable across the surface, and nowhere near breathable.

Their conclusion, stated plainly in the paper: terraforming Mars is not possible with present-day technology.

This is a wonderful example of how science settles an argument. Not by rhetoric about ambition, but by taking an inventory and finding the cupboard nearly empty.

And we cannot yet build a working ecosystem

One more, and it is the least glamorous and possibly the most important.

Biosphere 2 was a sealed habitat in Arizona, built to house eight people with a closed, self-sustaining ecosystem. It cost a great deal and was built by capable people.

Oxygen concentrations fell steadily and had to be injected from outside. Most vertebrate species and all pollinating insects went extinct inside it. The system did not close.

Nobody has since demonstrated a closed ecological life-support system that sustains people indefinitely without resupply, at any scale.

Which means that even setting aside the rocket equation, the radiation and the pressure, we currently cannot build the thing a self-sufficient settlement would consist of. We have never made a small Earth that works.

Which makes the backup-planet framing exactly backwards

Put the comparison honestly.

Take the worst plausible Earth. Severe warming, ecosystem collapse, even a nuclear exchange. That Earth still has: breathable atmospheric pressure, a magnetic field, an ozone layer, liquid water, soil containing a functioning microbial ecosystem, and a biosphere that has recovered from five mass extinctions.

Mars has none of those and cannot be given them with existing technology.

Any effort capable of making Mars habitable would more easily make a devastated Earth habitable. The backup-planet story survives only if you do not run the comparison.

This is not an argument against going. It is an argument against a specific bad reason for going, one that functions to make Earth’s problems feel optional.

The real case, which is stronger

Because here is the thing: space is one of the most valuable things our species does. The value is just pointed the other way.

Instruments looking down. Lesson 8.4 described India’s groundwater vanishing, measured by two satellites detecting a mass anomaly through microscopic changes in their separation. That measurement does not exist without spaceflight. Neither does the sea level record, the ice mass record, the global temperature record, or the atmospheric composition record. This course’s evidence base for chapter 8 is substantially made of satellite data.

Navigation, communication, weather. Cyclone warnings in the Bay of Bengal have saved enormous numbers of lives, and they run on satellites.

Science. Everything in chapters 1 to 4 of this course — the cosmic microwave background, exoplanets, the composition of other worlds — came from instruments above the atmosphere.

And planetary defence, which is the one genuinely existential service space provides, and which lesson 5.7 made the case for.

India’s programme illustrates the whole argument. Chandrayaan-1 carried an instrument that detected hydroxyl and water molecules distributed widely across the lunar surface — a first-rank scientific result that rewrote what was believed about the Moon. The Mars Orbiter Mission reached Mars orbit in 2014 for a fraction of the cost of comparable missions. Chandrayaan-3 landed in the lunar south polar region in 2023, a region that matters precisely because permanently shadowed craters there may hold ice.

That is what a space programme is for, and none of it depends on anybody emigrating.

So should we go?

Yes, and for the reasons that survive scrutiny.

Because the science is extraordinary. Because the engineering has produced enormous spin-off value. Because a species that can deflect an asteroid is safer than one that cannot. Because permanent presence on the Moon is genuinely plausible this century, and the Moon is three days away rather than nine months.

And because it is, on the whole, one of the better things we do with our attention.

Just not as a backup, and not as an alternative to fixing the thing we already have. Chapter 4 spent seven lessons on how improbably specific this planet is. Nothing in this lesson changes that. If anything it sharpens it: we have surveyed the neighbourhood, and there is nothing else remotely like home.

Deeper

The rocket equation, written out

The velocity change achievable is the exhaust velocity multiplied by the natural logarithm of the ratio of initial to final mass. Inverting gives mass ratio as an exponential in delta-v over exhaust velocity. For a delta-v to low Earth orbit of about 9.4 kilometres per second including losses, and a chemical exhaust velocity around 3.4 kilometres per second, the required mass ratio is roughly sixteen, meaning about ninety-four per cent of the initial mass is propellant. Staging helps by discarding empty tank mass, which is why launch vehicles are built in stages rather than as one tank. Higher exhaust velocity is the only lever with real leverage, which is why electric and nuclear propulsion are pursued despite low thrust.

Why heavy ions are the hard part of the dose

Galactic cosmic rays include nuclei up to iron, at energies where a single particle can deposit a dense track of ionisation through a cell. This differs qualitatively from the sparse ionisation of X-rays or gamma rays, and the biological effect per unit dose is larger and less well characterised. Risk models therefore apply quality factors with substantial uncertainty. Because human epidemiological data come from photon exposures, extrapolation to heavy ions is the dominant uncertainty in mission risk estimates, and it is unlikely to be resolved without accelerator-based radiobiology on a scale that has not yet been done.

What Jakosky and Edwards counted

The inventory covers the residual and seasonal polar caps, carbon dioxide adsorbed in the regolith, carbonate minerals identified from orbital spectroscopy and in meteorites, and the possibility of buried deposits. Constraints come from gravity and radar observations of the caps, neutron spectrometry, and mineral detections. The paper is careful about what it cannot exclude, in particular deeply buried carbonate that observations do not reach, and notes that hypothetical undiscovered reservoirs would have to be very large and would still require energy to mobilise. The conclusion is therefore about accessible carbon dioxide given present knowledge and technology, which is the relevant question.

What went wrong inside Biosphere 2

Oxygen declined from about twenty-one per cent toward the mid-teens over roughly sixteen months. The eventual explanation involved microbial respiration in organic-rich soil consuming oxygen, with the resulting carbon dioxide taken up by curing concrete rather than returning to the atmosphere, breaking the expected balance. This is worth dwelling on: the failure came from an unmodelled interaction between soil biology and building material. A closed system fails through the coupling you did not think to include, which is a general property and not a criticism of that project specifically.

Frontier

Partial gravity

The decisive experiment has not been done, and would require either a rotating facility in orbit or a long-duration lunar presence with careful physiological monitoring. Until it is, statements about human health during multi-year Mars surface stays are extrapolations between two data points with nothing in between.

Closing the loop

Life support research continues on higher-recovery water and atmosphere recycling and on food production in controlled environments. Systems in current use recover a large fraction of water and some oxygen, but none approaches closure, and the gap between high recovery and true closure is where all the difficulty lives.

Transcript and shot list

Target runtime about 10 minutes. Built to demolish the backup-planet story with numbers, then build the real case for space, which is stronger and points the other way. Shot types: ANIM is rendered from code with Manim, AI is a generated shot from Higgsfield with the shared style reference, REAL is agency imagery with credit, CAM is the narrator.

[00:00] S01 · CAM · Narrator direct to camera.

After a full nuclear exchange, Earth would still be more habitable than Mars. That is not rhetoric. It is a comparison of atmospheric pressure, radiation dose, water and soil, and Mars loses every single one. Let me show you the numbers, and then let me tell you what space is actually for, because the real answer is better than the story.

[00:28] S02 · ANIM · Delta-v as a price list for destinations. Then the rocket equation: propellant needed to accelerate propellant, mass ratio rising exponentially. The result drawn as a vehicle cross-section — 94 per cent propellant, a few per cent payload.

Every destination in space has a price, and the price is delta-v: how much you have to change your velocity to get there. And the reason that is brutal is the rocket equation. To go faster you carry propellant, propellant has mass, and carrying it means you need more propellant to accelerate the propellant. The required mass ratio grows exponentially. Practical consequence: getting to low Earth orbit means about ninety-four per cent of a rocket’s liftoff mass is propellant. Payload is a few per cent. That is not an engineering failure, it is arithmetic, and better engines move it a bit rather than removing it.

[01:17] S03 · ANIM · Earth’s two shields: the magnetic field deflecting charged particles, and ten tonnes of atmosphere per square metre. Then both removed, and galactic cosmic rays including iron nuclei arriving.

On Earth you are shielded twice. By the magnetic field, which deflects charged particles, and by ten tonnes of atmosphere sitting above every square metre. Leave, and you meet galactic cosmic rays: high-energy charged particles from outside the Solar System, including heavy nuclei like iron.

[01:37] S04 · ANIM · A dosimeter inside the Curiosity cruise capsule reading about 1.8 millisieverts a day; the round trip totalling several hundred before any surface time. Then a heavy nucleus striking a shield and fragmenting into a secondary shower, with the dose behind thin shielding shown as worse than none.

A dosimeter rode inside the capsule that carried the Curiosity rover to Mars and measured what an astronaut would take on that journey: roughly 1.8 millisieverts a day, behind real spacecraft shielding. A round trip is several hundred millisieverts before a single day on the surface. And here is the counterintuitive part. You cannot straightforwardly fix this with more shielding. When a high-energy heavy nucleus hits a shield it does not stop, it fragments, and it produces a shower of secondaries. Thin or moderate shielding can leave the dose worse than no shielding at all. Effective shielding means a lot of mass, and mass is precisely what the rocket equation charges you exponentially for.

[02:30] S05 · ANIM · The twin study: same genome, one in orbit for nearly a year, one on the ground. Changes in gene expression, immune function, microbiome, cognition, eye structure, telomeres. Most returning toward baseline within six months; some not. Then the data gap: points at 1 g and 0 g, nothing between, Mars marked at 0.38.

The most careful measurement we have compared an astronaut who spent nearly a year in orbit with his identical twin on the ground. Same genome, different environments. The findings were extensive: gene expression, immune function, the microbiome, cognitive test performance, the structure of the eye, telomeres behaving unexpectedly. Most of it returned toward baseline within six months of landing. Some of it had not. Which is about the right shape of answer — the body is more adaptable than pessimists expected and less than optimists hoped. And there is a gap worth naming. Nobody has ever spent a long period at partial gravity. We have data at one g and at approximately zero. Mars is 0.38. Whether that is enough to stop bone and muscle loss is completely unknown, and it is the biggest physiological unknown in any long-duration plan.

[03:35] S06 · ANIM · Mars audited: surface pressure under one per cent of Earth’s, below the boiling point of body fluids at body temperature; a CO₂ atmosphere providing no shielding and nothing to breathe; no global magnetic field; perchlorates in the soil; water present as ice at the poles and in shadowed craters.

Mars without the romance. Surface pressure is under one per cent of Earth’s, far below the level at which your body fluids boil at body temperature — so every moment outside needs a pressure suit, not an oxygen mask. The atmosphere is mostly carbon dioxide, which gives you no useful radiation shielding and nothing to breathe. There is no global magnetic field. The soil contains perchlorates, which are toxic and would have to be removed before you grew anything in it. Water does exist, as ice, and that genuinely matters — it is why the poles and the shadowed polar craters are the interesting places.

[04:23] S07 · ANIM · The terraforming inventory: polar caps, adsorbed regolith, carbonate minerals, everything the observations constrain, each counted. The total released, and pressure rising to a small fraction of what habitability needs. The conclusion stated as printed.

The standard reply to all of that is terraforming. Release the frozen carbon dioxide, thicken the atmosphere, warm the planet. In 2018 two researchers did the obvious thing and counted. They inventoried every accessible carbon dioxide reservoir on Mars: polar caps, gas adsorbed onto soil grains, carbonate minerals in the crust, everything the observations constrain. Release all of it and the pressure rises to a small fraction of what habitability requires. Not close. Not close enough for liquid water to be stable at the surface, and nowhere near breathable. Their conclusion, in the paper: terraforming Mars is not possible with present-day technology. That is a lovely example of how science settles an argument. Not with rhetoric about ambition. By taking an inventory and finding the cupboard nearly empty.

[05:22] S08 · REAL · Biosphere 2 exterior and interior (agency imagery; confirm credit). S09 · ANIM · Oxygen declining from 21 per cent toward the mid-teens; the eventual explanation — soil microbes consuming oxygen, the resulting CO₂ absorbed by curing concrete; oxygen injected from outside; pollinators extinct.

One more, and it is the least glamorous and maybe the most important. Biosphere 2 was a sealed habitat in Arizona built to house eight people with a closed, self-sustaining ecosystem. It cost a lot and capable people built it. Oxygen fell steadily and had to be injected from outside. Most vertebrate species and all the pollinating insects went extinct inside it. The system did not close. And the eventual explanation is worth hearing: soil microbes consuming oxygen, with the carbon dioxide they produced being absorbed by curing concrete instead of returning to the air. The failure came from an interaction nobody had modelled, between soil biology and building material. Nobody since has demonstrated a closed life-support system that sustains people indefinitely without resupply, at any scale. Which means that even setting aside the rocket equation and the radiation and the pressure, we cannot currently build the thing a self-sufficient settlement would be made of. We have never made a small Earth that works.

[06:38] S10 · ANIM · The comparison run honestly: worst plausible Earth — severe warming, ecosystem collapse, nuclear exchange — still holding breathable pressure, magnetic field, ozone, liquid water, living soil, a biosphere that has come back from five mass extinctions. Mars beside it with none of them.

So run the comparison honestly. Take the worst plausible Earth. Severe warming, ecosystem collapse, even a nuclear exchange. That Earth still has breathable atmospheric pressure, a magnetic field, an ozone layer, liquid water, soil with a functioning microbial ecosystem, and a biosphere that has recovered from five mass extinctions. Mars has none of those and cannot be given them with existing technology. Which means any effort capable of making Mars habitable would more easily make a devastated Earth habitable. The backup-planet story only survives if you never run the comparison.

[07:19] S11 · CAM · Narrator direct to camera.

That is not an argument against going. It is an argument against one specific bad reason for going — the one whose function is to make Earth’s problems feel optional. Because space is one of the most valuable things our species does. The value just points the other way.

[07:42] S12 · ANIM · Instruments looking down: the two GRACE satellites from lesson 8.4 weighing India’s missing groundwater; then sea level, ice mass, global temperature and atmospheric composition records, each tagged as satellite-derived. Then cyclone warning in the Bay of Bengal.

Lesson 8.4 showed you India’s groundwater vanishing, measured by two satellites detecting a mass anomaly through microscopic changes in the distance between them. That measurement does not exist without spaceflight. Neither does the sea level record, the ice mass record, the global temperature record or the atmospheric composition record. This course’s evidence base for chapter eight is substantially made of satellite data. Add navigation, communications and weather — cyclone warnings in the Bay of Bengal have saved an enormous number of lives, and they run on satellites.

[08:22] S13 · ANIM · Chandrayaan-1’s Moon Mineralogy Mapper detecting hydroxyl and water across the lunar surface; the Mars Orbiter Mission reaching Mars orbit in 2014 at low cost; Chandrayaan-3 landing in the lunar south polar region in 2023, with permanently shadowed craters marked as possible ice reservoirs.

And everything in chapters one to four of this course — the cosmic microwave background, exoplanets, the composition of other worlds — came from instruments above the atmosphere. Plus planetary defence, which is the one genuinely existential service space provides, and lesson 5.7 made that case. India’s programme illustrates the whole argument. Chandrayaan-1 carried an instrument that detected hydroxyl and water molecules spread widely across the lunar surface, a first-rank result that rewrote what was believed about the Moon. The Mars Orbiter Mission reached Mars orbit in 2014 for a fraction of the cost of comparable missions. Chandrayaan-3 landed in the lunar south polar region in 2023, and that region matters precisely because permanently shadowed craters there may hold ice. That is what a space programme is for, and none of it needs anybody to emigrate.

[09:25] S14 · CAM · Narrator direct to camera, then the calendar returns.

So should we go? Yes, for the reasons that survive scrutiny. Because the science is extraordinary. Because a species that can deflect an asteroid is safer than one that cannot. Because a permanent presence on the Moon is genuinely plausible this century, and the Moon is three days away rather than nine months. And because it is one of the better things we do with our attention. Just not as a backup, and not as an alternative to fixing the thing we already have. Chapter four spent seven lessons on how improbably specific this planet is. Nothing here changes that. If anything it sharpens it. We have surveyed the neighbourhood. There is nothing else remotely like home. One lesson left, and it goes all the way to the end.

Shot list

Zero AI shots, eleven code-rendered shots, one archive image, three narrator shots. Real mission imagery is available and free for most of this; generated planetary vistas would undercut a lesson whose entire argument is that the romantic version is wrong.