Chapter 8: The Dead Worlds and the Icy Ones

The last chapter made a claim about a system: that air, ocean, land, thermostat and chemical disequilibrium are not five separate gifts from volcanism but five aspects of one circulating process, and that breaking any part of the circuit degrades the rest.

The way to test a claim of that kind is to find a planet with a component missing.

We have four of them within reach of a telescope, and one within reach of a rented car, and between them they remove almost every component in turn. This chapter is that experiment. It then goes further out, to a set of worlds where the engine runs on entirely different fuel and erupts something other than rock — which forces a question about what we have been calling volcanism all along.

Register: the observations are empirical, often from a single mission and sometimes from a single instrument. The interpretations are theoretical. The final section is speculative and marked.

The simple cases

The Moon is the control with almost every variable set to zero. It is small, it cooled quickly, and its volcanic history is essentially over. The dark maria that make the face are flood basalts, erupted mostly between roughly four and three billion years ago into impact basins, and then the engine ran out of usable heat and the surface stopped changing except by impact.

There is a complication worth noting, because it is recent and it cuts against the tidy version: samples returned by the Chang'e-5 mission are considerably younger than the previously known mare basalts, extending lunar volcanism substantially later than the textbook account allowed. The Moon stayed warm longer than we thought. It is still, unambiguously, finished.

Mercury is more interesting, because it shows you the shutdown in progress. Its surface carries extensive volcanic plains, so it had a productive period. It also carries lobate scarps — long thrust-fault cliffs, running for hundreds of kilometres, formed as the entire planet contracted. That is what a cooling body does: the interior loses heat, the whole planet shrinks by some kilometres of radius, and the crust, having nowhere to go, buckles and thrusts over itself.

Mercury's surface is a record of a planet visibly closing down. It does still have a magnetic field, which means some part of its core remains fluid, so the process is not complete. But the direction is not in doubt.

Neither body has an atmosphere, a hydrosphere, or a carbon cycle. No engine, no circuit, nothing.

Mars: the engine that ran down

Mars is the case that matters, because Mars is big enough to have had a real geological life and small enough to have finished it.

It has about half Earth's radius and around a tenth of its mass, which means a much larger surface area relative to volume, which means it lost heat faster. It also began with less radiogenic fuel in absolute terms. The engine was always going to be shorter-lived.

Its volcanoes are the diagnostic, and the diagnosis is beautifully clean. Olympus Mons is the largest volcano in the solar system — an edifice some twenty-odd kilometres high and hundreds of kilometres across, sitting on the vast Tharsis volcanic rise.

The question to ask is not why it is so large. It is why Earth has nothing like it, given that Earth has more heat and more volcanism.

The answer is that Earth's plates move. A hotspot beneath a moving plate builds a volcano, and then the plate carries that volcano away and the hotspot starts another one. You get a chain — Hawai'i and the Emperor Seamounts, from Chapter 1 — with the volume distributed along it. On Mars the lithosphere does not move. A hotspot beneath a stationary lid builds one mountain, in one place, for as long as the supply lasts.

Olympus Mons is enormous because Mars has no plate tectonics. Its size is not a sign of a more vigorous engine. It is a sign of a stalled conveyor.

And there is abundant evidence that Mars once had liquid water at the surface: valley networks, catastrophic outflow channels, deltaic deposits, and clay minerals that only form by prolonged water–rock interaction. It had an atmosphere thick enough and warm enough for that. It now has a wisp — a fraction of a per cent of Earth's surface pressure, almost entirely carbon dioxide.

Where did the rest go? Some was stripped to space, and there is spacecraft data quantifying that loss in progress. Some was lost to impacts. But a substantial part of the answer is the one this book is about: the carbon that dissolved into Martian water and precipitated as carbonate had no way back.

On Earth, buried carbonate is subducted, cooked, and returned to the atmosphere by arc volcanism. That is the return path, and it is what closes the thermostat's loop. Mars has no subduction. Carbonate formed on Mars stayed formed. The sink ran, the source declined, and nothing recycled. A one-way carbon cycle is not a cycle; it is a drain.

Mars is not quite dead — there is evidence for geologically recent volcanism in the Cerberus Fossae region, and seismometry has detected activity there. But it is a stagnant-lid planet with a spent atmosphere, and it shows you what the loss of the return path does over four billion years.

Venus: the loop with no return

And now the case I promised at the end of the last chapter, which I think is the single most useful planet in the solar system for the argument of this book.

Venus is Earth's twin by the crude measures: nearly the same radius, about eighty per cent of the mass, next door, made of the same material. If planetary size and composition and rough distance were what determined habitability, Venus would be habitable.

Its surface is around four hundred and sixty degrees Celsius — hot enough to melt lead — under about ninety times Earth's atmospheric pressure, in an atmosphere of almost pure carbon dioxide, beneath permanent clouds of sulfuric acid.

Now the part that matters. Venus is not short of volcanism. It is covered in it: tens of thousands of volcanic edifices, vast lava plains, pancake domes, and the coronae that appear to be the surface expression of mantle upwellings. Its surface is geologically young and remarkably uniform in age, which implies large-scale resurfacing — whether by one or more global events or by continuous slow renewal is argued. And there is evidence that it is volcanically active now: reanalysis of decades-old radar data has identified a vent that changed shape between two passes, along with other indications of recent flows and variability in atmospheric sulfur.

What Venus does not have is plate tectonics. It has a single, unbroken, stagnant lid. No spreading ridges, no subduction zones, no return path.

So construct the thermostat from Chapter 7 on Venus and watch it fail. The source term works fine — volcanoes supply carbon dioxide in quantity. The sink requires liquid water: rain, weathering of silicate rock, dissolved ions to the sea, carbonate precipitation. And the return requires subduction.

Venus has neither the sink nor the return.

How it lost the water is itself instructive, and it is the mechanism that would eventually have threatened Earth too. As the Sun brightened, Venus — being closer — crossed a threshold. Water evaporated; water vapour is a powerful greenhouse gas; the warming drove more evaporation; and the feedback ran away until the oceans, if there were oceans, were in the atmosphere. Water vapour reaching the upper atmosphere was broken apart by ultraviolet light, and the hydrogen, being light, escaped to space.

We can check this. Hydrogen escapes more easily than deuterium, so a planet that has lost an ocean this way should be left enriched in heavy hydrogen. Venus's atmosphere has a deuterium-to-hydrogen ratio far above Earth's — a large multiple. That is the fingerprint of an enormous quantity of water that left.

And once the water was gone, the sink was gone permanently. Volcanic carbon dioxide kept arriving and nothing removed it. Ninety-two bars of it are still there.

Let me state the lesson as plainly as I can, because it is the correction that keeps this book's thesis from being facile:

Volcanism is not sufficient for habitability. Venus proves it. What Chapter 7 actually established is the necessity of the circuit — outgassing plus a sink plus a return path. Venus has one of the three, in abundance, and it is the hottest solid surface in the solar system.

Anyone who reads this book as saying "volcanoes are good for planets" has misread it. The eruption is not the point. The loop is the point.

What the comparison establishes

Four bodies, four ways of failing.

The Moon and Mercury ran out of heat: no engine, no atmosphere, no cycle.

Mars kept some heat but never had a return path, and lost its atmosphere over billions of years with the carbon draining one way into rock and space.

Venus has heat and volcanism in abundance and no return path, and turned into a furnace.

Earth has all three components and has held its surface in the liquid-water range for four billion years.

That is a sample of five, which is not a statistical argument and I will not pretend it is. But it is a set of natural experiments in which the components are removed one at a time, and the results are consistent with the mechanism proposed in Chapter 7 and difficult to explain without it.

Io: the same phenomenon, different fuel

Now outward, where the assumptions start to come apart.

Io, the innermost of Jupiter's large moons, is the most volcanically active body in the solar system by a wide margin. It has hundreds of active volcanic centres, lava lakes tens of kilometres across, and plumes that rise hundreds of kilometres above the surface. Its heat flow, per unit area, is enormously greater than Earth's. It resurfaces itself continuously; it has essentially no impact craters, because they are buried faster than they accumulate.

Io is roughly the size of our Moon. By the reasoning applied to Mercury and the Moon above, it should be geologically dead many times over. Its radiogenic budget was exhausted long ago.

Its heat does not come from radioactivity or from the leftovers of accretion. It comes from orbital mechanics.

Io is locked in a resonance with Europa and Ganymede — for every orbit Ganymede completes, Europa completes two and Io completes four. That resonance keeps their orbits slightly non-circular. A moon in a non-circular orbit around a massive planet experiences a tidal bulge that changes in size and orientation over each orbit, so the whole body is repeatedly flexed. Flexing rock dissipates energy as heat. Jupiter's gravity is doing work on Io's interior, continuously, and the resonance prevents the eccentricity from damping away.

Io is entirely silicate — it is rock volcanism, not ice — and it has no water, no meaningful atmosphere, and no prospect of habitability whatsoever. Which is worth saying, because it makes the same point Venus does from the opposite direction: this is the most volcanically active object in the solar system and it is a wasteland.

But it establishes something the rest of the chapter depends on. Internal heat does not have to come from radioactive decay or primordial warmth. Orbital configuration will do it. And orbital configuration does not decay on the timescale that isotopes do.

Enceladus, and a correction that keeps recurring

Enceladus is a moon of Saturn about five hundred kilometres across — small enough that you could drive its circumference in a couple of days, if there were roads and if the surface were not water ice at well below a hundred kelvin.

From fractures near its south pole, informally called the tiger stripes, it erupts. Continuously. Plumes of water vapour and ice grains, extending thousands of kilometres, feeding one of Saturn's rings.

The Cassini spacecraft flew through those plumes, repeatedly, and sampled them in flight. What it found is the reason this small moon has reorganized a great deal of thinking about where life might be.

Salt. Sodium-rich, which means the water has been in prolonged contact with rock — it is not melted pure ice, it is ocean water that has leached a seafloor.

Silica nanoparticles. These form under fairly specific conditions of temperature and chemistry, and their presence implies hot water interacting with rock at the base of the ocean.

Molecular hydrogen. This is the important one. Hydrogen in the plume indicates ongoing water–rock reaction producing hydrogen faster than anything is consuming it.

Organics, including complex macromolecular material, and methane.

Independent measurement of the moon's slight wobble as it orbits established that the ice shell is not attached to the core — there is a global ocean between them.

Now put that alongside Chapter 6. Warm water. Rock. Hydrogen. Alkaline conditions. Hydrogen plus carbon dioxide is an energy-yielding reaction requiring no light and no photosynthetic oxygen, and it is the metabolism of the deepest-branching organisms we know.

Enceladus appears to have a hydrothermal system driven by water reacting with rock, generating hydrogen — which is to say, it appears to have something closely resembling Lost City, on a moon of Saturn.

And here is the correction I have now had to make three times in this book, which I think means it is a real pattern rather than an accident. Serpentinization is not magmatism. The chemistry that makes Enceladus interesting is water reacting with rock, generating its own heat and its own hydrogen. Tidal flexing supplies the energy that keeps the ocean liquid and the rock fractured and accessible — but the life-relevant chemistry, the hydrogen production, is the same non-magmatic reaction I had to concede in Chapter 6.

The engine is upstream of the chemistry, again. It is not the chemistry, again.

A caveat that should not be buried: the energy budget is not fully understood. Estimates of the tidal heating available to Enceladus have had difficulty accounting for the observed output over the age of the solar system. Something in the accounting is incomplete.

Europa, briefly, because it is about to change

Europa is the case where I should write least, because the situation is likely to be different by the time many people read this.

It has an ice shell over a global salt-water ocean containing more liquid water than all of Earth's oceans combined. Its surface is criss-crossed with fractures and disrupted "chaos" terrain, and it has very few impact craters, which means it is young and being resurfaced. Beneath the ocean there is rock, and therefore the possibility of a water–rock interface of the kind that matters. Being considerably larger than Enceladus, it has meaningful radiogenic heating in addition to tidal.

Plumes have been reported from Europa on several occasions, from telescope observations and from reanalysis of old spacecraft data. They are not confirmed, and the detections are contested.

The Europa Clipper mission is en route, having launched in 2024, and will begin its campaign of close flybys around the end of this decade. It is designed to settle a good deal of this. I have written this section to be as robust as I can manage to being overtaken, and I expect it to be overtaken anyway.

Titan

Titan is the strangest surface in the solar system and the one that most stretches the vocabulary.

It has a thick nitrogen atmosphere, denser at the surface than Earth's. It has weather, clouds, rain, rivers, lakes and seas — made of methane and ethane, because at ninety-odd kelvin methane occupies the role water plays here. It has dunes of organic solids. Beneath the surface there is very likely a water–ammonia ocean.

There are features that have been proposed as cryovolcanic — domes and depressions with the right morphology. The evidence is suggestive and has not converged; radar data admits other interpretations, and the community has grown more cautious about the identifications over time rather than less. The Dragonfly mission, a rotorcraft designed to fly between sites on the surface, is intended to launch toward the end of this decade and will address some of this, though not soon.

What counts as a volcano

Which brings the definitional problem to a head, and I want to handle it properly rather than waving at it.

If volcanism means the eruption of molten rock, then nothing in the last four sections except Io qualifies, and this part of the chapter belongs in a different book.

The alternative is to define volcanism by process rather than by substance: the eruption of buoyant, mobile interior material through a rigid outer shell, driven by internal heat, resurfacing the body and exchanging material between its interior and its exterior. Under that definition, ice is the rock and water is the magma, and Enceladus is a volcanic world.

I think the process definition is the right one, and I think it is genuinely illuminating — it lets you ask comparative questions about heat transport, resurfacing rates and interior–surface exchange across bodies made of completely different stuff.

But I am not going to pretend the analogy is free, and there is a specific physical problem that ought to be stated rather than glossed.

Water is denser than ice. On Earth, magma rises because melt is less dense than the solid it came from; buoyancy does the work, as described in Chapter 1. On an icy body, the melt is heavier than the shell it would have to rise through. Simple buoyancy does not merely work less well — it works backwards.

So cryovolcanism needs a different driver: pressurization from the freezing of an enclosed reservoir, which expands and squeezes the remaining liquid; or dissolved volatiles providing gas-driven ascent; or ammonia, which lowers both the melting point and the density of the liquid; or fractures opened by tidal stresses connecting an ocean directly to a vacuum, which is more or less what appears to be happening at Enceladus.

The mechanisms are different in kind, not just in degree. Calling both "volcanism" is defensible and useful. Calling them the same thing is not.

The habitable zone, redrawn

Register: speculative, and brief.

The conventional habitable zone is the range of distances from a star where a planet with a suitable atmosphere could have liquid water on its surface. It is defined by starlight. It is a ring.

Tidal heating breaks that framing. Enceladus and Europa are ten times further from the Sun than Earth is, receive a hundredth of the light, and have liquid water oceans that have plausibly persisted for a very long time — heated by orbital mechanics, and insulated by their own ice.

A body's internal energy budget, in other words, need not have anything to do with its star. Which means the set of places where liquid water can exist is very much larger than a ring, and includes moons of giant planets anywhere in a system.

Two consequences follow, and they point in opposite directions.

The optimistic one is that there may be far more liquid-water environments in the galaxy than the classical accounting suggests, most of them in places nobody was counting.

The uncomfortable one, for this book, is that those environments would be almost entirely invisible. A subsurface ocean under kilometres of ice produces no atmospheric biosignature, no surface spectrum, nothing detectable across interstellar distances. If most habitable environments are of this kind, then most of them are undetectable by any method currently imaginable, and our surveys are looking at the small, visible minority.

And there is a sharper problem for the argument of Part III, which I would rather raise myself.

Everything in Chapter 7 was about a surface habitability maintained by a tectonic circuit — outgassing, weathering, subduction, thermostat. Enceladus has none of that. No atmosphere, no weathering, no subduction, no carbon cycle in any recognizable form. If Enceladus turns out to be inhabited, then a biosphere can exist without the entire apparatus I have just spent a chapter arguing is necessary.

That is not a small objection. It is, in fact, the central objection, and I have been building toward it for three chapters.

So the next chapter asks it directly: is a living planet necessarily a firing one? And I will tell you now that I do not think the honest answer is yes.

Draft notes — verification status

Standing convention. This chapter has the highest density of mission-derived facts in the book and several sections depend on single instruments or single reanalyses. It also contains time-sensitive material about missions in flight that will need updating before publication.

Pending verification — high priority:

• Venus D/H ratio relative to Earth's (draft: "a large multiple"; commonly cited as ~100–150×). This is the chapter's key evidence for ocean loss and the number should be stated precisely.
• Venus surface conditions: ~464 °C, ~92 bar. Confirm.
• Evidence for ongoing Venusian volcanism: Herrick & Hensley (2023) reanalysis of Magellan radar showing a vent changing between 1990–92 passes at Maat Mons; Venus Express VIRTIS emissivity anomalies; atmospheric SO₂ variability. Verify each independently — the "active now" claim rests on them.
• Venus surface age and resurfacing: crater-count-derived ages (~300–1000 Ma) and the global-catastrophic versus continuous-resurfacing dispute. The draft says the question is argued; confirm that characterization is current.
• Olympus Mons dimensions (~22 km height, ~600 km diameter) and the claim that it is the largest volcano in the solar system.
• Mars atmospheric pressure (~6 mbar, ~95% CO₂) and MAVEN's quantification of ongoing atmospheric loss.
• Cerberus Fossae: evidence for geologically recent volcanism and InSight's detection of marsquakes localized there.

Pending verification — standard:

• Lunar mare basalt age range, and the Chang'e-5 sample ages extending lunar volcanism later than previously established (~2 Ga).
• Mercury's lobate scarps and the total radial contraction inferred (several km).
• Mercury's magnetic field and the inference of a partly liquid core.
• Mars/Earth size and mass ratios as stated.
• Martian carbonate inventory and the argument that carbon was locked without a return path — check this is the mainstream reading rather than one position among several.
• Io: number of active volcanic centres, heat flow per unit area relative to Earth's, plume heights, and the Laplace resonance ratio (1:2:4 Io:Europa:Ganymede).
• Enceladus: diameter (~500 km); plume composition from Cassini (sodium salts, silica nanoparticles, molecular hydrogen, methane, macromolecular organics); the libration measurement establishing a global ocean; the tiger stripe terminology.
• The Enceladus heat-budget shortfall — verify this is still an open problem and characterize it accurately.
• Europa: ice shell and ocean thickness estimates; total water volume relative to Earth's oceans; surface age from crater counts; the status of plume detections (HST observations and Galileo magnetometer reanalysis) as contested.
• Europa Clipper: launch date (October 2024), current cruise status, and expected arrival/flyby campaign start (~2030). Update before publication.
• Dragonfly: current launch date and arrival estimate. These have slipped before. Update before publication.
• Titan: surface atmospheric pressure (~1.5 bar), methane hydrology, named seas, and the current status of cryovolcanic feature identifications (Sotra Patera, Doom Mons) — the draft says the community has grown more cautious; verify this.
• The density argument against simple cryovolcanic buoyancy, and the proposed alternative drivers (freezing overpressure, exsolved volatiles, ammonia lowering density, tidally opened fractures).