The Zen of Earth’s Water
Part 2 of Where Does Earth’s Water Come From?
“The only Zen you find on the tops of mountains is the Zen you bring up there with you.”
Robert M. Pirsig, Zen and the Art of Motorcycle Maintenance
The presence of water on Earth presents two mysteries. Why is there any water on Earth? And why is there a lucky, Goldilocks quantity of water on Earth, equivalent to about 0.02% of the planetary weight?
For a long time, exogenous delivery — that Earth’s water was delivered by meteorites from the outer solar system — seemed to be the most obvious explanation for the presence of water on Earth. It seemed obvious given the circumstances of solar system formation. Earth, and the three other rocky planets (Mercury, Venus, Mars) were formed too close to the young sun, on the wrong side of the snow line. The snow line was the distance from the young sun during solar system formation beyond which water ice could exist.
The leading candidate for the exogenous delivery is called the late-veneer hypothesis. This hypothesis posited that between about 4.45 and 4.35 billion years ago, even before the Moon formed, Earth was struck by a group of meteorites from the cold outer solar system that coated a dry Earth with its oceans.
For decades, this was the accepted orthodoxy because it solved three mysteries at once: the types of gases near our surface, the specific blend of hydrogen in our seawater, and the surprising presence of gold and platinum in our crust. But the theory fell apart when scientists tested the isotopic signature of a rare metal in Earth’s mantle called ruthenium (Ru). Instead of matching wet, outer-space asteroids, Earth’s ruthenium perfectly matches bone-dry rocks born close to the sun.
As such, we are forced to concede that the late-veneer delivery trucks arrived from the inner solar system rather than the outer, and that they were bone-dry. They brought us our gold, but not water, because they couldn’t possibly have.
If the exogenous delivery can’t explain Earth’s water, we are forced to look at the only remaining suspect: Earth itself. If the water wasn’t shipped here late, it had to be here from day one. Just as the Zen you find at the tops of the mountains is the Zen you brought up there with you, the water we find in the depths of the oceans is the water Earth had all along.
But how does a volatile compound like water survive a 3,000°C Hadean magma ocean and the cataclysmic, moon-forming impact of Theia? Surprisingly, the answer to this question seems to solve both mysteries: the presence of water on Earth and the very specific amount of water on Earth. Remember, if the Earth had any less or any more water, there might not have been any life.
The answer lies deep inside Earth, literally thousands of kilometers beneath the surface, in the inaccessible core.
Let’s descend.
1. The Core as Hydrogen Storage
For over a century, the foundational paradigm of planetary geology was built on a clean, binary division of compositional labor between Earth’s layers.
In this old model, the mantle was a silicate rock engine responsible for tectonics and volcanism and other interesting stuff Earth does from time to time. The core, on the other hand, was an uninteresting, sterile sphere of molten iron and nickel completely separated in composition and temperament from the mantle. Volatile elements like hydrogen, carbon, and nitrogen were classified as highly incompatible with the iron-nickel core.
The paradigm dictated that during planetary formation, these flighty, volatile gases were entirely forced upward into the early atmosphere, completely decoupled from the iron-loving (siderophile) heavy metals sinking into the interior. It was an elegant, orderly architecture. It allowed geochemists to model surface oceans and atmospheric evolution without ever having to worry about what was happening 3,000 kilometers beneath their feet.
But here’s what’s exciting about human knowledge: a paradigm only holds until the tools of measurement advance enough to probe its boundary conditions.
The structural collapse of this binary framework occurred when experimental geochemists began simulating the extreme thermodynamics of the deep Hadean Earth. At the core-mantle boundary, pressures scale to a crushing 130 gigapascals, and temperatures exceed 5,000 kelvins. That’s almost as hot as the Sun’s surface, and about a million times more pressure than at Earth’s sea level. Under these sci-fi parameters, the classical rules of chemistry completely invert. Volatility submits to pressure.
In early 2026, Huang and Murakami reported using laser-heated diamond anvil cells to replicate the precise environment of early core segregation. They demonstrated that under the conditions mimicking the Hadean era, hydrogen ceases to behave like an elusive gas. Instead, it becomes profoundly siderophile—it finds a love for iron.
As the molten proto-Earth was separating into layers, a titanic chemical vacuum cleaner was operating at the center of the planet. Sinking liquid iron ruthlessly scavenged hydrogen from the surrounding magma ocean, locking it into the metallic matrix as stable iron hydrides.
The scale suggested by this new paradigm is truly staggering: Huang and Murakami’s data implicate the Earth’s core in sequestering the single largest hydrogen reservoir on the planet, hoarding up to 0.36% of its mass. When converted to its water-equivalent total (WET) by pairing it with the mantle’s limitless oxygen supply, the maths says that the core contains between 9 and 45 times more water-equivalent mass than all our surface oceans combined.
2. The Multi-Layered Planetary Vault
There’s a classic narrative device in mythology: if an artifact is too important to risk a coming cataclysm, you break it apart and hide the pieces in separate, unreachable vaults. Only when the threat stabilizes can the fragments reconverge to fulfill their purpose.
It seems like Earth did something similar with water.
To save its water from the scorching fury of the Hadean Eon, Earth executed a perfect geochemical partition. It split the ingredients of water — hydrogen and oxygen — across a vast elemental chasm. It hid the rare, flighty hydrogen deep within the iron matrix of the core, while storing a near-infinite ocean of oxygen safely in the silicate rocks of the mantle.
This process is dictated by a chemical metric known as the partition coefficient (D). In geochemistry, a partition coefficient is the mathematical ratio that describes how an element distributes itself between two competing mediums under given temperature and pressure conditions. In the case of Water, the elements were Hydrogen and Oxygen, and the competing mediums were molten metal (core) versus liquid silicate rock (mantle).
The mathematical expression for the partition coefficient of an element partitioned between the metallic core and the rocky mantle is:
DElement = [Element]metal-core / [Element]rock-mantle under given temperature and pressure. This ratio will change when temperatures and pressures change.
For Hydrogen, this can be written as: DHydrogen = [Hydrogen]metal-core / [Hydrogen]rock-mantle. During the high-pressure, high-temperature conditions of Hadeon when core segregation took place, the value of D for hydrogen was overwhelmingly skewed toward the metal. The core sucked up hydrogen until it reached a precise state of thermodynamic equilibrium with the overlying mantle.
When the violent impacts ceased and the planet finally began to cool, the environmental conditions shifted, and the lock on the vault slowly turned. As pressure and temperature gradients adjusted, hydrogen began to outgas, leaking across the core-mantle boundary and rising through the rocky interior.
This is where the multi-part artifact reassembled. The ascending hydrogen rose as a raw kinetic force, scavenging oxygen from the mantle’s iron oxides on its way up. Like a phoenix rising from an iron furnace, these elements synthesized to form water at the finish line. The volcanic vents of the early Earth belched out a highly specific, mathematically calculated remainder: the exact 0.02% water-equivalent mass that the core could no longer hold under the new thermodynamic regime.
If Earth had been even slightly smaller or larger, the gravitational pressure at the core-mantle interface would have shifted the partition coefficient. A slightly different equation would have either left the hydrogen permanently entombed in the iron core, leaving a bone-dry desert planet, or forced a massive over-allocation to the surface, drowning the continents under a global ocean. The life-sustaining blue layer we observe today is the precise chemical remainder of a deep-Earth balancing equation.
3. Earth’s Sterile Siblings
If Earth’s hydrosphere is the result of a perfectly executed geochemical heist where the ingredients of water were split, safely sequestered, and seamlessly recombined, then the rest of the inner solar system represents a series of botched jobs. The Moon, Mars, Venus, and Mercury all possessed the same foundational ingredients. But their specific planetary (and lunary) constraints caused their water vaults to leak, freeze, or fail entirely.1
The Moon represents a vault that never closed. Remember, the presence of water molecules on Earth predates the Moon entirely. The Moon was formed when a Mars-sized impactor, Theia, slammed into the Earth around 4.4 billion years ago. The Moon should have inherited a proportional water budget from Earth. But it is famously dry.2 This is because of the Moon’s unusual formation process.
Unlike the planets, the Moon did not form via slow, high-pressure accretion. It was a child of interplanetary violence, condensed from the ultra-high-temperature and low-pressure ejecta disk whipped into orbit by the giant Earth-Theia impact. Without the crushing gravitational mass of a true planet, the Moon could not generate the pressures required to dissolve hydrogen into its small core. Lacking a heavy atmospheric or mantle clamp, the vaporized volatiles simply bled out into the vacuum of space before a vault could even be constructed.
Mars represents the next failure mode: building a vault but lacking the energy to run a locking mechanism.
Mars is only about 11% of Earth’s mass and half of Earth’s size. Because it is smaller, it has a higher Surface-Area-to-Volume ratio than Earth. Because it is lighter, its surface gravity is only about 40% of Earth’s. These factors combined caused Mars’s core to rapidly lose its primordial heat and freeze solid. After the core solidified, the planet’s protective magnetosphere shut down (you need a rotating metallic core for a magnetosphere). A magnetosphere is what protects a planet from solar wind, and in its absence, the solar wind ruthlessly stripped the Martian atmosphere. The solar ultraviolet (UV) radiation then split the surface water into hydrogen and oxygen and drove the hydrogen into space, leaving behind a geodynamically dead desert planet.
Venus is a tragic case of what could have been, considering it is Earth’s near-twin in mass and size. This similarity implies that its internal core-vault pressure and partition coefficient were nearly identical to our own. Which means it almost certainly resurrected its own volcanic water phoenix in its core. The tragedy occurred at its surface.
Because Venus is positioned closer to the Sun, its early outgassed water vapor sat in an intensely hot environment that never changed because Venus has a broken thermostat — i.e., it has no plate tectonics to run the carbon-silicate cycle. As carbon dioxide built up into a crushing, supercritical nightmare, the surface oceans boiled away completely. The vaporized water rose into the upper atmosphere, where solar UV systematically ripped the hydrogen molecules apart. The light hydrogen atoms escaped into space, leaving the heavier deuterium behind. Today, Venus’s atmospheric Deuterium-to-Hydrogen ratio is 100 times higher than Earth’s—the ultimate chemical fingerprint of a planetary vault that leaked its entire ocean into the void.
Mercury is the easiest to explain as it is the outlier. It is essentially a giant iron core wrapped in a minimalistic rock mantle. Because it formed so close to the young Sun, Mercury was subjected to blistering thermal radiation that left its pre-planetary dust profoundly depleted of volatiles from day one. Any primitive hydrogen that managed to partition into its massive core was subsequently lost when early giant impacts blasted away most of its original silicate mantle.
To summarize, each of the other major rocky bodies in the inner solar system besides Earth — the Moon, Mars, Venus, and Mercury — represents its own particular kind of failure in realizing their own Water Zen.
Earth, on the other hand, accreted wet, partitioned its ingredients across a vast thermodynamic divide, insulated its hydrogen inside a core-alloy safe-house to survive the Hadean cataclysms, and resurrected it as a surface hydrosphere via mantle outgassing. The equations balance and the failures of our sterile sister planets seem to confirm the chain of evidence.
But as any rigorous investigator knows, circumstantial evidence does not a conviction make.
4. The Final Forensic Frontier
The excellent 2026 paper by Huang and Murakami is a lab-analog study. It relied on laboratory proxies to simulate what could have happened in the Earth’s core by squeezing microscopic mineral grains between laser-heated diamond anvil cells to simulate the crushing pressures and temperatures of the deep Earth—then extrapolating those results on a planetary scale.
This study, and others like it, do not analyze actual samples from Earth’s core. Because they cannot. Any direct analysis of Earth’s deep interior remains out of reach of humanity’s current technological capabilities. The definitive proof of the endogenous model is locked behind a (currently) unpenetrable technological wall.
The absolute limit of Earth’s technosphere beneath our feet reaches about 12 kilometers or 7 miles. That is the depth of the Kola Superdeep Borehole. This is ironic considering that in the other direction, above our heads, we have reached multiple planets in the Solar System, and the farthest specimen of human technology, the Voyager spacecraft, is 24 light-hours away. But when it comes to the ground we stand on, we have barely scratched the surface, literally and figuratively.
The reason why we haven’t been able to dig beyond is that at depths of a few miles, the rocks begin to plastically deform under intense heat and pressure, and our drill bits fail. At its deepest, the Kola Superdeep Borehole is just 0.2% of the distance to the center of the Earth. We have designed probes that have broken through the heliopause into interstellar space. And yet we fundamentally lack the materials science required to penetrate our own planetary engine room.
When we figure out how to make an automated titanium-boron-carbide probe or something along those lines, capable of piercing through the mantle and bringing back solid core samples from 3000 kilometers deep, we can experimentally test the parameters of the partition model:
We can verify whether the core actually hoards up to 0.36 wt% hydrogen in its iron-nickel alloy matrix.
We can measure the deuterium-to-hydrogen ratio of the core’s iron hydrides. Because the core has been a closed system since the Hadean, its deuterium-to-hydrogen ratio should act as a perfect time capsule, preserving the unmodified signature of the primordial inner solar system. We would check if this deep signature matches the baseline for the inner solar system, allowing us to mathematically calculate exactly how much light hydrogen escaped during outgassing to form the surface oceans.
We can test the core sample to understand how other “contaminants” like Silicon, Oxygen, and Sulfur affect the solubility of Hydrogen, because DHydrogen is highly sensitive to the presence of other dissolved elements. This would allow us to add real-world nuance to this idealized model.
All this to say: Because direct physical sampling is an engineering impossibility today, the ultimate test of the elegant endogenous model remains postponed, waiting on technology to enable it. Until we can physically retrieve and audit an actual piece of the core, our understanding of where Earth’s water comes from will remain a theoretical triumph constructed from analog studies. The case is incredibly compelling, but the vault remains unbreached.
5. The Symmetrical Abyss
Our attempts to resolve Earth’s found-water mystery lead us to a bizarre, symmetrical irony.
The human species has always looked to the stars above to find its place in the cosmic wilderness. We used to build elaborate myths out of constellations. Now we build multibillion-dollar space telescopes to sniff the atmospheres of exoplanets light-years away, looking for the spectral signature of vaporized water. Now we dream of sending robotic submarines into the subsurface oceans of Europa or Enceladus, searching for a second genesis of biology. All of these are worthy endeavors, with the power to make human spirits soar and human consciousness expand. And yet.
And yet we treat the search for life in the universe as an exclusively outward journey to be hitchhiked on the backbone of the Night. But the geochemistry of our own world hints at another possibility: maybe the map of cosmic habitability is buried directly beneath us, thousands of kilometers deep in the crushing, hot, pitch-black violence of our own iron core.
The mystery of Earth’s water is cosmic in its implications. If the old exogenous orthodoxy were right, if water on Earth is a superficial coating delivered by a lucky late-stage cosmic bombardment caused by Jupiter’s wayward walk, then Earth is a freak of nature and habitable worlds are a statistical anomaly. Consequently, Life is the product of an orbital lottery that most, if not all, protoplanetary nebulae lose.
But if the endogenous theory is correct, then water is a standard structural feature of planetary birth. Any planetary birth. Instead of most rocky worlds in the galaxy being deserts because they never received an exogenous delivery, the galaxy is full of worlds that contain the elemental ingredients of water as an endogenous inheritance.
Habitability, then, depends on a perfectly executed, long-term thermodynamic balancing act. It is not a cosmic lottery. Every planet is born with a water vault, and variables of planetary physics determine which ones can manage to utilize it. And there can be multiple combinations of these variables, multiple ways of sticking the landing in the synchronized dance of planetary mass, solar distance, magnetic field, and tectonic machinery, that lead to habitability.
Any rocky planet of sufficient mass, accreting from standard inner-system dust, inherently possesses a deep metallic vault capable of sheltering the raw ingredients of an ocean from the fury of its birth. It means that right now, across billions of star systems in our galaxy, at least a few million silent, cooling planets are slowly unlocking their cores, turning their own unique combination locks on their own particular thermodynamic vaults, and sweating oceans into their mantles to resurrect their own volcanic phoenixes.
And here we are. Standing on a thin silicate crust, sandwiched between two vast, enticing oceans. Above us lies the cosmic ocean of space, which we are just beginning to sail. Below us lies the iron ocean of our core, which holds the primordial blueprint of our survival but remains out of our physical reach for now. To understand the infinite expanse above, we must also reconcile with the unbreached dark below.
This is the ultimate Zen of Earth’s water.
With thanks to Mike Riggs, Elizabeth Van Nostrand, Rhishi Pethe, Tina Marsh Dalton, Abby ShalekBriski, and Jannik Reigl.
This section emerged in response to huskercr’s thoughtful comment on Part 1.
As Neil Armstrong remarked on landing: “When you kick the surface, [the dust goes out in] a little fan which, to me, is in the shape of a rose petal. There's just a little ring of particles—nothing behind 'em—no dust, no swirl, no nothing. It's really unique.”
A fascinating exploration of why Earth has the water it does and why Mercury, Venus, the Moon and Mars do not. The jury is of course still out as to how rare Earth and life are in the cosmos, and your hopeful analysis at the end of the piece was encouraging.
Thanks for the two articles, I enjoyed the read.
A question. If this is true, it would imply that liquid oceans are ultimately determined by terrestrial planet mass; any planet formed within a proper region of a protostellar disk would sequester and then outgas Hydrogen and Oxygen to form water; and any planet of sufficient mass would have a geology that could preserve that liquid surface water.
If this is true, and plate tectonics are a function of liquid water differentiating continental and oceanic crust and providing mantle lubrication so the plates can move; then why didn't Venus develop plate tectonics and thus have a sequestration cycle that would prevent its runaway atmospheric buildup and water loss? Are plate tectonics more complicated than I realize, or did the amount of water on Venus result in too much of one kind of crust? Or is the thought that Venus's proximity to the Sun never allowed the surface to cool enough for oceans to pool and begin the tectonic sequestration cycle?