Epigraph

٢٧ أَأَنتُمْ أَشَدُّ خَلْقًا أَمِ ٱلسَّمَآءُ ۚ بَنَىٰهَا

٢٨ رَفَعَ سَمْكَهَا فَسَوَّىٰهَا

٢٩ وَأَغْطَشَ لَيْلَهَا وَأَخْرَجَ ضُحَىٰهَا

٣٠ وَٱلْأَرْضَ بَعْدَ ذَٰلِكَ دَحَىٰهَآ

٣١ أَخْرَجَ مِنْهَا مَآءَهَا وَمَرْعَىٰهَا

٣٢ وَٱلْجِبَالَ أَرْسَىٰهَا

٣٣ مَتَٰعًا لَّكُمْ وَلِأَنْعَٰمِكُمْ

“Which is harder to create: you people or the sky that He built, raising it high and perfecting it, giving darkness to its night and bringing out its morning brightness, and the earth, too, He spread out, bringing waters and pastures out of it, and setting firm mountains in it, for you and your animals to enjoy?” (Al Qurʾān 79:27–33)

Abstract

Water covers most of Earth’s surface and pervades its interior, yet the question of where and how our planet acquired its water remains among the most actively investigated problems in cosmochemistry and planetary science. This essay synthesizes the current evidence-weighted picture. Water’s constituent atoms have an ancient pedigree: hydrogen was forged in Big Bang nucleosynthesis, while oxygen was manufactured in the cores of massive stars and dispersed by supernovae; the water molecule itself formed on cold dust grains in molecular clouds, with deuterium enrichment recording a chemical history that predates the Sun. Because Earth accreted inside the solar nebula’s “snow line,” where temperatures were too high for water ice to condense, its wetness demands explanation. I review the competing and complementary hypotheses—late cometary delivery, carbonaceous-chondrite asteroidal delivery, inheritance from enstatite-chondrite-like building blocks, ingassing of nebular hydrogen into a magma ocean, and solar-wind implantation—and weigh them against the isotopic constraints provided by deuterium-to-hydrogen (D/H) ratios, nitrogen and noble-gas isotopes, and highly siderophile elements. Sample-return missions (Hayabusa2/Ryugu, OSIRIS-REx/Bennu, Hayabusa/Itokawa) and deep-Earth mineralogy (hydrous ringwoodite) sharpen the account. The emerging consensus favors a hybrid origin dominated by chondritic material—both dry inner-Solar-System building blocks and water-rich carbonaceous bodies—with comets a minor contributor and nebular gas a possible minor supplement. The essay closes with a reflective epilogue on water’s meaning.

1. Introduction: A Question Written in Isotopes

Earth is, by the standards of the inner Solar System, remarkably wet. Its surface oceans hold roughly 1.335 × 10²¹ kilograms of water—the standard value derived from NOAA’s ETOPO1 global relief model—and its mantle may hold as much again or more. That this should be so is not obvious. The standard narrative of terrestrial-planet formation places Earth’s assembly in a region of the solar nebula too hot for water ice to survive. The reconciliation of a dry birthplace with a wet planet is the central puzzle of this essay, and its resolution has been pursued largely through the forensic reading of isotope ratios—above all the ratio of deuterium (D, or ²H) to ordinary hydrogen (H, or ¹H) in water. Because D/H in water increases with distance from the young Sun and is set by the temperature at which the water’s ice condensed, it functions as a chemical fingerprint of provenance. Nitrogen isotopes (¹⁵N/¹⁴N) and noble-gas abundances provide independent, cross-checking constraints. This essay follows the water from its cosmic origins through its delivery to Earth to the physical evidence preserved in rocks, returned asteroid samples, and diamonds.

2. The Cosmic Origin of Water Itself

Water is a compound of two elements with radically different birthplaces. Hydrogen is primordial: essentially all of it was produced in the first minutes after the Big Bang, during Big Bang nucleosynthesis, which generated hydrogen, deuterium, helium and trace lithium and left the universe with no elements heavier than these until stars formed. Oxygen, by contrast, is a stellar product. It is synthesized in the cores of massive stars, in part through the CNO fusion cycle and helium burning, and is dispersed into interstellar space when those stars die as supernovae. Water therefore could not exist until the first generation of stars had lived and died.

Recent modeling suggests this may have happened startlingly early. Simulations by Daniel Whalen, Muhammad Latif and Bhaskar Agarwal (Whalen, Latif & Jessop, “Abundant water from primordial supernovae at cosmic dawn,” Nature Astronomy 9, 741–746, 2025) indicate that the deaths of the first (Population III) stars as core-collapse and pair-instability supernovae could have seeded their surroundings with oxygen that combined with residual hydrogen to form water at redshifts z ≈ 20—that is, roughly 100–200 million years after the Big Bang—with the water concentrated precisely in the dense cloud cores where stars and planets form. According to the study these early remnants reached water mass fractions only a factor of a few below those of the Solar System today, though the total masses were modest and much of the water was likely later dissociated by radiation. This is a single, model-based study and should be treated as a plausible frontier result rather than settled fact.

In the present universe, water forms efficiently on the surfaces of cold interstellar dust grains, where hydrogen and oxygen atoms meet and combine into ice mantles at temperatures near 10 K. It is one of the most abundant molecules in the interstellar medium and has been detected at every stage of star and planet formation—in prestellar cores, protostellar envelopes, protoplanetary disks, and Solar System bodies. Crucially, in the cold, dense environments of prestellar clouds, ion-molecule and grain-surface reactions preferentially incorporate deuterium into water, driving the D/H ratio far above the primordial interstellar value of about 2 × 10⁻⁵. This deuterium enrichment is a low-temperature signature.

A landmark modeling study by L. Ilsedore Cleeves and collaborators, published in Science in 2014, exploited this fact. The team asked whether the young Solar System’s protoplanetary disk could have produced, from scratch, the deuterium enrichment observed in Solar System water. Resetting the deuterium clock to zero and allowing disk chemistry to run for a disk’s lifetime (about a million years), they found it could not: the disk was too inefficient. The implication is that a substantial fraction of Solar System water—their simulations suggested on the order of 30–50% of Earth’s ocean water and 60–100% of cometary water at the high end—was inherited as ice from the interstellar medium, chemically predating the Sun. This inference has since been strengthened observationally; a 2025 Nature Astronomy study reported enhanced doubly-deuterated water (D₂O) in the outbursting disk V883 Ori, offering direct evidence of inheritance of pristine ices from the pre-stellar phase. Some of the water in a glass, in the most literal sense, is older than the Sun.

3. The Problem: A Dry Birthplace and the Snow Line

The solar nebula—the rotating disk of gas and dust from which the Sun and planets formed—was hot near its center and cold in its outer reaches. The snow line (also called the frost line or ice line) is the heliocentric distance beyond which it was cold enough for water to condense and be retained as solid ice grains. Inside the snow line, water existed only as vapor, and solids were restricted to metals and silicate rock; outside it, water ice roughly doubled the density of condensable solids, helping to build the massive cores of the giant planets. The snow line is generally thought to have lain within or near the present asteroid belt during planet formation, at a few astronomical units (AU), and it migrated over time as the disk evolved. The dichotomy between the dry inner planets and the ice-rich outer bodies is the observable legacy of this boundary; the modern asteroid belt preserves it, with dry S-type asteroids in the inner belt and hydrated, carbon-rich C-type asteroids farther out.

Earth accreted at about 1 AU, well inside the snow line. Its building blocks should therefore have been essentially dry. This is reinforced by the fact that Earth’s bulk chemical and isotopic composition closely matches that of enstatite chondrite meteorites—oxygen-poor meteorites believed to have formed in the hot inner nebula and long considered “dry.” The puzzle, sharply stated, is this: if Earth formed from dry material inside the snow line, why is it wet? Two broad classes of answer exist. Either water was delivered late, after Earth’s main assembly, from bodies that formed beyond the snow line; or water was present, in some overlooked form, in Earth’s building blocks all along; or—most probably—some combination of both.

4. Competing and Complementary Hypotheses

4.1 Late delivery by comets

Comets, being ice-rich bodies from the cold outer Solar System, were historically the most intuitive candidate for delivering Earth’s water. The isotopic test, however, has largely demoted them. The D/H ratio of Earth’s ocean water—Vienna Standard Mean Ocean Water, VSMOW—is (1.5576 ± 0.0001) × 10⁻⁴. Measurements of the classic Oort cloud comets returned values roughly twice this. Comet 1P/Halley, measured in situ by the Giotto spacecraft’s neutral mass spectrometer, yielded D/H = (3.06 ± 0.34) × 10⁻⁴ (Eberhardt et al. 1995, Astronomy & Astrophysics). Comet Hale-Bopp (C/1995 O1) gave D/H = (3.3 ± 0.8) × 10⁻⁴ (Meier et al. 1998, Science). The mean of the early Oort cloud comet measurements was about (2.96 ± 0.25) × 10⁻⁴—about twice terrestrial, and roughly an order of magnitude above the protosolar value. Such deuterium-rich water could not have been the dominant source of Earth’s oceans.

The picture briefly grew more hopeful in 2011, when Paul Hartogh and colleagues used the Herschel Space Observatory to measure the Jupiter-family comet 103P/Hartley 2—which originated in the Kuiper Belt—and found D/H = (1.61 ± 0.24) × 10⁻⁴, indistinguishable from ocean water (Hartogh et al. 2011, Nature). This suggested that at least some comets shared Earth’s water signature and reopened the debate.

That reopening was decisively complicated by ESA’s Rosetta mission. Its ROSINA mass spectrometer measured the D/H ratio of water in the coma of the Jupiter-family comet 67P/Churyumov-Gerasimenko—also of Kuiper Belt origin—and found (5.3 ± 0.7) × 10⁻⁴, more than three times the terrestrial value and among the highest measured in any comet (Altwegg et al. 2015, Science). Because 67P shares a dynamical origin with Hartley 2 yet has utterly different water, the results demonstrated that a comet’s dynamical family does not predict its D/H, and that comets span a wide isotopic range. Subsequent Rosetta analysis confirmed the value was stable over 67P’s orbit, with a mean of 5.01 × 10⁻⁴ (Müller et al. 2022, Astronomy & Astrophysics).

The strongest quantitative constraint on the cometary contribution comes from combining D/H with noble gases. Bernard Marty and colleagues, using ROSINA measurements of argon and other volatiles at 67P and treating the comet as representative of the cometary reservoir, concluded that comets contributed no more than about 1% of Earth’s water (Marty et al. 2016, Earth and Planetary Science Letters). Intriguingly, the same work and its successors argued that comets may nonetheless have delivered a substantial fraction—about 20% (Marty et al. 2017; Rubin et al. 2018 estimated 20 ± 5%)—of Earth’s atmospheric heavy noble gases such as xenon, and potentially prebiotic organic molecules. The current consensus is thus that comets were a minor contributor to the water budget but may have been important for the atmosphere and for organics. (More recent measurements have found some comets closer to Earth’s value—e.g., the Halley-type comet 12P/Pons-Brooks at D/H = (1.71 ± 0.44) × 10⁻⁴, reported in Nature Astronomy in 2025—underscoring the isotopic diversity of the cometary population.) Nature

4.2 Delivery by carbonaceous chondrite asteroids

The leading candidate for the bulk of Earth’s surface water is carbonaceous chondrite material from the outer asteroid belt. Carbonaceous chondrites, especially the CI and CM types, are water-rich—CI chondrites contain on the order of 10–13% water by mass, bound in hydrated silicates (clays)—and their water carries a D/H signature that matches Earth’s. A comprehensive analysis by Conel Alexander and colleagues of 86 carbonaceous chondrite samples found that their inferred initial water D/H ratios span and center near the terrestrial value (bulk carbonaceous-chondrite water D/H ≈ (1.4 ± 0.1) × 10⁻⁴), about a factor of two lower than the Oort cloud comets—implying that water was delivered to Earth chiefly by bodies that formed in the asteroid belt or sunward of Jupiter, not by outer Solar System comets (Alexander et al. 2012, Science). Carbonaceous chondrites uniquely match both Earth’s D/H and its ¹⁵N/¹⁴N ratio, a dual constraint that comets fail (comets are systematically ¹⁵N-rich). The C-type asteroids spectroscopically linked to these meteorites are the presumed parent population. This is the source most cosmochemists regard as the principal supplier of Earth’s surface water.

4.3 Water inherited from enstatite-chondrite-like building blocks

A complementary and, in recent years, increasingly influential idea is that Earth’s building blocks were not as dry as assumed. Enstatite chondrites (ECs)—which match Earth’s isotopic composition for oxygen, titanium, calcium, chromium and other elements and are widely taken to represent Earth’s dominant building material—were long considered essentially anhydrous because they formed in the hot inner nebula. In a pivotal study, Laurette Piani and colleagues measured the hydrogen contents and D/H ratios of 13 enstatite chondrites and found far more hydrogen than assumed (Piani et al. 2020, Science 369, 1110–1113). They reported that these meteorites contain sufficient hydrogen to have delivered to Earth “at least three times the mass of water in its oceans,” with a bulk hydrogen isotopic composition (δD roughly −100 to −150‰, i.e. D/H near 1.3–1.4 × 10⁻⁴) close to that of Earth’s mantle, and nitrogen isotopes also matching the terrestrial mantle. The least-metamorphosed enstatite chondrites contained a water-equivalent hydrogen abundance of about 0.44 ± 0.04 wt.%.

The implication is profound: Earth may have accreted much of its water with its earliest building blocks, “wet from the start,” rather than acquiring it all by late delivery. Piani and colleagues were careful to note that the enstatite-chondrite D/H sits slightly below ocean water, so a modest additional contribution of D-richer, CI-like outer-Solar-System water is still required to match the oceans and atmosphere precisely. Their model finds Earth’s ocean D/H is reproduced by a mixture of about 95% enstatite-chondrite-type water and roughly 5% delivered by comets or water-rich asteroids. This hypothesis does not displace carbonaceous chondrites so much as share the stage with them: the modern view is a mixture, with dry-formed inner material carrying more water than expected and water-rich carbonaceous material topping up the surface reservoirs.

4.4 Ingassing of nebular hydrogen into a magma ocean

A third source is the solar nebula itself. While the nebular gas persisted, Earth’s largest planetary embryos, heated by radioactive decay and giant impacts, would have been covered by magma oceans beneath primordial hydrogen-rich atmospheres captured from the nebula. Hydrogen from such an atmosphere can dissolve (“ingas”) into molten silicate and react with metallic iron. Jun Wu and colleagues modeled this process and proposed that Earth inherited chondritic water and additionally ingassed nebular hydrogen, with iron-hydrogenation reactions sequestering hydrogen into the core while fractionating hydrogen isotopes (Wu et al. 2018, Journal of Geophysical Research: Planets). Their model requires ingassing of a small amount—typically up to about half an ocean of nebular hydrogen—supplementing seven to eight oceans from chondritic sources, with about 60% of the total hydrogen entering the core. The nebular signature is deuterium-poor (protosolar D/H ≈ 2 × 10⁻⁵), and this hypothesis is supported by the discovery of anomalously low D/H in some deep-mantle materials and by the presence of solar-type neon and helium in the deep mantle (Williams & Mukhopadhyay 2018, Nature). Nebular ingassing is generally assigned second-rank status relative to chondritic sources but has independent support from noble-gas evidence.

4.5 Solar-wind implantation on dust grains

A further isotopically light reservoir was identified in returned asteroid samples. Using atom probe tomography, Luke Daly and colleagues examined an olivine grain from the S-type asteroid Itokawa (returned by JAXA’s Hayabusa mission) and found its solar-wind-irradiated rim enriched in water and hydroxyl by about 1 mol% (Daly et al. 2021, Nature Astronomy). The mechanism is the implantation of solar-wind protons (H⁺) into silicate surfaces, where they combine with oxygen to make water; the team confirmed the process experimentally. Because solar-wind-derived water is isotopically light (low D/H), it provides a candidate for the “missing” light reservoir needed to reconcile a purely asteroidal mixture with Earth’s precise ocean value. Daly and colleagues estimated the Itokawa regolith could hold on the order of 20 liters of such water per cubic meter, and that fine dust irradiated in this way—drifting into the accreting Earth—could have contributed a meaningful, isotopically distinct increment. As one author put it, building a habitable world took “a little bit of everything.”

4.6 The late veneer and highly siderophile elements

The concept of a late veneer—a final addition of chondritic material after Earth’s core had finished forming—rests on the highly siderophile elements (HSEs: the platinum-group elements, plus Re and Au). These elements have such a strong affinity for iron metal that core formation should have stripped them almost entirely from the mantle. Yet they persist in the mantle at levels far above expectation, and in roughly chondritic relative proportions. The standard interpretation is that about 0.5% of Earth’s mass was added as chondritic material after core formation, delivering these HSEs to the mantle. This late veneer is sometimes invoked as a vehicle for volatile delivery, including water.

The evidence weighs against the late veneer being the dominant source of Earth’s water, however. Morbidelli and Wood (2015) and others argue that mass balance and the H₂O/Xe ratio of the mantle make a dominant late-veneer water contribution unlikely: had objects with cometary or chondritic noble-gas contents delivered the oceans as a late veneer, they would have delivered too much xenon, with the wrong isotopic composition, relative to what Earth’s atmosphere contains. The late veneer is real and important for HSEs and probably contributed some volatiles, but most of Earth’s water was likely acquired during, not after, its main accretion. (A 2026 analysis further argues that HSEs delivered by impactors larger than about 1 km should sink to the core, complicating the mass estimates and leaving the total late-accreted mass poorly constrained.)

4.7 Dynamical delivery: the Grand Tack and Nice models

The hypotheses above address the source of the water; dynamical models address the mechanism that brought outer-Solar-System, water-rich material into the inner Solar System. The Grand Tack model, developed by Kevin Walsh, Alessandro Morbidelli, Sean Raymond and Avi Mandell, proposes that Jupiter migrated inward through the disk to about 1.5 AU before Saturn’s growth reversed the migration, sending both giants back outward (Walsh et al. 2011, Nature). This inward-then-outward “tack” simultaneously explains the small mass of Mars (by truncating the inner disk) and scatters the asteroid belt into two populations: dry S-types from the inner regions and water-rich C-types implanted from between and beyond the giant planets. Critically, the same scattering flings water-rich planetesimals inward across Earth’s feeding zone. Grand Tack simulations find these polluting bodies outnumber those trapped in the belt by roughly an order of magnitude and naturally supply Earth with about its current water budget, with C-type compositions consistent with Earth’s water signature.

The related Nice model addresses a later dynamical instability among the giant planets, associated with the reconfiguration of the outer Solar System and, in some formulations, a late bombardment of the inner planets. Together these models supply the “delivery trucks”: a physical means to transport carbonaceous, water-bearing material from beyond the snow line into the terrestrial-planet region. Alternative, simpler models (e.g., inward scattering during Jupiter and Saturn’s gas accretion, without a full tack, as in Raymond and Izidoro 2017) achieve similar water delivery, so the specific dynamical pathway remains debated even as the general picture of inward scattering is robust.

5. Recent Sample-Return Evidence

The last decade has transformed the debate by bringing pristine extraterrestrial material into terrestrial laboratories.

Ryugu (Hayabusa2, JAXA). JAXA’s Hayabusa2 returned 5.424 ± 0.217 grams of material from the C-type asteroid Ryugu, collected at two touchdown sites, in December 2020 (Yada et al. 2022, Nature Astronomy). Analyses established that Ryugu is a near-perfect match to CI (Ivuna-type) carbonaceous chondrites in mineralogy, chemistry and isotopic composition—indeed a more pristine sample of CI-like material than any meteorite, uncontaminated by terrestrial weathering (Nakamura et al. 2022, Science; Yokoyama et al. 2022, Science). Ryugu underwent extensive aqueous alteration on its parent body, producing hydrous silicates (serpentine, saponite), carbonates, magnetite and sulfides, and it is rich in carbon (~4%) and organic molecules, including amino acids. Oxygen-isotope evidence from Ryugu was interpreted as supporting early water delivery to Earth by CI chondrites (Nakamura/curation teams, Nature Astronomy 2022). Direct measurement of the D/H ratio of Ryugu’s hydrous minerals gave (165 ± 19) × 10⁻⁶, similar to CI chondrites and close to terrestrial water; on this basis the CI-like contribution to Earth’s surface water was estimated at only about 3%, and the water was inferred to derive ultimately from interstellar ice precursors that partly re-equilibrated with nebular hydrogen (Piani et al. 2023, Astrophysical Journal Letters). Ryugu’s water is slightly deuterium-rich relative to Earth, consistent with CI-like material being one contributor among several rather than a sole source.

Bennu (OSIRIS-REx, NASA). NASA’s OSIRIS-REx returned 121.6 grams of regolith from the C-type asteroid Bennu in September 2023—the largest asteroid sample ever collected in space, over twice the mission’s 60-gram requirement, and, per mission scientists, containing the highest abundance of carbon yet measured in an extraterrestrial sample. Initial characterization found the sample dominated (about 80% by volume) by hydrated phyllosilicates—magnesium serpentine and saponite—together with magnetite, carbonates, sulfides, and abundant organic compounds and nitrogen (Lauretta, Connolly et al. 2024, Meteoritics & Planetary Science). A striking discovery was magnesium-sodium phosphate and a suite of sodium-bearing salts (carbonates, sulfates, chlorides, fluorides) recording the evaporation of an ancient brine, implying that Bennu’s parent body hosted liquid water and may have been a fragment of a small, primitive ocean world (McCoy et al. 2024, Nature). Later work reported abundant ammonia and nitrogen-rich organics, and the hydrogen-isotope composition of Bennu’s organics indicated its parent body was not fully hydrated. Together, Ryugu and Bennu provide direct, ground-truth confirmation that water-rich, organic-bearing carbonaceous asteroids of the kind invoked to supply Earth’s water genuinely exist and carry near-terrestrial isotopic signatures.

Itokawa (Hayabusa, JAXA). As discussed above (§4.5), the S-type asteroid Itokawa’s samples revealed solar-wind-produced water in irradiated grain rims, adding an isotopically light source to the inventory.

Lunar samples. Analyses of hydrogen in lunar volcanic glasses and apatite have found water with D/H ratios overlapping those of carbonaceous chondrites, suggesting that the Moon—and by inference the Earth-Moon system after the giant impact—inherited water of carbonaceous-chondritic character present before the impact, reinforcing the view that water was part of the inner Solar System’s inventory early rather than delivered entirely afterward.

6. Timing: How Early Was Earth Wet?

The delivery question is bounded by a timing constraint: liquid water existed on Earth’s surface astonishingly early. The evidence comes from the Jack Hills zircons of Western Australia—detrital zircon crystals that are the oldest known terrestrial materials, the oldest dated at 4,404 ± 8 million years (Wilde et al. 2001, Nature). The oxygen-isotope composition (elevated δ¹⁸O) of these grains indicates that their parent magmas incorporated material that had interacted with liquid water at or near Earth’s surface, implying that liquid water, and possibly oceans, existed by about 4.3–4.4 billion years ago—within roughly 150 million years of Earth’s formation (Wilde et al. 2001; Mojzsis et al. 2001, Nature; Valley et al. 2002, Geology). A 2014 atom-probe tomography study confirmed the age of the oldest grain at 4.374 ± 0.006 billion years and supported the “cool early Earth” model (Valley et al. 2014, Nature Geoscience). The heavy-δ¹⁸O interpretation has been contested by some who attribute it to later alteration (e.g., Hoskin), so a degree of caution is warranted; but the weight of evidence favors a wet Hadean surface. Whatever delivered Earth’s water did so early, and its oceans have persisted, with interruptions, ever since.

7. Deep-Earth Water: The Hydrous Transition Zone

Water is not confined to Earth’s surface. A dramatic line of evidence for a wet interior came from a diamond from Juína, Brazil, within which Graham Pearson and colleagues identified an inclusion of ringwoodite—a high-pressure polymorph of olivine previously known on Earth only in meteorites and predicted to be a major constituent of the mantle transition zone between 410 and 660 km depth (Pearson et al. 2014, Nature). The inclusion was hydrous, containing about 1% water by weight (up to ~1.5%), providing the first direct terrestrial evidence that the transition zone is, at least locally, water-bearing. Because ringwoodite is so voluminous a phase at those depths, even ~1% water implies an enormous reservoir; estimates suggest the transition zone could hold as much water as, or more than, all the surface oceans combined—potentially close to three oceans’ worth if the hydration is widespread. This deep water bears directly on the origin question: it demonstrates that Earth’s water is not merely a surface veneer but is distributed through the planet’s interior, consistent with water having been incorporated during accretion and differentiation as well as delivered to the surface. As Pearson noted, water in the interior fundamentally shapes how the planet works—its magmatism, volcanism and plate tectonics.

8. Synthesis: The Evidence-Weighted Picture

No single source accounts for Earth’s water; the evidence points to a combination, with the contributions weighted roughly as follows.

The dominant reservoirs are chondritic. Earth’s dry-formed building blocks—enstatite-chondrite-like material—appear to have carried substantially more hydrogen than long assumed (Piani et al. 2020), plausibly supplying much of the planet’s water during main accretion and matching Earth’s mantle D/H and ¹⁵N/¹⁴N. Water-rich carbonaceous chondrite material from the outer belt (CI/CM types, sampled directly at Ryugu and Bennu) supplied additional water whose D/H uniquely matches both Earth’s hydrogen and nitrogen isotopes (Alexander et al. 2012), delivered inward by the dynamical scattering that models like the Grand Tack describe (Walsh et al. 2011).

Comets were a minor contributor to water—on the order of 1% or less—because their water is generally too deuterium-rich (Altwegg et al. 2015; Marty et al. 2016), notwithstanding the isotopic diversity revealed by Hartley 2 and 12P. Comets may, however, have contributed significantly to atmospheric noble gases and to the organic inventory.

Nebular ingassing of deuterium-poor hydrogen into early magma oceans (Wu et al. 2018) and solar-wind implantation on dust grains (Daly et al. 2021) are plausible minor supplements, both attractive because they supply the isotopically light hydrogen needed to fine-tune a mixture of heavier chondritic sources to Earth’s precise ocean value, and both supported by independent deep-mantle noble-gas and sample evidence.

Open questions remain. The exact proportions are unsettled, and depend on poorly constrained parameters such as the total water content of the deep mantle and core. The behavior of D/H as a function of heliocentric distance in the disk is model-dependent. The role of hydrogen sequestered in the core—potentially a vast hidden reservoir—is only beginning to be quantified. And the tension between the late-veneer HSE budget and the noble-gas water budget is not fully resolved. What is secure is the shape of the answer: Earth’s water is overwhelmingly of chondritic parentage, acquired largely during accretion and supplemented by inward-scattered carbonaceous material, with comets a garnish and nebular and solar-wind hydrogen possible seasoning.

9. Epilogue: The Waters Above and the Waters Within

There is a particular vertigo in learning that the water in a cupped hand is older than the sunlight that glints on it. The hydrogen in it was struck into being in the first three minutes of the universe, when there were as yet no stars, no planets, no eyes to see. The oxygen bonded to it was cooked in the heart of some massive, anonymous star that lived and died before the Sun was lit, scattering its ash across the dark so that, ages later, on the surface of a frozen grain drifting in a cold cloud, the two could meet and become water. That water rode into the collapsing nebula, survived the violence of a star’s birth, and was folded into the body of a forming world—some of it, perhaps, arriving in the dark hold of a carbonaceous asteroid flung sunward by a migrating Jupiter, some of it wrung from the very rock Earth was built from, some of it breathed in from the nebula and buried in the deep. It has been in the mantle and in the transition zone, locked in diamond, and in the oceans that were already lapping some Hadean shore more than four billion years ago.

To trace this provenance is not to diminish water into a mere logistics problem of the early Solar System. It is, if anything, to deepen the sense of water as the precondition of everything that matters to us. Every living thing on this planet is, chemically, an elaboration built around water; life as we know it is water’s way of becoming complicated. It is difficult to contemplate this long descent—from Big Bang to supernova to interstellar ice to asteroid to ocean to cell—without a kind of reverence, and it would be a poverty of imagination to insist that reverence and rigorous science must occupy separate rooms.

It is in that spirit, and not as any claim of proof, that one may notice a resonance in older ways of speaking. The Qur’an declares that God “made from water every living thing” (Q 21:30), and that “God created every moving creature from water” (Q 24:45)—verses that a modern reader, standing at the end of this long causal chain from the primordial fire to the living cell, may find quietly anticipatory. This is offered as resonance, not as concordist demonstration: the ancient text is not a cosmochemistry paper in disguise, and the science stands on its own evidence. But the human intuition, expressed across many traditions, that water is the wellspring and substance of life turns out to be, in a strict and astonishing sense, true. The waters that make Earth a living world were gathered from the whole history of the cosmos to arrive here. That we can now read that history in the deuterium of a raindrop is one of the quieter marvels of our age—and that the raindrop remains, for all our reading of it, the thing without which none of the reading would be possible.

References

• Alexander, C. M. O’D., Bowden, R., Fogel, M. L., Howard, K. T., Herd, C. D. K. & Nittler, L. R. (2012). The provenances of asteroids, and their contributions to the volatile inventories of the terrestrial planets. Science 337(6095), 721–723.
• Altwegg, K., Balsiger, H., Bar-Nun, A., et al. (2015). 67P/Churyumov-Gerasimenko, a Jupiter family comet with a high D/H ratio. Science 347(6220), 1261952.
• Cleeves, L. I., Bergin, E. A., Alexander, C. M. O’D., Du, F., Graninger, D., Öberg, K. I. & Harries, T. J. (2014). The ancient heritage of water ice in the Solar System. Science 345(6204), 1590–1593.
• Daly, L., Lee, M. R., Hallis, L. J., et al. (2021). Solar wind contributions to Earth’s oceans. Nature Astronomy 5(12), 1275–1285.
• Eberhardt, P., Reber, M., Krankowsky, D. & Hodges, R. R. (1995). The D/H and ¹⁸O/¹⁶O ratios in water from comet P/Halley. Astronomy & Astrophysics 302, 301–316.
• Hartogh, P., Lis, D. C., Bockelée-Morvan, D., et al. (2011). Ocean-like water in the Jupiter-family comet 103P/Hartley 2. Nature 478, 218–220.
• Lauretta, D. S., Connolly, H. C., et al. (2024). Asteroid (101955) Bennu in the laboratory: properties of the sample collected by OSIRIS-REx. Meteoritics & Planetary Science 59, 2453–2486.
• Marty, B., Avice, G., Sano, Y., et al. (2016). Origins of volatile elements (H, C, N, noble gases) on Earth and Mars in light of recent results from the ROSETTA cometary mission. Earth and Planetary Science Letters 441, 91–102.
• McCoy, T. J., et al. (2024). An evaporite sequence from ancient brine recorded in Bennu samples. Nature 637, 1072–1077.
• Meier, R., Owen, T. C., Matthews, H. E., et al. (1998). A determination of the HDO/H₂O ratio in comet C/1995 O1 (Hale-Bopp). Science 279(5352), 842–844.
• Mojzsis, S. J., Harrison, T. M. & Pidgeon, R. T. (2001). Oxygen-isotope evidence from ancient zircons for liquid water at the Earth’s surface 4,300 Myr ago. Nature 409, 178–181.
• Morbidelli, A. & Wood, B. J. (2015). Late accretion and the late veneer. In The Early Earth: Accretion and Differentiation, AGU Geophysical Monograph 212, 71–82.
• Müller, D. R., Altwegg, K., Berthelier, J. J., et al. (2022). High D/H ratios in water and alkanes in comet 67P/Churyumov-Gerasimenko measured with Rosetta/ROSINA DFMS. Astronomy & Astrophysics 662, A69.
• Nakamura, T., et al. (2022). Formation and evolution of carbonaceous asteroid Ryugu: direct evidence from returned samples. Science 379, eabn8671.
• Pearson, D. G., Brenker, F. E., Nestola, F., et al. (2014). Hydrous mantle transition zone indicated by ringwoodite included within diamond. Nature 507(7491), 221–224.
• Piani, L., Marrocchi, Y., Rigaudier, T., Vacher, L. G., Thomassin, D. & Marty, B. (2020). Earth’s water may have been inherited from material similar to enstatite chondrite meteorites. Science 369(6507), 1110–1113.
• Piani, L., Nagashima, K., Kawasaki, N., et al. (2023). Hydrogen isotopic composition of hydrous minerals in asteroid Ryugu. Astrophysical Journal Letters 946, L43.
• Raymond, S. N. & Izidoro, A. (2017). Origin of water in the inner Solar System: planetesimals scattered inward during Jupiter and Saturn’s rapid gas accretion. Icarus 297, 134–148.
• Rubin, M., Altwegg, K., Balsiger, H., et al. (2018). Krypton isotopes and noble gas abundances in the coma of comet 67P/Churyumov-Gerasimenko. Science Advances 4(7), eaar6297.
• Valley, J. W., Peck, W. H., King, E. M. & Wilde, S. A. (2002). A cool early Earth. Geology 30(4), 351–354.
• Valley, J. W., Cavosie, A. J., Ushikubo, T., et al. (2014). Hadean age for a post-magma-ocean zircon confirmed by atom-probe tomography. Nature Geoscience 7, 219–223.
• Walsh, K. J., Morbidelli, A., Raymond, S. N., O’Brien, D. P. & Mandell, A. M. (2011). A low mass for Mars from Jupiter’s early gas-driven migration. Nature 475, 206–209.
• Whalen, D. J., Latif, M. A. & Jessop, C. (2025). Abundant water from primordial supernovae at cosmic dawn. Nature Astronomy 9, 741–746.
• Wilde, S. A., Valley, J. W., Peck, W. H. & Graham, C. M. (2001). Evidence from detrital zircons for the existence of continental crust and oceans on the Earth 4.4 Gyr ago. Nature 409, 175–178.
• Williams, C. D. & Mukhopadhyay, S. (2018). Capture of nebular gases during Earth’s accretion is preserved in deep-mantle neon. Nature 565, 78–81.
• Wu, J., Desch, S. J., Schaefer, L., Elkins-Tanton, L. T., Pahlevan, K. & Buseck, P. R. (2018). Origin of Earth’s water: chondritic inheritance plus nebular ingassing and storage of hydrogen in the core. Journal of Geophysical Research: Planets 123, 2691–2712.
• Yada, T., Abe, M., Okada, T., et al. (2022). Preliminary analysis of the Hayabusa2 samples returned from C-type asteroid Ryugu. Nature Astronomy 6, 214–220.
• Yokoyama, T., et al. (2022). Samples returned from the asteroid Ryugu are similar to Ivuna-type carbonaceous meteorites. Science 379, eabn7850.