Water exists on the Moon in several forms and locations, from trace molecular films clinging to sunlit soil to buried ice deposits in permanently shadowed craters near the poles. The 2009 LCROSS mission confirmed ice in the lunar south pole’s Cabeus crater at a concentration of roughly 6% by mass in the regolith, and subsequent observations have mapped water signatures across broader areas of the polar surface. Far from the bone-dry world Apollo-era scientists described, the Moon turns out to host a subtle but dynamic water cycle, one that researchers are still working to fully characterize and that space agencies are betting on as a cornerstone of future exploration.
How We Know It Is There
The definitive confirmation came on October 9, 2009, when NASA deliberately crashed a spent Centaur rocket stage into Cabeus crater, a permanently shadowed region near the lunar south pole. A trailing spacecraft observed the debris plume with cameras and spectrometers. Near-infrared signatures of water vapor and ice, along with ultraviolet emissions from hydroxyl radicals, left little room for doubt. The instruments detected about 155 kilograms of water vapor and ice in their field of view, and from the estimated mass of excavated regolith, researchers calculated the ice concentration at the impact site to be roughly 5.6% by mass.1PubMed. Detection of water in the LCROSS ejecta plume Later modeling refined and confirmed those spectroscopic results, establishing beyond reasonable doubt that water ice sits inside permanently shadowed craters.2Icarus. Water within a permanently shadowed lunar crater: Further LCROSS modeling and analysis
More recently, NASA’s SOFIA airborne telescope detected a 6-micrometer spectral signature attributed to molecular water on sunlit portions of the lunar south polar region, suggesting that water is not confined exclusively to dark craters.3Geophysical Research Letters. Regional Map of Molecular Water at High Southern Latitudes on the Moon Using 6 μm Data From the Stratospheric Observatory for Infrared Astronomy Meanwhile, deep-learning analysis of near-infrared reflectance spectra has identified water ice in the south polar region as small crystalline particles, roughly 70 micrometers across, though amorphous ice may lurk beneath the optically sensed surface layer.
Where the Water Comes From
The Moon has no rain, no rivers, and no atmosphere to speak of. So the water had to arrive or form through processes that look nothing like Earth’s water cycle. Researchers have identified at least three major sources, each contributing in different ways and on different timescales.
Solar Wind
The Sun constantly streams charged particles outward, and among them are protons, which are simply hydrogen nuclei. When these protons slam into the oxygen-rich silicate minerals that make up lunar soil, they can bond with oxygen atoms to create hydroxyl groups and, in some cases, water molecules. Laboratory experiments simulating this process at solar-wind energies have confirmed that hydroxyl forms readily in minerals like olivine and silicon dioxide, with an initial conversion rate of implanted hydrogen into hydroxyl of about 90%.4Journal of Geophysical Research: Planets. Hydrogen implantation in silicates: The role of solar wind in SiOH bond formation on the surfaces of airless bodies in space Further experiments irradiating olivine, pyroxene, plagioclase, and volcanic glass together found that all formed hydroxyl after proton bombardment, with plagioclase capturing the most hydrogen.5Icarus. Experimental investigation of structural OH/H2O in different lunar minerals and glass via solar-wind proton implantation
Electron microscopy of interplanetary dust particles has gone a step further, revealing actual water sealed inside tiny vesicles within the radiation-damaged rims of silicate grains. This means that solar-wind weathering does not just produce hydroxyl on surfaces; it generates and traps water molecules inside mineral structures. The researchers who found this concluded that the process is ubiquitous across any airless body with silicate minerals.6PubMed Central. Detection of solar wind-produced water in irradiated rims on silicate minerals
Comet and Asteroid Impacts
The Moon has been pelted by comets and water-bearing asteroids for billions of years. Numerical simulations suggest the two types of impactors behave quite differently. Carbonaceous asteroids hitting at relatively low speeds (below about 10 km/s) can leave a significant fraction of their material in the crater without fully decomposing, meaning hydrated minerals can survive and deposit both free and chemically bound water.7Planetary and Space Science. Water delivery to the Moon by asteroidal and cometary impacts Cometary water, by contrast, tends to vaporize even at low impact speeds and disperses broadly across the surface. Models tracking how much of this vapor stays gravitationally bound to the Moon estimate that about 6.5% of the total impactor mass flux is retained. After accounting for migration and stability of ice at the poles, one study estimated that comets alone may have delivered between 130 million and 4.3 billion metric tons of water that remains stable at the poles over a billion-year window.8Icarus. Volatile retention from cometary impacts on the Moon
Ancient Volcanic Outgassing
Billions of years ago, when the Moon was volcanically active, eruptions would have released water vapor from the interior. Modeling suggests that roughly 41% of the total water mass erupted during this period could have condensed as polar ice. Each major eruption would have generated a short-lived collisional atmosphere lasting around 2,500 years, long enough for water vapor to migrate poleward and freeze. Over the Moon’s entire volcanic history, this mechanism may have sequestered as much as 8.2 trillion kilograms of water ice at the poles.9The Planetary Science Journal. Polar Ice Accumulation from Volcanically Induced Transient Atmospheres on the Moon Those transient atmospheres have long since dissipated, but the ice they deposited could remain locked in permanently shadowed craters to this day.10Icarus. Lifetime of a transient atmosphere produced by lunar volcanism
How the Water Moves and Where It Ends Up
Lunar water is not static. On sunlit portions of the surface, ultraviolet data from the Lunar Reconnaissance Orbiter show that trace amounts of water in the topmost regolith vary with both terrain type and local time. As temperatures rise during the lunar day, water molecules desorb from soil grains and hop across the surface ballistically until they either stick to a cooler spot or escape to space. When the Moon passes through Earth’s magnetotail, shielded from the solar wind’s supply of fresh protons, the surface water level does not drop, which tells researchers the water is migrating rather than being continuously manufactured and destroyed in place.11Geophysical Research Letters. Diurnally Migrating Lunar Water: Evidence From Ultraviolet Data
This hopping process gradually funnels water molecules toward the poles, where permanently shadowed regions act as cold traps. Temperatures in these shadows can fall below 100 Kelvin, cold enough that water molecules stick essentially forever. Modeling of the layered deposits inside south polar craters suggests that the oldest cold traps (those older than four billion years) hold two distinct zones: deep, buried ice-rich gigaton deposits, and a younger, more gardened mantle closer to the surface that has been stirred by micrometeorite impacts over time.12Geophysical Research Letters. Stratigraphy of Ice and Ejecta Deposits at the Lunar Poles Among south polar craters, Amundsen, Cabeus, and Cabeus B appear most likely to retain large ice deposits at depths up to 100 meters, shallow enough to potentially be detected by ground-penetrating radar.13Journal of Geophysical Research: Planets. Buried Ice Deposits in Lunar Polar Cold Traps Were Disrupted by Ballistic Sedimentation
Besides the large shadowed craters, the Moon hosts countless small-scale permanently shadowed pockets, sometimes called micro cold traps, created by boulders or small crater walls. These make up an estimated 10 to 20% of the total cold-trap area and are more accessible to future landers, though their permanently shadowed volume is much shallower, on average only about 0.5 to 1% as deep as the topographic features that harbor them.
Micrometeoroids and the Double-Edged Sword of Impacts
The lunar surface is constantly bombarded by tiny, fast-moving particles of interplanetary dust. These micrometeoroids play a complicated role in the water story. On the one hand, when a micrometeoroid hits hydroxyl-bearing regolith at speeds of 8 to 20 km/s, it can generate new water molecules, with production ranging from about 44% to 275% of the water that existed at the impact site beforehand. On the other hand, higher-speed impacts also blast water molecules off the surface and into space. The net gain in local water content from a single impact can be as low as 5% at high velocities or as high as 73% at lower ones.14Geophysical Research Letters. Molecular Dynamics Simulations of Water Formation and Retention by Micrometeoroid Impact on Lunar Surface
How hydrogen is distributed within the soil matters a great deal for this outcome. Molecular dynamics simulations show that when hydrogen is spread throughout the depth of the regolith rather than concentrated near the surface, nearly all the water produced during an impact stays in the substrate. But when hydrogen is near-surface, impacts eject much of the water into the exosphere.15Journal of Geophysical Research: Planets. Effect of Solar Wind and Micrometeoroid Impact on the Lunar Water Cycle: A Molecular Dynamics Study In other words, micrometeoroids simultaneously create and destroy lunar water, and the balance depends on conditions that vary from spot to spot.
Extracting Water for Human Use
If people are going to stay on the Moon for more than brief visits, local water is a game-changer. Hauling water from Earth costs tens of thousands of dollars per kilogram in launch mass. Water sourced on the Moon could be used for drinking and life support, split into hydrogen and oxygen for rocket propellant, or used in manufacturing processes. The challenge is getting it out of the regolith.
The most straightforward approach is thermal extraction: heat the icy soil and capture the escaping vapor. This can be done with relatively simple hardware, essentially an oven and a cold plate.16Planetary and Space Science. Thermal extraction of water ice from the lunar surface II – vapor yields for an improved regolith model A more energy-efficient alternative being tested uses microwave heating at 2.45 GHz to extract water from icy regolith starting at cryogenic temperatures around minus 80°C. In laboratory demonstrations, an integrated microwave system successfully heated simulated icy regolith, transmitted water vapor through a closed flow path, and collected liquid water in a cold trap.17PubMed Central. Massive Water Production from Cryogenic Icy Lunar Regolith by a Microwave Heating Method
Once extracted, the raw water needs purification. Lunar ice will not be pristine; volatiles like ammonia, sulfur compounds, and other contaminants co-deposited over billions of years will come along for the ride. Paragon Space Development Corporation has been developing a purification system that uses a cold trap as a freeze-distillation step, selectively depositing water ice from a vapor stream while rejecting contaminant volatiles. Testing showed that contaminant concentrations in the collected ice fell to levels consistent with theoretical predictions, demonstrating the concept’s viability.18International Conference on Environmental Systems. Experimental Proof of Concept of a Cold Trap as a Purification Step for Lunar Water Processing
There is also the question of how hard the icy regolith itself will be to dig. Laboratory testing of simulated icy lunar regolith at temperatures from minus 240°C to minus 10°C reveals that water content, dry density, and temperature all dramatically affect its compressive strength. Below about minus 180°C, the material’s strength becomes less sensitive to further cooling, but it is already extremely hard by that point. The water content matters too: samples are strongest when fully saturated.19Advances in Space Research. Analysis and prediction of uniaxial compressive strength of icy lunar regolith under extreme temperature Any mining operation at the poles will need equipment designed to cut through material that, at the temperatures found in permanently shadowed craters, may behave more like concrete than loose sand.
The Economics of Lunar Propellant
The most commercially ambitious vision for lunar water is splitting it into hydrogen and oxygen to make rocket fuel. A depot at the lunar south pole or in orbit could refuel spacecraft headed to Mars or deeper into the solar system, drastically reducing the mass that must be launched from Earth’s deep gravity well. Economic modeling suggests that the profitability of lunar propellant is physically possible and economically plausible, but it hinges on hardware performance parameters that still need real-world demonstration, particularly the energy efficiency of extraction and the throughput rate of processing equipment.20Acta Astronautica. Economics of in-space industry and competitiveness of lunar-derived rocket propellant The business case, in other words, exists on paper but has yet to survive contact with actual lunar conditions.
Spacecraft Contamination Is Already a Concern
There is an uncomfortable irony in the race to study and use lunar water: the very spacecraft sent to explore it could contaminate the deposits before we understand them. Simulations of spacecraft exhaust show that the water vapor component is globally redistributed across the Moon, with a significant fraction reaching permanently shadowed polar cold traps, where it can remain stable on geological timescales. The contamination scales with the mass of exhaust and the proximity of the landing site to the poles.21PubMed Central. The Evolution of a Spacecraft-Generated Lunar Exosphere
A concrete analysis of what a single SpaceX Starship landing near the south pole might mean puts this in stark terms. One study estimated that a Starship landing at 89° south latitude could deliver about 16 metric tons of water to the permanently shadowed regions, roughly 27% of the total mass of existing surficial icy regolith observed in those regions by current instruments. If a south polar base required one such landing per year, after just four years the accumulated exhaust-derived water would roughly equal the total intrinsic surface ice mass, making it very difficult to distinguish ancient lunar water from spacecraft byproducts.22The Planetary Science Journal. Possible Anthropogenic Contributions to the LAMP-observed Surficial Icy Regolith within Lunar Polar Craters: A Comparison of Apollo and Starship Landings Any artificially released volatiles migrating to polar cold traps could cause irreversible contamination for future scientific exploration.23Journal of Geophysical Research: Planets. Rocket Exhaust Derived Surface Water Migration at Lunar Landing Sites
This is not a hypothetical future problem. Missions are already being planned and flown to the polar regions, and the window to characterize pristine lunar ice may be closing.
How Old Is the Ice, Really?
For decades, the default assumption was that polar ice deposits are ancient, accumulated over billions of years in cold traps that have been stable since the Moon’s early history. That assumption carried scientific weight: if the ice is old, it could preserve a record of the inner solar system’s volatile history, essentially a frozen archive of what was flying around during the period of heavy bombardment. Recent modeling of the Moon’s orbital evolution, however, suggests the ice archives may be surprisingly young, fundamentally altering the search for volatile history in polar deposits. If the Moon’s axial tilt has shifted significantly over time, what we think of as permanent shadows may not have been shadowed for as long as previously believed, and the ice inside them may reflect more recent delivery events rather than a deep-time record.
Who Gets to Use It
As multiple nations and private companies target the lunar south pole, the legal framework for extracting resources in space remains ambiguous. The 1967 Outer Space Treaty prohibits national appropriation of celestial bodies but says little about mining their resources. The Artemis Accords, introduced by the United States in 2020 and now signed by dozens of countries, propose “safety zones” around lunar operations to prevent interference between missions and to facilitate resource extraction.24Space Policy. Artemis Accords: Are Safety Zones Practical for Long Term Commercial Lunar Resource Utilisation? These zones are controversial. Critics argue that establishing exclusive operational areas around extraction sites could effectively amount to territorial claims in all but name, violating the non-appropriation principle. Successfully implementing safety zones without crossing that line is considered essential to making the Accords’ broader resource framework work.25Journal of Space Safety Engineering. Implementing safety zones for lunar activities under the Artemis Accords
China and Russia, pursuing their own joint lunar base plans, are not signatories to the Accords. That means the most ice-rich region of the Moon could become the site of overlapping, possibly conflicting resource claims governed by different legal interpretations, a scenario with no precedent in space exploration and limited mechanisms for resolution.