Lunar Water Ice: From LCROSS Confirmation to Artemis ISRU

In the permanently shadowed craters near the lunar poles, temperatures fall to -230°C — the coldest natural environment in the solar system, colder than Pluto’s surface during its aphelion. These craters have not seen sunlight in billions of years. The solar wind cannot reach them, ultraviolet radiation cannot penetrate them, and the thermal environment is as stable as any natural feature in the solar system.

They are ice traps. Any volatile compound — water, carbon dioxide, ammonia, methane — that wanders into a permanently shadowed region is captured and preserved indefinitely. Over billions of years, a slow supply of water from cometary and asteroidal impacts, from solar wind proton reactions with oxygen-bearing minerals, and from volcanic outgassing has accumulated in these cold traps in an unknown but potentially substantial quantity.

The presence of water at the lunar poles has been suspected since the 1960s, computationally modelled since the 1990s, and definitively confirmed since 2009. Its strategic implications are still being worked out — but they are significant enough that water ice is increasingly the primary driver of lunar polar site selection for every major space agency with a surface exploration programme.

Key parameters

Parameter	Value
LCROSS ejecta plume water detected	~155 kg
Ice concentration (PSR estimate)	5–10% by mass (upper layer)
H₂O electrolysis energy requirement	237 kJ/mol
LOX/LH₂ from 1 tonne ice	~889 kg O₂ + 111 kg H₂
Permanently shadowed region temp	40–100 K

LCROSS: The Impactor That Confirmed Ice

The Lunar Crater Observation and Sensing Satellite (LCROSS) mission was designed by a small team at NASA Ames Research Center to answer the water question directly and unambiguously. Its approach was characteristically direct: throw something large at the Moon, look at what comes up.

On 9 October 2009, the spent upper stage of the Atlas V that had launched the LCROSS mission — the Centaur, massing approximately 2,300 kg — was guided into a controlled impact at 2.5 km/s into Cabeus crater, a permanently shadowed crater near the lunar south pole. The impact excavated approximately 1,500 cubic metres of lunar subsurface material and ejected a plume of debris to approximately 10 km altitude.

LCROSS, trailing 370 km behind the Centaur, flew through the plume, measuring its composition spectroscopically in the 4 minutes before it impacted the same crater. Nine scientific instruments documented what the Centaur excavated.

The results, published in Science in November 2009 by Anthony Colaprete and colleagues, were unambiguous: water ice was detected at approximately 5.6 ± 2.9% by mass in the excavated material, along with carbon monoxide, hydrogen sulphide, ammonia, ethylene, and a suite of other volatile compounds. Cabeus crater contained water. The permanently shadowed regions were volatile reservoirs.

Earlier and subsequent confirmation came from multiple instruments:

Chandrayaan-1’s Moon Mineralogy Mapper (M³, 2009): Detected hydroxyl (OH) and water (H₂O) absorption signatures in reflectance spectra of the sunlit polar regions — suggesting active production of surface OH by solar wind proton interaction with silicate oxygen, and possibly surface ice.

LRO/LAMP (2010): UV observations of permanently shadowed regions by the Lyman Alpha Mapping Project instrument detected the UV-dark signature of water ice at the surface in multiple polar craters.

LRO/Diviner (2009–present): Mapped surface temperatures in permanently shadowed regions to below -250°C, confirming the thermodynamic stability of water ice in those locations.

SOFIA airborne observatory (2020): Detected a water absorption feature at 6 μm in sunlit soil at 12°S latitude — raising the possibility that water exists not just in polar cold traps but in microscopic glass beads formed by micrometeorite impacts across the lunar surface, though in much smaller quantities (100–400 parts per million).

The ISRU Case: Why Ice Changes Everything

ISRU (In-Situ Resource Utilisation) is the use of locally available materials to reduce the mass that must be launched from Earth. For a lunar surface programme, water is the most valuable ISRU resource by a wide margin — not primarily for drinking or radiation shielding, but for propellant.

Water (H₂O) can be electrolysed into hydrogen (H₂) and oxygen (O₂), both of which are storable cryogenically and both of which are propellant components for liquid oxygen/liquid hydrogen engines — the highest-performance chemical propellant combination available. The Isp of LH₂/LOX is approximately 450 seconds (RL-10, SSME), compared to 340–360 s for hypergolic propellants.

The depot economics: Launching propellant from Earth to low lunar orbit costs approximately $25,000–50,000 per kilogram at current prices using existing launch vehicles (SLS at ~$10,000/kg to LEO, plus transfer costs; commercial heavy lift at lower per-kg cost but still adding up). If lunar ice can be extracted, processed, and stored as liquid H₂ and LOX at a propellant depot in lunar orbit, the economics of the entire cislunar economy change. A Lunar Gateway refuelling station fed by surface-extracted ice reduces the mass that each mission must launch from Earth.

This is why NASA’s Artemis programme explicitly identifies the lunar south pole as its landing target — specifically the Shackleton-de Gerlache ridge, which offers near-continuous solar illumination (important for power) within walking or driving distance of the permanently shadowed regions where ice exists. The 2024 Artemis III mission — the first crewed lunar landing since Apollo 17 — targeted this region.

What We Don’t Know About the Ice Deposits

The confirmation that ice exists at the lunar poles is not the same as characterising the ice well enough to mine it.

Depth: The LCROSS impact excavated to approximately 2–3 metres depth. The spectroscopic signatures indicate ice at that depth, but the total ice thickness and depth profile is unknown for any crater. Is there a 1 cm layer at 50 cm depth, or a 100 m deposit? This is a critical uncertainty for mining feasibility.

Distribution: Ice appears in discrete, cold-trapped deposits rather than being uniformly distributed across the permanently shadowed terrain. High-resolution radar mapping by LRO’s Mini-RF instrument suggests localised ice in specific cold pockets rather than a uniform layer.

Form: Is the ice crystalline or amorphous? Pure water ice, or ice mixed with regolith, carbon compounds, and other volatiles? Dirty ice changes the thermal properties, the electrolysis purity requirements, and the mechanical extraction challenges.

Accessibility: The permanently shadowed regions that contain ice are, by definition, never illuminated by sunlight. Operating rovers, drilling equipment, and processing plants in permanent darkness with no solar power, while communicating via line-of-sight relays because the crater floors are below the horizon, is a significant engineering challenge.

NASA’s MOXIE experiment aboard Perseverance demonstrated in-situ oxygen production from the Martian CO₂ atmosphere — a proof-of-concept for ISRU at planetary scale. Analogous lunar ice extraction has not been demonstrated. The NASA Commercial Lunar Payload Services (CLPS) programme includes several contracts targeting ice characterisation missions; the PRIME-1 drill demonstration on a lunar lander was intended to directly test extraction at depth.

Geopolitical Dimensions

The strategic value of lunar ice has not been lost on national space programmes. The Artemis Accords — bilateral agreements between the US and partner nations (as of 2025: 43 signatories) — establish principles for resource extraction, safety zones, and commercial operations on the Moon. China and Russia have not signed.

China’s Chang’e programme has targeted the lunar south pole for its Chang’e 7 mission (targeting 2026) and the International Lunar Research Station (ILRS) programme, a Chinese-led alternative to Artemis with Russian participation. The ILRS targets the Shackleton area — the same region as Artemis.

The Outer Space Treaty of 1967 prohibits national appropriation of celestial bodies, but is silent on resource extraction — a gap that the US addressed via the US Commercial Space Launch Competitiveness Act (2015) and that subsequent Artemis Accords signatories accept. Whether international law provides a durable framework for managing competing resource claims on the lunar south pole is an open question.

Water ice on the Moon is not just a scientific curiosity. It is potentially a strategic resource, a propellant depot, and the enabling infrastructure for a cislunar economy. How it is extracted, by whom, under what legal framework, and whether the deposits are large enough to support the aspirations placed on them — these are among the most consequential questions facing space policy in the next two decades.

For the broader challenge of landing at the lunar south pole with sufficient precision, the entry and descent challenges described in Mars EDL are instructive analogues for the precision landing problem on the Moon.