How to Mine the Moon: What ISRU Can Actually Do

Overview

Every kilogram of anything currently on the Moon was launched from Earth at enormous cost. In-situ resource utilization is the proposal to stop doing that — to make propellant, breathable oxygen and water out of material already present on the lunar surface.

The case rests on one measurement. When NASA deliberately crashed a spent rocket stage into Cabeus crater in 2009, the LCROSS mission analysed the resulting debris plume and found water ice at roughly 5.6% of regolith mass, with an uncertainty of plus or minus 2.9%.

Nearly two decades later, that remains the strongest direct measurement available, because no mission has yet successfully drilled lunar ice. The first attempt, in March 2025, ended with the lander on its side and its batteries flat before the drill could operate.

There is a real distinction inside ISRU that usually gets flattened. Extracting oxygen from ordinary regolith is a mature, nearly ready technology. Extracting water from polar ice is neither, and conflating the two makes lunar resource extraction sound far closer than it is.

What LCROSS Actually Measured in 2009

The LCROSS experiment was blunt by design. NASA steered the spent upper stage of its launch vehicle into Cabeus, a permanently shadowed crater near the lunar south pole, and flew a shepherding spacecraft through the debris plume to analyse what came up.

Direct measurement of that plume put water ice at 5.6% of regolith mass, plus or minus 2.9%. Subsequent modelling produced a range depending on assumptions about how densely packed the regolith is — around 8.2% by weight at a density of 1.5 g/cm³, dropping to roughly 4.3% at a density of 3.0.

That spread is not a flaw in the work; it is an honest reflection of inferring composition from an impact plume. But it means the foundational number for lunar water is a single site, measured indirectly, with real uncertainty attached. Everything downstream inherits it.

Why 5% Is Both Encouraging and Marginal

Five percent by mass sounds low and, for a space resource, is genuinely promising. It is also demanding once translated into machinery.

At that concentration, producing one tonne of water requires processing roughly twenty tonnes of regolith. Every tonne has to be excavated from material with the consistency of packed concrete, moved, heated, and have its dust separated from the vapour — by equipment operating in vacuum, at cryogenic temperature, with no possibility of maintenance.

Terrestrial mining handles far leaner ores routinely, but it does so with heavy machinery, abundant power and technicians. On the Moon every one of those advantages is inverted, which is why concentration alone was never the limiting factor.

Why Oxygen Is Ready Before Water

The most useful correction to the popular picture is that lunar oxygen and lunar water are separate problems at very different stages of maturity.

Oxygen does not require ice at all. Lunar regolith is roughly 40 to 45% oxygen by mass, chemically bound in metal oxides across essentially the entire surface. Processes such as molten regolith electrolysis liberate it from ordinary soil, and they work anywhere — in sunlight, at survivable temperatures, without hunting for a shadowed crater.

Water ice sits only inside permanently shadowed regions, which are the coldest and least accessible places on the Moon and receive no sunlight to power the equipment sent there. The assessment across the field going into 2026 is blunt: lunar oxygen is close to ready, and lunar water is not.

The First Ice-Mining Attempt Failed

NASA's Polar Resources Ice Mining Experiment flew to the lunar south pole aboard Intuitive Machines' IM-2 Athena lander, touching down on 6 March 2025. It was the first serious attempt to drill for and identify lunar ice in place.

The landing went wrong. Athena came down roughly 400 metres from its intended site near Mons Mouton and came to rest on its side in a small crater. In that orientation its solar panels could not sustain the spacecraft, and the batteries depleted before the drilling sequence could be completed.

The instruments were never given the chance to fail or succeed on their own terms — the mission ended at the landing, not at the science. That is an important distinction, because it means the hardware remains unproven rather than disproven.

What TRIDENT Was Built to Do

The prospecting package is worth understanding, because the same design is intended to fly again. TRIDENT is a one-metre auger drill built by Honeybee Robotics, designed not to extract usable quantities of water but to bring buried material up where it can be analysed.

Why VIPER Was Cancelled and Then Wasn't

The programme meant to answer the prospecting question at scale had an unusually turbulent history. VIPER, the Volatiles Investigating Polar Exploration Rover, was designed to carry a version of the same TRIDENT drill and MSolo spectrometer across the lunar south pole, mapping where ice actually sits rather than sampling one point.

NASA cancelled it, a decision that drew significant criticism given that the rover was largely built and that no comparable mission existed to replace it. It was subsequently revived, and is now being prepared for launch in late 2026.

The episode is a fair summary of lunar ISRU's status overall — technically credible, strategically agreed to be important, and repeatedly vulnerable to budget cycles. A resource base cannot be planned around a rover whose existence depends on the current appropriation.

The Chemistry That Makes Water Worth Mining

Water's appeal is that it is simultaneously life support and rocket fuel. Passing an electric current through it splits it into hydrogen and oxygen, and the mass divides roughly 11% hydrogen to 89% oxygen.

That lopsided ratio is the point rather than a footnote. Oxygen is the oxidiser, and in a hydrogen-oxygen propellant load the oxidiser dominates total mass. Producing oxygen locally therefore removes most of the propellant mass a mission would otherwise have to lift out of Earth's gravity well, even if hydrogen still had to be brought along.

It also explains why regolith-derived oxygen is strategically valuable on its own. A base that makes oxygen but imports hydrogen has already eliminated the larger share of its resupply burden — without needing a single successful ice drill.

Why Cold Is the Hardest Engineering Problem

Permanently shadowed regions are cold traps: crater floors near the poles where the Sun never reaches, sitting around -230°C. That is precisely why the ice has survived for billions of years, and precisely why extracting it is so difficult.

At those temperatures ordinary engineering assumptions collapse. Lubricants freeze solid, batteries lose capacity, metals become brittle and normal electronics stop functioning without dedicated heating — which itself demands power in a location that by definition receives no sunlight.

Equipment must then survive the transition between that cold and the temperatures needed for processing, cycling through hundreds of degrees while maintaining seals against vacuum and abrasive dust. Lunar regolith is sharp, electrostatically charged and clings to everything, and it has degraded mechanisms on every mission that has encountered it.

What Has to Happen Before a Lunar Gas Station

The distance between current capability and a working propellant depot is measured in successful missions, not in years or engineering drawings.

Prospecting has to succeed first — a drill needs to reach ice, a spectrometer needs to confirm it, and the result needs repeating across enough sites to establish that a deposit is extensive rather than a local anomaly. Then extraction has to be demonstrated at meaningful scale, which no one has attempted. Then liquefaction and long-term cryogenic storage, hydrogen being notoriously prone to boiling off and leaking through most containment.

Oxygen production from regolith may well arrive considerably earlier and is the more realistic near-term prize. The honest position for 2026 is that ISRU's chemistry is well understood, its oxygen pathway is close to ready, and its water pathway is still waiting on a first successful drill.

Conclusion

In-situ resource utilization really is the difference between visiting the Moon and staying there. A base that manufactures its own oxygen and propellant breaks the resupply dependency that has constrained every crewed mission ever flown, and the underlying chemistry is neither speculative nor difficult.

The gap is entirely in execution. One direct measurement of lunar water ice, from a single impact site, with meaningful error bars. One attempt to drill it, ended by a landing failure before the instruments could operate. One prospecting rover cancelled and reinstated, now targeting late 2026.

The nearer-term story is oxygen. It is present across the whole surface rather than only in the coldest craters, it needs no shadowed-region survival engineering, and it represents the heavier share of any propellant load. If a lunar gas station opens, it will most likely sell oxygen first — and the water may take considerably longer than the headlines suggest.

Data Source and Attribution

NASA PRIME-1LCROSS AnalysisTRIDENT Drill Paper

Water ice concentration figures come from the LCROSS mission's published plume analysis and subsequent modelling of the Cabeus impact. Mission hardware details, drill specifications and the PRIME-1 landing outcome come from NASA and from published documentation of the TRIDENT drill developed for the PRIME-1 and VIPER missions. Oxygen extraction maturity reflects current in-situ resource utilization assessments. Launch dates are targets and remain subject to change.

FactsFigs reviews, cleans, and cross-checks every source dataset before shaping it into a data story. Each visualization is created and designed in FactsFigs Design Studio — an internal tool developed and owned by FactsFigs — and is the original work of a FactsFigs author, not an AI-generated copy of any existing graphic. Individual assets within a visual may or may not be produced with AI tools, but the design of the visual itself is solely FactsFigs' own.

Figures are estimates at the time of publication, provided for information only — nothing here is financial advice or a guarantee of accuracy.

2026-07-20