Don't Get Your Hopes Up for a City on the Moon
The dawn of the space age in the 1950s and 60s arrived with a particular kind of audacious optimism. The futuristic zeitgeist of the era inspired sweeping scientific visions of humans not merely visiting the Moon, but living there — permanently, in sprawling cities beneath pressurized domes. Arthur C. Clarke, the legendary science fiction author and visionary futurist, envisioned outposts of inflatable, pressurized domes that would gradually grow into thriving urban centers. The U.S. Army's Project Horizon was not science fiction — it was a serious, formally commissioned feasibility study for establishing a military and scientific lunar base, complete with detailed logistics and cost projections. Other proposals followed from NASA, aerospace contractors, and independent scientists alike.
An eventual, permanent human presence on the Moon was considered an inevitability by many of the brightest minds of the 20th century. The Apollo missions seemed to confirm that humanity was on a fast track to becoming a multi-world species. Then the funding dried up, the political will evaporated, and the Moon — once humanity's most immediate cosmic destination — receded into the background of ambition.
Now, decades later, a renewed surge of interest in lunar colonization has emerged, driven in part by a remarkable scientific discovery: the Moon harbors a vast amount of frozen water ice, primarily locked in ancient, sunless craters near its poles. This revelation has reignited dreams of lunar colonies and cities. Prominent billionaires have shared grandiose visions of building permanent settlements on Earth's frigid natural satellite. But are these ideas grounded in scientific reality? Or are they little more than aspirational marketing dressed up in the language of exploration?
The Science of Lunar Water Ice
Before evaluating the feasibility of a lunar city, it is essential to understand the nature of the Moon's water resources. The Moon's water ice is not distributed like the oceans or aquifers of Earth. It exists primarily in Permanently Shadowed Regions (PSRs) — ancient craters near the lunar poles whose geometry shields their floors from direct sunlight for billions of years at a time. In these perpetually dark environments, temperatures can plunge to as low as –250°C (–418°F), cold enough to trap and preserve volatile compounds, including water ice, over geological timescales.
The existence of lunar water was long theorized but only confirmed with increasing certainty over recent decades. NASA's LCROSS mission in 2009 provided dramatic confirmation when it deliberately crashed a spent rocket stage into the lunar south pole's Cabeus Crater, and spectroscopic analysis of the resulting plume detected water vapor and ice. Subsequent missions, including India's Chandrayaan-1 and NASA's Lunar Reconnaissance Orbiter (LRO), have further mapped and characterized these ice deposits.
There are other potential water sources on the Moon beyond the PSRs — including hydroxyl (OH) molecules bound within the lunar regolith across much of the surface — but these diffuse deposits are not considered easily accessible or resource-rich enough to support large-scale human use. The PSRs remain the primary focus of discussions about extractable lunar water. One of the most prominent of these regions is Shackleton Crater, a large, 21-kilometer impact crater near the Moon's south pole. A 2007 study estimated that Shackleton alone could hold between 1.6 to 4.5 million tons of water ice — a striking range that illustrates just how poorly constrained our measurements of lunar water still are, and how urgently ground-truth measurements from surface missions are needed.
For further reading on lunar water discoveries, visit NASA's Lunar Science page and the European Space Agency's Moon exploration portal.
A New Study Cuts Through the Hype
A new paper published in Frontiers in Space Technologies applies rigorous scientific analysis to the question of lunar water sufficiency. The article, titled "No cities on the Moon: a billion tons of water is not enough for sustainability," is authored by Dr. Martin Elvis, from the Smithsonian Astrophysical Observatory at the Center for Astrophysics, Harvard & Smithsonian, and Dr. Jonathan McDowell, from the Space Research Centre at Durham University in the United Kingdom. Both authors bring decades of expertise in space science, and their analysis is methodical, data-driven, and — for those who have been enthusiastically nodding along to billionaire visions of lunar metropolises — somewhat sobering.
"The Moon, with an area of 37.93 million km², about 50% bigger than North America (24.71 million km²), is talked about as our '8th continent.' As a result, many authors consider the Moon's resources to be so vast as to pose no significant constraints on industry and settlement." — Elvis & McDowell, Frontiers in Space Technologies
The authors are not merely pointing out a misconception — they are highlighting a systemic failure of reasoning. Comparing the Moon to a continent, as if its surface area alone translates to abundant, accessible resources, is a category error of the highest order. Earth's continents are rich not because of their area, but because of billions of years of geological activity, hydrological cycles, biological processes, and a dynamic atmosphere. The Moon has none of these. Its "resources" are static, deeply frozen, poorly mapped, and extraordinarily difficult to extract.
The billionaire class has not been shy about their lunar ambitions. Jeff Bezos has suggested moving heavy industry to the Moon to protect Earth's environment — an idea that sounds visionary until one starts asking basic logistical questions. Elon Musk has spoken of self-growing cities on the Moon, as though the satellite's surface were a fertile substrate waiting to bloom. Elvis and McDowell's paper methodically dismantles these visions, not with philosophical objections, but with arithmetic.
"Presumably, in order to make an economic difference to the human economy, they are both imagining significant settlement of, say, at least 1% of Earth's population, i.e., ∼80 million people living on the Moon. These are grand plans." — Elvis & McDowell, Frontiers in Space Technologies
The Energy Question: A Solvable Problem
Any serious discussion of lunar settlement must grapple with two fundamental resource requirements: energy and water. The paper examines both, but reaches very different conclusions about each. On the question of energy, the authors are cautiously optimistic.
One genuinely favorable aspect of the Moon's geography works in the colonists' favor. The Permanently Shadowed Regions that contain the ice deposits are located in craters near the lunar poles. Crucially, the rims of these very same craters are among the best locations in the solar system for solar power generation. Because the Moon's axial tilt is only about 1.5 degrees — far less than Earth's 23.5 degrees — crater rims near the poles experience nearly continuous sunlight. Solar panels mounted vertically on these rims can rotate to track the Sun as it circles the horizon, maximizing energy collection.
According to the paper, these solar arrays could range from several meters to as tall as one kilometer in height, generating substantial power with relatively minimal infrastructure. Additionally, fission nuclear power represents a well-understood technology that could supplement or replace solar generation, particularly during periods when dust storms or equipment failures might compromise photovoltaic output.
Elvis and McDowell crunch the numbers carefully, accounting for both personal energy use and the far greater demands of industrial operations. Their conclusion: energy is not a limiting factor for populations up to at least one million people, assuming either advanced solar or fission power systems.
"If industry requires no more than double the personal use rate, then a million people can be supported on the Moon with either solar or fission power." — Elvis & McDowell
They do note that fusion power — still under active development here on Earth — would become a practical necessity if populations were to reach several million. But for realistic near-to-mid-term scenarios, energy appears manageable. The bottleneck, they conclude, lies elsewhere entirely.
The Water Problem: Fundamental and Unforgiving
Water is not merely a convenience for a lunar settlement — it is the sine qua non of survival. It is required for drinking, cooking, personal hygiene, growing food, manufacturing rocket propellant (via electrolysis into hydrogen and oxygen), producing breathable oxygen, industrial cooling, and countless other applications. There is no substitute, no workaround, and no reasonable prospect of importing it in sufficient quantities from Earth given the staggering cost of lifting mass out of Earth's gravity well.
The authors settle on a working baseline figure that is generous by current estimates: one billion tons of water ice in the Moon's Permanently Shadowed Regions. To put this in perspective, Earth's total water inventory is approximately 1.4 billion billion tons (1.4 × 10¹⁸ metric tons) — making the Moon's entire estimated water reserve a negligibly small fraction of what our own planet holds. One billion tons is not nothing, but in the context of supporting a large human civilization, it is dangerously finite.
The authors build their consumption model from several key figures:
- Personal water use: Approximately 125 tons per person per year, based on average adult consumption estimates for a U.S. resident, covering drinking, sanitation, and direct personal use.
- Food production water use: According to World Bank estimates, between 2,000 and 5,000 liters of water are required to feed one person per day — translating to roughly 730–1,825 tons per person per year, or approximately 6 to 15 times the direct personal consumption figure.
- Industrial and manufacturing use: Additional water demands for production, construction, and propellant manufacturing add further pressure to the resource budget.
Working with these parameters, the researchers calculate a stark result: a city of one million people would exhaust the Moon's entire estimated water reserve in approximately 2.4 years without any recycling. The number is almost absurdly small — and it frames the entire challenge with brutal clarity.
The Critical Role of Water Recycling
Of course, no serious proposal for a lunar settlement assumes zero water recycling. The authors acknowledge that recycling will be absolutely central to any viable lunar water economy, and they look to humanity's most sophisticated existing example: the International Space Station (ISS).
The ISS, out of sheer necessity, operates the most advanced closed-loop water recycling system ever deployed by humanity. Its Water Recovery System (WRS) processes urine, condensate from crew members' breath, and other wastewater back into potable water. Until 2023, the ISS achieved a water recycling efficiency of approximately 94% — an impressive technological feat. In 2023, NASA engineers achieved a remarkable breakthrough, pushing that efficiency to 98%. As the authors note, this may seem like a marginal improvement, but it is not — it represents a threefold reduction in water losses, a significant engineering achievement with profound implications for long-duration space habitation.
You can learn more about the ISS Water Recovery System at NASA's International Space Station page.
Applying these recycling efficiencies to lunar settlement scenarios, Elvis and McDowell produced a table of projected water exhaustion timelines for cities of varying populations. The results are illuminating:
- At 94% recycling efficiency, a city of one million people depletes one billion tons of water in approximately 40 years.
- At 98% recycling efficiency, that same city extends its water supply to approximately 100 years.
- A smaller settlement of 100,000 people at 98% efficiency could sustain itself for approximately 1,000 years — a timescale the authors describe as genuinely approaching sustainability.
- A "Moon Village" of approximately 1,000 people — comparable to the winter-over population of Antarctic research stations — could operate with relatively little constraint on water use for extraordinarily long periods.
"For such a large investment, a lifetime of about a century seems to fall short of the sustainable, long-term settlement beyond Earth that some advocate." — Elvis & McDowell
These figures are not merely academic. They define the envelope of what is physically possible given the Moon's known water resources, and they set hard limits that no amount of optimism or marketing language can dissolve.
Strategies for Extending Water Longevity
The authors do not simply present a problem and walk away. They explore several potential strategies that could extend the functional lifetime of lunar water resources, each with its own complexities and limitations.
Dietary Adaptation
One of the most direct ways to reduce water consumption is to shift away from water-intensive food production. Meat, particularly beef, is extraordinarily water-intensive — estimates suggest that producing one kilogram of beef requires approximately 15,000 liters of water. A low-meat or plant-based diet could substantially reduce the water footprint of food production. Of course, whether any form of livestock agriculture is practical on the Moon at all — given the radiation environment, microgravity effects, and spatial constraints — is itself a serious open question. Hydroponic and aeroponic plant cultivation systems would almost certainly form the backbone of any lunar food production system.
Importing Water from Space
Water-rich asteroids, particularly carbonaceous chondrite (C-type) asteroids, are known to contain significant quantities of water bound in hydrated minerals. In principle, water extracted from a well-chosen asteroid and delivered to a lunar depot could supplement the Moon's native ice reserves. However, the scale of engineering required to capture, redirect, process, and transport asteroid-derived water is truly enormous — and well beyond the scope of near-term mission planning. The authors acknowledge this possibility but note that its magnitude falls outside the bounds of their current analysis.
Finding More Water on the Moon
Perhaps the most realistic near-term solution is simply to find more water on the Moon itself. Current estimates are based on remote sensing data with significant uncertainty ranges. Ground-truth measurements from robotic or crewed missions could reveal that the Moon's water inventory is substantially larger than the conservative one-billion-ton baseline used in this study.
Underground reservoirs are a particular focus of speculation. While there is no confirmed evidence of subsurface liquid water on the Moon — unlike on Mars, where there are radar-based indications of possible subsurface brines — subsurface ice deposits that are more concentrated and extensive than surface PSR ice are not out of the question. Lava tubes, for example — vast underground tunnels created by ancient volcanic activity — could potentially trap and preserve ice. As the authors note: "In the case that an unexpected reservoir of water is found, then dou
