The discovery of water at the poles of the moon has inspired speculation about whether we can build there – cities, industrial neighborhoods, places where people work and live. But is there enough water, and can we use it efficiently enough? In their new article in Frontiers in Space Technologies, Dr. Martin Elvis and Dr. Jonathan McDowell do the math. Even at the most generous estimates of the available water and recycling efficiency, a city of one million people would only last 100 years before all the water runs out. In this guest editorial, which Frontiers in Space Technologies provided, Elvis breaks down what we can and can't do with the technology available.

The discovery of water in dark craters at the moon's poles has sparked a "moon rush" to establish bases there, and then villages, cities and heavy industry. How sustainable is this rush?

Soon after the moon was formed, asteroids bombarded it, creating its pockmarked topography. Near the moon's poles, there are dozens of crater floors that have not seen direct sunlight in about four billion years. Many of these cryogenically cold "pits of eternal darkness" are below a temperature of 110 Kelvin (minus 262 degrees Fahrenheit), forming "cold traps" in which water ice won't sublimate away by more than a millimeter in a billion years, even in the vacuum at the moon's surface. And there is water to freeze: some of those asteroids brought it along with them. Since 2013 we have known that that water is still there, possibly as much as a billion tons of it, thanks to a succession of moon-orbiting missions that have mapped out likely locations in increasing detail.

The presence of water at the lunar poles changes everything about human visits to the moon. Suddenly it is practical to envisage a "moon village" where people could survive without importing every last bit from Earth. They could grow their own food and have showers and functional toilets, unlike the astronauts on the International Space Station (ISS). This possibility has sparked ambitions from the likes of Jeff Bezos to move heavy industry to the moon and Elon Musk's aims to have "self-growing cities" there. These are enticing visions, but are they attainable?

The good news

We realized that we could make a first feasibility estimate simply by looking at how much power and water a large population would use on the moon, and how long a city of 100,000 or one million people could survive there. How sustainable is a city on the moon?

The good news is that power, even without nuclear reactors, is not a problem. That's because the rims of the craters containing the pits of eternal darkness are in almost permanent sunlight. (That's why they are sometimes, slightly incorrectly, referred to as the "peaks of eternal light.") Up on those highly illuminated rims it would be possible to generate three gigawatts of electricity using kilometer-tall towers covered with photovoltaic arrays. It should be possible to manufacture solar panels on the moon. There's certainly plenty of silicon. This could even make AI data centers on the moon possible, by placing them near these sunny peaks. That would be the start of a true lunar economy.

The bad news

The bad news is that, even assuming a generous one billion tons of water to start with, without recycling the answer to our question of sustainability is that the city would last only a few years. Even in the best-case scenario, with 98% efficient recycling of water — the same as achieved on the ISS — a one-million-person city will run out of water in just over a century. That is a problem for sustainable moon settlement. And today's best estimates of the amount of water on the moon are about 30 times less than a billion tons, shortening the time to water exhaustion by the same factor. This means that even a small city would run dry after just a decade or so. A moon village with a thousand people, or even a town with 10,000 people, would instead be sustainable for several centuries or more.

There are potential solutions. We could improve the efficiency of water recycling by a factor of five or more; lower water usage from new techniques such as vertical farming; import water, probably from accessible asteroids; or, simply, find more water. Of these it may be this last hope that is most promising. Current water surveying techniques do not reach below the surface more than a few meters, while the rubble-like rocks of the surface regolith typically extend tens of meters down and may contain water in the cold traps. If the ambitious plans of the space billionaires are to be realized, then finding more water will be the first step.

We've been studying the moon for a long time. NASA has gathered mountains of data over decades of lunar missions and studies. The data collected by the Lunar Reconnaissance Orbiter (LRO) alone, for example, exceeds that of all NASA's other planetary missions combined.

Analyzing all this information takes lots of time and effort from researchers. So, NASA and IBM have developed an AI model to help sift through the moon data — an advance the agency announced on Thursday (Sept. 10). The open-source model, called the NASA-IBM Lunar Foundation Model, is currently available for free at Hugging Face.

"NASA has spent decades building an extraordinary scientific record of the moon, but collecting data is only part of the job," NASA's Kevin Murphy, chief science data officer and acting chief data and AI officer, said in the announcement.

The new AI model can help researchers with tasks like mapping craters, finding young volcanic features, and modeling possible locations of water ice near the lunar poles, according to NASA.

In the announcement, Murphy said that NASA needs to make scientific data easier to explore and use. To this end, NASA has been developing and releasing science-focused AI models with IBM. Previous models include one focused on Earth observation and another for heliophysics.

This is part of a larger trend of AI use in planetary science. Space agencies and private space companies around the world have been looking into how AI models can help with analysis to increase scientific productivity, reduce costs and streamline workflows.

"The NASA-IBM Lunar Foundation Model shows what's possible when we bring AI to NASA's petabytes of scientific data," Murphy said in the announcement. "That's a real opportunity we see with AI: turning large-scale data into new discoveries."

The AI model was mostly trained on data from the LRO mission. Over 17 years, NASA's lunar orbiter has compiled a nearly complete mosaic of images that detail the moon's surface in high definition.