Let’s start from the beginning

Let’s build a train on the Moon! Imagine chatting with a friend and arriving at that conclusion after a few beers. Beyond the “how?”, the most obvious question is “why?”. The short answer is that, right now, for nothing… but I prefer the long answer. Imagine Ponce de León mapping the coasts of Florida back in 1515. As much as we can’t conceive the current state of Florida without transportation infrastructure, back then, the last thing they needed was to build roads to nowhere. Something similar happens to us right now. Therefore, my intention with this article is not to answer that question but to pose it from a different perspective. Let me know in the comments if I’ve convinced you.

Let’s get serious

I have several personal anecdotes with colleagues from work and university with whom I talked about space colonization. Most saw no interest or argued that we have other priorities to worry about on our planet. Others, the minority, advocated for a neutral space where space exploration and scientific development in international collaborative projects are prioritized, instead of a competition for resources or other strategic interests. A sort of “this time let’s do things right.” Finally, there were those who envisioned a future of spaceships and colonies on Mars… but without going into much more detail. A “how cool, but honestly, I have no idea what you’re talking about.” As a child, I was in this last group of naive and well-intentioned optimists, although now I see it more pragmatically: I believe that space exploration is an inevitable advancement, whether in 5 or 50 years, mainly for two interrelated reasons: material extraction and climate change. I believe both reasons are critical, and I have my arguments that I will detail in the next entry to not make this one too heavy. But now I’m going to ask you for an act of faith and assume a scenario in which space colonization is a reality. I don’t mean something like Columbus’s first voyage, where less than 100 bewildered people arrived in America and returned a few days later. Rather, like the second one, where 2,000 came with the clear intention of staying. With this premise, we can focus on what interests us, so let’s start with some important considerations:

1. We define a “transportation system” as the physical infrastructure and rolling stock necessary for moving people, materials, machinery, and products (which we’ll collectively call resources) between two points.
2. The need for a transportation system is a direct consequence of human activity. In other words, human activities require resource movement, and therefore, a transportation system.
3. There are many different transportation systems, but their characteristics make them more or less suitable for specific objectives. In the context of this thought experiment, the prioritized attributes are, in order: safety for people, transportable volume, and travel speed.

Point 1 serves only to clarify that we are talking about the same thing. It is point 2 that really interests me, since it logically implies that: if we assume the premise that space colonies with human activity exist, then a transportation system will be necessary in those colonies. Not convinced? Okay, I’ll try again. Remember that for now, we continue to assume the premise as true (I promise we will discuss it later), so I will focus on justifying the truthfulness of the argument.

According to the European Environment Agency, all human activity exerts pressure on the environment (link), defining pressure sources as “agents and means that cause alterations in the environment” (link). By definition, altering a physical environment means “changing its essence or form” or “damaging it”. In our lunar context, this means adding, removing, or modifying its components. Now, can you think of any way a human activity could alter an environment without moving either the modified object or the thing modifying it? Essentially, human activity alters the environment, which means bringing in people or machines to perform the activity and, in some cases, removing altered objects. Yes, it’s a bit convoluted, but rephrase the last sentence differently, and you’ll get point 2. The bottom line: if someone wants to do something, first, they have to get there, and last, they have to leave.

Okay, but why a train?

Simple: because it’s faster, safer for humans, more reliable, has a higher transport capacity, and offers lower operational and maintenance costs.

Returning to point 3: safety for people is even more critical on the Moon than on Earth because everything in the lunar environment can kill you in seconds. Whatever transport system is chosen must protect against extreme cold, vacuum exposure, and radiation levels that make Chernobyl look mild. Transportable volume is mainly about material transport, and the logic is simple—the more, the better. Lastly, travel speed ties into the first two criteria but remains a key factor because it includes time value and, therefore, efficiency.

Some might wonder why I haven’t included costs as a criterion. First, it’s hard to quantify economic alternatives due to a lack of accessible data. However, I’ll provide rough estimates where possible. Second, given the complexity of the problem and the current technological state, I believe functionality and feasibility are more important than strict cost constraints—unless costs become prohibitively high. With that said, let’s review the available options for transporting people and materials on the Moon today (if you know of others, let me know in the comments!):

Lunar Rovers

Perhaps the first thing that comes to mind when we think about driving around the lunar surface (I’m sure some Futurama fans feel called out). The most famous “model” today might be the VIPER from the Artemis mission, but to be honest, it’s not going through the best of times (link). So, let’s stick with a slightly older but much more iconic reference: the Lunar Rover from the Apollo missions. We could talk days about this marvel of engineering, but this post is already getting long, so I’ll leave you with a couple of great links that describe its capabilities (link, link y link in english and spanish). For now, I’ll focus on the following key data: A Lunar Rover could carry about 440 kg (970 lbs) of weight between crew and cargo, travel at a speed of 10–12 km/h (6–7.5 mph), and had a theoretical range of about 50 km (31 miles) (it was an electric vehicle). As for costs, each unit had a price tag of around 38 million 1971 dollars (link). According to this page (link), that’s equivalent to over 297 million dollars in today’s money (2024)… annyway, it was the Space Race, and getting to the Moon first was never going to be cheap. If you want a more modern reference, there are rough cost estimates for China’s Yutu-2 rover, which has been exploring the lunar surface since 2019. It’s not quite the same as the Apollo Rover, but it works as a miniature example, with an estimated cost between 73 and 172 million dollars (link). I want to highlight one key detail: If you’re imagining these vehicles as small golf carts capable of cruising all over the Moon… they’re not. In fact, they’re quite delicate and cumbersome. We’ve already seen that the Apollo Rover had a range of up to 50 km (which it never actually reached). Meanwhile, by January 2022—three years after landing—Yutu-2 had traveled “only” about 1,000 meters (0.62 miles). The Soviet Lunokhod-2 holds the distance record at 37 kilometers (23 miles), a record that stood for 40 years until it was surpassed by Opportunity on Mars in 2012 (link) … Yes, you’re absolutely right—these machines are engineering marvels, and their limited travel distances don’t take away from their achievements. But the simple fact is that they were never designed as a general-purpose transportation system for long distances or large cargo. They were built as transport solutions for astronauts and scientific equipment in a local environment. Maybe in the future, we’ll see rovers with greater capacity, but that technology doesn’t exist yet and will have to overcome major challenges—specialy the terrain (as far as I understand, it’s like trying to walk on flour… but we’ll talk about that in other post).

Rockets

Another classic of actual technology. Let’s say it complements rovers by focusing on the exact opposite: long-distance travel. This idea has its pros and cons. The positive side is that, due to lower gravity and the absence of an atmosphere, much smaller rockets would be needed compared to those launched from Earth. As a simple comparison, the Saturn V rockets that carried the lunar modules generated about 34,500 kN of thrust, whereas the ascent module used to return to orbit only required around 16 kN Link). The downside is probably obvious: it has very limited capacity. For example, the Apollo ascent module could carry only 108 kg (238 lbs) of samples, excluding the crew (link). Of course, new systems would offer greater capacity and significantly lower costs, such as the promising SpaceX Starship (link). However, in practical terms, we can consider this transportation system useful only for very long-distance travel and for transporting highly valuable resources (people, complex machinery, and refined minerals of great worth).

Conveyor belts

Now we move on to more “exotic” solutions. At first glance, the idea of conveyor belts seems quite logical for transporting materials, and in fact, NASA is already brainstorming with the FLOAT project (which they call the “first train on the Moon”—though I’d love to hear how exactly it resembles a train… link). The idea is to create “robotic trays” that, using magnetic properties, can float and move along magnetic tracks laid on the ground. Now, let’s get into the details: I can’t provide any cost estimates because I haven’t been able to find a single reference… though it seems relatively cheap, given that it requires little maintenance and minimal infrastructure (as long as we don’t count the cost of transporting it to the Moon). The highest cost seems to be the robots themselves. Their speed would be around 0.5 m/s (1.8 km/h), while their carrying capacity is about 33 kg (73 lbs). The key to this system is having a huge number of robots, allowing for the transportation of 100,000 kg (220,000 lbs) of regolith/payload over several kilometers per day (link). To wrap it up, one good and one bad thing about this system:

• The bad: It would only be useful for short-distance material transport (not for people).
• The good: It would consume very little energy (40 kW per day for the aforementioned performance).

Rail-guided systems

Think of a train, the kind that runs on rails, but on the Moon; that’s what we’re going to talk about in this option. Let’s start with the downsides: we need infrastructure (a lot of it), we need machinery to build that infrastructure, we need a locomotive and carriages capable of withstanding the harsh lunar environment, we need steel rails to guide the train (and I’d love to hear where we’re supposed to get that on the Moon), we need energy to power it… The good part? It would be a system capable of transporting passengers and tons of materials in a single trip, at high speeds, and for a very low cost. It is also a safe and fully autonomous means of transport.

Some visionaries have already proposed this type of system for space bases, including Daniel García Espinel with his Martrópolis project (link). However, when I started working on this topic in 2021, there weren’t many well-documented proposals on how to build such a system. But as I updated the information for this blog, I came across a new development: since early 2024, the Northrop Grumman conglomerate has proposed a very similar train concept (link)… although, as we’ll see, its design has some significant weaknesses. (As a side note, I find it interesting that a military contractor was the one to propose this idea).

Throughout this blog, I will go into more detail on how something like this could be built, based on my Master’s Thesis proposal. But to give you a taste of what’s possible, here are some numbers:

A lunar train could theoretically be built in two phases:

• A second phase with installed refineries and the ability to manufacture metal rails.
• An initial, more rudimentary version without refineries on the Moon—meaning no steel rails.

The flat-terrain speed in the second phase would be around 70.5 km/h (43.8 mph), based on the track alignment criteria considered, with a locomotive power requirement of 25 kW. The train’s length would be limited by the force transmission to the lunar surface, requiring many axles to distribute the weight. In fact, the maximum theoretical weight per axle is around 2.65 tons (yes, there are studies on the bearing capacity of lunar regolith). Regarding costs, my estimates have some limitations in quantifying specific expenses, but hey, they serve as a decent reference. My calculations suggest that a 21-kilometer (13-mile) section would cost around €620 million in direct execution costs, where:

• 57% of the cost is just for transporting the construction machinery from Earth.
• 37% goes to Safety and Health measures.
• The actual construction cost is less than 6%!

If the low labor costs surprise you, it’s because everything would be built by autonomous robots. We’ll explore this in more detail later, but for now, here’s a quick summary:

In short, a train requires a complex implementation of the associated infrastructure, but it has a capacity and speed that all other transport systems lack. Especially for medium distances and the transport of large volumes of resources.

Recap

The goal here isn’t to sell you on the idea. I just want you to see that the concept isn’t as absurd as it might seem at first. Let me sum it up in a few key points:

1. If we assume the existence of lunar bases, it will be necessary to implement transportation systems on the Moon’s surface.
2. Currently, there are very few options for such systems. While each has its strengths, none of them simultaneously provide: safe transportation for people, high-capacity material transport, good overall performance. Except for…
3. Rail-guided systems—in other words, trains. They theoretically (and I’ll prove this in future posts) offer greater advantages for moving large quantities of resources over distances of at least a couple of kilometers.