Vision 2040: The first million on Mars

As the 2020 Mars Society Convention has just finished, I’m publishing here my entry in the Mars City State Design Competition. Also available as a pdf. Congratulations to the winners team Nexus Aurora and all the other 176 competitors!

Vision 2040: The first million on Mars

Casey Handmer

Conceptual overview

Twenty pages is hardly adequate to describe the totality of any city, let alone the first city on Mars. Too much is uncertain or unknowable for me to be prescriptive. And yet, to chart our course we need some idea of a destination. The tools of science and the talents of a generation are easily equal to the task, provided only that we set out in the right direction. This design competition entry therefore places an emphasis on developing not answers but questions, as a step toward focusing our attention and, if we are lucky, sloughing off a layer of two of ignorance. Let us focus on the less intuitive aspects of Mars city design and seek useful insights. 

In these 20 pages, I present a cross section of the first city on Mars.

Technical design

What are the requirements? What functions is the city intended to perform?

It must:

  • Meet the human needs of its population.
  • Provide infrastructure for ongoing, rapid growth.
  • Achieve near self-sufficiency.

How are these functions to be executed? How can a good design serve and promote these core functions? On what questions can we base our trades?

Urban planning

Let’s lay out the city and determine what goes where. What aspects can be determined by analogy with organically-developed cities on Earth, and what has to be re-invented to meet the conditions on Mars?

Some functions ought to be collocated, such as living, food, healthcare, education, and entertainment. Some functions must be segregated, such as noisy or dangerous industrial processes. And some functions must necessarily be more remote, such as the space port, solar farms, or mines.

How these parts are laid out determines their interfaces. It is crucially important that both people and cargo are able to move efficiently throughout the city, even as population and traffic continues to increase. This requires adequate space to create wide thoroughfares, as well as dense, walkable environments that permit rapid, shirt-sleeves pedestrian movement between every area. 

Abolishing scarcity

Mars may be the second most hospitable planet in the known universe, but it is still a frozen poisonous cratered irradiated asphyxiating place that cares not if we live or die. A Mars city of any size needs to maximize productivity to increase the odds of survival. Some industrial factors, such as human labor, will always be relatively scarce. As much as possible, other key resources such as water, electrical power, living and working space, heat, and raw materials should be made abundant. Scarcity and rationing inflict enormous costs on any process, and the Mars city simply cannot afford them. 

How can we minimize avoidable scarcity and ensure that our factories exist in a comfortable buyer’s market? While aggressive recycling and waste minimization are an important part of the picture, the Mars city needs to generate ongoing surpluses of everything, even as demand continues to grow. Much of primary production is gradually being automated on Earth – on Mars, even the tooling manufacturing is routinely automated.

More generally, automation exists as an abstraction layer between human intention and actual manipulation of matter. To ensure constant increases in productivity relative to human labor inputs, automation has to continually ascend the value chain, from manufacturing robots to robot production robots.

Finally, pressurized volume is itself a valuable commodity. Inadequate supply of space in factories, farms, or living areas exacts an exponentially increasing toll. If the Mars city is to flourish in a surplus of space, the labor and material cost of generating more volume must be aggressively minimized.

What does this look like?

While I can be relatively certain about what the Mars city needs, I cannot be as certain about how to meet those needs. I offer here a partial sketch of how a Mars city might go about solving these problems but remind the reader that Vision 2040 is ultimately the product of the expertise of millions of people working for many decades.


  • The city is pressurized and climate controlled, with almost all its occupants working in an expansive shirt-sleeves environment.
  • The base unit of pressurized volume is a tensile Kevlar-reinforced ETFE membrane, similar to a high performance yacht sail, anchored to the ground at regular intervals. No domes, no excavation, no tunnels, no masonry arches, no dense rubble piles.
  • This approach maximizes usable area for given inputs of materials, labor, and energy. Known for their light weight and versatility, inflatable tensile structures have a long history and are not as exotic as they may sound. They are used for both Zodiac dinghies and inflatable mattresses. They have even been used to make a fully functional inflatable plane.
  • Steel ground anchors are installed with a pile driver. Membrane-ground interfaces are corrugated steel embedded in a trench. This approach accommodates thermal expansion, sealing, and local tensile loads without requiring major geoengineering.
  • Membrane material is transparent and can be welded/patched thermally or chemically in the field. The membrane requires a minimum of processing or finessing for installation.
  • Interior structures built to purpose, generally do not require thermal, pressure, or rain protection. Structures can also be suspended from the membrane.

Image: Rendering showing how periodic anchors transmit pressure load to the surface for effectively limitless pressurized volume at minimum cost.

Residential core

  • Crescent shaped residential core surrounding broad undeveloped, unenclosed area ensures that “wilderness” is readily visible from within the city.
  • Central residential/commercial area with walkable high density, lowrise neighborhoods. Population density comparable to Manhattan (25,000/km^2), less than Kowloon walled city (2,000,000/km^2). With a surface area of 50 square kilometers, parts of the city extend beyond the relatively close Martian horizon.
  • Functional urban design favors the use of “small multiples” where local neighbourhoods have local businesses, rather than massive centralized malls that require vehicles to access. On the other hand, functions such as the Starport would be centralized.
  • Non-intrusive infrastructure such as grid storage batteries and data centers are located in the core as well as industrial areas.

Radial industrial bays

  • Flexibly specified roughly rectangular pressurized areas radiate from the central core, each devoted primarily to specific industries. Each bay contains discrete factories and warehouses linked by a broad pressurized road, and is partitioned as necessary by bulkheads for additional isolation and atmosphere customization. 
  • Each bay is separated from the next by a broad unpressurized open space, which facilitates ready external access to all parts of the structure for, eg, maintenance or direct docking of excessively large mining vehicles.
  • Radial city design ensures all areas have room to expand as needed, while reducing congestion and travel time for nominal traffic patterns.
  • As industrial production matures, older obsolete factories are recycled and their footprint reconfigured to provide additional room for core expansion.
  • Copious pressurized space enables execution of standard, Earth-proven industrial processes at scale, with an emphasis on automation.
  • Several concentric ring roads permit travel between bays without congesting the core.
  • Some bays are devoted to primary production, including air mining, production of various gasses, liquids, and solids (glass, plastic, metal, concrete), recycling, and agriculture. 
  • Some other bays are devoted to secondary production and manufacturing. These include electronics, basic integrated circuits, electrical equipment, machinery, textiles, fabricated metal, furniture, chemicals, transport equipment, plastics, rubber, and animal products.
  • While 3D printing is essential for prototyping and specialty manufacturing, mature manufacturing will use the highest productivity methods, including traditional subtractive manufacturing. 
  • Bays are set up to permit high material flow and use rate through stockpiles, distribution via conveyor belts, automated trucks, and containerized/palletized freight.
  • If the factory is the machine that builds the machine, the industrial sector is the metafactory: The ecosystem that feeds the machine that builds the machine.

Remote sites

  • Remote sites are dedicated to mining and/or science. Mine operation is nearly fully automated.
  • Mines operate as open cut unpressurized, open cut pressurized, or subsurface in pressurized tunnels dug with automated tunnel boring machines. 
  • A thorough geological site survey has located sources of all strategic minerals. 
  • The Starport, with approach and departure clearance east and west, is located some distance to the north of the city. The Starport’s infrastructure is simple enough that it can be moved further away as the city expands.
  • Retired rockets are transported to the Rocket Park for materials recycling. 
  • Solar PV and wind turbine park. The city requires around 100 square kilometers of locally produced solar panels, an area comparable to the rest of the city. 
  • Nuclear power plants. While Vision 2040 does not depend on nuclear power, any plants would be located remotely from the city, Starport, and other major infrastructure.


Certain commodities are widely used and so “on tap”.

  • Water: Wells, reservoirs, filtration, distribution, sewer, treatment, and open space such as ponds and pools, which help with city thermal management.
  • Power: Solar, wind, storage with batteries and fuel cells, which generate electricity at the point of use with methane and oxygen.
  • Gases: Gaseous oxygen, methane, nitrogen, CO2 (below ambient pressure), partial vacuum for low pressure gas disposal and methane line jacketing. All gas and liquid lines have keyed non-congruent fittings to avoid inadvertent cross-plumbing.
  • Pipes are run at a set of different standard heights radially and circumferentially to avoid interference.
  • Telecommunications are performed with wifi/cell coverage on the front end, satellite for remote operations, and data centers for backend compute.
  • Life support systems are employed at varying scales and levels of complexity to enable scalability and systemic redundancy. For example, every isolatable space, no matter how small, has a rated life support capacity to function as an air shelter. Life support systems in smaller areas, such as rooms in dwellings or vehicles, are necessarily simpler in form. Distributed, multi-system overcapacity is essential to avoid single points of failure or unacceptable levels of centralization of control.

Image: Diagram of notional city plan, showing industrial bays radiating from the central core. All functions can be independently resized as needed with minimal disruption.

Strategy for achieving self-sufficiency

What does self-sufficiency look like for a city on Mars? A popular image of self-sufficiency provides a rugged, capable pioneer with a small plot and some animals building themselves up from nothing. While easy to articulate, a Mars city cannot bootstrap like this, because the environment is too hostile. 

Environmental hostility is a way of thinking about what will kill people and how quickly. Astronauts, oil rig divers, and mountain climbers all work in hostile environments, depending on advanced technology and rigorous procedural problem solving to stay alive.

While the Mars city encloses a large enough volume that the inhabitants can move around unencumbered by spacesuits, the system as a whole still embodies precarious advanced technology that is not capable of regenerating itself by default. Therefore, all the shiny life-supporting widgets must either last forever, be readily importable, or readily replaced. This is a tall order.

On Earth, with its habitable environment and billions of people, there are only five nation states that have achieved sufficiently advanced industry to “make everything”. They are China, Japan, South Korea, Germany, and the USA. A Mars city doesn’t need to make fighter jets but they need to make nearly everything that goes into one, including advanced robotics, computers, plastics, metals, composites, and tooling for advanced manufacturing.

While an early Mars base must import nearly everything, a more complete city necessarily has to make more things locally. How can we think about prioritizing local manufacturing?

Generally speaking, local production favors bulk raw materials that are both easy to make and must be provided in large quantities. Conversely, import favors complex technology that embodies large quantities of energy and labor, such as microchips. With a handful of exceptions, products with a lower cost per kilogram (as a proxy for manufacturing difficulty) and higher use rate would be made locally first. Market pricing mechanisms allow for “natural” prioritization without rigid central planning.

Local production for any given product begins with prototyping and moves into mass production and then fully automated production. It is critical that localizing production consumes proportionately less resources, in particular human labor, than they create.

Graph: Some industrial products by cost density and US per capita production. Asterisks mark products (water, methane, oxygen, CO2) with substantially different sources and usage patterns on Mars. Self-sufficiency starts at the bottom right (water, rubble) and moves towards the upper left (flash memory, morphine). Regions for Mars production, import, and export at a million people as discussed in the economics section below.

As an example, early production favors oxygen, nitrogen, water, electricity, methane, plastic feedstocks, masonry, and other bulk commodities that can be produced on Mars with only gas or liquid water feedstocks, or unprocessed dirt. These materials do not require dedicated facilities for remote ore extraction and processing.

Later production favors food (expensive in a place where arable land must be made from scratch), metals (some recycled from retired spacecraft), and a wide range of industrial chemicals.

Still later, secondary production (manufacturing) processes raw materials into discrete products beginning with mass-intensive structural parts of machines (booms, fasteners) and eventually trending into electronics and integrated circuits.

It cannot be overstated how difficult and ambitious this program is. Many nations on Earth have tried and failed to achieve this, despite higher populations and better resources. There are, however, a few considerations which could affect the overall difficulty of the enterprise.

Building for growth

As the Mars city grows, demand for products also grows. Factories are built to continually increase productivity, with design incorporating room for growth and for greater automation. There is a fundamental limit to the rate at which a human being can perform manual tasks, so over time, human workers are separated from the actual products by steadily increasing layers of automation, or robotic abstraction. In mature industries, even factory construction and machine calibration are automated or remotely operated from Earth.

Simplified and compacted industrial stack

The McMaster-Carr catalogue lists 550,000 discrete items. Beyond a certain point, provision of additional part diversity suffers diminishing returns. Mars-focused products are sourced from a simplified parts catalogue. Additionally, sub-industries that are irrelevant to Mars, such as coal steam turbines or container ship construction, need not be built.

Path dependency, lock in, and technical debt

More than a few megaprojects and industrial sectors have run into severe problems due to path dependency, lock in, and technical debt. This occurs when an early and apparently unimportant decision has unintended consequences that are both difficult to correct and difficult to live with. As an example, the internet was designed in an era when everyone on it knew everyone else, so security and attribution were afterthoughts at best. This has left our entire society with endless security issues in critical infrastructure! 

Technical debt is one of the canonical “hard problems”, since it affects every industry to some degree and there is no easy answer. That said, a degree of mindfulness when performing systems engineering may help prevent the deepest of regrets.

One concrete example is deciding what atmosphere to operate the city under. A lower pressure atmosphere reduces pressure loads in structures, as well as decompression effects when using space suits. On the other hand, it transports heat less effectively, meaning that every cooling fan and system has to be made bigger to dissipate unwanted waste heat.

Proficiency goal for one million people

A Mars city of a million people is able to produce nearly every resource it needs to survive. By mass, the cargo manifest is human immigrants, by a super-majority. That is to say, per immigrant cargo allotment has shrunk from perhaps 10 T at the outset to less than 10 kg, a reduction of a factor of 1000. By mass, more than 99.9% of products and resources used by Martians are locally sourced. 

This means that robust local surpluses exist for all gases, liquids, materials, food, water, precision machinery, vehicles, structures, bulkheads, infrastructure of all kinds, chemicals, bulk electronics parts (actuators, sensors, capacitors, circuit boards, batteries, solar panels, etc) and some local production of integrated circuits such as a generic rad hard x86 processor, FPGA, and flash memory unit. 

Imports, therefore, comprise primarily luxury goods and consumer computing devices such as tablets and mobile phones, along with a range of pharmaceuticals that have very low use rates.

Of the million people, perhaps half are devoted to tertiary services and facilitation ensuring that labor remains specialized and efficiently allocated across all sectors. Not everyone works in a factory or mine.

To put this into perspective, near autarky with a million people on Mars implies an improvement in per-capita disposition of resources equivalent to all the advances since the industrial revolution, again. No miracles are required, only a lot of hard work. Not only is this physically possible, it can be achieved from our present technological state with only steady incremental advances.

Economic design

Despite my strong temptation to design and specify a command economy, the Mars city has adapted best to a regulated open market and free enterprise. In any system beyond some small critical size, the distributed asynchronous mechanisms of capitalist buyer’s market economies scale much better than the alternatives, such as any form of centralized control or artificial pricing. 

That said, the thousands of ambitious entrepreneurs in the city face some unusual conditions which merit discussion.

Refining the success condition

The Mars city is economically successful if its economy is capable of sustaining the industrial capacity to exceed the material wants and needs of the people who live there. That is, it has achieved robust prosperity despite its relatively small size, hostile external environment, and high shipping costs from the nearest developed market.

Economic success of a human city in space is sometimes defined as achieving wild profitability for traditional Earth-based speculative investors. Without the allure of a get-rich-quick scheme, we are told, it will be impossible to fund this sort of development. I may be a member of a small minority that believes that no-one who wants to make an easy buck should look to space, whether it be mining the Moon, asteroids, or building a profitable space hotel.

This is why net profitability is a counterproductive success condition. While building a city on Mars can’t generate net wealth for all Earth-based investors, it is meaningful to ask how much progress might be made for a given investment. Indeed, nearly all space exploration, whether using rockets or telescopes, depends on either private philanthropy or government expenditure. Since von Braun’s Das Marsprojekt, even exploration missions to Mars have come with an impossibly steep price tag. Mars Direct showed how to reduce the cost by perhaps three orders of magnitude, while SpaceX’s reusable Starship architecture aims for a further improvement of the same scale. If these transportation innovations are successful, Vision 2040 could be built for a total cost measured in hundreds of billions of dollars over several decades, which is affordable enough for a global civilization in full bloom.

What are the implications of the given transport costs of $500/kg to Mars and $200/kg from Mars?

It is worthwhile to explore what assumptions underlie the competition specifications.

One passenger, their luggage and life support supplies weigh about 400kg, implying a one-way ticket cost of $200k. Once on Mars, an adult human consumes about a tonne of food, water, and air per year. If imported, this would cost $500k, but around 99.9% of this is locally produced/recycled by the time the population reaches a million people. Finally, the cost saving of someone staying on another 26 months until the next launch window is about $300k. All together, this implies that supporting workers on Mars is about ten times more expensive than is typical on Earth. That is, even with a million people on Mars their net productivity has to be at least ten times greater than one would reasonably expect on Earth. This also meshes nicely with expectations for the level of automation and self-sufficiency, since the Mars city must have an industrial stack and product manufacturing diversity more in line with a country of a hundred million people. 

Regarding self-sufficiency, a cargo import cost of $500/kg is prohibitive for commodities that cost much less than this (such as water and food) but relatively insignificant for products that cost much more. These include industrial machinery, certain chemicals, long lived radioactive isotopes, integrated circuits, and other stuff that routinely travels by air courier. Nevertheless, if a certain product can be obtained more cheaply from local manufacturers, imports necessarily diminish. Local manufacturers get, in effect, a $500/kg import duty which has to be balanced against a 10x human labor cost increase. Robotic labor on Mars is more expensive than on Earth but relatively cheaper than human labor. This means local manufacturing costs are between one and ten times more expensive than on Earth, depending on the process and material inputs. As an example, a product that requires $55/kg of labor input on Earth would cost $550 to produce on Mars. Given import costs, production on Earth or Mars is equally favored at this price. Since relatively few tangible products require that much hand labor, a Mars city of a million people makes nearly everything locally.

Let’s explore cost implications for exports. For a SpaceX Starship flying 100 T to Mars, approximately 15,000 T of propellant would be burned, costing perhaps $10m, or 20% of the overall ticket price. The rest covers overhead, including amortization of the Starship’s manufacture. A returning Starship can return about 20T of cargo for 1200T of propellant burned, requiring about 500 megawatt-days (that is, one megawatt for 500 days) of electricity to synthesize from CO2 and water. If 50% of the return ticket cost is fuel, then wholesale electricity prices on Mars are about 16c/kWh, comparable to electricity prices in California in 2010 before solar crushed everything. Today solar costs are around 2c/kWh including storage, so we find that Mars power cost increase is consistent with the 10x labor price increase. This is slightly troubling as labor scarcity would prefer to exploit electric power where available. However, for all but the most power intensive processes the electricity cost is not very important. These power intensive processes include desalination, electrolysis, aluminum smelting, sodium production, and in particular propellant production, which is why shipping stuff back to Earth remains expensive despite other advances.

The terms of the competition invite contestants to examine the potential for exports. Broadly these may be divided between physical products, which need to be physically transported, and knowledge products, which may be sent by radio or laser.

Let’s examine the trade balance. A million people on Mars each costing $500k/year implies a total GDP of $500b. If 5% of GDP ($25b) is spent on imports with an average price of $1500/kg including shipping, then the city imports 16,000T of cargo (160 Starships, in addition to perhaps triple that number carrying externally-funded migrants) per year. Since most humans on Mars stay, let’s say 500 Starships are available per year for exports, with a total capacity of 10,000 T, and costing $2b to fuel. If the city aims to earn back that $25b in exports then it needs to sell $27b of goods, implying a value density of $2700/kg. This is substantially higher than the make/buy threshold of $55/kg discussed above. This disparity rules out essentially any commodity product, as they are available more cheaply on Earth. As far as material exports go, the city needs Mars-unique branded premium goods or, potentially, rare minerals of highly unusual local abundance.

At about $50,000/kg, technically gold and platinum-group metals could qualify, if and only if their local production costs were, for some reason, far lower than on Earth, and Earth demand remains strong. For example, Earth’s annual production of gold is about 2500 T. Even if Mars could produce gold at a competitive price, it couldn’t export more than about 250 T/year without demand elasticity causing the price to fall. 

The implications of this discussion for local production, import, and export are displayed pictorially in the product graph in the previous section.

As an aside, fueling 700 Starships a year (or 1500 per launch window) would require about a gigawatt of electricity. Assuming that fuel production consumes 30% of the city’s power, the total area of solar panels is about 100 square kilometers, or 100 square meters and 3 kW per person. (This is assuming 500 W/m^2 insolation at Mars, 20% panel efficiency, 0.3 capacity factor. Roughly twice the US per-capita consumption.)

The other potential export is information. Traditional suggestions include Martian IP to be licensed on Earth. In general, however, the high relative cost of labor is a strong forcing function for all non-essential knowledge labor to be non-local. That is, for any task that doesn’t require either physical access or temporal immediacy, Martian companies would be well served to hire contractors on Earth and import (not export!) their software and databases. In the early days of the city, large dedicated teams on Earth would support individuals on Mars to optimize their productivity and tele-operate machinery. By the time the city reached a million people, this had largely given way to free enterprise. Meanwhile the falling cost of Mars labor justifies the hiring of Earth-based assistants, but not necessarily individualized teams. Another model might be distributed organizations where Earth-based engineers develop products inspired by observation and operation of facilities on Mars, then share profits as a form of ongoing R&D investment.

How does the Mars economy work?

The Mars economy is structured similarly to any country on Earth that specializes in mining and manufacturing, such as Germany, South Korea, Japan, or the US. Diverse capital markets with sophisticated risk management enable aggressive investment and expansion of critical infrastructure. There is no need to reinvent the wheel here.

Fiscal policy is streamlined to maximize the exchange of value mediated by money, with the US dollar being an obvious and safe choice. Inevitable market failures are managed through a combination of reactive taxation and direct subsidy. Policy is improved based on quantitative assessment of effectiveness, rather than ideological theorizing. A successful economic policy is mostly invisible, while unnecessary complexity increases transactional overhead and impedes the flow of value.

Much ink has been spilled speculating as to the ultimate source of funding for a Mars city. Who pays for it all? It is worth remembering that the most valuable thing migrants bring to Mars is not the paper in their wallets, but the skills in their mind, the strength in their body, and the flinty determination in their eye. Enable the economic and physical mechanisms to permit mass migration and the rest will follow.

Social/cultural/political aspects

What should Martian society be like? What is the human experience on Mars? Some visions of life on the frontier foresee abridgment of human freedom and hard, short lives. Yet throughout history economic growth and productivity has always walked hand in hand with facilitation of individual excellence for all. The single greatest asset of Vision 2040 are the people who choose to build it, and their celebration of human capability reassures us that life on Mars, while difficult in some ways, is exciting and empowering.

Social consequences of self-selective migration

A million people on Mars, and every single one surmounted substantial hurdles to be there. Like other elite self-selective communities with a high barrier to entry, Mars society is shaped by ambition, celebration of achievement, and a strong work ethic. From the Peace Corps to Everest, SEAL Team Six to Grad School, there is something different and special about existing in a community where drive and determination replace sloth and angst. The same on Mars, only more so.

Moving to Mars and building a new world isn’t for everyone. But for people who live for a challenge and love the frontier, it is the only place to be. It is the most focused concentration of passion and feverish innovation in the history of humanity. A place for doers to get stuff done unencumbered by the legacy of fifty centuries of business as usual.

Freedom of labor

Everyone who moves to Mars has both skills and the ability to develop them, but continuing development presents two unusual challenges. First, maintaining high morale and productivity in a workforce that cannot “rotate home” requires freedom to vary employment, alternate gigs, and develop new skills on the job. Second, very few jobs on Mars do not change rapidly as automation and AI steadily consume the industrial stack. Therefore, the design of the collectively lived environment must be continually iterated through open contributions to maximize learning and performance improvement. This means the normalization of hacking reality to enable technological miracles.

Any Mars city will suffer a continuing labor shortage, ensuring that employers compete to attract and retain the best workers. By removing barriers to competition in the labor market, we can ensure optimal alignment between interests and needs. 


High costs drive labor outsourcing to Earth, where air is free. Any task that doesn’t require physical proximity or real-time interaction are mostly done by professionals on Earth. Such jobs include software development, planning, remote operation of mines and other machinery, and environmental monitoring. People on Mars work closely with assistants on Earth who monitor their work environment and implement constant improvements via augmented reality interfaces or backend software improvements. Imagine waking up to find that “the fairies” have fixed the previous day’s problems!


Ensuring long term high productivity of the labor force precludes 20 hour work days, so Martians have plenty of time to engage in non-work activities. Whether art, music, sports, cooking, literature, or any of a million other things, the Martian do-ocracy enables well-motivated people to build whatever they want or need to perform their activities. As a result, the culture is explosively creative and varied, like Burning Man but a hundred times bigger. 

With readily available Kevlar-reinforced ETFE to create pressurized volumes, there’s no reason to cram everyone together. Most people live in dense walkable neighborhoods to facilitate easy access to the needs of life, but there are few practical limitations on pressurized volume. Fly a section of roof a kilometer high, plant a forest of giant redwoods, and export Martian lumber at $3000/kg. Throw a tent over a nearby mountain, keep it cool, and operate a ski slope. Pressurize a volume to 4 bar and have human-powered ornithopters fly in the low gravity, while executing a game of 3D golf. Perfect a Martian pizza recipe. Build a ranch and farm mutant dwarf buffalo. With a nearly automated industrial stack, an embarrassing surplus of nearly all material resources and an unbuilt planet, there’s no reason to think small or slow.


Science fiction city design is always a good opportunity to flog some personal hobby horse and governance is no exception. It’s always easier to identify problems from a distance than to do the dirty work of actually building peaceful consensus among us lightly-evolved apes. So instead of dictating how Martian governance functions on some untested theoretical level, I will instead interrogate the very notion of government. 

Why have one? What sort of functions does leadership perform?

While an early Mars base can exist as a self-organizing anarchic collective much like scientific research stations in Antarctica, as organizations grow their management difficulties also grow.

The functions of industrial development being largely devolved to subject-specific corporations, the responsibility of government is to safeguard peace and prosperity. This requires: 

  • Performing diplomacy with sovereign representatives on Earth and other bodies.
  • Regulation/legislation, with an emphasis on continual compression of the code.
  • Promoting liberty and administering justice.
  • Promotion of prosperity.
  • Provision of public goods.
  • Self-regulation and improvement.

In accordance with enlightenment views, the leadership should govern by consent and maintain accountability for actions performed in the public service.

Of the six major functions listed above, the one that varies most from typical political or corporate governance models is self regulation and improvement. As the Mars city grows the demands on governance continually change. It is highly unlikely that an optimal governance structure can be generated by induction on the first try. Instead, mechanisms and practices must be continually tweaked and updated to ensure that the government remains a nimble servant of the public’s needs. The single most important function of the government is to maintain and improve the mechanism by which it improves itself.

I do not regard myself as an expert on systems of governance but I can imagine worse places to start than a bicameral representative democracy. It has, afterall, worked in Iceland for nearly 1100 years.

Corporate governance is intentionally not prescribed. The strongest political and economic systems are syncretic, which is to say diverse and inclusive. For example, some industrial functions may be well-served by a traditional corporate governance structure, while others may function best as worker-owned cooperatives following the Mondragon model, or anything in between. 

Better than Earth?

What is the success condition? Beyond industrial autarky and meeting material needs, how does a city of a million people know they’ve “made it”? Certainly migration appeal turns steadily more mainstream, but consider instead the lived experience of Martian-born children.

There is no reason to suppose that children could not exist and have happy lives on Mars from the very earliest days, but as far as labor goes, importation is much faster and cheaper, on Mars, than making new humans from scratch. Indeed, even on Earth it is generally considered easier to hire people to perform tasks than to make them oneself. We no longer have children to ensure financial security and care in old age. Like the Mars city itself, the objective is to perform a worthy activity and minimize accompanying financial losses, rather than execute with the expectation of profit for external investors. It turns out that the set of things that make money overlaps incompletely with the set of human activities that are worthwhile. 

Thus the economic senselessness of rearing children on Mars is a microcosm of the overall economic senselessness of building the Mars city in the first place. Since we agree that a Mars city must be built despite its inevitable consumption of enormous quantities of treasure, we may sensibly ask: Why is Mars the best place to be a child?

Children are the future. A child raised today on Earth may not believe that the best of human civilization is yet to come. The old frontiers are closed, the population is rapidly aging, and many institutions are calcified around a stable consensus view of the way business is done. As beings that grow into our future, our future on Earth is not as unlimited as it once was.

A child growing up on Mars is as separated from authentic wilderness as any Earthborn city kid, but they have the benefit of being around adults who believe powerfully in the future, knowing that their world needs them and has a meaningful place for them. 

Who could take that from a child?


A nearly self-sufficient city of a million people on Mars is a stupendously ambitious project. Technically, economically, and socially it is possible, that is to say, not forbidden by the laws of physics. But building the Mars city “Vision 2040” requires more than physical possibility. It needs millions of people to make this project their life’s work. And that requires something else.

All successful large scale collaborative projects obviously had sufficient technical execution. But they also evince love and celebration of beauty. Wikipedia, Linux, the Internet, the American experiment. It is not enough to be a good idea, or to assemble some patchwork constituency who kind of like it. It has to also inspire the profoundly human response of collective nurturing.

We have come to the Vision 2040 aesthetic. Vision 2040 is both a physical place and a powerful idea. It may be perceived through the senses and through the mind. These factors reinforce harmoniously to invoke a sense of the numinous. A sense of vertigo, that humanity is collectively teetering on the brink of a significant moment, a birth of history and a death of our confinement to the planet of our origin. It is this feeling that motivates my Terraformed Mars art project, where I have made planetary scale renders of a Mars with life and water.

The power of this vision has been employed by Elon Musk in his recruitment of idealistic genius engineers at both Tesla and SpaceX, but for Vision 2040, it must go further. Moving to Mars is a significantly bigger and more permanent commitment than moving to a foreign country. Vision 2040 exerts powerful magnetism on the ambitious and idealistic. At $200,000 per ticket, the pitch has to be better than “maybe come to Mars, maybe you won’t die”. 

Beauty in form – the physical environment

Let’s imagine the process of becoming a Mars migrant. 

We have a good life on Earth among family and friends, but we remember our earliest memory of contemplating the stars on a chill fall evening. Like everyone, we’ve followed the last two decades of progress on building a Mars base. Early setbacks. Improbable victories. Now, it looks like it’s going to stick. Little by little, we realize that our vision for Mars includes us being there.

Of course, even now private migration is only just possible. Get recruited, sell everything, get a loan. We don’t remember ever having seen that much money, let alone spending it on a single thing. It’s like a briefcase stuffed with cash.

Not that the money matters, not really. Plenty to be made on the other end, or in any number of jobs back here on Earth. Beyond a certain point, additional money just buys anxiety rather than freedom. No, the real investment isn’t a distillation of personal possessions and a tearful goodbye. It’s putting our body and mind out there, adding our voice to the swelling chorus filling this splinter of humanity on the dusty Arcadia Planitia. 

The launch window approaches, a relentless schedule of tasks necessary to shut down a life at 1 AU and restart it somewhat further out. Our friend drives us to the airport, we give them our car keys as we haul a carefully weighed duffel of mostly old teeshirts into the bowels of the transport machine. Will we meet again?

The usual buffeting as the suborbital electric jet drops through the sound barrier over Brownsville. Seemingly minutes later up the elevator, across the gantry, and into the Starship through its scorched and still warm hatch. So, that’s what Earth looks like from space. Surprisingly shiny in the sun. Smooth at this scale. Then four months in deep space, about which said the less the better.

Mars, a bright star, grows to a turning disk, the city lights just visible beneath the dawn terminator. Patches of ice on the higher mountains. A few moments of tension, then mere minutes of noise and force. Landing with a bump. 

A spiral shaped vehicle access gantry locks on, a crowded corridor of faces, small spaces reflected in myopic eyes. Stepping over the threshold blinking into the day’s red light, we are refugees from Plato’s cave. 

Tented roads stretch from the Starport back towards a crescent shaped city complex, with various satellite facilities and enormous fields of solar panels. A hint of green beneath the shiny plastic in the distance. The road takes us past older landing pads and starships, now being subsumed into the growing city.

The road tent passes through a steel bulkhead and opens up into a cavernous volume filled with giant redwoods, their dark evergreen needles fluttering noiselessly as we, the newest Martians, stare.

We alight at a central plaza surrounded by four and five story structures, windows open to the mild air. We had arranged living quarters in a modern apartment. A compact and cozy place with a decent view and common facilities for eating and entertainment. It looks like it had been finished about two weeks before, and it probably had. Our shift began that afternoon, so we resolved to walk there by the least direct route. Light gravity feels like flying. On the way we grabbed a tasty snack from a venerable looking food truck emblazoned with the proud words “established in 2027”.

At work, a placard reminded us that doubling productivity in two years requires only 2.7% improvement per month, or 0.1% per day. Better get going!

Beauty in function – the power of an idea

We’ve lived on Mars for 500 days now. They’ve passed in a blur, and yet in that time the city has changed noticeably. One section is kept how it was at the beginning, and sometimes a bewhiskered old timer tells war stories about how it used to be, back when the dirt beneath our feet was exposed to the vacuum of space. 

We thought this day would be a big one. A go/nogo decision whether to return to Earth or stay at least another two years until the next launch window. Most employers offer a rotation bonus to stay, because transport is so expensive, but we didn’t give it a second thought.

We’ve only had one birthday since leaving Earth and yet that seems like a previous life. Our whole life on Earth could be fit into about two weeks here. We see tangible evidence of our progress as we wrest order and life from the chaos that has been here since the beginning of the universe.

Less evangelizing. It’s not as obvious if you’re not living it. What is happening here? A million people – more than we will ever meet, all moving to the same rhythm. The challenge is clear, it confronts us every day and taunts us. Will we establish a permanent foothold or will we slip into the void?

We monitor progress in all kinds of ways. The most concrete is the actuarial table which shows us how long it would take to run out of essential supplies in the event of supply chain degradation. Right now, we could survive 10% degradation indefinitely, and 100% degradation for seven Earth years. And that’s the best it has ever been. 

Can you imagine how it feels to be in this position? On the one hand, only two doses of bad luck from oblivion. And on the other, complete autonomy and empowerment to do whatever we can to help the situation. When I look around at the million here solving problems every day I have a tangible grasp of the inherent capability of humanity. We have the audacity to abandon the dysfunctional old ways and try something new. Experiment. Unleash creativity.

It is hard to explain but easy to see. Look around. It turns out that the frozen dead Martian soil was a fertile substrate for our dreams. That’s why four of my old Earth friends are already on their way.


All images, graphs, and data © the author unless otherwise attributed. These were originally footnotes – check the pdf for context.

For more on industrialization, check out my Mars Society 2018 talk

Domes suck:

Contrary to popular belief, lack of radiation shielding is not a showstopper. 

Inflatable plane:

Highest density human habitation ever:

For Earth-Mars internet synchronization, see Mars Colonies. F. Crossman (ed). The Mars Society, 2019. p. 163. (J. Greenblatt and A. Rao.)

Mole, A, and Frank Williams. Baseline Design for a Mars Colony. Website: p. 7, for specifics on Mars nuclear power.

Most likely obtained from a “rodwell” melted into subsurface ice. Wooster, Paul. Personal communication, 2020. See also:

For incompatible keyed interconnections, see e.g. 

For a thorough Mars city simulator focused on material cycles, see SIMOC:

Mars Colonies (ibid), 2019 p. 89. (C. Plevyak and A. Douglas).

For a great summary of ore processing chemistry, see: Mars Colonies (ibid), 2019. pp. 57-66. (J. D. Little).

Notable autarky failures include Cuba, Albania, North Korea, Cambodia, Brazil, Yugoslavia, and Romania. Most didn’t even get close, but all had agriculture, air, warmth, and more than a million people.

McMaster-Carr Catalogue:,98%25%20of%20items%20from%20stock.

Simplified industrial parts catalog suggested by Marinova, Margarita. Personal communication, 2020.

Aldrin, B. Mission to Mars. National Geographic, 2013. p. 176.

Economic difficulty of exploiting space resources:

MacDonald, Alexander. The Long Space Age: The Economic Origins of Space Exploration from Colonial America to the Cold War. Yale University Press, 2017.

Werner von Braun’s “Mars Project”

Zubrin, R and R. Wagner. The Case for Mars. Touchstone, 1996. p. 37.

For a discussion of MarsSpec standardization, see: Mars Colonies (ibid), 2019. p. 141 (K. Nebergall). See also 

Mars working conditions:

For historical examples of exceptional innovation and speed of execution, see:

Less conventional commercial structures:

Zubrin, R. Entering Space. Putnam, 1999. p. 114.

Explored in Mars Colonies (ibid), 2019. pp. 193-194. (A. Dworzanczyk).

But why?

The power of love:


Explored in Mars Colonies (ibid), 2019. p. 423. (S. Schur). Housing built according to demand.

Zubrin, R. The Case for Space. Prometheus, 2019. p. 116.

21 thoughts on “Vision 2040: The first million on Mars

  1. That canopy design idea is fascinating, and I wonder why I’ve haven’t seen it get more attention versus domes or buried structures. Reminds me of the “roof world” concept with paraterraforming, and the extra space would be valuable (and useful in another way – the more air above you in the canopy, the more shielding you have from radiation).

    Although I think by the time you’ve got a million people on Mars or so, you probably won’t be flying to Mars in SpaceX starships anymore. If there’s the demand for it, you can build bigger space-only interplanetary spacecraft and refuel them on both ends.


      1. I suppose that would make sense. If they’ve already built offshore platforms for large Starships, then you’ve already mitigated some of the concerns of launching very large rockets.

        One other concern I did have was about the perchlorate salts. You mention in one of your other essays that they dissolve in water, but you then have to run the water through a filter to get rid of the perchlorates – the water doesn’t destroy them. You’d have to very thoroughly soak whatever soil is under the plastic canopy, or use an alternative method to break it down (either heat or biology).


  2. Yay, a fun big-picture post.

    I don’t expect any animal products. At first they will be unaffordable, and later they will be non- traditional.

    I don’t really see why the core area would be crescent shaped.

    Domes. Three of them. I still think some monumental architecture is going to be worth having. The only thing a very young Martian city can sell to Earth is tickets, and some iconic things that are visible on approach are a good bet for branding. Of course it doesn’t have to be domes, but they are visually striking. Unlike a giant air mattress.

    I think that keeping dust out is going to be the main function of the outermost layer, in a lot of minimally pressurized area. Machines will outnumber people, and they don’t need a human-breathable atmosphere.

    More later.


    1. Crescent shape is a nice way to ensure compact, walkable urban architecture and a sense of open space. I like the idea of the city enclosing wilderness that’s kept more or less as it was.


  3. Quite a read. I like it! My thoughts are:

    1. A radiating circle may be best for enlarging sections / uninhibited growth but isn’t necessarily that efficient (is it?) for transport / communication between the spokes – the distances get very big quickly. But maybe that is the case with any layout – ie some things can be closer but others will be further apart what ever approach is taken – except perhaps multiple smaller cities. Ie set up on a disaggregated smaller town basis… 2. Children. No group of people of any size will not start having / wanting children. I don’t see it as something that will be controllable in a sense. And as you say so eloquently – why would you not have kids on Mars! In fact the closer the Martian society reflects the best of human civilization and society the better and that includes children. I happen to think that most people improve through the experience / responsibility. 3. Whether self-selected/ paying or not, I suspect that a Martian city will have many and possibly some new levels of problem / crime, as on earth. All those driven people. Not all will be bright/sane/etc. I felt your blog skated lightly over this but maybe that is sensible.



  4. I don’t find vast forests or other wide open spaces compelling, but I do like the idea of having a grove of trees. With the lower gravity, the trees could be taller than is possible in 9.8 m/s2. If the Martians want to be really extravagant, they could have high pressure for extra-dense air, so that people can fly among the trees in pedal-powered personal aircraft.


  5. Today I think that the launch window will be obsolete before the Martian population reaches a million. Some things will go from Earth to Mars on slower trajectories. Some things will go both directions on fast trajectories that cost an amount of delta v that now seems impossible. We will have some non-rocket options. Maybe we will catch and reuse reaction mass. Maybe we will have mass drivers at all the Lagrange points: Earth-moon, Earth-sun, Mars, Venus, Jupiter, capable of throwing payloads at speeds where interplanetary travel is essentially line-of-sight, with precision that lets them hit the tip of a rotating tether from three AU away.

    In other words, a million sounds to me like a lot of people, so today I’m guessing that it won’t happen all that quickly.


  6. Wow, I had no idea about any of this I just came across the article while doing research for a client of mine that’s a solar manufacturer. I’ll have do more research, I’m so interested! Thanks for sharing.


  7. Why go back down a Hole??
    Everything is more DIFFICULT on Mars than in space
    Why not go to one of “Earth’s other Moons”
    100 Billion tons of raw materials
    Use tethers and spin your habitat for gravity
    Lots of material for radiation shielding and manufacturing
    Your power supply (sun) stays in one place and does not run away and hide
    Everything is easier on Cruithne than on Mars


    1. Except recruiting.

      What can a newly founded city in space profitably sell on Earth? Tickets. How many millionaires can you find who want to fund a trip to Mars, and how many who want to fund one to Cruithne?


  8. Re. Graph of what materials are importable/exportable, in particular the part about exporting to Earth:
    Assuming there is a market for any type of megastructures in space, once a martian city is quite self-sufficient it could sell those to Earth – as getting them to many places in the solar system from Mars is not only cheaper (lower delta-v), but also easier, since thinner atmosphere should make it easier to get bigger parts out of it.
    This is probably not a great market to compete in, since existence of a Martian city is a proof that getting things to space from Earth got a lot cheaper, but it is a market that sufficiently developed Mars can theoretically compete in quite well against Earth.

    Liked by 1 person

  9. Today I imagine that the initial layout of the city will be shaped by the difficulty of surface transportation on Mars.

    At first, there won’t be a space port. The first few starships will land on relatively flat ground, with almost no site preparation. There will be preliminary surveys, done by fully autonomous rovers and satellites, between launch window zero and launch window one. There will be some work done by prototypes testing out what can be done with tele-robotic systems that tolerate the full light-speed lag from operators on Earth. But the work that can be done before the arrival of the first Martians will be very limited.

    They’ll have a bunch of starships land as close together as possible. That’s the distance where a failed landing of one starship won’t quite destroy the nearest previously landed one. Then they’ll do the stuff that uses the contents of each starship where that starship is. If they pack the starships in as closely as possible, the city will be laid out on a hexagonal grid.

    I still don’t believe in distant mines. I think it will be easier to extract everything from ordinary rock (along with air and ice) than to ship stuff without oceans and ships.

    I think gritty regolith will make components wear out if they try to transport stuff over the surface. Once they have enough processes going that they can make all the wear-exposed parts locally, that’s ok. But when they’re still dealing with Earth-to-Mars delays and transport costs for replacement parts, they will plan around most of the high-wear activities as much as possible.

    A few starships will be closer together. The ones landed first will have cargo that will have decent salvage value if there’s some unplanned geobraking next door, instead of having it be a total loss.

    Later development will be done in ways that make sense for building on to earlier facilities. I don’t foresee neat layout like a crescent with radiating sectors. I imagine stuff accreting around other stuff, in ways that connect, building by building, rather than following any grand scheme.


  10. That might be true for stuff directly needed at Mars, but the opposite doesn’t have to hold. The average engineer at Facebook now makes something like $460k. Aside from a few communication difficulties, the company couldn’t care less that some might want to work from another planet. The same holds for a lot of the tech space. In fact, that combined with those companies wanting to participate in the new society might push a surprising amount of work to Mars that makes its revenue on Earth.


  11. The canopy idea needs a product name or concept name so you can refer to more easily and specifically. Without resorting to neologisms, I’ll offer “baldaquin”, but it’s your word to sell.


  12. Thinking more about it, I’m not sure if I understand the electricity price calculation.
    Price is where the supply and demand curves meet. We know that given SpaceX’s fixed costs, they would like to buy electricity at 16c/kWh or less to keep costs equal to Earth’s. But that doesn’t mean there will be supply at that price, given how much it could cost to produce and/or manufacture panels and batteries. It could be that electricity is simply much more expensive than that.
    The provided 2c/kWh is promising. Solar panels are lightweight but I don’t know how much batteries would be needed. It would be:

    Solar_Power * (NoSunlight_Time + Night_time*Night_usage) = Storage needed

    Solar_Power: output of the panels;
    NoSunlight_Time: Time during the day when there is no sun because of dust storms;
    Night_time*Night_usage: Night hours times a multiplicative factor to account to how much less energy is used at night.

    Preferably all during winter.


  13. My other comment shows up but this one doesn’t:

    I loved the calculation of how much electricity will cost!

    How do you think the tax regime will be? The Mars government obviously should have low taxes, but at least at first I expect Mars-city to be part of the US. Do you think the US government will make Mars-city a federal-tax free zone?

    A few thoughts:
    You mention how the Mars economy would favor goods and services that can’t be done remotely. Two things come to mind:
    1. Crytocurrency mining, taking chips to Mars is cost effective and while you might not have the highest margin you may still have a profit. It’s also cheap to transport back to earth, albeit the 10 minutes block interval of bitcoin might be a problem at first;
    2. Film production, a capital intensive enterprise that is hard to replicate on Earth (as long as the script takes place on Mars). The equipment can be divided in expensive cost effective to transport cameras, heavy equipment such as cranes and cheap stuff like props both of which can be manufactured o Mars.

    Martian children:
    As you pointed, maybe not at first, but eventually the Martian people will have a lot of free time due to the labor/intellect intensive nature of their jobs. That and I expect Martians for at least some time to have a joyful outlook of life.
    Those two things combined seem like the perfect combo for baby making, see post-war baby-booms and economic-prosperity baby-booms.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s