If you can do something in space, you invariably can do it more cheaply and easily on Earth. This isn’t only a financial problem or a dying-in-space problem—space proposals are environmentally dubious too. Rockets are about 80% propellant, all of which is burned on the ride to space. Advocates sometimes note that in principle, propellant can be generated using only renewable energy, but in practice, this isn’t yet happening. Adding vast numbers of megarockets to the needs of a grid already struggling to go green seems, to say the least, more likely to contribute to climate change than solve it.

Despite what many of its proponents believe, space is neither a solution to current environmental challenges, nor an escape hatch in case of environmental calamity. We’re here to cast some skepticism on the most popular ideas, not because it’s what we want to believe, but it’s where the data led us. 

This environmental argument for space settlement was popular in the 1970s. In its most basic form, the logic is this: To ease the load on Earth’s biosphere, we should reduce the amount of bio—namely, Homo sapiens—in the sphere. 

The challenge is, well, getting people to space is daunting. The largest rocket that might be available in the next few years is SpaceX’s Starship. Still in the testing phase, the spacecraft might eventually carry 100 people per launch. In 2022, Earth added about 80 million people. To simply hold Earth’s population steady, rockets would need to send up 220,000 volunteers per day. How much would that cost? And does 2,200 daily launches of skyscraper-sized vehicles mostly made of rocket propellant seem apt to reduce the burden on the biosphere?  

Alright, so moving the humans will be tricky. Perhaps we should consider a different approach—just move heavy industry.

Move mining and manufacturing out of our atmosphere and dispose of their nasty byproducts in the vast landfill that is the solar system. As former Amazon CEO and Blue Origin founder Jeff Bezos says, “Earth will be zoned residential and light industrial.” Is this plausible? Let’s consider just one heavy industrial product, which is a serious contributor to climate change: cement. Cement manufacturing contributes at least 8% of the planet’s carbon dioxide emissions. Just imagine if we could make it off planet.                             

Technically, most of the components of cement exist on the Moon, but they won’t be easy to dig up. For starters, there’s the obvious challenge of running a construction site with no air. To make matters worse, lunar days and nights are each 14-Earth days long, with temperature swings from 120°C to ‐130°C at the equator. Also, without the protection of an atmosphere or magnetosphere, construction equipment and crew would be exposed to high doses of radiation. Even if there’s a way to cope with the extreme environment, the moon’s soil is nothing like Earth’s. There’s no wind, no rain, no life on the Moon—just a surface that has been thwacked by space objects and radiation for eons. The result is not soil but “regolith,” meaning “blanket of rock,” which are more like jagged bits of stone and glass. As a fun bonus, the regolith is charged, so it clings to surfaces, almost like a living thing. 

With current technology, this probably won’t work at all, let alone work at scale. Earth currently requires over 3.5 billion metric tons of cement per year. Just imagine how much fuel is required to move industry workers to space and move moon-cement back down to Earth. Does that really seem like the most clean, convenient, and cost-effective way to save our planet? 

A more superficially plausible approach is to produce green energy in space. After all, there’s a whole sun’s worth of sunlight in space—no annoying atmosphere or weather or rotation of the planet to cause problems. Without the unpredictability of cloudy days and with a large enough fleet of interstellar solar panels, perhaps we could build a resilient space-based renewable grid. The idea has been pitched by governments and private companies alike, who dream of beaming clean energy back down to Earth.

The energy is certainly there—space-based solar energy could provide an order of magnitude improvement in the amount of energy produced per solar panel. That sounds like a lot—until you consider how much it costs to field and maintain solar panels in space. Right now, the cheapest method of getting stuff to orbit costs around $1,500 per kilogram. A typical individual solar panel weighs around 20 kilograms. So, even if you assume fantastic drops in the cost of launch and in the weight of panels, you’re still talking hundreds, likely thousands of dollars, just to get them where they need to be.

That alone is pretty much a dealbreaker, but the problem becomes even more acute when you start to think about maintenance. Solar panels are huge sheets of glass. In space, they’ll be bombarded by meteorites, pummeled with radiation, and heated from the full exposure to the sun. They will need massive radiator systems because in space there is no water or air into which to dump excess heat. They’ll also require highly trained astronauts or an army of autonomous robots to keep them functional.

And that’s just to get the power. You also have to beam it down to Earth—but not at too high of energy, lest it risk zapping birds and planes. You’ll also need a large receiver station, and you’ll lose quite a bit of energy transmitting between space and that receiver. 

Maybe let’s cover Arizona in solar panels before we decide this is a good idea. 

A more fatalistic line of thought assumes that at some point soon we’ll destroy this planet, and so building these space cities is an exit hatch. We call this the “Short-term Plan B.” 

It’s also a bad idea.

Listen, we like humans. Some of our best friends are people. They’re not perfect, but they have such potential. The problem is that if we screw up this planet during the next century, space is unlikely to be a refuge.

Why? Because space sucks.

Over the short-term, space settlement won’t help with any catastrophe you’re imagining right this second. Not global warming, not nuclear war, not overpopulation, probably not even a dinosaur‐style asteroid event. 

That’s because space is so terrible that in order to be a better option than Earth, one calamity won’t do. An Earth with climate change and nuclear war and, like, zombies and werewolves is still a way better place than Mars. Staying alive on Earth requires fire and a pointy stick. Staying alive in space will require all sorts of high‐tech gadgets we can barely manufacture on Earth.

The least-bad option for a Plan B world is Mars. But Mars is still garbage if the goal is survival: toxic soil, minimal protection against radiation, a dim sun, and dust storms that can engulf the whole planet. If you step outside without a spacesuit, you die. We have no idea if it’s possible for humans to even reproduce there. What little data we have on space baby-making comes from a grab bag of experiments on various animals and gametes from orbit—a non-trivial problem if the goal is sustaining our species without the home planet.

We also don’t know how to recreate the Earth’s biosphere in a sealed container at scale. The closest we’ve ever come is the Biosphere-2 project from the 1990s, in which eight human beings spent two years in a greenhouse, growing most of their food and generating most of their oxygen. It worked okay in the sense that they survived. It worked badly in the sense that they were starving by the end, at one point nearly suffocated, and early on had split into a pair of hateful factions. The project was supposed to repeatedly run for two years at a time, but amidst infighting and cost overruns, it was canceled after the first “voyage.” Since then, only a few similar experiments have been done, all in smaller facilities and without completely closed systems. That’s a problem for a Short-term Plan B, because any surviving Mars settlement looking to avoid inbreeding will want more than eight humans.

Few analysts have been brave enough to try to guess how many humans would be required to survive loss of contact with Earth. The lowest estimate we found, relying on extremely generous assumptions, put the figure at 100,000. Others have put the number at a billion. To get an intuition about why, consider how many services Earth provides for free or cheap —air you can breathe, clean water, oceans to efficiently transport products between coastal cities, soil that isn’t laden with toxins. We’d need huge amounts of effort to get baseline Earth-like conditions, nevermind developing the advanced economy required to supply a civilization utterly dependent on technology for survival.

That’s not even to mention who will get access to air, to drinkable water, and to clean land. If we can’t justly apportion these resources on Earth, where they’re free and more abundant, what makes us think we can do so in space, where they’ll be rationed and in short supply? Surely, our societal flaws on Earth—the ones that led us into the climate crisis in the first place—will follow us into the cosmos, too. 

Any plan that requires space to deliver some environmental benefit very soon is apt to fail. But, there is an idea we call “the Cathedral of Survival.” If our goal is to create a second reservoir for humanity, to help guarantee the survival of our species in the long-term, another planet might be a good way to go. 

If so, like the builders of a cathedral, we may want to start putting pieces in place even if no one living today will see the spire placed on top. This work will not help us against any 21st century calamity, but it might help us in the 22nd. Even if it doesn’t, much of the science required to survive on alien worlds may prove beneficial as we battle an ever more alien climate. 

The meticulous ecological science required may lack the urgent excitement of Musk’s City on Mars or Bezos’s deindustrialized Earth, but when it comes to real environmental challenges, we ought to be, well, grounded.