Let’s be optimistic and assume that we manage to avoid a self-inflicted nuclear holocaust, an extinction-sized asteroid, or deadly irradiation from a nearby supernova. That leaves about 6 billion years until the sun turns into a red giant, swelling to the orbit of Earth and melting our planet. Sounds like a lot of time.
But don’t get too relaxed. Doomsday is coming a lot sooner than that.
The Earth is, in some ways, in a precarious spot in the solar system. There’s a range of orbital distances inside which a planet can have both liquid surface water (which is believed to be necessary for life) and enough atmospheric CO2 to carry on photosynthesis. This range is called the photosynthesis habitable zone. The Earth orbits barely within the sun’s zone. Some scientists estimate that the inner edge lies just 7.5 million kilometers away, which is only 5 percent of the distance between the Earth and the sun.
And that inner edge is moving out. Our sun is a massive ball of gas held together by its own gravity. At its center, the intense pressure and heat fuse hydrogen nuclei together to form helium. It takes four hydrogens to make one helium nucleus. As the number of nuclei in the core of the sun decreases by three with each helium nucleus formed, the outward pressure of the core also decreases (because the pressure is proportional to the number of nuclei per volume). In response, the outer layers of the sun squeeze the core harder, increasing its pressure, temperature, and fusion rate, leading to a 10 percent growth in brightness every billion years.
The Earth responds to this increasing brightness, in part, by decreasing the thickness of its warming blanket, which is provided by atmospheric CO2. Rising temperatures on Earth speed up reactions between water and silicate-rocks, which in turn draws CO2 out of the atmosphere. Sort of like changing your winter down comforter for a cotton sheet in the summer, this helps to keep the surface temperature of the planet habitable.
Eventually, though, the warming sun will cause CO2 levels to fall so low that plants will start to die. First to go will be the C3 plants, so called because their photosynthesis process involves a molecule containing three carbon atoms. Most plants are of the C3 type, including wheat, rice, barley, oats, soybeans, peanuts, coconuts, bananas, potatoes, cotton, and most trees. In about 200 million years, when the CO2 concentration drops below 150 parts per million (it is at 400 today), C3 plants will disappear.
Civilization relies mostly on C3 plants, which make up about 85 percent of global agricultural production by dollar value. Another class of plants are C4 plants, which employ a form of photosynthesis that involves four carbon atoms and is more efficient under many conditions. Although C4 plants make up only about 3 percent of plant species, they account for about 25 percent of the total photosynthesis on Earth. In order to survive the die-off of C3 plants, we will probably turn to genetic engineering to expand the list of C4 plants. Today, C4 plants include corn, sorghum, millet, and sugarcane, as well as some grasses and weeds.
Not even cockroaches will survive.
Efforts are already underway to convert rice from a C3 plant into a C4, which would result in about 50 percent larger harvests under present-day conditions. At 100 ppm CO2, the only rice on Earth would be C4 rice. C4 plants are thought to have evolved in response to the ongoing depletion of CO2. It is likely that as C3 plants die off, C4 species will continue to expand naturally to fill the newly opened niches in the ecosystem.
Unfortunately, about 300 million years after C3 plants disappear, C4 plants will die out too. When the CO2 concentration drops below 10 ppm, neither type of plant will remain.
Once plants disappear and cease replenishing atmospheric oxygen, animals will begin to die out. Currently, there are no known significant non-biological sources of oxygen on Earth. Once the plants are gone, oxygen will be a non-renewable resource. Large animals would probably asphyxiate within a few million years. Smaller and hardier creatures might last longer. Some types of microbes will probably survive by using metabolic reactions that do not rely on carbon dioxide or oxygen, but instead use materials such as sulfates or iron.
So that’s the number: In a paltry 500 million years or so, no humans will remain on the surface of the Earth—at least, not outside of some hypothetical controlled environment. And things get worse from there. After the atmospheric CO2 is gone and no longer able to regulate Earth’s surface temperature, things will start to get very hot. In about a billion years, the average surface temperature will increase to above 45 degrees Celsius from the current 17 degrees Celsius. Important biochemical processes turn off at temperatures above 45 degrees Celsius, leaving most of the planetary surface uninhabitable. Animal life will need to migrate to the cooler poles to survive; but by 1.5 billion years from now, even the poles will be too hot. Not even cockroaches will survive.
Now, there are a few things we can do to stay our execution. We could, for example, move the Earth’s orbit. If we fired a 100 km wide asteroid on an elliptical orbit that passed close to the Earth every 5,000 years, we could slowly gravitationally nudge the planet’s orbit farther away from the sun, provided that we don’t accidentally hit the Earth. As a less precarious alternative, we could build a giant solar sail behind the Earth with enough mass to drag the planet away from the sun. Such a sail acts like a kite, where the photons from the sun are the wind and the gravity between the solar sail and the Earth acts as the string. The sail would need to have a diameter 20 times that of the Earth but a mass only about 2 percent that of Mt. Everest, a mere trillion metric tons. Strategies like these could, in principle, keep the Earth in the habitable zone until the sun expands into a red giant. (If some other civilization has already built such a large solar sail, we could detect it using the same photometric techniques that are currently used to find exoplanets.)
Another survival choice is more complicated—or simpler, depending on your perspective. The future Earth will actually be a pleasant home for non-biological life—better than it is today. For one thing, the brighter sun will provide more abundant solar power. The space weather will also be nicer. The sun is a dynamo spinning on its axis about every 24 days, generating giant magnetic storms that disrupt communication networks, overload power grids, and damage orbiting satellites. Robots today need fear that their circuits could be fried by a solar storm, such as the large solar storm in 1989 that caused a power failure across most of Quebec. Currently, such storms are estimated to occur about once or twice per century. But as the sun ages, this rotation slows down and the magnetic storms will abate.
Given these facts, we humans might simply decide to upload ourselves into machines, which would be relatively comfortable on the dystopic future Earth. This would require advanced computing resources and a deeper understanding of neuroscience than are currently available, but there is no known fundamental reason why we could not exchange our biological hardware for robotic replacements. We could probably figure it out in the next few hundred million years.
Michael Hahn and Daniel Wolf Savin are astrophysicists at Columbia University.
This article was originally published in our “Aging” issue in May, 2016.