Last October I asked whether you could engineer a human, or any mammal, to run on sunlight the way a plant does. I expected a biology answer, something about genes and organelles and immune rejection.
What I got was arithmetic, and the arithmetic ends the argument before biology gets a turn.
The answer everyone gives is the weakest one
Search this question and the first thing you will read is that humans cannot photosynthesize because we have no chloroplasts and produce no chlorophyll.
That is true. It is also close to irrelevant.
Chloroplasts are a biology problem, and biology problems have a habit of falling over eventually. We already move genes between kingdoms. Sea slugs steal working chloroplasts from algae and keep them running for months. If the only obstacle were “you do not currently have the machinery,” this would be a question about how many decades, not whether.
The reason a photosynthetic human is impossible is that you are the wrong shape and you use too much power. Hand someone a perfect set of chloroplasts tomorrow, install them flawlessly, solve the immune response, and they will still starve.
You are a 97 watt appliance
Start with what you cost to run.
A 2,000 kilocalorie day is 8,368,000 joules. Spread that across 86,400 seconds and a human being is a continuous draw of about 96.9 watts.
Sit with that for a second. You idle hotter than an old 60 watt light bulb, every hour, awake or asleep, forever. That is what the sunlight has to cover.
Now the sunlight, and I will be generous
The solar constant is about 1,361 watts per square metre. Spread that over a rotating sphere and the top of the atmosphere averages a quarter of it, around 340 watts per square metre. Roughly half of that survives clouds and air to reach the ground.
I am going to use 200 watts per square metre as the average power arriving on a horizontal surface, day and night, summer and winter, all year. The true global figure is nearer 185. Every rounding in this article runs in the human’s favour, and it will not save us.
Next, efficiency. Photosynthesis is not a solar panel. Zhu, Long and Ort worked out the ceiling in 2008: the maximum conversion of solar energy into biomass is 4.6 percent for C3 photosynthesis at 30 degrees and present-day carbon dioxide, and 6 percent for C4. Those are theoretical limits that no plant sustains across a season. Real crops, averaged over a year, land nearer 1 percent.
The thing nobody accounts for
Here is where the popular version of this calculation quietly cheats.
Every article I read reaches for the surface area of human skin, about 1.8 square metres, and multiplies. But skin is not a collector. You are a cylinder, and at any given moment roughly half of you faces away from the sun. Lie down and spread out and your projected area, the shadow you cast at noon, is something like 0.8 square metres. That is the number that catches light.
Nobody makes this correction. It costs you a factor of two before you start.
So: a human lying spread-eagle in the sun, twenty-four hours a day, forever, never getting up to find shade.
| Scenario | Power harvested | Shortfall |
|---|---|---|
| Real-world efficiency (1 percent), projected area | 1.6 W | 61 times short |
| C3 theoretical maximum, projected area | 7.4 W | 13 times short |
| C4 theoretical maximum, projected area | 9.6 W | 10 times short |
| Lit on every side at once, C3 maximum | 16.6 W | 6 times short |
| Lit on every side at once, C4 maximum | 21.6 W | 4 times short |
The bottom two rows are physically impossible. They describe a person somehow illuminated from all directions simultaneously, at an efficiency no living plant has ever achieved, and they still fail by a factor of four.
That is what makes this different from an engineering problem. Better chlorophyll does not close a tenfold gap, and neither does lying very still.
So how much leaf would it take?
Turn the question around. Do not ask what a human can collect. Ask what area a human would need.
At the theoretical C4 maximum, 8.1 square metres. At the C3 maximum, 10.5. At the efficiency plants actually manage over a year, 48.4 square metres, which is a square about seven metres on a side.
Seven metres by seven metres of flat green surface, held up in the light, connected to a slow metabolism that does not run anywhere.
That is not a description of a person. It is a description of a tree.
When I got here in the original conversation I typed a sentence I have been thinking about ever since: a human powered by the sun is just a tree, once you pull all the levers and max out the physical constraints. The equations do not merely say no. They hand you a blueprint, and it is an oak.
Why every sunlit animal is small, flat, or lazy
This generalises, and once you see it you cannot stop seeing it.
Energy demand scales with body mass, which grows as the cube of your size. Light collection scales with surface area, which grows as the square. Double an animal and its appetite grows eightfold while its solar panel grows fourfold. Photosynthesis is a game you can only win by staying small, or by going flat, or by refusing to move.
Warm blood makes it worse. Keeping a constant body temperature costs a mammal several times what a similar-sized lizard spends. Of all the animals you could pick to run on sunlight, an endothermic mammal is close to the worst candidate on the board.
Look at what actually manages it. Corals and giant clams host photosynthetic algae, and they are sessile, aquatic, and living in bright shallow water. Green hydra and some flatworms do it, and they are tiny. In 2011, Kerney and colleagues found green algae living inside the cells of spotted salamander embryos, the first known case of a photosynthetic symbiont inside vertebrate cells. The salamander grows up and gives it up.
Not one of them is a mobile warm-blooded animal, because there are none, because there cannot be.
The sea slug, which is not the exception you want it to be
Elysia chlorotica is the animal everybody reaches for. It eats the alga Vaucheria litorea, keeps the chloroplasts alive inside its own gut cells for as long as nine or ten months, turns bright green, and survives for weeks without eating.
I wanted this to be the triumphant counterexample. It is not, and the reason is delicious.
Researchers kept the slugs in total darkness, and blocked photosynthesis outright, and the animals survived about as well as the ones basking in the light. Whatever is getting them through the hungry months, the evidence that it is sunlight is a good deal thinner than the popular story suggests. The most famous photosynthetic animal on Earth may be a very green creature living off its own stored reserves.
Even the poster child, when you look closely, is mostly just not eating very much. Which is, again, the whole point. The way to live on sunlight is to need almost nothing.
What ChatGPT got right, and what I want to add
The original answer was good. It found the 97 watts, it used 200 watts per square metre, it landed on 48 square metres at 1 percent efficiency, and my numbers agree with it exactly. It also listed the immune problems, the ultraviolet damage, the heat, and the awkward fact that we wear clothes and live indoors.
But it compared that 48 square metres against the surface area of skin, and skin is not what faces the sun. The real gap is twice as bad as it said, and the impossible best case still loses by four.
I keep finding this shape. In the router post the numbers contradicted each other outright. With the freight train every digit was right and the physics was half a step short. Here the answer is correct and the conclusion is understated. Confident, competent, and quietly not pushing hard enough on its own argument.
There is a nice symmetry with the pistachios, where the same surface-area-to-volume ratio runs in reverse. A single nut sheds its heat and stays cool. Pile ten tonnes of them together and the interior cannot lose heat fast enough, and the cargo can catch fire. Being small saves the nut. Being small is also the only way an animal ever lives on light.
The short version
You cannot photosynthesize, and chlorophyll is not the reason. You burn about 97 watts continuously, you present less than a square metre of shadow to the sun, and photosynthesis converts a few percent of what lands on it. Lying naked in a field forever, at an efficiency real plants never reach, you would harvest under ten watts.
To close the gap you would need roughly fifty square metres of leaf, no legs worth mentioning, and the metabolism of something that has never once needed to run away. Every animal that lives on sunlight has accepted at least two of those terms.
Nature did the experiment. It ran for a few billion years and the answer came back as a tree, and separately, later, as something that eats the tree.
Sources
- Zhu, X. G., Long, S. P. and Ort, D. R. (2008). What is the maximum efficiency with which photosynthesis can convert solar energy into biomass? Current Opinion in Biotechnology 19(2). The 4.6 percent C3 and 6 percent C4 theoretical ceilings.
- Kerney, R., Kim, E., Hangarter, R. P., Heiss, A. A., Bishop, C. D. and Hall, B. K. (2011). Intracellular invasion of green algae in a salamander host. PNAS 108(16), pages 6497 to 6502. Also free to read via PubMed Central.
- Elysia chlorotica. Wikipedia. Chloroplast retention for nine to ten months, survival without food, and the experiments in darkness that complicate the photosynthesis story.
- Explainer: why can’t humans photosynthesise? The Conversation. Reaches the same conclusion through glucose and ATP rather than watts, and uses total skin area.