Is the Amazon really "the lungs of the Earth"?
― The forest makes oxygen, but uses almost all of it itself
"The Amazon rainforest makes a fifth of the world's oxygen." Every time there's a wildfire in the news, this line does the rounds again. But a mature forest uses up almost as much oxygen as it makes. And the air already holds so much oxygen in reserve that the whole forest burning down wouldn't make a dent. So does that mean the forest doesn't matter? Not at all. It matters for a different reason entirely ― one that has nothing to do with oxygen.
On TV, the Amazon rainforest is burning, smoke rising into the sky. Across the bottom of the screen: "The lungs of the Earth are on fire."
Watching it, you might feel a little short of breath. "If the forest keeps shrinking, will we run out of air to breathe?"
And yet, spend days deep in a forest and the oxygen never gets so concentrated it makes you dizzy. Stand in the middle of a desert and you can breathe just fine too. The share of oxygen in the air near a forest and in a desert is almost exactly the same. Why is that?
Two reasons "the lungs of the Earth" is an overstatement
Trees make oxygen by day, but they use oxygen through respiration both day and night. On top of that, the fungi and insects that break down fallen leaves and dead trees also use oxygen. In a mature forest, the amount made and the amount used are nearly in balance.
About 21% of the air is oxygen, and the sheer quantity is almost too large to picture. Even burning every plant on land would use up only about 0.1% of that reserve.
The first point is: "the forest is an oxygen factory, but it isn't shipping any out." The second is: "even if the factory stopped, the warehouse has thousands of years' worth of stock." Let's take them one at a time.
A forest's oxygen gets used up inside the forest
Plants use light to build sugar from carbon dioxide and water. Oxygen comes out as a byproduct. That's photosynthesis ― exactly what you learned in school.
But plants also live off the sugar they make. Breaking that sugar down with oxygen to release energy is respiration ― a reaction that runs exactly opposite to photosynthesis. Plants themselves are thought to use roughly half the sugar they produce this way, through their own respiration.
The rest of the sugar becomes leaves, trunk, and roots. But leaves eventually fall, and trees eventually topple. When fungi and insects eat and break them down, oxygen gets used again. Tropical forests are warm and humid, so this decay happens very fast.
So in a mature forest, "oxygen released by photosynthesis" and "oxygen used up by respiration and decay" end up almost equal. If a forest kept growing heavier and heavier, a gap would remain ― but the total weight of a forest that's been standing for centuries barely changes at all. No change in weight means what goes in and what comes out are balanced.
Researchers' estimates agree: the net oxygen the Amazon adds to the air, on balance, is close to zero. During the major fires of 2019, ecologists in the UK pointed this out, and it became a talking point.
Burning every plant on land would cut atmospheric oxygen by only 0.1%
So what if the forest vanished entirely? How much would atmospheric oxygen actually drop? We can calculate this. When wood burns, the carbon it contains combines with oxygen in the air to form carbon dioxide. The amount of oxygen used is set by the amount of carbon burned.
The carbon held in all of Earth's land plants adds up to roughly 450 billion tonnes. Burning all of it would use about 1.2 trillion tonnes of oxygen. That sounds huge ― but the atmosphere holds about 1,200 trillion tonnes of oxygen, a thousand times more.
Try moving the slider in Figure 1. The left bar is the share of oxygen in the air; the right bar is carbon dioxide. Even burning every plant on land, the left bar barely moves. 21% drops to only about 20.93%.
In Figure 1, it's really the right-hand bar that stands out. Carbon dioxide is such a tiny slice of the air that the same added amount shifts its share a lot. The problem with a burning forest isn't "oxygen going down." It's that the carbon stored in the forest gets released into the air as carbon dioxide.
The answer: plants and plankton that died and got buried without decaying. When a dead organism decomposes, the oxygen it produced gets used right back up. But if it gets buried at the bottom of the sea or a swamp and escapes decay, that oxygen stays behind in the air. Pile that up over hundreds of millions of years, and you get today's 21%. Coal and oil are exactly what got buried back then. Burn them, and you're slowly using back up the oxygen left over from way back. Indeed, observations show atmospheric oxygen has been very slowly declining.
Roughly half of all photosynthesis on Earth is thought to happen in tiny ocean plankton. That part is true. But in the ocean too, most of the oxygen produced gets used right back up by the plankton's own respiration and by the creatures that eat them. "Amount produced" and "net amount left over" are different things ― just as with forests.
In short
A mature forest uses up, through its own respiration and leaf decay, almost all the oxygen it releases through photosynthesis. And the air's oxygen supply is on such a different scale that burning every plant on land would only cut it by about 0.1%. Calling it "the lungs of the Earth" is an overstatement, at least when it comes to oxygen. But a forest is an enormous "vault" of carbon. Burn it or clear it, and that carbon escapes into the air as carbon dioxide. On top of that, forests sustain the water cycle that brings rain, and provide homes for countless living things.
A forest is less an oxygen factory than a carbon vault.
What we'd lose isn't the air we breathe ― it's what's locked in the vault.
For more on how much carbon a forest stores, see "How many trees does it take to absorb one person's CO₂ emissions?"; for leaf decay, see "Why don't fallen leaves pile up into mountains in a forest?" And for a breakdown of the weight of all life on Earth, see "If you put every living thing on Earth on a scale, what would weigh the most?"
- Put aquarium pondweed (like Egeria) in a clear glass of water. Set it in sunlight, and tiny bubbles will rise from the leaves ― these are bubbles of oxygen from photosynthesis.
- Cover the glass with a box to darken it, and check back later. The bubbles almost stop. Even in the dark, the pondweed keeps using oxygen through respiration.
- Adding a little BTB solution (from a school science lab) to the water makes it even clearer: in light, CO₂ drops and the water turns bluish; in the dark, respiration raises CO₂ and it turns yellowish.
Inside a single sprig of pondweed, "making" wins by day and only "using" happens at night. Add that up across a whole forest, over a day and a year, and the two nearly cancel out ― that's the whole story of this article.
Want to go deeper? ― Terms, formulas, and how this connects to the curriculumWe've labeled each section by level, from middle-school science to university specialist courses
- MSCovered in middle-school science
- HSCovered in high-school Biology, Chemistry, or Earth Science
- HS+Advanced high-school content, or a textbook sidebar topic
- Univ.Not covered in high school ― university-level specialist content (ecology, geochemistry)
- ResearchNot yet settled even at university level ― an active research question
MSTerms: this phenomenon has names
- Photosynthesis: the process by which plants use light energy to build sugar from carbon dioxide and water, releasing oxygen.
- Respiration: the process by which living things break down sugar with oxygen to release energy, producing carbon dioxide. Plants do this day and night.
- Decomposers: fungi, bacteria, and small insects that eat and break down fallen leaves and dead organisms, using oxygen in the process.
MSHSCheck with a formula: how much would burning all land plants cut oxygen?
The quantity we're after is "the share of atmospheric oxygen lost if plants are burned." We work out the weight of oxygen used in combustion, then compare it to the total weight of oxygen in the air.
| In symbols | C + O₂ → CO₂ , M(O₂) = M(C) × 32 ÷ 12 |
| In words | Weight of oxygen used = weight of carbon burned × (weight of one O₂ molecule ÷ weight of one carbon atom) |
| Where this comes from | In combustion, one carbon atom combines with one oxygen molecule (the chemical equation). Since atom counts are conserved, the weight ratio comes out as 32 to 12 (conservation of mass) |
| Carbon contained in land plants | about 450 billion tonnes (450 gigatonnes), estimated |
| Total atmospheric oxygen | about 1,200 trillion tonnes (1.2 million gigatonnes), estimated |
| Share of oxygen in air (by volume) | about 20.95% |
| Current CO₂ concentration | about 420ppm |
| Carbon corresponding to 1ppm of atmospheric CO₂ | about 2.12 gigatonnes |
Here, M(C) is the weight of carbon burned (gigatonnes), and M(O₂) is the weight of oxygen used (gigatonnes). A gigatonne is one billion tonnes; ppm stands for parts per million.
| Carbon weight × 32 | 450 × 32 = 14400 |
| ÷ 12, giving oxygen used (gigatonnes) | 14400 ÷ 12 = 1200 |
| As a share of atmospheric oxygen | 1200 ÷ 1200000 = 0.001 |
| Drop in oxygen share (percentage points) | 20.95 × 0.001 ≒ 0.021 |
| Oxygen share after burning (%) | 20.95 − 0.021 ≒ 20.93 |
| Rise in CO₂ (ppm) | 450 ÷ 2.12 ≒ 212 |
| CO₂ after burning (ppm) | 420 + 212 = 632 |
Oxygen falls by only about 0.1% (one part in a thousand) of its total, dropping from 20.95% to 20.93%. The same amount of carbon pushes CO₂ up by half, from 420ppm to about 630ppm. Each oxygen molecule used corresponds to one CO₂ molecule produced, so the "number" of molecules gained and lost is the same. They look so different only because the starting quantities differ by a factor of about 500.
HSHS+Separating "amount produced" from "amount left over"
HSIn an ecosystem, the total organic matter plants build through photosynthesis is called gross primary production. Subtract the plants' own respiration, and you get net primary production. Subtract further the amount lost as fallen leaves and dead branches, and the amount eaten by animals, and what's left is the forest's growth ― the increase in its weight.
HS+In a mature forest (a climax forest), growth is close to zero. The organic matter built as net primary production flows, via fallen leaves and the like, to decomposers, and returns to carbon dioxide through respiration. Treating the whole forest as a single box, oxygen going out and oxygen coming in balance, and the net contribution to the air all but disappears ― that's this article's conclusion.
Univ.Net primary production, net ecosystem production, and the atmospheric oxygen budget
In ecology, gross primary production (GPP) minus plant respiration is called net primary production (NPP). Subtract further the respiration of heterotrophs such as decomposers, and you get net ecosystem production (NEP) ― the quantity directly tied to net exchange with the atmosphere. Observations using methods like eddy covariance suggest the NEP of mature tropical forests is close enough to zero that it swings positive or negative from year to year. Atmospheric oxygen itself is thought to have built up over geological time through the burial of organic carbon and pyrite in sediments. In today's atmosphere, precise measurements confirm that the ratio of oxygen to nitrogen (the O₂/N₂ ratio) is falling very slightly each year, due to fossil fuel combustion.
📖 For the full derivation and further reading: Carbon cycle (Wikipedia, Japanese) ／ Scripps Institution of Oceanography Atmospheric Oxygen Research Program
ResearchWhat's still not fully understood
- Is the Amazon still absorbing carbon dioxide today? In the southeastern regions, where logging and drying have advanced, observations have reported the forest has flipped to becoming a net source of carbon dioxide. Whether this holds for the forest as a whole is still under debate.
- Why has the oxygen share stayed stable at around 21%? One hypothesis holds that too much oxygen makes wildfires more likely, which then reduces plant cover and thus oxygen supply ― a kind of self-regulation. How much this actually matters remains unclear.
- How much rain does a forest actually generate? The Amazon is thought to evaporate water from its leaves and carry rain inland. Estimates of the tipping point ― how much forest loss would break this rainfall mechanism ― still vary widely.
In other words, even this article is "the best explanation we have for now." The true value of forests, beyond oxygen, still isn't fully measured.
Where this fits in the curriculum (by level)
| Level | Subject / Unit | Where in this article |
|---|---|---|
| MS | 2nd-year Science, "How plant bodies function" (photosynthesis and respiration) | Reason 1, observation block |
| HS | Chemistry Basics, "Chemical equations and quantitative relationships" / Biology, "Ecosystems and matter cycles" | Checking with a formula, gross vs. net production |
| HS+ | Biology, "Climax and succession" | Budget of a mature forest |
| Univ. | Ecosystem ecology, geochemistry (carbon cycle, atmospheric oxygen budget) | Net primary production and net ecosystem production |
| Research | Tropical forest carbon-budget observation, long-term atmospheric oxygen trends | Research section |
| ― | Connecting to daily life | Rereading the news phrase "lungs of the Earth" as a story about carbon |
- Bar-On, Y. M., Phillips, R., Milo, R. (2018) The biomass distribution on Earth. Proceedings of the National Academy of Sciences 115(25)
- Scripps O2 Program (Scripps Institution of Oceanography Atmospheric Oxygen Research Program)
- Keeling, R. F., Shertz, S. R. (1992) Seasonal and interannual variations in atmospheric oxygen and implications for the global carbon cycle. Nature 358
- IPCC (2021) Climate Change 2021: The Physical Science Basis. Chapter 5, Global Carbon and Other Biogeochemical Cycles and Feedbacks
- Wikipedia, "Carbon cycle" (炭素循環, Japanese)
※This article is a general-audience science explainer. The figures given are approximate, meant to help convey the underlying mechanism. Estimates of forest carbon storage and total atmospheric oxygen vary by several tens of percent depending on the source.