Climate.ruthner.at← ruthner.atWeather Amstetten →DEEN

CO₂: amount, origin and memory of the atmosphere

How much CO₂ is in the air, where does it come from and how long does it stay? Plus the Keeling spiral and the seasonal cycle.

How to read: Measurement measured values · Reconstruction from indirect data such as ice or tree rings · Model computer model · Own calculation calculated by me from raw data. Technical terms are explained in the glossary.

The Earth’s energy balance: why CO₂ matters

The Earth absorbs solar energy and radiates heat back out. In balance the two amounts match; greenhouse gases hold back part of the heat radiation. This natural greenhouse effect raises the mean surface temperature to about 15 °C, more than 30 degrees warmer than without an atmosphere (NASA). More CO₂ strengthens the effect.

Schematic with rounded values. The 1361 W/m² is the solar constant at perpendicular incidence; on global average about 340 W/m² arrive according to NASA, about 29 % of it is reflected, the Earth absorbs about 240 W/m² and radiates almost the same amount back. The back radiation corresponds roughly to the incoming sunlight (about 340 W/m²). A small surplus of a little over 0.8 W/m² remains. The individual reflection values (77 and 23 W/m²) are approximations of the illustration. Without an atmosphere the mean temperature would be about −18 °C (NASA gives −20 °C). The value for Amstetten (+0.66 °C per decade, 1993–2025) comes from the GeoSphere station data of this page. Source: NASA Earth Observatory, “Climate and Earth’s Energy Budget”.
CO₂ · amount, sources, memory

How much CO₂ is in the air, where does it come from, and what would a country change?

First the order of magnitude (how little CO₂ there is in the air and what it is made up of), then the countries with their own calculation scenarios, then the “memory” of the atmosphere: the seasonal cycle at Mauna Loa and how long a tonne of CO₂ stays in the air. The sources are given below each chart.

Amount and origin
MeasurementOwn calculation

How small is 0.04 per cent?

💬 In plain language
  • Air consists almost entirely of nitrogen and oxygen. CO₂ is a very small part of it.
  • Out of 10,000 air particles, only about 4 are CO₂ particles.
  • It used to be slightly fewer, about 3. Today it is about 4. The increase comes mainly from people burning coal, oil and gas.
  • Nature also produces a lot of CO₂ every year. Plants, soils and oceans release it and take it up again. That is about 17 times as much as people emit. These flows largely balance each other, however; what people emit comes on top. That nature turns over more therefore does not mean that the human share is small.
  • However, nature takes up more in total than it releases. It even absorbs part of the CO₂ that people emit. The rest stays in the air. That is why there is more of it there.
  • How much warmer this makes it is estimated with computer models. More on this in the tab “Countries & scenarios”.
Detail of the upper left corner, magnified 40 times
remaining air (nitrogen, oxygen, argon …)CO₂ already present before industrialisation (278 ppm)CO₂ increase since 1750 (147 ppm)
Out of 10,000 air particles, about 4.3 are CO₂ particles: 0.0426 % by volume (426 ppm, annual mean 2025). Each square stands for one hundredth of a per mille of the air. Source: NOAA Global Monitoring Laboratory (global annual mean); pre-industrial 278 ppm according to IPCC AR6.
Annual exchange: who “produces” the CO₂?
Each year roughly 200 billion tonnes of carbon move back and forth between air, plants, soil and sea (respiration, decay, outgassing). Human emissions (combustion, cement, land use) amount to 11.6 billion tonnes of carbon (2024). That is about 6% of the natural exchange (11.6 of 200) or about 5.5% of the total flow of roughly 212 billion tonnes shown in the chart. Depending on the estimate of the natural exchange (about 180 to 220 billion tonnes) it is 5 to 6%. The natural exchange is therefore by far the larger item.
Of 100 tonnes of CO₂ from human emissions: where do they go?
Each square is one of 100 tonnes of CO₂ that people emit (average 2015 to 2024). Around 29 are taken up again by the ocean, around 21 by plants and soils. The remaining 50 or so stay in the air. That is why the CO₂ content of the air rises only about half as fast as humans emit it. Figures for 2024 in billion tonnes of carbon: emissions 11.6; the ocean takes up 3.4, the land 1.9; 7.9 are added to the air; a small remainder of 1.7 cannot be allocated. Source: Global Carbon Budget 2025.
The CO₂ in the air today
The amount in the air has risen from 278 to about 426 ppm. The increase of about 147 ppm corresponds to roughly 35% of today’s level. It is smaller than what humans have emitted, because sea and plants absorb part of it (see next chart).
Where did human CO₂ go? 1850 to 2024
Total emissions 1850 to 2024: 751 billion tonnes of carbon (503 from combustion and cement, 248 from land use). Of this, 39% stayed in the air, 27% went into the sea, 23% into land ecosystems; 11% cannot be allocated in the balance. Source: Global Carbon Budget 2025 (Friedlingstein et al.).

The global CO₂ cycle at a glance

Schematic with rounded orders of magnitude. The figures for the atmosphere (426 ppm correspond to about 905 Gt C) and for the 2024 emissions (11.6 Gt C, about 42.5 Gt CO₂) are matched to the documented values of this page (Global Carbon Budget 2025, NOAA). The arrows for photosynthesis and respiration (about 120 Gt C each) and for ocean uptake and release (about 90 Gt C each) are gross fluxes that nearly balance; their size and the stocks of land and oceans are approximate values of the illustration and are not individually documented in the charts above. What matters is the small surplus of uptake, the net sink: in 2024 the ocean took up about 3.4 Gt C net and the land about 1.9 Gt C (Global Carbon Budget 2025), together about half of the human emissions.

CO₂ is heavier than air – why does it still rise into the atmosphere?

That is true: a CO₂ molecule (44 g/mol) is about one and a half times as heavy as an average air particle (29 g/mol). Yet in the open atmosphere it does not sink to the ground. The reason is constant mixing: wind, turbulence and rising warm air keep stirring the lower atmosphere, near the ground within hours, up to the upper troposphere within days to weeks. Gravity cannot compete with this stirring: it pulls each single particle down only extremely slowly.

That is why the air is well mixed up to about 100 kilometres. Even the heavier argon (40 g/mol) is present everywhere at 0.93%. Only above that, where the air is so thin that particles hardly collide any more, do gases separate by weight. Aircraft measurements accordingly show only a few ppm difference between ground and great height. Within one hemisphere CO₂ spreads in weeks to months, between the northern and southern hemisphere it takes about a year. That is why the level is almost the same worldwide; the differences the globe shows result from sources, plants and seasons.

There is one exception: in windless, closed hollows (fermenting cellars, wells, pits, mofettes) CO₂ really does collect near the ground because nobody stirs it. That is a local safety issue and has nothing to do with the content of the atmosphere.

Assessment based on textbook atmospheric physics (mixing up to the turbopause); the vertical distribution is shown by aircraft profiles, for example from NOAA.

How is this calculated, and what does “from humans” mean?

Share of the exchange (about 5%). This is a ratio of annual amounts: human emissions (11.6 bn t C) to all flows between air and nature (roughly 200 bn t C per year according to IPCC AR6). It answers “who releases how much each year?”.

Share of the stock (about 35%). This is the increase in the amount of CO₂ in the air since 1750 relative to today's amount; it is not the share of “human” molecules among all CO₂ molecules. It answers “how much more is there than before?”. Both figures are correct; they answer different questions.

How the origin is attributed. First, human output (751 bn t C since 1850) is larger than the increase in the air (292). Second, carbon from fossil sources carries its own isotopic signature (no carbon-14, less carbon-13); this ratio in the air has shifted accordingly. The balance does, however, have a gap of 80 bn t C (11%) that is not allocated.

Ppm means “parts per million”: 426 ppm is 0.0426 per cent by volume. Conversion: 1 ppm CO₂ corresponds to 2.124 billion t of carbon or 7.78 billion t of CO₂.

The memory of the atmosphere
MeasurementOwn calculation

Seasonal cycle, spiral and residence time

💬 In plain language
  • The Earth “breathes” in the rhythm of the seasons.
  • In spring and summer, plants in the Northern Hemisphere (for example in Europe, Asia and North America) grow. They take CO₂ from the air. That is why the CO₂ level falls until autumn.
  • In autumn the trees lose their leaves, and plant remains decay. This releases CO₂ into the air again. That is why the level rises in winter and spring.
  • This up and down is about 7 ppm in size. On top of that comes a rise of about 2.4 ppm a year. This rise matches what people emit, because sea and plants absorb only part of it.
  • A tonne of CO₂ that enters the air today stays for a long time: after 20 years about half to two thirds is still in the air, after 100 years about 40 per cent.
Keeling spiral: Mauna Loa since 1958
Each loop is one year, running clockwise from January (top). The radius is the CO₂ level.
Monthly means from Mauna Loa (Hawaii), 3,400 m high in the middle of the Pacific, hence little local pollution. The rings move outwards because the level is rising, and wobble with the seasonal cycle. Source: NOAA Global Monitoring Laboratory.
The seasonal cycle (mean 2015 to 2025)
In spring and summer in the Northern Hemisphere, plants take up CO₂ and the level falls until October; in winter it rises again. The swing between May and September is about 7 ppm, the rise over the year about 2.5 ppm. The seasonal cycle thus balances out each year, while the rise remains.
How long does a tonne of CO₂ stay in the air?
Share of a one-off CO₂ pulse that is still in the air after t years (model function after Joos et al. 2013). Part of it disappears quickly, part stays for centuries. The “memory” is therefore the long residual amount. Hover over the line with the mouse.

See it on the globe: How CO₂ is distributed over the Earth across the seasons is shown by the globe in the mode “CO₂ in the air”, with a monthly animation.

2025
2025
Annual mean, anomaly from the 1961–1990 mean (°C)
−1,50+1,5
Earth surface: NASA Blue Marble (GIBS) · Coastlines: Natural Earth · Land: Berkeley Earth · Sea: NOAA ERSSTv5

Before about 1900 the data are sparse (light area on the left): few ship measurements, large ocean areas filled in statistically; the identification of El Niño and La Niña years is uncertain there. El Niño year · La Niña year · Volcano: major eruptions. Sulphur-containing gases rise into the stratosphere and form fine particles there that reflect sunlight; the Earth then cools by a few tenths of a degree for one to three years. Krakatau (Indonesia, August 1883), Santa María (Guatemala, October 1902), Agung (Bali, March 1963), El Chichón (Mexico, March/April 1982), Pinatubo (Philippines, June 1991, the strongest eruption of recent times, with up to about 0.5 °C of global cooling) and Hunga Tonga–Hunga Haʻapai (Pacific, January 2022, an underwater eruption that released mainly water vapour; the effect on temperature is small and still disputed). After Pinatubo the curve stays low for several years (1991/92 is also overlaid by an El Niño); in 1982 the eruption coincided with a strong El Niño, so the effect there cannot be separated out.

CO₂ over Europe through the year 2024
412416420424428432436440ppmJanFebMarAprMayJunJulAugSepOctNovDecJanuary: 432.0 ppmFebruary: 432.2 ppmMarch: 432.4 ppmApril: 431.6 ppmMay: 429.9 ppmJune: 426.6 ppmJuly: 421.6 ppmAugust: 418.9 ppmSeptember: 421.0 ppmOctober: 428.0 ppmNovember: 433.3 ppmDecember: 436.0 ppm418.9436.0EuropeWhole Earth

In summer plants take up CO₂: lowest value over Europe in August (418.9 ppm), highest in December (436.0 ppm).

Monthly means near the ground (lowest model layer), Europe = 35–70 °N, 10 °W–40 °E, area-weighted. For comparison the mean of the whole Earth. Source: NOAA CarbonTracker CT-NRT.v2025-1 (transport model matched to measurements), monthly data 2024. The values apply to model cells of 3° × 2°, not to single places.
Why does CO₂ fluctuate over the year?

The plants of the Northern Hemisphere “breathe” along. Most of the land area, and thus most of the forest and farmland, lies in the Northern Hemisphere. From about May to September the plants are growing: through photosynthesis they take CO₂ from the air and build leaves, wood and fruit from it. The CO₂ level in the air therefore falls over several months and reaches its low in September or October.

In autumn and winter this reverses. The leaves fall and photosynthesis rests. Soils, leaf litter and dead wood, however, continue to be broken down by fungi and bacteria, releasing CO₂. In winter there is also more burning for heating. The level therefore rises again from October and reaches its peak in May, just before the plants sprout again.

How big is it? At Mauna Loa the peak and the low are about 7 ppm apart, around 1.6% of the level. Near the ground over Siberia or northern China the swing is clearly larger (over 10 ppm, see globe), and over the South Pacific almost zero, because there is hardly any land there and the seasons run in opposition.

Why the low point is in September and not in July: Until late summer, plants take up more than soils and leaf litter release. Only when photosynthesis declines does release predominate, and the concentration rises again.

And humans? The rise of currently about 2.4 to 2.5 ppm per year is superimposed on the seasonal cycle. The seasonal cycle balances out within a year: what plants take up in summer, they mostly release again in autumn and winter. The rise, by contrast, goes in the same direction every year.

Data as of November 2026. The data series end in different years depending on the source (mostly 2024 or 2025); the end point is given in the source note under each chart, an overview is on “Data and methods”.

Feedback

Values wrong, an idea or a question? The message goes directly to the author of the page. Name and email address are optional and are only used for a reply.