La lune


Documentation > Greenhouse effect > Greenhouse gases : what are they ? > What gases are greenhouse gases? How much do they contribute to the greenhouse effect?

What are the greenhouse gases ?

last modified : august 2007

website of the author : - contact the author :


The definition of a greenhouse gas is at the same time very simple and very complicated for the ordinary man (or woman !) : such a gaz is "just" a gas mixed in the atmosphere that absorbs the infrared radiation emitted by the earth's surface. We are not accustomed to these gases because neither nitrogen nor oxygen, the two most abundant gases of the atmosphere (78% and 21%, respectively), that many of us have heard of, have this ability to intercept infrared radiation. But once this is said, what are these gases exactely ? And are we the sole emitters for these gases ?


The "natural" greenhouse gases

The two main gases responsible for the greenhouse effect (and not noly its recent increase) are :

water vapour (H2O),

carbon dioxide (CO2).

There are others such gases, and even many others. Some of them are "natural", which means that they were present in the atmosphere before the apparition of men, and other can be called "artificial", in the sense that they are present in the atmosphere only because of us.

Beyond water and CO2, the other important "natural" greenhouse gases are :

methane (CH4), which is nothing else than the cooking gas we use in our stoves,

Nitous oxyde (N2O), the scholarly name for..... laughing gas (which is not so much amusing here),

ozone (O3), which molecule comprises 3 oxygen atoms (the molecules of the "regular" oxygen gas have only 2 atoms of oxygen).

When we say that these gases are "natural", it does not mean that men did not play a role in the amount we can find in the atmosphere today. It just means that there are also natural sources (or natural cycles). For these 3 above mentionned gases, humanity "simply" adds its part to natural emissions and therefore significally increases their concentration in the air.

All these "natural" gases are taken into account in the international negociations (like the Kyoto Protocol, for example), except ozone, because as it has no direct emissions. Ozone results from a subtle chemistry taking place in the air, involving "precursors" which are regular pollutants - NOx, hydrocarbons - with the help of sun rays. Calculating - even roughly - the amount of ozone emitted by a country is today clearly very difficult.


The "industrial" greenhouse gases

The main "industrial" greenhouse gases are halocarbons (generic formula CxHyHalz, where Hal represents any halogen) : it designates sa vast familly of gases obtained by substituting, in a hydrocarbon molecule (propane, butane, or even octane, than can be found in car gas, are hydrocarbons), all or part of the hydrogen by a halogen gas (fluorine, chlorine, iodine...). The molecules obtained that way have two important properties for our purpose :

They are generally highly efficient to absorb infrared radiation, much more than CO2 (their absorption bandwiths are large).

Some of them (perfluorocarbons par example) are very "solid" : they are extremely stable, and only the high energy ultraviolets or the cosmic rays can "break" the liaisons of these molecules once they are in the atmosphere. As these degrading processes happen slowly and far from the ground, halocarbon molecules generally have very long residence times in the air, because it is necessary to wait until they get to the stratosphere - even though they are (very) heavy molecules - before they are degraded, and that can require thousands years.

Among the halocarbons we will find a well known sub familly : the CFCs (for chloro-fluro-carbons). Not only they are potent greenhouse gases, but they also lead to a decrease in the stratospheric ozone. Their production has been progressively banned, by the way of the Montréal protocol, signed in 1987, and that does not concern the other greenhouse gases.

There is another "industrial" gas that is often mentionned among the experts, sulfur hexafluoride (SF6). It is used, for example, to fill transformers (that require gases that stay inert in extreme conditions) or...double glazings. It is not emitted in large quantities, but is even more potent than any halocarbon and its degradation requires several thousand years.


What are the gases that generate the higher greenhouse effects and where do they come from ?

If we do not bother about the origin (natural or anthropic) of the greenhouse gases, the one that generate the highest greenhouse effect is....water vapour.

Breakdown of the "natural" greenhouse effect by contributing gas. As halocarbons are industrial gases they are not represented here.

From IPCC, 1992 (the relative importance of each gaz has not varied much today).

But if we only consider the greenhouse effect of human origin, sometimes called "additional" greenhouse effect (because it goes on top of the natural one), or anthropic greenhouse effect, the relative importance of each gas is totally different :

anthropic water emissions are negligible. Indeed, on a planet which is covered by water for its two thirds, and taken into account that water does not accumulate in the atmosphere where its residence time is roughly a week, direct human emissions of water vapour do not have a significant impact on the global water cycle. Men can definitely cause major perturbations to the water cycle on a local scale (through deforestation, irrigation, creation of dams, etc) but that doesn't have significant repercussions on the average proportion of water vapour in the atmosphere on a global scale, and therefore on the global greenhouse effect coming from water vapour.

This explains why water vapour is not taken into account when measuring the greenhouse gas emissions caused by human activities, except in some very particular cases.

CO2 generates a little over 55% of the human induced greenhouse effect. There are of course natural CO2 emissions (respiration of animals, plants and humanity, decaying of biomass, natural forest fires, ocean emissions...). The anthropic CO2 comes :

from fossil fuels use (coal, oil, natural gas) for the largest part,

from industrial processes for a small part (when referring to these processes we exclude combustion, but only take into account other chemical reactions), for example cement production,

from deforestation for a non negligible part, particularly between the tropics (see the page on the sinks for explanations).

Methane generates a little over 15% of the human induced greenhouse effect. Methane is nothing else than the main component of "natural gas" (and also the cooking gas of most people, and...the firedamp so feared by coalminers) and is formed as soon as any organic compound decays (through putrefaction or fermentation) in the absence of air (actually in the absence of oxygen), for example under water or under the ground. Natural gas reserves were formed exactely that way, by the decay, a long time ago, of terrestrial or marine biomass. Finding methane in the atmosphere is therefore totally normal, as a result of the existence of swamps, and...termites !

But man has added its part, through :

biomass combustion, especially in the tropical zone. Wood burning is almost always an imperfect combustion, that releases in the atmosphere various unburnt or partially burnt hydrocarbons, including methane,

cattle raising (cows, sheep, goats, yaks, and more generally any ruminant), because the food they ingest ferments in theur stomachs, and that leads to methane emissions in the atmosphere (another conclusion is that all ruminants burp but that horses, that are not ruminants, just fart). It might interest the reader to know that there are roughly 1,4 billion cows on earth : they weight more than humans ! And their methane emissions are far from marginal.

rice paddles, that are humid zones just as swamps (where bits of dead plants fall under water, and decav there in the absence of oxygen),

dumpyards (putrefaction once again) and compost production,

fossil fuels production, because of leaks (oil and gas production) or ventilation of mines (coal).

halocarbons generates a little over 10% of the human induced greenhouse effect (no natural sources). These gases are used :

as fluids for filling fridges and more generally any device that generate cold (air con, domestic or in a car, freezers, etc). Emissions then happen because of leaks during use (almost any cold device leaks at least a little) or when dumped.

as propellants in sprays : the all too famous CFC constitue a subset of halocarbons ; the Montreal Protocol has decided their progressive ban because, in addition to being potent greenhouse gases, they are also responsible of the destruction of high altitude ozone,

in a number of industrial processes (manufacturing of plastic foams, but also...of semiconductors : anybody who reads this sentence or used a cellular phone is indirectly at the origin of halocarbon emissions).

Nitrous oxide (N2O) generates roughly 5% of the human induced greenhouse effect. This gas is a by-product of microbian activity in the soil (and is obviously to the nitrogen cycle), and therefore also has natural sources, mostly humid zones. The human part comes from :

the use of fertilizers in agriculture,

some chemical industries (no surprise : nitric acid production is among them).

Ozone (O3) generates roughly 10% of the human induced greenhouse effect. Ozone is a variant of the "regular" oxygen molecule (it has 3 atoms of oxygen instead of 2 for the "regular" oxygen gas) and is naturally pressent in the atmosphere. We like it (a lot) or not, depending on where we find it :

in the higher atmosphere, were it is named stratospheric ozone (the stratosphere is the layer of the atmosphere comprised between 10 and 50 kms above the ground), it stops the ultraviolet radiation coming from the sun, which have the ability to break down some weak chemical liaisons in the organic molecules. In this place we therefore like it a lot, because its apparition allowed advanced life forms to come out of the oceans (and if this ozone suddently disappeared, it is questionnable whether advanced life forms could remain for long on emerged land),

in our cities, it is a very aggressive pollutant, and we'd rather not see it there ! The tropospheric ozone (the troposphere is the lowest layer of the atmosphere, the one that "touches" the ground) is a classical component of air pollution, and indirectly comes from hydrocarbon combustion. The contribution coming from tropospheric ozone is therefore mostly a consequence of road and air transportation.

Since the beginning of the industrial age, that is 1750, all the additionnal greenhouse gases that we have poured into the atmosphere have created a "radiative forcing" that amounts to roughly 1% of the incoming solar energy.

In other words, through its greenhouse gases emissions, man has modified his environment "exactely as if" the incoming solar energy had increased by 1%. 1%, that may seem very little. But, given the considerable energies that are at play, the fragile balance of many subsystems of the climate machine, and the fact that this enhancement, once made, remains for very long periods of time, it is very significant for our future, as we will see later on.


How long do greenhouse gases stay in the atmosphere ?

Thoses gases, once in the atmosphere, do not remain there forever. They can be removed by various means :

it can be the consequence of a physical process. For example rain, which is the consequence of condensation (a physical process), removes water vapour from the atmosphere.

it can be the consequence of a chemical reaction happening in the atmosphere. It's the case for methane, for example, which is essentially removed from the atmosphere by reacting with OH radicals, resulting in CO2 among other things. It's also the case for ozone, a very reactive gas that disappears from the atmosphere in a matter of hours or days, being able to combine with many other compounds (including our lungs !),

the removal process can be a physico-chemical reaction, at the border of the atmosphere and of another compartment of the planet. CO2, for example, is reduced by photosynthesis (atmospheric/ground border), or is dissolved in the ocean then converted into bicarbonate and carbonate ions (CO2 is chemically stable in the atmosphere),

as a consequence af a radiative process. For example, the cosmic rays and the most energetic ultraviolet radiations emitted by the sun are able to breack down the laisons of many molecules in the higher atmosphere. Part of the halocarbons and part of the nitrous oxide "disappear" that way. Halocarbons being heavy molecules, it takes a long time for them to diffuse to the stratosphere, so this way of removal only concerns those that are chemically stable in the lower atmosphere (saturated halocarbons, practically). When halocarbons keep enough hydrogen, they generally get removed like methane, through chemical reactions with hydroxile radicals.

But there is a very bad surprise here : as opposed to water vapour, which is removed pretty quickly after being emitted, other gases take a very long time to go away once emitted. It is not easy to know accurately how much time is necessary for a given gas to be removed, because the atmosphere is a very complex system, involving a tremendous number of physical and chemical reactions (it doesn't seem so when we just look up !), with many feedbacks, and predicting the residence time with good precision for a given gas is a tricky exercise.

That being said, it is nevetheless possible to give a rough estimate of the residence time (called atmospheric lifetime sometimes), that is the time necessary for a surplus of gas to be removed. Of course this time is valid today, but may well cease to be if the conditions change much tomorrow.


Approximative time of residence in the atmosphère


100 years


12 years

Nitrous oxide

120 years


from several weeks up to 50.000 years

We can immediately see above that the vast majority of the greenhouse gases that we emit today, including the CO2 that we just emitted this morning while commuting in car, or while turning on the central heating (coal, fuel oil or natural gas), will remain above our heads and that of our kids and grandkids in one century or two. All that time that they will remain in the atmosphere, they will enhance the greenhouse effect.


How do they compare ?

In order to be able to compare them (what is essential for mitigation : as long as we don't know if it's better to prevent the emission of one kg of CO2 or of one kg of methane, priorities difficult to establish and therefore our hands are a little tied up...), it is possible to calculate, for each of team, a "global warming potential", or GWP in short, that allows to know how much additional greenhouse effect we generate when we emit one kg of a given gas.

The global warming potential of a gas is defined as the "radiative forcing" (that is the additionnal radiative power that the gas is sending back to the ground) of a given quantity of gas, cumulated over a given period, generally 100 years.

In short it is a notion that enables to apprehend at the same time the "instant power" of the gas, that is the ability to intercept infrared and transmit heat to the atmosphere (which varies with the number of frequencies the gas can intercept), and its lifetime in the atmosphere.

This value is actually never given as an absolute figure, but relatively to CO2. The GWP of a gas is therefore "how much more" (or how much less) it "enhances the greenhouse effect over 100 years" (that is how much additionnal energy it sends back to the ground) compared to a similar quantity of CO2 emitted at the same time. We then mention the "relative GWP". In mathematical terms it court be written as follows (F = radiative forcing and N is generally equal to 100 years) :

But it must be noticed that there are overlappings in the ranges of frequencies that the various greenhouse gases absorb. Methane and nitrous oxide, for example, absorb the same infrared around 7 microns). As a consequence, the result of the emission of a given gas in the atmosphere cannot be predicted independantly of the pre-existing quantities of the other gases that have absorbtion ranges in common.

Percentage of the radiation absorbed (vertical axis) depending on the wavelength in micrometres (horizontal axis) for methane and nitrous oxide in the atmosphere. Source Gérard Lambert, Revue du Palais de la Découverte.

It is also important to notice that the residence time may vary a lot depending on the conditions. For example, if the sinks that now remove CO2 from the atmosphere get saturated in the future, the residence time (of CO2) will increase, or even become infinite if the sinks turn to source. Assuming a constant rate of removal of CO2 from the atmosphere being precisely contrary to the known conclusions (this rate will change) it can be said that this GWP notion is approximative by nature.

If we wanted to be precise, each GWP should be a function not only of the absorption possibilities of the gas, and of its lifetime in the pressent conditions, but also of the concentration of all the other gases with overlapping ranges, the evolution of these concentrations in the future, and the evolution of the sinks for the given gas for the coming centuries !

This is of course impossible (not with explicit formulas, anyway). But as imperfect as it may be, an approximative comparison remains preferable, for action, to no comparison at all.


Here are the relative GWP of the 6 gases of gas famillies covered by the Kyoto protocol (Perfluorocarbons and Hydrofluorocarbons are sub-famillies of halocarbons),



Relative GWP / CO2 (100 years)

Carbon dioxide






NItrous oxide





7400 to 12200



120 to 14800

Sulfur hexafluoride



From IPCC, 2007


The above table means that if we emit today 1 kg of methane in the atmosphere, we will increase the cumulated greenhouse effect, for the coming century, just as much as 25 kg of CO2 (also emitted today). In short, we could say that a kg of methane "generates" 25 times the greenhouse effect that a kg of CO2 generates over a century, or that methane is a greenhouse gas 25 times more potent than CO2.

In the same idea, regarding the greenhouse effect, a kg of sulfur hexafluoride equals 22,8 (metric) tonnes of CO2, that is the annual emissions of 3 French ! (or of one American). Fortunately, the emitted quantities remain low for the moment (see below).

At last, rather than mentionning the mass of carbon dioxide, engineers and economists often use the carbon equivalent. Just as lenghts are measured in meters, greenhouse gas emissions are measured in carbon equivalent.

By definition, a kg of CO2 is worth 0,2727 kg carbon equivalent, that is the weight of the sole carbon in a kg of CO2.

For the other gases gaz, the carbon equivalent is definded as :

carbon equivalent = relative GWP x 0,2727

It may seem very complicated, but actually it is very simple, because that way when we burn one kg of pure carbon, we obtain one kg of carbon equivalent in CO2. This then enables to know how much carbon equivalent a hydrocarbon will produce when burning, by simply measuring the weight of carbon per kg of hydrocarbon burnt (burning hydrogen produces water, not taken into account as explained at the top of the page).

And then the carbon equivalent of a kg of hydrocarbon that is burnt is simply the weight of carbon in this kg. Simple, I said !

For our main greenhouse gases, the carbon equivalents are therefore the following :



Carbon equivalent per kg

Carbon dioxide






NItrous oxide





2 015 à 3 330



34 à 4 040

Sulfur hexafluoride


6 220

Why go through all this pain ? If some states set up a "carbon tax", as some of them consider doing to discourage the greenhouse gas emissions, then it woud be logical to adjust this tax to each gas depending on his "potential to harm", or...its GWP. If the ton of carbon equivalent (sometimes noted tCe) is taxed 1000 euros, then the emission of one ton of CO2 would be taxed 273 euros, the emission of a ton of methane 6820 euros, the emission of a ton of nitrous oxide 81300 euros, etc.

Once we have a comparison basis (otherwise it is not possible !), we can then give a breakdown of the global human emissions by gas, except for ozone (which, as explained above, has no direct emissions) :

Breakdown of the anthropic greenhouse gas emissions by gas, in billion tons carbon equivalent, in 2004. Source : IPCC, 2007

Same as the bar graph on the right, but as a pie chart with the share of each gas. Source : IPCC, 2007



In addition to greenhouse gases, humanity emits many other substances, including what is called aerosols or aerosol precursors.

An aerosol is basically a suspension in the air or liquid or solid particles. We can see everyday an example of aerosol : coulds. But a dust cloud also fits in this caterigy : when we sweep energically, we create an aerosol.

If we stick to humankind, aerosol emissions include :

tiny particules emitted when we burn oil or caol (the famous "black smoke" we can see coming out of the exhaust pipe of a diesel car),

tiny particles emitted when we burn wood (wood accounts for 10% of the global energy consomption on earth, in rough figures),

dust directly lifted by road trafic, or stone mining...

A precursor is something that preceeds : aerosol precursors are therefore gases that, after several chemical or physical transformations, lead to the formation of aerosols. For our climate change purpose, the most important of them are :

sulfur dioxide (SO2), a well know local pollutant that is emitted as soon as we burn something containing sulfur, the main sources being coal and oil combustion (these fossil fuels always contain sulfur when extracted from the ground). This dioxide is then transformed in little sulphate (SO4) particles, that are solid.

to a lesser degree, nitrous oxyde emissions (NOx), essentially coming from africultural practices, that will lead to the formation of nitrate particles.

Aerosols have two effects on the climate system :

they reflect or absorb light, depending on the color of the particles they are composed of,

they provide "condensation nucleus", which means that the tiny particles they are composed of helps the atmospheric water vapour to condensate into droplets and therefore modify the formation of clouds that are themselves, as we have seen, aerosols of a particular kind. Clouds haver two opposed effects on, the climatic system.

How do clouds influence the climate system ?

composed of water and water vapour, they contribute to the greenhouse effect (see above),

but, on the other side, by preventing light to reach the ground, they have a "cooling effect" on the surface.

The precise balance of these two antagonist effects is of major importance for the temperature rise that will happen in the 21st century, and an accurate representation of clouds in climate models is clearly one of the major challenges of science nowadays.

It is nevertheless established that the greenhouse effect wins over the cooling effect for high altitude clouds (cirrus, and...plane contrails), that are translucid enough to let light through in significant quantities, but allready pretty opaque to infrared, but that the cooling effect wins over the greenhouse effect for low altitude clouds (cumulus, stratus...).

As SO2 has a tendancy to favorize the formation of low coulds, in addition to the contribution of sulphate particles to light reflexion, it is considered as a "climate cooler". This being said, it is also the coumpound at the origin of acid rain, that have strong negative effects on the environment, and our lungs dont taste it that much. It is therefore difficult to consider emitting massive quantities of this pollutant to oppose the effect of greenhouse gases ! Actually all public policies in the world are generally aimed at diminishing SO2 emissions, with effective results as it can be the ice of the polar caps.

Evolution of the sulphate concentration in the ice of the polar caps (milligrams of SO4 per ton of ice ; vertical axis on the right) since 1800. The vertical axis on the right gives the deducted annual SO2 emissions, in millions of tonnes of sulfur.

From IPCC, 2001

As for the greenhnouse gases, nature also knows how to emit aerosols, particularly through volcanism. For example, the Pinatubo eruption, that has sent in the high atmosphere a massive quantity of dust that remained there for a long time, has generated a perceptible decrease (0,1 - 0,2 °C) of the world average temperature for a couple of years.

In short, aerosols have direct effects on the radiation exchanges, and indirect effects through favorizing the apparition of clouds. Their precise contribution, globally negative (meaning that they globally "cool" the climate), is still at the core of an intense debate in the scientific community.


Greenhouse gases against aerosols : who wins ?

Part of the "heating" caused by the greenhouse gases that we accumulate into the atmosphere is therefore offsetted by the cooling of the aerosols and aerosols precursors that we also put in the atmosphere.

But the aerosols have an important characteristic that cause their effects not to last for long after they have been emitted : their residence time into the atmosphere is only a couple weeks (the water contained in a cloud eventually turns into rain and does not stay for years in the sky ; dust falls back on the surface of earth just as the dust send upwards by our broom will fall back on the floor - and the furniture ! - pretty quickly).

As a consequence, they don't accumulate in the atmosphere (as opposed to the greenhouse gases that do).

It is therefore certain that the effects of aerosols cannot offset the effects of greenhouse gases on the long term.

In addition, the atmosphere requires only a couple months to homogeneize its composition (for gases). Greenhouse gases, wherever emitted, therefore spread to the whole atmosphere in a matter of months (what is short compared to their residence time), and as a consequence the place where they are emitted is of no importance for their effect on the climate. This explains why international negociations are non avoidable.

It is not the case for the aerosols, that have an atmospheric lifetime of a couple weeks only, and that produce most their effects over the places where they have been emitted or formed (everybody can see that a cloud over Sydney is of little influence for New-Yorkers).

The diagram below gives the "radiative forcing" (that is the additionnal energy sent to the ground) for all the modifications induced by man that influence the energy exchanges in the atmosphere (in Watts per square meter).

From to to bottom it pictures :

the contributions of CO2,

the contributions of other "long lived" greenhouse gases : CH4, N2O, halocarbons, see above,

the contributions of stratospheric (that constitutes the famous "layer") and tropospheric ozone (that causes the "pollution peaks"),

the contribution of additional water vapour in the stratosphere resulting from additional methane oxydation in the upper troposphere,

the consequence of albedo change (albedo is what measures the reflexivity of a surface : a mirror has an albedo of 1 - all incoming light is relfected - and a black body an albedo close to 0 - all incoming light is absorbed) resulting from land use change. For example, when turning a forest into an agricultural soil, the albedo increases, because a forest generally reflects less light that crops or bare ground.

the direct contributions of various aerosols : sulphate, black carbon, dust...

the indirect effects of aerosols on cloud formation and lifetime

the indirect effects of contrail formation from aviation (that is excluding exhaust fumes other than water vapour).

though it is not anthropic, the consequences of the variations of solar activity (that is the variation of the energy the sun sends us).

For all these effects, the rectangles represent the most probable values, and the lines the error range. When the line is very streched compared to the lenght of the rectangle (or wwhen there is no rectangle at all !), it means that we have only a vague idea of the numerical value, let alone of the physical processes involved (bottom line of the diagram).

From IPCC, 2007


Back to climate change index
Back to the top