par Eric Steig (traduit par Gilles Delaygue)
Note :Ceci est une mise-à-jour d’un article précédent, que beaucoup ont trouvé trop technique. L’original, ainsi qu’une série de commentaires, se trouvent ici.
Pendant les 150 dernières années, la concentration en dioxyde de carbone (CO2) a augmenté de 280 à 380 parties par million (ppm). Le fait que cette augmentation soit due pratiquement entièrement aux activités humaines est si bien établi qu’on le voit rarement remis en question. Pourtant, il est tout à fait raisonnable de se demander comment nous le savons.
(suite…)
L’une des façons de savoir que les activités humaines sont responsables de l’augmentation en CO2 est tout simplement de regarder les enregistrements historiques de ces activités. Depuis la Révolution Industrielle, nous avons brulé des combustibles fossiles, coupé et brulé des forêts à un rythme jamais atteint auparavant, et ces processus convertissent du carbone organique en CO2. Une comptabilisation minutieuse de la quantité extraite et brulée de combustible fossile, et de la surface déforestée, montre que nous avons émis beaucoup plus de CO2 qu’il n’en reste actuellement dans l’atmosphère. Les 500 milliards de tonnes de carbone, approximativement, que nous avons produit, sont suffisants pour augmenter la concentration atmosphérique du CO2 jusqu’à presque 500 ppm. La concentration n’a pourtant pas atteint ce niveau, parce que les océans et la biosphère ont la capacité d’absorber une partie de ce que nous produisons.* Néanmoins, c’est le fait que nous produisons du CO2 plus vite que les océans et la biosphère ne peuvent l’absorber qui explique l’augmentation observée.
Une autre méthode, complètement indépendante de la précédente, de savoir que ce sont bien les combustibles fossiles et la déforestation qui sont responsables de l’augmentation du CO2 pendant les derniers 150 ans est obtenue par la mesure des isotopes du carbone. Des isotopes sont simplement des atomes différents ayant le même comportement chimique (isotope signifie “même type”), mais de masses différentes. Le carbone est formé de trois isotopes différents, 14C, 13C and 12C. 12C est le plus commun. 13C représente environ 1% du total. 14C n’est présent qu’une fois sur mille milliards d’atomes de carbone.
Le CO2 produit par la combustion de bois ou de combustibles fossiles (hydrocarbures, charbon) a une composition isotopique très différente du CO2 atmosphérique. C’est à cause de la préférence des plantes pour les isotopes légers (12C par rapport à 13C), ainsi elles ont des rapports 13C/12C plus faibles. Comme les combustibles fossiles sont issus de plantes disparues, les plantes et les combustibles fossiles ont tous à peu près le même rapport 13C/12C –environ 2% plus bas que celui de l’atmosphère. Comme le CO2 de ces matières est émis et se mélange dans l’atmosphère, le rapport moyen 13C/12C de l’atmosphère décroît.
Les géochimistes des isotopes ont développé des séries temporelles de variations des concentrations atmosphériques en 14C et 13C. L’une des méthodes utilisées est de mesurer le rapport 13C/12C des cernes d’arbres, et d’utiliser ceci pour estimer ce même rapport pour le CO2 atmosphérique. Ca marche parce que durant la photosynthèse, les arbres absorbent leur carbone à partir de l’atmosphère et fixent ce carbone en matière organique sous forme de cernes, fournissant un instantané de la composition atmosphérique de cette époque. Si le rapport 13C/12C du CO2 atmosphérique monte ou descend, il en est de même pour le rapport 13C/12C des cernes d’arbres. Ce qui ne veut pas dire que les cernes d’arbre ont la même composition isotopique que l’atmosphère –comme mentionné précédemment les plantes ont une préférence pour les isotopes légers, mais tant que cette préférence ne change pas beaucoup, les variations des cernes d’arbre vont suivre celles de l’atmosphère.
Des séquences de cernes annuels d’arbres longues de plusieurs milliers d’années ont maintenant été analysées pour leur rapport 13C/12C. Puisque l’âge de chaque cerne est connu précisément** nous pouvons faire un graphe du rapport 13C/12C dans l’atmosphère au cours du temps. On trouve que jamais sur les derniers 10000 ans le rapport 13C/12C dans l’atmosphère n’a été plus bas qu’aujourd’hui. De plus, le rapport 13C/12C commence à diminuer de façon importante juste quand le CO2 commence à augmenter –vers 1850. C’est exactement ce à quoi on s’attend si l’augmentation du CO2 est bien due à l’utilisation de combustibles fossiles. De plus, nous pouvons suivre l’absorption du CO2 par les océans en mesurant le rapport 13C/12C des eaux de surface. Même si les données ne sont pas aussi complètes que pour les cernes d’arbres (nous avons seulement commencé ces mesures depuis quelques décennies), nous observons ce qui est attendu : le rapport 13C/12C des eaux de surface diminue. Des mesures de 13C/12C sur des coraux et des éponges –dont les squelettes carbonatés reflètent la composition chimique de l’océan comme les cernes d’arbres enregistrent celle de l’atmosphère– indiquent que cette diminution a commencé à peu près au même moment que dans l’atmosphère, c’est-à-dire lorsque la production anthropique de CO2 commença réellement à s’accélérer.***
En plus des données provenant des cernes d’arbres, il y a aussi des mesures du rapport 13C/12C sur le CO2 emprisonné dans les carottes de glace. Les données issues à la fois des cernes d’arbre et des carottes de glace montrent que la variation totale du rapport 13C/12C de l’atmosphère depuis 1850 est d’environ 0,15%. Ce qui parait très faible mais est en fait très important par rapport à la variabilité naturelle. Les résultats montrent que le changement total glaciaire-interglaciaire du rapport 13C/12C dans l’atmosphère –changement qui a pris plusieurs milliers d’années– était à peu près 0,03%, soit environ 5 fois moins que celui observé sur les derniers 150 ans.
Pour ceux qui désireraient plus de détails, voici quelques publications pertinentes :
Stuiver, M., Burk, R. L. and Quay, P. D. 1984. 13C/12C ratios and the transfer of biospheric carbon to the atmosphere. J. Geophys. Res. 89, 1731-1748.
Francey, R.J., Allison, C.E., Etheridge, D.M., Trudinger, C.M., Enting, I.G., Leuenberger, M., Langenfelds, R.L., Michel, E., Steele, L.P., 1999. A 1000-year high precision record of d13Cin atmospheric CO2. Tellus 51B, 170-193.
Quay, P.D., B. Tilbrook, C.S. Wong. Oceanic uptake of fossil fuel CO2: carbon-13 evidence. Science 256 (1992), 74-79
—————————
Notes
* Combien ils pourraient absorber sur le long terme est une question scientifique intéressante et importante, discutée en détails dans le Chapitre 3 du rapport du GIEC. Clairement, la cause fondamentale de l’augmentation observée depuis la période pré-industrielle provient de notre capacité à produire du CO2 plus rapidement que les océans et la biosphère ne peuvent l’absorber.
** Le développement de séries continues de cernes d’arbres remontant sur plusieurs milliers d’années, en utilisant des arbres qui se recouvrent dans le temps, est connu comme la dendrochronologie (voir les pages internet du laboratoire de l’Arizona sur les cernes d’abres pour plus d’informations).
***Un graphe illustrant les données des éponges se trouve ici. Merci à F. Boehm d’avoir indiqué ce lien.
You probably saw the review of Crichton’s book in the New Yorker recent issue. It was critical and alarming.
I was very pleased to see your web site. It is needed.
Thanks I had just read this today. It is a well-written review, and is available on line at the moment even if you don’t subscribe.
http://www.newyorker.com/talk/content/?050103ta_talk_kolbert –Eric
is it? pollutant? http://www.nationalcenter.org/2004/12/gerald-marsh-co2-no-pollutant.html
Response: It depends. Like the definition of ‘weed’ as a plant growing where you don’t want it to grow, CO2 in the atmopshere is clearly reaching undesirable levels. The same plant maybe a fine addition to a garden somewhere (as CO2 is a necessary ingredient for photosynthesis), but for that time and place and concentration, it is unwanted. Therefore it is clearly a pollutant in that respect. However, in the US, the word ‘pollutant’ has a legal meaning due to its inclusion in the laws governing the EPA. In essence, the EPA can only regulate ‘pollutants’, and so whether they have juristriction of CO2 emissions depends on the definition of CO2 as a pollutant. Hence the agressiveness with which certain people in the US pursue seemingly inconsequential semantic arguments with no real point. – gavin
Thanks for your initiative.
I would suggest that you give a link to Weart’s “The Discovery of Global Warming”, on the site of the American Institute of Physics. http://www.aip.org/history/climate/
(A Hyperlinked History of Climate Change Science)
Weart’s history is quite a remarquable work. It is also well-written and easily accessible for many. I believe that for many people the historical description of how one has come to a particular insight, (for example the question “How do we know Co2-increases are due to human activities”)might be enlightening. Especially in view of the current “debate” where many viewpoints and conclusions are presented by the press as if they came out of a hat.
Is there any evidence that carbon dioxide concentrations in the atmosphere, prior to the industrial revolution, ever approached the concentrations that we are seeing now?
[Response CO2 levels are currently higher than for any time when we have direct measurements (directly, from 1950; before that, from air trapped in ice cores), which amounts to the last 780,000 years (see, e.g., a picture here for the last 400 kyr). Various considerations suggest that in the far past CO2 levels were considerably higher. From memory, the last time CO2 levels exceeded present was about 40 million years ago – William]
This is Science, of course! Bravo!
Can you point me to the details of the calculation mentioned in this statement:
” The roughly 500 billion metric tons of carbon we have produced is enough to have raised the atmospheric concentration of CO2 to nearly 500 ppm.”
I’m especially interested in finding out (1) the extent (volume and mass) of the atmosphere into which the carbon mass was added and (2) the ranges of uncertainty associated with both the extent of the atmosphere and the mass of carbon produced.
[Response: the CO2 produced is spread throughout essentially the entire atmosphere (by mass at least). Atmospheric pressure is approx 100,000 Pa/m2 and from that you can work out the mass (about 5×10^18 kg). 500 billion tons CO2 = 5×10^14 kg ~ 1.5×10^15 kg CO2, so by mass that much CO2 increases the atmos by a/b, ie 3×10^-4, which is about right. These done off the top of my head, so please check (JBS?). The extent of the atmos (and of the bit which CO2 is spread through) is very well known, as are fossil fuel emissions – William]
To me the opening sentences of the second paragraph:
“One way that we know that human activities are responsible for the increased CO2 is simply by looking at historical records of human activities. Since the industrial revolution, we have been burning fossil fuels and clearing and burning forested land at an unprecedented rate, and these processes convert organic carbon into CO2.”
are not very clear as follows. The addition of carbon into the atmosphere by human activities does not automatically imply that the atmospheric concentration must necessarily increase. Might the natural sinks of carbon at some times be able to absorb the human additions and result in no net increase?
[Response: If we see CO2 increasing in the atmosphere, and humans emitting enough CO2 to account for that rise, then you have to go through some odd contortions to avoid a connection. You would have to postulate a suddenly increased natural sink (to remove the human CO2) and then a suddenly increased natural source (to increase the atmospheric CO2) – William]
Thanks for your assistance.
I don’t think there is any question that 6 billion living people, and rising, will have some effect on CO2 levels (not to mention CH4 levels). What seems clear, however, is that we don’t know exactly what that effect is. From the article above, the expected rise to 500 ppm hasn’t occurred. Nowhere near. But the precise reasons for the apparent shortfall are not known. So we really can’t say how much of an effect humans are having and whether they might be responsible for all or part of the current rise. The concensus that humans are responsible for all of the rise seems to be intuitive rather than scientific. I’d agree with the intuitive conclusion that humans are responsible for most of it, but that isn’t the story going out.
Response. How can I word my response strongly enough? NO, NO, NO. Intuition has nothing to do with this. This is one of the best understood aspects in the entire science of climate change, and it really is not that complicated! We understand very very well why the level hasn’t (yet) reached 500 ppm. If you add CO2 to the atmosphere some of it will have to go into the ocean. This is very basic chemistry. Imagine taking a bottle filled half with water and half with pure air that has no CO2 in it. Add CO2 to the air in the bottle. Some of the CO2 will dissolve in the water, resulting in less CO2 in the air that you have originally put in. We expect this to happen, and the isotope measurements demonstrate it IS happening, and at what rate.
In response to the question whether the natural sinks of carbon might compensate for the CO2 we are putting into the atmosphere, the answer is yes, but not very quickly. In the short term, the ocean cannot simply, magically, absorb all the excess CO2. If you try to pack all this excess CO2 in the surface ocean, it will come right back out. Again, that is what chemical equilibrium demands — there is no way around this. In the long term, the deep ocean will (probably) absorb much more of the CO2 we have put into the atmosphere, but this is the long term. If humans stopped producing CO2 today, it would take around 700 years to come back down to the original value. This is essentially because the timescale of ocean circulation is of order 1000 years.
If you really want to understand the details here, the college-level textbook by Kump and others “The Earth System” would be a good place to start.
–eric
Thanks for the response, eric. I hadn’t realised that the rate of absorption of all possible carbon sinks was extremely well known. Most reporting indicates that it isn’t. It seems that accurate reporting will be crucial in getting the right message across.
Response You are missing the point. If it is 4:00 p.m. in Los Angeles, we don’t conclude that rush hour is just about to end just because we don’t know precisely how many cars are getting off the road. All we need to know — to accurately predict that rush hour is just beginning and that we’ll be stuck in traffic for a while — is that the number of cars getting off the road is much smaller than the number getting on. See my comment below.
Tony,
Just to clarify further, the point is that all of the possible important carbon sinks are well known enough that we are certain we are not going to see the carbon we put into the atmosphere disappear any time soon. The numbers we’re talking about are very large, and there are uncertainties in those numbers that are themselves large enough to be interesting scientifically. Furthermore the uncertainties are important politically, because the accounting used in Kyoto tries to give credit to countries with larger sinks (e.g. the US, which argued that the regrowth of New England forests is an important sink, which is probably true — but not enough to outweigh the US sources!).
Let me give you some numbers:
Three numbers we know well are the following (data are from Takahashi, Science, Vol 305, Issue 5682, 352-353, 16 July 2004).
a) Emissions from fossil fuel combustion plus cement production is estimated to be 244 +/- 20 GT C (GT = 1 billion metric tons).
b) The amount of carbon in the atmosphere that is greater than the 1850 level in the atmosphere of 165 +/- 4 GT as of 1994.
c) The amount of C emitted to the atmosphere from land use changes (burning forest land, etc.) is about 140 +/- 40 GT
d) The ocean has taken up 118 +/- 19 GT of this carbon.
If you add these numbers up, you’ll find that 100 GT +/- 60 of C has gone into new growth in the terrestrial biosphere. Calculated this way, the size of that number is quite uncertain (we know it no better than +/-60%), but it is still clearly much smaller than the amount of C that has gone into the atmosphere and ocean combined . Independent calculations of how much C has gone into the terrestrial biosphere concur with this calculation.
Thanks for that eric. I didn’t like your analogy of the LA rush hour, but I take your point.
I may not be as good at math as I used to be but the figure I got, from what you wrote was 101 GT +/-83 of new terrestial growth of C. Excuse my ignorance, but how does this figure being much smaller that the amount that has gone into the atmosphere and ocean (283 GT C +/-23) give certainty that all atmosphere increases are anthropogenic? And are there any offsetting additional amounts of terrestial carbon (such as new forests or plant growth, or is this accounted for in the 3rd figure?).
(Unfortunately, I can’t check the Science source until next week)
Response The point is that more is being produced by humans than is winding up in the atmosphere and this is what we expect to happen. Sure, it is possible that not “all atmospheric increases are anthropogenic.” I never claimed otherwise. But to suggest that the natural environment has arbitrarily decided to start increasing the flux of CO2 into the atmosphere, right around the same time we are doing it is bizarre. Let me turn this around: what is the evidence that would suggest this is happening, and why?
This is a dynamic system and CO2 would not be constant even without human influences. There is CO2 continually leaving the ocean and entering the atmosphere in places where ocean upwelling brings carbon-rich waters to the surface. There is CO2 entering the atmosphere from plant and animal respiration. Indeed, on an annual basis, CO2 increases during N. Hemisphere winter because plant respiration exceeds photosynthesis. The points though are that a) CO2 began to rise when we starting producing it in earnest, b) its isotopic signature demonstrates it comes predominantly from fossil fuels, and c) such an increase has not happened in at least 800,000 years as far as we can tell.
As I said before, please read the scientific literature on this.