Perte d’ozone et réchauffement climatique

par Gavin Schmidt, traduit par Vincent Noel

Dans le débat sur le changement climatique, l’une des erreurs les plus courantes est la confusion entre la diminution de l’ozone et le réchauffement planétaire. Ceci n’est pas vraiment surprenant, compte tenu de la qualité discutable des informations propagées par les medias.

Depuis plusieurs années maintenant, les climatologues et météorologues tentent d’expliquer au public que si la planète se rechauffe, ce n’est pas parce qu’il y a davantage de rayonnement solaire qui passe par le trou dans la couche d’ozone. Les deux phénomenes sont cependant étroitement liés, ce qui complique l’adoption d’une stratégie pour s’attaquer à l’un ou l’autre de ces problèmes. Cette semaine, par exemple, un nouveau rapport de l’IPCC s’est interessé au possible effet de serre des produits chimiques (HCF et HCFC) utilisés pour remplacer les CFC dans les bombes aérosols et les systèmes de refroidissement dans le cadre du protocole de Montréal (et de ses amendements).


La dépendance climat / ozone a lieu dans les deux sens : Tout d’abord, les CFC, HFC, etc sont des gaz à effet de serre, tout comme l’ozone (qui de plus absorbe les radiations solaires dans l’UV). Un changement dans la concentration de ces gaz affecterait le transfert des radiations à travers l’atmosphère, et donc le climat. D’autre part, les processus chimiques qui contrôlent la perte d’ozone sont très sensibles à la température et l’humidité à l’échelle locale. Comme le changement climatique affecte directement ces propriétés, il affectera également la perte d’ozone.

Les CFC d’origine sont des gaz à fort effet de serre (leur forçage représente à peu près 0.34 W/m2 depuis 1850), et même en tenant compte du refroidissement créé par la perte simultanée d’ozone stratosphérique (-0.15 W/m2), l’effet moyen reste le réchauffement. La suppression des CFC contribue donc à réduire en même temps la perte d’ozone stratosphérique, et le forçage radiatif à l’origine du réchauffement climatique. A l’heure actuelle la concentration en CFC se stabilise, et on prévoit qu’elle devrait diminuer dans les décades à venir. Cependant, bien que les gaz utilisés en remplacement des CFC (par exemple le HFC-23) n’affectent pas l’ozone, leur effet de serre reste potentiellement important. On estime le forçage total de ces gaz faible par rapport à l’augmentation du CO2, mais toute réduction dans ce forçage reste bonne a prendre, car elle aiderait à contre-balancer l’impact du CO2. Heureusement, certains gaz de remplacement (par exemple l’ammoniaque) n’affectent ni l’ozone, ni l’effet de serre. Le remède contre la perte de l’ozone n’est finalement pas pire que le mal !

D’un autre côté, dans l’Arctique les effets du changement climatique sur l’ozone sont à la fois chimiques et dynamiques (voir ici). D’une part, l’augmentation du niveau de méthane et d’humidité stratosphérique affecte directement la chimie locale. De plus, le refroidissement stratosphérique (provenant de l’augmentation du niveau de CO2) accélère indirectement la perte d’ozone, en affectant la vitesse des réactions chimiques destructrices. Au niveau dynamique, l’activité des ondes planétaires et de gravité (reliées a la convection ou au courants jets, par exemple) affectent l’équilibre stratosphérique et controlent la circulation de Brewer-Dobson. Des changements dans cet équilibre pourraient modifier la circulation et la stabilité des vents stratosphériques. De tels effets dynamiques peuvent mener a des changements locaux de température (surtout aux hautes latitudes) plus importants qu’aucun changement d’origine radiative, en jouant notamment sur l’intensité du vortex polaire (Shindell et al, 1998).

Alors que nous en apprenons toujours plus sur l’ozone stratosphérique et le changement climatique, ces deux problèmes à l’origine clairement distincts deviennent de plus en plus dependants l’un de l’autre. Il semble clair que les climatologues devront s’habituer a expliquer comment les deux problèmes interagissent.

19 comments on this post.
  1. worthywads:

    It would appear from this article that the difference between ozone depletion and GHG isn’t even clear among some scientists.

    Idea: Genetically modify trees to grow 2-3+ times faster than normal to suck up co2 faster. Yet the final comment is “Clearly, a forest that grows twice as fast consumes twice as much and contributes to the shrinking of the hole in the ozone.”

  2. matt:

    And – not the only time I’ve come across this – why not throw tsunamis into the global warming mix too, as the article linked in (1) does?

    Is this just to grab the attention of the reader or are some supposedly educated people fundamentally ignorant of the natural environment? If the former, it is a nice illustration of how these confusions might come about…

  3. Lynn Vincentnathan:

    I read 4 years ago that NOx from driving & farming was also thinning the ozone layer over the mid-latitudes, especially in summer during agricultural season. I think there was also a mention of methyl bromide in farming as harming ozone. Are these points true? I think NOx is also a greenhouse gas as well.

  4. Eli Rabett:

    #3. The NOx contribution to the greenhouse effect is mostly indirect through production of ozone near the surface. You can find more than you want to know at Near surface ozone is a problem because of health issues, both for humans and plants, but is relatively unimportant in terms of UV exposure.

    Enough methyl bromide survives to make it into the stratosphere that it effects the ozone layer. On a molecule for molecule basis, bromine is much more effective than chlorine in catalyzing ozone destruction, although natural production is also large. Another information overload can be found at Somewhat more accessible is

  5. grundt:

    There is a feedback, indeed.
    I have read about that since about 1996-1998. Well, about the time when the link (Shindell) you post was published.
    Methyl bromide was going to be phased out some time ago (or will it be in 2010?). It is tremendously harmful. As Eli Rabett points, unfortunately has also natural sources, it would not be enough to ban it.

    Some people (mainly NGO), who were gathering before IPCC (in Argentina), 1998, denied the connection of GW and Ozone depletion.

    The Montreal Protocol gave the Industry the opportunity to replace CFC, and, as always happens, they hurried to develop new compounds that ended being harmful also. It is fortunate ammonia and other comps. are an alternative.

    The story with fossil fuels is a bit different.
    The post, Gavin, is very good. Neverending story.

    gavin “..So, as we learn more about stratospheric ozone and climate change, what were once two separate problems have become more and more entwined. It therefore appears unlikely that meteorologists are going to get a break anytime soon from explaining exactly how the two issues do, and don’t, connect.”
    Maybe you, meteorologists, will never have a break. Too many chemicals synthesized every year…..

    Good Luck

  6. Francis MASSEN:

    Ground ozone may be a health problem when concentrations are extremely high (say > 300 ug/m3 =~ 150 ppb), but rarely under normal conditions; most horror news on ozone related health issues boil down to heat-stress related problems, as high ground ozone nearly always happens in June/July/Aug when temperatures + solar irradiances are very high. I spoke several times to pneumologists, and none ever had a patient with an accute health problem that could be clearly tied to extreme ground ozone levels (contrary to some research papers and daily papers articles). Making the problem worse is that ozone prone situations go together with very unconfortable urban air conditions caused by traffic + industry emissions and urban heat trap: ozone as such becomes the sole culprit in the media, because this is easy to understand. Germany has been a model in this ozone hysteria during the last years ( GW has put this on 2nd range now… ). What makes the problem interesting, is that the highest ground ozone levels are usually measured in pristine non urban locations (remember that natural VOC’s (terpenes) + clean air + high UVB is all you need to get plenty O3 !), and these high levels are only rarely caused by upwind located industry . Here in Luxembourg (where I am measuring O3 since about 8 years, see for live data), the European alarm level of 180ug/m3 is based on the readings of the most rural station, assuring as such several “healthy” alarm situations per year( = good for the ego of the clean air bureaucrats); if the measurements of the city of Luxembourg (where most people live) would be taken, alarm events would be less frequent!
    Nevertheless, ground O3 levels should be carefully monitored (as they are increasing everywhere), without hastily launching ill conceived or simply silly mitigation strategies, and without publishing horror studies that may make a good frightening reading, but are scientifically non-sense.

  7. dave:

    Two Simple Questions:

    Stratispheric ozone depletion seems to link up with GHG forcing in the troposphere and concomitant lower stratispheric cooling. OK. From your cited Will spring 2005 be a bad … post, a quote:

    First, let’s go over some background. The recipe for massive springtime ozone loss in the polar regions, such as the annual ozone hole seen over Antarctica during the past two decades, is fairly simple. The two main ingredients are reactive halogen gases such as chlorine or bromine and sunlight. To prepare, keep the halogens at extremely cold temperatures, typically below -78 C (195 K). Use a strong polar vortex to mix the halogens to help achieve the required temperatures. When the mixture has been properly chilled, add sunlight and you’ll get rapid ozone destruction.

    So, my first question is this: Why is ozone depletion most pronounced at the poles and not nearly so serious in the mid-latitudes? Is it because the high latitude stratispheric temperatures are so much lower? Or, are the reactive halogens concentrated at the poles?

    And my second question is: if statispheric ozone is absorbing solar UV radiation and there is less ozone due to halogens and stratospheric cooling, what is the gain in solar insolation at the poles and is this significant?

  8. Joel Shore:

    In regards to the first question in #7: Go to and look at the “Scientific Assessment of Ozone Depletion: 2002” (in particular, question #10 in the “20 questions” section). The basic answer is that the chemistry of ozone depletion is such that it occurs much more readily where temperatures are very cold…in particular, cold enough to form what are called “polar stratospheric clouds”.

  9. Eli Rabett:

    A reply to #6.

    While VOC + UVB can produce ozone, adding NOx to an NOx free environment significantly increases ozone concentrations. Chemical radicals formed in the decomposition of the VOCs convert NO to NO2, whose photolysis leads to formation of O3. As in everything else details count.
    see pp 13 for a photochemical model, but the entire document is interesting and on point.

    At least in North America the highest concentrations are measured in urban areas. The maximum 1 hr US federal standard is 120 ppb (0.12 ppm) and 0.085 ppm over an eight hour period. These are often exceeded in the summer months, and in some cases have reached above 200 ppb. As I recall the area around Frandfurt/M airport is no bowl of cherries either. There is a significant literature on increases in asthma attacks correlating with ozone exceedances. A search of Pubmed turns up a raft of these, so, at least for concentrations typical of LA, Atlanta, Houston, etc. tropospheric ozone is a problem.

    A summary of the the number of exceedances can be found at
    and a summary for one of the “hot spots” the LA basin can be found at

    It is interesting to compare Charlotte and Atlanta to the Great Smokies (OK different years, but I think the principal is clear.

  10. paul gosling:

    At the risk of sounding really stupid. Has the decline in stratospheric ozone got anything to do with increased surface temperatures or is the increase in UV reaching the surface too small to have an measurable effect?

    [Response: Actually it’s the other way around. Stratospheric ozone loss has lead to a net cooling (due to the competing effects of ozone as a greenhouse gas and an absorber of UV). However, the CFCs which mainly caused this loss are powerful greenhouse gases, so the net effect of their emission has been a warming. This is significant, but still small compared to the effects of CO2 and CH4. – gavin]

  11. Hans Erren:

    re: #10

    The Hansen GHG forcing graph

    doesn’t contain ozone…

    [Response: This graph is for well-mixed greenhouse gases only. Ozone doesn’t count (since it is not well-mixed. The ozone effects can be seen in the graphs in instead. -gavin]

  12. dave:

    Re #11

    So, Ozone is not well-mixed in the troposphere, right? But is well-mixed in the stratosphere? Or at least was until people emitted CFCs,

    Anyway, (from the post)

    These dynamic effects can often lead to local changes of temperature (particularly in high latitudes) much greater than any radiative change e.g through possibly changes to the strength of the polar vortex (Shindell et al, 1998).

    I wonder what is meant by “local” here. Do these temperature changes go in both directions? I am wondering about these effects vis-a-vis observed increased melting in the polar glaciers and major ice sheets (Greenland, West Antarctica).

    [Response: Ozone isn’t well mixed in the strat either, and wasn’t before CFCs either. Well mixed means approx constant concentrations, which ozone clearly doesn’t have, because its short lived – William]]

  13. Ferdinand Engelbeen:

    re the use of ammonia and other non-GWP replacements:

    CFC’s were originally developed to replace extremely toxic ammonia in cooling equipment, as non-toxic, non-flammable alternatives. Because of their ozone depleting potential, they are replaced now by HFC’s and hydrocarbons like pentane. Ammonia was and still is used in industrial installations, but is strictly forbidden in area’s where non-professional people may be present, including residential areas.

    Hydrocarbons are far less toxic than ammonia, but have their own problem: highly flammable, they caused already several deadly accidents where used as foaming agent and during filling of installations.

    If used in tight applications with effective destruction at the end-of-life, HFC’s may be the best option.

    A similar problem exists with brominated halon’s: they are very effective in fire suppression, while non-toxic for humans, so one has a better chance to escape from fires in (computer) rooms, filling with halon. The CO2 alternative is deadly toxic for humans above a few % in air, that is before it becomes effective as fire suppressant… In this case there (still) are no effective non-toxic alternatives available.

  14. Ferdinand Engelbeen:

    Re #3,

    One need to make a difference between nitrous oxide (N2O) and the other oxides of nitrogen: NO, NO2,… mostly reffered to as NOx. They both have the same sources: fossil (and wood) burning, and fertilizer use. N2O is rather stable and can reach the ozone layer. It is considered a greenhouse gas and has ozone depletion potential (see: ). In contrast, NOx are strong oxidants and are reacting rapidely with organic material in the lower troposphere, causing smog and forming ozone at the wrong place…

  15. Ferdinand Engelbeen:

    In reply to #7:

    About your first question:
    Ozone is mostly formed in the tropics, where UV/light is at maximum input. Less ozone is formed toward the poles and no ozone at all is made during polar nights. There is some flow of air (and thus ozone) in the stratosphere toward the poles, which works until there, as long as the polar vortex doesn’t prevent that. Ozone depletion (natural and man-made) is more uniformly distributed all over the stratosphere, thus the effect is larger at high latitudes and highest at the poles. Natural ozone depletion was and is mainly (besides explosive volcanic eruptions) from methylchloride, which is formed by fungi and algue. While readily degraded in the atmosphere, the amounts are such large that this compound alone is responsible for 20% of the current ozone depletion, but most depletion now is from CFC’s.

    The ozone hole is specific for extreme cold temperatures (where chlorine condenses on particulate and is not reactive anymore) and the polar vortex which prevents air/ozone from beyond to replace the cold air within the vortex. When spring starts, chlorine is released at once, starting it’s ozone depleting work. After a few months the vortex is weaker, allowing fresh air/ozone to flow to the poles…

    About your second question:
    UV is less than 1% of the energy from the sun, reaching the earth. At the poles even much less, as most of it is filtered out during the much longer travel through the stratosphere, compared to the equator (the difference in UV at the surface between the equator and North Norway is 5 times, or 500%!). Thus even during the months of the ozone hole, the difference in energy reaching the surface is negligible. Which doesn’t imply that there are no important differences in effect on plankton and changes in the stratosphere (temperature, air circulation)…

  16. Lynn Vincentnathan:

    RE #14, thanks. It must have been N2O (not NOx) I read about that is thought to be thinning the strat. ozone over mid-latitudes during ag season & acting as a GHG. But then that’s during summer & over mid-latitutdes, so the cold needed to affect O3 wouldn’t apply as much. Still, perhaps the sheer quantity could be having an effect.

  17. Eli Rabett:

    A minor disagreement with Ferdinand Engelbeen on the mechanism for forming a polar ozone hole

    Much of the chlorine in the stratosphere is “stored” as ClONO2 where it is not available to react with ozone. Ordinarily this limits Cl catalytic ozone destruction. Polar stratospheric clouds are composed of nitric acid trihydrate (NAT) particles. ClONO2 is chemically converted to nitric acid and Cl2 or HOCl on these particles

    Much of the NO2 in the polar vortex stratosphere is converted to nitric acid. The ClO, which normally was stored as ClONO2 is then free to participate in ozone destroying cycles when unfrozen. Cl2 and the HOCl formed from ClO on the NAT are easily photolyzed at first light. All of this produces a catastrophic pulse of Cl and ClO which destroys ozone during the polar spring forming the ozone hole.

    In addition to the effect of a colder stratosphere produced by higher greenhouse gas concentrations in the atmosphere, Margret Tolbert
    demonstrated that the efficiency of NAT for denitrification of ClONO2 is greatly enhanced by incorporation of water vapor in the solid NAT. If
    the upper troposphere is wetter because of increased methane emissions and warming of the tropopause this would increase the depth of the polar ozone holes after cold enough winters.

  18. Francis Massen:

    a reply to #9:
    Ok, photolysis of NO2 creates O3, BUT the chemical reaction between NO and O3 is destructive: ( very crudely [O3] = k*[NO2]/[NO]). This can clearly be seen in comparing daily O3 variations in urban and rural areas: there is no photolysis at night, but heavy traffic in urban areas and minor only in rural areas: as such big nightly emissions of NO in urban, negligeable NO in rural areas: the result: during high O3 events, O3 levels fall down rapidly in urban areas ( e.g. reach morning levels at ~ 24:00), and much slower in rural country (maybe 05:00). I am not sure if anyone made a comparison between yearly urban and rural O3 doses ( dose = integral( [O3]* time) )… I stick to my statement that here in Luxembourg O3 levels are highest in rural areas. But sure, Luxembourg City is tiny, and its traffic minuscule compared to LA or Frisco Bay!

  19. Eli Rabett:

    Francis Massen makes good points. A place on the net where these issues are dealt with in some detail is especially section 1.3 which deals with the chemistry. Whether a particular situation is more VOC or NOx sensitive depends on the local loading of each (and the nature of the VOCs). One point to be made is that night-time destruction of O3 goes by O3+NO –> NO2 + O2, and, of course when the sun rises, the NO2 photolysis produces O3 again, so on net you are in the same place you started wrt O3 but 8 hours later.