The two degrees C “dangerous interference” threshold – a meaningless metric

For some time now the world’s emissions-reduction strategies have been guided by the belief that global temperatures must not be allowed to exceed two degrees C above the pre-industrial mean (or slightly more than 1˚C above current temperatures) if we are to avoid “dangerous interference with the Earth’s climate system”.

Politicians are in general agreement that limiting warming to less than 2˚C is necessary to avoid redlining the Earth’s climate:

“The world (must) keep the global temperature increase from climate change below two degrees.” Ed Davey.

“Our joint goal (is) limiting global warming to under two degrees celsius.” German Environment Minister Barbara Hendricks.

“The U.S. continues to support the 2˚C goal.” Todd Stern, America’s top climate diplomat

And also that bad things are likely to happen if the climate does redline:

“experiencing global warming of as much as three or four degrees (could) lead to catastrophe, if not war.” François Hollande.

“A rise of 4˚C would be enough to wipe out hundreds of species, bring extreme food and water shortages in vulnerable countries and cause catastrophic floods that would displace hundreds of millions of people.” Ban Ki-moon

Clearly a lot depends on whether the world can keep global mean temperatures below the 2˚C “danger threshold”, assuming that 2˚C really is the danger threshold. Which raises the question, how good is the 2˚C estimate? Given the trillions of dollars that have already been spent trying to stay below it one imagines that it must be very good – a product of exhaustive research and the application of hard scientific principles.

But is it?

Unfortunately, it isn’t.

The 2˚C threshold originated with a study published by the German Advisory Council on Global Change for the guidance of the delegates who attended the UN Framework Convention on Climate Change in Berlin in 1995, and it hasn’t been changed since. The study itself, however, can only charitably be described as scientific. It begins with two questionable assumptions:

    • The late Quaternary period “has shaped our present-day environment, with the lowest temperatures occurring in the last ice age (mean minimum around 10.4°C) and the highest temperatures during the last interglacial period (mean maximum around 16.1°C).”
    • If this temperature range is exceeded in either direction, “dramatic changes in the composition and function of today’s ecosystems can be expected.”

Proceeds to a questionable conclusion:

    • The “tolerable temperature window” is therefore 10.4˚C to 16.1˚C. (The authors then widened the window by 0.5˚C in both directions to be “generous”, raising the upper limit to 16.6˚C, 1.3˚C  above the 15.3˚C global mean in 1995 and ~2˚C above the pre-industrial global mean).

And backs up its results up with no data:

    • No evidence is presented to support the claim that 1.3˚C more warming will cause “dramatic changes in the composition and function of today’s ecosystems”.
    • No evidence is presented to support the claim that the “Quaternary range of fluctuation” represents the “tolerable temperature window”.
    • The source of the 10.4˚C mean minimum and 16.1˚C mean maximum Quaternary temperature estimates is not given.
    • The justification for expressing Quaternary global mean temperatures to the nearest tenth of a degree is not discussed. (Realistically we would be lucky to get within a couple of degrees.)

By defining 9.9˚C-16.6˚C as the “tolerable temperature window”the study also implies that while the Earth and its ecosystems can’t tolerate more than 1.3˚C of additional warming they can weather an unusually frigid Ice Age without serious difficulty.

One would think that a study that reached conclusions like this without providing any backup data would have come under scrutiny, but the science, or lack of it, behind its 2˚C threshold estimate was never seriously questioned (although there was a lot of later wrangling over the amplitude, with some countries that felt particularly threatened, such as the Pacific Island nations, demanding 1.5˚C to be on the safe side). But after being formally adopted by the European Council in 1996 the rest of the world fell gradually into line, and the 2˚C threshold finally became the world’s official target at the Cancun Conference in 2010.

The study that gave birth to it, however, was published almost twenty years ago. What advances in scientific understanding of the “danger threshold” have there been since then? Well, science has concluded that the danger threshold is unquantifiable; you can’t put a number on it. Where you put it depends on how you define “dangerous” and how you assess risk, and risk assessments are sensitive to small changes. The “burning embers” diagrams of Smith et al 2011, for example, show risk levels increasing significantly between the 2001 IPCC Third Assessment Report and the 2007 Fourth Assessment report simply because the IPCC was more confident in 2007 than it was in 2001 that a risk existed (up from “more likely than not” to “likely” etc):

Smith et al. “burning embers” diagrams

And because risk assessments of this type are purely subjective (my purely subjective risk assessments are much less alarming than Smith et al’s, incidentally) they don’t give us any hard numbers, and even if they did these numbers wouldn’t be used because UNFCCC negotiators are already locked into 2˚C. As RealClimate puts it: If one wanted to sabotage the chances for a meaningful agreement in Paris next year, towards which the negotiations have been ongoing for several years, there’d hardly be a better way than restarting a debate about the finally-agreed foundation once again, namely the global long-term goal of limiting warming to at most 2°C.

Personally I think it would be an excellent idea to sabotage the Paris negotiations and start afresh with a global emissions-reduction plan that had some chance of working, but that isn’t going to happen. The politicians and policymakers are going to press on regardless, meaning that yet more trillions will be spent trying to stay under a 2˚C “danger” threshold that dates from a 20-year old study of highly dubious scientific merit and which for all we know may not be anywhere close to the true danger threshold. Hopefully the world’s emissions-cutters will at some point come to their senses and realize this, but don’t hold your breath.



This entry was posted in Climate change and tagged , , . Bookmark the permalink.

28 Responses to The two degrees C “dangerous interference” threshold – a meaningless metric

  1. edhoskins says:

    The 2degC threshold is entirely unlikely to be reached. This note explains why

    and this note shows how little man can do about it

    Never forget that the last millennium was the coldest of the current Holocene interglacial about 3 degC below the “Holocene climatic optimum” according to ice core records.

    That “optimum” was substantially exceeded some 120,000 years ago during the previous Eemian interglacial by about a further 3degC +

    The world thrived during the Eemian and Richard Tol’s assessment says that there is no world economic downside till beyond +2degC from now.

    As the Holocene interglacial is now cooling and its end at 10,000 years is overdue, a more likely future would be for natural cooling.

    • Hugh Sharman says:

      @edhoskins, I am glad you raise the ever interesting subject of the previous and warmer interglacials, during which life on earth, including our ancestors, thrived.

      On the other hand, since we are talking of geological time-scales, it must be noted that CO2 concentrations have not been so high for several million years and really we have no clue about its possible consequences. These may be benign, neutral or adverse for humanity; nobody knows for sure. All else is guesswork.

      I also agree that the Holocene interglacial must be due to end, “quite soon”. That will be curtains for industrial civilisation as we know it and catastrophic for humanity.

      However, ummm….”10,000 years overdue”?

      • Hugh:

        I also agree that the Holocene interglacial must be due to end, “quite soon”. That will be curtains for industrial civilisation as we know it and catastrophic for humanity.

        Not according to the German Advisory Council. They claim the Earth can tolerate a ~5C temperature decrease but not a ~1C temperature increase.

        It would be interesting to know where they would have set the lower tolerance threshold if CO2 caused cooling rather than warming. 😉

      • Raff says:

        Quite soon? As in, within the next millennium? Or century? Or decade? What are you saying here and how do you judge “quite soon”?

    • Raff says:

      That the world “thrived” during the Eemian is irrelevant to our predicament. What we are or should be worried about is the effect of changes on the 6 billion humans who were not around then.

      Also if you are going to quote Tol, please say whether you are quoting: Tol+gremlins – you know, the version of Tol’s paper that gremlins attacked and, amongst other errors, switched -2 to +2 to give the wrong answers, including positive effects of warming – which sceptics gave a lot of coverage. Or Tol minus gremlins – in which Tol fixed his gremlins’ work and updated to include more studies – you know, the version that showed no positive effect from warming – which sceptics gave more or less zero coverage.

  2. Graeme No.3 says:

    The ancestors of monotremes, marsupials and mammals were living 120 million years ago in the Cretaceous. So for that matter were birds. The ‘global temperature’ at that time is now estimated at 21℃. Previous guesses show that we haven’t a clue what it was, but that the Cretaceous was certainly warmer that today, with a wider range of habitable land, and micro-climates.
    The Quaternary Period encompasses the most recent 2.6 million years, and there were at least 60 rapid fluctuations in temperature. In the last half million years there were 4 interglacials warmer than the present, up to 3℃. The Holocene optimum about 9 to 7,000 years ago must have been warmer and wetter, especially for the Sahara Desert. Tassili frescos.

    All this is known to any scientist who spends a few hours (at most) checking the records. It looks increasingly obvious that Climatologists should not be called scientists.

  3. Sam Taylor says:


    I would like to know what factors, precisely, your subjective risk assesment is based on. The ecosystem is an incredibly complex system, and frankly it’s almost impossible to know how subtle changes in CO2 concentrations or temperature will alter various processes.

    As an example, a study ( found that elevated CO2 levels can alter the dynamics of a symbiotic relationship between a particular fungus and it’s host grass. The fungus produces ergovaline, which is toxic to cattle and thus acts to protect the grass from being overgrazed. Under elevated CO2 the fungus produced significantly higher amounts of this toxin, amounts which would definitely be in the harmful range for cattle.

    Given that our food system is dependent on many such subtle interrelationships. Failing to apply the precautionary principle seems, to me, foolish. But then I’m a member of the generation who’ll have to live with the consequences of this experiment, so perhaps I’m a bit biased.

    • Roger Andrews says:


      ….frankly it’s almost impossible to know how subtle changes in CO2 concentrations or temperature will alter various processes.

      That was entirely my point. We have no way of knowing. All we can do is make value judgments, and value judgments involve personal biases, and my personal biases would cause me to place the “danger threshold” higher than you would. That’s not to say that I’m right and you’re wrong. We’re probably both wrong. No one can put a hard number on it.

      But the politicians and policymakers have. They say it’s two degrees C. But they are no more likely to be right than you or I. And if this number is wrong they are going to squander a lot more money chasing the wrong target.

  4. Hans Erren says:

    Another concern is the Bern Model which assumes a sink saturation with high emissions, but so far the airborne fraction still doesn’t show any sign for an increase.

    This has ernormous implications for future warming, if combined with low climate sensitivity per lewis and Curry:

  5. Euan Mearns says:

    Find United Nations in 405 E 42nd St, New York, New York 10017 😉

    • That’s where the instruments are assembled. Component parts come from factories located at:

      Norwich Research Park, Norwich, Norfolk NR4 7TJ, United Kingdom

      Fitzroy Rd, Exeter, Devon EX1 3PB, United Kingdom

      2880 Broadway, New York, NY 10025, United States

      Telegrafenberg 31, 14473 Potsdam, Germany


  6. Euan Mearns says:

    In the “mind” of climate science, average global near surface troposphere temperature has become synonymous with climate and a change in that unmeasurable metric with climate change. I believe this is a totally wrong approach.

    For a start I believe one of the many processes at work is changing spectral output from the Sun with time that satellite measurements are only just beginning to see. This has impact upon where in the atmosphere incoming solar radiation is absorbed. One effect appears to be the geometry and activity of the sub-polar jet.

    This in turn I believe has the possibility to change the climate in many places without necessarily altering the global average near surface temperature. Put simply, some places may get warmer, some colder, some wetter and some dryer with the averages not changing much.

    I’m trying to write a post on CO2 greenhouse for beginners that is turning out to be much more complex than I thought. Tentative conclusion thus far is that CO2 greenhouse may affect the upper troposphere more than the lower troposphere. This may affect climate in ways as yet very poorly articulated by the IPCC and its followers. But its possible that Richard Lindzen has been barking up that tree.

  7. Raff says:

    Rising CO2 levels will increase temperatures by some amount. The amount could be as little as 1C or as much as many degrees C. The effect of a rise of many degrees C is also unknown to any reasonable accuracy but could be severe. And the risk of a bad outcome are high enough to make reducing CO2 a priority. To address this risk, an arbitrary best guess of 2C has been set as a target and to motivate action. The amount of CO2 we need to cut to achieve a 2C limit is not really known, what with all the uncertainties. And the willingness of countries to cut CO2 is based upon political considerations not on any accurate calculation. But you are are quibbling because the 2C limit might be wrong. Does that about summarise the post?

    • You agree that we don’t know how much warming CO2 might cause. You agree that the effects of whatever warming it might cause are “unknown to any reasonable accuracy.” You state that the 2C target is an “arbitrary best guess” (it’s actually pseudoscience, which is worse). You agree that “the amount of CO2 we need to cut to achieve a 2C limit is not really known, what with all the uncertainties”. Yet you seem to think it’s OK for the world to establish 2C as a target and spend trillions of dollars chasing it. Words fail me.

      • Raff says:

        Your attempted parody ignored the essential part – that the risk of a bad outcome is high enough to make reducing CO2 a priority. Do you have a method of establishing that the risk of a bad outcome is very small? Your method should, if you want it to be credible, not involve throwing away any sensitivity study that gives too high values. It should also involve accepting that all studies, even those of Lewis and co., give probability distributions that have long tails into high risk.

        You also assume that spending money on decarbonisation is necessarily bad. Prove it (and not just that is will be bad for FF companies).

        • Roger Andrews says:

          The point of the post was simply to demonstrate that 2C is a lousy metric to target. You did an excellent job of backing this conclusion up in your previous comment. I suggest you leave it at that.

          • Raff says:

            Really? I thought its purpose was to help “sabotage the Paris negotiations” by undermining the 2C target. Silly me.

            Targets are usually arbitrary. Their utility is in existing, not their exact details.

        • Graeme No.3 says:

          There is a risk that because it is made of green cheese parts of the moon will fall off and crash onto the flat earth raising disastrous tidal waves. The chances seem slim to me, but using your logic we should send tams to the moon to apply superglue.

          BTW do you realise the The Flat Earth Society believes in Climate Change. “By their company, ye shall know them”.

          • Raff says:

            If you can point to dozens of studies over twenty years pointing to a significant risk of the moon breaking up and then miraculously parts of it defying the normal laws of conservation of energy and momentum and somehow falling into the earth, then I would support your concern, if not your solution. If, as seems more likely, your example does no more than indicate your level of knowledge of physics, then I would doubt your opinions on this and any related subject.

        • A C Osborn says:

          Raff, you are a typical self opinionated warmist.
          You push the Precautionary Principle when simple logic says we should adapt as we have always done in the past.
          Then you say the that the OP should prove his assumption, when he does not have to prove anything to you.
          You also suggest that a post on a small UK blog (no offense to Euan) about a “Target” that was plucked out of thin air was to ““sabotage the Paris negotiations””.
          Conspiracy Theorist?

          • Raff says:

            In our situation, the precautionary principle is, if there is a risk we are in a hole, stop digging.

            On assumptions, it is clearly an article of faith amongst climate sceptics that action to reduce CO2 will be damaging (alarmists say it will lead to economic disaster). You may be prepared to believe that without any evidence, but I am not.

            And on sabotage, I said Roger’s aim was to *help* sabotage the Paris negotiations. Just like the various sceptical bloggers *help* delegitimise climate science, scientists, public bodies, learned societies etc by constantly throwing accusation, insinuation and so on. It seems to be a team sport.

          • A C Osborn says:

            Raffn that is the whole point, we are not in a hole, 18-26 years of increasing CO2 and no increase in Temperature and you talk about Sceptic “faith”.
            In the world that I inhabit, wasting $1.2Trillion on a form of energy that is too expensive and unreliable is definitely heading towards economic disaster.
            You do know that the USA has a national debt of over $17Trillion, have you worked out how much that represents to each USA person.
            You consider $1.2Trillion a good investment, I consider it wasted and could have been spent on improving the lot of humanity.
            But that you have already shown you do not care about that.

          • Raff says:

            AC how would you know what I care about? If you can think of some ways to improve the lot of humanity, head over to Bishop Hill where there is a discussion thread ( asking how to spend $1.7tn – you’ll be surprised to hear that people can’t name any ways. Since you know how the money could be spent, why not head over there and help them out.

            You classify the $1.2tr as wasted but can you explain what you think “wasted” means economically? If I buy some new shoes and after wearing them for a day decide they are too small, I’ve wasted my money. But for the economy, the money was not wasted. It had the same effect as if the shoes had fit well.

            On 18-26 years, you are wrong.

          • Euan Mearns says:

            Raff, I cannot see that you contributions here are adding anything by way of insight or analysis. I think it important to have a range of informed opinion and in the past year I have only banned 5 commenters. You are about to join them.

          • Raff says:

            Hi Euan, I thought we were having some interesting discussions and have no wish to be banned. I don’t believe I have been rude or given offence, but if I have broken blog rules or norms please tell me how I should respond more in line with expectations. Regards, Raff.

          • Euan Mearns says:

            Raff, your response to my comment is very reasonable and that buys some time. I don’t like banning commenters since it is seen as a form of censorship and in order to have a decent argument, which I enjoy having, we need someone to argue with.

            Your comments are polite but are too frequent and largely devoid of substance. At my time of writing you have 9 comments on this thread from a total of 25, i.e. 36%. And for that what have you said?

            Asking questions is always a good way of staying polite. But asking questions of every commenter on every detail is hogging bandwidth. It is best to 1) ask a question when you genuinely want to learn something or 2) ask a question where you believe someone has posted mis information.

            My advice to you is to limit yourself to a couple of comments per thread, focussing on the main, on-topic issues.

    • Craig: Thanks for the link to your thoughtful article. Just one comment. I wasn’t trying to portray climate policy experts as “fact twisters”. “Fact proof” would be closer, but since there are essentially no facts for them to be proof against “fact-indifferent” would probably be the best description.

Comments are closed.