Betting on climate change

A translation in Romanian is available here.

125 comments on this post.
  1. Dan Allan:

    I’m afraid I have significant issues with efficient market theory, both logically and empirically. I will forego the empirical cases where it has failed, and stick to what I perceive as the logical problem.

    Efficient market theory presumes, in effect, that, given the incentive making money (and the fear of losing it) a “correct” consensus will inevitably arise. With enough informed players, so the theory goes, every publicly traded equity will arrive at its “correct” price.

    Unfortunately, the theory fails to consider the possibility of all players misperceiving a situation in the same direction. Consider a simple case, the familiar math problem: “how many people are required in a room such that the likelihood is greater than 50% that two will have the same birthday?” If you give this problem to a room of math students, you will not only get many incorrect answers, but they will fail to congregate around the correct solution, because most who get it wrong will conclude that the number is a good deal higher than it is (if my memory serves, the answer is 21, which is paradoxically low. If I’m misremembering, I’m sure someone will be happy to correct me). There is nothing magical about allowing people to bet on their answers that will force the market “price”, i.e., the concensus response to come out to 21.

    Climate science is obviously orders of magnitude more complex than this simple probability problem, so the possibility of miscalculation is presumably that much greater. And there is no logical reason to assume that the miscalculations will somehow “average out”.

    I am in no way a skeptic about global warming. But I am very much a skeptic about efficient market theory.

    [Response:I'm not going to defend efficient market theory to the death. But it seems to me that your example doesn't work. If enough people bet on the wrong answer, people who bet on the right answer will make money/be rewarded. In that case, people would pretty quickly learn who to trust for the right answer - William]

  2. Chris Randles:

    I know what you are getting at Dan re #1.

    I suspect people betting on 184 would not be happy to part with much money on that basis. I did a simple but wrong calculation that said 52% chance with 20 people. A better calculation came out at 50.7% with 23 people but that is also wrong. Anyway it is possible that the distribution of opinions could be skewed in the opposite direction to what you intended.

    Yes I agree there is a problem with the possibility of lots of people misperceiving the problem. However there would be a powerful incentive to spot such situations and this is likely to lead to fast absorption of new information. Research may become more focused which may be good for decision making but bad for the science.

    How should sceptics, who believe the uncertainty is greater than that implied by the IPCC, act? Perhaps they might be more inclined to take out insurance bets rather than to gamble to win in the manner suggested. Could this be what we are seeing from the mentioned people who refuse to enter sensible bets rather than them actually being more alarmist than the IPCC?

    Perhaps we also need a market to estimate the amount of uncertainty in the estimate of warming and/or perhaps the amount of natural variability over a multi-decadal period.

  3. Tom Fiddaman:

    To amplify #1, individuals perform poorly on dynamically complex cognitive tasks, where feedback is remote in time and space, even in the absence of noise. In ordinary tasks this is not a big problem as one can learn through repetition and experimentation. Markets are similar – even if people make systematic errors in valuation, cash flow ultimately reveals the truth and errors are selected out within a time scale of at worst a decade or so.

    In a climate market, the revelation of information about the true state of the world is going to be slow, so foolish ideas could persist for a long time. Think of how oil markets have performed as signals of long run scarcity over the past few decades. There’s every reason to think that foolishness would arise if market participation is broad, given that people apparently don’t appreciate even the first-order dynamics of carbon accumulation – see Assessing Public Understanding of Global Warming.

    Seems to me that another big challenge would be to determine the methods and authority by which global mean temperature is judged, and maintain its integrity and institutional stability. The USDA maintains very strict control of information prior to crop reports, but the time scale of that problem is short. Perhaps the RealClimate team would agree to be sequestered on a remote island for 25 years if we gave them a really nice supercomputer.

    I don’t want to rain on the idea of a market though – if nothing else it could serve as a motivating aggregator of information and barometer of public opinion. That the skeptics won’t take fair bets is rather compelling.

  4. Doug:

    If I were a famous scientist, I would not take the bet for fear that it would make me look silly. In the end, the money won wouldn’t be worth it.

    In a case like this with the high odds suggested by Lindzen, it sounds reasonable to me. If the alarmists have 95% confidence of warming, they should have 98% confidence of no cooling, so 50 to one odds is an even bet. (I’m being approximate of course, I’m no statistician, but by an even bet I intend to mean the expected value of the bet equals the size of the bet.)

    So I guess the question is: Just how confident are the alarmists?

    Probability of 2 people with the same birthday:

    For 23 people .507
    For 30 people .706
    For 50 people .970
    For 60 people .995

    Birthday problem answers from Fundamentals of Probability by Saheed Ghahramani p 41 ex 2.6

  5. Chip Knappenberger:


    A couple of things.

    1) The bet proposed in World Climate Report ( that you refer to as “Michaels” was made in December 1998 and pertained to the trend in monthly UAH 2LT global temperature anomalies from 1998-2007. As you mentioned on your blog (, we cherrypicked the starting point to illustrate just how warm the average El Nino-elevated global average temperature was in 1998 ( More than five years later you have come to us with an interest in taking the bet, and then act somewhat surprised in our decision to ‘pass’ on it. Obviously, it would be quite advantageous to be able to get in on a bet at halftime while holding the house to its pre-game line. Had you approached us in 1999 things would have been different.

    2) As far as betting what the next 20 years holds in store in terms of temperature change the time period is too short to really start differentiating between trends that are too close together. For instance, in your scenario of a 20-yr temperature change of 0.3ºC +/- 0.18ºC, assuming a natural noise level (observed standard deviation of detrended annual global temperatures from 1977-2004) of 0.085ºC, a statistically significant difference in the trend that leads to the lowest end of your range (a change of 0.12ºC) and the trend that leads to the highest end of your range (0.48ºC) doesn’t begin to rise above the level of noise until around year 16 or 17. Thus, given natural variability, 20 years is only enough time to start tell apart (in a statistical significant fashion) trends that are at least disparate by about 0.15ºC/decade.

    But, perhaps that is enough for wagering purposes.

    Recall that in their 2001 Third Assessment Report, the IPCC gives a range of temperature increase between 1990 and 2100 of 1.4 and 5.8ºC based upon the simulated output from 7 different climate models run under 35 different emissions scenarios – each of which the IPCC claimed as having an equal probability of occurrence. At World Climate Report, we believe, that instead of having an equal likelihood of occurrence, that the temperature rise during the next 50 to 100 years will lie closer to the low end of the IPCC projected range than to the high end of the range and thus the overall impacts will tend towards the modest rather than the extreme. We have written this opinion numerous times (most recently here, However, the high end of the IPCC range (or even higher) is often waved about in order to hype the issue and draw “concern” from the general public such that they stand behind efforts to limit carbon dioxide emission. We contend that the high end of the range in unreasonable. So perhaps there is room for a wager.

    The IPCC temperature change range from 1990 to 2100 (11 decades) represents a range of rates of temperature change of 0.13ºC/decade to 0.53ºC/decade. Given the theoretical 20-yr signal-to-noise threshold of 0.15ºC/decade established above, in 20 years, a temperature trend that lies between 0.12 and 0.25ºC/decade (in the bottom portion of the IPCC range) should be able to be statistically differentiated from a trend that lies between 0.40 and 0.53ºC/decade (in the top portion of the IPCC range).

    Since the IPCC contends that all emissions scenarios are equally likely, there should be equal likelihood that the rate of temperature change should fall within either of the two ranges I just defined. Therefore, we need not take the odds to be any different from 1:1.

    In fact, let’s extend the low end of the low range infinitely downward, and the upper end of the high range infinitely upward (this should be amiable since there are numerous papers who suggest the high end of the IPCC range is too low and yet hardly extend the low end of the range downward) – doing so will allow the ultra alarmists and the ultra skeptics to have their positions included as well.

    Thus, the wager that I propose is as follows. Even odds. I (or whoever is on my side) get a future warming rate (as determined over the next 20 years) of 0.25 ºC/decade or less – indicating that I (we) believe that future climate change will be modest. And you (or whoever is on your side) get a warming rate of 0.40ºC/decade or greater-indicating that future climate change will be extreme. Anything in the middle (between 0.26 and 0.39ºC/decade) is a wash (since we can’t tell those values apart anyway).

    This should effectively separate out the alarmists from the realists.

    Does this seem reasonable?

  6. Brian S.:

    I don’t understand the estimate “0.3+-0.1C/decade for anthropogenically-forced warming over the next 20 years” – the linked IPCC site ( says .1 to .2 C over the next several decades. Can someone clarify?

    [Response:I suspect that it should read 0.3 over 20 years, not /decade. Fact-checking is in progress...; completed. "/decade" should not have been there and is now struck out - William]

  7. Ed:

    If there is a perception that necessary information is lacking, it devolves from making an investment decision into pure gambling based on chance. A skeptic of human induced global climate warming can reasonably view a bet on temperatures in 20 years as a game of chance, not an investment decision. That a handful of skeptics choose not to participate in this market may mean only that they do not play games of chance.

    [Response:Information isn't lacking - there is lots and lots of it. Knowing which bit is correct is lacking. As to your skeptic, it would depend on what sort of skeptic they were. If they were a skeptic that claimed to have information - many of them do, many of them claim to "know" that future warming will be lower than the IPCC consensus - then it should be possible to find odds for a bet - as the post says. If they were a skeptic that claimed that we have no idea at all about the future - warming is as likely as cooling - then again, there is room for a bet, since if you think warming/cooling is 50/50, then presumably your odds on warming-by-0.3-over-20-years would be far less than 50%, which again means there is scope for a bet. That all the skeptics (and calling them a handful seems unreasonable - it includes some quite prominent ones) so far contacted won't put their money where their mouth is seems quite significant - William]

  8. Tom Cecere:

    Surowiecki argues in “The Wisdom of Crowds” that if the bettors are sufficiently independent that the result is nearly always equal to or superior to the result of the panel of experts. The auction market does not depend upon classical efficient market theory as much as it depends upon people’s natural caution — the fact that losing hurts more than winning gives pleasure. This tends to tone down “wild” claims that normally take precedence either in a group of similarly-minded people or in the race to be noticed by the media.

    You can have dollar limits on the bets, not take them from corporations, etc., but all you’re doing is interfereing with the information flow that helps the accuracy. If, for example, Big CO2 Inc. wanted to put a billion dollars down on global cooling, 2 things would happen — the web would be full of “sign up and take some money from Big CO2 Inc.” emails, and, more importantly, some of the people who think it’s a slam dunk for global warming might temper their predictions just a bit because some clown put so much money down on the opposite outcome.

    None of this, of course, changes the science; but policy depends upon good predictive models.

  9. Thomas Palm:

    Tom, the problem with your scenario is that Big CO2 Inc might very well put down a billion dollar to push down the estimate, even if they suspected they would lose, because this reduced estimate of warming might mean less regulation and them saving more than a billion dollar.

    In theory one may consider that small CO2 inc might bet on high warming, because their more efficient industry emits less CO2 and would thrive with demands for reduced CO2 emissions, however, companies that profit from the current situation will have a lot more money available than companies that might profit from any future situation.

  10. Franz:

    I think it is a good idea, but it is by far not a solution to every problem that global warming creates. One drawback is, that you can only bet against the consensus. That means if you fear your island is going to be submerged by the rising seas, and everybody expects that it will be, there is no way you can be reimbursed for your losses. One can only hedge against outcomes which are still unknown.

    It also leaves out those who are creating the situation in the first place. One idea might be a global carbon tax that buys insurance against identified and not identified risks of global change. Of course there is little hope to get anything like that implemented in the current political climate.


  11. Michael Jankowski:

    Lindzen may have put his foot in his mouth (I don’t see a direct quote in the “Reason” article, so I guess it’s possible his wager invitation was taken out-of-context), but what about the rest?

    Most skeptics believe not only that much/all of the recent warming is natural but that we should be warming as part of the ascent from the LIA. Such people would want to leverage their bet in case “natural warming” occurs/continues. If the issue at hand is anthropogenic global warming, a skeptic will either want to remove the potential for natural warming from the equation or expect to be compensated more heavily for being correct.

    [Response: One of the classic fallacies advanced by climate change contrarians involves the notion of a 'recovery from the Little Ice Age (LIA)' or some variant on this theme. This is based on the simply false supposition that the LIA represents some perturbation to the climate from which it must naturally rebound. In fact, quite the opposite appears to be the case. As discused elsewhere on this site, modeling studies indicate that the modest cooling of hemispheric or global mean temperatures during the 15th-19th centuries (relative to the warmer temperatures of the 11th-14th centuries) appears to have been associated with a combination of lowered solar irradiance and a particularly intense period of explosive volcanic activity. When these same models are forced with only natural radiative forcing during the 20th century [see e.g. Crowley (2000)] they actually exhibit a modest cooling trend. In other words, the same natural forcings that appear responsible for the modest large-scale cooling of the LIA should have lead to a cooling trend during the 20th century (some warming during the early 20th century arises from a modest apparent increase in solar irradiance at that time, but the increase in explosive volcanism during the late 20th century leads to a net negative 20th century trend in natural radiative forcing). In short, given natural forcing factors alone, no ‘rebound’ from the LIA should have been expected at all. The only way to explain the upturn in temperatures during the 20th century, as shown by Crowley (2000) and many others, is indeed through the additional impact of anthropogenic forcing, on top of any natural forcing. – Mike]

    Let’s say I’ve got a friend who thinks he can bring rain with a song-and-dance and is willing to bet he can make it rain tomoorw. The forecast calls for an 80% chance of rain tomorrow – should I wager? Of course not. I’m pretty sure his rain dance won’t do the trick, but if Mother Nature is sending signals that rain is likely, I don’t want her rain to be confused with that of the rain dance. I’d much prefer to take that wager when the forecasts suggest a slim chance of rain (although 50% would be perfectly “fair”). Along the same vein, those who feel global warming is most/all natural variability shouldn’t/won’t take a straight-up wager unless the indications are that natural variability will be neutral or provide for a cooling effect. Are there any such forecasters, such as the solar cyclists?

  12. Dano:

    Back to the point, no one wants to put their money where their mouth is, irrespective of whether a person believes in efficient market theory. This is the point: when cornered, septics won’t back their beliefs.



  13. Franz:

    It may take some time until a market is set up, and betting about temperature might not be the most interesting one, because its connection to the damages is not yet fully understood. But it might be worthwhile to figure out the bets that are already going on.

    For example Exxon is extremely visible in lobbying against any carbon regulations and in financing climate sceptics. One could say that Exxon’s managers are gambling with the future of the company. Suppose there would be large damages that could be attributed to global change in 30 or 50 years. The scientist then could easily dissect any of the arguments which the sceptics are spreading now. There will be lots of documents and memos showing that the executives themselves did not believe the nonsense whose spread they payed. The McFagans of 2050 will happily go to court and bleed them dry. So how about put options on Exxon stock that expire in 30 or 50 years?

    If one wants to make money from expected climate change one could check the results of climate models. Which countries and areas might profit, where are the losers? How about put options upon real estate that we expect to become desert or on beach front property? The opposite ist probably harder to do. Can we identify areas that are barren now, and which might become green. Of course they should be stable enough that you can invest there.

    I think there should be some demand in consulting fonds in long term strategies that hedge against damages by global warming. It will take some time, but I think you are on the right track

  14. Dan Hughes:

    This is somewhat off-topic, but along the lines of #5, #6 and #7.

    Who or what is “Big Co2 Inc” and “Small CO2 Inc”? IMO that are no such things.

    It is all of us who demand products and services that require production of CO2 that are the cause of the “problem”. And in particular those of us in the “first world” countries, and within these countries those of us with above-average incomes consuming our above-average number of unnecessary products and services (computers, books, CDs, Internets, individual transportation, recreation, etc. etc etc.). We are “those creating the situation in the first place”.

    In the end, all of society pays for all regulations; businesses do not pay for regulations in a vaccuum. And we determine which businesses profit and which do not. If you don’t want to support “Big CO2″ don’t buy their products and services. That way you will not be among “those creating the problem in the first place” and won’t be paying a carbon tax either.

  15. Tom Cecere:

    Thomas Palm -

    Thanks for your comments. It seems to me, however, that no board of directors would approve a market dominating purchase because it would either ruin the market and simply tie up the billion dollars until the bet is over, or, more likely, the odds would be so attractive that tens of thousands of us would trip all over ourselves to put down $100 to get $10,000 on what looks like a sucker bet.

    And, adding insult to injury, as the years go on and global warming becomes harder to deny, BIG CO2 will have the regulatory/legal costs as well. I just hope I’m around long enough to see the day!

  16. dave:

    Re #6 decadal temperature changes

    Richard Kerr in Getting Warmer, However You Measure It (Science, Vol 304, Issue 5672, 805-807 , 7 May 2004) gives the surface trend as 1.7C/decade – reporting on Fu et. al. 2004. Fu himself cites the result of Jones and Moberg in the Tropical Glacier Retreat topic here at realclimate; their result gives 1.3C/decade in the tropics.

    I don’t know where Kerr got his 1.7 figure from. What is the “fact checking” William is doing to get this 3.0/20 years = about 1.5/decade number? How is this number baselined? This temperature calculation should also be rising decade to decade? Is it?

    [Response: This is obviously a slip of a decimal point. The warming trend over the last 30 years is 0.17 C/decade. -gavin]

  17. dave:


    Please convert my numbers in this post. 1.7C should be 0.17C, etc.



  18. Glen Raphael:

    Lindzen’s sort of lopsided bet seems appropriate and useful to me. Maybe a little agressive, but not hugely so. Bets like that would help us gauge how much confidence the alarmists have in their prediction. The big fear of a skeptic taking a bet is that the alarmists might turn out to be right by accident. The most reasonable skeptical view isn’t certainly in a particular prediction but merely some degree of confidence that you don’t have the certainty you claim and that the certainty you actually have isn’t sufficient to justify extreme measures.

    If alarmists are 95% confident that the IPCC temperature range encompasses the likely possibilities, they should be willing to offer 20:1 odds that the temperature will be in that range. If they’re only 90% confident, 10:1. If they’re only 66% confident, 2:1. But to offer no better than even odds on warming vs no warming is to agree with the strongest possible skeptical position – it’s saying that in your opinion we have no information better than flipping a coin as to whether it will get warmer. To turn around your criticism of Lindzen, your mouth claims 90% but your wallet claims 50%.

  19. Terry:

    Looks like we’re in luck! Junkscience is working out the terms of exactly the type of bet you are talking about. Looks like they are pretty serious. They are actively soliciting comments, so you should contact them directly.

    Please keep us posted on what bet you arrive at.

    See: … search for “Lindzen.”

  20. Joel Kuni:

    Your article is very interesting! If I understand your wager correctly, you want to bet that anthropogenic influences will be responsible for at least a tiny fraction of the warming of the earth that most people expect to occur in the next few decades.

    The fact that nobody has taken you up on this bet suggests the possibility that you are fighting a classic straw man. In other words, perhaps there are no scientists anywhere who assert that the activities of humans won’t cause at least a trivial, inconsequential, temporary amount of warming.

    Or maybe nobody is willing to accept your wager because nobody thinks you will be able to prove the extent to which any warming that occurs will be due strictly to the activities of humans and definitely would not have occurred if the last human on earth had dropped dead a century ago.

    Personally I would not accept a mathematical model as “proof” of such an extraordinary claim unless the model can accurately predict the daily temperatures in every major city on the globe at least five years in advance.

    All the models I’ve seen rely on the assumption that an increase in atmospheric greenhouse gases will necessarily increase the long-term average temperature of the globe and that all the other mechanisms that cause or counteract warming are understood and modeled fairly accurately. Yet that is the very issue under dispute.

    The fact that a particular model accurately predicts a warming trend does not in any way prove that the model accurately explains WHY the warming actually occurred.

    If my model predicts that my house will warm up after I light a match and the house does in fact warm up, it does not prove to me that the warming would not have occurred anyway due to some other factor I neglected to model such as the heat generated by the refrigerator. Nor does it prove that the house does not contain an air conditioner that will prevent the temperature from ever rising above a certain acceptable level.

    Without relying on a mathematical model to demonstrate that some of the warming of the globe has been due to human influences, what proof do you propose to offer when it comes time to settle the wager?

    By the way, it has not escaped me that nobody has accepted the wager offered by Chip in #5. His proposed wager and method of proof seem sound to me. Would those of you who refuse to accept Chip’s wager care to explain your reticence to the rest of us?

  21. Phillip:

    One of the problems with this analysis is the statistics. I recall this MIT paper that maintaine that the probability distribution is highly skewed towards higher temperatures.

    Although it looks at the IPCC horizon of 2100, rather than simply 20 years ahead, it implies there is effecively zero probability that the global mean temperature might be equal to or lower than that recorded in 1990.

    On the basis of the MIT study, if I was in Richard Lindzen’s shoes, I too would be demanding much longer odds than the author in this instance seems willing to offer.

    To put it another way, the author should quickly snap up the 50:1 odds on offer over a 20 year horizon.

    Massachusetts Institute of Technology
    Joint Program on the Science and Policy of Global Change
    Uncertainty Analysis of Global Climate Change Projections:

  22. Steve Bloom:

    Re #18: The alternative bet offered by Lindzen (warming over the next twenty years of less than .2c vs. greater than .4c, not “warming vs. no warming” as you misquoted it) was stated to be even odds per the IPCC. Note that the .2c figure just misses the low end of the IPCC scenarios. Yet he still wanted 2 to 1 odds? Presumably his version of an even-odds bet would allow him to win even if the IPCC is correct.

  23. geniusNZ:

    > If alarmists are 95% confident that the IPCC temperature range encompasses the likely possibilities, they should be willing to offer 20:1 odds.

    Haha you think the best things these people have to do with their money is make bets they are not making any money on? that would jsut be plain stupid Besides if you are lets say 90% sure and that is the corect odds no one will take a 95% bet that doesnt mean the politicians shouldnt act on a 90% probability – letting the market set the rate is more appropriate.

  24. Mark Bahner:

    James Annan writes, “My quick and dirty estimate above based on the IPCC TAR suggests that they would put the probability (of cooling) at more like 10%, so his offer actually appears to affirm the IPCC position.”

    Your “quick and dirty” estimate is completely wrong. The IPCC TAR puts the odds that the world will cool from 1990 to 2030 at MUCH less than 1%. By my “quick and dirty” estimate, it’s less than 0.2% (i.e., less than 1 in 500).

    People interested in a scientific analysis of the probabilities given in the IPCC TAR should see the analysis performed by Thomas Wigley and Sarah Raper, which was published in Science magazine in 2001 (“Interpretations of High Projections for Global-Mean Warming”). Interested readers can read the article for details about how the analysis was done, but basically Wigley and Raper presented probability density functions, based on all of the IPCC TAR scenarios. James Annan apparently hasn’t read this article, but here are the the probability values from Table 1 of that paper, for temperature increases from 1990 to 2030, in degrees Celsius.

    Percentage Temperature Rise (deg C)

    1% <0.36
    5% <0.48
    25% <0.66
    50% <0.80
    75% <0.95
    95% <1.17
    99% <1.31

    If James Annan truly thinks the IPCC TAR is correct, he should be willing to give odds of more than 500-to-1 against cooling from 1990 to 2030.

  25. Dan Hughes:

    My comment #14. I must have been looking at the wrong page. The citations to previous comments should be to #s 8, 9, 10. And now #15 too.

    I again say that we are “Big CO2 Inc”.

  26. Roger Pielke Jr.:

    James- Very nice work. For those interested in interesting questions about how IPCC contributors, non-IPCC experts, different disciplines, the public, etc. a prediction market can be a fantastic tool. I urge caution in interpreting very public bets among prominent people, which brings in a whole bunch of other factors other than those at play in a market setting (like the difference between a stock price and a stock picker). I really think that this could be an exciting development in understanding expert opinion on climate issues. Last comment, it’d be useful to focus the bets on variables of direct interest to decision makers. Global average temperature is not particularly relevant.

  27. Roger Pielke Jr.:

    One interesting point relevant to policy about bets focused on global average temperature (GAT) — it appears from this discussion that there is unanimous agreement among those who disagree on what future GAT will be that there is no need to discount for policy interventions. That is, the IPCC projections are not conditioned upon assumptions about policies being implemented related to emissions. A functioning market would make this more apparent (e.g., if the market price is equal to the IPCC midpoint scenario), of course, but it appears to me that in the debate motivated by James so far that there is no expectation that emissions reductions policies will have a discernable effect on the climate (as measured by GAT) by 2030 (the terminal point of the bet). From a policy perspective, this suggests a number of possibilities, among them, (a) that the benefits of emissions reduction (as related to GAT) lie beyond 2030, (b) that everyone implicitly agrees that emissions reductions are unlikely (I doubt this is the case), (c) that emissions reductions of any conceivable amount will have no noticable effect on GAT, and so on. If bettors expect that emissions reductions will have a discernable influence on GAT over this period, then it should be apparent in the market value for GAT, which would be less than the IPCC’s non-policy mid-point projection.

    [Response:I took me several reads of your first few sentences to make sense of them... but now I have, I partially agree. I would say that out to a short horizon like 2030, you may be mostly right. I think "the IPCC projections are not conditioned upon assumptions about policies being implemented related to emissions" is wrong, since there are a variety of projections, which are conditioned explicitly on the emission scenarios. Well, its in the SPM (of course it is, everything is :-)) fig 5. Which shows that though the emission scenarios vary a lot out to 2030/50, CO2 levels vary much less (of course). And then there is fig 5d, which I always misinterpret. Anyway, I wouldn't discount (b) as much as you seem to. And, of course, there is a lot of mental inertia behind IS92a - William]

    [Response: Actually, Roger is correct. IPCC SRES scenarios do not take into account any regulatory or legislative action to specifically curtail emissions. Their mandate was to provide plausible storylines for how emissions may develop in the absence of any such action. Like William though, I agree with conclusion (a). I doubt that the other statements are widely supported by anyone serious. - gavin]

  28. James Annan:

    Thanks to all for the comments.

    I’ll reply to some of the specific points in turn, but firstly, as Brian (#6) spotted, the 0.3+-0.1C estimate from the IPCC was of course the 20 year total, as I hope the linked page and rest of the calculation made clear.

    #4: Your sums are the wrong way round, if I believe cooling has a 10% chance (as per my rough calc) then I would certainly choose the cooling side of Lindzen’s offer. (I’m assuming that the probability of precisely 0 temperature change being either zero, or at least very very small – NASA GISS quotes to 0.01C.)

    #5: Sorry Chip, I should have included your name in that list. I’m glad to see that you are now at least prepared to consider a new bet.

    You seem to have gone to quite some length in creating an impressive straw man, which doesn’t really need dealing with in detail, since I can simply jump to the end-point and agree that warming in excess of 0.325C/decade is indeed unlikely over the next few decades.

    I hope that innocent bystanders have noticed how useful the betting paradigm has been in generating consensus – me, Chip Knappenberger and the IPCC report all agree that warming is unlikely to be greater than 0.325C/decade in the immediate future (probably “very unlikely”, in IPCC-speak). Wasn’t that easy?

    I’ll correct one detail of the straw man, since it seems to crop up a lot. Chip says:

    35 different emissions scenarios – each of which the IPCC claimed as having an equal probability of occurrence

    but the IPCC Special Report on Emissions Scenarios quite explicitly contradicts that claim, eg here:

    Preferences for the scenarios presented here vary among users. No judgment is offered in this report as to the preference for any of the scenarios and they are not assigned probabilities of occurrence

    This is reinforced again in the summary, even more explicitly (with my bold emphasis):

    Probabilities or likelihoods are not assigned to individual SRES scenarios. None of the SRES scenarios represents an estimate of a central tendency for all driving forces and emissions, such as the mean or median, and none should be interpreted as such. The statistics associated with the frequency distributions of SRES scenarios do not represent the likelihood of their occurrence.

    #18: Obviously the odds on the bet have to lie somewhere between the honest beliefs of the opponents. With Lindzen estimating 50% warming and me at about 90%, there is a lot of room (my calc is very rough but given the gulf in opinions, high precision should not be required). I actually offered 2:1 (=67%), not 50% as you claim. Lindzen would not drop below 98%, at which odds I would prefer his side.

    #24: Oh, hi Mark, good to hear from you again. Mark, have you replied regarding your CO2 bet yet? The one I explicitly ACCEPT although I would like to up the wager to something worthwhile? I do look forward to hearing from you on this.

    As for your ref to Wigley and Raper, I refer you to what the IPCC SRES actually says, as quoted in my reply to Chip. Their choices for probabilities of emissions scenarios are for them to defend if they wish. I’m not going to get involved in debating every paper in the field, or even some of the more important ones – the issue is to find what the market consensus is on global warming over the next couple of decades. So far, I’ve had a handful of offers which all apparently endorse the IPCC TAR and none which contradict it.

    #26,27: Sure, there are lots of ways of developing the idea in more policy-relevant ways (including different variables, and perhaps more interestingly, climate change bets which are conditional on policy), but simple SAT seemed like the most obvious place to start. Robin Hanson’s pages have a wealth of ideas.

  29. No de Qur&#039;Tuba:

    Putting Your Mouth Where Your Money Is
    In a discussion on Tim Lambert’s blog, Daniel Davies remarks:
    [blockquote]If anyone’s interested in what large and wel…

  30. Chip Knappenberger:


    Thanks for the comments in #28.

    As you point out, the IPCC does not explicitly state that the SRES scenarios all are of equal likelihood, but, given what they do tell us (as you indicated) 1) we should not rely on the frequency distribution to develop probability of occurrence, and 2) “No judgment is offered in this report as to the preference for any of the scenarios and they are not assigned probabilities of occurrence,” it seem to me that the best we can do is to make the simple assumption that they are equally likely (with departures from equal probability randomly distributed). If you have a better idea, let me know.

    As far as your call for a wager, it seems as if you are only interested in calling out the skeptics – a group in which you consider me (and Pat Michaels) to be part of. As we have written numerous times, our stance is clear – we believe that the warming during the next 50 to 100 years will lie close to the low end of the IPCC projected range.

    [Response:This seems an odd thing to write. For you, I don't know: could you quote some instances? For Michaels, it seems to be untrue: this is 3/4 oC, by implication over the coming century, which is well below the IPCC range. Could you quote something from M in support of your statement? - William (update: oops, I'm wrong: that was 50 years not 100: see #35 - William)]

    Thus, if you consider this to be in agreement with the IPCC, then apparently you can still be a skeptic and agree with the IPCC. Further, since you agree with us that the warming rate during the next several decades will be below 0.325ºC/decade, then, as I have pointed out, due to the level of natural variability, a 20-yr time period is too short to really differentiate between your beliefs and ours (if there exist any). This being the case, I welcome you to the group of “skeptics”, as evidently, you can’t prove yourself not to be.

    Since we (you, me, the IPCC) all are in agreement as to the likelihood of the rate of (near) future temperature change, the real impetus in your call for a wager should be geared towards calling out the alarmists – those folks who entertain the idea that the IPCC extreme temperature change scenarios are the most probable. After all, it is these folks who keep driving the press coverage and the political wrangling over this issue of anthropogenic climate change. So let’s see how many of them are willing to put their money where their mouths are.

    I propose to join with you and together we’ll issue a call to wager along the lines of which I outlined in comment #5 – even odds for those who believe that future climate change will be modest (an average 20-yr temperature trend of less than 0.25ºC/decade) vs. those who think it will be extreme (an average 20-yr temperature trend of greater than 0.40ºC/decade).

    What do you say?

  31. Mark Bahner:

    James Annan writes to me, concerning my comment (#24) on the probabilities for different degrees of warming in the IPCC TAR: “As for your ref to Wigley and Raper, I refer you to what the IPCC SRES actually says, as quoted in my reply to Chip.”

    Please provide a copy/paste of the exact paragraph of “what the IPCC SRES actually says.” Please do not simply link to the page, and please do not provide a copy/paste of only a portion of a sentence.


    [Response:From SRES Box 6.4

    There is no single most likely, "central", or "best-guess" scenario, either with respect to other SRES scenarios or to the underlying scenario literature. Probabilities or likelihoods are not assigned to individual SRES scenarios. None of the SRES scenarios represents an estimate of a central tendency for all driving forces and emissions, such as the mean or median, and none should be interpreted as such. The statistics associated with the frequency distributions of SRES scenarios do not represent the likelihood of their occurrence. The writing team cautions against constructing a central, "best-estimate" scenario from the SRES scenarios; instead it recommends use of the SRES scenarios as they are.


  32. Mark Bahner:

    Chip Knappenberger writes, “As you point out, the IPCC does not explicitly state that the SRES scenarios all are of equal likelihood,…”

    To quote from Wigley and Raper’s 2001 Science article I referred to in my comment number #24 (see page 452 of Volume 293 of Science, dated 20 July 2001, first paragraph):

    “For emissions, the SRES rport (2) states that ‘There is no single most likely, ‘central,’ or ‘best guess’ scenario, either with respect to the SRES scenarios or to the underlying literature’ and does not assign probabilities or likelihoods to individual scenarios [see also 93)]. WE THEREFORE ASSUME ALL 35 EMISSION SCENARIOS TO BE EQUALLY LIKELY.”

    Emphasis added. :-/

  33. Mark Bahner:

    James, please provide the COMPLETE quote of the warming rate given in the IPCC TAR summary.

    [Response: (Comment edited): From IPCC TAR:

    On time-scales of a few decades, the current observed rate of warming can be used to constrain the projected response to a given emissions scenario despite uncertainty in climate sensitivity. Analysis of simple models and intercomparisons of AOGCM responses to idealised forcing scenarios suggest that, for most scenarios over the coming decades, errors in large-scale temperature projections are likely to increase in proportion to the magnitude of the overall response. The estimated size of and uncertainty in current observed warming rates attributable to human influence thus provides a relatively model-independent estimate of uncertainty in multi-decade projections under most scenarios. To be consistent with recent observations, anthropogenic warming is likely to lie in the range 0.1 to 0.2°C/decade over the next few decades under the IS92a scenario. This is similar to the range of responses to this scenario based on the seven versions of the simple model used in Figure 22.


  34. Steve Bloom:

    Re #30: The betting issue aside, being one of those “alarmists” I guess I don’t share your confidence that (projecting out the high end of your figure) 2.37c in 2100 would be “modest” in the sense that we can reasonably presume it will lack major consequences. Hopefully you’re right. On the other hand, I buy insurance policies for risks I consider to be much lower.

  35. J. Sperry:


    In your response to #30, you appear to misrepresent P. Michael’s article, in which he clearly points out that the 0.75 oC estimate is for the next 50 years (from Oct. 2003), not over the coming century. And of course, this estimate is based on NASA’s Jim Hansen, who is not exactly in the skeptic crowd. Chip’s characterization that this is the low end of the IPCC range (min. of 1.4 by 2100) doesn’t sound odd to me.

    [Response: I concur - Michaels' estimate is for 2050. However, comparing it to Hansen's estimate is misleading and I discussed why very early on. Basically, Hansen is saying that is the smallest possible rise in temperature to be expected given sufficient regulatory and legislative effort (particularly to reduce methane, black carbon and CO2 emissions). Michaels' is claiming that this is the most that can be expected given no action whatsoever. I would hardly call that argeement. -gavin]

    [Response:I confess - Michaels is for 2050. I'm surprised: why is Michaels considered such a skeptic when he agrees with the IPCC (OK, I know, he's only just within the range and there is the tone to be considered). JA is right: this *is* bringing out how close everyones positions really are - William]

  36. Mark Bahner:

    James Annan has repeatedly represented the “IPCC consensus” for warming over the next few decades to be 0.1 to 0.2 degrees Celsius per decade.

    When I asked him to provide the complete quote supporting his representation, Gavin Schmidt provided this quote:

    “To be consistent with recent observations, anthropogenic warming is likely to lie in the range 0.1 to 0.2°C/decade over the next few decades under the IS92a scenario.”

    But all Gavin Schmidt’s provided quote does is show clearly and unequivocally that James Annan is WRONG. Let me repeat the quote, in it’s entirety:

    “To be consistent with recent observations, anthropogenic warming is likely to lie in the range 0.1 to 0.2°C/decade over the next few decades UNDER THE *******IS92a******* SCENARIO.”

    Emphasis added.

    That 0.1 to 0.2 degrees Celsius over the next two decades represents the warming ONLY for the IS92a scenario (under different climate sensitivities). This fact should be **obvious** to anyone who bothers to read the sentence carefully!

    I request that James Annan please admit his error. Failing that, I request that Gavin Schmidt, William Connolley to acknowledge James Annan’s error.

    [Response: You're wrong, as JA points out, since much of the warming over the next few decades is contrained by commitment and current levels and doesn't much vary by scenario. Now, how about you decide whether you're going to accept JAs bet or not? - William]

  37. Joel Shore:

    I saw Michaels give a talk in the physics department at the University of Rochester a few (~3?) years ago and the title of his talk was something like: “Global Warming: No Big Deal.” After a general trashing of various things including surface observations and climate models, he admitted that his prediction for the globally-averaged warming (of ~1.5 C by 2100) is within the IPCC range…albeit at the low end. However, he claimed that this warming would be no big deal because:

    (1) Most of the warming would actually occur near the surface in areas with shallow cold dry air masses, such as in Siberia and northern Canada where it would not have a large effect. [I forget if he had anything to say regarding melting of land ice and glaciers raising sea levels.]

    (2) Various other arguments that basically amounted to saying that the effects of the warming were being exaggerated.

    I don’t know if this has been his opinion all along or an opinion that he has evolved to as it has become less and less credible to claim that little or no warming will occur.

    [It is also interesting to note that Michaels conveys a quite different tone in such seminars as he does in his generally more strident editorials for the masses. In particular, he basically admitted that his was a minority viewpoint within the climate science community but mumbled various things about Thomas Kuhn and how hard it is to overturn an established paradigm...and something about public choice theory.]

  38. Chip Knappenberger:

    Gavin has, on several occasions, contended that Pat Michaels has taken Hansen’s (PNAS, 2001) warming estimate for the next 50 years out of context (for the latest see #35). However, Gavin’s characterization of Hansen’s estimate is wrong. Hansen IS NOT saying that a rise of 0.75 degC is “the smallest possible rise to be expected given sufficient regulatory and legislative effort.” Instead, Hansen says that 0.5ºC is the minimum expected warming value by 2050 and that the maximum realized warming will be 1.0ºC (assuming some public concern and action). He thus arrives at a expected warming value of 0.75 ºC +/- 0.25 ºC by 2050 given how he expects things (including public concern and action) to evolve over the next 50 years.

    Both sides keep claiming that the other has taken some of Hansen’s text out of context, so I include the whole pertinent section below:

    - Global Warming, 2000-2050. A byproduct of the above analysis is the conclusion that future global warming can be predicted much more accurately than is generally realized. We show elsewhere (8) that a forcing of 1.08W/m2 yields a warming of 3/4 °C by 2050 in transient climate simulations with a model having equilibrium sensitivity of 3/4 °C per W/m2.

    - We contend that a forcing much smaller than 0.85 W/m2 is unlikely, because fossil fuels are expected to be the primary energy source for at least several decades. Rapid introduction of nonfossil energies or CO2 sequestration might reduce the forcing by a few tenths of 1 W/m2. However, much of the warming in the next 50 years will be from presently “unrealized warming” caused by the existing planetary radiative imbalance of at least 0.5 W/m2 (8, 37). Slowing CO2 emissions in the second quartile of the century, although crucial for stabilizing atmospheric composition later in the century, would have only a small effect on the warming in 2050. These considerations suggest a minimum warming of 0.5°C by 2050.

    - At the other extreme, CO2 growth exceeding exponential at 1.5%/year would be inconsistent with historical trends and with the negative feedback caused by human concern about climate change. Thus the maximum CO2 forcing is 1.28-1.54 W/m2 (Table 1). BC and O3 are unlikely to be much greater in 2050 than today. Indeed, China has already begun to reduce its air pollution (38) and other developing countries are probably near their limits. Continued global warming would produce at least moderate public concern, thus limiting added forcing to about 1.5 W/m2 and realized warming to about 1°C.

    - Given these constraints on climate forcing trends, we predict additional warming in the next 50 years of 3/4 +/- 1/4°C, a warming rate of 0.15 +/- 0.05°C per decade. A slower warming rate will occur in the second half of the century, assuming that the climate forcing growth rate begins to trend downward before 2050.

    Quoted from: Hansen, J.E., and M. Sato, 2001. Trends of measured climate forcing agents. Proceedings of the National Academy of Sciences, 98, 14778-14783.

    [Response: I recommend that anyone interested read the whole paper (linked above). The CO2 minimum forcing estimate of 1.08 W/m2 by 2050 assumes flat emission growth (i.e. no further increases in CO2 emissions), and thus is the absolute minimum (and something I would be willing to bet against!). With continued growth of emissions, forcing estimates rise to up to 1.54 W/m2 but may be somewhat less. Other terms such as from CH4, O3 and black carbon are posited to decrease substantially as per the 'alternative scenario' outlined by Hansen in a number of publications. Unlike the IPCC scenarios, Hansen is actually trying to make a prediction of what is likely to happen. This is different to SRES because of a) the inclusion of the effects of policies specifically designed to reduce emissions and b) he implies that there is a probablity assigned to his scenario - i.e. he claims this is more likely that any particular SRES storyline. This is a perfectly valid exercise and does have use in the policy debate, however, it cannot be used to argue that anthropogenic forcing is not a problem and that no efforts should be made to tackle it (as Michaels implied recently on CNN). It is precisely because he thinks that unrestrained emissions growth would be a problem that Hansen feels confident in positing that some efforts will be made to restrain emissions.
    This is an entirely different argument than those that have been made by Michaels. In 1987 and again in 1997 (referred to here), Michaels claimed that climate sensitivity was between 1.0 and 1.5 deg C (for a doubling of CO2), thus projected increases of GHGs would only lead to a small change. Since the 1998, Michaels has insisted that temperature trends can be simply be linearly extrapolated out to 2050 and further and that the resulting (small) changes are nothing to be concerned about. Since then his argument has remained consistent, however the rate of warming has not. In 1998, Michaels was projecting 0.125 C/decade as the likely long-term trend (0.13 a year later), and now he claims 0.17 C/decade (~0.75/4.5). I would be interested in knowing what Michaels now considers to be the best estimate of the climate sensitivity, since if he remains convinced that it is much smaller than the number used by Hansen above (around 3 C for a doubling of CO2), then any claim of 'agreement' is coincidental in the extreme. -gavin]

  39. Mark Bahner:

    William Connolley writes in comment #35, “I confess – Michaels is for 2050. I’m surprised: why is Michaels considered such a skeptic when he agrees with the IPCC (OK, I know, he’s only just within the range…”

    I think you have a mispelling…don’t you mean, “septic?” (Isn’t that the word you routinely use?)

    [Response:You're thinking of this. Be more discriminant. Michaels is marginal for his k: he agrees with the IPCC (I and many other discover to our surprise) but does his best to disguise it by his choice of words - William]

    Isn’t the reason Pat Michaels is considered a “skeptic” obvious? It’s that he doesn’t think governments need to pass any new regulations to “stop” (or even slow down) global warming.

    [Response:Not as far as I'm concerned, since I'm talking about the science not the politics. Michael Tobis had a nice post which might help you - William]

  40. Steve Latham:

    Re #20 and some others: WAIT A MINUTE!

    “…you want to bet that anthropogenic influences will be responsible for at least a tiny fraction of the warming of the earth that most people expect to occur in the next few decades.”

    You mean that if the earth warms in the next few decades as the IPCC models predict (at least within the bounds of the prediction), this debate will be no closer to being resolved? Crap! On what non-anthropogenic basis do the skeptics expect the earth to warm over the next few decades? What’s the non-anthropogenic mechanism? Since the earth is now warmer than it has been for the last couple thousand years (or, less controversially, now that it is warmer than average), I would have thought that those who think anthropogenic global warming is bunk would expect the GAT to go back to a more ‘normal’ value.

    Maybe the skeptics will argue that there is some kind of inertia to the latest warming trend. On the other hand, in a laypersons debate on another site, the argument was made that the earth is now on a cooling trend since 1998.

    I think the ‘main debate’ should be whether or not there is anthropogenic global warming. People who agree that there is, but lean toward the low predictions, are not really skeptics at all, in my mind. After the ‘main debate’ is over, which I believe it should be, then you can haggle all you want about how much warming will occur.

    It’s not for climatologists to determine whether X degrees Celcius is worth worrying about or not. The fish I study (and the livelihoods and cultures of those who fish them) are already threatened by high water temperatures and the lack of glacial moderation of those temperatures. They might not be able to handle just the continuation over time of the recent high temperatures, never mind some additional modest warming.

    Please, ‘skeptics’, define your positions. It would be best not to lump y’all into a single category. And if you think that anthropogenic global warming exists and is in part responsible for recent trends, let the public know so that folks like the ones I deal with can properly lay at least some of the blame.

  41. Brian S.:


    It would either further the consensus or facilitate betting if you clarified that you do NOT think the temperature rise will be less than the IPCC TAR likely minimum increase of .1 C for the next 2 decades. In other words, either you agree with the floor (yeah consensus!) or there’s a significant spread between your likely range of temperature increases, and the IPCC range of .1 to .2 C/decade, which means there may be something worth betting on.

  42. Terry:


    What did Junkscience say when you contacted them? Have they responded yet? Do you think the bet he is suggesting is fair?

  43. Mark Bahner:

    “…and the IPCC range of .1 to .2 C/decade,…”

    The IPCC range is not 0.1 to 0.2 degrees Celsius per decade. That value only refers to the climate sensitivity around IPCC scenario IS92a:

    “To be consistent with recent observations, anthropogenic warming is likely to lie in the range 0.1 to 0.2°C/decade over the next few decades UNDER THE IS92a SCENARIO.”

    Emphasis added.

  44. dave:

    Re #16, 17:

    Gavin told me that in #16 that I had got the decimal points wrong, which I had acknowledged in the unfortunately posted #17. I had hoped you would do some editing.

    OK, all that aside, I STILL WANT TO KNOW, where do these decadel figures come from – Gavin says it is 0.17C/decade, the same number published by Kerr as cited in #16.

    OK – how is that being calculated?

    LOOK, if you’re going to present an entire thread talking about betting on a twenty year increase 0.3 over two decades, PLEASE tell us how these figures are being caluculuted, OK? And, I don’t want to hear about the IPCC (2001) estimate that states the that decadel changes will be between .10 and .20 per decade. That was 2001; this is 2005.

    What’s the latest stuff and how is that being calculated? If I’ve missed something in cited material, tell me so, and — otherwise respond.

    [Response:I presume you've been pointed to this before, but the answer for the numbers is in the IPCC TAR SPM. If you haven't read that document before, and are interested in climate change, then you should read it now. If you follow links to the future-climate-change chapter, you'll find where the numbers come from.
    You don't want to hear about the IPCC TAR? Well thats tough because its the best summary we have, until the AR4 comes out in a year or two - William]

    [Response: An additional note. Download the GISS temperature anomalies from and calculate the trend over the last 30 years. -gavin]

  45. Lynn Vincentnathan:

    These economic/gambling games are fun and revealing, esp. the odds the skeptics are willing to bet on that warming or cooling will occur. However, we should be cognizant of what the stakes are, what GW means, including loss of human life from GW, which is happening even now — 160,000 per year according to WHO. Even though economists can calculate a price on human life, e.g. for court settlement purposes (with a greater price for the rich and important people), and chemists can figure a human to be worth $2 in chemicals (that was 40 years ago, so it must be at least $10 now), there is no amount of money that can bring back a life. Also, money (whether in paper or coin or plastic card form) is not good when stranded on a desert island, or globally warmed earth that is greatly diminished of supportive biota and other conditions for life.

    As Roy Rappaport (“Distinguished Lecture,” American Anthropologist, 1993, p. 299) states: “Monetization…forces the great range of unique and distinct materials and processes that together sustain or even constitute life into an arbitrary and specious equivalence. Phenomena that relate to each other essentially in terms of their qualitative distinctiveness are represented and understood in terms of a logic that reduces all qualitative distinctions to mere quantitative differences, a logic that…attempts to “bottom line” the world”

    Nevertheless, if efficient market theory and betting on GW help build consensus, that’s great. I doubt poor drought-stricken farmers in India (& now in Australia), who have a high rate of suicide due to inability to pay loans, would be able to participate & hedge their future losses. We will also need to figure some monetary ways of facilitating GW mitigation & compensation/penalties. It’s just that none of this can bring back the dead or replace actual net agricultural losses due to GW each year.

  46. David:

    Chip your choice of language is “interesting”, when descibing change as “modest” “in the middle” and “extreme”. The smallest warming/sea level rise in TAR figure 5 will place a wide range of human and natural systems under very considerable pressure (and based on estimates of the melt-down point for greenland place us teetering on the edge of dangerous climate change). The mid range warmings will tip well into the realm of dangerous climate change, and the upper range; well it is pretty much game over for the world as we know it. You have also brushed over the fact that the warming rate increases rapidly in the mid and upper scenarios so using the centenial trend applied to the next 20 years is inappopriate.

    >However, the high end of the IPCC range (or even higher) is often waved about in order to hype the issue and draw “concern” from the general public such that they stand behind efforts to limit carbon dioxide emission.

    You could be accussed of the same. Your language suggests coping will be easy; we can only hope. For climate scientists I know, the low end is troubling enough; their ain’t a need to hype the upper end.


  47. Joel Kuni:

    In #40, Steve asks “On what non-anthropogenic basis do the skeptics expect the earth to warm over the next few decades? What’s the non-anthropogenic mechanism?”

    My answer – On the basis of billions of years of historical global warming periods, some of which happened rapidly and without any human influence whatsoever – and whose specific causes remain a mystery. For example:

    Around 14,000 years ago (about 13,000 radiocarbon years ago), there was a rapid global warming and moistening of climates, perhaps occurring within the space of only a few years or decades. In many respects, this phase seems to have resembled some of the earlier interstadials that had occurred so many times before during the glacial period. Conditions in many mid-latitude areas appear to have been about as warm as they are today”

    [Response:You've misunderstood the question. If you don't believe that CO2 forces climate change, then why do you think the temperature should go up in the future, rather than down? As of 150 years ago, the thousand-year trend was downwards if anything: . If you believe in random natural shifts, you presumably believe they are as likely to be downwards as upward. And the billion year scale is irrelevant. And the various YD and D-O type events probably only occur when the Laurentide ice sheet is around - William]

  48. Joel Kuni:

    Lynn, in #45 you refer to losses of human life that are supposedly already occurring due to global warming. Don’t forget to add back the lives that are being saved because there are fewer deaths due to hypothermia, frostbite, etc. Most of the warming occurs in the coldest parts of the earth and at night, not in the warm regions of the planet.

    Also, the longer growing season and the increased concentrations of CO2 obviously improve crop yields, thus preventing deaths due to malnutrition and starvation.

    The WHO apparently wants us to believe they know of people in New Delhi who died because, due to global warming that has already occured, the high temperature was 108.37 degrees rather than the pleasant 108 degrees it would have been otherwise. I guess they think we’re stupid.

    I would like to know how the WHO justifies showing pity for overheated equator-dwellers while showing no pity at all for the Inuit victims of frostbite and malnutrition in northern Canada.

  49. James Annan:

    Ok, round two.

    # 16,17,39 (dave): Sorry I thought your earlier retraction meant you’d found out. NASA GISTEMP is the obvious source of surface temperature data. As for new results, you (and I) will have to wait for AR4, I’m afraid. I would be surprised if it was very different to the 2001 TAR version as far as the next few decades go, but I certainly have no priviledged information about that.


    it seem to me that the best we can do is to make the simple assumption that they are equally likely (with departures from equal probability randomly distributed). If you have a better idea, let me know.

    You can make any assumptions you like, but the scenarios are essentially a question of socioeconomic forecasting and that is not my field. If the people writing the SRES had written your comment above, I would have probably taken their word for it. As it is, I view the scenarios as much as a choice as a random input – ie, how much carbon do we want to burn? Which is one reason for my focus on the 20-30 year time scale, which is essentially a matter of climate science and committed change.

    I’m sorry if you feel slighted by being bracketed with some of the more outlandish forecasts that cooling was more likely than not (Corbyn, Jaworowski) or even close to a 50% chance (Lindzen). It would perhaps advance the debate if you explicitly distanced yourself from these sort of statements. After all, you did offer to bet on significant cooling in 1998, a bet which you would have lost had anyone taken you up on it. Certainly, if you agree that ongoing warming (of 0.15C+-0.05/decade over the next few decades) is likely, then you don’t disagree with the IPCC TAR on that matter at least. But I think it would be very hard to reconcile your extended prediction of ~+1.5C in 2100 with your assumption that all scenarios are equally likely.

    I will happily take on any alarmists who say anything that I believe to be false (ie extreme predictions which I believe are substantially less likely than they claim). I would be interested (genuinely) to hear of any examples that you think provide a bettable proposition.

    Mark: I’m still waiting for your confimation of full details (including the max stake you are comfortable with) on this bet (see the last section) :-)

    As for the IS92a thing: yes, it is a reasonably middle-of-the-road scenario, although now a bit outdated. For the purposes of a short forecast, it doesn’t make a whole lot of difference – the IPCC TAR says 0.1-0.2C/decade with IS92a, and Wigley and Raper, using a different set of scenarios, get….about 0.2C/decade after 4 decades. Sorry, but I’m not going to get excited over a discrepancy (if you can call it that) of that magnitude. I guess there may be a gnat’s crotchet of disagreement if you look closely. If you want to bet against them, good luck to you. If you want to bet against me or the TAR, you’ll have to bet against what we (it) actually says, rather than what you want to pretend it says.

    So, calling for bets has apparently stopped people making unreasonable predictions, and replaced it with people putting unreasonable predictions into the mouths of their opponents. I guess we can call that progress, of sorts :-)

  50. Jim Carson:

    This thread seems a little close to getting “involved in political or economic implications,” but what the heck, it’s fun.

    Since no one’s actually addressed the legitimate point made by Michael Jankowski and others that while the bet might prove warming, it can never prove GHG-induced warming, how about we make a bet that does?

    James Annan, et. al. have increasing faith in their climate models, so let’s bet on that. James and company can put forth a model that uses any data set they like, so long as it’s independent, consistent, wide-ranging (good approx of global) and expected to remain so for the duration of the bet. Calculation of R-squared is made on the raw and modeled temperatures for the prior twenty years. The bet will then be whether the R-squared for the next twenty years will be higher or lower. The only serious point of contention should be the level of R-squared which proves the hypothesis of anthropogenic global warming. You might be surprised how low an R-squared I’d be willing to bet $1000 against. As for contracts, I’d suggest up-front purchase of US T-Bills, held by an interested law firm.

    This type of bet has several benefits:

    1) We’d actually be betting on GHG-induced warming, not just warming.
    2) We can shorten the duration of the bet, or place multiple bets (5, 10 and 20 years, e.g.)
    3) We can have a jolly old time charting the progress on this and other websites.

    For Lynn (post 45): I swear I don’t intend any sarcasm when I say, “Cheer up!” If you’re trying to persuade people, this is probably not the best forum for your message. If you really believe all that, you need to find a way to stop thinking about it because that much negativity will destroy a person.

  51. John Finn:

    Regarding the original article, Richard Lindzen is quite right to expect favourable odds. Betting odds are not just based on the likelihood of a particular result happening; they also reflect the popular appeal, i.e. the number of people who actually take (or are likely to take) the bet. These are not necessarily the same thing (e.g. England in the World Cup)

    As there is such a strong consensus for AGW (reportedly 98% at least), Lindzen’s 50 to 1 – although a bit high perhaps – is in the right area.

    On later comments

    I’m not sure why there is so much caution among the global warming advocates. James Hansen (along with Gavin) and others have recently reported that there is 0.6 deg C warming in the pipeline. The majority of this must become evident in the next decade or so. Atmospheric CO2 levels will continue to increase anyway. Barring an unprecedented run of volcanic eruptions the pro-warmers would appear to be the red-hot favourites. The odds need to reflect this

  52. Terry:


    Why no mention of the bet that Junkscience is proposing?

  53. Michael Jankowski:

    RE#11 response,

    We’re talking about skeptics making bets. A skeptic is going to wager based on what he/she believes, not what you believe. If someone believes we’re naturally coming out of a cool period (which Lindzen, etc, may believe), they’re much less likely to take a wager that we’ll be cooler 20 yrs from now or will want better odds than a flat wager. You may think you have supporting information that the LIA didn’t exist or wasn’t abnormally cool, that 20th century temps were abnormally high, etc. But if a skeptic disagrees with you on some of those points (and many/most would), then they very well could be expecting global average temps to warm naturally over the next 20 yrs.

  54. Joel Shore:

    Re #47 and 48:

    (1) You have given an example of a sudden warming from a glacial state to interglacial or near-interglacial conditions. However, what we are talking about is a sudden warming from what are already “balmy” interglacial conditions…a warming that, if it continues, will likely surpass the temperatures in previous somewhat warmer interglacials (like the one immediately previous to the current one ~100,000 years ago…when the sea level is believed to have been several meters higher than it is today).

    (2) Your example of a heat wave in New Dehli being 0.37 degrees warmer is a red herring. What we are talking about is not a warming that is completely uniform in space and time but one that makes extreme weather events such as strong heat waves considerably more likely. I.e., I do not think it is correct to simply conclude that you add 0.37 degrees, or whatever amount, to each daily temperature.

    (3) While your concern for the Inuit is touching, those folks are well-adapted for the cold climate that they live in and, in fact, they are one of the groups that is most concerned about severe negative consequences to their way of life due to global climate change.

  55. Dano:

    So, amidst all the hand-waving, has anyone determined whether Bahner indicated that he’ll take the bet? Has anyone actually nailed him down?

    I must admit I enjoy his spectacle and will miss the show when he disappears.

    So, anyway. Has he taken a bet?


  56. Christopher M. Nicol:

    I am a simple working class individual living in the suburbs. My question pertaining to global warming is common. I wonder how long it may be before the worlds societies are forced to drastic “near overnight” change? I am especially curious about the United States, as this is where I reside. How will this affect our lives, and more importantly at about which point should we expect catastrophic societal events. I ask this answer not in terms of years, but rather in generations.

  57. Brian S.:

    I can’t find the bet offer that’s supposed to be on Can anyone give an exact URL and/or reproduce it in the comments here?

  58. Michael Jankowski:


    I assume you’re trying to link GW to the droughts in Australia. I, too, read about the farmer suicide rates there. The only information I can find is a copy the same article at a variety of sites (such as this one, and it isn’t specific about the regions sufferent from droughts. It does, however, mention the New South Wales Famers’ Association and someone in Barellan (also in NSW). I also verfied here that NSW is under severe drought conditions thanks to low autumn rainfall .

    If this drought is due to GW, we should be able to see similar events due to warm global temps in the past, correct? Well, the best Aussie rainfall data I could find was here , and it’s limited to the period of 1910 through 1995. But I think in general that when you use the interactive site and look at rainfall for NSW and Australia as a whole both over autumn and the entire year, you don’t find a very good correlation between warmer global temps and droughts. You will also see that there is a tremendous amount of variability from year-to-year. The Aussie Bureau of Meteorology states: “Drought is also part and parcel of life in Australia, particularly in the marginal areas away from the better-watered coasts and ranges” and that “Many, but by no means all, droughts over eastern and northern Australia accompany the El Niño-Southern Oscillation phenomenon.”

    Also, see this other Bureau of Meteorology link showing the annual rainfall over Australia from 1900-2002. It would be tough to blame GW for a widepread Aussie drought based on that chart, which suggests that the 1990s – the warmest decade of the century (or in history, according to some) shows some of the more consistent rainfall totals from year-to-year and which appear to be overall well above the 20th century average. Just to check, I downloaded the data, and it’s true. For 1900-1999, the avg was 453mm with a standard deviation of 80mm (for 1900-1989, those figures are 450 and 80). For 1990-1999, those figures are 485 and 72, respectively. So if anything, the warmest decade of the century/history brought more rain and less variability. As the website states, “The high year-to-year variability of Australian rainfall dominates any background trends. Some of this variability can be accounted for by the El- Niño Southern Oscillation.”

    Last-but-not-least, I came across this Aussie study of suicides which shows a link between suicides and sunlight in Aussies. So maybe the reduction of global dimming is to partly to blame!

  59. Jeffrey Davis:

    I’ve followed this site for many months, and I have to say that this thread takes the cake for esoterica. I’ve no idea who is agreeing to what or what ideas are being advanced. Every phrase is hedged and qualified to the nth degree. All over imaginary money!

  60. Hans Erren:

    There is of course also the Nenana Ice breakup bet, which is held every year in April (or May)

    [Response: This is about climate, not interannual variations, which are effectively weather - William]

  61. Glen Raphael:

    The article has scrolled off the page, but you can still find it in the google cache. It didn’t specify a specific bet. It did point to a good recent Reason article.

    Bahner has officially posted a bet offer here. His prediction is that “the actual warming of the lower troposphere from 1990 to 2100, as measured by satellites, will be less than 1.94 degrees Celsius.” Splitting the difference between Michael Crichton’s prediction and one interpretation of the mean IPCC prediction.

    [Response:Bahner is weaselling, because his bet is only open "to members of IPCC" whatever that means. Insisting on going out to 2100 makes it less interesting/immeadiate than going to 2030. Giving any credibility to Crichton shows you where he is on the science. And putting in a value that is within the IPCC range is rather revealing - William]

  62. Steve Latham:

    To Terry,

    I went to the junkscience website too and searched for Lindzen (as you instructed) and couldn’t find the bet either. Was that just a trick to get some hits at that website? For something that’s supposed to be debunking bad science, the discussion of the science is not very prominent.

  63. dave:

    Re: Decadel Ranges and #49 James’ comment

    # 16,17,39 (dave): … you (and I) will have to wait for AR4, I’m afraid. I would be surprised if it was very different to the 2001 TAR version as far as the next few decades go, but I certainly have no priviledged information about that. [emphasis added]

    How can this be when each passing year (since the IPCC TAR) gets into the top 5 warmest years on record? Outside 1998, currently #1 because of the unusually strong El Nino, the next top 4 are 2002 (2nd), 2003 (third) and 2004 (fourth). And Hansen predicts 2005 will surpass 1998 even in the absence of a strong El Nino. Even if it doesn’t, it will probably displace some year in the top 5 (another wager!)

    We’re at 0.17C right now and I am assuming a 30 year moving average. 2 more years of this warming trend will surely move the decadel range higher than the TAR range of 0.1 to 0.2 for coming decades in the AR4. Assuming no really big volcanoes, of course! Given converging results from models for climate sensitivity (2.5 to 4.0C) and increasing GHG levels, decadel trends and predictions can only go up. Decreased albedo feedbacks and increased warming at high Northern latitudes alone guarantee your bet. I can only conclude that if you’re betting on climate futures, the time to buy is now. How about a AR4 oredicted decadel range of 0.14 to 0.24 C in 2007 for future decades? Any takers on that one?

  64. beth:

    Sorry, this isn’t science but catching up: I remember Chip K. from…

    [Response:This comment was interesting but, regrettably, doesn't come within the site policy so had to be removed - William]

  65. Dano:

    Glen, that’s an old bet.

    We want Bahner’s answer to James. Will he do the standard Bahner hand-wave and eventually go away, or will he actually allow himself to get nailed down to a specific?

    What’s his answer to James? What’s the bet?


  66. Terry:

    Here is the Junkscience discussion of the bet. I am sure James is deep in discussions with them as we speak. Keep us posted James. This has been the most interesting thread on this site yet.

    Hmm… how could such a wager be fairly structured? What positions could advocate and “contrarian” take? How would we determine the winner? What metrics could we use? Let’s see what we can come up with:

    Representing industrial-strength Anthropogenic Global-warming in the blue corner, let’s call them AG (please, no blind Patagonian Sheep jokes) and, in the red corner, we’ll have Usual Situation (US).

    So, we would have to assume AG would take the position of at least the median IPCC estimate of +1 °C over the 1990 figure by 2025 (our 20-year bet) and we’ve already hit a problem. The GISS Global Surface Air Temperature Anomaly (C) (Base: 1951-1980) suggests 1990 was a tad warm so we may be unfairly biasing against AG by using this figure. Let’s try the 5-year mean from that same record and make the bet on the 5-year mean for 2025 (not available until after 2027 because it averages 2 forward and 2 back from the target year). This looks a better proposition (ooh! sorry) since it will reduce El Nino and other transient effects.

    Now, we don’t have quite the number of records we’d like but -0.14 °C is the mean of the first 5 years of the record and is probably close enough for our requirements. US contends +0.5 °C over 11 decades is none too exciting and a continuation of that gentle recovery from the chill conditions of the Little Ice Age could lead to near enough +0.6 °C over 13.5 decades (to 2025, in other words) so 1990′s 5-year mean +0.11 °C.

    Defining our figures then we have AG punting for a target global mean temperature of 1951-1980 mean (14 °C*) + 1990 5-year mean of +0.36 + 1 °C for an aggregate of 15.36 °C and US tipping 1951-1980 mean (14 °C*) + 1990 5-year mean of +0.36 + 0.11 °C for an aggregate of 14.47 °C (rounded to 15.4 °C vs 14.5 °C here for convenience).

    Now we come to the problem of that darn “*” seen above. The estimated global mean absolute surface temperature 1951-1980 is 14 °C ± 0.7 °C. It’s like that because we’re really not that flash when it comes to determining the actual temperature of the planet yet and we could even be getting worse due to urban heat island effect (UHIE) corrupting the near-surface temperature amalgam (records such as GISTEMP are beginning to race ahead of atmospheric measures and the discrepancy appears to be worsening).

    Our situation then would appear to be AG punting a range of about 14.7 °C through 16.1 °C versus US with 13.8 °C through 15.2 °C and a mutually-claimed zone of 14.7 °C through 15.2 °C.

    What to do?

    We could declare the overlap a “no joy” zone, I suppose (should the house win or all bets off?). Then there’s unforeseen transients like explosive volcanic events in the preceding year or two – might need to declare such things as bet invalidators too…

    Climate’s not really a binary situation is it, rather more nuanced than usually presented and, given our precision difficulties, trying to work out just what is happening turns out to be looking through a glass darkly.

    What do you think? If you believe you can work out a fair wager for climate over 20 years, drop us a line. Workable entries may be published.

    [Response: I would just point out that the introduction of the mean global temperature and it's uncertainty is simply a distraction to the principle point here which is related to temperature anomalies. As JS knows full well, the anomaly field in any of the data sets being discussed is much better characterised than the mean. To paraphrase the offer, they appear willing to bet on a continuation of the linear warming trend calculated from 1900 (around 0.06 C/decade) compared to James's estimate of 0.1 - 0.2 C/decade. This may be doable.... -gavin]

  67. Eli Rabett:

    Anyone who takes this bet is a fool,

    “Bahner has officially posted a bet offer here. His prediction is that “the actual warming of the lower troposphere from 1990 to 2100, as measured by satellites, will be less than 1.94 degrees Celsius.” Splitting the difference between Michael Crichton’s prediction and one interpretation of the mean IPCC prediction.”

    given that the method/algorithm for determining the lower tropospheric warming is not specified. There are about 5 of them, and they vary from 0.08 to 0.24 C/decade. Which is the best representation is a good start to what passes for a bar room brawl. Note that Anan is tying his proposition to the GISS surface record, which, whatever else you want to say about it, is well described and stable.

  68. Joel Kuni:

    Most of the wagers proposed above seem to depend on future temperatures changes graphed as a function of time. As was noted by a couple of people, this implicitly assumes there won’t be any volcanic eruptions or any moderation of the increase of greenhouse gasses in the atmosphere. In my opinion, that puts too much uncontrollable risk into the wager.

    I propose that our wager should involve the behavior of temperatures as a function of the concentration of GHG in the atmosphere, rather than as a function of time, since that’s the foundation of anthropogenic global warming theories.

    I also propose we throw out any 12 month period following a volcanic eruption or the beginning of an El Nino event, neither of which are theorized to be caused by humans.

    Furthermore, since global warming models uniformly predict that the greatest anthropogenic warming will occur in the winter months in the driest, coldest climates, how about if we only look at average winter temperatures in the region of Anarctica, plotted as a function of mean GHG concentrations in the atmosphere of that continent?

    If global warming alarmists are correct, when we graph temperature as a function of GHG we should expect to see not only that a least-squares trend line fitted to multi-year data has a strong positive slope, but more importantly we should see a convincingly large value of R squared.

    I propose (tentatively) that I should lose the bet if the slope of the trend line is positive AND if the value of R squared is greater than 50%. (If it’s less than 50% it would imply that less than half the increase in temperature can be explained by the presence of greenhouse gasses, which implies to me that the earth has adequate defenses to counteract any warming forced by humans, in the long run.)

    If this wager seems fair to Those Who Know About Such Things, then I have a question. Don’t we have enough historical data to settle the bet right now, based on observations collected during the past 100 years or so? What do you suppose we will find if we look?

    Based on the skimpy data I’ve been able to locate (in a search lasting upwards of three minutes) I suspect I will win my bet handily. If I do, will that persuade those of you who tend to side with the alarmists?

    If I lose my bet, I will readily concede that Inuits should reduce their greenhouse gas emissions if they place less value on the human lives global warming will save from frostbite than they place on their “way of life” (which involves GHG-belching snowmobiles, according to ).

  69. Stephen Berg:

    Re: #48,

    “Lynn, in #45 you refer to losses of human life that are supposedly already occurring due to global warming. Don’t forget to add back the lives that are being saved because there are fewer deaths due to hypothermia, frostbite, etc. Most of the warming occurs in the coldest parts of the earth and at night, not in the warm regions of the planet.”

    This statement is utterly ridiculous! Those who live in these colder climate regions have adapted and really have little chance of dying of hypothermia and frostbite. The Inuit of northern Canada and Alaska are completely adjusted to the cold climate, such that a warming of their climate would result in the destruction of their way of life (many animal species on which they rely for food and clothing will be drastically reduced), which may result in the reduction in numbers.

    “Also, the longer growing season and the increased concentrations of CO2 obviously improve crop yields, thus preventing deaths due to malnutrition and starvation.”

    Another ridiculous comment, since the “improved crop yields” in some fields would be offset, and then some, by desertification and the spread of disease and severe weather which will likely result (and is possibly resulting today).

    “The WHO apparently wants us to believe they know of people in New Delhi who died because, due to global warming that has already occured, the high temperature was 108.37 degrees rather than the pleasant 108 degrees it would have been otherwise. I guess they think we’re stupid.”

    Ummm. No. Look at the heat wave in Europe a summer or two again. That is what is happening and will likely get worse. Tens of thousands died there from the heat.

    Also the 108.37 degrees was the climatic AVERAGE, not the high temperature on a day-to-day basis. This means that there were likely extremes of daily temperature (perhaps 125 or so and then 85 or so), which would have resulted in the deaths of thousands.

    “I would like to know how the WHO justifies showing pity for overheated equator-dwellers while showing no pity at all for the Inuit victims of frostbite and malnutrition in northern Canada.”

    Again, the Inuit are not victims of frostbite! This comment shows extreme ignorance of any sort of reality.

    However, the Inuit are likely beginning to suffer from malnutrition due to the decreasing populations of healthy animals on which they rely for food as a result of the local climatic effects of global warming.

  70. Lynn Vincentnathan:

    RE #58, I wasn’t necessarily attributing current Indian & Australian droughts to GW, only saying that farmers already in hock may not be able to hedge their future losses from GW. And I don’t think WHO is attributing GW-linked suicides & other deaths more indirectly linked to GW, so my guts tell me they may be underestimating the number of deaths. I have the feeling that as science improves & looks into all angles in the future WHO will look back on our era & say that deaths due to GW were much higher than they had estimated (even subtracting lives saved by GW).

    I do know that the warmth of the ocean does play a role in whether the monsoon drops rain over the ocean (when the ocean is a degree or so warmer) or drops rain over India, giving enough rain for its crops. I do know that GW is expected to have a tremendous impact in the future – also from melting Himalayan glaciers, which will increase flood damage in winter & leave them no water in summer for irrigation (maybe not even for drinking). Of course, there are many other environmental problems harming India & elsewhere, in addition to GW.

    I’m not sure if the ocean temps likewise affect Australian droughts. Does Austalia get monsoons???

    I guess I’d say that if the current droughts in Australia & India are not in any part due to GW (& I think they might be), then we can only expect much worse in the future. So why are we playing RUSSIAN ROULETTE re GW, when we could be saving mucho money from energy efficiency & conservation. Luckily I’m not a gambler, I’m a frugal person, so I’d rather save money AND reduce my GHGs. Win-win is much better than zero sum games.

  71. Joel Kuni:

    Re #69:

    “Traditionally, Arctic hunters were at risk of hunting accidents (e.g. animal attacks, shooting and boating accidents) and death from exposure or hypothermia.”

    Please provide the source for this scientific statement: “Also the 108.37 degrees was the climatic AVERAGE, not the high temperature on a day-to-day basis.”

    I am astonished to learn that the average climactic temperature has been so high without my having noticed it!

  72. Stephen Berg:

    Re: #71,

    “Traditionally, Arctic hunters were at risk of hunting accidents (e.g. animal attacks, shooting and boating accidents) and death from exposure or hypothermia.”

    Arctic hunters were, but those who stayed in the home were not, unless they ventured out while a winter storm was occurring. The Inuit were, and are, aware of what the harsh weather could do to themselves.

    They passed along knowledge and tales of those who survived and those who perished, which educated them at an early age of the dangers of their surroundings. This improved their odds of surviving.

    However, now, with the increases in temperature over the last 150 years or so, their environment is changing, which is increasing the risk of their culture being destroyed and will also likely render this past knowledge useless.

    Talk to Sheila Watt-Cloutier and ask her whether global warming is likely to be positive or negative. (Very negative. I’ve heard her speak and she tells of the probability of the Inuit way of life being extinguished due to the activities of people to the South, and NOT because of their own activities.)

    Talk to those who live on the coasts of Alaska by the Bering Strait and ask them how much their lives are improving. (They are not improving, but becoming increasingly threatened. Read Mark Lynas’ book “High Tide” for a story on Alaska and the Alaskan Aboriginals.)

    “Please provide the source for this scientific statement: ‘Also the 108.37 degrees was the climatic AVERAGE, not the high temperature on a day-to-day basis.’

    I am astonished to learn that the average climactic temperature has been so high without my having noticed it!”

    I was just quoting your post from earlier. It was if you had said New Delhi experienced 108.37 degrees instead of the 108 degrees from before climate change. I assume you made this scenario up, now.

  73. Eli Rabett:

    This thread appears to be a fine example of putting your mouth where your money is not. Where are the bets? (who could resist?, feel free to trash this post)

  74. Mark Bahner:

    I will be commenting much, much, MUCH more on the subject of “Betting on Climate Change,” as my free time allows. I’m currently working on posts for my own weblog, Roger Pielke’s “Prometheus,” and the Google alt.sci.environment discussion group.

    However, I’d like to address this issue:

    Glen Raphael commented (#61), “Bahner has officially posted a bet offer here. His prediction is that “the actual warming of the lower troposphere from 1990 to 2100, as measured by satellites, will be less than 1.94 degrees Celsius.” Splitting the difference between Michael Crichton’s prediction and one interpretation of the mean IPCC prediction.”

    William Connolley responded, “Bahner is weaselling, because his bet is only open “to members of IPCC” whatever that means.”

    I wonder why William Connolley didn’t go to the link Glen Raphael gave, and then go to the “Discuss this prediction” section? That would be the LOGICAL thing to do, if one had the question, “What does ‘members of the IPCC’ mean?”.

    [Response:You wonder why I didn't follow a tedious chain of non-obvious links, when there was a big message stating very clearly "open to IPCC only?". Oh come along - William]

    If William Connolley HAD gone to the bet discussion section, he would have seen this comment, which I posted two days ago (June 15):


    “When I made this bet, I stated that it was only open to ‘members of the IPCC.’ I meant ‘any authors of the IPCC Third Assessment Report (TAR).’”

    “However, when I wrote that, I thought that Gavin Schmidt and William Connolley of the website “Real Climate” were primary or secondary authors of the IPCC TAR. My understanding now is that Gavin Schmidt was not, and I don’t know about William Connolley. In any case, this bet challenge is specifically open to both of them. I am predicting that the warming of the globe, as measured by satellite measurements of the lower troposphere, will be less than 1.94 degrees Celsius. I challenge both of them to either agree or disagree with this prediction.”

    So there you have it. I specifically challenge you, William Connolley, to agree or disagree with my prediction. (If you disagree, I have $200–the minimum bet on “Long Bets”–that says I’m right and you’re wrong.

    [Response:Take away the "IPCC only" disclaimer: why are you so shy: James Annans bets are open to anyone. And clarify the satellite series, which vary by a factor of three: there are so many: by 2100 there will probably be even more. And pick a time horizon, like JA when we will still be alive: see - William]

    Of course, I can definitely understand if you’ll “weasel out,” William Connolley. After all, on your own blog, you point with pride to how high your blog comes up when one Googles the word, “Stoat.” (Here in the U.S., “stoats” are commonly known to be “weasels.”)

    [Response:Yes, stoats and weasels are very hard to distinguish. I'm very fond of both. Its a bit tricky when I use the word as an insult, because its simultaneously a term of endearment. I'm sure you can cope - William]

  75. Terry:

    One way to solve this whole debate is to simply include the skeptic’s predictions into the IPCC range of predictions. Since the skeptics seem to be predicting around 0.6 degrees over 100 years (not too far below the current low end of the IPCC range), just make that the lower end of the range. Then, we have a “consensus” which runs from 0.6 degrees to the upper limit of the IPCC range.

    The skeptic’s estimate is based on a simple extrapolation from the recent trend. This is a straightforward model that seems to be within reason. I, personally, don’t see any reason to exclude it from the set of predicitons.

  76. Mark Bahner:

    In comment #65, “Dano” writes, regarding my bet challenge on Michael Chrichton vs the IPCC:

    “Glen, that’s an old bet.”

    It’s only “old” in the sense that I made it many months ago. But it’s still BRAND NEW in the sense that no IPCC TAR primary or secondary author has stepped forward to accept my challenge.

    And it’s particularly new, in the fact that I made clear two days ago that the bet is SPECIFICALLY open to William Connolley and Gavin Schmidt.

    Dano continues, “We want Bahner’s answer to James.”

    All in good time, my pretty. All in good time. I have other work to do.

  77. Henry Molvar:

    Thanks to James Annan for this post and to all who have commented. This truly clarifies many issues that were troubling me. What I find here is:
    1) Climate scientists and skeptics (some of whom are climate scientists) agree that global surface temperatures are rising.
    2) Both also agree on the approximate rate of increase and the projected approximate rate of increase over at least the next few decades.
    3) The only influential skeptic who falls outside 1) and 2) is Michael Crichton who, I contend, doesn’t have any actual beliefs regarding GW. He states his “beliefs” in order to promote his book(s) and movie(s) through whatever controversy that he can generate.

  78. Michael Jankowski:


    “This statement is utterly ridiculous!”

    Which part – the deaths saved to hypothermia, or the idea that the colder, drier air masses warm first and the most? I think the latter is well agreed upon by both sides of the AGW argument. As for the former…

    “Those who live in these colder climate regions have adapted and really have little chance of dying of hypothermia and frostbite.”

    Not that simple according to the CDC : ***…the highest hypothermia-related death rates in the United States occur in northern states, where winter is characterized by moderate to severe cold temperatures (e.g., Alaska and Montana)…”

    And can we not adapt to warmer temps, too? See Davis, R.E., et. al., 2003, “Changing heat-related mortality in the United States, Environmental Health Perspectives.” You’ll see that “excess summer mortality rates” (the same thing used to quantify the deaths in Europe you spoke of) in 28 US cities had dropped in the 90s to 25% of what it was back in the 60s. And the 90s were the “hottest decade of the century” (or “all-time,” according to some) while the 60s were much cooler.

    “Look at the heat wave in Europe a summer or two again. That is what is happening and will likely get worse. Tens of thousands died there from the heat.”

    For starters, you’re linking that summer heat wave to GW, which very well could be a stretch and the kind of thing a previous poster was arguing against. Night and winter temps should be showing the greatest increases, not summer daytime temps. Secondly, shouldn’t Europeans be “adapted” to hot summers by now, just like the people in cold climates are (according to you) adapted to cold weather? Lastly, how about these tens of thousands? ***Exposure to cold is estimated to cause 30,000 deaths a year in the UK.***

    You also need to note exactly how that “tens of thousands” number was calculated for the European heat wave. The vast majority of those deaths were not directly attributed to the heat – they were tallied based on the difference in mortality from that summer to the previous one. The method is spelled out in this link Note that Germany and Spain, which had not yet applied that methodology, had recorded a very small amount of heat-related deaths.

  79. Brian S.:

    It’s worth pointing out that Mark Bahner offerred the CO2 bet to Annan, Annan accepted the offer, and now Bahner is playing games as to whether a bet has been made. In this context Bahner should be extremely careful about throwing the word “weasel” around, as it is very close to hitting himself in the face.

    Several more days without an affirmation from Bahner, and we can say he does not stand by his bets.

  80. Jeffrey Davis:

    The kinds of bets being proposed are a kind of conspicuous consumption: none of us will be alive to witness the outcome of bets due in 2100. It’s doubtful even if the escrow agent would exist or paperwork survive. To make a bet like this is simply to demonstrate that one has extra cash.

    Congratulations on having enough cash to light your stogies with.

  81. Mark Bahner:

    Brian S. writes, “It’s worth pointing out that Mark Bahner offerred the CO2 bet to Annan, Annan accepted the offer, and now Bahner is playing games as to whether a bet has been made.”

    You’re “pointing out” something that is completely FALSE.

    Here is what I actually wrote on my weblog (Proposed bets for James Annan, Regarding IPCC TAR

    “Methane concentrations: The methane atmospheric concentration in 1990 was approximately 1700 ppb. The IPCC TAR projects a 50 percent chance that the methane atmospheric concentration will be more than 2060 ppb in 2030. That is obvious nonsense. In fact, if the atmospheric methane concentration is more than 2060 ppb in 2030, I will give you $50. But if it’s less than 2060 ppb, you give me $1. In other words, I’m giving you 50-to-1 odds on something that, if the IPCC TAR was correct, should be even money!

    Industrial emissions of CO2: Industrial emissions of CO2 in 1990 were approximately 6.0 gigatons as carbon. The IPCC TAR projects a 50 percent chance that emissions will be more than 13.2 gigatons as carbon in 2030. If emissions are more than 13.2 gigatons as carbon, I will give you $25. If they are less, you give me $1. In other words, I’m offering you 25-to-1 odds on something that, if the IPCC TAR was correct, should be even money.

    CO2 concentrations: The CO2 atmospheric concentration in 1990 was approximately 354 ppm. The IPCC TAR projects a 50 percent chance that the CO2 atmospheric concentration will be more than 438 ppm in 2030. If the CO2 atmospheric concentration is more than 438 ppm in 2030, I will give you $25. If the CO2 atmospheric concentration is less than 438 ppm in 2030, you give me $1. Again, I’m offering you 25-to-1 odds on something that, if the IPCC TAR was correct, should be even money.”

    In other words, I offered bets to James Annan on methane atmospheric concentrations, CO2 emissions, and CO2 atmospheric concentrations. ALL those bets would be viewed as giving James Annan EXTREMELY favorable odds, if the IPCC TAR projections were reasonable (which the most certainly are NOT).

    As everyone can see from the last paragraph quoted above, I offered James Annan ***25-to-1*** odds in his favor on something that SHOULD be, if the IPCC TAR projections were legitimate (which they are not) a 50/50 bet!

    By the way, my estimation of the IPCC TAR midpoint CO2 concentration in 2030 was based on the general method followed by Wigley and Raper to get the “50 percent probability” temperature increase in 2030 (relative to 1990), of 0.80 degrees Celsius. Anyone interested can see the IPCC TAR CO2 projections for the various scenarios here:

    IPCC TAR Summary for Policy Makers, Figure 5

    James Annan accepted that offer…if the CO2 atmospheric concentration is below 438 ppm in 2030 (which there’s approximately a 96 percent chance it will be, in my opinion), then James Annan will pay me $1. If the CO2 atmospheric concentration is above 438 ppm in 2030, I will pay James Annan $25.

    James Annan did have one question as he agreed to the bet. He asked if the results would be judged based on CO2 measurements at the Mauna Loa volcano in Hawaii?

    That is fine with me: CO2 concentrations at Mauna Loa, Hawaii, by CMDL

    The concentration for 2030 will be the average concentration over the entire year, assuming this is acceptable to James Annan.

    So James Annan and I definitely have a bet. If the CO2 concentration at Mauna Loa, averaged over the year 2030, is greater than 438 ppm, I will give him $25. Once again, I will give him ***25-to-1 ODDS*** on something that, if the IPCC TAR projections were correct, should be only even money.

    P.S. To my knowledge, James Annan has never accepted the ***50-to-1*** odds I gave him on the IPCC TAR’s midpoint 2030 atmopheric methane concentration.

    And he has never accepted my 25-to-1 odds on the IPCC TAR’s midpoint 2030 industrial CO2 emissions level.

    That kind of makes me wonder if Mr. Annan is being completely honest when he says he thinks the IPCC TAR projections are good. (Especially since I already know that they are not.)

  82. dave:

    Could someone (Gavin?) relate these bets about warming by 2030 to the climate system’s inertia (warming lags) with respect to forcing as described in Earth’s Energy Imbalance: Confirmation and Implications by Hansen et. al. 2005. Here’s the quote:

    The lag in the climate response to a forcing is a sensitive function of equilibrium climate sensitivity, varying approximately as the square of the sensitivity, and it depends on the rate of heat exchange between the ocean’s surface mixed layer and the deeper ocean. The lag could be as short as a decade, if climate sensitivity is as small as 0.25C per W/m2 of forcing, but it is a century or longer if climate sensitivity is 1C perW/m2 or larger. Evidence from Earth’s history and climate models suggests that climate sensitivity is 0.75 to 0.25C per W/m2, implying that 25 to 50 years are needed for Earth’s surface temperature to reach 60% of its equilibrium response.

    We investigate Earth’s energy balance via computations with the current global climate model of the NASA Goddard Institute for Space Studies (GISS). The model and its simulated climatology have been documented, as has its response to a wide variety of climate forcing mechanisms. The climate model’s equilibrium sensitivity to doubled CO2 is 2.7C (~2/3C per W/m20.

    Now, suppose we stop all forcings now, taking equilibrium response at 380ppm C in the atmosphere plus the other GHGs. We still have 0.6C warming in the pipeline. Suppose further that 60% of that warming is manifest after 25 years (we’re at 2030) using the low estimate from the quote. That is 0.36C warming over the next 25 years and works out to 0.144C/decade. Now, this is obviously unreasonable since we’ve used the low estimate. Using the high estimate we could say that 60% of the pipeline warming is manifest after 50 years, yielding 0.36 warming over 50 years yielding 0.072C/decade over that period and 0.175C by 2030.

    Now, obviously GHG emissions are continuing to rise and the current decadel increase is 0.17 C reflecting an on-going transient response to increasing GHG levels in the past.

    I know that these calculations are naive, but how could anyone possibly think that we would not reach 0.30/C in twenty years?

    [Response: Well someone could claim that the model, the current radiative forcing and the ocean heat content changes they were validated against are all wrong, but I basically agree with your calculations - the current imbalance and the warming 'in the pipeline' imply even in the absence of further emission increases temperature trends are unlikely to be much below your estimate, (although you have to allow for some potential solar/volcanic/intrinsic variability). -gavin]

  83. Glen Raphael:

    By that argument the Kyoto accord is also a kind of conspicuous consumption: none of us will be alive to witness the outcome of that bet either.

    To make a hundred-year bet is to demonstrate confidence in the continued viability of human institutions. If you don’t live long enough to collect, you can still sell your position to somebody else at whatever seems an appropriate price. If it starts tending to look like you’ll win, the value of the bet will increase over time. If not, not.

  84. Lynn Vincentnathan:

    RE #48, you mention that the difference between 108 & 108.7 F in Delhi is not so great, but we need to understand that while temperature may increase incrementally, the damage due to GW may not follow a linear or incremental function. For instance, the last few inches added from a flood that breaches a levy and floods a huge metropolitan area are much worse than the first few inches from that flood. Likewise, the last .07 F increase in temp may cause a lot more heat deaths than the first .7 increase (say from 92 to 92.7 F). And the ice shelf break off is very sudden compared to the slowly rising temps.

    In the 70s I read about “catastrophe theory” in math. Perhaps that could be applied to GW damage scenarios. I remember seeing a potato chip function – I think meaning that you increase X just a bit more, & Y goes haywire.

  85. R. T. Pierrehumbert:

    A small correction to the post by “Doug,” who wrote:

    “If alarmists are 95% confident that the IPCC temperature range encompasses the likely possibilities,”

    Such people, of course, are “scientists,” not “alarmists.” To qualify as an alarmist you’d have
    to believe in something ridiculous, like a 50% chance of a Venus-type runaway greenhouse, or
    the supposition that adhering to Kyoto would bankrupt the US economy.

  86. Joel Kuni:

    In his Response appended to #47, William wrote:

    “As of 150 years ago, the thousand-year trend was downwards if anything: . If you believe in random natural shifts, you presumably believe they are as likely to be downwards as upward. “

    There’s no article in the link when I click it. Did you mean to give a different link?

    With regard to your second question, I am aware the trend of the past few years is upward. Why would I want to wager that the trend will not continue upward? I think the geologic record demonstrates that CHANGES in trends are infrequent and are therefore unlikely to occur in my lifetime.

    [Response:The link is correct and works (...ah. Now. It was correct when I pasted it in, but evidently WordPress is not smart enough to exlcude trainling periods from URL-ification). If you follow it, you'll find a pic showing... that changes in the trends do indeed occur - William]

  87. SteveF:

    I’d say the record for the late Pleistocene demonstrates repeated changes in overall trends. Take the various late glacial climate changes; the Bolling-Allerod and the Younger Dryas in the warming from the LGM. Or in the cooling towards the LGM there were the Dansgaard-Oeschger fluctuations.

  88. DrMaggie:

    Re #86:

    I believe the link contains a spurious period at the end, e.g. it should actually read !

    [Response: Thanks. All previous links have now been fixed. - gavin]

  89. Stephen Berg:

    A report about desertification as a result of climate change and unsustainable development:

  90. Steve Latham:

    This is mostly for Joel Kuni but I’d like to solicit a comment or two from a RealClimate climatologist and a skeptical climatologist. The most relevant posts are #s 20 (JK), 40 (me), 47 (JK), 53 (Mike Jankowski), 54 (Joel Shore), 68 (JK), and 86 (JK).

    Dear Joel, I really think you are being disingenuous here. When asked why you think that a null hypothesis regarding non-anthropogenic-GHG forced temps should include continued increases, you have responded that (a) poorly understood fast termperature increases have occurred at the ends of ice-ages, (b) compared to the last several billion years, recent temperatures aren’t very high, and (c) temperature trends don’t change frequently, so you should not expect another change in your lifetime. I guess in response I should say (a) poorly understood rapid declines in temperature have also occurred (could you find a citation for rapid warming during an interglacial?); (b) some skeptics deride paleoclimatological data in support of worrisome GHG (+’ve feedbacks) but here you’re claiming support from estimates of temperatures a billion years ago; and (c) I have seen skeptical comments in the media that it was hotter in the ’30s, there was a global cooling scare in the ’70s, and we’ve been in a cooling phase since 1998 — these don’t jive with your assessment of trends. Okay, I know that my criticisms rely on others’ words, and truthfully I doubt the helpfulness of further explanation of why you don’t expect a return to recent mean temperatures (although I wouldn’t mind seeing more on your definition of a trend). Instead I’d like to use your comments in responding to your bet proposal.

    In #68, you say that you would need to see a statistically significant positive slope for temperature over the next number of years AND a model correspondence of 50%. I think it’s silly to demand an arbitrary r^2=0.50. Also, why focus on one part of the globe when we’re talking about global warming, and why focus on Antartica when the IPCC models suggest relatively little warming there? Instead what you (and the skeptic crowd in general) need to do is generate your own model for climate change that excludes positive forcings by anthropogenic GHG. Then that model or a family of them have to be run against IPCC models over a period of time so that the two can be compared to see whether including anthropogenic-GHG information adds to predictive or explanatory power. This suggestion relates to a general complaint I made in #28 here ( As of now I would presume that your model predicts a continuing monotonic increase in temperature at the same rate as has been occurring since the current warming “trend” began (please define “trend”), with a stochastic annual variance about that trend as estimated for the last billion years or so. Alternatively, you could adopt as your model an IPCC model with anthropogenic forcings due to GHGs zeroed-out (force-fit to historic data for the recent “trend” to re-calibrate the other forcings) and carried on into the future. I don’t expect that either model is a horse you’d bet on, but that’s really the kind of thing I think the skeptics need to provide. I reckon that if you can’t predict or explain climate change at least as well as the IPCC, then you should shut up about how bad the IPCC predictions are. The proof isn’t in the money you put up for silly bets and it’s not in how loudly you object “I’m not convinced!” — it’s in the abilities of an alternative model.

    Prior to Joel Kuni’s rebuttal, perhaps a RealClimate climatologist who understands these models could comment on errors I’ve made in my conjecture regarding climate models. My being inaccurate or stupid won’t advance the debate, but perhaps there is the germ of a valuable statement here that can be confirmed. Likewise. because demonstrating Joel’s position to be ridiculous would also accomplish little (his views seem extreme relative to most skeptics’ views presented here), perhaps one of the participating skeptical climatologists could also comment and educate me on whether this argument is relevant at all to most skeptical climatology.

    [Response:JK said since global warming models uniformly predict that the greatest anthropogenic warming will occur in the winter months in the driest, coldest climates. This appears to be a common illusion. I don't know of any evidence for it. As you point out, the models don't show a great deal of warming there. In fact, the models show (broad brush here) the greatest warming where the ice-albedo feedback is most active, ie in regions of seasonal snowcover or seaice.]

    [Response:As to predictions, JKs favoured "model" now appears to be... current trends will continue. Disregarding his reasonning (which I think is wrong) this is nonetheless a simple easy-to-use model, which can either predict global or local (if you chose to use it that way) temperatures. Run globally, it produces results compatible with the IPCC range, though towards the low end, so we have (yet again) the skeptics agreeing with the IPCC results (if not the method of obtaining them). Run locally, I suspect it would be quite wrong - William]

  91. Michael Jankowski:


    If a skeptic chalks up most of the 20th century change to “natural variability” (as it seems the IPCC did with the warming of the first half the century),

    [Response:This is wrong. Read what they say: Simulations of the response to natural forcings alone (i.e., the response to variability in solar irradiance and volcanic eruptions) do not explain the warming in the second half of the 20th century. However, they indicate that natural forcings may have contributed to the observed warming in the first half of the 20th century. NF "may have contributed" is a long way away from chalking up most of it to NF - William]

    then it’s hard for one to come up with a model for future temps.

    [Response:No not really. If your postion is that nat var causes most of the changes, then you are in a position to predict that warming is as likely as cooling (or poss that cooling is *more* likely, if you regard the current warming as unusual). At one point, this appeared to be Lindzens position, but now we discover its not - William]

    And the skeptical position of many is that trying to model our climate system to any level of accuracy 10, 20, 50, or 100 years down the road is a fruitless exercise in itself. I do feel certain that if you (or someone else) were to establish a research facility/program for the purpose of creating and maintaining such models, you would have no trouble finding people to fill positions.

    You have some good points, but this is bogus: “I reckon that if you can’t predict or explain climate change at least as well as the IPCC, then you should shut up about how bad the IPCC predictions are.”

    When I had some work done on my driveway years back, the finished product had several problems that justifiably required correction (even as admitted by the contractor). Now I can’t tell someone how to pave a driveway, nor can I do it myself (beyond something not fit to look at or drive/park on). By your logic, I should’ve just shut-up about it. Was I in the wrong for complaining and having the work re-done properly?

  92. Joel Kuni:

    Re #90,

    Steve wrote:

    “you have responded that (a) poorly understood fast termperature increases have occurred at the ends of ice-ages, (b) compared to the last several billion years, recent temperatures aren’t very high,”

    With all due respect, no, I didn’t. The link I provided describes a multitude of warming events that were not caused by humans, and not all of them coincided with the end of an ice age.

    Furthermore, I don’t recall ever writing a statement similar to your item (b) (did I?) but I think it’s interesting that the above link contains this information:

    ” Following the sudden ending of the Younger Dryas, about 11,500 years ago (or 10,000 14C years ago), forests quickly regained the ground that they had lost to cold and aridity. Ice sheets again began melting, though because of their size they took about two thousand more years to disappear completely. The Earth entered several thousand years of conditions warmer and moister than today; the Saharan and Arabian deserts almost completely disappeared under a vegetation cover…”

    Is this the terrible consequence of global warming we’re supposed to fear? That the Saharan desert might disappear under a vegetation cover?

    You asked if I can find a citation for rapid warming during an interglacial. The same source cited above provides a good one:

    “A particularly widespread cool event associated with relatively wet conditions seems to have occurred in many parts of the world around 2600 years ago (van Geel et al. 1996). A general pattern in climate during the Holocene has been detected from high-resolution cores in the north Atlantic. It seems that at least in the North Atlantic region, and possibly globally, there was a warm-cold cycle with a periodicity of around 1500 years (Bond et al. 1997). In the north Atlantic region, and probably adjacent oceanic areas of Europe, the change from peak to trough of each period was about 2 deg.C , a very substantial change in mean annual temperature (though only a small fraction of the change between glacial and interglacial conditions).”

    My choice of Antarctica for a region to study was made because I am attempting to focus on the regions of the earth that should be warming fastest, if the IPCC models are correct. The models uniformly predict that the greatest anthropogenic warming should occur in the winter months in the driest, coldest climates of the globe. Wintertime in Antarctica seems to fit this description perfectly. If the models cannot explain the climate in Antarctica, then I think we have to seriously question their validity and all the conclusions drawn from them.

    I am surprised by your assertion that “the IPCC models suggest relatively little warming” in Antarctica. This is news to me. Please provide your source. I am eager to correct my misconceptions.

    [Response:Are you really? How nice. And yet, you've been told something quite directly, and still haven't bothered to look it up, nor can you be bothered to cite the sources of your misconceptions. If you want to know about the IPCC TAR projections of future climate, then a good place to look might be... the IPCC TAR: . The rest of your post is similarly littered by misconceptions, which people have corrected you on, but I rather doubt you're going to listen or learn - William

    My choice of r-squared=50% was not arbitrary. If nobody is willing to wager on a value this high, then we can probably make the following statement, which I think will come as a gigantic shock to most of the general public and the reporters who pander to them:

    Climatologists are in unanimous agreement that the link between greenhouse gasses and temperatures is so weak that less than half the changes in the earth’s climate can be attributed to mankind’s emissions of greenhouse gasses.

  93. William:

    A reminder re the comment policy: pure economics is out. We’re fairly flexible about this but a line has to be drawn somewhere. Repetetive long comments are also at risk – William

  94. Dano:

    Since the poster didn’t do so in #92, can anyone provide a link to the shocking, likely untrue phrase Climatologists are in unanimous agreement that the link between greenhouse gasses and temperatures is so weak that less than half the changes in the earth’s climate can be attributed to mankind’s emissions of greenhouse gasses.?

    Has Arrhenius been audited and found wrong?

    Thank you in advance,


  95. Michael Jankowski:

    Re#91 William,

    Yes, it does say that in the TAR (I guess I was using the ’94 SAR interpretation), but there are points throughout the TAR that I think conflict with that statement. It will take some time and length for me to find them and present them.

    As far as your comments with the model issue and 50/50 cooling vs warming idea, I addressed those in another post. But to repeat: if a skeptic believes we are still recovering from the last ice age (or at least the LIA), or at least believes we are in a naturally warm cycle that may not completely reverse itself back to current temps within the next 20 yrs, then a skeptic would not take a flat wager on cooling from 2005-2025. To repeat another point (in case it has arisen again): what you or a non-skeptic thinks about the LIA being a regional phenomenon and/or 20th century being unnatural warm is not necessarily what a skeptic believes. A skeptic would base his/her wager accordingly.

  96. SteveF:

    r.e. post #32.

    Just a quick point, but the greening of the Sahara wasn’t the result of a warming climate. Rather it was the result of the particular orbital configuration during the early-mid Holocene and a bunch of other processes (not yet fully understood) including land surface feedbacks and the effect of the adjacent ocean.

    Also, with regards to warming in the interglacial, the notion of Holocene stability is indeed something of a myth. It mainly derives from the lack of action (8.2 ka event aside) in the ice cores and as ice core results are pretty trendy and have a large say (rightly so) in directing research questions, a lot of people think of the Holocene as kind of boring. The Bond paper is cited all over the place, however on land these periodicities haven’t really been replicated and I’m not sure if they have been fully replicated in the oceans either. Climate changes (precipitation more significantly than temperature) have undoubtedly occurred in the Holocene though.

  97. SteveF:

    r.e. #95,

    I sincerely hope that no reasonably well informed skeptic has ever said that recovering from an ice age could be an explanation for what we are currently seeing. If they have, please post a link so that I can laugh at them.

  98. Michael Jankowski:

    Re#94 Dano,

    Not the same, but the following TAR statement doesn’t seem that far off to me with regard to temperatures:

    “In the light of new evidence and taking into account the remaining uncertainties, most of the observed warming over the last 50 years is likely to have been due to the increase in greenhouse gas concentrations.”

    Most people would translate that to indicate that the IPCC is saying there’s >50% chance that >50% of the warming of the last 50 years is due to increasing GHG concentrations, but it depends on what your interpretation of the words “likely” and “most” are. The link between GHG concentrations and the warming of the 1st half of the 20th century is almost certainly weaker, is it not? So it isn’t that much of a stretch to thereby say the IPCC cannot conclusively attribute over 50% of the 20th century temperature rise to anthropogenic GHG emissions – and that’s just the 20th century.

    And, of course, the original post says, “changes in the earth’s climate,” which may not necessarily only be referring to temperature.

    [Response:You are making what I think is a common mistake - to assume that anything not definitively attributed to GHG is attributed to something else. Thats wrong. The TAR, as I (and now, it seems, you) read it, attributes last-50-y to GHG's, mostly. But fo the first 50 y, attribution statements are much weaker. A mixture of various forcings could be at work and its rather hard to disentanlge them. But there is nothing in there (AFAIK) that says "we can definitively exclude GHG's as the major contributor to the warming in the first half of the 20th C" - William]

  99. Michael Jankowski:

    RE#98 William,

    I didn’t “assume that anything not definitively attributed to GHG is attributed to something else.” That is not part of #92, #94, or #98. The question was whether or not half of “climate change” could be attributed to GHG. I just pointed out that even the IPCC couldn’t definitively say that over half the warming of the latter part of the 20th century was due to anthropogenic GHGs, and that figure would certainly be lower for the 1st part of the 20th century, along with certainly the vast majority of temperature changes prior to the 20th century. I think this is along the lines of what was said in #92.

    ***But there is nothing in there (AFAIK) that says “we can definitively exclude GHG’s as the major contributor to the warming in the first half of the 20th C***

    There is also nothing that says GHG’s were the major contributor to the warming in the first half of the 20th century, and my post #98 (and the original post of #92) was concerned with only how much was attributed to GHGs. If you’re going back to what I said in #95, I’ll repeat that I have to set aside time to go thru the TAR and pick-out the conflicting information.

  100. Eli Rabett:

    WRT the back and forth between William and Michael, I think we are into defining levels of proof rather than discussing science. While this is not science, but law, the levels of proof in the law form a useful construct for analyzing the situation. In the US these are reasonable suspicion, probable cause, preponderence of the evidence, clear and convincing, and beyond reasonable doubt. There are some explanations of these terms at although most probably have a good idea. WRT the reality of significant and increasing human driven climate change the question is where on this scale are you, and where do you think the IPCC is. I’m at clear and convincing and I think that is where the IPCC is. I should point out that this is the level of proof required in a child custody cases. Given that we are responsible for the Earth, it is appropriate.

  101. Dano:

    And with the above post, Eli brings this thread full circle in the context of betting: will someone put their money where they think the reasonable proof is? Thanks for that focus, Eli.

    So, can we structure the bet around these levels of proof [saaaay...where'd Bahner go?]?

    And, a related topic to these bets and levels of proof is policy. Policy-makers ‘bet’ on whether the proof is good enough by making policy, and they use different ways of knowing that are not all “Science” – experience, judgement, hunch, etc.

    Our job is to make climate levels of proof understandable for policy-makers who do not always use science in their decision-making.


  102. John Finn:

    But there is nothing in there (AFAIK) that says “we can definitively exclude GHG’s as the major contributor to the warming in the first half of the 20th C” – William

    Apart from the fact that the increase in GHGs (from pre-industrial levels) is nowhere near enough to account for the rise in temperature between 1880 and 1940.

  103. Joseph O'Sullivan:

    A bet is an understandable way to demonstrate probabilities to the public and non-scientist. The use of U.S. legal standards is also a useful construct (Eli Rabett #100) to translate a complex topic. However the way things work in science (once my field) and in law (my current field) are very different. Science is about collecting data and interpreting data to find factual information. The legal system, especially in courtroom situations, is about persuading people to your viewpoint, sometimes even if facts contradict your viewpoint.

    The way lawyers try to change juries’ perceptions is very much like the way politicians, lobbyists and public relation types try to change the public’s perceptions. There is a public relation/political campaign machine pushing the skeptic viewpoint. I realized this in after reading RealClimate motivated me to do some research on the politics of the science.

    On the political side anthropogenic climate change (ACC) is promoted as a belief of environmentalist types. All environmentalists are cast as extremists and alarmists, and so the idea of ACC must also be extremist and alarmist, but not scientific fact.
    The politically motivated skeptics also questioned the science. First the position was there is no evidence. When evidence for ACC was produced their position became there might be some evidence but it is uncertain and is overruled by other causes. There is a newer tactic that I have seen on TCS and other sites to admit there is evidence but to denounce all of the evidence pointing to ACC as promoting unrealistic catastrophic ACC.
    A common theme of extremism/alarmism is prominent in the skeptic arguments. R. T. Pierrehumbert notices this also (#85). The conclusion of these arguments is no political action should be based on extremism or alarmism so no climate change regulations should be made.

    Michaels accepting the low IPCC projection does not surprise me. The skeptics have been careful to change their position as the science information has changed. The idea is to question the science but in a plausible way. The skeptics can’t completely reject the science when the evidence is overwhelming (they would lose credibility), but also don’t want to be seen as embracing the science (this could be used to promote regulation). This is a common rhetoric trick. You first present yourself as trustworthy and unbiased then promote your biased viewpoint. Even used-car salesmen do this by first saying they don’t care if you buy a car but then start to urge you to buy a car.

  104. Lynn Vincentnathan:

    RE #100, glad you brought in the judicial perspective. I’ve been following the “medical model.” The doctor does not say to the patient, “there is only 94% certainty that the lump is cancerous, so we’ll wait & see if it gets to 95% certainty before considering an operation.” I started reducing my GHGs in 1990, way before the first studies started coming in with 95% certainty on AGW in 1995. That’s the only reasonable, moral, logical, correct response (unless you’re a CEO of Exxon; & even Dick Cheney is into alternative energy). I started with efficiency & conservation that did not reduce my living standard (actually increased in a couple of areas), and was able to reduce 1/3 to 1/2, & now am on 100% wind power, and saving $1 per month over conventional energy.

  105. Joel Kuni:

    If one follows the link William provided in his response to #92 one finds three pictures collectively labeled “figure 9.8.” The bottom left-hand corner of the uppermost picture seems to indicate that the IPCC anticipates ground-level warming in the interior of Antarctica will indeed exceed the global average.

    [Response:Err, no: if you look at that piccy you'll clearly see that the warming in the NH north of 30N exceeds that in Antarctica. The warming shown in that pic for Antarctica is between 1-2 degrees over most of it, and I can't see how you manage to interpret that as more than the global average. Even the warming right at 90S (just over 2 oC) isn't clearly greater than the global average, and is probably less - William]

    It also confirms my prior belief that temperature increases in this region are predicted to be significantly higher than in equatorial regions, but only at ground level in the interior of the continent.

    The response appended to #90 asserts that “the models don’t show a great deal of warming there” (in the world’s driest, coldest climates). With regard to Antarctica this appears to be a true statement only if one looks at temperatures throughout the entire troposphere, apparently because the oceans surrounding the continent efficiently soak up a lot of excessive global heat. To that extent, I was clearly mistaken and I stand corrected.

    But the response in #90 is clearly false if one looks only at ground-level temperatures in the interior of the continent.

    The IPCC link also points out that excessive warming in the northern arctic regions “is connected with a reduction in the snow and sea-ice cover.” Obviously, snow and ice would melt in response to warming regardless of the reason for that warming. This melting and the resulting increase in temperatures it causes in the future does not provide any evidence that the initial warming and melting was caused by greenhouse gasses.

    Since the IPCC seems to be saying that excessive future warming in northern arctic regions will largely be a result of ice melt, future temperature changes in that region do not seem particularly useful for identifying anthropogenic influences. In my estimation, the best candidate region for that purpose remains the portion of Antarctica in which the IPCC models predict significant warming.

    Based on this information, I suggest the wager I proposed in #68 could be settled by referring strictly to ground-based temperature readings in the Antarctic interior. If the average annual temperatures in this area during any consecutive 50 year period (excluding years of volcanic and El Nino events) are plotted against atmospheric GHG concentrations and a least-squares trend line fitted to this data has a value of r-squared exceeding 50%, I will have to concede defeat.

    But I’m betting that no climatologists will accept such a wager.

  106. Joel Kuni:

    Re #103:

    You assert that “The skeptics have been careful to change their position as the science information has changed.” What do scientists do when scientific information changes? They do the same thing. It is an admirable thing to do, yet you imply the opposite. Astonishing.

    You claim there is a public relation/political campaign machine pushing the skeptic’s viewpoint. There is also heavily-financed campaign machine pushing the environmentalist/alarmist viewpoint. It is being pushed by people whose future incomes are dependent on convincing the general public that dangerous amounts of anthropogenic global warming will occur in the future.

    The groups whose incomes depend most heavily on successfully promoting this belief are employees of environmentalist institutions, climatologists who derive income from government grants, and a very large number of politicians and bureaucrats.

    If we are to dismiss the arguments of skeptics because they receive income from institutions that want to advertise a certain viewpoint then, for exactly the same reason, we must reject the views of most climatologists.

  107. Joel Kuni:

    Re #104: In this case the skeptics are like doctors who advise “There is only a 50% chance the lump in your head is malignant rather than benign, and there is a 90% chance you will be severely disabled if we attempt to remove it.”

    When I say “benign” I am referring to the fact that global warming will save a large number of people who will otherwise die of frostbite, hypothermia or malnutrition.

  108. Jeffrey Davis:

    Re: 106 and the “astonishing” comment. At what point, do self-described skeptics then discard the word “skeptic”?

    The groups whose incomes depend most heavily on successfully promoting this belief are employees of environmentalist institutions, climatologists who derive income from government grants, and a very large number of politicians and bureaucrats.

    Very large? How many is that, roughly? Hundreds? Thousands?

    I say I’m a skeptic on that, but I’d be willing to see the evidence.

  109. Michael Jankowski:

    RE#103, “Michaels accepting the low IPCC projection does not surprise me.”

    It shouldn’t surprise you, either, considering that he said over two years ago that he projected an increase from 1990-2001 of 1.6 deg C, which is 0.2 deg C higher than the IPCC’s low-end

    His arrival at that number may be overly-simplistic, but it should be clear as to why he accepts the IPCC low-end.

  110. SteveF:

    r.e. post #106

    “You claim there is a public relation/political campaign machine pushing the skeptic’s viewpoint. There is also heavily-financed campaign machine pushing the environmentalist/alarmist viewpoint. It is being pushed by people whose future incomes are dependent on convincing the general public that dangerous amounts of anthropogenic global warming will occur in the future.”

    Firstly, whilst I don’t have any figures, I am somewhat dubious that ‘environmentalists’ have a larger budget going into the global warming debate than the skeptics (i.e. industry).

    Secondly, climatologists derive their income from being climatologists. Their income is not dependent upon persuading the general public about anything. There were climatologists before global warming and there will be climatologsts regardless of what happens to the climate in the future. In addition I happen to know a few of the figures involved in the global warming debate (I’m in grad school in the palaeoclimate world). Your implication that the research they produce is coloured by the neccesity to earn money is flat out wrong. End. Of. Story.

    r.e. post #104

    “When I say “benign” I am referring to the fact that global warming will save a large number of people who will otherwise die of frostbite, hypothermia or malnutrition.”

    Hmm, potentially saved by global warming? The odd inuit. Potentially killed by global warming? The vast number of people who live in coastal regions around the world.

  111. Stephen Berg:

    Re: #107,

    “When I say “benign” I am referring to the fact that global warming will save a large number of people who will otherwise die of frostbite, hypothermia or malnutrition.”

    How about the millions who will die as a result of HYPERthermia and related problems?

    As for your prediction that fewer will die as a result of malnutrition, that is incorrect. Loss of arable land as a result of climate change (due to desertification) and suburban sprawl will significantly reduce the food supply for this planet, which is calamitous, since the population continues to explode. As a result, there will be far fewer people able to eat enough to survive, causing massive loss of life and desparation (resulting in greater violence).

    I have no idea of how you came up with this, Joel.

  112. Brian S.:

    Like the illustrious Mark Bahner, I now have a global warming bet offer at, although unlike Mr. Bahner, my bet is open to anyone who wants to take me up on it.

    My bet is that the IPCC TAR is more likely to be correct over the next 20 years than Richard Lindzen. Choosing a change of .3 degrees Celsius for the IPCC over 20 years as compared to Lindzen’s no change, the mid-point between the two positions, and my bet offer, is that temperatures will increase at least .15 degrees Celsius.

    The prediction and bet offer is here: solves the problem of guaranteeing that your betting opponent will pay up, but at a price – you don’t get the money. Instead the money goes to a charity of your choice.

  113. Michael Jankowski:

    Re#110 – “Hmm, potentially saved by global warming? The odd inuit.”

    Southeast England had an annual excess winter mortality rate of 1265 per million from 1992-2001

    According to this , the mortality for England and Wales goes up by 3,500 per winter for a 1 degree drop in temperature.

    Check out the winter mortality for England and Wales – 20,000 to nearly 50,000 people a year. And that’s during the 90s – “the warmest decade of the century/history.”

    Keep in mind these stats are for England and Wales alone. Not only is the odd iniut not included in those numbers, but neither is the rest of the world. FWIW, I believe the Scandanavian nations have an excess winter mortality rate of about half of England’s, so one could start counting from there.

    Here’s the Hadley Centre, citing an empirical model which shows that urban mortality would drop in 4 of the 5 cities mentioned under certain GHG emission scenarios (while admittedly ignoring that these cities could adapt to warmer temps, which is something that cities in the US have been shown to have done in recent decades).

    Re#111 – You’re starting to sound like Paul Ehrlich. You also seem to imply the world’s population and resources are going to go drastically beyond some sustainable limit and then suddenly snap like the stock market back in ’99. More realistically, the ties between population, agriculture, violence, etc, would likely contend with each other towards some end rather than overshoot the end and erupt into chaos.

    Can you (or someone else) tell me which areas will suffer desertification in this warmer (and supposedly wetter) world so that I can plan accordingly? I particularly would like to buy some large tracks of arable land for agricultural use. I can wall it up to protect it from suburban sprawl and the oncoming violence. The last I checked, the GCMs had strongly confliction information on which areas would be wetter than today and which ones would be drier, particularly when it came to the extremes (desertification vs widespread flooding).

  114. Joseph O'Sullivan:

    RE #106

    I did not imply that changing one’s position is bad. Frankly I admire the politically motivated skeptics for their ability to run an extremely effective public relations/political campaign, even though I do not approve of their goals or their machiavellian methods.

    What I think is interesting is how the skeptics changed. At first they said anthropogenic climate change (ACC) was nonsense with no basis in fact. As this position started to publicly lose traction, they changed to there might be evidence but it is very uncertain (uncertainty is a major theme). Now that evidence supporting ACC is getting stronger they have imported a tactic that they were already using. They know that the public is put off by extremism so they cast their opponents and their positions as extremist. They now portray the evidence that supports ACC as alarmism/extremism/catastrophism. In the legal/political world such attacks are a sign that the facts are not on someone’s side, so they use methods to move the discussion away from the facts and instead disparage their opponents to reduce public credibility.

    It is not just my claim that the skeptics are running a public relations/political campaign. It is a fact. Such campaigns have always been a part of environmental politics. What is noteworthy is how sophisticated, well-funded and negative the campaign is and how the skeptic campaign targets not just environmentalists but science and scientists. The driving forces behind this campaign are industries that could lose financially and political conservatives.

    For example Luntz is a public relations firm they are employing. See:
    The campaign that is being run and the funding behind it. See:
    The lobbyists hired by the government to advance the skeptic argument. See:
    The major media outlets that are supporting the skeptics. See:,2933,159835,00.html

    There is not anything like the skeptics campaign on the other side of the issue.
    Environmental groups are campaigning but they don’t have the money, the unquestioning support of some major media outlets and do not use the same negative tactics like the skeptics. Environmental groups are openly political and freely admit this while many of the organizations that support the skeptics present themselves as independent academic institutions but really are industry/conservative lobbying groups.
    In government the bureaucrats (I was once one of them) do not stand to personally gain from regulations. Climate change regulations just means more work for agencies that are already understaffed and under funded. More regulations could result in more science grants, but climatologists are not just out to get more grants.
    Such accusations by the skeptics are a rhetoric tactic that tries to cover up their actions by falsely blaming the other side. It’s like arguing kids saying “I know you are but what am I”.

    I am also not dismissing the skeptics’ argument just because they receive funding from institutions that want to advance a viewpoint. The questions are what is their viewpoint, why do they advance it and how they are advancing their viewpoint.
    The politically motivated skeptic viewpoint is they do not want a new regulatory scheme for financial or ideological reasons. There is nothing wrong with this position and I agree that complex regulations should be avoided if possible. The scientific consensus about anthropogenic climate change does support some kind of action and that action would really have to be a new regulatory scheme. In environmental politics in the past industry/conservatives lost on the science side and that led to their political defeats. Realizing this they decided to go after the science in the climate change debate.

    Questioning science that backs policy is a good thing. The best scientific knowledge should be used to formulate policies. How someone questions the science is the key question. Honest questions about the accuracy of science are good, but disparaging valid science just because someone does not like the political implications is not. Much of the material that Michaels and other politically motivated skeptics present to the public has dubious scientific value and is really just political attacks. That is why I question their claims.

  115. Dan Allan:


    I am not a climate scientist, but it seems to me that there are a couple of serious problems with your wager. The most glaring is that climate science predicts that there are positive feedbacks that result from anthropogenic warming. So do we attribute the increases in temperature that are due to these positive feedbacks to anthropogenic factors? Of course, logically we must. Without the anthropogenic kick-start, they would not occur. However, your r-squared method does not account for this, and you could actually win the bet precisely because these positive feedbacks occur as predicted.

    The other problem is one of time slices in your r-squared. How frequently are you testing global temperature, and how frequently is it reasonable to test to develop a trend line. The shorter the time scales the greater a role will be played by natural variability, ENSO, etc.

  116. Joel Kuni:

    Re #115 – I agree there are a multitude of feedbacks, both positive and negative. But I don’t think it’s logical to say “Without the anthropogenic kick-start, they would not occur.” Something started the current warming, but it wasn’t necessarily humans. As others have pointed out (in #102 for example), the cause of the warming that kick-started the whole cycle cannot necessarily be blamed on mankind.

    With regard to your question about time scales, my proposal is that temperatures should be graphed as a function of atmospheric greenhouse gas concentrations, not as a function of time. I think that should prevent the problem you pointed out.

    Re #110, which says “Your implication that the research they produce is coloured by the neccesity to earn money is flat out wrong. End. Of. Story.” I will be happy to concede this point, just as quickly as the rest of you concede that the same statement applies to scientists working at Exxon.

    Or maybe, based on the rather shocking anecdotal evidence in this link, I should not be so hasty to concede this point.

  117. Lynn Vincentnathan:

    RE #107 (ref to #104), the medical analogy breaks down in that surgery costs money & could be risky. Energy & resource efficiency & conservation, and some alternative energy (esp. when subsidies to fossil fuels are eliminated or a level playing field is allowed) $AVE MONEY, & strengthen the economy. And they also reduce many other harms: local pollution (ground, water, air), regional pollution (acid rain), ozone depletion, reliance on foreign energy, military costs/losses associated with such, resource depletion. Many smaller, efficient cars are not only safer for those inside the car (see DOT info), but also make roads safer for all. Bicycling & walking (to offset a few auto trips) are good for the health & spirit, reduce crime, cause less wear & tear on roads, saving taxes. Unfortunately I had to give up my bike when we moved to a place not good for cycling. People with a generous spirit can find thousands of things to do to reduce GW, and save money to boot. If some are not feasible to some people, there are many other things to do.

    Reduction in GHGs entails not only reducing energy, but all other resources & products that entail energy consumption, including water, esp. hot water. For instance, a low-flow showerhead, costing $6, can save $2000 over its 20 year lifetime in water & heating it. Following REDUCE, REUSE, & RECYCLE principles can save money and save the environment, without lowering living standards or our economy. In fact, these could even improve our economy (see Amory Lovins’s works, NATURAL CAPITALISM,, and As Hunter Lovins (Amory’s wife & partner) says, “The national energy policy comes down to the cracks around your windows,” and “the poor can’t afford to be energy inefficient and wasteful.”

    My approach is a win-win-win-win game. We need not continue to play the GW lose-lose-lose-lose game (with perhaps a few lives saved from reducing cold, against many more lost.

    RE malnutrition, the experts say GW will cause net loss in world agricultural output. For instance, while plants do better in a CO2 enriched world, weeds greatly outstrip food crops, and food crops are less nutritious, causing insects to eat more. The result is greater loss of crops. They did an experiment, and the results actually surprised them. If you add in water problems & high heat & other weather problems associated with GW, the loss will be much greater. There may be some crop increase in some places, but the world output is slated to decrease.

    There is also the possibility of very serious problems, with tremendous lost of life; even if this is only 5%, why should we risk it when abating GW (even if GW were not happening) is such a good thing in so many ways, as outlined above. If I had not reduced my GHGs with such success, I perhaps wouldn’t be so sure, but I am totally, 100% certain on this! Remember our roots: Ben Franklin said, “A penny saved is a penny earned.”

    Don’t listen to those anti-American people at TCS. They would have us get into such trouble with GW, that severe regs would have to be imposed, and we’d have to suffer under a totalitarian regime. May I suggest reading THE GALILEO SYNDROME, a futuristic novel about a world that refused to abate GW. We shouldn’t want to risk material harm by letting GW get out of hand, and we shouldn’t want to risk losing our democratic freedoms.

  118. SteveF:

    r.e. #113,

    I was being somewhat flippant with my Inuit remarks. Of course people, especially old people, die because of the cold. On the other hand, they also die because of excessive heat – take the recent warm summer (2003?) that knocked off so many old people in France. A warmer climate means more summers like that one – I recently read that such summers would be the norm come 2040.

    Compared to the potential deaths from drought, coastal flooding etc, I doubt that reducing the deaths from cold in the winter would make much of a dent on this toll. Hence my flippancy.

    As an aside, I wonder if the UK winter fuel allowance has cut into deaths during winter?

  119. Dan Allan:

    Re #116,

    Saying that natural causes could also kick-start positive feedbacks in no way refutes my first point. The bottom line is that the bet you are proposing assumes that climatologists expect a linear relationship between GHG concentration and temperature (excluding natural factors), and this is not what anyone is forecasting. So, again, temperatures and GHG concentrations could both match their forecasts pretty closely and, because of positive feedbacks, they could still lose the r-squared bet to you.

    Your second point I don’t quite follow, but maybe it is me. Aren’t you sampling correlation between GHG concentration and temperature over time? Annually? Monthly? Once per decade? The frequency of observations will of course affect the value calculated for r-squared, and my guess is that the more frequent the data-points, the lower the correlation you are likely to see because for shorter time-scales, the change in temperature is far less likely due to GHG concentrations than for longer timescales.

  120. Mark Bahner:

    William Connolley and James Annan have repeatedly asked me whether I will accept “James Annan’s bet.”

    For those who don’t know the history, I proposed (and James Annan accepted) a bet on the atmospheric CO2 concentration that is the year 2030 “midpoint” in the IPCC TAR’s analysis: 438 ppm. If the IPCC TAR is to be believed, there is about a 50/50 chance that the CO2 concentration in 2030 will be above 438 ppm. I offered James Annan $25 if the concentration in 2030 was above 438 ppm, where he would only have to pay $1 if the concentration was below 438 ppm. In other words, I offered 25-to-1 odds on what should be an even-money bet.

    After accepting the 25-to-1 odds on the $1 bet, James Annan proposed that he should also get 25-to-1 odds on a $100 bet. If the concentration was above 438 ppm, I’d give him $2500, if the concentration was below 438 ppm, he’d give me $100.

    Of course I’m not interested.

    Since the “offer” itself seems to reflect a lack of understanding of basic economics, let me explain. James Annan’s offer can be viewed in two ways, which both produce the same basic answer (that the offer is a farce):

    1) The worth of the bet to me is equal to the likelihood I’ll win times the amount I’d win, minus the likelihood I’ll lose times the amount I’ll lose. Since the likelihood I’ll win is 24/25 and the likelihood I’ll lose is 1/25, the value to me of James Annan’s proposed bet is = (24/25 * $100) – (1/25 * $2500) = -$5. (Note: The value of the bet in which I give James Annan $25 to his $1 is also negative to me. But I made the offer to prove a point, not to win money.)

    2) Another way to look at the bet is as an investment. James Annan is offering me a 96% chance that I’ll win 4% “interest” (i.e. $100) on my “principal” of $2500 dollars. But there’s a 1 in 25 chance that I’ll lose my entire principal.

    Have James Annan and William Connolley ever made an “investment” where they could earn at most 4% interest on the investment, but had a 1-in-25 chance of losing the entire principal? If not, why–other than possible complete ignorance of basic economics–would they make the “offer” to me?

    But just to show I have no hard feelings about James Annan’s insulting “offer” I have the following two counter-offers:

    1) James, I’ll be happy to bet you an extra $25 at even money. In other words, if the concentration is below 438 ppm, you owe me $26 ($1 from the original bet, plus $25 from this new offer), but if the concentration is above 438 ppm, I’ll owe you $50. I’m sure, since you think the IPCC TAR’s projections are accurate, you’ll be happy to accept this new offer. ;-)

    2) But I’ve got an even better offer for you both…a New and Improved Free Money Offer. I have boosted my offer to an incredible $8 per IPCC TAR author’s vote on my bet #181 on Long Bets (that my projections are more accurate than the IPCC TAR’s). That’s for up to a maximum of $200 (25 votes). I can’t think of any reason why y’all would refuse this offer. You can bet *I* wouldn’t, if y’all made the offer to me!

    [Response:Oh dear. Mark, the reason we won't even bother read your new and improved offer is that you keep weaselling out. You offered a bet, at certain odds, and now you say that at those odds you expect to lose money so you don't want to take it up. So it was a silly bet to start with. To those interested in this further: you are advised to follow it up on the newsgroup sci.environment - William]

  121. Terry:

    Re: #114

    “There is not anything like the skeptics campaign on the other side of the issue. Environmental groups are campaigning but they don’t have the money, the unquestioning support of some major media outlets and do not use the same negative tactics like the skeptics.”

    You’re kidding right?

    We are trying to sway the center in this debate (think of them as the jury), so credibility is very important. Obviously counter-factual assertions like this just alienate people who are potentially persuadable.

  122. Terry:

    So whatever happened to the bet proposed by JunckScience? We still haven’t heard from James.

  123. Michael Jankowski:


    As I’ve posted elsewhere, the “excess summer deaths” in all of Europe in 2003 were roughly that of the typical number of “excess winter deaths” annually in Britain alone. It’s hard to suggest the British could get much more acclimated to cold winters by now. On the other hand, studies have shown that people in US cities have adapted to warming over the past 3+ decades (as I believe I referenced somewhere else on this site). I’ve also linked elsewhere to a study showing that a change of 1 degree C in Britain’s winter equates with 3,500 deaths/lives.

    Assuming global warming does increase sea levels, we can adapt to them. We already have in many ways with floodplain construction restrictions, protective barriers, etc. I’m also not sold on the idea that our future warmer and supposedly wetter climate will produce more droughts. Regardless, much of the world is already equipped to handle droughts. Countries which cannot handle droughts have other serious problems at their forefront – disease, potable water and adequate sanitation, etc – which seem likely to take more lives than climate change.

    I do not see how the 2003 summer in Europe will be the “norm” in 2004. Temps were well above normal for prolonged periods of time. With the IPCC predicting the almost certain average global temp change by 2040 to be no more than 1-2 degrees from 1990, it’s hard to imagine that the “norm” summer would heat several times above that. That would require the other seasons in 2040 and beyond to be much cooler than 1990 in order to compensate.

  124. Joel Kuni:

    Re #119 -
    You wrote that climate scientists do not believe there is “a linear relationship between GHG concentration and temperature.” I think you are correct, but I also think members of the media and the general public have been led (by alarmists) to believe there is such a linear relationship.

    The wager I proposed is an attempt to probe the limits of what climatologists really think they know. Specifically, I am attempting to learn whether the following hypothesis is correct:

    Climatologists are in unanimous agreement that the link between greenhouse gasses and climate is so weak that less than half the yearly changes in the earth’s average temperature can be attributed to mankind’s emissions of greenhouse gasses.

    I propose to test this hypothesis with the following experiment.

    I propose to offer climatologists a wager. The wager will be that they cannot fit a trendline to data consisting of average temperatures plotted as a function of atmospheric GHG concentrations in such a way that the resulting value of r-squared exceeds 50%.

    I will make several concessions:

    The trend line can be any polynomial of degree three or less, or it can be a exponential, logorithmic, hyperbolic, trigonometric or power function (but it cannot be a mixture of such functions). This should make adequate allowance for any feedbacks that might magnify (or diminish) the effects of GHGs. I think you are contending that these feedbacks might turn a linear function into a more-steeply increasing function such as an exponential. I accept that possibility, which is why I don’t insist on a linear function.

    The climatologists may exclude data collected during any 12 month period following a major volcanic eruption or an El Nino event, even though this will make it much harder for me to win the bet because eliminating data from these years will automatically increase the value of r-squared.

    The data may be limited to average temperatures in any area of the globe covering at least 1 million contiguous square miles in a roughly circular pattern, and in any continuous size slice of the atmosphere, even though this will make it harder for me to win the bet because climatologists can test hundreds of possibilities looking for the one needle in the haystack that will prove my conjecture wrong.

    The temperature data may be collected during any period of each year, so long as the period lasts three months or longer, and so long as the same period is used in every year.

    I will accept any number of data points equalling or exceeding 50, so long as they are collected from consistenty reliable and accurate sources during consecutive years (excluding volcanic and El Nino years), and so long as they include the year 2004. Within this limit, any combination of past and future data will do.

    The GHG concentration data must be collected from the same general region as the temperature readings but they do not have to be for the same time period. They can be for any constant number of months before or after the temperature readings. This will allow for the possibility that temperature changes lag behind changes in GHG levels.

    If no climatologists will agree to offer me even-money on this wager, then my hypothesis cannot be rejected.

    What do you think? Would this experiment accurately test the hypothesis? If not, how can I improve it?

  125. Midas Oracle .ORG » Blog Archive » James Annan on Midas Oracle:

    [...] into a $10,000 global warming bet with two Russian scientists. (More general info here and here and here.) To decide who wins the bet, the scientists have agreed to compare the average global surface [...]