RealClimate logo


Unforced variations 3

Filed under: — group @ 21 October 2010

Here’s an open thread for various climate science related discussions, to prevent more off-topic clutter everywhere else. We have some good posts coming up, but if you want to discuss something you read in the media, saw in a press release or just wanted to ask about, this is the time.

Some interesting things we’ve seen recently include discussions on the epistimology of climate modelling, Andy Dessler’s adventures in debate land and his new paper on water vapour trends, and a review of trends in the Columbia glacier. Have at it.

Addendum: Kevin McKinney has beaten us to the mention of this, but another recent article of importance is a thorough review of the state of knowledge of drought, past and future, by Dai.  The article is open access here.

573 Responses to “Unforced variations 3”

  1. 101

    October Leading Edge report features Cuccinelli’s ‘Witch Hunt” regarding his attempt to undermine publicly funded science by attacking Michael Mann

    #96 TimTheToolMan

    You seem to consistently miss some crucial points, most likely based on fundamental misunderstandings of how science is done. But also the general non sequitur nature of your reasoning.

    What if Gavin turned out to be an alien from another planet? That would not overturn our understanding of how to build a road.

    Skill is ability to be successful. The models are skilled and the physics matches up reasonably with the skill, and the quantitative addition industrial based GHG’s fits the amounts measured in the atmosphere, and the sensitivity is skillfully showing to be relatively correct, though when slow feedbacks kick in may prove to be to low, even at 3º C, and the effects on a planetary scale are also matching up to all of the above.

    When you add that all up, the error bars get smaller. Think about the odds of so many disciplines all coming to the same general conclusion with separate analysis and data observations? The odds that they would all point in the same direction is actually pretty slim.

    I’d bet that if you did the statistical analysis form that perspective, the likely hood that humans are responsible for the warming would weigh in at higher than 95%.

    But I’m a common sense kinda guy. When looking at the preponderance of the evidence, I just don’t see a lot of flaws.

    And when people such as your self talk about some mysterious flaw I have to wonder why we are not giving PhD’s to cabbage patch dolls.

    Take a look at the scientific method and try to figure out how skill is achieved. Then you might realize the basic flaw in your argument. The skill is robust. Your argument is weak, and this global warming event is human caused.


    Fee & Dividend: Our best chanceLearn the IssueSign the Petition
    A Climate Minute: Natural CycleGreenhouse EffectClimate Science HistoryArctic Ice Melt

  2. 102
    TimTheToolMan says:

    I dont particularly want to get into an argument about that models can and cant do but as you’ve listed some examples we may as well look at a couple of them

    Gavin says : “They have successfully predicted, the cooling after Pinatubo (before it happened) (and the change in water vapour, radiation, circulation), ”

    And I would expect them to do so in some cases. Adding a forcing that doesn’t impact too severely on any “wrong bit” mightn’t cause the model to go too far wrong. And just to be clear, I’m not coyly implying any particular problem in this discussion, I’m talking generally.

    You mention a couple of other successful feats of the models and they seem to be circular arguments to me. What else verified the satellite corrections needed to be made? Who was around measuring the ocean temperatures during the ice ages?

    And then there are the hindcasting predictions you mention but aren’t the models built and tuned to most accurately model known and measured events?

    I dont want to dis the models too badly. As a tool they’re great. Its just that they’ve now BECOME the science and that is at best too risky to building climate science upon and at worst potentially just plain wrong.

    [Response: No, your last sentence is very wrong. The models, as in any science, are just attempts to synthesize and integrate what’s known about a system, and their purposes vary depending on the nature of the specific model in question. Climate science is way bigger than just modeling. Very fundamental–Jim]

  3. 103
    adelady says:

    Tree planting. Holding water in the soil?

    One thing that should be borne in mind is that young trees take up vast amounts of water. Melbourne (Australia) is suffering from lack of water into its water storages because replanting, to replace logging, in watershed areas is taking out water that would otherwise flow or percolate through soils into creeks. This of course has been exacerbated by undertaking these activities during a drought.

    If precipitation is expected to be more unreliable in the future, careful thought, much more careful thought, will have to be given to placement and rates of planting of trees in all sorts of places.

  4. 104
    Didactylos says:

    True that. Trees cause major soil shrinkage, at least in temperate climates.

    I’m more interested in tropical climates, where rainforests have a bigger effect on climate, but are surprisingly bad at improving soil. The fertile layer is incredibly thin.

    We have a lot to learn from terra preta soils.

  5. 105
    TimTheToolMan says:

    @Ray Ladbury “OK, Tim, ’splain me this: Where on Earth did you guys get the idea that if somehow the models were found to be flawed, the climate crisis would go away?”

    Why would I do that? How about you ‘splain to me what that has to do with a discussion centred on the risk associated with using model results to build our science upon?

  6. 106
    apeescape says:

    I’m curious what RC thinks of some statistical developments (hier. models) in paleoclimate. I have a hard time imagining the mean reconstructions will change much, but would it be a worthwhile investment to investigate its fidelity against Reg-EM (or whatever is popular now), instead of (say) focusing on regional reconstructions of the climate?

    Tingley et al. (2010) PIECING TOGETHER THE PAST: STATISTICAL INSIGHTS INTO PALEOCLIMATIC RECONSTRUCTIONS. Technical Report. Department of Statistics. Stanford University.
    http://statistics.stanford.edu/~ckirby/techreports/GEN/2010/2010-09.pdf

    Li et al. (2010) The Value of Multi-proxy Reconstruction of Past Climate. Journal of American Statistical Association.
    http://www.image.ucar.edu/~nychka/manuscripts/JASALiPaleo.pdf

  7. 107
    TimTheToolMan says:

    @Jim : “No, your last sentence is very wrong. The models, as in any science, are just attempts to synthesize and integrate what’s known about a system, and their purposes vary depending on the nature of the specific model in question.”

    Well I disagree and there are a number of papers that have been produced that perform “experiments” wholy within models.

    [Response: Of course! That’s one of the main things you do with models. It doesn’t contradict what I said, nor prove your point about their singular importance in climate science.–Jim]

  8. 108

    105 (TTTM),

    Its just that they’ve [the models] now BECOME the science

    False. Please provide any evidence whatsoever to support the belief that the models are the science.

    …using model results to build our science upon…

    False. Please provide any evidence the model results are the foundation of climate science.

    The models are one fraction of the science, and they are used in an appropriate role, not as the keystone. Your inappropriate emphasis on them is what Ray is quite properly questioning.

  9. 109
    Patrick 027 says:

    Re 68 aphillips – There was likely generally more CO2 and CH4 (due to less O2) earlier in Earth’s history (Not familiar with the specifics but knowing some basic principles, aside from the negative feedback of chemical weathering that would tend to increase CO2 in response to cooler conditions over long time periods, and aside from the effects of evolution and the effects of climate and topography on organic C burial, more CO2 could be sustained by increased geologic outgassing). There have been some very cold periods in the Proterozoic eon.

    Re 102 TimTheToolMan – see RC’s two “FAQ” posts on climate models. At least generally, they are not tuned to fit climate changes; they are tuned to fit an average climate for some time period. There are some basic underlying physics that constrains the possibilities even if the tuning is wrong; some parameterizations can be based on observations of processes occuring on shorter time periods (as I recall, weather).

  10. 110
    John E. Pearson says:

    TimTheToolMan said: “How about you ’splain to me what that has to do with a discussion centred on the risk associated with using model results to build our science upon?”

    There is absolutely nothing other than models upon which to base science.
    All science is models. Indeed all human understanding is models. Hawking does a brilliant job explaining this in “The Grand Design” with Leonard Mlodinow, (Kindle location 411) . This is a brief snippet. Hawking discusses this at some length. Models are in no way unique to climate science.

    “There is no picture- or theory-independent concept of reality. Instead we will adopt a view that we will call model-dependent realism: the idea that a physical theory or world picture is a model (generally of a mathematical nature) and a set of rules that connect the elements of the model to observations. This provides a framework with which to interpret modern science. “

  11. 111
    Damien says:

    A new denialist argument I heard the other day floored me: that “water vapour was dropping”, which is contrary to the little I know about climate science.

    Fortunately, Skeptical Science covered it a few days after the claim was made: http://www.skepticalscience.com/Climate-cherry-pickers-Falling-humidity.html

    What I’m more interested in is **why** the Paltridge 2009 paper shows what it does. It’s partially covered on that post:

    “So why does Paltridge 2009 show decreasing humidity? The authors of Paltridge 2009 themselves point out the well-documented problems with radiosonde humidity observations in the upper troposphere. Comparisons of Paltridge 2009 with satellite measurements (NASA’s Atmospheric Infrared Sounder – AIRS) find the Paltridge 2009 reanalysis has large biases in specific humidity in the tropical upper troposphere. Additionally, Paltridge 2009 doesn’t show any large increase in specific humidity during the 1998 El Niño. Direct measurements indicate the tropical atmosphere does indeed moisten during El Niño events and such moistening is seen in the other reanalyses.”

    Was there a post on RC that I may have missed on this topic, either the radiosonde issue or the datset in question? Have there been any follow-up literature to this paper that goes into detail on the dataset?

  12. 112
    TimTheToolMan says:

    @Jim : “It doesn’t contradict what I said, nor prove your point about their singular importance in climate science.–Jim”

    You’re right of course and my previous comment wasn’t really responding to your point. So I agree that not all (climate) science relies on model results. However conversely how many papers produced these days rely on models? Answer (that I dont think would be disputed) is a lot!

    So my main point stands that if something was ever found to be seriously wrong with the models or our understanding of the climate that has been built into the model significantly changes then a lot of science either collapses or at least becomes questionable.

    The answer to this is not simple. I’m not offering a solution simply putting the idea up for discussion.

    Unfortunately it seems most people here cant grasp the idea of risk and concepts of risk mitigation. Then there are others who simply dont believe its possible for the models to be “wrong” in the manner I’m describing.

    [Response: Tim, all systems in which manipulative experiments are impossible rely on a symbiotic interplay of models and observations. Yes, in any system if you find a serious flaw then it’s of course going to weaken the power/skill of your model and send you back to the computer. The question is, what’s the likelihood of that happening, given the sum total of current knowledge. And to conclude that the science necessarily “collapses” or “at least becomes questionable” is not warranted without the specific details of what exactly you got wrong.–Jim]

  13. 113
    TimTheToolMan says:

    Jim : ” Yes, in any system if you find a serious flaw then it’s of course going to weaken the power/skill of your model and send you back to the computer.”

    Yes, but the risk multiplies when you take a previous step as unverified but “works in the model” and build on it for the next step. If and when the model fails it would potentially “take with it” all the steps back to the original. Such is the risk of doing unverifiable science.

    [Response: Whoa–big misunderstanding right there. I didn’t say unverifiable, I said non-experimental. There’s an enormous difference. You do understand that, right?]

    Again, I need to stress that this is in the very nature of using models and there is no simple solution but it needs to be addressed in some manner because it doesn’t seem to be at the moment.

    [Response: Very hard to follow exactly what you don’t get. Your last clause is way wrong–how and why models are used has had a huge amount of attention paid to it.–Jim]

  14. 114
    TimTheToolMan says:

    Jim : “[Response: Whoa–big misunderstanding right there. I didn’t say unverifiable, I said non-experimental. There’s an enormous difference. You do understand that, right?]”

    *I* said unverifiable. Do you agree that model results are unverifiable?

    [Response: I know that–and it’s a misinterpretation of what I said. Do I agree that model results are unverifiable? Of course not! You don’t mean to tell me….do you?-Jim]

  15. 115
    TimTheToolMan says:

    Jim : “[Response: I know that–and it’s a misinterpretation of what I said. Do I agree that model results are unverifiable? Of course not! You don’t mean to tell me….do you?-Jim]”

    No, what you said had nothing to do with my reply. I didn’t misinterpret you at all. I had in fact taken it for granted that model results were unverifiable but clearly you disagree.

    Frankly I find that idea bizarre when you yourself essentially said they couldn’t be measured. I mean you cant measure the result from the idea that CO2 doubles therefore the result cant possibly be verified.

    [Response: Tim this is sort of a classic example of why there’s so much confusion among the public. I said no such thing about measuring anything–you misunderstood what I said, because you appear to have certain preconceptions about what models are, and do. Since you go back and forth between generic and specific uses of the term “model” it’s impossible to follow at what level of generality you are talking–but overall you seem to be saying that all modeling is a flawed endeavor. The idea that model results are inherently “unverifiable” is just flat wrong (and that not even accounting for the difference between verification and validation, two very different things–you are actually referring to validation I assume). Maybe some others can enlighten you better than I.–Jim]

  16. 116
    Hank Roberts says:

    TTTM, you understand that all models are wrong, and some are useful, right?
    Finding something wrong in a useful model isn’t a problem, it’s a development.

  17. 117
    TimTheToolMan says:

    I think I’m best responding to Hank intially

    “Finding something wrong in a useful model isn’t a problem, it’s a development.”

    …because my point has obviously been lost. The problem isn’t a “something wrong” that means the model needs a “corrective tweak” the problem is that something major is found that breaks it. And much of the science that relied on it along the way.

    To Jim, ” overall you seem to be saying that all modeling is a flawed endeavor.”

    Of course. All GCM’s are flawed. But that doesn’t mean they aren’t useful. There is an enormous difference between a flaw that is a not quite perfect simplification vs a flaw that is a fundamental mis-modelled process.

    Possibly I mean that models cant be validated rather than verified although I probably mean both. But if I confused you having my terms mixed, then I apologise. At the end of the day, its semantics though because you appear to understand my point.

    [Response: No, those particular terms aren’t the problem–I knew what you likely meant. The point is that unless you are evaluating a purely mathematical construct uninformed by anything in the real world, the idea that models cannot be validated is wrong–otherwise what would be the point in developing them? Yes, it’s of course implicit that you will want to validate only certain key concepts, not every detail. A key point here relates to the types of evidence that really “count” toward the overall validation–and that requires an intimate understanding of the model’s structure/function in relation to the suite of available observational data. That’s why Gavin fired off a list of things that GCM’s have predicted ~ right. What’s the chance of that happening if basic physics are not represented correctly? That, without including the longer term (paleo) data that specifically informs your earlier questioning about evidence related to the effect of CO2 changes–Jim]

  18. 118
    Edward Greisch says:

    Thanks much for the link to the alarming paper by Dr. Aiguo Dai and the discussions. It looks like BPL was too close to right on.

    Maya mentioned fighting wars. See:
    http://ampedstatus.com/the-covert-origins-of-the-af-pak-war-the-road-to-world-war-iii

    We aren’t being told the truth by the standard media. I am getting nervous. Food will be a cause of war if there are any bombs or people left by that time. Dr. Dai’s article should be immediately pointed out to the Senate. We are in a really tangled mess. Fossil fuels are at the heart of both the the wars in “Pipelineistan” and GW. Getting off of fossil fuels ends both tangled webs, except for grudges.

    If Homo “Sap” survives this century, it will be by the skin of his teeth.

  19. 119
    TimTheToolMan says:

    Jim : ” The point is that unless you are evaluating a purely mathematical construct uninformed by anything in the real world, the idea that models cannot be validated is wrong”

    Well I guess thats the point of this discussion. There is no point in continuing if you are going to disagree that there is any possibility that the models are significantly flawed.

    Jim writes : “A key point here relates to the types of evidence that really “count” toward the overall validation–and that requires an intimate understanding of the model’s structure/function in relation to the suite of available observational data.”

    Surely it also relates to the fact that the model is correct in the first place. It doesn’t matter two hoots if everyone agrees that process X is modelled correctly if process X doesn’t apply in nature.

    In my opinion there is a very real possibility that the models are seriously flawed simply because the climate processes are so complex and many of them are not yet well understood.

    [Response: Hopeless. You don’t get it and don’t want to. On to other conversations we thankfully go. My mistake everyone, I thought this guy was serious.–Jim]

  20. 120
    HenryP says:

    I have a question for the panel here. I am an ordinary chemist.

    here is a paper that confirms to me that CO2 is (also) cooling the atmosphere by re-radiating sunshine:
    http://www.iop.org/EJ/article/0004-637X/644/1/551/64090.web.pdf?request-id=76e1a830-4451-4c80-aa58-4728c1d646ec

    they measured this radiation as it bounced back to earth from the moon. So the direction of this radiation was: sun-earth-moon-earth. Follow the green line (for CO2) in fig. 6, bottom. Note that it already starts at 1.2 um, then one peak at 1.4 um, then various peaks at 1.6 um and 3 big peaks at 2 um. You see these peaks all back in fig 6 top.
    This paper here shows that there is absorption of CO2 at between 0.21 and 0.19 um (close to 202 nm):
    http://www.nat.vu.nl/en/sec/atom/Publications/pdf/DUV-CO2.pdf
    There are other papers that I can look for again that will show that there are also absorptions of CO2 at between 0.18 and 0.135 um and between 0.125 and 0.12 um.
    We already know from the normal IR spectra (chemistry!)that CO2 has big absorption between 4 and 5 um.

    So, to sum it up, we know that CO2 has absorption in the 14-15 um range causing some warming (by re-radiating earthshine) but as shown and proved above it also has a number of absorptions in the 0-5 um range causing cooling (by re-radiating sunshine). Clearly, this cooling happens at all levels where the sunshine hits on the carbon dioxide same as the earthshine. The way from the bottom to the top is the same as from top to the bottom.
    So, my question is: how much cooling and how much warming is caused by the CO2? How was the experiment done to determine this?

    All things being equal as to weather and water, I am trying to establish the net effect of the radiative warming and radiative cooling of CO2. Bear in mind that the warming carries on 24/7 and cooling for 12 hours per day.

    I hope therefore to see some results in W/m2/m3 air with —ppm CO2/24 hours, both for the warming and the cooling.

  21. 121
    Dappled Water says:

    Damien @ 111 – check the Skeptical Science link again, Kooiti Masuda @ 23 sums up the situation.

  22. 122
    meteor says:

    Gavin

    I read “atmospheric CO2: principal knob governing earth’s temperature”

    In your experience you are zeroing all the non condensable gases and all the aerosols.
    Why all the aerosols?
    The aerosols have a GH effect OK but they have also an effect on the clouds (is this the reason of the sharp rising of the cloud cover?)
    but what is the effect of this rising on the cloud feedback effect and finally on the temperature?

    Can you read the article on my blog and the replies?

    thanks

  23. 123
    aphillips says:

    Patrick027,

    Thanks for that. Lindzen seemed quite sure that the early faint sun paradox is a genuine paradox, and that greenhouse gases don’t fully explain it.

    This is pure unresearched speculation, but I wonder whether sheer geothermal forces account for any of the early warmth.

  24. 124

    @Fred 56,

    Thanks for the response, I know that the paper is crap… I have written more about it in Swedish. However, the “skeptics” here in Sweden had a call out in newspapers about new articles that proved IPCC wrong… this is the only one of those that have not been refuted in a scientific journal (lindzen was an other).

    I understand that it seems like a waste of time for the top researchers to refute thees kind of silly papers, but it is the easiest way to stop the “sceptics” to reach media (at least in Sweden and other parts of Europe)… and also to later show the journalists, look here now that was what i told you, are you going to trust them again?

  25. 125
    MalcolmT says:

    Re Septic Matthew #16, Mojave Desert: Nice pics, indeed, but the map at the end isn’t very useful. The desert has an area of 25 000 sq miles (65 000 sq km). What percentage of that does the proposed power plant occupy? How is this an environmental disaster greater than the extra CO2 we would pump into the air if we didn’t build the extra solar capacity?

  26. 126
    TimTheToolMan says:

    Jim : ” You don’t get it and don’t want to.”

    What is it I dont get? Is it that I dont undersatand the models represent as complete an understanding about climate process as we have and that the physics is essentially solid for the components?

    Do you perhaps think I believe they dont work?

    Because I haven’t said either of those things. And they’re not even relevent to the discussion except as relates to the risk and consequences associated with a flaw being discovered.

    What I see is you saying is that for various reasons you believe the models ARE correct and so to entertain the idea that they might be seriously flawed is irrelevent and unnecessary.

  27. 127
    Jason B. says:

    TTT: Why can’t you specifically state what you think is flawed about specific models?

    You keep saying the same thingover and over, that models could be flawed therefore they cannot be useful or informative at all.

  28. 128
    Didactylos says:

    I think possibly the fact that TimTheToolMan doesn’t grasp is that ALL models are “wrong” (as Hank says above). As a matter of necessity. They wouldn’t be useful if they were full scale replicas of the real world, because we wouldn’t be able to run them or do anything useful with them.

    All models make some simplifications, they have boundaries beyond which they don’t operate. Models (model realisations, at least) must be finite.

    And scientists know about these limitations. It’s fundamental to how models are used and the results interpreted. Knowledge of these limitations is absolutely key to getting useful, solid results from models in spite of their limitations.

    So it’s no wonder scientists have little patience with people who say “but what if the models are wrong?”

  29. 129
    Jesse says:

    Saw Gavin quoted briefly in the Scientific American article on climate change. (Interestingly this issue had more science writers on the feature pages than usual, as opposed to working scientists who usually fill it). Anyhow, I’d be curious for any expansion on your thoughts of Judith Curry’s work with the “Skeptic” blogs.

    Seems to me that while in principle what she says is right, her problem with the IPCC stems from not getting that the IPCC reports aren’t really science documents, in that the whole thing had to be massaged to make many governments happy as it was. The IPCC reports are pretty good, but again, speaking as an outsider, non-scientist, it seems there were a few too many cooks in that kitchen from the get-go (which is, to my mind one reason there were as many errors as there were. I am an editor and when 30 people edit the same thing –even when they have expertise in the subject matter, or perhaps, especially when they have such– the result isn’t always pretty). But maybe I am completely wrong, so if so, tell me.

  30. 130
    Dan H. says:

    Very good Didactylos,
    I agree with for once. Models, by definition, are a simplication of the whole. The models are as good as our understanding of what is being modelled, and future models will change with greater understanding.
    Not everyone’s climate models are the same. Hence, we get a bell type curve of climate sensitivity with a hump between 2 and 4 (roughly). Several models fall on the tail ends of this curve. Obviously, not all these models can be correct, and the sensitivity could change as conditions change. Individually models arrive at a concise value, but the collective is often displayed as a curve. Of course the curve can be influenced by the choice of models to include, but that is all part of the science.

  31. 131
    TimTheToolMan says:

    “TTT: Why can’t you specifically state what you think is flawed about specific models?”

    Because I’m not trying to point to a particular flaw.

    The discussion I’ve been wanting to have surrounds the implication to climate science if a significant flaw is found and the risk to science that the increasing reliance on models has created. It seems that discussion cant move beyond the start gate on this.

    Perhaps its my fault for not driving it more but I was really interested in what other people thought without me influencing too much.

  32. 132
    ghost says:

    Well, TTT, suppose the GCMs are “flawed” (or blind to a yet undiscovered factor) as to dramatically underestimate the amount of AGW, and that the observed result under non-mitigation will be to eliminate most life on earth in an unexpectedly short time. What do you propose? You see, most “the models are flawed” proponents that we encounter assume that the models overstate the AGW threat. If you know models as well as you appear to state, then you know that underestimation probably is as likely as overestimation, however likely those are. Following that line appears to lead to this: (i) the GCM consensus tells us to ditch fossil fuels in an accelerated fashion; and (ii) an AGW underestimation tells us to ditch fossil fuels in an accelerated fashion. Augment those with the reality that fossil fuel supplies are finite, dwindling, and increasingly expensive. That would seem to lead to one bright conclusion: we should begin replacing fossil fuels with renewables in haste, and use part of the probable carbon budget to fuel that build-out. We’re going to have to do it eventually anyway, so why succumb to the carbonites’ delay-for-profit propaganda now, and why wait until fossil fuels are prohibitively expensive? As much as the kooks like comfort, things ain’t as rosy under the current carbon regime as they would like to believe. For those who think the GCMs’ output will be nullified by the coming of the Rapture, or the arrival of aliens bearing “To Serve Man” books and bestowing fusion technology upon us, I suggest stopping their “Cheap Domestic Beer and Mystery Science Theater 3000” diet.

  33. 133

    126 (TimTheToolMan),

    …the idea that they might be seriously flawed is irrelevent and unnecessary.

    I’m with you. They just don’t get it.

    I mean, what if some day someone discovers that the human race doesn’t even exist? Why do they discount that possibility as if it’s irrelevant? If the human race doesn’t exist, there’s no way we could be causing climate change. It rips the A right out of AGW.

    And yet people ignore this obvious fact. The fanatical alarmist believers want to take moderate but substantive steps to reduce CO2 emissions and forestall what appears will be a civilization altering event in the next fifty to one hundred years.

    Paleoclimate studies, atmospheric physics, direct temperature observations, myraid indirect physical and biological observations resulting from temperature changes, and dozens of different, independent, complex climate models all agree on and point towards that same outcome.

    But if humans don’t exist, none of that matters!

    If I don’t exist, why should I be asked to change my behavior? Why should I in any way ameliorate the situation by impinging on an almost magical, fairy tale, unnecessarily pampered lifestyle that I enjoy, albeit almost certainly at the horrific expense of my children and grandchildren?

    People just don’t get it. It’s so obvious, it’s as plain as the nose on their non-existent faces.

  34. 134
    Didactylos says:

    Dan H.: Models are just one source of estimates for climate sensitivity. If you look at this figure of climate sensitivity, you will see that the best estimates from multiple disparate lines of exploration all cluster around the 3 degrees values.

    What is the probability that all these completely different approaches have unrelated errors that all act in the same direction? Nope. It’s not likely. As research progresses, I expect to see the uncertainty narrow.

    We must act on the basis that the best estimate is likely, and that lower or higher values are possible. We must not act on the hope that the likely value is the lowest and everything else is wrong. That would be foolish, and terrible risk management.

  35. 135
    Didactylos says:

    TimTheToolMan said: “The discussion I’ve been wanting to have surrounds the implication to climate science if a significant flaw is found”

    So why have you studiously ignored my comments (and those from others) that address this very question? Could it be that you don’t like that answer, and want to cling to a fantasy of huge but unsuspected errors?

    That’s a nice conspiracy theory, but it just doesn’t fly in scientific circles.

    Please go back and read my post at #40, Chris Colose at #24 and Ray Ladbury at #29.

  36. 136

    Todd Albert 53,

    Let me know if you want to see a draft of my book, “The Case for Global Warming.”

  37. 137

    aphillips 68: What about the weak early solar paradox, or whatever it’s called? Lindzen mentioned this. Apparently, the sun was something like 25% weaker, but the earth was not frozen, and therefore climate sensitivity is weak. What’s the story on that?

    BPL: The “Faint Young Sun Paradox,” first explicitly stated by Carl Sagan in 1972. Yes, the sun was about 30% dimmer 4.5 billion years ago, so early Earth should have been frozen over. But geological data (e.g. 4.4 GY old zircons) indicate open water on the surface.

    The paradox is resolved if greenhouse gases were much higher in the earlier atmosphere.

  38. 138

    Dan H 70: I have seen claims of up to 10C/doubling of CO2

    BPL: The only one was Fritz Moller, 1963 (9.6 K/doubling), which was quickly shown to be due to a conceptual error.

  39. 139
    Hank Roberts says:

    TTTM, significant flaws don’t particularly surprise or bother anyone.

    Models aren’t expected to be all that good to be useful — they’re gross approximations.

    Our atmosphere isn’t like a “slab model” — like a stack of pancakes, each one a little fluffier than the one below it. But the models work.

    Say we find some mysterious unknown physics is causing the warming we see.

    That would be a flaw. Well, but the models have been working, so we then look for a second mysterious unknown that has been cancelling out the warming we know comes from CO2. Or a handful of others.

    You find one flaw in a model, the model keeps working, so the odds are you will find another flaw that more or less counterbalanced the first flaw.

    You find a ten cents difference in your bank balance and a thousand dollar error in a check deposited, you have to keep looking.

    You might find one countervailing error to explain the $9,999.90 difference. More likely you’ll find smaller errors in various directions before you get your checkbook model of your bank balance to be an exact match.

    But your checkbook model was useful even with the errors.

    You’ve got “overthrow the founder and the whole thing collapses” as your model of reality.

    That model isn’t useful — for science.

    It works for religion. It works for politics. It works in the schoolyard.

    It doesn’t work in science. Science works with flawed approximate models because that’s what people can make. We don’t duplicate reality.

    All models are flawed. Some models are useful.
    Look that up. Read about the guy who said it.

  40. 140

    131 (TimTheToolMan),

    …the implication to climate science if a significant flaw is found and the risk to science that the increasing reliance on models has created.

    Ditching my previous sarcasm, the problems that everyone has with this are:

    “…the implication to climate science…” are virtually none. Models are one small part of the science, and are themselves built upon myriad other aspects of science (physics, chemistry, biology, etc.). No one is going to find a flaw that suddenly, dramatically overturns the model results. It’s a denial fantasy, and ridiculous to even consider discussing.

    “…if a significant flaw is found…” which is so unlikely as to be insignificant. That fact that you have no specifics, and are just thinking of a completely intangible hypothetical is the first clue. This is a huge if, which you offer with no ideas or error bars whatsoever. There’s no reason to consider it.

    “…increasing reliance on models…” is a silly denial meme with no foundation in fact. You’ll pick it up from WUWT and other denial-echo-chamber “sources” where it’s taken as incontrovertible truth only because everyone there repeats it over and over. Do a google scholar search on climate papers and look at how many there are, and how few of those are based on models. The number and complexity of interwoven disciplines and diverse areas of study in climate science is mind boggling, and to claim something like a “reliance on models” is absurd.

    If you stop trying to convince everyone that you know something they’re too silly to realize, and instead invest time in actually studying the science and the huge volume of information available on the subject, you’ll be a lot better off.

  41. 141

    Tin Man 117: the problem is that something major is found that breaks it. And much of the science that relied on it along the way.

    BPL: Except that nothing of the sort has been found. And if it were found, you’d have to explain how the models have “wrongly” gotten so many predictions right. All the big, glaring flaws just happened to cancel out? What are the chances?

  42. 142
    Radge Havers says:

    Tim @ 131

    “if a significant flaw is found and the risk to science that the increasing reliance on models has created”

    Your faulty assumption is that somehow those silly scientists are just blindly modeling their poor little hearts out without thinking about the direction their epistemological proclivities might be taking them. If you don’t get from the responses so far that this isn’t the case, perhaps you could spell out what you would consider to be convincing evidence that they actually know what they’re doing. Then you could be provided with an answer that you could either verify for yourself (given sufficient energy on your part) or you could prove once and for all to all and sundry what fools they’ve been for not bowing to the wisdom of the wing-nut-o-sphere. No goal post moving allowed.

    At least it would serve as a point of conversation. Otherwise it looks like you’re just aimlessly messing about, fishing for something that doesn’t exist, but hoping you’ll get lucky.

  43. 143

    henry P 120: I am trying to establish the net effect of the radiative warming and radiative cooling of CO2

    BPL: Pick up a copy of John T. Houghton’s “The Physics of Atmospheres,” read through it, and work the problems. Then do the same with Grant W. Petty’s “A First Course in Atmospheric Radiation.” That should give you a pretty solid foundation in this.

  44. 144

    Tin Man 126: What I see is you saying is that for various reasons you believe the models ARE correct and so to entertain the idea that they might be seriously flawed is irrelevent and unnecessary.

    BPL: What part of “they’ve repeatedly made predictions that panned out” did you not understand?

  45. 145

    Tin Man 131: The discussion I’ve been wanting to have surrounds the implication to climate science if a significant flaw is found and the risk to science that the increasing reliance on models has created. It seems that discussion cant move beyond the start gate on this.

    BPL: Why do you think that is? Sheer meanness on the part of the posters here? Or do you think there might be something flawed about “The discussion [you’ve] been wanting to have” itself?

    Example: You go to a convention of Egyptologists and tell them, “I’d really like to have a serious discussion on how it would impact what we know about ancient Egypt, if it turned out that aliens built the pyramids.”

  46. 146

    TTT: “Because I’m not trying to point to a particular flaw.

    The discussion I’ve been wanting to have surrounds the implication to climate science if a significant flaw is found and the risk to science that the increasing reliance on models has created. It seems that discussion cant move beyond the start gate on this.”

    How can we discuss “a significant flaw” without any qualification as to what that flaw might be, what effects it would have or what portions of “the science” it would affect? Beyond, that is, saying “well, scientists will look at the affected portions, and figure out the implications”–which is more or less what someone already said upthread.

    The devil is always in the details.

  47. 147
    Rattus Norvegicus says:

    Maya,

    A quick search of the met office site turned up this:

    http://www.metoffice.gov.uk/climatechange/guide/quick/doubts.html

  48. 148
    Ray Ladbury says:

    Tim@131, The problem with your question is that it is entirely too broad. The implications of “a problem” depend entirely on exactly WHAT the flaw is. There is far more likelihood that a mistake would raise sensitivity estimates that there is that it would lower them.

    The thing is that we know that the planet is warming well outside of anything we’ve seen in the record. We know ice is melting. We know it is placing strain on ecosystems. Without models, we are flying blind, and the only responsible thing to do would be to put the breaks on HARD.

    What you seem to fail to comprehend is that climate models have been astoundingly successful in terms of both explanatory and predictive power. Because these models are based on physics rather than a mere statistical fit, it would be very hard to achieve such agreement erroneously.

    Now, again, why do you think a failure of the models would be of benefit to the denialist side?

  49. 149

    #112 TimTheToolMan

    Tim, as to your point “Then there are others who simply dont believe its possible for the models to be “wrong””

    Let me be clear for you. I think all models are wrong.

    http://www.ossfoundation.us/projects/environment/global-warming/myths/models-can-be-wrong

    Are you happy now?

    As to your statement: “Unfortunately it seems most people here cant grasp the idea of risk and concepts of risk mitigation.”

    Not only is this a totally arrogant statement, I’m at least somewhat confident that it is wrong. I think many here are examining risk mitigation in their considerations pertaining to climate change. . . or were you referring to something else other than risk mitigation?

    Another fatal flaw in your reasoning follows:

    Let’s say you are right and all the models, maths, physics are completely wrong, or even mostly wrong. That leaves a problem. It would mean that the earth is warming and we don’t have a clue why. . .

    . . .unless of course you have an alternative theory? Maybe a large alien ship cloaked above our atmosphere with a large infrared gun heating our planet?

    Please do enlighten us as to how the temperature trend has started warming with no apparent explanation, because all our models are wrong?

    Please hep me out on this one. I’m desperate for your qualifiable answer that explains this?


    Fee & Dividend: Our best chanceLearn the IssueSign the Petition
    A Climate Minute: Natural CycleGreenhouse EffectClimate Science HistoryArctic Ice Melt

  50. 150
    Dan H. says:

    Barton,
    Someone has to pick up the baton here. If a serious flaw was discovered, obviously the model would have to scratch many of the results. It is possible that the models which currently make up the tail ends (either higher or lower) could then become the norm. Alternatively, even these could be flawed in that the contained the error, but in a lesser contribution.

    Just because the model would be flawed, does not mean that we would have to throw out all we know. In the pyramid case, they were still used as tombs. The political implication would probably be much greater than the scientific, and science would simply discard them and move on. Politically, it could be a death sentence, due to the heavy reliance on the models.

    We will see if people respond with enthusiasm or meanness.

    [Response: This is fantasy land. The probability that something was so wrong that everything would be tossed is infinitesimally small, and combined with your fervent desire to have all understanding about climate change risks be related to models (it isn’t), you have a recipe for unhinged speculation that is completely pointless. – gavin]