Climate Science Glossary

Term Lookup

Enter a term in the search box to find its definition.

Settings

Use the controls in the far right panel to increase or decrease the number of terms automatically displayed (or to completely turn that feature off).

Term Lookup

Settings


All IPCC definitions taken from Climate Change 2007: The Physical Science Basis. Working Group I Contribution to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Annex I, Glossary, pp. 941-954. Cambridge University Press.

Home Arguments Software Resources Comments The Consensus Project Translations About Donate

Twitter Facebook YouTube Pinterest

RSS Posts RSS Comments Email Subscribe


Climate's changed before
It's the sun
It's not bad
There is no consensus
It's cooling
Models are unreliable
Temp record is unreliable
Animals and plants can adapt
It hasn't warmed since 1998
Antarctica is gaining ice
View All Arguments...



Username
Password
Keep me logged in
New? Register here
Forgot your password?

Latest Posts

Archives

How reliable are climate models?

What the science says...

Select a level... Basic Intermediate

Models successfully reproduce temperatures since 1900 globally, by land, in the air and the ocean.

Climate Myth...

Models are unreliable
"[Models] are full of fudge factors that are fitted to the existing climate, so the models more or less agree with the observed data. But there is no reason to believe that the same fudge factors would give the right behaviour in a world with different chemistry, for example in a world with increased CO2 in the atmosphere."  (Freeman Dyson)

Climate models are mathematical representations of the interactions between the atmosphere, oceans, land surface, ice – and the sun. This is clearly a very complex task, so models are built to estimate trends rather than events. For example, a climate model can tell you it will be cold in winter, but it can’t tell you what the temperature will be on a specific day – that’s weather forecasting. Climate trends are weather, averaged out over time - usually 30 years. Trends are important because they eliminate - or "smooth out" - single events that may be extreme, but quite rare.

Climate models have to be tested to find out if they work. We can’t wait for 30 years to see if a model is any good or not; models are tested against the past, against what we know happened. If a model can correctly predict trends from a starting point somewhere in the past, we could expect it to predict with reasonable certainty what might happen in the future.

So all models are first tested in a process called Hindcasting. The models used to predict future global warming can accurately map past climate changes. If they get the past right, there is no reason to think their predictions would be wrong. Testing models against the existing instrumental record suggested CO2 must cause global warming, because the models could not simulate what had already happened unless the extra CO2 was added to the model. All other known forcings are adequate in explaining temperature variations prior to the rise in temperature over the last thirty years, while none of them are capable of explaining the rise in the past thirty years.  CO2 does explain that rise, and explains it completely without any need for additional, as yet unknown forcings.

Where models have been running for sufficient time, they have also been proved to make accurate predictions. For example, the eruption of Mt. Pinatubo allowed modellers to test the accuracy of models by feeding in the data about the eruption. The models successfully predicted the climatic response after the eruption. Models also correctly predicted other effects subsequently confirmed by observation, including greater warming in the Arctic and over land, greater warming at night, and stratospheric cooling.

The climate models, far from being melodramatic, may be conservative in the predictions they produce. For example, here’s a graph of sea level rise:

Observed sea level rise since 1970 from tide gauge data (red) and satellite measurements (blue) compared to model projections for 1990-2010 from the IPCC Third Assessment Report (grey band).  (Source: The Copenhagen Diagnosis, 2009)

Here, the models have understated the problem. In reality, observed sea level is tracking at the upper range of the model projections. There are other examples of models being too conservative, rather than alarmist as some portray them. All models have limits - uncertainties - for they are modelling complex systems. However, all models improve over time, and with increasing sources of real-world information such as satellites, the output of climate models can be constantly refined to increase their power and usefulness.

Climate models have already predicted many of the phenomena for which we now have empirical evidence. Climate models form a reliable guide to potential climate change.

Basic rebuttal written by GPWayne

Last updated on 26 February 2014 by LarryM. View Archives

Printable Version  |  Offline PDF Version  |  Link to this page

Further reading

Update

On 21 January 2012, 'the skeptic argument' was revised to correct for some small formatting errors.

Comments

Prev  1  2  3  4  5  6  7  8  9  10  11  Next

Comments 301 to 350 out of 505:

  1. Muon I guess you will have to purchase the article to see what I mean. I also notice you do not address the other abstracts. Here is a hint, on google scholar, google books, and in many journals, missing, flawed and uncertain physics, are discussed. Dikran I am done posting here
    You are simply mistaken
  2. "... in many journals, missing, flawed and uncertain physics, are discussed"

    That is an utterly devastating argument. Of course, one could equally say 'in many journals, complete, correct and convincing physics are discussed'; another utterly devastating argument.

    No, rebuttal of science must be made with science and not with vague generalization. This is an excellent example of the poverty of argument in denierdom: At some point, the denial always reduces to merely another version of 'No, its not. Because I said so.'

    An interesting non-technical review of the state of climate science modeling appears in the Winter 2010 Tau Beta Pi magazine:

    ... climatology is a young science. Its practitioners rarely work in laboratories. They must rely on highly variable field measurements and complex mathematical models that have very visible limitations.

    Arrayed against them are a smaller number of scientists and engineers. Only some have degrees in climate-related sciences. They charge that governments and climate activists have a pro-global warming agenda that stifles true scientific debate and that climate data and models are flawed.

    Many of these so-called skeptics have a clear agenda. They seem bent on denying climate change at any cost. Few do original research or publish in peer-reviewed climate journals (some submit articles to friendly journals in unrelated fields). Nor do they propose research to resolve the contradictions they claim to find, a common practice among the climate scientists whom they also claim lack skepticism.

    It is a recipe for controversy. And on the Internet, these scientific debates take on a life of their own. ...


    But key to the question here,

    If models raise so many questions, why does anyone trust them? The answer is that they do a surprisingly good job of predicting climate.
  3. Dikran Marsupial at 20:45 PM on 24 February, 2011, re "Large uncertainty does not imply unreliability. In fact it means that models are more likely to be reliable as the model projections cover a wider range of possibilities."

    Whilst that may satisfy the academics, the question that arises for those looking for something worthwhile to work with, is at what point is any usefulness lost?

    As an example, in Australia, BOM and CSIRO found the secret to increasing realibility of their medium to long term forecasts was by issuing them in terms such as one frequently offered "there is a 50% chance of above average rainfall".
    The classic however was a seasonal forecast of a 40% chance of above average rainfall.

    However, as is painfully obvious, but as was also observed in a recent Parliamentary inquiry, such forecasts are somewhat less than useful.

    Reliability is meaningless if it has been gained at the expense of usefulness.
  4. I would say that models have reliably predicted global climate trends(but not weather) in that observed climate variable have tracked prediction within the bounds of uncertainty. Is that useful? They are telling you it will be expensive implications if GHGs continue to be emitted at current rate. Sounds a useful prediction to me.
  5. 299. CO2 Lags PS.

    Sorry, everything went wanky when I tried to paste the Berner citation. I think it is in Breeker anyway.

    I wanted to cite a specific example. Berner, in the failed citation, makes the grand arm waving statement " A large Devonian drop in CO2 was brought about primarily by the acceleration of silicate rock weathering by the development of deeply rooteds plants in well drained upland soils."
  6. To get the discussion restarted on a more appropriate thread...

    Regarding the climate projection being "wrong", it is important to realise that the mean of an ensemble of model runs is not intended to be a projection of the actual observed temperatures over some period in the future. It is a projection of only the "forced component" of the climate, i.e. what happens in response to the change in forcings define in the particular scenario, assuming the effect of "internal variability" is negligible. This means that the projection isn't "wrong" if it doesn't exactly match the observed temperatures, because it isn't intended to. This is because the effect of "internal variability" is unlikely to be negligible, but it is of no interest as it is (i) quasi-cyclical and averages out to zero in the long term and (ii) irrelevant for determining the consequences of e.g. fossil fuel use. Where the "internal variability" is considered is in the spread of the model runs, which gives the stated "margin of error" of the projection for the purposes of comparing with observed temperatures. If the observed temperatures lie within the spread of the model runs, then the observations are consistent with what the models say is plausible.
  7. Gilles

    In a comment you made in the weather and climate thread, you said "there is some implicit selection of "good" parameters behind" Are you saying that models are bad because parameters that reflect reality are used?

    Gilles and HR,
    Correctly me if I am wrong, but both of you seem to have the opinion that

    1) climate is sensitive to parameters/physical processes in the model, and without knowing precisely what theses parameters and unknown processes are, the outputs don't reflect reality.

    2) the models that are able to replicate reality do so because the modellers found a set of parameter that works.

    would these two statements reasonably describe your views?
    Response: Sorry, I accidentally deleted your other thread's pointer to here. I asked John to restore it. [DB] Comment restored.
  8. Playing devils' advocate for a moment, if the models are as sensitive to the parameters that they can explain essentially any historical phenomena (as implied by Poptech), then why has no skeptic produced a model with a set of parameters that can explain the climate of the 20th century without CO2 radiative forcing? Has this been done?

    I suspect the reason is simple, the models are not that sensitive to the tuning of parameters, especially as the parameters are often constrained by knowledge of the physics, so they can't be set to completely arbitrary values.
  9. What gets me about this issue is the "skeptics" only use it as a way to instill doubt about climate change. They offer no alternate solution.

    No one claims that models can perfectly predict future climate but they can give us very important perspective on what we are doing to the climate. Does anyone expect that any specific area on the planet will change exactly the way models show? No. But we get strong indications of what ways the planet is going to change. Enough of an indicator to seriously look for ways to deal with the problem.

    If the "skeptics" were serious about this issue they'd offer up something better. How would they propose to test changes in the climate system? Modelers are doing what is necessary to test conditions. If "skeptics" think they're getting it wrong somehow then they need to produce their own models instead of just playing the doubt game.
  10. Perhaps time to revisit Hansens model updated with observed data up to Feb 2011, putting six more years worth of real data onto the chart in the intermediate version of this post:



    I’ll offer few comments other than the obvious but easily and often missed point that observed average global temperatures have in fact gone up since 1988, and that they have not levelled off, and they certainly aren’t going down. Limitations of this particular model have been discussed elsewhere (it slightly over-predicted CO2 levels). A massive amount of work has been done since. Hansen had many critics at the time.
  11. In reply to Poptech:

    "...my point [is that] if your initial conditions are uncertain that makes the results uncertain.

    Uncertain is a damn sight removed from worthless, don't you think?

    I said:

    "In climate models, if you are expecting them to perfectly model the exact evolution of the atmosphere, then you are essentially expecting them to be a perpetual weather model."

    You said:

    "That is exactly my point. If they cannot do this, their results are meaningless. Calling computer code a "climate model" does not change how computers work."

    So we've gone from uncertain to meaningless? Well, weather models also can't perfectly model the evolution of the atmosphere. Are their results meaningless? If you have ever taken an unbrella with you because the weatherman said it would rain, there's a good chance you did it on guidance that you yourself consider 'meaningless'.

    I said:

    "Additionally, to be 'perfect', a climate model would need both perfect initial knowledge of the entire ocean, cryosphere and biosphere, and perfect knowledge of the future evolution of all things that affect climate: solar output, GHGs, aerosols, volcanoes, etc etc etc. And what's more, the 'perfect' model is a pipe dream because models treat continuous time and space as finite blocks."

    You said:

    "Exactly, which is why computer climate models for predictions are worthless."

    Climate models are supposed to estimate the avolution of the climate for a given scenario. One thing they aren't is a deterministic forecast. They provide a projection, not a prediction. That's not just a matter of semantics, there is a distinct difference in meaning.

    Oh, and as you agreed with me in all of the above except interpretation, you have some hubris to claim that:

    "Stu you display ignorance of computer systems and computer science." - care to point out where?
  12. i didn't find "Schwartz" in a cursory search of this thread. It is two of the entries on PT's list (the 2007 paper plus the 2008 response). In the 2007 rebuttal by Foster et al, http://www.jamstec.go.jp/frsgc/research/d5/jdannan/comment_on_schwartz.pdf the climate system time lag was claimed to be irreducibly complex: "Such a multicomponent physical system cannot be expected to act with a single time scale. Even if the system evolves according to an AR(1) process, it must be a vector AR(1) process with many distinct time scales" They criticize Schwartz's short time constant by a inference to the long time constant of the deep ocean.

    But there is no set of independent variables on which to perform a vector AR1, all of the other temperature variables (deep ocean, mixed ocean, cryosphere) are functions of global average surface temperature. Schwartz is correct in his rebuttal http://www.ecd.bnl.gov/pubs/BNL-80226-2008-JA.pdf that the time constants in the models can be reduced to a single time constant which can be estimated using AR1. The critics (Foster, Annan, Schmidt and Mann) conditionally accept the AR1 premise, then work backwards from their models with high climate sensitivity to prove that the time AR1 time constant can't be as short as Schwartz claims. But that's really putting the cart before the horse.

    To specifically answer Rob (#313) the model I would propose is a low sensitivity, short time lag model in which CO2 and natural forcing creates about 1.2C sensitivity (defined as temperature change for a doubling of CO2 or 3.7W/m2)
  13. Eric,

    "To specifically answer Rob (#313) the model I would propose is a low sensitivity, short time lag model in which CO2 and natural forcing creates about 1.2C sensitivity (defined as temperature change for a doubling of CO2 or 3.7W/m2"

    The sensitivity of a model is an emergent property, not pre-programmed. That's why one function of models is to estimate sensitivity.

    You could certainly tweak a model to produce a high or low sensitivity, but the ultimate goal should be to make a model that reproduces the entire climate system as realistically as possible, and then see what sensitivity it has...
  14. Eric... But how could you possibly rationalize sensitivity that low? The only papers that make such a claim (Lindzen 2009) have been shown to be questionable, and are not supported by paleoclimate estimations.
  15. Stu, I agree with the progression from physics to model to sensitivity. Schwartz agrees too, his energy balance model is basically an AR1 equation with parameters derived from empirical data (the temperature record and external forcings). The model has some debatable characteristics, 1) it is linear, 2) external forcings that act differently on parts of the climate system (e.g. solar forcing into ocean warming) are not treated separately, they are all combined into one variable. But his critics did not use empirical data the same way, but ran part of it through their model which by its particular parameterization of weather has a resultant high sensitivity. It is not programmed to be high.

    Rob, it looks Schwartz lengthened the time constant to 8.5 years by fixing a mistake (still not sure what the mistake was in the original 2007 paper). That yielded sensitivity of close to 2C per doubling. I had read that a while ago, but forgot about it when I wrote my previous post.
  16. IanC, sorry for being late , I missed your question.


    "Gilles

    In a comment you made in the weather and climate thread, you said "there is some implicit selection of "good" parameters behind" Are you saying that models are bad because parameters that reflect reality are used?"

    all models are approximate, so I don't really know what you're calling "bad" or "good". My question would rather be ; are they reliable (good predictive power)? in other facts : is the fact that they correctly fit past data enough to believe in their predictions ? and my answer is : no.

    "
    1) climate is sensitive to parameters/physical processes in the model, and without knowing precisely what theses parameters and unknown processes are, the outputs don't reflect reality."

    same remark : they always reflect a part of reality. The only question is if it's good enough to make reliable predictions - and how we can assess that. Many people seem to think that seeing a set of models superimposed to data is enough to believe them- I don't.
  17. "a set of models superimposed to data is enough to believe them- I don't. "
    No, but if they didn't match it would be good reason to disbelieve them. There is no way to "prove" a model is reality, but continuing success of model does increase confidence. Model validation is done in rather more complex ways than just global temperature trends including testing the physics of all the components. However, could any paper or data cause you to change your mind and decide we did need to act to limit CO2 - or you would always just find debating tricks to excuse such an action?
  18. " then why has no skeptic produced a model with a set of parameters that can explain the climate of the 20th century without CO2 radiative forcing? Has this been done?"

    This is a point that needs some further emphasis. It would be a telling blow to climate science if you could fiddle with the parameterization so as to reproduce historical temperature records without a CO2 influence. Considering the rubbish that opponents do fund, wouldnt attempting this be a better bet than dubious disinformation? AR4 model will run on a modern desktop. Even a TAR model would be devastating, so not that difficult if parameterization is so tunable. Petroleum companies certainly have the resources - hey they could contract my institute to attempt it! That would be fun.

    Back in real world, this has happened because it cant. The tuning argument is from those that dont understand the process. Its FUD created to rationalize debelief in a message that they dont want to hear.
  19. I find it difficult to believe that anyone could consider the argument posed in posting 312 and endorsed in 322 as in anyway persuasive. The argument against climate modeling is essentially that no computer model of a non linear dynamic system of the complexity of the global climate can accurately predict the future. (read chaos by James Gleick) The fact that no-one has built a model that does not include co2 forcing is not relevant to the point.
    In particular the models are not capable of guaranteeing that if the carbon dioxide produced can be cut by x% then it will have y degrees implact on reducing the temperature at the end of the century. If the models can not provide these types of guarantees then they are not a valid basis for public policy initiatives involving spending trillions of dollars of ordinary taxpayers money on carbon taxes trading schemes and the like
    Response: [DikranMarsupial] Weather is chaotic, that does not mean that climate (long term average behaviour) is also chaotic. GEP Box said that "all models are wrong, but some are useful", whether models can "accurately" predict the future depends on how you define "accurate". Secondly, it is irrational to require a guarantee before taking action. I have car insurance, but I didn't take action to buy it because there was a guarantee that I will need it. We all make such probabilistic judgements every day, this is no different.
  20. The argument against fluid dynamic modeling is essentially that no computer model of a non linear dynamic system of the complexity of any man-made object moving through a fluid can accurately predict the safety and efficiency of the real thing.

    Models are not capable of guaranteeing that if the mass and drag can be cut by x% then it will have y degrees impact on reducing the operating costs. If the models can not provide these types of guarantees then they are not a valid basis for public policy initiatives involving spending trillions of dollars of ordinary taxpayers money on planes, ships, trains, automobiles and the like.

    If we destroy the wind tunnels and computers we can save the taxpayers a lot of money. It will also create jobs in the sabot sector.
  21. "The argument against climate modeling is essentially that no computer model of a non linear dynamic system of the complexity of the global climate can accurately predict the future. (read chaos by James Gleick)"

    The argument you refer to is one against the assertion that models are "tuned" through parameterization to reflect the biases of the modeller.

    Your argument is different. Weather is chaotic but that doesn't imply climate is chaotic. It remains an open question but the mathematical systems used in climate modelling are not chaotic in the formal sense.

    For more on this, see this argument

    It might also be good John included the climate model FAQ from realclimate (here and here) in the "Further reading" part of this article.
    Response: [DB] Added the RC climate model FAQ links per your suggestions. Thanks for taking the time to make them!
  22. scaddenp at 10:20 AM, I notice your using of the "weather is chaotic" meme and am wondering if enough thought is given to whether or not it's regular use as a catch-phrase is indeed still valid or justified.

    I believe the basis of the term lies not within the nature of weather itself, but with man's ability to understand the combination of factors that create seemingly complex processes.

    There is no doubt that to some of those who undertake predicting the weather, their results would appear to indicate that weather is indeed chaotic, and as such provides an excuse for the failure of their predictions, so maintaining the meme is extremely useful for them.

    However as most of us know, the reliability of weather predictions is constantly improving, the full extent at any point in time perhaps not realised by those who rely on the many free services available rather than the specialised professional services.
    It is not in the interests of such professional services to describe weather to their clients as chaotic. In fact it is the opposite they must emphasise, that being that weather is in fact quite predictable, and that the advantage that they are able to provide is that they have introduced more relevant data into their modeling to achieve that higher degree of predictability.

    So maybe the time has come for the term to be retired and a more appropriate catch-phrase developed, for those who rely on such things that is.
  23. Formal "chaos" is a descriptor of a mathematical system not a physical system. Weather is described as chaotic because the mathematical system used to model it has this character. What it tells you is that even the model perfectly captures the physical system, small errors in describing the initial state (and in weather you can only sample the initial state and all measurements have error) will eventually propagate to the point where predictability is lost.

    The better you can quantify the initial state, then the longer the forecast will accurate and I understand the improvements in grid resolution are also helping to make better predictions of the regional expression of weather systems. However, there is no escaping that the underlying mathematics used for modelling in weather are chaotic.

    Weather in a climate model is chaotic, but climate isnt. I recommend the realclimate FAQ for more on the subject.
  24. johnd - And here I was, thinking that "weather is chaotic" meant "highly dependent upon initial conditions", and that better data, better sensor coverage, more accurate measures, and faster computation led to better weather predictions by taking more data, more initial conditions into account. Silly me.

    Weather is the very definition of chaos, johnd. That's been rigorously determined mathematically. Many non-linear systems are; weather absolutely is. Any error in initial state will lead to divergent predictions down the line at some point.

    I certainly cannot speak to individual "subscription" weather services that do not publish their methods; but I suspect that if they did indeed provide a consistently better prediction than the normal weather bureaus they could make a lot of money supplying data to them - and they don't.

    So - back to the "climate models" thread? Where we're discussing systems limited by boundary conditions, not initial states?
  25. Upon consideration, any discussion of chaos and climate should really be directed to the Chaos theory and global warming: can climate be predicted thread.
  26. KR wrote : "I certainly cannot speak to individual "subscription" weather services that do not publish their methods; but I suspect that if they did indeed provide a consistently better prediction than the normal weather bureaus they could make a lot of money supplying data to them - and they don't."


    Very true. It would appear that such services rely on an aura of self-styled knowledge to do with the fact that they are so good that only those who pay enough can be allowed to share in the knowledge. That aura is helped by having others rave about how good private services such as these are, without ever having to reveal how those services line-up against reality - unlike the public services, which are constantly tested, accused and belittled.
    As you say, if those private services WERE so much better than the national forecast services, there would be no choice than to pay that money and receive that better service. The fact that this is not the case, speaks volumes...to those who are interested in the facts.
  27. KR at 12:27 PM, why do you suspect professional weather services would supply their services to the weather bureaus when the needs, and rewards, are with private enterprise?

    Anyway, aren't those bureaus always claiming high success rates using their own resources, the only exceptions being the unexpected events that "nobody" could foresee.

    Insurance, agriculture, mining and exploration, construction are all examples of industries that have to address weather dependent risk.

    Daily, decisions are being made which require accurately determining the possibilities of unfavourable weather impacting on the program being planned.

    At the smaller end of the scale, operators of agricultural enterprises are constantly having to anticipate the season ahead, and for even a small operator a wrong decision can mean differences measured in hundreds of thousands $.
    Do you think those decisions are made from reading the newspapers?

    At the other end of the scale in offshore construction where programs extend through all seasons and delays are measured in millions of $ per day, with programs that might have to run through cyclone seasons, and require weather windows to enable milestones to be met.
    Do you think the Captain of a derrick barge sticks his head out of the window on the day to determine whether it is safe to do an installation, and that they are not instead facing conditions that could result in a catastrophic loss?

    The need for reliable long range predictions begins when the contracts are being negotiated.
    What will be the expected weather delays if the program starts at a certain time of the year?
    Should the program start early or be delayed?
    What completion bonuses or delay penalties are likely to be earned or incurred?

    In the mining industry, you don't think that a mining company planning on opening a new pit might consider an reliable prediction very valuable if it meant that delaying the new pit avoided spending months pumping it empty of water, or should they have instead relied on the TV weather?
  28. Logicman
    The fluid mechanics analogy is a poor one. This is a much simpler system and the computational tasks involved here and the data collection required to produce usefull results is trivial compared to the data capture or computational difficulty in modeling global weather systems. Also the results produced by computer models can be tested in wind tunnels with clay models or prototypes to validate the results. We don't have the luxury of these things when modeling the global climate.

    I do not have a problem with using models to enhance or understanding of what is happening which is essentially what the fluid mechanics models are used for. But I am very concerned when countries set emmissions targets on the basis of climate models and politicians believe that achieving these targets will avert disasters that are predicted by the same models. This is suspension of disbelief on a massive scale.
  29. Ggf, I have a problem with your prediction that the warming will not exist or won't be bad. And I have a problem with your support of political action that prevents attempts to mitigate. And you must admit: your modeling is much less robust than the IPCC's. Seriously: upon what basis do you contradict not only the models (which try to tell us "how much when") but their foundational science (which tells us "what")? If you don't have a problem with the science, then you are forced to say, "It's going to happen." At that point, it would be damnable to say, "but we don't need to worry about when, because models of fluid, dynamic conditions are somewhat inaccurate." If you do have a problem with the science, then take it to the appropriate thread.

    Perhaps we can apply the same argument to the human climate. Politicians and governments, after all, are modelers of dynamic, fluid situations that, even worse, do not have stable structures, laws, etc. We should just give it up: no government! Oh wait . . . sigh.
  30. Ggf - there is the equivalent of wind tunnel for climate - natural forcings. Do the models predict correctly what will happen with volcano erupts, sun dims, etc. Why do you think there is so much interest in PETM?

    It absolute basic physics that if you increase net forcings, you will increase the temperature. If you apply heat to pot of water, its temperature will increase though you will have extreme difficulty predicting the evolving convection system. How fast it will warm is difficult but politicians have to make the choices based on best info available. That is a climate sensitivity of around 3 is consistent will model and empirical evidence. You seem to be arguing for doing nothing because we dont have perfect knowledge. Is that really sensible risk management?
  31. DSL
    I have not suggested that warming will not happen or that the consequences will not be bad. I am saying that i do not believe that climate models are capable of making predictions as to what will happen with a sufficient level of confidence to spend large amounts of money based on their predictions. What i am also suggesting is that unless we can be confident that what is being proposed in terms of emissions cut will work we should be cautious about spending large amounts of money on what may be the wrong response. I am not convinced that we need to rush into setting targets now at large cost to the economy when we can't be sure that the proposed interventions will work. Governments are being spooked into committing large amounts of money on the basis of unreliable models. A more measured response is required which considers a broader range of possible responses
  32. Ggf #335: "not convinced that we need to rush into setting targets now at large cost to the economy ..."

    You've not provided any evidence that anyone is rushing into so-called interventions. Nor have you provided any evidence that such interventions will cost more than doing nothing -- when there is evidence in a 2010 study that doing nothing costs more in the long run. See the thread Plan for 100% renewables for a discussion of these alternatives.

    "Governments are being spooked into committing large amounts of money on the basis of unreliable models."

    Spooked? Which governments are you referring to? Nor have you provided any evidence that models are unreliable. Your argument in #323, "no computer model of a non linear dynamic system of the complexity of the global climate can accurately predict the future," is overly general and far from convincing. So far, without evidence, what you have is just hearsay.

    "A more measured response is required which considers a broader range of possible responses "

    A more measured response than doing nothing is ... ?
  33. At the other extreme, evidence suggests that models are quite reliable at climate prediction, and that the cost of doing nothing is more expensive than taking action. Furthermore, estimates of climate sensitivity from empirical means also provide alarm bells.

    Suppose the climate models are wrong and we take action needlessly. What are consequences? Now consider the more likely possibility that models predictions are accurate (lets hope they dont underestimate) and that we do nothing?
  34. Ggf: "I have not suggested that warming will not happen or that the consequences will not be bad. I am saying that i do not believe that climate models are capable of making predictions as to what will happen with a sufficient level of confidence to spend large amounts of money based on their predictions."

    You have to establish a position on the science before you can establish a position on the modeling. If you understand the science, then you understand the basis for what (and how) gets put into the models. If you understand things this far, then you shouldn't be saying, "I'm not saying that . . ." Instead, you should be saying, "Yes, warming must occur based on the physics, and feedbacks (positive and negative) must occur because nature is all of a piece, and the warming will happen in this time period if we do business-as-usual because of the residence time of GHGs and the nature of the feedbacks." If you understand that and you're still engaged in trying to find a reason not to mitigate, then you're in questionable territory. Again, the models tell us when and where the ICBM we fired will hit. They might be wrong by a few miles or a few minutes--indeed, significantly wrong, but that doesn't mean we're not responsible for the firing, nor does it mean that the ICBM will not eventually cause destruction.

    Unless, of course, you're suggesting that the models are totally wrong, and that brings us back to the science.

    "What i am also suggesting is that unless we can be confident that what is being proposed in terms of emissions cut will work we should be cautious about spending large amounts of money on what may be the wrong response."

    Have you ever had convincing evidence that a very expensive solution would work? What would convince you in the case of GW? How long would the models have to be within or near their error bars? If the modeled trend is below the observed reality over the next five years, is that a model failure that would also cause you to say, "No mitigation!" The models have been a little off in this way with Arctic sea ice--too much too soon.
  35. I just found the NASA publication (Oct 2010)that I think may be the basis for the link at the top of this site about CO2 being the most important control knob. (This link did not work for me)

    Wow! The GISS model predicts that if all the CO2 in the atmosphere were neutralized GAT would drop to the level of Wisconsin glaciation in the first year! And would drop to levels capable of producing sea ice near the equator in a decade!

    As a paleo guy I had to put my feet up and think about this one. On one hand the result is utterly absurd and a clear indication that the model parameterizes far too many magical properties to CO2. On the other hand we actually need a mechanism to explain why climate is so bloody unstable. The forams are telling us that even during the prevuously supposed halcyon periods like the Mesozoic the were wild swings in temperature.

    The F-15 is an airplane whose "forcings" are so strong that no human has the reaction time necessary to fly it and computers are required to smoothe. Perhaps we have an F-15 climate? Water is a likely candidate for the wild thing with two phase transitions within earth temperature range and attendant latent heat as well as "feeds" back, forth, up, down and sideways. Whether CO2 is the moderating agent is debatable.

    A serious model of earth climate has to account for three things that vary in both duration and amplitude:

    On a millenial time scale the DO ocillations.

    On a million year scale the glacial-interglacial episodes within glacial periods.

    On a billion year scale the glacial periods themselves with 200 million year intervals of higher temperatures.

    No model even approaches this standard and ( -invective snipped- ).

    We know that temperature and CO2 have tracked together like mutt and jeff for 800,000 years. The computer models are hypotheses that CO2 has controlled temperature. I am going to ( -invective snipped- ) by advancing a hypothesis that at least until our own efforts may have decoupled them that over all of the Phanerozoic temperature has controlled CO2. What data have we to contradict this? After all, it is what the ice cores have been trying to tell us if we could only break the chains of our preconceptions.

    Any computer model that predicts temperature as whatever convoluted function of CO2 will always be right because they ALWAYS correlate.
    Response:

    [DB] Please stick to the science and reduce the level of invective.

    CO2 Control Knob links:

    1. Richard Alley (Video Presentation) 2009
    2. Lacis et al 2010
  36. trunkmonkey writes: "Any computer model that predicts temperature as whatever convoluted function of CO2 will always be right because they ALWAYS correlate."

    Excellent.

    By this statement we would have to conclude that the current rise in CO2 must result in a correlating rise in temperature.

    Glad to see you are coming around.
  37. Sorry about the double post. Don't know how to retract it. All I did was refresh.

    340.

    I am saying that until we mucked it up the rise in temperature produced the rise in CO2.
    Response: [DB] Refreshing after posting will result in a double (triple, quadruple, etc) posting. You are not the first, nor will you be the last.
  38. No truckmonkey, we know that isnt true. The isotope ratios for fossil fuel produced CO2 is different from that produced by carbon cycle feedbacks. If you look at the isotopes in CO2 from ice core bubbles, the increased CO2 during warming is from carbon cycle. If you look at isotope ratio in current atmosphere, you see increase is due to fossil fuel. At the moment, the carbon sinks are cleaning up about half our emissions. Over longer time, this will reverse.
  39. 342.

    All that happened after we mucked it up.

    Look, we are in uncharted territory. I know perfectly well that the millions of tons of CO2 we have dumped are producing some warming.

    Did you read 339?

    What I am saying is that hindcasting may be misleading.

    You may recall that I am haunted by a feeling that we are missing a piece to the puzzle, whether it be before our noses and we are overlooking it, or something beyond our current understanding. I fully realize that I, who have repeatedely invoked Occam, am ignoring him here.

    But nothing we have comes close to explaining the three things in 339.
  40. There is a difference between "this is unexplained" and "there are multiple explanations and current data doesnt constrain them".

    "nothing we have comes close to explaining the three things"

    What??? I really dont get this. You invoke million year and 200 million year "cycles" which are controversial to call cycles. I dont actually see anything in any your points which challenge current climate science. There are multiple possible explanations, none applicable to current warming.

    And yes, perhaps we have missed something but the energy flow to produce current warming shouldnt be hard to miss. Meanwhile we have perfectly good physical theory which accounts for we see and ignoring that while looking for something fanciful seems like extremely poor risk management to me.
  41. According to NASA the greenhouse effect of CO2 is 10 w/m^2 or 20% of the total greenhouse effect. I don't know where they got this, if it contains feedback assumptions, or if it is just the net absorbtion at 385 ppm.

    My understanding is that NASA is the custodian of the GISS model. I have effectively zero knowledge of numerical modeling so I haave to treat the model as a black box.

    First they tell me that Co2 is only 20% of the greenhouse effect, and then they tell me that when they take this 20% out of the box GAT drops six degrees in the first year.

    I have been lead to believe that the model was tested using hindcasting. I suspect that much of the source code was written before the ice cores were drilled.I assume (this is the weakest, honestly) that during this hindcasting the presumption was that CO2 had temperature on a leash.


    My suggestion is that until we discovered how much easier life can be if we burn that nasty black stuff, temperature had CO2 on a leash.

    Apologies for the excessive breadth of 339. I don't really believe in the "cycles". It's just how they are commonly referenced. The tendency on this website has been to say that because there is spectral significance for precessional cycle in the first half of the Pleistocene and eccentricity in the second half, that paleoclimatology is a done deal and it's all Milankovitch.

    Milankovitch is irrelevant in both the millenial DO and the billion year "cycles".

    I believe these "cycles" are actually more like the ENSO.
  42. A lot to learn here. For the attribution question, see
    Schmidt et al. There is two issue though. One is current state - how much of the greenhouse effect is attributable to each gas in their current concentration in the atmosphere. The other is what happens when change CO2. The other greenhouse gases do not stay in same concentration (especially water varies with temperature) so feedback must be accounted for. Temperature change affects albedo as well and to lesser extent aerosols so this is not a trivial calculation.

    Why would you suspect something about what you dont know? Firstly, the modelling is physical not statistical. Codes like calculating the absorption of GHGs are improved slowly but code write and rewrite happens all the time as computer speed allows more and more physics to be added to the model. Realclimate has a good FAQ on modelling; I suggest you read it, rather than suspecting.

    Milankovich is irrelevant to DO and there is a large literature on what the causes actually are. But relevance to modern climate is???
  43. trunkmoney wrote: "First they tell me that Co2 is only 20% of the greenhouse effect, and then they tell me that when they take this 20% out of the box GAT drops six degrees in the first year."

    Earth's Effective Temperature = ~255 K
    Earth's Actual Surface Temperature = ~288 K
    288 - 255 = 33 K greenhouse effect
    33 * 20% = 6.6 K greenhouse warming from CO2

    That said, the '20%' figure for CO2 contribution to the greenhouse effect is somewhat arbitrary and probably not how they got to the 'six degrees' figure you cite.

    The absorption spectra of the various greenhouse gases (GHGs) overlap. If we take the percentage of greenhouse warming which CO2 would cause if it were the only GHG over the total warming it comes out to about 26%. However, if we consider only the portion of greenhouse warming which CO2 causes that would not also be caused by other GHGs then it drops to about 9%.

    Thus, that 'six degrees in the first year' is more likely 33 K * 9% = 2.97 K immediate cooling plus a similar amount from immediate water vapor feedback effects and a bit more from the start of ice albedo feedback changes.

    As scaddenp noted, climate models are based on observed measurements. For instance, satellite readings of atmospheric temperature and water vapor content over time have been used to calibrate water vapor feedback. Greenhouse forcings of various gases have been calculated from their absorption spectra. Albedo differences of snow, ice, land, and ocean have all been measured. Et cetera.

    When you can then plug all these values and equations into a climate model and get results which closely follow the measured temperature trend since 1880... AND paleoclimate reconstructions... AND climate on other planets, it becomes somewhat difficult to claim that the model is not robust. There may be (indeed, certainly ARE) many details missing, but either the broad strokes are all included, everything is matching due to implausibly remote coincidence, or the modellers are committing massive scientific fraud... with (in many cases) open code and data.
  44. trunkmonkey,

    I just read your series of posts, starting with 339. They contain a fair number of gross misunderstandings about the physics of climate as well as how models are constructed and what they do.

    First, your surprise that temperature drops 6˚C in the first year if all CO2 is removed is understandable, but in the wrong direction. I'm shocked that it only drops 6˚C, but then, that speaks to the incredible heat capacity of the massive volume of water in the Earth's oceans (which keeps the temperature up, despite the loss of CO2).

    Second, you seem to have this feeling that models are somehow based on parametrization and statistics, and that the ongoing work on those models does not completely dwarf what was done in prior decades.

    My suggestion is that there is a lot of information out there on both of these subjects. Certainly, much of it is incorrect and as such leads to unnecessary confusion. I would be very, very careful about choosing your source of information. Go with things written by scientists and professors, and avoid bloggers (and engineers) of all flavors.

    But the answers to all of your questions and doubts are out there. I'd gladly answer them for you, except that a proper treatment of either subject would fill pages and pages of comments, and still come up short.

    One very well written starting point which use less math and a more narrative approach, and so is more palatable to most, is Spencer Weart's A Discovery of Global Warming. It is highly recommended to all.

    Please, please, please go find the answers to your questions, not by immediately looking for the answers, but instead by first building the foundation knowledge that will help you to appreciate the answers when you get there.

    From Jurassic Park, spoken by the "chaotician" Malcom (played by Jeff Goldblum):
    The problem with scientific power you've used is it didn't require any discipline to attain it. You read what others had done and you took the next step. You didn't earn the knowledge yourselves, so you don't take the responsibility for it. You stood on the shoulders of geniuses to accomplish something as fast as you could, and before you knew what you had, you patented it, packages it, slapped in on a plastic lunch box, and now you want to sell it.
    Take the time to build the foundation. Great leaps made to skip deep chasms lead to wrong conclusions.
  45. Physics does not handle emergent properties well. Life itself, while not violating the letter of the second law of thermodynamics, definitely violates the spirit.

    At the risk f heresy, I am confessing straight away that I believe there may be a law or two of thermodynamics undiscovered. Particularly as regards life and emergent properties.

    I can understand how you guys feel that if you have the physics nailed down, the paleo stuff barely matters. It's a brave new world, the Anthropocene anyway.

    When I hear, "The data are all bounded."; "The broad brush stokes have been made."; I caution the hubris of the cowboy who arises to find his charges have jumped the fence. Such are emergent properties.

    (I'm restraining myself here due to the excessive breadth of previous posts)
  46. trunkmonkey @349:

    First, the primary long term consequence of life to the physics of Earth is that Entropy is increased by is presence by a significant degree more than it would have been in its absence. So from the physics perspective, life very much acts according to the spirit of the 2nd Law of Thermodynamics.

    Second, the rest of your post reads like nothing other than an example of the hubris of ignorance - the belief of a person that because they themselves do not know particularly much on a subject, therefore nobody knows much more. If this is all the argument you have to offer - the vaunting of ignorance, then you have no argument, only an unwillingness to follow the evidence.
  47. 349, trunkmonkey,
    I can understand how you guys feel that if you have the physics nailed down...
    Since you obviously care about the issue, I would urge you to study until you also feel you have the physics nailed down. At that point where you have a viable understanding, you would also have a viable opinion. Until then, your entire position is based on feeling, not knowledge or reason.
    ...[feel that] the paleo stuff barely matters.
    Quite to the contrary, all of the evidence matters, and any contradictory evidence must be resolved one way, or the other. The problem is, there is no contradictory evidence. Observations, physics, models, paleoclimate of many... each of them in multiple, varied, forms... it all agrees, which is why there is talk of a consensus. But the real consensus isn't just among the scientists, but also among the data. All of the data is in agreement.

    The only contradictions that I've ever seen come from people simply making stuff up, or purposely misinterpreting or misrepresenting it to make it try to seem like there are contradictions.

    There are gaps, mind you. There are certainly things we don't know, or things we think we know but the margin for error is so wide that there is uncertainty.

    But for what we do know, everything fits.

    And we know a whole lot.
  48. So truckmonkey, you think it is rational policy to continue adding CO2 in the face of all known physics in the hope that somehow there is unknown laws of nature at work?

    You are aware that emergent phenomena occur in models? And that life and other emergent phenomena do not break the laws of thermodynamics - only some people's misinterpretation of them?

    Of course paleo matters - a theory of climate must account for the past. However, there are many puzzles there that cant be solved because of lack on constraining data not because of a problem with the theory. However, if it is difficult to distinguish between solar forcing, ocean forcing and ice melt dynamics in sorting out DO, it doesnt mean that is an issue with climate for next 100 years, because none of those causes are in effect now.
  49. 352.
    No. I know we have a problem. I just don't know how big it is, but I'm not yet convinced you do either.

    As Sphaerica has discerned, I care deeply about this. It may be the defining issue of our time.

    We paleo guys have been beating our heads against the wall forever and have a profound sense of how difficult climate is.

    I have seen many projections of the models into the future. You claim that the models are hindcast, but I have never seen a graphic to demonstrate.

    I am not surprised you find emergent phenomena in models.

    Surely DO didn't disappear just because we added 150ppm CO2...

    Where can I get my own AR4 level model?
  50. trunkmonkey wrote: "Surely DO didn't disappear just because we added 150ppm CO2..."

    We haven't added 150 ppm CO2 (yet) and 'Dansgaard-Oeschger events' haven't disappeared... read up on 'Bond events' and you'll see what I mean.

    Not that this is at all relevant given that DO/Bond events are redistributions of heat between the two hemispheres of the globe (i.e. one gets warmer while the other gets cooler) and we are instead currently seeing accumulation of heat in both hemispheres.

Prev  1  2  3  4  5  6  7  8  9  10  11  Next

Post a Comment

Political, off-topic or ad hominem comments will be deleted. Comments Policy...

You need to be logged in to post a comment. Login via the left margin or if you're new, register here.

Link to this page



The Consensus Project Website

TEXTBOOK

THE ESCALATOR

(free to republish)

THE DEBUNKING HANDBOOK

BOOK NOW AVAILABLE

The Scientific Guide to
Global Warming Skepticism

Smartphone Apps

iPhone
Android
Nokia

© Copyright 2014 John Cook
Home | Links | Translations | About Us | Contact Us