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 Support

Twitter Facebook YouTube Pinterest MeWe

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
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.

Mainstream climate models have also accurately projected global surface temperature changes.  Climate contrarians have not.

Various global temperature projections by mainstream climate scientists and models, and by climate contrarians, compared to observations by NASA GISS. Created by Dana Nuccitelli.

A 2019 study led by Zeke Hausfather evaluated 17 global surface temperature projections from climate models in studies published between 1970 and 2007.  The authors found "14 out of the 17 model projections indistinguishable from what actually occurred."

There's one chart often used to argue to the contrary, but it's got some serious problems, and ignores most of the data.

Christy Chart

Basic rebuttal written by GPWayne


Update July 2015:

Here is a related lecture-video from Denial101x - Making Sense of Climate Science Denial

Additional video from the MOOC

Dana Nuccitelli: Principles that models are built on.

Last updated on 9 September 2019 by pattimer. View Archives

Printable Version  |  Offline PDF Version  |  Link to this page

Argument Feedback

Please use this form to let us know about suggested updates to this rebuttal.

Further reading

Carbon Brief on Models

In January 2018, CarbonBrief published a series about climate models which includes the following articles:

Q&A: How do climate models work?
This indepth article explains in detail how scientists use computers to understand our changing climate.

Timeline: The history of climate modelling
Scroll through 50 key moments in the development of climate models over the last almost 100 years.

In-depth: Scientists discuss how to improve climate models
Carbon Brief asked a range of climate scientists what they think the main priorities are for improving climate models over the coming decade.

Guest post: Why clouds hold the key to better climate models
The never-ending and continuous changing nature of clouds has given rise to beautiful poetry, hours of cloud-spotting fun and decades of challenges to climate modellers as Prof Ellie Highwood explains in this article.

Explainer: What climate models tell us about future rainfall
Much of the public discussion around climate change has focused on how much the Earth will warm over the coming century. But climate change is not limited just to temperature; how precipitation – both rain and snow – changes will also have an impact on the global population.

Update

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

Myth Deconstruction

Related resource: Myth Deconstruction as animated GIF

MD Model

Please check the related blog post for background information about this graphics resource.

Comments

Prev  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  Next

Comments 301 to 325 out of 1311:

  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.

Prev  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  21  22  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

THE ESCALATOR

(free to republish)


© Copyright 2022 John Cook
Home | Translations | About Us | Privacy | Contact Us