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

While there are uncertainties with climate models, they successfully reproduce the past and have made predictions that have been subsequently confirmed by observations.

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)

There are two major questions in climate modeling - can they accurately reproduce the past (hindcasting) and can they successfully predict the future? To answer the first question, here is a summary of the IPCC model results of surface temperature from the 1800s - both with and without man-made forcings. All the models are unable to predict recent warming without taking rising CO2 levels into account. Nobody has created a general circulation model that can explain climate's behavior over the past century without CO2 warming.


Figure 1: Comparison of climate results with observations. (a) represents simulations done with only natural forcings: solar variation and volcanic activity. (b) represents simulations done with anthropogenic forcings: greenhouse gases and sulphate aerosols. (c) was done with both natural and anthropogenic forcings (IPCC).

Predicting/projecting the future

A common argument heard is "scientists can't even predict the weather next week - how can they predict the climate years from now". This betrays a misunderstanding of the difference between weather, which is chaotic and unpredictable, and climate which is weather averaged out over time. While you can't predict with certainty whether a coin will land heads or tails, you can predict the statistical results of a large number of coin tosses. In weather terms, you can't predict the exact route a storm will take but the average temperature and precipitation over the whole region is the same regardless of the route.

There are various difficulties in predicting future climate. The behaviour of the sun is difficult to predict. Short-term disturbances like El Niño or volcanic eruptions are difficult to model. Nevertheless, the major forcings that drive climate are well understood.

A paper led by James Risbey (2014) in Nature Climate Change takes a clever approach to evaluating how accurate climate model temperature predictions have been while getting around the noise caused by natural cycles. The authors used a large set of simulations from 18 different climate models (from CMIP5). They looked at each 15-year period since the 1950s, and compared how accurately each model simulation had represented El Niño and La Niña conditions during those 15 years, using the trends in what's known as the Niño3.4 index.

Each individual climate model run has a random representation of these natural ocean cycles, so for every 15-year period, some of those simulations will have accurately represented the actual El Niño conditions just by chance. The study authors compared the simulations that were correctly synchronized with the ocean cycles (blue data in the left frame below) and the most out-of-sync (grey data in the right frame) to the observed global surface temperature changes (red) for each 15-year period.

The red dots on the thin red line correspond to the 15-year observed trends for each 15-year period.  The blue dots show the 15-year average trends from only those CMIP5 runs in each 15-year period where the model Niño3.4 trend is close to the observed Niño3.4 trend. The grey dots show the average 15-year trends for only the models with the worst correspondence to the observed Niño3.4 trend.  The size of the dots are proportional to the number of models selected.  The envelopes represent 2.5–97.5 percentile loess-smoothed fits to the models and data. Figure 2: Red: 15-year observed trends for each period. Blue: 15-year average trends from CMIP5 runs where the model Niño3.4 trend is close to observations. Grey: average 15-year trends for only the models with the worst correspondence to the Niño3.4 trend. The sizes of the dots are proportional to the number of models selected. From Nature Climate Change

The authors conclude,

When the phase of natural variability is taken into account, the model 15-year warming trends in CMIP5 projections well estimate the observed trends for all 15-year periods over the past half-century.

It's also clear from the grey figure that models that are out-of-sync with the observed changes in these ocean cycles simulate dramatically higher warming trends over the past 30 years. In other words, the model simulations that happened not to accurately represent these ocean cycles were the ones that over-predicted global surface warming.

Climate models have also been accurately projecting global surface temperature changes for over 40 years.  Climate contrarians have not:

Figure 3: 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."

Uncertainties in future projections

A common misconception is that climate models are biased towards exaggerating the effects from CO2. It bears mentioning that uncertainty can go either way. In fact, in a climate system with net positive feedback, uncertainty is skewed more towards a stronger climate response (Roe 2007). For this reason, many of the IPCC predictions have subsequently been shown to underestimate the climate response. Satellite and tide-gauge measurements show that sea level rise is accelerating faster than IPCC predictions. The average rate of rise for 1993-2008 as measured from satellite is 3.4 millimetres per year while the IPCC Third Assessment Report (TAR) projected a best estimate of 1.9 millimetres per year for the same period. Observations are tracking along the upper range of IPCC sea level projections.


Figure 4: 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)

Similarly, summertime melting of Arctic sea-ice has accelerated far beyond the expectations of climate models. The area of sea-ice melt during 2007-2009 was about 40% greater than the average prediction from IPCC AR4 climate models. The thickness of Arctic sea ice has also been on a steady decline over the last several decades.

Figure 5: Comparison of observed September minimum Arctic sea ice extent through 2008 (red line) with IPCC AR4 model projections.  The solid black line shows the mean of the 13 models, and dashed black lines show the range of the model results.  The 2009 minimum was calculated at 5.10 million km2, the third lowest year on record and still well below the IPCC worst case scenario.  (Source: Copenhagen Diagnosis, 2009)

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

Do we know enough to act?

Skeptics argue that we should wait till climate models are completely certain before we act on reducing CO2 emissions. If we waited for 100% certainty, we would never act. Models are in a constant state of development to include more processes, rely on fewer approximations and increase their resolution as computer power develops. The complex and non-linear nature of climate means there will always be a process of refinement and improvement. The main point is we now know enough to act. Models have evolved to the point where they successfully predict long-term trends and are now developing the ability to predict more chaotic, short-term changes. Multiple lines of evidence, both modeled and empirical, tell us global temperatures will change 3°C with a doubling of CO2 (Knutti & Hegerl 2008).

Models don't need to be exact in every respect to give us an accurate overall trend and its major effects - and we have that now. If you knew there were a 90% chance you'd be in a car crash, you wouldn't get in the car (or at the very least, you'd wear a seatbelt). The IPCC concludes, with a greater than 90% probability, that humans are causing global warming. To wait for 100% certainty before acting is recklessly irresponsible.

Intermediate rebuttal written by LarryM


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 29 December 2018 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

1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  Next

Comments 1 to 25 out of 1329:

  1. Look at plate 1 in Hansen's 88 paper, the model includes the oceans. Hansen's Scenario C is the one that most closely matches the "Land – Ocean" temperature. John Cook wrote: "A way to test the accuracy of models is through hindcasting - see whether they successfully predict what has been observed over the past century." Not true for any model. All that shows is they can fit the model to the history. That is beside the point as the IPCC does not claim that the models can predict anything. John Cook wrote: "The key point is that all the models fail to predict recent warming without taking rising CO2 levels into account." Given enough "tunable parameters" that should come as no surprise. The modelers also assume that there is some positive feedback, there is no proof that this is the case. Here is one for you straight from the IPCC, Chapter 8, page 596: "The number of degrees of freedom in the tuneable parameters is less than the number of degrees of freedom in the observational constraints used in model evaluation." IOW, the models are nothing more then sophisticated curve fits. Calling the models "predictions" does not instill confidence that you have done your homework. Kevin E. Trenberth http://blogs.nature.com/climatefeedback/2007/06/predictions_of_climate.html "In fact there are no predictions by IPCC at all. And there never have been. The IPCC instead proffers “what if” projections of future climate that correspond to certain emissions scenarios." And from the same letter: "Even if there were, the projections are based on model results that provide differences of the future climate relative to that today. None of the models used by IPCC are initialized to the observed state and none of the climate states in the models correspond even remotely to the current observed climate. In particular, the state of the oceans, sea ice, and soil moisture has no relationship to the observed state at any recent time in any of the IPCC models." John Cook wrote: "Satellite measurements show that the troposphere is warming" The models predict that the troposphere should warm faster then the surface, it isn't.
    Response: Re tropospheric warming, I recommend reading Satellite show little to no warming in the troposphere. The argument over "prediction" vs "projection" is semantics. Kevin Trenberth is merely saying we don't know with certainty what future emissions will be so we make predictions based on various emission scenarios. However, lest it be a stumbling block, I'll update the text. Thanks for the feedback!
  2. Models are the biggest gun in the arsenal for AGW. What people like Dyson are telling us is that the models use assumptions that are not validated by observation and that cannot account for many known effects. The models might be right but they haven't got a good track record except in hind sight. (After they've been fudged to fit the past) Someday they will probably be good they are better than 20 years ago.
  3. I recommend this paper and it references for this section as well http://members.iinet.net.au/~glrmc/2007%2005-03%20AusIMM%20corrected.pdf
  4. Leaving aside the silly notion that you can 'prove' a model's accuracy by checking it's fitting to the historical record--I mean honestly, you are aware that these models are tweaked *until* they fit the historical record, aren't you? The past is not the problem. The Hansen forecast sounded impressive, so I looked over the paper and did some googling. There is definitely a different spin on the accuracy of the forecast. Discussed here: http://www.climateaudit.org/?p=796 which demonstrates that scenario B is nowhere near the perfect fit implied by your article or Hansen. Hansen could be right, but he doesn't seem to explain where he is getting his data from. I can only find vague references to 'Station Data' and 'Land-Ocean'. What data is it he is using? How has it been adjusted? At least the sceptical article above is up front on where the data is coming from. This doesn't prove that Hansen is wrong. But it doesn't leave one with a high degree of confidence either.
  5. Well, here is NASA telling us there is no meaningful comparison of models to observed global temp change "The analysis by Hansen et al. (2005), as well as other recent studies (see, e.g., the reviews by Ramaswamy et al. 2001; Kopp et al. 2005b; Lean et al. 2005; Loeb and Manalo-Smith 2005; Lohmann and Feichter 2005; Pilewskie et al. 2005; Bates et al. 2006; Penner et al. 2006), indicates that the current uncertainties in the TSI and aerosol forcings are so large that they preclude meaningful climate model evaluation by comparison with observed global temperature change. These uncertainties must be reduced significantly for uncertainty in climate sensitivity to be adequately con- strained (Schwartz 2004). Helping to address this chal- lenging objective is the main purpose of the National Aeronautics and Space Administration (NASA) Glory mission, a remote sensing Earth-orbiting observatory" http://ams.allenpress.com/archive/1520-0477/88/5/pdf/i1520-0477-88-5-677.pdf
  6. Here is an interesting quote from IPPC's AR4 found in chapter 1: "The strong emphasis placed on the realism of the simulated base state provided a rationale for introducing ‘flux adjustments’ or ‘flux corrections’ (Manabe and Stouffer, 1988; Sausen et al., 1988) in early simulations. These were essentially empirical corrections that could not be justified on physical principles, and that consisted of arbitrary additions of surface fluxes of heat and salinity in order to prevent the drift of the simulated climate away from a realistic state. The National Center for Atmospheric Research model may have been the first to realise non-flux-corrected coupled simulations systematically, and it was able to achieve simulations of climate change into the 21st century, in spite of a persistent drift that still affected many of its early simulations. Both the FAR and the SAR pointed out the apparent need for flux adjustments as a problematic feature of climate modelling (Cubasch et al., 1990; Gates et al., 1996). By the time of the TAR, however, the situation had evolved, and about half the coupled GCMs assessed in the TAR did not employ flux adjustments. That report noted that ‘some non-flux adjusted models are now able to maintain stable climatologies of comparable quality to flux-adjusted models’ (McAvaney et al., 2001). Since that time, evolution away from flux correction (or flux adjustment) has continued at some modelling centres, although a number of state-of-the-art models continue to rely on it." A 'flux adjustment' is where you discover that the model's predictions start to vary so much from the historical record that you have to go in and change the values inside the software to re-fit the model to what's actually happening. Very confidence inspiring. And what does 'a number of' mean? 50%? 20%? 80%? How many of these models are manually fiddled with to get them to continue to work...?
  7. Here is another posting assessing Hansen's model work in a not very favourable way: Whether these alternate assessments of Hansen's work stand up is a separate issue. I would point out we should not accept them blindly any more than we should blindly accept Hansen's paper on how brilliant Hansen's previous work was, as this naive article does...
  8. "The models might be right but they haven't got a good track record except in hind sight. (After they've been fudged to fit the past)" "Leaving aside the silly notion that you can 'prove' a model's accuracy by checking it's fitting to the historical record--I mean honestly, you are aware that these models are tweaked *until* they fit the historical record, aren't you?" Nonsense. Are you saying that Hansen, way back in 1988, was able to travel in a time machine to 2006 and back, so that he could make the adjustments to his 1988 models to make them agree all the way to the present? The denialists have nothing but nonsense.
  9. Oh, and ClimateAudit is a barrel of laughs: http://scienceblogs.com/deltoid/2008/01/climate_audit_comedy_of_errors.php
  10. And besides, if models can be "fudged" to fit anything -- as our `skeptics' claim -- why are the _same_ `skeptics' saying that they can't get Hansen's model to fit the data? Can it be because our `skeptics' are simply full of junk?
  11. No, we are saying that Hanson's model from 1988 does not fit the present, even his conservative projections are significantly high of actual observation at this point. (High relative to the ground based measurements and wildly high compared to satellite and balloon measurements to be more specific) If a model can't take past conditions and produce results that fit current reality it would be obviously useless. However since modelers are not simpletons that isn't the problem that was being discussed! The problem is just because current models have been changed so they can somewhat be used to fit past observations that doesn't mean those changes were the correct changes, therefore it doesn't mean that they are making correct predictions. The models still contain assumptions for various parameters that have not or perhaps can not presently be varified. Freeman Dyson is correct here, Models are improving but they have a long way to go before they are better than educated guesses. You should read Dyson's entire statement this is a bit out of context.
  12. Wondering Aloud: "because current models have been changed" You're clearly off spouting rubbish you don't know a thing about. Look at the temperature predictions in Hansen et al. (2006) and Hansen et al. (1998). They are _exactly_ _the_ _same_. The 1998 model has _not_ been changed at all, and it still agrees all the way to 2006. All your talk about "fudge factors" can't explain that.
  13. And you say I'm the one who clearly doesn't know what he's talking about!
  14. Models are as reliable as the data put into them.
  15. I thought this comment was interesting and relevant. It is taken from the US Senate Committee on the Environment and Public Works - http://epw.senate.gov/public/index.cfm?FuseAction=Minority.SenateReport#report Physicist Dr. Freeman Dyson, Professor Emeritus of Physics at the Institute for Advanced Study, in Princeton, is a fellow of the American Physical Society, a member of the US National Academy of Sciences, and a fellow of the Royal Society of London. Dyson called himself a "heretic" on global warming. "Concerning the climate models, I know enough of the details to be sure that they are unreliable. They 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 behavior in a world with different chemistry, for example in a world with increased CO2 in the atmosphere.," Dyson said in an April 10, 2007 interview. Dyson is also a fellow of the American Physical Society, a member of the US National Academy of Sciences, and a fellow of the Royal Society of London.
  16. I'll also raise the question whether anyone really believes this extract (from above) which appears to be a basic premise for the page: "This betrays a misunderstanding of the difference between weather, which is chaotic and unpredictable and climate which is weather averaged out over time. While you can't predict with certainty whether a coin will land heads or tails, you can predict the statistical results of a large number of coin tosses. Or expressing that in weather terms, you can't predict the exact route a storm will take but the average temperature and precipitation will result the same for the region over a period of time." It's a false analogy. Random *independent" events provide statistical certainty over a period of time. The climate does not fit this description. Can anyone provide some evidence -peer reviewed citations - that long range climate forecasting is more accurate than weather forecasting? I know the IPCC claimed it in their report but they didn't backup the claim.
    Response: That's the problem with using analogies - the comparison always breaks down at some point when you compare it directly. The main point is the contrast between short term, random unpredictability and long term, statistical summations. While weather is chaotic and non-linear, long term climate trends are discernable and predictable. As is seen in these peer reviewed studies analysing the success of climate forecasts.
  17. A common comment regarding GCMs is that they do not account for clouds very well. This is a substantial weakness. There have been assertions that warming increases atmospheric water vapor which, through a feedback mechanism, increases warming. Certainly increased atmospheric water vapor would produce more, lower-level clouds. How do the GCMs account for this? A simple cloudy-planet point model where standard atmosphere tables are used to get average cloud temperature vs altitude shows that a change of average cloud altitude of 305 meters would result in an eventual average earth temperature change of 0.75C. Many other factors known to influence cloud formation are not accounted for in the GCMs.
  18. Frankbi said: "And besides, if models can be "fudged" to fit anything -- as our `skeptics' claim -- why are the _same_ `skeptics' saying that they can't get Hansen's model to fit the data? Can it be because our `skeptics' are simply full of junk?" I am surprised that John did not reply to this, but I assume he does not have the time to respond to every nonsensical claim that appears on his board. Frank, if you do not know that GCMs (and many other models) are "tweaked" to fit past data, then you have no place attacking others. It is common practice, and there is little that is nefarious about it, though it may appear as such. You probably know that there are many uncertainties and complexities in the climate system, and in attempting to model such a system, you must used what has already been observed to better your understanding and accuracy. If models were based purely upon theory for such a complex system, they would appear wildly inaccurate and worthy of no utilization. -Robert
  19. I'm repeating here what I've said in another place on your blog: The IPCC summary of computer simulations you link above only go back to 1850 and blurs out problems with individual models by replacing the spaghetti curve with a grayed out region. (Errors in the simulations are highly correlated from year to year, the figure makes it seem they are not, which is false and misleading.) Also did you notice the huge 0.3°C anomaly around 1940-1950 that the models, even with the fuzzing provided by IPCC, are unable to explain? Where did that warming come from? I would conclude from that, that we aren't at the place yet, even for a 150-year period with a lot of fudge factors thrown in, where we can accurately describe past climate, let alone accurately predict future climate. Secondly did you notice that there was very little anthropogenic forcing before 1970, according to the models? Have you ever considered how disingenuous it is, given this fact, to compare glaciers from e.g. 100 years ago to current, when the models say that almost all warming prior to 1970 was natural?
  20. Robert S: Yes, I do know that model parameters are usually adjusted according to some past data, _and_ the resulting model has to be validated with data that are _not_ used to configure the models in the first place. If I didn't make this clear enough, my apologies. From my understanding, this approach of tweaking and holdout validation is what climate scientists have been doing. And it's perfectly good science, of course.
  21. Poptech When I was in college we were taguht Fortan IV, even though it had already been supplanted by Fortan 77. I did not realize that anyone was still using it. My own last experience was in SAS and that was in the 90s. Are you saying that these climate models are being coded in Fortran?
  22. Wow Poptech, what a rousing, impassioned, statesman-like speech. Unfortunately, it contains no verifiable concrete facts.
  23. Poptech, nice job of trying to help people understand what computer modeling is and what it can do. Folks, if a climate model doesn't predict past data 100% perfectly then it's useless. You can create an infinite number of different mathematical models that will predict any data series 100% perfectly. To deserve any respect these climate models must predict the previous data perfectly as a start, none should even be thought about unless it does that, and then it has to predict the future better than a simple polynomial fit that also perfectly predicts past data. Frankbi, all the facts in Poptech's post are verifiable. I learned them in school. His analysis is spot on.
  24. "all the facts in Poptech's post are verifiable. I learned them in school." I don't think that's what "verifiable" means. As always, the "criticisms" of climate models are devoid of any concrete, testable facts. -- bi, International Journal of Inactivism, http://frankbi.wordpress.com/
  25. Here is a new study that evaluates the accuracy of climate models: D.Kutsoyiannis,N.Mamassis,A.Christofides,A.Efstratiadis,􀈱S.M.􀈱Papalexiou Department of Water Resources and Environmental Engineering National Technical University of Athens (www.itia.ntua.gr) http://www.itia.ntua.gr/getfile/850/3/documents/2008EGU_ClimatePredictionPrSm.pdf Or to sum up the study: "Climatic models generally fail to reproduce the long term changes on temperature and precipitation."

1  2  3  4  5  6  7  8  9  10  11  12  13  14  15  16  17  18  19  20  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