Climate Science Glossary

Term Lookup

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


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


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

Bluesky Facebook LinkedIn Mastodon MeWe

Twitter YouTube 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...

New? Register here
Forgot your password?

Latest Posts


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)

At a glance

So, what are computer models? Computer modelling is the simulation and study of complex physical systems using mathematics and computer science. Models can be used to explore the effects of changes to any or all of the system components. Such techniques have a wide range of applications. For example, engineering makes a lot of use of computer models, from aircraft design to dam construction and everything in between. Many aspects of our modern lives depend, one way and another, on computer modelling. If you don't trust computer models but like flying, you might want to think about that.

Computer models can be as simple or as complicated as required. It depends on what part of a system you're looking at and its complexity. A simple model might consist of a few equations on a spreadsheet. Complex models, on the other hand, can run to millions of lines of code. Designing them involves intensive collaboration between multiple specialist scientists, mathematicians and top-end coders working as a team.

Modelling of the planet's climate system dates back to the late 1960s. Climate modelling involves incorporating all the equations that describe the interactions between all the components of our climate system. Climate modelling is especially maths-heavy, requiring phenomenal computer power to run vast numbers of equations at the same time.

Climate models are designed to estimate trends rather than events. For example, a fairly simple climate model can readily tell you it will be colder in winter. However, it can’t tell you what the temperature will be on a specific day – that’s weather forecasting. Weather forecast-models rarely extend to even a fortnight ahead. Big difference. Climate trends deal with things such as temperature or sea-level changes, over multiple decades. Trends are important because they eliminate or 'smooth out' single events that may be extreme but uncommon. In other words, trends tell you which way the system's heading.

All climate models must be tested to find out if they work before they are deployed. That can be done by using the past. We know what happened back then either because we made observations or since evidence is preserved in the geological record. If a model can correctly simulate trends from a starting point somewhere in the past through to the present day, it has passed that test. We can therefore expect it to simulate what might happen in the future. And that's exactly what has happened. From early on, climate models predicted future global warming. Multiple lines of hard physical evidence now confirm the prediction was correct.

Finally, all models, weather or climate, have uncertainties associated with them. This doesn't mean scientists don't know anything - far from it. If you work in science, uncertainty is an everyday word and is to be expected. Sources of uncertainty can be identified, isolated and worked upon. As a consequence, a model's performance improves. In this way, science is a self-correcting process over time. This is quite different from climate science denial, whose practitioners speak confidently and with certainty about something they do not work on day in and day out. They don't need to fully understand the topic, since spreading confusion and doubt is their task.

Climate models are not perfect. Nothing is. But they are phenomenally useful.

Please use this form to provide feedback about this new "At a glance" section. Read a more technical version below or dig deeper via the tabs above!

Further details

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 shown 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. Sea level rise is a good example (fig. 1).

Fig. 1: 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. A 2019 study led by Zeke Hausfather (Hausfather et al. 2019) 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."

Talking of empirical evidence, you may be surprised to know that huge fossil fuels corporation Exxon's own scientists knew all about climate change, all along. A recent study of their own modelling (Supran et al. 2023 - open access) found it to be just as skillful as that developed within academia (fig. 2). We had a blog-post about this important study around the time of its publication. However, the way the corporate world's PR machine subsequently handled this information left a great deal to be desired, to put it mildly. The paper's damning final paragraph is worthy of part-quotation:

"Here, it has enabled us to conclude with precision that, decades ago, ExxonMobil understood as much about climate change as did academic and government scientists. Our analysis shows that, in private and academic circles since the late 1970s and early 1980s, ExxonMobil scientists:

(i) accurately projected and skillfully modelled global warming due to fossil fuel burning;

(ii) correctly dismissed the possibility of a coming ice age;

(iii) accurately predicted when human-caused global warming would first be detected;

(iv) reasonably estimated how much CO2 would lead to dangerous warming.

Yet, whereas academic and government scientists worked to communicate what they knew to the public, ExxonMobil worked to deny it."

Exxon climate graphics from Supran et al 2023

Fig. 2: Historically observed temperature change (red) and atmospheric carbon dioxide concentration (blue) over time, compared against global warming projections reported by ExxonMobil scientists. (A) “Proprietary” 1982 Exxon-modeled projections. (B) Summary of projections in seven internal company memos and five peer-reviewed publications between 1977 and 2003 (gray lines). (C) A 1977 internally reported graph of the global warming “effect of CO2 on an interglacial scale.” (A) and (B) display averaged historical temperature observations, whereas the historical temperature record in (C) is a smoothed Earth system model simulation of the last 150,000 years. From Supran et al. 2023.

 Updated 30th May 2024 to include Supran et al extract.

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

Last updated on 30 May 2024 by John Mason. 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.


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

Denial101x videos

Here are related lecture-videos from Denial101x - Making Sense of Climate Science Denial

Additional video from the MOOC

Dana Nuccitelli: Principles that models are built on.

Myth Deconstruction

Related resource: Myth Deconstruction as animated GIF

MD Model

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

Fact brief

Click the thumbnail for the concise fact brief version created in collaboration with Gigafact:

fact brief


Prev  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  Next

Comments 1026 to 1050 out of 1321:

  1. SCE...  (We have a rule here about no piling on. I see we have four against one going here, so apologies. Just say something if you'd prefer to drop this back to just one specific line from one commenter and the rest of us will oblige.) 

    For my part, I'd just like to point out that the discussion keeps going the same direction to focus on possible reasons for lower CS. Again, this may turn out to be the case, but there are as many (if not more) reasons to believe that CS might be higher than IPCC central estimates. 

    When the Montreal Protocol was implemented industry had been screaming that it would be a business killer. Entire industries were going to be decimated by this attempt to regulate SO2 emissions to address acid rain. After it was implemented, quite the opposite happened. There were costs involved but it ended up being far less economically impactful than estimates. 

    We have a far more critical situation with CO2. Even if CS is 2°C, that would only mean we have an additional decade (maybe) to address the problem. If CS ends up being higher... then we're really behind the 8 ball. By all estimates the most rational response is to agressively start addressing this asap. The scientific community has been saying this for a long time and politicians have failed to respond, and they've failed to respond specifically because of efforts by the fossil fuels industry to seed doubt in the minds of the general public. Curry and Lewis are very much part of that effort.

  2. Michael... I'm also old enough to remember it too. People born after the 1970's when the Clean Air Act was implemented just don't grasp what it was like. What I remember was both the thick black air and the smell of the local plastics factory, which was located right next to our neighborhood. 

    And I didn't grow up in NYC or LA. This was out in East Tennessee! The river that ran through town stank from the raw sewage that was being dumped there. The air was filty, especially in the winter when everyone was burning coal in basement furnaces to heat their uninsulated homes. 

    Ultimately, the cost of cleaning these things up is less than the costs they are imposing on the economy in the first place. I'm certain CO2 is going to be very much the same.

  3. "Rather that only in the satellite era do we have a spacially dense data set adequate for capturing most of the relevant phenomena that must be captured and calibrated in the models."

    I would not doubt for a moment that satellite data is a monumental advantage for understanding climate. However, examining the error bars on pre-satellite trends, I do not agree with your assertion that these are not useful for validation/calibration of models. If you mask your model output with same coverage that was observable then, and the model output does not match observation within those error bars, then the model is wrong. This is very much approach that must be taken for paleoclimate or even post-industrial period.

    With temperature in particular, the strong spatial correlation of temperature anomolies does compensate for poor coverage to some degree.

  4. Glenn Tamblyn @1021 & 1023 and scaddenp @1028

    The paleoclimate data is interesting, but I have concerns about how relevant it is for our world today and for predicting climate dynamics into the future. My biggest concern is that climate sensitivity is likely highly dependent on the prevailing ocean circulation patterns, since these dominate the heat exchange between the atmosphere and the oceans, thus amplifying or moderating greenhouse gas effects.  For this reason I would be very skeptical of any conclusions based on data older than ~30 million years, since at that time the modern continents were still forming and ocean circulation must have been different. Presumably, there are other such factors that would make even more recent data of questionable value, though I'm no expert on the topic and would be interest to hear from anyone with such knowledge.  Specifically, how relevant is paleoclimate data to today's world and how far back do we still have a modern climate system (eg. modern ocean and jetstream circulation patterns)?

    My next concern is the accuracy and precision of the available climate record.  During the instrument era (~200 years), we have daily, monthly, and yearly data, accurate to within tenths of a degree.  By contrast, the error bars for paleo reconstructions are surely much larger, probably on the order of degrees or even tens of degrees.  Furthermore, depending on the particular proxies, they often represent annual or at best seasonal averages.  Thus, it becomes hard to distinguish a short-term period of extreme temperatures from a longer bout of moderate temperatures.

    Finally, as I mentioned before, the paleo data is somewhat geographically sparse.  So, what is interpreted as a large gobal climate shift may simply be a local, temporary abberation.

    Now, I'll shift back to the topic at hand, which is"How reliable are climate models?". The problem is that if the data set against which you are validating the model has large error bars, significant uncertainties in the temporal resolution, and large spacial gaps, it becomes too easy to tweak the model such that it fits the data, but for the wrong reasons.  For example, one of the biggest challenges for the models is handling cloud coverage.  The unit cells are often much larger than individual clouds, so parameterizations must be used to represent cloud coverage.  This means the entire cell has some "average cloud effect", which may or may not reflect reality.  The result is that the model includes a "fudge factor" of cloud coverage, which cannot be independently verified. 

    Another such factor involves aerosols and particulates (think Sea-spray or Dust storms). In these cases, you need to know both the variation in the particle and aerosol densities, as well as the sensitivity of the climate to these densities.  Clouds, particulates, and aerosols are important phenomena, which are known to impact global temperature and climate, so you can't ignore them, but we are just beginning to understand the factors that drive them, so by necessity our current models are quite crude.  This means, we can probably "fit" many models to match paleo data, but it doesn't mean we are doing it correctly or that those models will be able to predict the future climate.

    Only in the satellite era are we beginning to get the proper instrumentation, so that we can monitor cloud, aerosol, and particulate densities so as to verify that the assumptions that are put into the models are reasonable.  Because of this, I would say that our current models are not very reliable, but there is hope that within the coming decades they will become much better.


    [PS] You make the assertion "My biggest concern is that climate sensitivity is likely highly dependent on the prevailing ocean circulation patterns, since these dominate the heat exchange between the atmosphere and the oceans, thus amplifying or moderating greenhouse gas effects", implying unmodelled feedback effects. To me this looks like hand-waving in extreme so please provide some sources for supporting that opinion.

  5. michael sweet @1024

    Why should I cite a peer reviewed report to back a claim that I never made? I never said that changing to renewable energy will be bad for the economy.

    I stated that, "the social costs of planning for an overly pessimistic ECS would also be tremendous."

    So, let's go for a worst-case IPCC ECS estimate of 6degC/doubling. How much would the U.S. need to cut it's CO2 emissions to avoid a 2degC increase in global temperatures? What would such a reduction cost?

    Now don't forget that in the U.S. we don't make the best choice, rather we make the most politically expedient choice. So, how would we likely make up the difference? Probably by expanding fracking of natural gas (basically, what we've already been doing) and by expanding our subsidy to the Corn industry for ethanol-based fuels. Of course, the latter is especially troublesome, since it would cause food prices to spike, which would hit the most vulnerable in our society and because of the intensive use of hydrocarbons in agriculture, it would probably do little to reduce CO2 emissions.

    It's important to get the science right and make reasonable attempts to adapt on reasonable time-scales.

    As for renewable energy, it has it's place, but I'm skeptical it can be implemented at sufficient scale to replace our existing infrastructure in a timely manner. Large-scale adoption of wind and solar will also run into vital material shortages (eg. niobium for generator magnets).  Finally, government subsidies to encourage adoption of renewables are tricky and can lead to unintended consequences (see Arizona's attempt at encouraging alternative fuels:

    Finally, with respect to renewables, I think we'd be better off investing in Nuclear as it's a proven technology with scale sufficient to meet our electricity needs,  but that's a debate for another time.


    [DB] Off-topic snipped.  Please post on the topic of the OP of this thread.  Other topics and threads can be found using the Search function.

    Please note that posting comments here at SkS is a privilege, not a right.  This privilege can be rescinded if the posting individual treats adherence to the Comments Policy as optional, rather than the mandatory condition of participating in this online forum.

    Please take the time to review the policy and ensure future comments are in full compliance with it.  Thanks for your understanding and compliance in this matter.

  6. Latest climate model vs observations update - courtesy of Zeke Hausfather @ Berkley Earth:

  7. SCE,

    Simply repeating a false claim that you read on the internet does not make it an argument.  Your claim that "the social costs of planning for an overly pessimistic ECS would also be tremendous" is simply false.  

    The US Academy of Science warned President Johnson in 1965 that AGW would be a crisis for mankind in the future.  How long do you want to wait??  Scientists have been sure since the 1990's that this is a critical issue that must be addressed immediately and you want to wait? 

    Let us look at the worst case . We start to seriously build out renewable energy (WWS).  After 10 years it is more expensive than we expected and the consequences of AGW are less than we expected.  We stop the build.  We end up wasting a few hundred billion dollars on the energy build.  The money goes to pay workers who spend it and improve the economy.  It is a wash overall.  

    Since every time the IPCC puts out a new report it increases the damage already caused it is extremely unlikely the worst case will happen.  The report I linked above (which you do not appear to have read) documents billions of dollars damage that are being caused already.  To mention only today's newspaper, wildfires caused by AGW have caused several hundred million dollars damage already and the fire season has barely begun.  The 11 million people who will lose their homes from sea level rise in Florida alone can go live in your house.

     We wasted over two trillion dollars unsuccessfully trying to steal Iraq's oil. If we switch to WWS we will not need half the military we currently have. Jacobson estimates that WWS will cost less than conventional energy. The cost of solar and wind have plummeted since Jacobson made these estimates so his cost estimates are too high.  You must support your claim that fossil fuels have a comparable cost to WWS.  I have not seen any serious analysis that suggest fossil fuels are as cheap as WWS when all costs are included.

    The cost of wind and solar have plummetted the last few years and they are now the cheapest form of energy period.  You are arguing to preserve more expensive energy sources for what??  Jacobson identified the need for alternate materials for the turbine magnets years ago.  Iron nitride magnets are one alternative.  Others are being developed.   Where are you going to get the beryllium you need for all those nuclear power plants?

    No mater what is done, fossil fuels will run out in a few decades.  When do you think we should start the switch to WWS?

    You are skeptical of WWS.  Perhaps if you read more about them you would be less skeptical.  Read the Jacobson 2015 article I linked above.

    Several hundred thousand people per year die from air pollution in the USA alone from your preferred energy source.  All these people will live longer if we switch to WWS.  The cost of medical care caused by coal pollution alone is over $40 billion per year.  We will save  all that money.

    I am amazed that you claim (without any supporting data or citations) that WWS is a risky choice and then you are in favor on Nuclear.  Toshiba is on the verge of bankruptcy from its nuclear adventures.  No market economy will build any nuclear for decades.  Raw materials (like beryllium and uranium) do not exist for a widespread nuclear effort.


    [PS] This is wandering completely offtopic. Unless there is comment on modelling, please use the search box to find a more appropriate thread and continue there.

    SCE - please avoid stating opinions and hand-wavey arguments without providing the data/sources which form the basis of your opinion. And stick to topic.

  8. SCE,

    Here is a more suitable thread to continue this discussion.


    [PS] Thank you

  9. PS inline @1029, SemiChemE's speculation is, I think, fairly obvious.  Examples of changes in ocean circulation that would effect climate sensitivity include the opening of the Drake Passage between Antarctica and North America, the opening of the gap between Australia and Antarctica,  the closure of the Central American Seaway between North and South Americas (with consequent strengthening of the Gulf Stream), and even the northward drift of Africa which is slowly strengthening the leakage of warm, saline water from the Indian ocean to the Atlantic Ocean, which leakage may have a significant effect on deglaciations.  The extent of the effect on climate of these events is disputed in each case, but that they have some effect is, I think, incontrovertible.  Each may have some role in the onset and/or intensity of recent glacial cycles; a sign that they have increased climate sensitivity over the geological average.

    Nor are they the only geological changes that can impact climate sensitivity.  Geological evidence shows that in periods with glacial caps climate sensitivity is greater.  It follows that geological events that significantly increase or decrease CO2 concentrations will, respectively, decrease or increase climate sensitivity.  The draw down of CO2 concentrations that resulted from the formation of the himalayas, therefore, has also likely increased climate sensitivity.

    Despite agreeing with SemiChemE's speculation, however, I do not agree with his conclusion.  We have climate sensitivity estimates across a very wide range of continental positions and glacial conditions.  While the estimate of climate sensitivity from paleo data for one particular time may not be a reasonable analog of the modern Earth, we can expect the range of climate sensitivities across all times to constrain likely climate sensitivities today.  In particular, we can expect modern climate sensitivity to be above within the range of climate sensitivities found in glacial conditions (as we are in a glacial condition).  


    [PS] Understodd. Thank Tom, and apologies to SCE for not following his line of reasoning.

  10. Here are a few references, discussing the importance of Ocean-effects on global Climate Sensitivity.

    Balmaseda, Magdalena A., Kevin E. Trenberth, and Erland Källén. "Distinctive climate signals in reanalysis of global ocean heat content." Geophysical Research Letters 40.9 (2013): 1754-1759.

    Meehl, Gerald A., et al. "Model-based evidence of deep-ocean heat uptake during surface-temperature hiatus periods." Nature Climate Change 1.7 (2011): 360-364.

    Raper, Sarah CB, Jonathan M. Gregory, and Ronald J. Stouffer. "The role of climate sensitivity and ocean heat uptake on AOGCM transient temperature response." Journal of Climate 15.1 (2002): 124-130.

    There are many, many more such papers. Clearly, the oceans are very important to climate Sensitivity.

    As for the evolution of Ocean Currents over time, here's a nice summary:

    As you can see, prior to ~30 million years ago, the arrangement of the continents was different, which had a dramatic impact on ocean circulation patterns. The Atlantic Ocean was much smaller and the isthmus of Panama had not formed. Given that Atmospheric-Ocean coupling is a major factor in determining climate sensitivity (see references above) and that due to geological changes in the configuration of the continents the Ocean circulation patterns were different, it is entirely reasonable to believe that climate sensitivities prior to 30 million years ago may have been different from those today. Thus, as I stated before, I would be very skeptical of the relevancy of paleoclimate data from >30 million years ago for predicting the modern climate.


    [PS] Fixed link

  11. DB inline @1030, I appologize for letting another poster pull me into an off topic discussion.  My interest is in modelling and I will try to constrain my posts to that topic.

  12. SemiChemE @1035, the three papers you cite all deal with the Transient Climate Response, formally defined as the temperature achieved at 70 years after a 1% increase in CO2 per annum over the 70 year period.  The Equilibrium Climate Sensitivity (ECS) is the condition that obtains when quasi equilibrium between energy in and energy out is obtained at the Top Of the Atmosphere.  The difference between the two is a function of how much energy must be stored at the Earth's surface (primarilly in the oceans) for the temperature to rise sufficiently to obtain an energy balance.  Because of this, the three papers have no bearing on the value of the ECS except in that the TOA energy imbalance is used in energy balance estimates of ECS to change an estimate of the TCR to an estimate of the ECS.  (That wording may be confusing, for which I apologise. If necessary I shall explain more clearly later when I have more time.)  The papers certainly have no bearing on the issue of estimating the ECS from paleo data. 

  13. Tom Curtis @1034, Thanks for understanding the point I was trying to make and giving a better explanation than I could have (see post #1035) for why paleoclimate data from >30million years ago may not be useful for predicting the earth's climate sensitivity to CO2 in modern times.

    As for my conclusion, your post suggests I was not clear in stating my conclusion, since your argument appears to be about the likely range of climate sensitivities. I did cite a paper (or papers) that reflect a lower climate sensitivity, but my point in doing so was to highlight potential flaws in the models that might cause them to make improper predictions about future climate trends.

    My intended conclusion was that climate models are still quite crude and unreliable for predicting the future climate. I do have hope that the models will get better over time, especially in light of modern data collection techniques (eg. Satellites, Argo sensors, etc...), which will enable modellers to reduce the acceptable ranges of the parameters that are currently used to adjust the model outputs.

    I also argued that paleoclimate data is not sufficient to completely validate any given model due to
    1. Limitted accuracy and precision
    2. Poor temporal resolution
    3. Significant gaps in global coverage
    4. Limited visibility to important historical factors, including cloud behavior, aerosol and particulate variations, Ocean Currents, etc...

    Finally, while I believe my statements about Paleoclimate data to be true, I am certain there is a literal army of climate scientists working to address these shortcomings and I would welcome any suggestions for a good summary on the latest state of the art in understanding our planet's climate history.

  14. Skeptical questions from a lay person:  What if the accuracy of climate models does not continue to improve as is claimed, and the current error rates in predictions of global temperature each year continue at their current rate?  Is it possible that the aggregative upshot of serial errors in temperature prediction could lead to a very different result than that which is currently being predicted by the present day models?  And isn't the only relevant question for members of the public whether the climate models can accurately predict what happens in the future?

    I am not denying that the physical science and math and statistics that goes into climate models are not scientifically valid and independently accurate in other applications.  What I am questioning is whether they have ever been demonstrated to have the level of predictive value which would be necessary to project policy 50 years into the future and beyond.  

    An analogy:  In the realm of medicine, prior to a treatment or test being administered it must be shown that the treatment or test is effective.  When we are talking about a particular method which is in essence a test (to predict increased planetary temperature) the test must be capable of predicting what it is meant to predict.  For example, the law does not allow pregnancy tests to be placed on the market, when such tests have not been consistently shown effective at predicting that a woman will eventually have a baby in actual real world clinical trials.

    In my mind, climate science is similar.  Climate science is an amalgum of scientific techniques and human judgments that can be thought of as a particular test (albeit much more complex than a pregnancy test) which is being used in order to predict the planet's future temperature.  The lay people of the world are being asked to make serious policy changes with far-reaching negative economic ramifications on the basis of this particular "test" or methodology.  Therefore it stands to reason that this "test" of climate change must be able to demonstrate that it has a record of being successful in predicting global temperature changes.  Can we really say that?  The discrepancy in the above graph between predicted and actual seems to belie that the "test" is really there yet.  By the way, the same problem of lack of sufficient demonstrated predictive value for the purposes asked also exists in the political world, where everyone was wrong about Trump's chances.

  15. [I noticed a few typos in my previous post, in last paragraph it should read "Therefore it stands to reason that this "test" of climate science"

    In second paragraph should read "What I am questioning is whether *their use in combination with other methods in climate science as it is currently practiced* has ever been demonstrated....]

  16. I am not quite sure what you mean by "error rates in prediction"? What do you regard as "error"? Strawman arguments are favourite denier attack - eg . "climate model predict this (cue cooked graph), actual is this, therefore climate science is wrong". The important point to consider is what do climate models actually predict here. The answer is climate, not weather. They have no skill at decadal level prediction and dont pretend to. What they do predict with considerable skill is what 30 year weather averages will do. In the graph at the above comment  you will see the grey area is "weather uncertainity". What this means is that any wriggly line in that grey area is consistant with the climate models. The solid black line is model average. You can see this more clearly on this AR4 graph.

    Every single line is an individual run of the model. Every one of them, a possible climate future. Discussed in more detail here.

    Furthermore, whatever their imperfections, (and modellers would be first to list them), they remain our best predictor for what future climate will look like. Yes, we would very much prefer to know whether climate sensitivity is closer to 2.5   or 4.5, but this is best possible at moment. Dont assume errors will be on the side of least effect. Uncertainity is not your friend.

    "with far-reaching negative economic ramifications". Hmm, sounds like drinking the FF propoganda to me.  Certainly negative for some industry sectors, but what are you using as the basis your assertion on negative consequence? Want to compare them with the costs on doing nothing and even a low sensitivity of say 2.5?

    Also, please dont confuse physical models for predicting the future with statistical models (eg polls). Not a lot in common for functionality.

  17. You also bring up medical test - the idea of what is "effective" is complex and the conclusions from testing can be very murky indeed. However, we dont see joe public (on the whole) thinking they know better than the medical consensus. Judging those results requires considerable domain knowledge, just like climate science. The scientific consensus might be wrong - plenty of cases where it is - but scientific consensus is what should guide public policy.

    Also, treating the models as test of climate science is also are rather dubious assumption. There are numerous direct tests of climate science instead. Models though are limited by computer power, measurement gaps, subscale issues etc and underlying chaotic processes. This is not test of the science. The correct question to ask is "are the models skillful?". The answer is yes. They sure do better than the entrails of chicken or assuming climate is unchanging, and frankly, very very much better. Broecker 1975 practically nailed the 2010 temperature.

  18. DavidS @1039,

    you are overlooking a vitally important point, which is :- global warming is already occurring now in a big way.  There are no Ifs or Buts or Maybes.  The signs (like with an advanced pregnancy) are very obvious, right now.  The higher surface temperatures, the rising sea level, the loss of glaciers and plummeting arctic ice volume, and so on.  All that is very plain and obvious.  And since the known cause (of warming) is still operational and actually increasing — by around around 30 billion tons per year — the warming will continue strongly.

    Even without any model projections for the future, it is clear to the reasonable man [as the lawyers describe him] that action needs to be taken now, rather than postponed for many decades.  In a half-century or less, fossil fuels can be phased out, with little or no real economic cost (bearing in mind the already high price of health costs and other less obvious costs of operating a coal/oil fuelled economy).

    The presence or absence of model projections is a relatively trivial matter, which may be of greater relevance to the lattermost decades of the 21st Century : but has minimal importance in practical terms of what actions we need to take both now and in the near future.

    DavidS, your "pregnancy test" analogy is a very long stretch.

    Perhaps a slightly closer analogy is this :- A young woman reports she has had no periods for 4 months : and her abdomen shows a smooth midline lump reaching almost to her umbilicus.  Plus several other signs are present, indicative of pregnancy.

    Yes, there is a 0.1% chance she is not pregnant.  But the reliabillity and accuracy of a urine pregnancy test is almost moot.  Her real interest and need, is to make decisions about practical things she should do now and in the coming months.  She would be wasting her time if she agonises over whether the urine test kit is 95% or 98% accurate.

    Timely practical decisions are needed, in tackling our rapid global warming.

  19. DavidShawver:

    Something for you to chew on...

    Researcher's 1979 Arctic Model Predicted Current Sea Ice Demise, Holds Lessons for Future

    Study from decades ago proved remarkably accurate in showing how global warming would affect the Arctic's sea ice, currently in steep decline.

    by Sabrina Shankman, Inside Climate News, Feb 20, 2017

    Claire Parkinson, now a senior climate change scientist at NASA, first began studying global warming's impact on Arctic sea ice in 1978, when she was a promising new researcher at the National Center for Atmospheric Research. Back then, what she and a colleague found was not only groundbreaking, it pretty accurately predicted what is happening now in the Arctic, as sea ice levels break record low after record low.

    Parkinson's study, which was published in 1979, found that a doubling of atmospheric carbon dioxide from preindustrial levels would cause the Arctic to become ice-free in late summer months, probably by the middle of the 21st century. It hasn't been ice-free in more than 100,000 years.

    Although carbon dioxide levels have not yet doubled, the ice is rapidly disappearing. This record melt confirms the outlook from Parkinson's 1979 model.

    "It was one of these landmark papers," said Mark Serreze, director of the National Snow and Ice Data Center. "She was the first to put together the thermodynamic sea ice model."

  20. David... Other's have offered quite a lot, but here's one more element that you might consider:

    With climate science you essentially have one subject on which to perform any experiments or analysis. The subject has a very slow metabolism and a very long lifespan. And, all the relevant physics and effects only occur on or near a very thin surface area on the subject.

    This presents both advantages and disadvantages relative to medical research. But moreover, it merely means the methods that scientists must take to understanding the subject matter has to be approached in very different ways. It's very hard to compare the two in any meaningful manner.

  21. David @1039... You appear to misunderstand the nature of medical testing and medical prognostication.

    You pregnancy example misses the point. Pregnancy is unusual in medicine, because it is an almost pefect binary condition - someone cannot usually be "a little bit pregnant" (ambiguous cases can actually occur, but they are rare). And of course, that binary nature partly reflects that pregnancy is not even a disease, but is instead a highly evolved biologically programmed physiological state. With pregnancy, there is little of the conceptual messiness that is usually associated with defining a disease and deciding which cases to lump together under the same categorical label. Pregnancy testing is also unusually accurate compared to nearly any other medical test you could have named.

    Most medical conditions are less well-defined, and cannot be modelled with any accuracy. Although it is often known that, say, treatment A will be more effective than placebo, it is often not even known whether treatment A will be better than treatment B. Moreover, it is rarely the case that the precise disease course for an individual patient can be plotted predictively. For most cancers, for instance, a specialist will often quote an approximate median survival, which is no more than the time interval within which they expect half the patients with that cancer to die. For the individual patient, the actual surivival time is likely to diverge substantially from that median. Other times, the specialist may quote the expected 5-year survival as a percentage, but for an individual patient, 5-year survival will either be 100% or 0%, so the crude 5-year survival model does not apply.

    Insisting on perfect prognostication before acting would be foolish in a medical context. If even one oncologist reported that a lung mass was an early-stage cancer, and that removing it would be associated with greatly improved median survival, then most people would have the mass removed. If a second, third and subsequent opinion is concordant, then it would be crazy to leave the mass in place, refusing to cooperate until the oncologist provides an accurate chart of its projected growth. It would be crazy to wait and confirm that the cancer really was capable of spreading to other organs, etc.

    For climate science, we have the added problem that there is only one planet, and this is the first time that AGW has occurred, so we have to act before fine-tuning the prognostic model.

    Don't confuse uncertainties in the fine points of prognostication with uncertainties in the diagnosis. There is no serious doubt about the planetary diagnosis at this point, and it is obvious what we need to do to fix it.

  22. I would first like to state that I have finally found a website that is balanced on this very emotional issue. I also want to thank SemiChemE and Tom Curtis (along with a few others) who have engaged in a very fascinating discussion on climate models.

    My intention is to pose a question on climate models in keeping with this blog, however because this is my first post, I would like to explain my background. I am a lawyer by training and have a very limited base in physics (took Latin in Grade 12 rather than physics) although I always did well in science. I will also disclose that I do have an involvement in the Canadian oil and gas industry notwithstanding that I live in Vancouver, BC.

    Ever since the issue of global warming came to the fore in the late 1990’s and since, I have to admit that I have tended to accept the “scientific consensus” if only because I had no reason to question it. Climategate shook my confidence in 2009 but if Neil DeGrasse Tyson still believes that the principal causes are anthropogenic then far be it for me to question it. However, it always seemed logical to me that a first step in reducing the effects of CO2 should be to move from oil and coal to natural gas (especially for electrical generation) which puts about one half of the pollutants into the air compared to coal and oil. After spending enough holidays in France, I have also thought that a switch to nuclear energy made more sense than disfiguring our planet with massive wind turbines and great areas of solar panels. Driving from LA to Palm Springs is not a pretty sight. But I do appreciate that there are real concerns relating to disposing of nuclear waste and issues of terrorists getting their hands on nuclear fuel. However, someone as significant as James Hansen believes that we will not achieve our goals without a turn to nuclear energy.

    In any event, my recent interest in the causes of global warming really came about because I have two sisters who are just about no longer on speaking terms owing to their disagreements on global warming. When one sister called me asking where I stood on climate change and if I truly believed that this was all a “global conspiracy of the left” to increase taxes and government control over our lives, I promised her that I would buy some books on both sides of the argument and get back to her. Needless to say, I do not believe in conspiracy theories of any sort.

    So the two books I located were The Science & Politics of Global Climate Change by Dessler and Parson and Climate Change the Facts edited by Alan Moran.

    By the time I was finished with Dessler’s book I was convinced of the science. Then I read the essays in the Moran book and found myself at least questioning some things.

    I actually then went back and re-read Dessler’s book to see where the gaps were. I have to say that when I found that Mark Steyn had an essay in the Moran book I almost did not read the book because of his extreme views. Just to make my political views clear, I think Donald Trump poses a major threat to liberal democracy in the US and to the world in many ways. But it does look like the US institutions may be able to withstand him and his cohorts. I also follow Sam Harris’s podcasts “religiously”.

    Since reading these two books I have largely pursued my research on the web even reading the submissions of the four climatologists on March 25, 2017 to the House Committee on Science Space and Technology.

    Based upon Judith Curry’s reference in her submission, this led me to the most fascinating discussion of the topic of climate models by a panel of physicists formed by the American Physical Society (APS) which posed questions to six (6) well-known climatologists having “different perspectives”. Three (3) of them (Collins, Santer and Held) are IPCC climatologists and the other three (3), Curry, Christy and Lindzen are on the other side of the debate. This was the 2014 Workshop sponsored by the APS as part of its 5 year review of its Climate Change Policy Statement.

    As a lawyer, I have to admit that if I treated (i) the IPCC 2013 Assessment as an appellate lawyer’s factum, (ii) the Workshop Framework posed as questions from the bench, and (iii) the 600 page transcript of the panel hearing as the “give and take” between the judges and lawyers during the oral argument of the appeal, I would have predicted a “win” for Curry, Lindzen and Christy and a “loss” for Collins, Santer and Held. Both the Workshop Framework questions and the transcript are on the website. Just search “Climate Change Statement Review”. If anyone has read any legal transcript of a hearing you know it is a simple read so don’t be put off by the “600 pages”.

    The APS panel consisted of six (6) arm’s length physicists (with no axe to grind) chaired by Steve Koonin who were asking hard questions of both sides. What actually struck me as very astounding was how honest Koonin was about his previous lack of understanding as to how uncertain climate science is owing to the uncertainties underlying the climate models.

    This panel hearing took place in February 2014. By November 2015, the judgment of the Board of Directors of the APS was in. The connection between increases in CO2 and global warming was “compelling”. However, the APS did acknowledge that there were significant uncertainties in the science and urged sustained research in climate science.

    Where my comparison with an appellate hearing breaks down is that no appellate court would render a significant judgment without providing its reasons. We do not get any reasons from the panel as to why it recommended to the Board of Directors (as I assume it did) that the APS “stay the course” with its policy statement notwithstanding the serious reservations you could see in Koonin’s and other panel members questions to Collins, Santer and Held and the weak answers provided by them. The IPCC climatologists in effect admitted that Christy’s now famous chart showing how far apart the average predictions of the climate models were from actual observations was “old information” and did in fact represent the existing state of models predictions versus observations. See Santer page 504. The IPCC climatologists effectively said that they could not trust the observations! Koonin’s rhetorical question to Held to this "observational" response earlier was: “So the ability then to reproduce historical data is neither necessary or sufficient to predict the future. Is that what I understand?” See page 453 of the transcript. Held effectively avoids answering the question. See page 454. Read it yourself and see if you disagree with my view of his response.

    So here is my question.

    From everything that I have read so far, other things being equal, a doubling of CO2 in the atmosphere from pre-industrial levels of CO2 of around 280 ppm to 560 ppm will increase the global surface temperature by about 1 to 1.2C and the balance of the predicted range of 1.5C to 4.5C of the IPCC 2013 Assessment is based upon “positive feedbacks” resulting from increased water vapour that is assumed will form arising out of the 1C increase by CO2. I accept (or understand) that the 1C increase is “solid physics” or “hard science”.

    Is it “solid physics” that:

    1. Water vapour will in fact increase as modelled?
    2. Water vapour will cause the predicted additional increase in temperature by a factor of 2 to 3 times?

    Although this next observation is not specifically focused on the climate models, what also troubles me in everything that I have read so far on climate change is the following:

    1. The Mediaeval Warming Period had temperatures for at least 200 years in at least Greenland and Northern Europe close to or equal to our present temperature.

    2. During the 1600's and 1700's there was a "Mini-Ice Age" when they were skating on the Thames.  We were just coming out of this cold period at the beginning of the American Revolution (good timing).

    3. From 1990 to 1940 we experienced about .3C warming; then from 1940 to 1975 there was a levelling off or cooling period; then from 1975 to 1998 we experienced .5C warming; and then there was a levelling off (termed the “hiatus” by the IPCC) of now about 17 plus years that may or may not have ended in 2015 (El Nino event 2015-2016). I appreciate that 1998 was an El Nino year but the IPCC 2013 Assessment recognized the “hiatus” up to that time.

    If climatologists cannot explain why these other warming and cooling periods occurred which, other than the 1975-1998 period, were primarily or completely caused by natural climate change, then why can they so confidently claim that this one warming period was primarily caused by the CO2 rise?  Just because there was a concomitant rise in CO2?  What about the rise of CO2 from 1950 to 1975?

    The models predicted that our temperature would increase on a linear basis. There were no “waves” in the models. I guess based upon Michael Mann’s most recent testimony the most recent peer reviewed papers are now suggesting that we will be going up in steps or waves. Can we now expect that the new models with show the steps or waves?

    So where did the “warming” go during this “hiatus”? If the answer is into the oceans, then why did the “warming” not come from the oceans during the period 1975-1998? Could we have had a cooling period during the “hiatus” that offset the warming from CO2 during this period? What are the impacts of a decrease or increase in low clouds caused by natural factors which impacts the amount of sunlight hitting the earth? Because of computer capacity issues, we can only make “parameterizations” of clouds in the models.  These are the kinds of questions that make me question the validity of the models.

    When we talk of the difference between weather and climate we say we cannot predict weather but we can predict climate because we know that next July it will be warm. But why do we know that? Not from models, from observation. If observations and models do not correspond, when do we admit that the models do not have sufficient predictive value to be relied upon? It is OK for science to say “we just do not presently understand the science sufficiently to make reasonably accurate predictions”.

    On the other hand, here are the major science societies of the world like the APS, the US National Academy of Sciences and the UK Royal Society coming out strongly in support of the proposition that man-made global warming is a serious problem and is going to get worse. My worry is that they got on a band wagon in the early 2000’s before the “hiatus” was apparent and they now find it very difficult to get off even when they see that these models are not predictive.

    I apologize for such a long-winded initial blog. If you think I have to reduce it, please advise.


    [PS] Welcome to Sks. We really like to keep things on topic, (see the comments policy) so we would prefer you put your questions about MWP on the appropriate thread after first reading the article. (your questions are pretty well answered in the IPCC WG1 as well). Ditto, other questions not about modelling. Could responders please stick to modelling questions only on this topic? Thank you for your cooperation.

  23. NorrisM, initially I'm going to assume you are who you claim to be, though the content of your post makes me suspicious--very suspicious--that you are one of SkepticalScience's fake-skeptic, trolling, chronic sock-puppeteers, and one in particular.

    Your statement

    The APS panel consisted of six (6) arm’s length physicists (with no axe to grind) chaired by Steve Koonin who were asking hard questions of both sides. What actually struck me as very astounding was how honest Koonin was about his previous lack of understanding as to how uncertain climate science is owing to the uncertainties underlying the climate models.

    is incorrect. Steve Koonin is a notorious fake skeptic, who has both the background and the subsequent, repeatedly delivered, information to know that most of what he says and writes is factually and drastically incorrect. Christy has and continues to make claims that are factually incorrect, and is motivated primarily by political and religious beliefs. Christy's partner in crime is Roy Spencer, who is a member of the Cornwall Alliance that claims human-caused global warming is impossible because God promised Noah there would not be any more floods. Really. LIndzen's pet theory about the "iris" mechanism that self-regulates the Earth's temperature conclusively and repeatedly has been proven wrong (obviously, since Earth's temperature has varied drastically--Snowball Earth, ice ages,...) but that has had no effect on his opinion, and he very much resents and takes personally the criticisms. Curry once was an adequately productive climate scientist, but for reasons I won't speculate on here, has become quite the opposite.


    [PS] That is uncalled for tone. I know all too well how tiring the denier trolls are but polite and reasonable enquiries are to be encouraged, as are substantive responses.

  24. NorrisM, your claim

    The models predicted that our temperature would increase on a linear basis. There were no “waves” in the models.

    is incorrect.

    All the models projected (not "predicted") inconsistent ("wavy") temperature increase. None projected monotonic ("linear") increase. The mean of models usually is shown in graphs, but all the individual model runs that contribute to that mean are "wavy." For ease of viewing, sometimes graphs of climate models show only the models' mean, or show the individual runs' spread as a shaded area. But the actual runs look like spaghetti.

    We do not expect the actual temperature to fall exactly on the "ensemble mean" hindcast and forecast, because that mean has far too little variability. In fact, we expect the temperature to have wild ups and downs as you see exemplified by the orange and blue skinny lines in Figure 2 here, because we expect there to be El Ninos and La Ninas, variation in solar radiation, volcanic eruptions, changes in human-produced aerosols due to varying economic activity, and a slew of random and semi-random factors. It would be downright shocking if actual temperature followed the ensemble mean, because that would mean all those variations in forcings and feedbacks were far less variable than we have observed so far. So we judge the match of the real temperature to the projected temperature by whether the real temperature falls within the range of the entire set of individual model runs. Even so, we expect the real temperature to fall within that range only most of the time, not all of the time. The range you see drawn as a shaded area around the ensemble mean usually is the 90% or 95% range, meaning the set of individual model runs falls within that shaded range 90% or 95% of the time. That means, by definition, we fully expect the real temperature to fall outside that range 10% or 5% of the time. So occasional excursions of the real temperature outside that range in no way invalidate the models, when "occasional" means 10% or 5% of the time.


    [PS] An example graphic of real output of models and discussions is here. Realclimate also has a FAQ on models - written by the modellers themselves so not heresay.

  25. NorrisM, you asked

    Is it “solid physics” that:

    1. Water vapour will in fact increase as modelled?
    2. Water vapour will cause the predicted additional increase in temperature by a factor of 2 to 3 times?

    Answers: Yes. I assume you are familiar with the concepts of relative and absolute humidity, at least from weather reports you've read and seen your entire life. Perhaps you've noticed that cold air (e.g., winter) is drier than warm air (e.g., summer). For more details, read the Basic and then the Intermediate tabbed panes here.


    [PS]"In addition, water vapor concentrations have increased throughout the troposphere at about 1.2% decade−1 since the ERBE period (Trenberth et al. 2005, 2007b). " from here. Also see Clausius–Clapeyron relation for seriously settled science. Water vapour is not the only feedback.

Prev  32  33  34  35  36  37  38  39  40  41  42  43  44  45  46  47  48  49  50  51  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


(free to republish)

© Copyright 2024 John Cook
Home | Translations | About Us | Privacy | Contact Us