Skip to content

Latest Climate Models Still Running Far Too Hot

February 24, 2021

By Paul Homewood



The narrative that global warming is largely human-caused and that we need to take drastic action to control it hinges entirely on computer climate models. It’s the models that forecast an unbearably hot future unless we rein in our emissions of CO2.

But the models have a dismal track record. Apart from failing to predict a recent slowdown in global warming in the early 2000s, climate models are known even by modelers to consistently run hot. The previous generation of models, known in the jargon as CMIP5 (Coupled Model Intercomparison Project Phase 5), overestimated short-term warming by more than 0.5 degrees Celsius (0.9 degrees Fahrenheit) above observed temperatures. That’s 50% of all the global warming since preindustrial times.

The new CMIP6 models aren’t much better. The following two figures reveal just how much both CMIP5 and CMIP6 models exaggerate predicted temperatures, and how little the model upgrade has done to shrink the difference between theory and observation. The figures were compiled by climate scientist John Christy, who is Director of the Earth System Science Center at the University of Alabama in Huntsville and an expert reviewer of the upcoming sixth IPCC (Intergovernmental Panel on Climate Change) report.

Models CMIP5.jpg

Models CMIP6.jpg

Both figures plot the warming relative to 1979 in degrees Celsius, measured in a band in the tropical upper atmosphere between altitudes of approximately 9 km (30,000 feet) and 12 km (40,000 feet). That’s a convenient band for comparison of model predictions with measurements made by weather balloons and satellites. The thin colored lines indicate the predicted variation of temperature with time for the different models, while the thick red and green lines show the mean trend in degrees Celsius of warming per decade for models and observations, respectively.

The trend for CMIP6 models is depicted more clearly in Christy’s next figure, which compares the warming rates for 39 of the models. The average CMIP6 trend in warming rate is 0.40 degrees Celsius (0.72 degrees Fahrenheit) per decade, compared with the actual observed rate of 0.17 degrees Celsius (0.31 degrees Fahrenheit) per decade – meaning that the predicted warming rate is 2.35 times too high.

Models CMIP6 warming rate.jpg

These CMIP6 numbers are only a marginal improvement over those predicted by the older CMIP5 models, for which the warming trend was 0.44 degrees Celsius (0.72 degrees Fahrenheit) per decade, or 2.75 times higher than the observed rate of 0.16 degrees Celsius (0.29 degrees Fahrenheit) per decade (for a slightly different set of measurements),

Full story here.



Despite misinformation from alarmist scientists to pretend this is not the case, it was the IPCC itself in AR5 which confirmed the woeful performance of models in 2013:



  1. Malcolm Bell permalink
    February 24, 2021 10:50 am

    I watched Sir David getting near hysterical on BBC last night about how we are doomed unless we do something dramatically fast. Poor old
    man has got a bad eye.

    Why does a graduate geologist who
    spent a lot of his early life catching wild
    animals to put in zoos and then introduced BBC2 Colour to Snooker
    and finished doing voice overs for naturalist films made by others think
    he has the authority to shout at us.

    He is just a “celebrity” like so many others.


    • February 24, 2021 12:53 pm

      I don’t blame him in particular, the BBC operates a sieve, anybody who goes off-message falls through, such as David Bellamy.

    • February 24, 2021 1:34 pm

      More, his degree in geology should preclude him from talking any of the tosh that he has been coming out with vis-a-vis climate

      • ThinkingScientist permalink
        February 24, 2021 2:09 pm

        Hasn’t stopped the Geological Society of London falling for the koolaid and spouting the same nonsense.

      • dave permalink
        February 24, 2021 6:47 pm

        We have discussed this before. It seems unlikely that he actually HAS an earned degree. He entered Clare College, Cambridge in 1945 and was called up to the armed forces in 1947, and did not return.

        It is only possible, in two years, to obtain Part I of the Bachelor’s degree. When I was at Cambridge, 1962-65, you HAD to be resident for three years
        and take Part II.

        He is always very vague when referring to Clare. Thus:

        “Learning about the natural sciences which I studied while at Cambridge was exciting…”


        “…all of us who once [sic] belonged to Clare…”

        A graduate ALWAYS belongs to his College – if he graduates!

    • Cyan permalink
      February 24, 2021 4:28 pm

      Watch ‘Horizon’ on BBC2 tomorrow, in which Dr.s Chris and Xand van Tulleken will be blaming the COVID pandemic on climate change and habitat destruction.

      Probably the only way to get on the beeb these days is to blame something on climate change

  2. It doesn't add up... permalink
    February 24, 2021 11:24 am

    If theory doesn’t agree with experiment, then it’s wrong.


    If theory doesn’t agree with experiment let’s just adjust HADCRUT


  3. Allan Shelton permalink
    February 24, 2021 11:27 am

    Computer models that have CO2 as a cause of warming are wrong at the outset.
    They should be ignored. The GHG Theory has been debunked.

    • Malcolm Bell permalink
      February 24, 2021 11:31 am

      Exactly so – except no one in power
      has acknowledged that – especially
      the EU.

  4. Chaswarnertoo permalink
    February 24, 2021 11:36 am

    If your model cannot predict reality it is worthless.

  5. Mack permalink
    February 24, 2021 11:52 am

    Don’t you just love a little ‘settled science’ in all its’ technicolor glory? I bet a decent bookie could give you a better forecast than most of the climate modellers, but then again they work with real world data, observations and form, not the metrics of Fantasy Island.

  6. Patsy Lacey permalink
    February 24, 2021 12:25 pm

    This article by John Lee is actually about Covid but it screams its application to climate change. It talks about the PM’s ignorance of the fundamental limitations and uses of modelling.

    • Gerry, England permalink
      February 24, 2021 2:33 pm

      With somebody as dim as Boris as PM it is not difficult for the likes of Dim Whitty, Ferguson et al to run rings around him. Not known for doing detail it will all be above his head so best his concentrates on dressing up for his next visit somewhere and coming up with his next childish statement in the forlorn hope that it will be considered Churchillian.

      • February 24, 2021 3:59 pm

        Wicked, but so true!

      • Robert Jones permalink
        February 24, 2021 4:02 pm

        Your contribution to the serious matter of correcting the widespread misunderstanding of so-called ‘global warming’ achieves what exactly?

  7. February 24, 2021 12:49 pm

    If your climate models work better without any ‘CO2 forcing’ factor, face reality and get rid of it — and stop wasting everybody’s time and money.

    • A man of no rank permalink
      February 24, 2021 12:58 pm

      I find it odd that when the physicists, Ned Nikolov and Karl Zeller, show that the greenhouse effect was little to do with absorption of radiation, or CO2, that this impressive science is just ignored. They found that ‘compression heating’ was involved and this follows the atmospheric pressure and solar radiation. In a true scientific manner their work was reviewed, published online for all to see and challenge, and it cited over 130 previous papers from the scientific literature.
      In my world these climate modellers should be asked to falsify Ned and Karl’s work and if they cannot do this, they should be shut down as they are heading in the wrong direction. Moreover, the BBC’s emotional campaigners, Harrabin and Attenborough, should also be invited to challenge this proper science – and when they fail they should exit stage left!

      • Ed Bo permalink
        February 24, 2021 8:07 pm

        AMONR: Nikolov and Zeller are not physicists, and they do not even understand basic high school physics. I have taught technical subjects at a major university, and if a first-year undergraduate turned in a paper like theirs, I would reject it out of hand.

        Their grasp of basic mathematics is similarly wanting. Thinking you can demonstrate anything by fitting a curve with as many parameters as data points is a classic mistake of a weak student in an introductory class.

      • Richard Hart permalink
        February 25, 2021 9:16 am

        Hi, I notice Ed No below didn’t establish university they lectured at, or wha bin, or what flaws exist in Nikolov and Zeller’s maths. Tunnel say that a peer reviewed paper is weak and unworthy an undergraduate! Oh and by the way, Ed, undergraduates don’t turn in papers. I call BS on your whole comment.

      • Ed Bo permalink
        February 25, 2021 5:02 pm

        OK, if you insist. The university is UCLA. The glaring problem with N&Z is that static pressure (as opposed to dynamic compression) transfers no power on any ongoing basis.

        If you’re paying attention in high school physics, you learn that the energy transferred by a force (pressure x area) is:

        Energy = Force x Distance

        in terms of power, the equation is:

        Power = Force x Velocity

        In the expansion stroke of a piston engine, the piston is moving from the force, so the velocity is non-zero, and hence the power transfer to the mechanical system is non-zero.

        But in the case of the force on the earth’s surface from the pressure of the atmosphere, the velocity is zero, so the power transfer is zero. This means that the pressure CANNOT provide the additional power needed to maintain higher surface temperature (and match the additional losses) than solar input alone would create).

        If you believe that static pressure by itself can maintain higher temperatures, then you believe that the compressed air in a tire (which initially has a higher temperature than ambient) will maintain that higher temperature indefinitely. Do you really believe that???

        The N&Z analysis was roundly rejected by real journals. They had to publish under pseudonyms because they were so widely considered completely ridiculous.

        And yes, undergraduates do turn in papers at many universities. But in whatever form — papers,reports, problem sets, tests — this analysis would simply be rejected as completely wrong.

    • February 24, 2021 1:41 pm

      Given that there is neither empirical “proof” of the human component to global warming or indeed of the existence of the greenhouse effect as it is claimed one would think any other group would look in anouther direction. The problem is that in the IPCC founding documents they begin their experimentation with a conclusion, that man by releasing CO2 back to the Carbon Cycle is causing the problem. They have spent all these years trying to prove what basis physics ( and geological hisotry confirms) is impossible. The deliberate ignorign of the physics is a criminal act against science.

  8. Simon Derricutt permalink
    February 24, 2021 1:27 pm

    As far as I can tell, the assertion that only CO2 (and Methane, NOx, etc.) affects the downwelling radiation (basically, radiation from the ground level that is reflected back towards the ground) is wrong. The H2O in the air is ignored, and it shouldn’t be.

    The reason for this is the absorption length of IR radiation in the atmosphere, which is in the region of 20m or so – I need to find a plot of absorption length versus frequency to really pin this down. What this implies is that nearly all the reflection of that ground-emitted radiation will occur in the first few hundred metres from ground level.

    This effect can most-easily be experienced when driving in a fog – your headlamps get reflected back at you, and the lights of a car coming towards you can’t be seen until they are pretty close.

    Therefore, the stratospheric water and CO2 concentrations really don’t matter, and the reflection of that ground-emitted IR will be done by the total amount of radiative gas within the first few hundred metres of ground-level, and most of that will be due to the H2O. H2O concentration varies between around 20,000 and 40,000ppm, so about 50-100 times the CO2 concentration. There’s also only one band around 4.7 microns wavelength where the water absorption bands don’t overlap the CO2 absorption bands, and where the CO2 has any extra effect. Still, that effect will be of the order of 1% of the effect from H2O when it comes to reflecting energy, and it’ll be in saturation anyway and thus the actual change in “downwelling” radiation will be minimal and probably below the threshold of measurement.

    It thus seems logical that the concentration of CO2 in the air has no measurable effect on temperatures except in the few places where there is very little water vapour in the air. Even then the difference in the absorption length relative to the height of the atmosphere implies that any such effect will be minor.

    Going back in history over the instrumental temperature record (since 1659) also implies no dependence on CO2 concentration. If we go back further (using proxies for temperature and CO2 concentration, or ice-cores when available) then again we see that there’s no absolute correlation, though a rise in temperature preceded a rise in CO2 concentration, implying that the temperature changes caused the CO2 concentration changes (as would be predicted from Henry’s Law and the solution of CO2 in the ocean).

    I could do with feedback/discussion on this. This does seem to me to be a major failing of the physics in the climate models, in that only the CO2 is recognised as the main greenhouse gas whereas they should be looking at the total quantity of radiative molecules which includes the water vapour.

    • February 24, 2021 1:34 pm

      The “Left” harps on CO2 because it can be taxed — and that is the insidious gateway to the Paris Agreement — the Green New Deal — the New World Order — Communism.

  9. February 24, 2021 1:37 pm

    A “model” in this context is an experiment. A thinker or scientist formulates an idea. The way that idea is tested is by experimentation. When the experiment fails that proves the idea is WRONG, no matter how elegantly or complex the ideas forming the theory is crafted, it is WRONG WRONGF WRONG and should be discared. But what do we have? These clowns average WRONG models and then dare to sell us it as a doom laded prediction!

  10. Broadlands permalink
    February 24, 2021 1:51 pm

    The Earth’s natural variability (volcanic activity, jet streams, ENSO and other oscillations) are each inherently unpredictable. How, then, is it possible for any model to accurately predict the next 30 years…climate by definition?

    • Mike Jackson permalink
      February 24, 2021 3:56 pm

      “The climate system is a coupled non-linear chaotic system, and therefore the long-term prediction of future climate states is not possible.” (IPPC, AR3) They said it themselves. They are the experts. Why is no-one listening?

  11. devonblueboy permalink
    February 24, 2021 2:21 pm

    The GIGO principle applies as much to climate modelling as it does to Covid-19 mortality modelling

    • Mack permalink
      February 24, 2021 2:45 pm

      From a brief perusal of the mortality figures, It would seem that Covid 19 is another ‘magic molecule’ from the Co2 stable as it has, apparently, almost wiped out influenza and pneumonia as a primary cause of winter excess deaths in the elderly, frail and medically vulnerable.

  12. Richard Greene permalink
    February 24, 2021 2:31 pm

    They are climate computer games, not climate models.
    Models make accurate predictions, and are repeatedly refined to be even more accurate.
    These computer games make inaccurate predictions, and do not get more accurate over time. They do not deserve the name “climate models” — they don’t model the climate on our planet.

  13. Cheshire Red permalink
    February 24, 2021 2:48 pm

    Why can’t someone take government to court over a policy decision based on obviously wrong data?

    Likewise why can’t someone use the courts to force NASA / MET Office / HadCRUT etc to testify under oath?

    It’s as plain as day models don’t match observations.

  14. William Birch permalink
    February 25, 2021 11:39 am

    It is a very worrying trend in modern science when computer models are unable to be verified by robust scientific observations, that those in positions of power still insist in believing the computer models. We are entering some crazy, fantasy dystopian world where western world leaders have all agreed to take a series of draconian measures which manifestly will have a direct negative impact on the heath, well being and lifestyle of millions of their people but will be of no use whatsoever in dealing with the alleged problem..

    • Penda100 permalink
      February 25, 2021 1:35 pm

      Might this be the reason. H L Mencken said: “The whole aim of practical politics is to keep the populace alarmed (and hence clamorous to be led to safety) by an endless series of hobgoblins, most of them imaginary.”

Comments are closed.

%d bloggers like this: