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."
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.
Last updated on 30 May 2024 by John Mason. View Archives
dvaytw It sounds to me as if your firnd may be thinking of "flux corrections" but modern climate models have improved to the point where they are no longer necessary. Sadly unless he can give a concrete example, it is difficult to address his concerns more directly. The modellers do however perform experiments to determine the sensitivity to the parameters, which I understand are called "perturbed physics experiments". IIRC there are on-line repositories that contain the results of such experiments resulting from a sort of SETI@home type project where people can use their spare processor time to run climate models (I'll see if I can find it again).
This gets raised at RC from time to time. You might like to look at the response to this comment. Sounds like an objection based on dated information and overstating the problem.
Also note that the "flux corrections" discussed by Dikran are specific to models that incorporate simulations of both atmospheric and oceanic circulations (note the AOGCM acronym in the wikipedia page Dikran references - AO means Atmosphere/Ocean General Circulatation Model). Such flux corrections do not exist in any other categories of climate models, and they are not in all AOGCMs.
Simulating both atmospheric and oceanic circulation at the same time is a complex problem: you need small time steps to deal with the rapid atmospheric changes, but you need long simulation times to deal with the slower motions and changes in the ocean. Various tricks (oh, there is that awful word again) are used to deal with the mathematical complexities.
Thanks guys. My friend now makes the following objection in defense of his refusal to actually simply appear in the comments section and ask his questions himself:
It strikes me that I've never paid too much attention to credentials here, but would y'all respond to this?
dvaytw @654.
Your physicist friend-of-a-friend asks an exceedingly narrow question (presented @650) given he is describing his reasons for entirely dismissing climate modelling. And now he impunes SkS blog-writers & blogs and accuses commentors of wielding Okhams broom.
All in all, folk who speak with such 'authority' do not do so in such a dismissive way if they are genuine. (The apparent incoherence is not untypical of a 'technophile' but not entirely a good sign.) So I fear that that broom is in the hands of you physicist
You could ask him to (or may be you could yourself) describe the "many of the points Tom and (he) have been stating (t)here (that) have been brought up, and if you watch closely, ... get brushed aside, as if paying too much attention to them would spoil the game." It would allow all to get the measure of each other without his personal appearance here at SkS which he is so afeared to do.
dvaytw's friend wrote "What I am saying is that your belief in climate models is based on accounts written by science journalist, none of whom actually do the research themselves. In fact, looking through Skeptical Science, I am left wondering if the writers of the blogs actually can understand the original studies they are reporting on."
Actually, I have worked on statistical downscaling of climate models, which is a task that requires a reasonable working knowledge of the operation, mathematical and statistical principles involved. The paper was published in the international journal of climatology here:
M. R. Haylock, G. C. Cawley, C. Harpham, R. Wilby and C. M. Goodess, Downscaling heavy precipitation over the United Kingdom: A comparison of dynamical and statistical methods and their future scenarios, International Journal of Climatology, volume 26, issue 10, pp. 1397-1415, August 2006. (doi)
However, given he starts his question with an ad-hominem and just continues it, and adds little detail of the scientific question he wanted to ask, I rather doubt that any answer will change his opinion.
Credentials are irrelevant in science, what matters is the strength of the evidence and the internal consistency of the argument. Now if your friend is really concerned about credentials, ask him why he doesn't accept the views of the IPCC on models, as the authors of the IPCC report have as good a set of credentials as you could possibly want to see. I suspect the reply will be another ad-hominem, but do ask him.
MA Rodger & Dikran Marsupial:
It is possible that dvaytw's "friend" is a ruse by dvaytw to get around the SkS Comments Policy. You should keep this possibility in mind when responding to dvaytw's "friend." I and other Moderators need to keep this possibility in mind when reviewing dvaytw's posts.
JH, surely nobody would stoop to that sort of behaviour on climate blogs? Perish the thought! ;o)
John Hartz - I would strongly disagree based on dvaytw's presentation of SkS information on other forums. He indeed is arguing the science with support from discussions here.
dvaytw - I would strongly suggest including such links (if possible) to these various discussions whenever referring questions from deniers? As in, every time? Not only does that make your position clearer, but provides context that might greatly aid answering such questions. In fact, that might cut short a pattern I've seen with your questions from several back-and-forths to a briefer conversation that more directly addresses the denier myths involved.
As it is, your questions feel like a game of Telephone...
[JH] Thanks for the feedback re dvaytw's credentials. Your advice to him is also appreciated.
Moderators - My apologies, the link in my previous post should be "dvaytw's presentation of SkS information on other forums".
Echoing KR@659, I too have occasioned upon dvaytw engaging the denialista. My only criticism would be his (once) suggesting that I was an "actual climate scientist" which I am not. Then as @565 "Credentials are irrelevant in science, what matters is the strength of the evidence and the internal consistency of the argument. "
Could anyone explain why Antarctic climate models would be harder to get right than full global climate models?
[DB] Please support your assertion with a link to a credible source establishing that claim.
Climate models predict more snowfall than ice melting during the next 50 years, but models are not good enough for them to be confident about the prediction.
The British Antarctic survey team confirm this but their site is down at the moment
[TD] Part of the answer is that the smaller the geographic region, the more difficult it is to project its climate. The whole globe is easier than any subregion of the globe. That's because difficult-to-predict factors regionally, tend to cancel out those same factors in other regions. This difficulty of projecting at short geographic scales is similar to the difficulty of projecting at short time scales--even when the "region" is the whole globe. See the post The Difference Between Weather and Climate. See also the National Academy of Sciences' excellent series of short videos, Climate Modeling 101.
For a given size of region, naturally some regions are more difficult to project than others are, but I don't know about the difficulty of Antarctica versus other regions of similar size.
First of all, copying the work of others (as you do) without the usage of quotes is considered plagiarism, FYI. The only remaining question is whether you copied it from Wiki or one of the many denier sites parroting that phrase.
Further, I see no mention in the BAS site of the phrase you use. Feel free to look yourself.
Daniel, for what it is worth, wikipedia's (and hence Vonnegut's) claim is based on the following statement by the BAS:
The only problem is that the statement comes from a page that was taken down by the BAS sometime between Feb 7th, 2006 and June 7th, 2007. Ergo the statement precedes the IPCC AR4, let alone AR5.
The nearest recent equivalent (from the page you are currently redirected to) reads:
(Current to at least Jan 17, 2014)
Of course, that does not support Vonnegut's claim.
Vonnegut @666:
1) Antarctica does not have a simple climate, not even in relative terms.
2) Quoting a 2004 interview and assuming comments made regarding models in 2004 are relevant now represents a specious argument. If you cannot find a relevant modern quote, your Find a relevant modern quote or your questions are without basis.
A problem I have with this entry supporting climate models is that it seems to skirt what is most critical of them. Most current climate models completely missed the current warming pause. This is an issue that has spawned a host of investigation. A recent article about stronger trade winds in the media for example. Also, it also does not address that ALL models of complex systems mispredict to some extent. Given climate models are long term projections of where we are going, it is reasonable to expect that they will miss shorter term climate movements. From my point of view, it is more indicative that climate models seem to still say we should be warming even after the past decade or so of flat atmospheric temperatures is taken into account. Those who would deny that CO2 causes ANY warming or that its effect is small and not significant may point to the past decade as proof of future results. I'd love to sell them a bridge in NY for that risky logic.
knaugle "Most current climate models completely missed the current warming pause."
This is actually completely unsurprising, however to see why this is, you need to understand how GCMs operate, which in trun will show what they can reasonably expect to be able to predict and what they can't.
Imagine we had a quantum mirror which we could use to visit Earths in parallel universes. Say we could choose only those where the climate forcings (e.g. solar, volcanic, aerosols, CO2 etc.) are exactly the same as they are on our Earth, but the initial conditions are slightly different (e.g. a butterfly flapped its wings on one, but decided not to bother in another). In this case, the response to the forcings will be exactly the same on the parallel Earths, but each will have a different set of variations in climate due to sources of "internal variability" such as ENSO.
Now you could have no better climate model than this, even in theory, as the parallel Earths have exactly the same laws of physics and infinite temporal and spatial resolution. Would they have predicted the "warming pause"? Well yes and no. The warming pause is likely the effect of sources of internal variability, in fact such sources of variability are completely sufficient as an explanation of the lack of warming (e.g. Foster and Rahmstorf). These are chaotic, and while these periods of little or no warming will be observed occasionally on the parallel Earths, they won't generally ocurr at the same time as the pause on this Earth.
So what can we predict from the model? Well if you take the average of the temperatures of the Parallel Earths, you will get an estimate of the forced response (i.e. the reaction of the climate to the change in the forcings, e.g. CO2). This is what is relevant to climate policy, not the effects of internal variability which are quasi-cyclical and have little long term effect on the climate. The forced response does not show pauses, but that doesn't mean the models are not predicting that pauses will happen everynow and again (but without being able to specify when)
Now, lets return to real climate models, Easterling and Wehner show that these pauses are also found in the output of individual model runs as well, but again the timing of the pauses is unpredictable. So it is unfair to say that the modells have missed the pause. They have said we should expect them to happen, but can't predict when they will happen.
Why should we expect real climate models to be able to predict something that a theoretically perfect model could not?
A recent article in the WSJ by John Christy and Richard McNider (I'm sure many of you know about it) claims that the models predicted more warming than was actually observed. I'll leave it here so that other contributors can explain why it is either misleading or flat-out wrong. The image I am referring to is entitled "Warming Predictions vs. the Real World". I would just add the image, but I can't figure out how, so here's the article.
[RH] Hotlinked url.
jsmith - Christy and McNider are using the same "on-year baseline" trick that they have been using for quite a while. This is a method used to make the difference between the models and the observations look bigger than it actually is, for details see dana's recent blog post at the Guardian (don't be put off by the title - the stuff about Christy and McNider is the second half of the post after the stuff anout Roy Spencer's unfortunate meltdown). Of course this won't fool anybody that understands how the models work and what the ensemble method does, but it does make great fodder for the media and "skeptic" blogs.
jsmith, they are essentially using the very same trick I already explained to you here.
jsmith, I think it's the same tricks as described on HotWhopper i.e. they picked a nice big spike in the observation data to "align" the models starting point with. My understanding is that models should be started at a multi-year average temperature - not a particular cherry picked start year e.g. 1979 in this case.
sapient fridge - the start date isn't cherry picked in this case as 1979 is the start of the satelite record, ut the use of a single year baseline is still incorrect for the reasons I demonstrated to jsmith on the previous thread.
I would be shocked if I was the first to bring this up, but a recent article in Nature Climate Change contends that the models predicted more global warming (as defined by global surface temperatures only) than actually happened. To wit: "Global mean surface temperature over the past 20 years (1993–2012) rose at a rate of 0.14 ± 0.06 °C per decade (95% confidence interval)1. This rate of warming is significantly slower than that simulated by the climate models participating in Phase 5 of the Coupled Model Intercomparison Project (CMIP5)." Does this in any way validate those who state that the models can't be trusted?