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
cruzn246... What you say might be interesting if it were based on any facts.
Do you think for even two seconds that the many thousands of scientists who work on climate issues have not considered the relative forcing of solar vs GHG's?
Global Mean Radiative Forcings
The image from the link above:
cruzn... Please look again at the chart I linked to before. It clearly shows that there is a low LOSU (level of scientific understanding) related to solar. But that doesn't mean it could be absolutely anything. Still the likely range solar forcing is a fraction of GHG forcing, which has a high LOSU.
You do realize, of course, we have satellites which are measuring solar variation, and you were even quoting radiative forcing figures, data that comes from those satellites.
cruzn246:
"Tell me why it has essentially stopped warming just when the solar torch let up?"
See the SkS climate myth #2 "it's the sun". The first graph does not support your statement above. Also, what do you mean, specifically in @628 when you say "it has stopped warming". Specifically what do you mean when you say "it".
Finally, I think it is a strech to imply that the modelers and those that utilize them to aid in understanding climate change "think they have it all figured out". However, you sound like that. And in my opinon have much less of a leg to stand on in your assertions.
[DB] Please note that cruzn246 has an extensive history of sloganeering and trolling in this venue.
(-snip-).
[DB] Please comport yourself with this site's Comments Policy. In it you will note under the Sloganeering section that if you wish to differ with established science, you will need to bring reputable evidence to support that chosen differing. Mere assertion, as has been your wont, fails to rise to that burden of proof.
Sloganeering snipped.
cruzn... I think you'll actually find that people here are willing to entertain contrarian ideas, but you have to be able to substantiate your position far better than you're currently doing.
Just because you think something is obvious does not make you automatically correct. As the moderator is saying, you must back your claims with actual science and research.
Kishoreragi, to add to Michael's comment: that is the way general circulation modeling is done. The Earth itself is one big experiment, and it has been well-demonstrated that the minor variables vary within certain ranges and rarely, if ever, end up leading global or even regional climate by the nose long enough to significantly alter major elements of general circulation. There is a strong tendency to regress to the mean, and the mean is driven by the major elements of the climate: continental drift, orbital variation, vulcanism, solar output, collision with significant extra-terrestrial objects, and, now, artificial enhancement of the greenhouse effect. Everything else is a feedback: something that reacts but is unlikely to change on its own -- biosphere (with rare exceptions), snow/ice albedo, ocean carbon cycle, clouds, water vapor, natural greenhouse effect, etc. Feedbacks vary and are integrated in different ways, but they regress to the mean of net forcing, with minor temporal variations driven primarily by ocean thermal capacity.
Thus it's a little inaccurate to describe the climate system as comprised of components that vary but are all of equal or near-equal power in shaping the future of the system. And, in addition to the major forcings, modeling does take into account many of the major and minor feedbacks. Of course, in the short-term (months to years), the interplay of major and minor feedbacks can produce significant but temporary anomalies, but the resonance of those anomalies across the long-term trend is ultimately insignificant. Arctic sea ice (ASI) is a great example. ASI is never, with all forcings stabilized, going to vary strongly and consistently to the extent that a glacial cycle is initiated. Only orbital or solar variation (or a one-timer) can do that (partially through the mechanism of snow/ice albedo feedback).
Given that climate models can closely approximate past climate, I'd think the onus is on 'skeptics' to show some reason why we should assume that they will not be equally accurate for future climate.
Just saying, 'there are too many factors involved to ever model climate' doesn't cut it given the established reality of models which already do successfully match past climate. Heck, we even know most of the causes of short term variation... such that if you plug major volcanic eruptions, variations in solar output, ocean cycles, and other such 'unpredictable' factors into model runs of past climate they then match not only the long term trends, but even the short term fluctuations around the trends. That's shockingly accurate for something which is supposedly 'impossible'.
kishoreragi at 17:00 PM on 18 September 2013 9 (found in What's causing global warming? Look for the fingerprints)
Your “a forest is like a climate model” analogy would benefit if you could identify the critical variables omitted from the models or issues with how climate models utilize them.
Yes, there are a lot of different animals moving around in a forest. Yes, modeling everything going on in a forest would require understanding how sensitive each thing is to changes in each other thing. But it’s not a valid analogy to just list some of the different things going on in a forest, and then jumping straight to the insinuation that climate models omit critical features. In short, you don’t provide any solid rational that climate models are not useful with respect to predicting changes in climate.
[JH] Thank you for responding to kishoreragi on this thread.
Roger D
I don't want to go predict big things right away with incomplete models, but I want to eloborate what I wanted to inform here.
I have given this analogy because there is a need to understand each and every animal (here it is physical process) is related (friendly/hostile, here for climate, how processes are intricately mixed- diminishing/strengthening) on forest variables (We can choose any variable here) on specific part of forest(Any region of the world). So that little by little, we can understand about the comple forest and their inter-relation.
My main point is that the science has progressed much further, but in wrong path. The simple fix for this is to make the system simple and see the intricacies among the processes, leaving the comparision with observations, on regional climate variables so that in the FAR future, we may be in a position to see the BIG picture like AGW without hesitation from anybody (skeptics/supporters) and with clear understanding. As far as I understand(ofcourse I am still reseach student), there is no other way as climate system super complex !!!
kishoreragi, you only need to model every conceivable input if you need to know every conceivable output... which isn't the case. To take your forest example, if the goal of the model is to determine how the forest will grow then the actions of deer and bears are largely irrelevant... they might impact a tree here and there, but they are not going to change the overall growth pattern of the forest. Instead, you are going to look at weather, human logging, beavers, and other factors which can actually have a significant impact.
Ditto climate models. No, they cannot possibly model every individual cloud and gust of wind... but there is absolutely no reason they would need to, because those things are not going to impact the overall climate trends.
Again, you do understand that climate models already work, right? They can successfully model the past and even the relatively primitive climate models in use 30 years ago produced results consistent with the past 30 years. You are arguing that something which has been done, cannot be done. You're wrong before you even get started.
Thank you all for reading my views and their valuable comments/suggestions.
CBDunkerson, thank you for your valuable comment on my analogy, and insights into climate modeling. I understand that climate models work. I have been trying to study precipitions but, I felt, with supervisor's advise, like I was cheating science because model resolution is coarse and cloud physics is not yet well understood. Hence, I have changed my thought to study other variables. May be climate models worked for few variables even before 30 years but, climate is not of those few and rest of them won't work even 21st century. May be you are looking at those few variables(I agree with you here) but, I always see other side with completely different view to progress further.
I will try to correct myself (that is why I have been focusing on literature from all corners) before get started my research as well begun is half done !!! I am off on this thread. Thank you all !!!
kishoreragi - I think what you are overlooking is that there is an approriate level of detail for studying anything. Details below the scale of GCM's are parameterized, treated as blocks that have known (as in, tied to observations) responses to inputs, and that physically based parameterization works just fine for global and regional level modeling.
More detail would be needed if you wanted to look at microclimates and the chances of a particular bush getting wet during a closely timed rain shower. But that's not the level of study for GCM's, and if subscale responses are reproduced well a GCM will give a fine answer at the scale it is actually studying.
These models aren't looking at the level of individual trees and gusts - hence they just don't need to simulate at that level to get a good answer for the regional/global scales studied.
I just saw "Overestimated global warming over the past 20 years", by Fyfe, Gillett and Zwiers in Nature Climate Change, Vol 3, Sept 2013, page 767. They contend that the divergence of models and observation is statistically significant. In trying to rationalize the discrepancy it seems to me that they don't consider the possibility that atmosphere-ocean heat transfer coupling methodology (particularly below several hundred meters depth) in the models may not yet be up to par.
What is the SkS take on this article?
tcflood's source in pdf, in case SkS can't access it by other means.
@tcflood it looks like there was indeed less warming over the period 1997-2012 than the models predicted, however, what "skeptics" often fail to mention is that "the surface warming trend from 1993 to 2007 was significantly higher than projections." See here for more information: http://www.abc.net.au/environment/articles/2013/09/27/3857704.htm
@jsmith. Thanks for the response. I read somewhere that many of the GCMs actually do produce ENSO episodes but that the timing of their appearance (as in nature) is chaotic. If modeling results are reported as an average of many models, wouldn't the effect be to remove the ENSO effects thereby insuring some departures of observation from the models when real ENSOs would occur? So, hypothetically, if there were some way to trigger an episode at the same time in all the models, the match between models and observation might be much improved. I don't pose this as a realistic modeling strategy, but rather as an accurate and effective counterargument to this common cavil in the denier echo chamber?
Also, the denier rant that called my attention to the above-cited paper said Zwiers is a vice chair of the IPCC. I infer from your comment that he is nonetheless a skeptic. The paper certainly reads like he is.
tcflood "wouldn't the effect be to remove the ENSO effects thereby insuring some departures of observation from the models when real ENSOs would occur? "
Essentially true. See for instance the graph here. Forcing ENSO in model to match what happens is somewhat like Kosaka and Xie did. See discussion here.
Comment from Mike at RC on Zwiers.
"[Response: Francis is a top-rate scientist of the highest integrity. I strongly suspect that he has been misquoted and mischaracterized quite a bit lately. -mike]
Further to Zwiers. Any good scientist is a skeptic - a real one. Fake skeptics are only skeptical about things they disagree with and will swallow any kind of junk uncritically if it supports their notions.
"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."
So if we look at the past 30 years we should get the climate trend. According to skepticalscience's trend calculator from 1983 to 2013 the earth's temp has increased by 0.17 C per decade (GISTEMP moving average 12 months). Since CO2 has risen at about a linear rate since 1980 (according to: http://www.esrl.noaa.gov/gmd/ccgg/trends/), is it reasonable to predict a best guess rise of 1.7C from decades (1990 - 2000) to (2090 - 2100)? This would point to the low scenario from IPCC 2007, which has a best guess of 1.8C.
Engineer, the issue is estimating climate sensitivity is non-linear feedbacks. Decreasing ice and increasing methane emissions would be two examples. Models attempt to build these influences in whereas linear extrapolation does not.
@ Scaddenp,
Of course, but I thought climate sensitivity can be considered constant for small changes in temp. ∆T = k * 5.35*ln (C/C0), where k is climate sensitivity. I graphed 5.35*ln(C/C0) and it looks approximately linear in the range for C: (275 ppm, 550 ppm) and C0=275ppm. Is climate sensitivity not supposed to be considered constant even for small changes in temp?Thanks.
But you would only expect temperature rise to be linear if transient climate sensitivity was same as equilibrium climate sensitivity. Emperical determinations of ECS suggest it is reasonably robust over quite a temperature range. However this is an observation, not an assumption.
You would also need to separate out the other forcings at play if trying to establish k by simple fits. I would recommend Chpter 10 of the newly out AR5 WG1 which has a section on estimation of TCS and ECS, along with references to the papers which attempt this from various observational sets. The section of estimation from the instrumental record looks like it would interest you most.
I have a friend of a friend who is a physicist. He is ~not~ an AGW denier; he does however have objections to models as he claims they are used in climate science. He fears that, to quote him:
I suggested he post this question himself, but since he didn't feel it worth his time (which, I pointed out, is maybe why, though ignorant myself, I shouldn't take his objection too seriously), I told him I'd post it and see what people had to say. If this question doesn't explain his position clearly enough, I have some longer expositions of it I can post as well.