This is something I never expected to see in print. Climate modeler Dr. Gavin Schmidt of NASA GISS comments on the failure of models to match real world observations.
Source:
[ http://twitter.com/ClimateOfGavin/status/340605947883962368 ]
While the discussion was about social models, it is also germane to climate modeling since they too don’t match real world observations. Below is an example of climate models -vs- the real world; something’s clearly not right.
Graph source: IPCC AR5 draft
Is it maths or assumptions (or both) that cause the divergence?
UPDATE: In comments, I had a discussion with reader “jfk” which I think is worth sharing. He made some good points, and it helped hone my own thinking on the issue:
jfk says: Submitted on 2013/06/01 at 8:40 am
Well, I still think it’s a bit unfair to Gavin (and I am no fan of his). But hey, it’s Anthony’s site.
For a good review of the many failures of statistical modeling in social sciences (and one or two successes) see the book “Statistical Models: Theory and Practice” by David Freedman. Whether or not climate modeling has devolved to the point where it is social science rather than physics, well, I hope it’s not quite that bad…
REPLY: And I think it is more than a bit unfair to us, that if he believes what he tweets, he should re-examine his own assumptions about climate modeling. We have economies, taxes, livelihood, etc. hinging (or perhaps failing) on the success of these models to predict the climate in the future. The models aren’t working, and Dr. Schmidt knows this. Unfortunately his job is tied to the idea that they do in fact work. I feel no regrets at making this comparison front and center. – Anthony
UPDATE2: RussR in comments, provides this graph below showing Hansen’s modeled scenarios against real world observations. He writes:
Here’s an excel spreadsheet comparing observed temperatures vs. model projection from: Hansen (1988), IPCC FAR (1990), IPCC SAR (1995) and IPCC TAR (2001), in pretty charts.
It can be updated as more observations are added.
https://dl.dropboxusercontent.com/u/78507292/Climate%20Models.xlsx
UPDATE3: Dr. Roger Pielke Sr. adds this in comments.
Climate models are engineering code with quite a few tunable parameters, and fitting functions in their parameterization of clouds, precipitation, land-atmospheric interfacial fluxes, long- and short-wave radiative flux divergences, etc. Only a part of these models are basic physics representations – the pressure gradient force, advection, the Coriolis effect.
The tunable parameters and fitting functions are developed by adjustment from real world data and a higher resolution models (which themselves are engineering code), but only for a quite small subset of real world conditions.
I discuss this issue in depth in my book
Pielke Sr, R.A., 2013: Mesoscale meteorological modeling. 3rd Edition, Academic Press, in press. http://www.amazon.com/Mesoscale-Meteorological-Modeling-International-Geophysics/dp/0123852374/ref=sr_1_2?ie=UTF8&qid=1370191013&sr=8-2&keywords=mesoscale+meteorological+modeling
The multi-decadal global climate model projections, when run in a hindcast mode for the last several decades are showing very substantial errors, as I summarize in the article
http://pielkeclimatesci.files.wordpress.com/2013/05/b-18preface.pdf
I sent N. N. Taleb, “Your twitter with Dr. Gavin Schmidt has become topic @ur momisugly Watts Up With That?” That he may speak for himself.
Ryan says: “… if you read the comments here it is very clear that your readers think he was talking about climate.”
Buzzzz, wrong. Anthony was clear as to what Gavin was Tweeting about, and readers here understood Gavin was referencing social models.
Roger–I think the official response to that awful tragedy in Woolwich tells us that reality that differs from the assumptions needed to justify the UN’s restructuring simply get ignored. With the media going largely along.
I am going to explain the media aspect by the way soon. The UN really does want the media to be its partner in selling the desired beliefs and new values. Basically education creates the desired new mindsets with few accurate facts and lots of erroneous supplied concepts that filter daily perceptions in predictable ways. Then media reenforces by what it covers and does not. Of course I had to track to servers near the ends of the Earth to get those particular blueprints from 2010. But I have them now.
The illusion of catastrophic climate change is front and center as the selling point to bring in what Irina Bokova calls the New Humanism. And it bears a striking resemblance to Uncle Karl’s human development model. And his small c social and economic model.
And I wish I was kidding or being a tad hyperbolic.
Well I don’t think it matters a hill of beans, whether Dr Schmidt was discussing social “science” (there’s a laffer ), of finance “science” (another one) or climate science.
Math is almost never wrong. After all, we made it all up ourselves, and defined how it was to be used. Well you can simply fail to add correctly, and you could use the wrong maths, but so long as you use the wrong maths correctly it will give the answers you would expect to get when you use the wrong math correctly.
But clearly the problem is that the model assumptions do not mimic reality.
If the sun beats down on a clear sky earth, at about 1kW/m^2 normal to the sun vector; but in your model, you use 240 W/m^2, or some other made up fictional number, that absolutely nobody, ever actually measured, anywhere on earth; well wake me up if you get the correct result, that matches reality.
A tornado with 200 mph winds can do a lot of damage, but if you use 50 mph, instead of the measured 200 mph, you won’t get much damage at all in your model.
So nyet Gavin; nothing wrong with maths, but maybe wrong maths; but first you have to model what is real.
@jfk says:
June 1, 2013 at 7:44 am
Wait a minute, is Gavin talking about climate models or social science models? Let’s try not to put words in his mouth that he didn’t say:
++++++++++++
Wrong: He is talking about climate models. A quick search will take you to his NASA page here:
http://www.giss.nasa.gov/staff/gschmidt/
He writes:
“Research Interests
My main research interest lies in understanding the variability of the climate, both its internal variability and the response to external forcing…”
If this is about his main research, that is what he is talking about. Don’t be confused by words and their meaning. There comes a point where human beings have to be able to judge words into a context. He was talking about climate models, not models in general.
Perfect maths plus bad assumpsions equals BS is not enough. If one is using computing to evaluate chaotic systems, then even if the maths is perfect the prediction will be wrong because computers truncate during calculations, hence they push the time evolution onto a false track at each step and the result will be more and more wrong as time passes. Hence any attempt at say regional predictions will be wrong. Frequency of events will be unpredictable. Threats of flood, drought, hot, cold for a region will be wholly unjustifiable in longer term predictions. Just try and predict how the jetstream will behave and hence the climate of Europe..
Robin says: “No because the purpose of theories and models in the social sciences is not to reflect reality but to change it.”
Almost same purpose behind theories and models of Global Warming.
“if your prediction is wrong then your hypothesis is wrong. Period.” – Richard Feynman
“No theory is carved in stone. Science is merciless when it comes to testing all theories over and over, at any time, in any place. Unlike religion or politics, science is ultimately decided by experiments, done repeatedly in every form. There are no sacred cows. In science, 100 authorities count for nothing. Experiment counts for everything.” – Michio Kaku, a professor of theoretical physics at City College of New York
“If your experiment needs statistics, you ought to have done a better experiment.” – Ernest Rutherford
http://pathstoknowledge.net/think/
Consensus starts to fade, understanding blossoms.
Jimbo says:
June 1, 2013 at 12:57 pm
———————————–
Not buying that higher CO2 played much role in seesawing climate during the termination events, but stomatal data themselves are revelatory.
Thanks!
The model with 0 GHG growth is the same as real world observations. Then the GHGs they are counting are not the GHGs that matter. Perhaps water vapor (effectively a constant) is the only GHG that matters, from which we easily infer CO2 is of negligible importance to climate. Except CO2 is vital to keep plants alive, having the primary role of stabilizing the atmosphere. And for that purpose, more is better.
Where is Mann? So much debunking of global warming this past week and I haven’t seen any tweets from him. Is he on vacation?
jim2 says:
Dr. Schmidt is stating the obvious and frankly I am aghast that Taleb would make such a obviously idiotic statement. Math is exact, but as Gavin points out, essentially, garbage in –> garbage out.
Dear Jim2:
Having read Nassims fine popular publications (none of his formal papers)…I think that I can defend what he says. Let’s put this way. Do you know what a Fourier Series is? Do you know what a SQUARE WAVE is…..? How MANY terms in a Fourier Series does it take to PERFECTLY match and create a SQUARE WAVE? Answer: An infinite number.
But, if you want an approximation, you can use a “few” terms. Now in the INFINITE DIMENSIONAL situation of doing a climate Model (and that is what it is, essentially an “infinite” number of variables…) how does one determine one has “summed enough terms” to have a “useful” construct? That really is a good question to ask. Schmidt is inherently IMPLYING that there exists some sort of “perfect” model. And if they just get the “assumptions” right, they can have perfect modeling.
AND THAT CONCEPT is what Nassim Taleb is denigrating. More power to him. (In fact, I’d recommend 10^78, or there about…the number of atoms estimated in the universe. Give or take a few 10^50 assemblies to 10^50…)
Analogy with things we study in social issues? How about that wonderful word ‘inoperative’ and its rich history? Or the streaker’s defence – it seemed like a good idea at the time. Or LBJ’s Theorem which here becomes ‘it’s better for politicians to have the greens inside the tent p***ing out than conversely’.
What is the mystery here? Correct me if I’m wrong, but don’t all these models have a dependent functional relationship between increasing CO2 causing increasing atmospheric temperature?
So of course there would be an increasing air temperature regardless… they were depending on the broken clock to be correct for the next 25 years. No?
Max Hugoson says:
June 1, 2013 at 2:28 pm
Max, it worse than that… they have to get ALL THE BOUNDARY conditions right too. Which in reality must be matched by the direction of change in each boundary variable for the first step of the first calculation. And as we all know they are also trying to model a chaotic & stochastic system. Aaaaahhhhhhhhhhhhh!
The models and assumptions worked fine, until they hit reality. Once they came out from the shadows and into the sunlight they become garbage. Sunlight is a wondeful disinfectant, especially if your propaganda and ideology is based on science and that science is false. Tricks with numbers won’t get you far either. If they stay naked in the sun too long there are going to be some pretty bad burns, they had better scurry back to the shadows or cover up with reality.
Mike jarosz (June 1, 2013 at 1:23 pm) wrote:
Nothing happens in the U.S. as far as treaties without 2/3 of the U.S Senate approval. Last vote in 1997 was 95 to 0 against destroying the U.S. economy to save the planet. They may be stupid liberals, but they’re not crazy.
No treaties needed when the Executive is allowed to rule by bureaucratic fiat (i.e. the EPA).
goldminor (June 1, 2013 at 2:21 pm) suggested:
“Consensus starts to fade, understanding blossoms.”
We’ll see — ( not holding my breath … )
___
Greg Goodman (June 1, 2013 at 12:41 pm) wrote:
“[…] 1883 1902 1912 1963 1982 1991 […]
[…] most of those dates seem to fall on a 9 year grid […]”
Are you going to keep ignoring the role of the sometimes-synchronized ~11 year grid?
Hint: Do a 1/4-cycle gaussian smooth of VEI sorted by solar cycle phase.
Then behold: the simple ellipse.
Dead simple, but beyond the reach of mainstream modeling “science” (which appears intent on drowning society & civilization in a cesspool of darkly ignorant &/or deceptive assumptions).
The models and assumptions worked fine, until they hit reality. Once the sunlight was let in it revealed garbage science, it is an excellent disinfectant. People are going to get badly burnt and they can either scurry back to the shadows or cover up with some reality. It will be interesting who chooses to do neither and stay in the sunlight naked.
well let s admit math are correct. there are assumptions, but there are uncertainties as well .how can error be estimated after thousands of calcultations?
well there are assumptions but in case of modelisation you have mathematical uncertainties as well,,the point is ..you don’t know them.
What I meant to add to my last post is that there will be three types of climate scientists, I think Mann will scurry, Dr Scmidt looks like a scurrier but I think will in the end cover up and Dana and Cook are going to stay naked and burn.
Have a look at the green line on the NASA et al projection. It assumes no CO2 increase after 2000 and it shows no temperature rise after 2000. That means it is saying there is no long time constant, no locked in inevitable further catastrophic rise, any response to rising CO2 is immediate. Even if all the temperature rise of 0.6C is due to rising CO2 (exceptionally unlikely given the steady temperature despite rising CO2 for the last 15 years) a projection of 560ppm by 2100 would only give a further 0.6C rise!!! This plot by itself is enough to debunk the CAGW argument.
Max Hugoson says: June 1, 2013 at 2:28 pm “AND THAT CONCEPT is what Nassim Taleb is denigrating. More power to him. (In fact, I’d recommend 10^78, or there about…the number of atoms estimated in the universe. Give or take a few 10^50 assemblies to 10^50…)”
Archimedes’ Number (The Sand Reckoner) and Eddington’s Number is 10^80 nucleons and is related to Dirac’s LNH
Reminds me of this old joke on the shortcomings of mathematics:
Problem: If a girl of 16 can go in the forest and pick 4 pounds of berries in 5 hours and a boy of 17 can pick 5 pounds in 6 hours, how many are they going to pick if they work together for 7 hours?
Answer: probably not very many.