A frank admission about the state of modeling by Dr. Gavin Schmidt

The task ahead is not to debunk several assumptions individual computational climate models may be based on, but to debunk the entire modelling paradigm as it is practiced in contemporary climate science.
What they are doing is to fit multiple, mutually inconsistent computational models of high Kolmogorov complexity to a single run of a unique physical instance (the “climate system”), which is far too large to fit into the lab, therefore can never be studied experimentally. Even observational evidence of the system’s unrepeatable history is patchy at best and for basic parameters (like “Planck weighted average thermal IR optical depth of the atmosphere” or “short wave planetary albedo”) are lacking.
This is why the concept of experimentum crucis can never be applied in this field of inquiry to incrementally eliminate models of poor performance and choose unequivocally between the few remaining competitors.
With enough free parameters to draw detailed representations of superheroes in comics, any such model can be fitted to any history, which means even if their basic assumptions are mutually exclusive, none of them can be falsified. This dire situation is supposed to be rectified by the slogan “All models are wrong, but some are useful”, with no operational definition of the concept “useful”, of course. Which makes choosing the useful ones arbitrary, to be based on preferences unrelated to the actual subject matter.
No wonder predictions of these models diverge widely. They would even more, if “useless” models were included, whose predictions utterly defy (political?) expectations.
The climate community responded to lack of falsifiability, mutual inconsistency & diverging predictions of models in a most curious way. They have invented the concept of “ensemble average”, which is supposed to “wash out” individual quirks of models and produce an outcome which, in a mysterious way, comes closer to reality than any member of said ensemble. However, we can see clearly, that ensemble average can only mirror the fuzzy concept of “useful” at best (preferences of “experts”), nothing else.
The other device introduced to fight uncertainty is the concept of “projection” (as opposed to prediction), which is justified by the existence of several different future “scenarios”, but which, in fact, is good for nothing but to kick unpredictability under the rug. In a sane world it could easily be replaced by genuine predictions over the field of all possible scenarios (emission trajectories & whatnot), but that would make them all too vulnerable to falsification.
The entire hullabaloo is justified by the phrase “there is no other way to make predictions in climate science than by modelling”, with the additional hidden assumption that with modelling it is indeed possible. But, as we have seen above, this hidden assumption is untenable. Which, if we suppose the the first proposition concerning models was true, would imply the unpredictability of climate. But that conclusion is never drawn.
The grave process described above has effectively slipped climate science down on a slippery slope to the level of a full fledged pseudoscience, where outcomes are not determined by nature, but wishful thinking and expectations.
________________________________________
Note all pseudosciences are organized around issues of utmost practical importance (health, fate, inexhaustible energy). On the other hand genuine science never even tries to be practical. The Earth with all its wonderful weather phenomena is something, but who’d care how optical depth determines the statistics of temperature distribution in a fluid, enclosed in a semitranspatent container, put onto a thermally well insulated rotating table in a pith black vacuum chamber whose walls are kept cool by liquid nitrogen wile irradiated by powerful lasers or such? No one but a bunch of weirdo scientists.
Still, that’s the only path that holds some chance to understand climate, because practical appliations usually come quite late in the game, if ever.
For genuine science does right the opposite climate modellers are trying to do. It seeks a single consistent model of low Kolmogorov complexity (a.k.a. “simple”), applicable to multiple runs of a wide class of physical entities, with certain members of said class being suitable to be studied experimentally in the lab.
With no general understanding of closed non-equilibrium quasi steady state thermodynamic systems with a vast number of nonlinearly coupled internal degrees of freedom, radiatively coupled to their environment (one of the many gaping holes in semi-classical physics) one has no hope to make any progress in bringing back climate science to the narrow path which may (or may not) lead to salvation.

Robin says:
June 1, 2013 at 8:54 am
“I just finished reading the UN’s Post-2015 plans for all of us so I wouldn’t get too excited that anyone is seeing the light. They are using climate change literally as the basis for remaking societies, economies, and new mindsets and values.”
There is currently a huge lovefest of ICLEI / Agenda 21 in Bonn, Germany, and you might not be surprised that all they talk about is how The Climate Change is the most serious problem for the development of cities (and that while the planet hasn’t warmed in 15 years)
You can’t make it up:
http://www.iclei.org/en/our-activities/our-agendas/resilient-city/resilient-cities-2013-live-blog/day-2.html

Unlike most here, I’m something of a fan of Gavin’s despite his CAGW promotion efforts. He is usually very precise in his use of language, as a good scientist should be. When Gavin says ‘assumption’, I think he means exactly that.
Assumption – something taken for granted
As opposed to a theoretical prediction.
Gavin is saying that something outside of the theoretical basis for the model, we haven’t really considered, is an important determinant of what happens. In the context of climate models, he is saying is, that important stuff, we don’t know much about, is missing from the climate models.
Which is a position many here could agree with.

The point that Anthony is making here is that Gavin is holding others to higher standards than he holds himself to, so Anthony’s entire post is right on the mark.

The problem with the Gavin’s like models is both mathematical and physical.
These models fail to properly reconstruct the climate oscillations at multiple scales such as the multidecadal-to-millennial ones. Thus, they are missing proper physics. The missing physics has to do with astronomical forcings of the climate and with the climate response to them because the missed oscillations are synchronous to astronomical oscillations related to the sun, to the moon and to the planets in general.
The models also fail mathematically because to recover the 20th century warming despite the missing physics, they have exaggerate the GHG effect by arbitrary parameterization of the internal feedbacks of the models.
All these things are clearly explained in my papers. For example in
Scafetta N., 2012. Testing an astronomically based decadal-scale empirical harmonic climate model versus the IPCC (2007) general circulation climate models. Journal of Atmospheric and Solar-Terrestrial Physics 80, 124-137.
http://www.sciencedirect.com/science/article/pii/S1364682611003385
and more recently in
Scafetta N., 2013. Discussion on common errors in analyzing sea level accelerations, solar trends and global warming. Pattern Recognition in Physics, 1, 37–57. (open access)
http://www.pattern-recogn-phys.net/1/37/2013/prp-1-37-2013.pdf
In particular, in the latter paper (in particular in section 2) there is a detailed discussion of the mathematical/physical errors made also by Gavin to exaggerate the man-made global warming.
There it is demonstrated that the GISS model underestimate the solar signature on climate by a very large factor both at the decadal and millennial scale.

Mike jarosz says:
June 1, 2013 at 1:23 pm
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.

You are assuming that they did not vote against the treaty because it would have taken away power from them.
As for volcanos, don’t make the same mistake as climate scientists . Do not assume that the last 100 years is a normal iteration in anyway NATURALLY to the state of affairs on a 4 billion year old planet that has cycles volcanically, weather, and in its climate system that we can not even dream of. In other words, you do not have enough data to make any assumptions about future volcanic eruptions. Any predictions based on 100 years of data are nothing but conjecture and in essence a blind-ass guess which is what climate scientists do. Please don’t scare people in other words on something you do not know anything about. Perhaps there will be an eruption in 2018. Perhaps CAGW will hit and prove us sceptics all wrong. Perhaps a unicorn will fart in China and cause a huge hurricane to hit the US. I have not studied any of these topics to give you a probability based on educated guesses, but I can tell you that the probability of having a volcanic eruption in 2018 is the same as it is for the year 2017 and just a tad lower than that of 2016 when leap year gives an extra day via our calender to make the event just that slightly more likely.
So if you are a betting man, I will bet on 2016 and take my chances with the extra day. If I was you, I would do the same thing if you are betting on the future.

The folks that spend their time studying ENSO statistical and dynamical models are having the same discussions. The old standby statistical models have, in my opinion, out-performed the newer dynamical models in predicting what the ENSO pattern will be. What is interesting is that there is a group of scientists who meet to discuss the whole enchalada and then make a “consensus” guess as to what the oceans and atmosphere will do. This week the consensus dipped below both the dynamical models (which always seem to favor boiling oceans), and the statistical models (which are never as El Nino-ish as the dynamical models are). Could it be that models have lost their shine in all climate circles?

Dere’s eben more smoke commin’ out da machine