by Judith Curry
[W]e presently find little evidence of trustworthy predictions at fine spatial scale and annual to decadal timescale from climate models. – Hargreaves and Annan
A new article in WIREs Climate Change:
Can we trust climate models?
JC Hargreaves and JD Annan
Abstract. What are the predictions of climate models, should we believe them, and are they falsifiable? Probably the most iconic and influential result arising from climate models is the prediction that, dependent on the rate of increase of CO2 emissions, global and annual mean temperature will rise by around 2–4∘C over the 21st century. We argue that this result is indeed credible, as are the supplementary predictions that the land will on average warm by around 50% more than the oceans, high latitudes more than the tropics, and that the hydrological cycle will generally intensify. Beyond these and similar broad statements, however, we presently find little evidence of trustworthy predictions at fine spatial scale and annual to decadal timescale from climate models.
WIREs Clim Change 2014. doi: 10.1002/wcc.288 Available online [here].
In the first two pages, they pose the following questions:
What are the predictions of climate models, should we believe them, and are they falsifiable?
Are the models sufficiently wrong that we should anticipate reality falling outside the range of model results?
On the agreement among climate models:
It cannot be argued on a rigorous basis that climate model agreement necessarily implies correctness, largely because of the ad hoc origins of the ensemble members, and unclear characterization of their inter-relationships. Models have shared code and ideas according to their origins,10 and therefore rather than considering them as independent sources of evidence concerning the climate system, it may be more realistic (albeit still perhaps optimistic) to interpret the ensemble collectively as representing (at least approximately) our range of beliefs and uncertainties regarding the behavior of the climate system.
On the falsifiability of climate models:
Nevertheless, one could argue that some combination of social pressure, and convenience, results in models sharing too much theory and even code, to the extent that they are little more than replicates. The validity of these competing arguments can hardly be decided on the basis of rhetoric, but there is yet little progress on how they can be assessed by analysis of the ensemble or other methods. Thus we consider this to be a particularly important area for future research. If it is the case that the pressure to conform in climate science has led to a serious disruption of the scientific process, then attention should indeed be focused toward building alternative models based on fundamentally contrasting physical hypotheses that perform equally well or better for the modern and past climates. However, proponents of such an exercise should note that while encouraging diversity in model design seems a laudable goal, alternative ideas cannot easily be conjured up from nothing, but are instead typically provoked by failures of the existing paradigm
Despite understanding the basic processes underlying the physics of the climate system, it is clear that the state-of-the-art climate models are not ‘good enough’, if we desire high resolution predictions with high temporal and spatial resolutions over coming decades. Thus far we seem to have only built sufficient confidence in the broad scale response of temperature and precipitation. The large-scale understanding of the physics seems to be sufficient, but the details are either not well understood, or are not being sufficiently well approximated by the model code. Given the spread of model results at the local scale, the issue is not so much one of falsification, but rather that current models do not provide much of a guide as to future climate change. Research to address this deficit in the models is required in order for the models to become truly trustworthy, but it is not clear when, if ever, this will be achieved.
JC comments
For background, a number of posts at Climate Etc on climate models are found [here]; my 2010 post What can we learn from climate models? provides the best introduction to my take on this issue.
Hargreaves and Annan’s article is an Opinion piece, not a scientific article; as such it is a relatively superficial treatment of the subject. In any event, I’m pleased to see their article published. The main thing that I disagree with is the statement in their abstract:
Probably the most iconic and influential result arising from climate models is the prediction that, dependent on the rate of increase of CO2 emissions, global and annual mean temperature will rise by around 2–4∘C over the 21st century. We argue that this result is indeed credible, as are the supplementary predictions that the land will on average warm by around 50% more than the oceans, high latitudes more than the tropics, and that the hydrological cycle will generally intensify.
They do not justify this statement in the main text; well since this is an opinion piece, I guess they can just state their opinion without justification. I guess it’s ok to be highly critical climate models, as long as you believe the 21st century prediction and the 20th century attribution.
I would like to see some serious thinking and discussion about assessment of fitness for purpose of climate models, on whatever timescales. Hargreaves and Annan say 100 years for ‘falsification’, and then dismiss the idea since we can’t wait in terms of decision making.
And finally, someone needs to get serious and discuss alternative climate model structural forms – not just adding more chemistry to the models, but rethinking the structural form of the dynamical core. Otherwise, we are spinning some very expensive wheels and potentially misleading decision makers.
Models forecast 0.2C per decade. Gaia did not agree. One of the models or Gaia are wrong.
Go figure.
I get a “Forbidden” message when I click on the link to the paper
http://onlinelibrary.wiley.com/doi/10.1002/wcc.288/pdf
This link works
I don’t know if I am being stupid, not unknown. However, according to the 2009 ‘Energy Budget’** the Earth’s surface emits to its atmosphere, average IR at the ‘black body’ rate, 396 W/m^2 (16 deg C) plus 97 W/m^2 = 493 W/m^2 convection and evapo-transpiration.
This is ~3x real 160 W/m^2 SW ‘thermalisation’ at the surface. Additionally, in discussing this with Met Office Modellers, I have ascertained that they claim they can apply at ToA Kirchhoff’s Law of Radiation. In the ‘two-stream approximation’. This counts as a negative heating flux in the ‘two-stream approximation. However, you can’t do that for a semi-transparent atmosphere where most IR energy is emitted from within the atmosphere!
The ‘black body’ claim is that 396+97 = 493 W/m^2 mean real energy is transferred from the Earth’s surface to its atmosphere. This is ~3x the real 160 W/m^2 SW thermalisation at the surface. The Kirchhoff’s Law claim offsets this by 238.5 W/m^2, giving a final 1.6x real lower atmosphere warming.
According to G L Stephens****, in hind casting the models use double real low level cloud optical depth in hindcasting to offset the extra warming of the atmosphere, about 30% extra cloud albedo.
As a humble engineer, I do not believe this is right. Am I wrong, or right?
**http://journals.ametsoc.org/doi/abs/10.1175/2008BAMS2634.1
***http://www.gewex.org/images/feb2010.pdf
Sorry: ‘the Earth’s surface emits to its atmosphere, average IR at the ‘black body’ rate, 396 W/m^2 (16 deg C) plus 97 W/m^2 = 493 W/m^2 convection and evapo-transpiration.’ is wrong.
It should be ‘the Earth’s surface emits to its atmosphere, average IR at the ‘black body’ rate, 396 W/m^2 (16 deg C) plus 97 W/m^2 convection and evapo-transpiration = 493 W/m^2’.
A few more passages to discuss. If you’re looking for a discussion on the hiatus, the upper-right-hand passage is the only implied mention of it in the entire opinion paper (other than decadal accuracy in general).
https://twitter.com/USthermophysics/status/477896956488863744
Judith, there is a frustrating lack of information in your commentary here. What, precisely – no, perhaps even generally – do you mean by “someone needs to get serious and discuss alternative climate model structural forms”?
At the moment, it just looks like vague disapproval for the sake of it. What are your concerns about the structural forms of the dynamical core?
(Doug McNeall)
Doug, I am intending to do a post on the hawkmoth effect (was hoping for a guest post, which hasn’t materialized). See
http://www.lse.ac.uk/CATS/Talks%20and%20Presentations/Talk%20Abstracts/Consequences-of-the-Hawkmoth-Effect-Thompson-Smith.pdf
The Earth’s climate is a complex many-dimensional dynamical system with feedbacks, nonlinearities, and long time scale internal variability. The Hawkmoth Effect is a description of structural instability in dynamical systems (sensitivity to the mathematical structure of a model), analogous to the manner in which the better-known Butterfly Effect describes dynamical instability (sensitivity to initial conditions). The Hawkmoth Effect can be paraphrased in the following way: “You can be arbitrarily close to the correct equations (model structure), but still not be close to the correct solutions (future trajectories)”. Here we illustrate the possible consequences for model-based climate research, drawing together a number of observations about modelling, data assimilation, climate model calibration, and numerical solution of partial differential equations. We contrast and then synthesise this with the statistical/stochastic modelling viewpoint and the sophisticated Bayesian frameworks that have been proposed to interpret model output. We conclude that the primary uncertainties in long term climate projections may well be the implicit subjective assumptions that dynamical modelling is appropriate and adequate for the predictive task, and that the Hawkmoth Effect will not be experienced. Good Bayesian practice involves quantification of such prior assumptions. The likelihood that model runs are adequate for predictive purpose will vary with the space and time scales of the task, and is necessarily quantified outside the limitations of the modelling framework; thus, intermodel diversity alone may not be informative about the probability of adequacy, nor about the timescales on which this assumption may become invalid. Intuition based on the underlying physics of the situation is, fortunately, likely to be of use, and expert elicitation frameworks exist for quantifying such judgements. Acknowledging and explicitly incorporating these subjective probabilities into research output, as the IPCC have already begun to do, will continue to improve the internal consistency, relevance and usefulness of climate science as a support for policy and make it more robust to anti-scientific attack.
I’ll do a post on climate model structural uncertainty next week, with ideas on alternative structural forms
Judith, is this basically saying transient global average temperature response and climate responses are moving targets and that there is no such thing and equilibrium sensitivity?
(but we can probably make good estimates of TCR on some short time scales)
An issue that gets far too little attention is the review of models. Models seem to be considered secret intellectual property of the creators, and therefore not subject to inspection by others. As you point out, climate models are complicated, consisting of many parts (modules, let us call them) that must interact with each other, and which include feedback loops that involve multiple modules. How do we know that the code of these modules accurately represents the “known” physics relevant to the module, that there is not additional code and constants representing “informed guesses” about relevant physics, chemistry, etc., where less than complete science is unable to provide the needed information, how scientific or arbitrary the interconnections between the modules are, and a host of other questions many would like to inspect the modules to determine?
It is close to criminal that large, complex bodies of code, mostly developed with government funds, can remain hidden and unexamined by multiple independent investigators. If we are to make major societal decisions about dealing with AGW based on these models, they deserve to be analyzed in depth by independent reviewers, so that we have some basis for believing they are sound. The model builders themselves should be calling for this, because otherwise if society takes actions based on what the models say, they will be responsible for any erroneous actions society takes.
In my view, any climate model developed with public funding should be made available to anyone willing to spend the effort to examine it.
“any climate model developed with public funding should be made available to anyone willing to spend the effort to examine it.”
Paraphrasing IPCC lead author Prof Phil Jones :
Why should I show you my model when I know you’ try and find something wrong with it?
The defeaning silence from government climate scientists in response to this flagrant anti-science comment, clearly illustrating that the prevailing ethic in govenment climate science is there to find excuses for more government, not get to the truth of the matter.
Doug, There are so many issues, it is hard to know where to begin. Turbulence modeling, artificial dissipation, planetary boundary layer, finite element methods vs. finite difference, etc. i gave a more complete discussion at James’ on a sensitive matter and at ATTP, even though I have little hope anyone understood it there. Basically, there is simply no such thing as an “accurate” simulation of a vortex street at even moderate Reynolds’ numbers. Vortex dynamics are basically the same at all scales above the turbulent scales so the “physics” is essentially the same and the methods have advanced tremendously since the 1960’s.
There is a vast literature on these issues in the numerical analysis and engineering world and it is just not credible that one would be able to predict the atmosphere’s details on long time scales.
Doug,
Natural variability in climate science really isn’t that hard to understand or to deal eith. The oscillations may look erratic or even chaotic but they are usually caused by very simple fluid dynamics. As an example, all of ENSO comes down to a nonlinear interaction between a nearly periodic forcing function and a natural sloshing resonance in the Pacific Ocean:
http://contextearth.com/2014/05/27/the-soim-differential-equation/
http://contextearth.com/2014/06/17/the-qbom/
Climategate emails and official responses to those 2009 disclosures show that we cannot trust current climate models.
In 2007 Smith et al of the Met Office published a paper saying that they predicted (not projected) a warming of 0.3C over the period 2004-2014, accompanied by a press release and media coverage. That prediction turned out, of course, to be completely wrong.
In 2009, Lean and Rind made a prediction/jection based on a simple empirical model of 0.15C from 2009-2014, again completely wrong.
The climate science community has failed to acknowledge these failures. It is telling, for example, that Hargreaves and Annan do not cite either of these papers.
JC said:
“…not just adding more chemistry to the models, but rethinking the structural form of the dynamical core.”
——
Yes, for the models to improve, their dynamical core must evolve. Any model that hasn’t, for example, been altered as a result of the “hiatus” is in danger of not being as useful. Any model that can’t explain dynamically, for example, how the deeper ocean can warm at a faster rate than the surface is in danger of not being as useful. In short, models are almost like living things and need to use feedback to evolve. The hiatus is feedback, ocean warming is feedback, increased Antarctic sea ice is feedback, and these all need to be accounted for in the evolved dynamical core of models.
Isn’t a simplier summary-
Any model that does not reasonably accurately match observed conditions should have little to no value in establishing government’s policies.
The problem is one of temporal resolution and natural variability versus a long term gradual forcing. Suggesting models that forecast most accurately at multidecadal to century level scales have no value in policies that will affect those future generations in absurd. Future generations will not forgive us if our excuse not to act was because the models had poor short term temporal resolution.
So . . . exactly how do we know that the models forecast accurately at multidecadal to century level scales? See my uncertainty monster paper for concerns re circular reasoning in 20th century attribution assessments. Also, the climatology of many of the models is as much as 2C from observed climatology
Gate writes-
“Suggesting models that forecast most accurately at multidecadal to century level scales have no value in policies that will affect those future generations in absurd.”
How do you know that they are even reasonably accurate over multidecadal to century level scales?
“So . . . exactly how do we know that the models forecast accurately at multidecadal to century level scales?”
Gates just knows, although to be fair he didn’t say that. What he’s really saying…or doing…is invoking that favorite of alarmist tools, the ever useful precautionary principle…which somehow I never see applied to the possibility that any supposed cure might turn out to be worse….perhaps much, much worse… than the alleged disease….
They assume that the models will work out in the long run, because they assume their assumptions are correct. Some would call that circular.
“Suggesting models that forecast most accurately at multidecadal to century level scales have no value in policies that will affect those future generations in absurd.”
This is the CAGW equivalent of the Our Father in Christianity – the fundamental prayer of the religion. It is a statement of religious faith, not a description of the physical world.
We know for a fact that the models are completely (and increasingly) inaccurate as far as temps for the last 17+ years. But we will still assume that they are “most accurate to century level time scales.” Why? Because it will take hundreds of years to ave any data.
It’s like “finding” the “missing heat” in areas of the globe where we have virtually no measurements.
I thought the new rule was “predict nothing that can be proven wrong in less than 30 years”
They’ve learned a lesson from the failed models, just not the right lesson.
Because the models are inaccurate at a multi decadal scale we should trust them to be accurate at a century scale, Gates?
Heh!
Gates > Suggesting models that forecast most accurately at multidecadal to century level scales have no value in policies that will affect those future generations in absurd.
So where are these models and the data that validates them then?
Models cannot forecast anything. Theycan only be used to check algorithms by comparison with observation of reality.
How do we know, or why would we be led to believe the models are more accurate at multi-decadal to century scale timeframes? It really goes back to fundamental thermodynamics and the nature of chaotic systems under an external forcing. A very simple analogy that I’ve used many times is one of a raindrop following a trail down a glass window pane. Despite the fact that virtually every bit of the physics is known (gravity, intermolecular forces, the momentum of the raindrop based on initial mass, etc) there is no model that can tell you exactly the path that raindrop will follow to the bottom. There are too many variables involved that can’t be fully measured (i.e. Natural variability). Yet a good model could tell you if a drop will make it to the bottom, and even within a range, when and where it will reach the bottom. The exact path however, is completely subject to deterministic chaos, and no model (except by luck) could tell you the exact path.
The path of future tropospheric temperatures, subject as they are to natural variability, unpredictable volcanic forcing, actual human emissions of GH gases and aerosols, etc. is completely unpredictable, yet within a range, assuming GH gas forcing continues, the models ensemble forecast at longer time frames gives us the best idea we have of what the general trajectory is from Anthropogenic forcing. Better still, we can look at the paleoclimate data, looking at analogues for the last time a similar combination of forcings were affecting the climate. The wealth of data coming now from Lake El´gygytgyn in northeastern Siberia is showing increasing convergence between models and the paleoclimate data. Better still, by default, the paleoclimate data contains all the negative and positive feedbacks, both known and unknown. Thus, in broad terms, a century from now we can have a reasonably high degree of confidence of a world that is warmer (likely much warmer in the Arctic), with less net global glacial ice, higher sea levels, seasonably ice-free Arctic, etc. This all represents the destination. The exact path we go to get there is a matter of deterministic chaos, natural variability, and perhaps a series of dragon-king events.
RG said:
One of the significant research findings on nonlinear interactions is that a periodic external force acting on a chaotic system can destroy chaos and as a result a periodic regime results.
[1]G. V. Osipov, J. Kurths, and C. Zhou, Synchronization in oscillatory networks. Springer, 2007.
So much for blaming what might happen on the great unknown. ENSO is not as unpredictable as it is made out to be — it actually goes through quasi-periodic limiting extents.
Gates. Where are the validation data that show the models to be accurate at multi-decadal and/or century time frames? Your hand waving above really doesn’t convince me.
You are arguing that the climate is as stable as a rain drop running down a pane. Why do you believe this? See the comments of Lorenz in this blog. We have no reason to belive the climate is stable and constant and that we will evolve to any such steady state like the drop of rain. It is a bogus analogy.
This sort of reasoning might convince non scientists, but it only serves to irritate people who actually understand the concepts of model reliability and mathematical proof. It gives your cause less credibility with those from math and engineering, not more.
Joey Nickels,
Lorenz probably was the one primarily responsible for diving in to the rabbit hole and getting everyone to follow.
ENSO is not as chaotic as everyone is lead to believe.
Btw, how did your comments at SkS go over, eh?
“ENSO is not as chaotic as everyone is lead to believe.”
This proof that climate is predictable is just about as rigorous as all the other ones I’ve heard, SKS included. Proof by intuition. Proof by “personal conversation”, proof by flawed analogy.
Its all the same nonsense.
“ENSO is not as chaotic as everyone is lead to believe.”
This one is “proof by I said so”.
Chaos isn’t even the issue, that is overkill. Numerical integration error alone ends all claims of predictability. Any problem that has an unstable manifold will yield it unpredictable, it doesn’t have to be chaotic.
Its called math. Their are people who spend their careers studying it and becoming experts at it. A subfield, known as numerical analysis in particular is very useful.
These people actually dedicate their career to learning about these topics! Amazing, eh?
Errors don’t propagate if there is a sufficient forcing that gets applied periodically to resynchronize the behavior.
You are solving the wrong problem.
If the models are based on an assumption that there is a free transfer of heat from the atmosphere into the ocean then they are bound to be wrong. The simple fact is that heat from a gas cannot enter the ocean through its surface via the convection process. The heat is completely blocked by surface tension even the surface remains stone cold. If a metal floating object is place on the surface the water will absorb heat because the surface tension is cancelled.
You can not put heat into the ocean in ADDITION to the heat that goes in via the sun’s radiation. That is the answer to Trenberth’s problems and it also means that the ocean cannot be boiled away. Water is interesting stuff.
RMB to perform oxygen evolution chloroplast experiments we used to use a slide project lamp as our photon source. As he tungsten filament is a lot cooler than the sun the lamp puts out a lot more IR to SW radiation than does the sun. Without a IR filter the oxygen electrode heats up, so we use an IR filter. What we use is a round bottomed flask filled with water, the shape allows us to focus the beam in the electrode chamber and ‘cool’ the beam, stripping the IR. The water heats up, as one would expect as the water is black in the IR, and glass is about 90% clean.
If models cannot be tested within meaningful time scales how useful are they for supporting policy decisions and what caveats should be clearly stated?
As a starting point with regard to testing, could an analysis of how models are configured to match temperature histories (the post-dictions) reveal something of their strengths and weakness for ‘prediction’? For example, what are the ‘fudge factors’ used to help configure the models, how many are there, what sensitivities do these cause, what happens if we progress systematically through single and multiple fudge factors taking them out, etc.
But they have been ‘tested within meaningful timescales’ and they have failed pitifully.
We know – by observation – that they cannot predict the most important climate number – the sensitivity – and hence fail to get the temperatures right.
If you cannot predict temperatures, you cannot predict anything about ‘global warming’ or ‘climate change’ and from these failures all others flow as night follows day.
Do you suppose making a lot of long runs of models, and comparing their level of internal variation to what we can deduce of pre-20th century internal variation, might offer some weeding out?
I have to say that if the models can’t get major features like EL Nino right, it’s hard to believe their treatment of internal variation could be expected to be similar to reality.
Otherwise, we are spinning some very expensive wheels and potentially misleading decision makers.
We should take out one word and add a few.
Otherwise, we are spinning some very expensive wheels and misleading decision makers, and everyone else who believes Chicken Little.
“Otherwise, we are spinning some very expensive wheels and potentially misleading decision makers.”
No, completely wrong. There’s no misleading of decision makers, they’ve made the decision and simply pay to have these models produced to support them.
Yes the basic problem is that models are paid for by a single institution – the state – that has an enormous vested interest in having them predict dangerous warming. So barring some unlikely conspiracy in their ranks to try and get to the truth of the matter, the modellers and their friends will thus always just ignore and hide ideas and data that don’t support an alarmist narrative.
We’ve known all this from as long ago as Climategate.
“2–4∘C over the 21st century.”
So far, since 2001, cooling, not warming.
So far, since the satellite era, 1.3 to 1.5 C per century – still off the mark.
So far, the peak 30 year rate since 1880 – 1.8C per century – STILL off the mark.
More significant than the GAT number is the error of the missing hot spot in the tropical upper troposphere. It reflects the errors we’ve known all along – gross error in the parameterizations of non-linear atmospheric processes.
We acknowledge the climate models don’t work, but nevertheless we believe their predictions of increasing warming is correct.
OK.
It’s worse than ‘we believe’ Jim – they claim “We can therefore be confident that the broad features of the climate system response to anthropogenic forcing are reasonably represented by current models.”
The logic leading up to this claim, however, is non-existent.
Paul, you write “The logic leading up to this claim, however, is non-existent.”
Precisely. But until the warmists, including our hostess, admit this, then the people who matter will assume that this logic DOES exist.
More important, we believe the consequenses of warming.
Good question by Judith Curry, concrete answer by James Hansen.
• 21 January 2014: James Hansen predicts: “The “climate dice” became noticeably “loaded” by the first decade of the 21st century … it appears that there is substantial likelihood of an El Niño beginning in 2014, and as a result a probable record global temperature in 2014 or 2015”
• 20 June 2014: Tropical Tidbits Ocean Analysis webpage reports: an unprecedented one-week upsurge in the El Niño index
Conclusion Yes, we already have solid reason to trust thermodynamic climate-change models, that have successfully predicted decade-after-decade global-scale ocean heating, sea-level rise, and polar ice-melt — without pause or obvious limit.
Whereas we have much *LESS* reason to trust cycle-seeking statistical analyses, on the grounds that the community of cycle-seekers commonly violates the fundamental principle “avoid backtest overfitting”:
The increasing dominance of thermodynamic models relative to cycle-seeking models is obvious to *EVERYONE* … especially young scientists and mathematicians, eh Climate Etc Readers?
And a skeptical community that cannot persuade young scientists and mathematicians, is doomed to extinction in a single generation, eh Judith Curry?
> (Hansen’s) Conclusion Yes, we already have solid reason to trust thermodynamic climate-change models, that have successfully predicted decade-after-decade global-scale ocean heating
An amazing feat, given that there is no data to support this.
> sea-level rise
Which has been rising inline with trends long before the CO2 age
> polar ice-melt
Now reversing, especially in antarctic.
IOW, Hansen’s “conclusions” are just his usual blind, politically-motivated prejudices dressed up as science, repeated here by an even blinder groupie dimwit.
I could predict an increasing surface temperature and sea level rise with my calculator. I’m not concerned over whether they happen as such, what worries me is how much and by when. And in this department the climate models need a lot of improvement.
Our hostess writes “I would like to see some serious thinking and discussion about potential falsification of climate models, on whatever timescales. Hargreaves and Annan say 100 years, and then dismiss the idea since we can’t wait in terms of decision making.”
NO, NO, a thousand times NO. You have it backwards. The models have NEVER been validated, and should never have been used to pretend that they can be used to tell us ANYTHING that is going to happen in the future.
If the people who initiated the hoax of CAGW had been honest scientists, and applied The Scientific Method ab initio, they would have got the right answer 50+ years ago.
NO-ONE, and I mean NO-ONE has the slightest idea what happens to global temperatures when more CO2 is added to the atmosphere from recent levels.
Exactly.
I agree. I would add that no one knows what exactly happens to the atmospheric CO2 content either, when more CO2 is added to the atmosphere.
Fortunately, nature is conducting the experiment for us at just the right time, with the drastic reduction in solar activity and/or the shift in natural climatic variability. Just a little patience (about a decade) and we’ll know much more.
of course we have an idea.
if you add c02, the planet will warm.
There is no evidence it will cool
there is no physics that suggests it will cool.
Everything we know points toward warming.
So, of course we have an idea of what happens when we add C02
Here is the idea we have: it will warm. Not cool.
pretty simple.
If you jump from the empire state building I have no idea how many pieces you will break up into. I have no idea whether all the bones in your body will break or just 100 or so.
But everything I know from physics and human physiology says you will
not survive. And I dont even have to test it to know this.
Who knows maybe Jim cripwell could survive. There is no proof that you would not. There is no controlled experiment where we dropped your twin
to give us good data. using all the information we have we come to a decision. its not a good idea for you to jump..
Yes, all other things being equal, the planet will warm with more CO2. But over the course of the 21st century, we have no idea how important the CO2 warming will be relative to natural variability (sun, volcanoes, internal variability). And even if natural variability is a wash over the 21st century, we don’t know how much the planet will warm from CO2, because we don’t have confidence in the sign/magnitude of the cloud feedback and also the water vapor feedback. So is +2-4C credible as a prediction for 2100? Not to me it isn’t.
Steven, you write “of course we have an idea.”
Once again you are playing with words, and not addressing the physics. Yes, if I jump off a high building, I will not survive. But if the drop I jump off is only 1 foot high, I will survive. We cannot discuss the physics unless and until we know the value of climate sensitivity.
And furthermore, all we need to do is wait for the next episode of repeated El Ninos and the warming will resume.
Not necessarily.
Yes there is. Currently approaching 20 years of flat temps with a slight cooling trend recently – meanwhile atmospheric CO2 up 40%. That is evidence.
Of course there is.
No it doesn’t.
And demonstrably, pretty wrong. Hence the OP’s accurate statement that we have no idea.
False analogy. The physics of a falling body are a relatively simple system. The Earths climate system is decidedly not, by several orders of magnitude. When you have to resort to such asinine comparisons to make your point, your point is wrong.
And even though the physics of your false analogy are orders of magnitude simpler than the Earth’s climate system, “everything you know” about it is not enough to prevent you making false claims of knowledge.
In point of fact, people have survived jumps from the top (the observation deck – at ~1,000 ft., the highest point that the general public can typically access) of the Empire State Building. The most recent jumper not only survived, but walked away with only minor injuries. Even for the relatively simple system of a jumper off the ESB, you get it wrong. That is because “everything you know” is an oversimplification of that already very simple system.
In In further point of fact, many people have survived jumps from much higher altitudes than the 1,200 or so feet of the ESB – several miles up. Meanwhile, many people have died from slipping in the bathtub.
Yeah, ya do. Ya think you don’t, but you are wrong about that. Even for the very simple system of a falling body. Without testing it, you don’t know what the winds and other air currents in NYC can do to a jumper, for example. You didn’t think that was important enough to include in “everything you know”, but it turns out that it is.
Much more egregiously, you oversimplify the question of CO2 in the vastly more complex and recursive beast that is the Earth’s climate system. You have no idea.
Jim, is water vapor a greenhouse gas?
Judith, is there a credible range of temperatures that you would assign to a doubling of CO2?
For equilibrium climate sensitivity, at a very likely level, I would put the range between 0.5-4C. Very likely confidence level implies 10% chance that the true sensitivity lies outside this range (I make no assumptions about shape of the distribution). Note my very likely range is shifted 0.5C below the IPCC’s likely range. Note, this has evolved since my previous statement on this, I think i previously said 0-10C at the very likely level. That said, I am not sure how useful the concept of equilibrium climate sensitivity is.
In terms of a prediction for 2100: I would put the range 0-2C, likely confidence level
Thanks, Judith.
Judy of course 2-4C its credible for me, because I accept different assumptions and because I oppose credibility to incredibility.
Its not incredible to assume that cloud feedback and water vapor feedback would be such that 2-4C could occur.
That assumption is clearly credible. It’s not contrary to any known physics.
2-4C is a possible scenario. If this is what you mean by ‘credible’, then ok. but ‘credible’ also has the connotation ‘worthy of belief or confidence’ – the issue is the confidence that is associated with these numbers
Jim
‘Once again you are playing with words,”
no. you claimed that we have no idea.
Of course we have an idea.
I just told you what it was. so obviously some of us have an idea.
Judith, you write “For equilibrium climate sensitivity, at a very likely level, I would put the range between 0.5-4C.”
The mind boggles. Here we have a scientist giving a range of numeric values for a quantity that has NEVER been measured. I never thought I would live to see the day when this sort of horrible guess was presented with some sort of claim to having a scientific basis.
Steven, you write “Of course we have an idea.”
You have given no scientific basis whatsoever, for believing any numeric value of climate sensitivity is correct. Or are you still claiming that there is no categorical difference between estimates and measurements?
Don, you write “Jim, is water vapor a greenhouse gas?’
Of course. So what?
Judith
To clarify did you mean the estimate Made at 3.07 was o.5 to 4 degrees C? If so that is surely useless?
It is a bit like some authority saying they expect the sea to increase in height by 5 cm to 4 metres. It is of no use at all in sorting out what the response should be.
Have I misinterpreted the estimate?
Tonyb
Well you are exactly right to infer that this is of no use at all in sorting out what the response should be. that is why i refer to situation as ‘decision making under deep uncertainty’, and have discussed strategies for making decisions (no regrets, robust) under such conditions
JC:
”
Hargreaves and Annan’s article is an Opinion piece, not a scientific article; as such it is a relatively superficial treatment of the subject.
”
And Judith Curry’s and Steven Mosher’s and Jim Cripwell’s personal beliefs concerning climate sensitivity are in the form Blog comments, and as such they are absolutely superficial treatments of the subject.
As if “credible for me” opinions are anything but.
Let’s all compare our post-modern partial narratives!
JC:
”
I guess they can just state their opinion without justification.
”
Indeed.
You people are too funny.
Heh, tb, she’s both narrowed and lowered her previous range. Since she gets attacked at both ends of her range, this is both brave, and as you say, relatively meaningless.
========
Kim
Your way with words has deserted you. It is not ‘relatively’ meaningless, it is completely meaningless.
We have had thousands of climate scientists working on this for over 20 years running cumputsrr models that would have taken millions of hours of human time and after this titanic and highly costly effort we have nothing better than a wild guess?
What purpose does such a wild guess serve?
I hope I have misconstrued Judith’s figure.
Tonyb
Jeb, you’d be funny haha too if you weren’t so peculiarly bitter.
=============
Reverend, you write “And Judith Curry’s and Steven Mosher’s and Jim Cripwell’s personal beliefs concerning climate sensitivity are in the form Blog comments, and as such they are absolutely superficial treatments of the subject.”
I am stating a fact. Climate sensitivity has never been measured. Are you claiming that it has been measured?
Tony, you write “What purpose does such a wild guess serve?
I hope I have misconstrued Judith’s figure.”
You and I know that all numeric values of climate sensitivity that anybody, including our hostess, quotes are meaningless. However, these numbers have been in the literature for so long, that no warmist ever has to admit that they are meaningless.
Jim
If you were to give me thousands of scientists, millions of hours of computer time and huge sums of money I would hope I could reconstruct the past temperature of the earth more accurately than a vague guess of such a wide temperature range that serves no useful purpose.
Tonyb
Mosher, WRT your determination that CO2 will warm the planet, a two-part question:
Part I: Is the nature of your determination:
1. Prediction
2. Projection
3. Expectation
4. Anticipation
5. Or what?
Part II: Was the method of your determination:
1. Measurement
2. Calculation
3. Estimate
4. Warm tingle running up your leg.
5. Metaphysical ESP
6. Not sure.
Just wanted to know if you got that part, Jim. No further questions. Thanks for your reply.
Steven Mosher: of course we have an idea.
if you add c02, the planet will warm.
There is no evidence it will cool
there is no physics that suggests it will cool.
Actually, as I have pointed out, there is physics to “suggest” that, starting now, adding CO2 to the atmosphere will result in net cooling. The “suggestive” evidence is the evidence and unexplored phenomena related to the non-radiative transport of energy within the system. There is certainly evidence from the historical record that cooling is possible.
Equally important, there is evidence, from biology and agricultural research, that cooling will be more harmful to life than warming, unless the warming is considerably more than the best evidence suggests to date.
You claim that we face a choice as stark as jumping from the Empire State Building, but that is absurd. It’s more like we have been bouncing on a trampoline and the question is whether an adjustment to the trampoline (or an increase in energy from some source such as increased wind) will make us bounce to higher highs or lower lows or perhaps make no noticeable change.
Using all of the evidence that we have now, it is a really good idea to prepare to collect more evidence for decades, and to prepare for both increases and decreases in temperatures, and to prepare, upgrade and expand flood control and irrigation infrastructure, and to develop a more diverse energy generating technology. There is no case to support immediate drastic actions such as California’s AB32.
Can we trust the models to help us make the proper investment decisions now? At present, they are not informative at all.
Mosher and Curry have informed ideas that are in the ballpark. The rest are arguing from ignorance.
Steven Mosher: I oppose credibility to incredibility.
What does that mean?
Jim Cripwell: Here we have a scientist giving a range of numeric values for a quantity that has NEVER been measured.
It hasn’t been measured *yet*. When Michelson began his work, the speed of light had not been “measured”, but Michelson knew of the history of estimation, including an astronomically based calculation that got the value of about 2/3 its currently estimated value. There is a large history of estimates of values that had not been measured *yet* at the time important decisions were made, or better measurements/estimates proposed.
> The “suggestive” evidence is the evidence and unexplored phenomena related to the non-radiative transport of energy within the system.
I’d like to know more about that noumena.
It could also
Don
Is an estimate as broad as this really in Any sort of ballpark?
http://judithcurry.com/2014/06/20/can-we-trust-climate-models/#comment-599749
Tonyb
Matt, Mosher did not say we face a choice as stark as jumping off the Empire State Building. You might want to back up and regroup. Mosher is not a catastrophist.
Tony B: We have had thousands of climate scientists working on this for over 20 years running cumputsrr models that would have taken millions of hours of human time and after this titanic and highly costly effort we have nothing better than a wild guess?
It is a little better than a “wild” guess. Other than that, Prof Curry’s paragraphs that you responded to there puts the case nicely. And as she noted earlier, that is the “equilibrium” climate sensitivity, which has only a vague notional analogous relationship to changes in Earth climate.
It’s a spacious ballpark, tony. Not conducive to hitting home runs. Mosher and Curry are smart people who have studied this question for some time. If they are not able to narrow the range down fine enough for you, try another blog.
Don Monfort: Mosher did not say we face a choice as stark as jumping off the Empire State Building.
He wrote of consequences as disastrous as breaking up into pieces. Exactly how close an analogy, metaphor etc that was intended is unclear from rereading the whole, but he clearly did not write as though it were the same as jumping a foot onto grass.
Don Monfort practices squishy logic.
Matthew, you write “It hasn’t been measured *yet*. ”
I agree. However, the no feedback climate sensitivity is IMPOSSIBLE to measure, yet it is still quoted as a valid number. No decision involving billions of dollars, and our standard of living should be made UNTIL the measurement has been made.
Willard(@nevaudit): > The “suggestive” evidence is the evidence and unexplored phenomena related to the non-radiative transport of energy within the system.
I’d like to know more about that noumena.
Start with the energy flow diagrams of Trenberth et al, and Graeme Stephens, et al. (for example, in “Atmosphere, Clouds and Climate” by David Randall, p 24.)
It wasn’t an analogy, Matt. It was an illustration. He is trying to school Cripwell, without any success. Et tu?
jimmmy2,
I used to be a denier. However, I have an open mind and I am capable of learning. I have evolved to lukewarmish lukewarmer. You appear to be stuck, like so many others. Until you get in the ballpark, you can’t play ball.
Jim Cripwell: However, the no feedback climate sensitivity is IMPOSSIBLE to measure, yet it is still quoted as a valid number.
It is like the atomic mass of lithium, or the energy dissipated by a typhoon.
Don Monfort: It was an illustration.
It was also absurd.
You think it is absurd, Matt, because you think Mosher is likening jumping off the Empire State Building to not taking action on CO2. He isn’t. You are wrong. I am surprised. You usually have more sense.
“Steven Mosher: I oppose credibility to incredibility.
What does that mean?”
That means exactly what it says.
credible means “able to be believed”
I oppose this this to
“impossible to believe”
so the models give a credible prediction. I am able to believe it if I accept certain assumptions
A simple EMB gives a credible prediction
climatology gives a credible prediction.
These are all “believeable” none are crazy.
they are all well reasoned.
they are internally consistent
They all depend on assumptions and simplifications
The question is.. which one is right?
that might be the wrong question.
Steven Mosher: The question is.. which one is right?
that might be the wrong question.
ok, I got it. That is one way of using “oppose”.
I frequently maintain that “Which one is right?” is the wrong question. Since all models have omissions and are approximations, the “right” question is “Which model is (or models are), shown to be sufficiently accurate for a given purpose”; which purpose may be guiding the design of the next experiment to increase the accuracy of an estimate.
No model of climate is demonstrably accurate enough to guide energy investment over the coming century. The GCMs discussed in the header are clearly too inaccurate to date for a claim of accuracy in the distant future to be “credible”.
Don Monfort: You think it is absurd, Matt, because you think Mosher is likening jumping off the Empire State Building to not taking action on CO2. He isn’t.
You have persuaded me that the Empire State Bldg “illustration” was merely an irrelevance, not an absurdity.
You are making progress, Matt. Do some more thinking. You could ask Mosher to help you.
Steve Mosher, what you say has obvious truth in it, but is simplistic and incomplete. You have little to no idea where we are on the ” CO2/temperature dose-response curve. Let me give you a pharmacological example: if I infuse 10ug of norepinephrine (an alpha adrenergic agonist) into one of your arteries I will elevate your blood pressure, among other things. If I continue to infuse this dose over a few days I notice that you rise in BP becomes progressively less pronounced until at some point I have to increase the dose so much to get the same effect that I will probably kill you for other reasons. You see Steve, the alpha receptor responds by be phosphorylated and norepi can no longer trigger it (tachyphylaxis). So it matters a great deal where you are initially on the dose response curve and feedbacks are critical to response. Clouds may be to CO2 driven warming as tachyphylaxis is to adrenergic stimulation.
We have had thousands of climate scientists working on this for over 20 years running computer models that would have taken millions of hours of human time
The descent of man is well documented eg Levy-Leblond
Fig 3 captures the information well.
http://www.udppc.asso.fr/national/attachments/article/559/energie_levy_leblond.pdf
Mosher equated the certainty of increased CO2=warming with the certainty of jumping off the ESB=death.
You people amaze me.
I have to add, I have no idea what part of the CO2 side he was comparing with the analogy of the uncertainty in how many resulting pieces/broken bones there would be from the tinkerbell leap.
That one threw me.
> Start with the energy flow diagrams of Trenberth et al, and Graeme Stephens, et al. (for example, in “Atmosphere, Clouds and Climate” by David Randall, p 24.)
Got the Trenberth diagram, MattStat.
Now, where’s the unobserved phenomena?
Willard, “Got the Trenberth diagram, MattStat.”
Since that is the only one you have you may be part of the phenomenon :)
One the K&T budgets, all of them, the atmospheric window is estimated as a fixed 40 Wm-2. That window is what is supposed to close somewhat as CO2 and even more water vapor reduce the size. If you look at the Stephens et al budget they estimate the window at close to 22 Wm-2 or about 18 Wm-2 of potential warming has already occurred. Science of doom has an interesting post on the Trenberth “window” and the water vapor continuum.
http://scienceofdoom.com/2013/02/02/kiehl-trenberth-and-the-atmospheric-window/
Now see if you can find the phenomenon.
Willard(@nevaudit): Now, where’s the unobserved phenomena?
Consider the energy and water flows in the wet thermals: How much increase or decrease will be caused in those flows on the non-dry surfaces of the Earth as the CO2 concentration in the atmosphere doubles? “Unobserved phenomena” without elaboration is an oxymoron: here I refer to something that can be studied better than it has been studied up to now, a change in flows (almost for sure) induced by the increase in DWLWIR hhitting the non-dry surfaces; that is something that I claim has to be known in order to know the “surface temperature sensitivity to a change in CO2.”
The equilibrium hypothesis entails the that the increased radiation flow will last long enough to raise the temperature (and its new wet lapse rate) to the new equilibrium value, but it does not take into consideration the non-radiative heat flux from the lower to the upper atmosphere, or the daily limited duration of the maximum insolation. Thus, the standard calculations produce a biased estimate, biased upward, of how much the surface temp can rise and must rise in order to rebalance the energy fluxes after CO2 increases.
As a non-scientist, I look at the output of the models (usually the ensemble average, as spaghetti graphs do nothing except remind me of my years in Italy) and I see a broad agreement with the temperature record, just scaled too high.
I don’t think anyone denies that models overestimate warming as exhibited to date and I personally would not be surprised to see that overestimation continue. But the broad shape of the curve on decadal scales doesn’t look bad.
And that’s a big achievement for models. Our problem is that we are asking models to do something that they cannot. Look at very minor changes on short term time scales.
We shouldn’t blame the models for the imperfections of our impetuous policy desires. Grow up and be patient.
If I remember rightly, Zeke Hausfather exhibited the decadal warming trends since 1940 over at Lucia’s in response to a request of mine. I believe the highest was 1.9C, although I don’t really recall which decade it was for. Most decades were significantly lower.
When we have about 38 decades of that kind of measurement record in hand, we will be able to make statistically significant comparisons to model projections. Until then we really cannot.
The consensus tendency to place religious faith in model outputs is clearly based on the outputs delivering an answer that pleases them. But I don’t consider that to be the fault of the models.
The climate has warmed over the past 130 or so years. It seems to correlate well with our emissions. Logic and physics strongly suggest that our activities have contributed to that rise in temperatures. The fact that the rise hasn’t been even should surprise no-one. The models chart that rise fairly faithfully, but overestimate the actual amount of heating.
Big whoop.
Don Monfort: You could ask Mosher to help you.
No need. He pitches in when he feels like it.
> “Unobserved phenomena” without elaboration is an oxymoron: […]
This is why I referred noumena, MattStat.
Thank you for detailing your working hypothesis.
Thank you for your explanation.
Jim2
The basis is physics. Watch Linden last talk
For the details.
Our best physics says more co2 will warm the planet
Not cool it.
Pretty simple.
Mosher,
Far too simplistic.
Climate changes abruptly, not smoothly as depicted by the climate models. Sudden cooling is due. Will human’s GHG emissions reduce the probability or magnitude of the next abrupt cooling event, OR delay it, OR prevent it all together?
If you believe i\ our GHG emissions will prevent it, will man’s emissions cause an abrupt warming event? Will the impacts be net good or net bad for humanity?
What are the probabilities of the time to, direction of and magnitude of the next abrupt climate change. What are the probabilities of the impacts of such a change?
Jim > Once again you are playing with words,
Mosher > no. you claimed that we have no idea.
Of course we have an idea. I just told you what it was. so obviously some of us have an idea.
Ah, I see the problem here. Mosher (English not his native tongue perhaps?) does not understand the idiom “have an idea”, which of coourse means to have advanced or reliable knowledge. Instead he takes it literally, to mean that he is in possession of a concept of indeterminate rigor.
Steve, re jumping off a high building. I worked in Millbank Tower in London, 338 feet high (over 100 metres). Shortly before I went for lunch one day, someone jumped off a ledge near the top, over 300 feet. He did not die on hitting the ground.
I admit that this does not strictly contradict your point, however. The poor fellow was beheaded when his neck hit a rail about four feet above ground level.
Tasteless, tasteless.
What is your Mosher Tongue?
================
Matthew R Marler @ 6.14 & 7.43, good summation of the essence of the issue.
Mr Mosher, there are several forces acting besides CO2. CO2 forcing is postulated to have positive feedback, but the amount and nature of feedback is questionable. I like to point out to your friends that AT THIS TIME the energy imbalance appears to be much lower than predicted by the global climate models. Not too long ago I suggested to Dr Lacis there may be a benefit if they focus on this point, and I’m writing this AGAIN to point out I’m seeing a disconnect between the model results and the total system energy imbalance.
Why is this important? It points out there’s a need to improve the models. It also shows the climate sensitivity isn’t as high as estimated AT THIS TIME. But we see a nearly hysterical push to adopt renewables based on what is clearly a set of iffy models and an even iffier engineering and economic analysis.
I come from the engineering/economics side, and as an experienced adult I consider the overall performance of the climatology and physics communities to be below par in their approach to this problem. The results are misrepresented, the community pretends to dictate solutions in matters it isn’t qualified, and many community members tend to pontificate.
As for the community of trollers and agitators, they are good entertainment, but their impact is negligible. They are engaged in an old style cafeteria food fight.
Standing at the top of the Empire State building you can argue that thermodynamically, you should already be at the bottom. In pieces.
Kinetics says otherwise.
Before he went down Mosher probably had to jump up, over a barrier.
“if you add c02, the planet will warm.”
But it didn’t, did it?
For around 2 decades, despite the highest rate of addition of CO2 in recorded history.
Even worse for your alarmist hypothesis, the very necessary increase in atmospheric water vapour concentration hasn’t occurred either, in fact Solomon et al seems to indicate it has actually decreased while the CO2 was increasing in the decade post 2000.
Clearly there are other factors involved, and your problem goes way, way beyond the simple, very recognisable fact that the models are totally without skill.
In any case, my thoughts on the folly of attempting to model non-linear systems of effectively infinite complexity and beset by extreme sensitivity to initial conditions have been stated before, on several occasions, anyone who claims they can pull off that trick is deluded at best and a computer salesman at worst.
The Earth is not a flask in a laboratory, and anyone who believes that tit can be assumed to behave like one is clutching at straws.
Steven Mosher: I oppose credibility to incredibility.
Do you also oppose credulity (or credit) to credibility? Just curious.
It’s a credible hypothesis that the measured speed of light is independent of the relative velocity of source and measuring instrument; and it’s a credible hypothesis that the measured speed of light is dependent on the relative velocity. More credit at this point goes to the independence.
Something similar can be said for the luminiferous aether: its existence and its non existence are equally “credible”, but the credit now goes to its non-existence.
I think it is a “credible” hypothesis that an increase in CO2 will speed up the hydrological cycle by a non-negligible. It seems from your writing that you give the hypothesis no credit, but I give it a lot of credit.
Likewise, it is a “credible” hypothesis that the equilibrium-based calculations are sufficiently accurate for informing public policy choices. By now, I personally give that hypothesis little credit, but I guess from your writing that you give it a lot of credit.
Mosher writes, “if you add c02, the planet will warm.
There is no evidence it will cool
there is no physics that suggests it will cool.” Perhaps biology suggests the planet will cool via natural sequestering after the addition of CO2 initially raises temperatures. Here is an article describing one possibility from 49m years ago. (Read past the first 4-5 paragraphs.)
http://www.economist.com/news/science-and-technology/21604532-finding-genome-extraordinary-plant-aquatic-alfalfa
Mosher,
Until we understand clouds, you are whistling in the dark. You many know some of the words, but the rest is just a guess.
Do we know what happens to climate when we have a massive volcano eruption? What are the long term results? Are there negative feedbacks? How do they operate?
And CO2 additions create what feedbacks?
I’m on board – CO2 should cause warming.
But how much?
Emissions are already LESS than the low end scenario.
With forcing and commensurate warming at this low end, they can and evidently are being superseded by natural variability for the observed 21st century.
Whether there’s a reversion to the mean -or- whether the increased albedo IS a reversion to the mean, this is not the stuff commensurate with calamity.
Where is a rigorous analysis of uncertainties in models? There are many sources of error: a rough grid; approximations used for physical processes (turbulence and cloud formation are particularly vexing); mathematical approximations in equation solvers.
Three years ago I reported a 2.5% error in a CAM 5.1 model of energy transport by evaporation from tropical seas. It has a potential to severely distort the model results in as little as 40 hours. Response? “It is a potential problem”.
We can not trust model. Even worse, we can not trust modelers.
I agree – then the real science can begin.
As a layman, the debate about models appears quite strange. As they’re presented as the basis for predicting a global emergency, for which only large-scale changes in public policy can save us, how have they been validated?
The logic step, seen in countless films describing such situations, calls for assembling a well-funded multi-disciplinary team of independent experts to review the models. Statisticians, software engineers, specialists in the specific physical sciences involved, etc. Open the lid and examine the machinery of the models.
While expensive, if done in 2000 it probably would have produced more useful results than the many more superficial projects done by the IPCC, climate agencies, and science associations during the past 15 years — at a similar or probably smaller cost.
This would not substitute for ongoing climate science research, but might help answer some of the vital questions asked by policy-makers and the public.
Any ideas why something like this has not been done?
Editor, you write “Any ideas why something like this has not been done?”
It has not been done, because it is the wrong thing to do. The warmists don’t want to do it; it would cast doubt on their certainty. The skeptics know that non-validated models are useless for making any sort of prediction.
All we ought to say, as scientists, is that there is no physics to tell us what happens to temperatures as we add more CO to the atmosphere from recent levels.
Jim
You know that a climate model can never be fully “validated”. Probably the best that can be reasonably expected is a history of matching observed conditions for the parameters the model is designed to simulate.
Isn’t the real issue how close do the models need to match oberserved conditions for each parameter they are designed to forecast over what time scales for them to be believed to forecast future conditions. Then the next question would be how long into the future should they be trusted.
Rob you write “You know that a climate model can never be fully “validated”.”
Of course, but I would leave out the word “fully”, Let me know when our hostess and all the other warmists agree. Then, and only then, can we start talking science when it comes to climate models..
Rob Starkey: Then the next question would be how long into the future should they be trusted.
Until a butterfly flaps its wings.
> The logic step, seen in countless films describing such situations, calls for assembling a well-funded multi-disciplinary team of independent experts to review the models. Statisticians, software engineers, specialists in the specific physical sciences involved, etc. Open the lid and examine the machinery of the models.
This naively ignores the far bigger issue overshadowing all of this : who would be PAYING these ‘independent’ experts? Once you know that, you will know what their conclusions will be.
Just follow the money.
The problem isn’t only the models. There’s also a need to have a much more formal and sophisticated work flow. This means the model inputs require rationalization. For example, the IPCC has the case they call RCP8.5. The case is fed into the GCM ensemble and the GCM results are used for other studies as well as propaganda. But the fundamental basis for RCP8.5 appears to be flawed. It’s internally inconsistent, and some of the input tables appear to have been prepared by a couple of sixth graders.
Pingback: Climate Models | Transterrestrial Musings
I read the Hargreaves and Annan opinion piece differently, as a weakly disguised apologia for CMIP5 failure to predict the pause.
I also find their core logic faulty given their professed faith in 2-4C end century. Nonlinear dynamic model divergence increases with time. We have known that since Ed Lorenz decades ago. To say CMIP5 models are adequate for temperature predictions 100 years out, but inadequate for a decade out, is backwards. Perhaps H&A meant the boundary conditions ‘envelope’ argument Lacis used in the Ellison thread on this subject a few days ago. I find that a very weak argument for the following reason (unless one has in mind the precautionary principle, taking precautions for the worst case no matter how unlikely–in which direction lies bankruptcy). The climate strange attractor (in n-1 Poincare space) is provably high dimensional. Let us suppose climate remains on one lobe for the next hundred years (no ‘abrupt tipping point’ transition to another lobe). There are still a very high dimensional number of possible solutions (trajectories around the attractor lobe) that are possible, giving a range of future temperatures combined with the other factors. The ‘envelope’ is too imprecise with respect to a single dimension like temperature to be useful, when the debate is over 1.5, 2 (adaptation), or 3 (mitigation) expected degrees while natural variation alone produces variations of about 1.5 degrees (Lanser reconstruction) just comparing the MWP to the LIA. (I ignore for brevity sake here the accompanying and important rate of change issue).
It seems to me that the way forward for practical policy purposes is not through tweaking the dynamical core of GCMs. It is through other modeling techniques entirely, whether simple net feedbacks 1/(1-f), stadium waves, notch delays, whatever. Loehle recently published an interesting example for the purpose of sensitivity estimation (in 2014 Ecological Modeling 276: 80-84). Simpler parameterizations using the observable data we have, for however long we have it, do do what we can. Sure these will not be ‘perfect’, but they might be fit enough for whatever purpose is intended, like sensitivities.
The reason for thinking other modeling paths entirely are necessary is basic and simple. One of the most fundamental processes is tropical convection, which produces thunderstorms and precipitation, mostly over oceans. These must thermoregulate, something along the lines of Lindzen’s adaptive iris (but not necessarily exactly that). There is no foreseeable advance in computing power that will enable small enough grid cells to capture convective cells, independent of the nonlinear dynamics of Navier Stokes equations. So no way to truly model water vapor transport to the upper troposphere (UTrH), its washout via precipitation (UTsH) releasing latent heat near to, at, or above the GHG optical windows, clouds, and so on. That is IMO why CMIP5 still produces a tropical troposphere hotspot when none exists observationally. Because of this impossibility, such processes have to be parameterized anyway, the parameters tuned to produce ‘adequate’ hind casts. (And mistuned to a period of at least partly naturally rising temperatures, so they all now provably run hot).
Better to start with simpler tractable models and parameterizations in the first place. Heck of a lot less expensive. Easier to explore a wider range of hypotheses about basic climate processes and their interactions than in the ‘everything at once’ GCM messes. Convert the climate model supercomputers to weather forecasting, and to work on regional or seasonal stuff. What I personally really want to know is the growing season in Wisconsin by the winter before. Hybrid seed companies would like to know it by the start of the previous growing season, so they can grow the best adapted hybrid maturities. A lot of seed corn grown under center pivot irrigation (so drought proof) in the lower Wisconsin River Valley where an abundant water table is just 15 feet down, but you never know whether you can plant by mid April or have to wait to late May.
Rud, Very well said!
I see it very much the same way Rud.
When the modellers say they need better start condtions to make better prediction they fool themselves. Not even weather models can do more than a week, and after one year or 10 years and averaging over the globe and year, i can’t see why the weather at some specific time should be important.
The models will not improve with higher resolution in time and space and processes, but just be so complex that no one knows what really goes on, and completely looses the overview of whats important and not important.
I once thought that the models could be used to analyse special parts of the climate and gain some new knowledge, but i have grown sceptical of that too.
The best would be to start over again with extremely simple models of the climate, and decouple them from weather models.
I believe Lorentz has proved that you can not predict a chaotic system, and averaging in time and space will not help.
I never believed that a GCM could tell me anything over and above what some simple first-principles calculations could. As an example of something that is actually not easy to do with first-order approximation is the electronic band structure of a semiconductor, such as silicon.
With that knowledge of how something that is seemingly so erratic may be the result of a mix of much less erratic periodic components, I give you the behavior of ENSO
http://contextearth.com/2014/05/27/the-soim-differential-equation/
This is still not trivial, but it can be described in a Wolfram Alpha one-liner.
Climate is not some “angry beast” — it just looks like that way to to us inconsequential ants. To an observer far away, the climate looks like a slight perturbation around a quiescent temperature point a few hundred kelvin above absolute zero. Easy to solve these kinds of systems.
WebHubTelescope: I never believed that a GCM could tell me anything over and above what some simple first-principles calculations could.
“To an observer far away, the climate looks like a slight perturbation around a quiescent temperature point a few hundred kelvin above absolute zero.”
+1
Rud Istvan: I also find their core logic faulty given their professed faith in 2-4C end century. Nonlinear dynamic model divergence increases with time. We have known that since Ed Lorenz decades ago. To say CMIP5 models are adequate for temperature predictions 100 years out, but inadequate for a decade out, is backwards.
That was a good post, start to finish.
‘Non linear dynamic model divrgencies Increase with time.’
http://climateaudit.org/2013/09/30/ipcc-disappears-the-discrepancy/
The IPPC Fig. 1.4 models programmed to overestimate warming
ain’t no measurement just a guesstimate with oh so costly
ramifications.
“I also find their core logic faulty given their professed faith in 2-4C end century. Nonlinear dynamic model divergence increases with time. We have known that since Ed Lorenz decades ago. To say CMIP5 models are adequate for temperature predictions 100 years out, but inadequate for a decade out, is backwards. ”
—-
Conflating weather prediction models and climate models is what is backwards. They are completely different in design and function. Whereas, you are right, the chaotic nonlinearities will create weather model divergence rapidly, climate models may do poorly over decadal or less timeframes (owing to the fact that unpredictable volcanoes and ENSO dominate these short periods), ultimately the net long term climate forcing and those dynamics will provide the longer-term accuracy of climate models.
And yet, Gates, the further along model projections one goes the further they have deviated from measured temperatures. So reality is on Rud’s side.
Gates : ultimately the net long term climate forcing and those dynamics will provide the longer-term accuracy of climate models
Your reasoning of late seems to be that since it is possible that models may well show an underlying long-term warming, despite short term noise (?) fluctuations, therefore they will.
Blindly ruling out any possibility of being wrong, and that just maybe there are factors at play not captured by the models.
Classic IPCC / Consensus conceit ?
Temperature is not climate. And we only care about climate in that we care about the weather.
We are concernded with how warming subtly affects changes in climate and how that affects society. The models get this even more wrong than temperature.
And we should be concerned with how climate is changing regardless of cause.
The model dependency was the bet AGW advocates made. Like Napoleon belief capturing Moscow would win the war. They were politically and rhetorically useful but they just weren’t built for the long-term conflict and standing up to actual science standards. Models were a tactic, they chose to undermine classical empirical science (hypothesis, evidence, reproducibility) for the same reason they decided to make up polar bear population counts. They were politically useful and can be sold as “science”.
Again, all of this goes to the point of who the model advocates are. The usual green driven suspects selling something. CO2 as the Earth’s “thermostat” was a pretty lame concept out of the gate yet thousands signed on for many different reasons but largely out of political identity. That much of academia has sunk into a political enclave of the worst sort is plainly evident to many. Since the entire field of “climate science” was spawned off largely a one-party group think culture with limited (what an understatement) gate-keeping of objectivity, the abstractions of AGW model based “consensus” should be dismissed.
It isn’t empirical, predictable even to trend. The contrivances of models as “proof” of anything has been exposed. The lesson? If you invade Russia you should order winter clothing or if you want to overthrow the classical “scientific method” you need a better meme. We simply discount all the huckster pushing and shoving that brought the AGW movement this far along to this point. Models are a smoking gun of past “expert” failures and were in themselves a desperate act at the time. It’s hard to think of what is left of “science” to manipulate, hype and distort going forward, they’ve tried everything already. So now all they are left with is chasing bad weather and making emotional pleas that people are to blame and should be punished with more government. They bet on “warming” and now they morf to “change”, models are just an unpleasant history to be purged or doctored from the meme going forward. This isn’t going to impact the hardcore advocates or followers, the left-wing base, they never required actual “science” no matter how much they loved the adopting of the word for their purpose.
You are entirely right to be skeptical cwon14.
The best defense against “Goat Bubble” economic devastation is insistence upon disciplined mathematical and scientific practice (particularly in regard to cycle-seeking/cherry-picking backtest overfitting).
That’s mathematical, economic, *AND* scientific common sense, eh cwon14?
@ cwon14
“They were politically useful and can be sold as “science”.”
Exactly!
The purpose of the models was never to predict the earth’s ‘climate????’ a century–or more–from now; their sole raison d’être was to provide ‘scientific’ justification for progressives to establish political control over all energy production and consumption.
Judged on that objective, they have been one of the most successful scientific ventures ever.
You will note that no matter HOW the models perform against observations, the progressives pronounce them valid, ex cathedra, and they continue to be cited as justification for such insanity as the recent coal power regulations.
Exactly Bob. Note how Dr. Curry validates it as a reasoned question in regard to model validity. Without directly endorsing conclusions she backs the basic validity of models as authority.
It’s horrendous.
Judith I agree with the one thing you disagree with.
The models do provide a credible prediction.
Of course it comes down to what one counts as credible.
I would also find a linear extrapolation of temps since 1850 to be a credible
prediction.
I would find the prediction of an Energy balance model (using ECS from observational studies) to be credible.
Clearly none of these are incredible. All are based on the knowledge we currently have, and all make certain assumptions. we can of course have issues with the knowledge they are based on ( nothing unique here) and we can question the assumptions ( nothing unique here either). Those questions go to the weight
What we are left with is a range of credible predictions. What is unclear is
how much credibility each has. what is unclear is how we use these credible
predictions. And I haven’t see a credible prediction that indicates no warming.
Let’s operationalize this. If you asked me to predict 2100 temps. I’d probably give you three approaches: ( see above ) and I’d give you the best arguments for each one, along with caveats. And I’d warn against averaging them. What you have is the range of prediction from various methods, none of which is perfect, all of which involve assumptions, and that just is what it is.
Are they credible enough to base policy on? Of course. The EPA is doing just that. Are they credible enough for more skeptical people? of course not. That just means that notions of credibility and notions of how much evidence is required to make a decision are rather loose. Sorry.
Well the problem is that the IPCC and climate models only provide one method for making predictions. Guessing, climatology, stadium wave, extrapolation of trend, are all methods that might produce better predictions, and these are ignored/dismissed by the IPCC and policy makers in favor alarming predictions from climate models
With regards to policy making, we are facing conditions of deep uncertainty in terms of what will happen in the future re climate; I have written numerous times here about the dangers of optimal decision making approaches using highly uncertain model predictions
Yes. I would think the best thing to do is assemble various methods of predicting. as for how these approaches are weighted or used?
the opportunities for politicization is front and center
I proposed multiple methods for decadal/regional predictions here
http://judithcurry.com/2014/03/19/uk-us-workshop-part-v-broadening-the-portfolio-of-climate-information/
> stadium wave, extrapolation of trend, are all methods that might produce better predictions
I’m not sure how the stadium wave hypothesis is a method.
you can project forward using the stadium wave. Personally, I am not using to predict beyond a half period of the wave
“I’m not sure how the stadium wave hypothesis is a method.”
well, read harder.
““The stadium wave signal predicts that the current pause in global warming could extend into the 2030s,” Wyatt said, the paper’s lead author.”
A method– which is nothing more and nothing less than a list of steps–
is used to understand the past.
Then those steps are applied to produce a prediction.
all hypothesis are methods when you un pack them.
still playing dumb.
> A method– which is nothing more and nothing less than a list of steps– […]
Which means that an hypothesis is a list of steps, no doubt. One has to appreciate when science becomes an algorithmic lists of steps.
Let’s see two examples how the concept of “method” is used in the Stadium Wave comment thread:
http://judithcurry.com/2013/10/10/the-stadium-wave/#comment-396937
The other hit:
http://judithcurry.com/2013/10/10/the-stadium-wave/#comment-396937
The method is an algorithm of some curve-fitting exercise that gets implemented in some code. An hypothesis is the result of that “method”, the other way around, or the result of some feedback loop between the two. But one is not the other.
Nice try.
Nice try.
“Guessing, climatology, stadium wave, extrapolation of trend, are all methods that might produce better predictions, and these are ignored/dismissed by the IPCC and policy makers in favor alarming predictions from climate models.”
——
From a energy dynamics and net forcing perspective, we’ve not yet seen the full effects of even 400 ppm of CO2. Simple linear extrapolation of trends or “guessing’ won’t cut it. Nonlinearities are to be expected, and if the models show “alarming” predictions later this century based on nonlinear dynamics that are not obvious from guessing or simple linear extrapolation, then the very least we should do is understand the source of these nonlinearities. With or without anthropogenic forcing the climate is never linear in progression, and indeed, some 3rd or 4th degree polynomial function usually provides the best fit to almost any variation.
“notions of credibility”
So climate science is subject to this?
Color me skeptical, then.
Andrew
all knowledge is.
“all knowledge is”
So this is your defense of climate science? That all knowledge is subject to notions of credibility. Like I said earlier, this is naught but a joke.
Andrew
a defense of climate science?
no.
since you havent launched a credible attack, no defense is required.
“since you havent launched a credible attack”
OK here’s one: If Global Warming is a reality, then show me where I can find it.
Be a scientist on this one (instead of an English major) and try to give me an answer.
I can’t wait.
Andrew
is that like the challenge “show me a monkey turning into a man”?
is that like the challenge “show me a monkey turning into a man”?
Is it supposed to be?
Andrew
lolwot, if one took away all,of the evidence of “a monkey turning into a man”, would you still believe it?
Or, put another way, do you believe an alligator turned into a man?
“Or, put another way, do you believe an alligator turned into a man?”
You are what you eat?
I’ve seen an alligator turn into a driveway. This one was fondling a garage door opener with a mischievous look in its eye. I thought it was a good time to vote for a new President.
================
After the EPA did a pretty good job of cleaning up a great deal of conventional pollution in America, they no longer had a strong basis for continued existence. But like all bureaucracies, their fundamental and highest priority was to further and continue its operations. Then, came along CAGW. That gave them a new life. The EPA is not concerned with credibility; only survival.
Jerry Pournelle’s Iron Law of Bureaucracy: “In any bureaucracy, the people devoted to the benefit of the bureaucracy itself always get in control and those dedicated to the goals the bureaucracy is supposed to accomplish have less and less influence, and sometimes are eliminated entirely.”
Steve,
Good post. RGB is making some points with you. Your “of course” comment on basing policy confuses me to degree. If we assume man does influence temperature then there are many potential impacts (population growth, road construction, farming, deforestation, etc.). Why only focus on CO_2 which is not really reacting as modeled?
I would also find a linear extrapolation of temps since 1850 to be a credible
prediction.
I love the way you Warmies insist on deriving a linear trend from a cherry-picked section of what is clearly a periodic function and extrapolating it to Armageddon.
Hi Judy – This issue is really important and I am glad it is finally being addressed in the literature, even if an opinion piece.
My latest discussion of this subject is in my House testimony
https://pielkeclimatesci.files.wordpress.com/2014/05/nr-150.pdf
Here are my main points (which I document in terms of peer reviewed papers in the write up):
The 2013 IPCC WG1 report and the 2014 US National Climate Assessment present a set of projections from global and downscaled regional climate models as the basis for projecting future societal and environmental impacts, offered as a guide to the future for decision makers.
However, these projections may not be reliable guides to the future. In fact, we are unable to accurately quantify their reliability. The IPCC and NCA did not adequately discuss the skill run in hindcast predictions over the last several decades when fossil fuel emissions, and other climate forcings, are actually known.
Except for limited exceptions the models cannot accurately represent over the last several decades the temporal evolution of major atmospheric circulation features over multidecadal time periods such as El Niño and La Niña, the Pacific Decadal Oscillation, and the North Atlantic Oscillation. These major factors determine which regions have drought, floods, tropical cyclone tracks, and other societally and environmentally important weather events.
.
The models have an even greater challenge in accurately predicting changes in the statistics of these major atmospheric circulation features over multi-decadal time scales.
The IPCC and NCA needs to more accurately report the importance of these model limitations, that were available to them in the peer reviewed literature. By not alerting them to these limitations, they are giving decision makers who face decisions at the regional and local level a false sense of certainty about the unfolding climate future.
Best Regards
Roger Sr.
WHAT CYCLE-SEEKERS, SKEPTICS & DENIALISTS
ALL GET WRONG
With respect, that is the wrong question Judith Curry!
A better question — for young scientists, family-starting voters, conservationists, religious leaders, and politicians — is this: “What Bayesian evidence can we find to diminish our confidence that ‘James Hansen’s climate-change worldview is fundamentally correct‘?”
The IPCC models are thus properly viewed as a sustained global effort to find any climate models, that are compatible with known physics and the prior climate-record, that give us any substantial reason to be skeptical of Hansen’s worldview.
This effort has failed … every realistic dynamical model of the climate supports Hansen’s worldview. The Bayesian probability that ‘Hansen is right’ must therefore be assessed as far greater than fifty-fifty (for me it’s about 85%).
Recognizing that cycle-seekers, skeptics, and denialists all reject Bayesian reasoning that ‘Hansen is right’, the plain fact is that young researchers overwhelming embrace this reasoning … for sound scientific reasons … isn’t that right Judith Curry?
Conclusion Now that the pause is ending, so is the viability of weak-science cycle-seeking and weak-politics denialism.
And that’s why family-starting voters, conservationists, religious leaders, politicians (even centrist Republican conservatives!) all are accepting that ‘Hansen is right’ … isn’t that plain to *EVERYONE*, Climate Etc readers?
The Common-Sense Question
That Cycle-Seeking Skeptics NEVER Answer For young researchers and young voters, what is a reasonable working estimate of the Bayesian probability that ‘Hansen’s climate-change worldview is correct’?
The world wonders!
‘Now that the pause is ending’
Did I miss something?
Question by Latimer Alder, answers by NOAA/TropicalTidbits/Mother Earth herself!
Apparently you were missing a lot of color.
FOMD,
I generally agree with a majority of your posts (and agree with Hansen’s world-view of AGW, although I do quibble with his 2100 estimates of 4C+ warming), but I think it’s WAY too premature to definitively state that “the ‘pause’ is ending”.
Even if there’s a El Nino this year (all signs are pointing to “yes”, although the intensity is still unknown), I still think that we should wait at least another year or so to determine whether this year’s El Nino kicks-off another round of 0.2C/decade warming. Case in point is 2010: that year’s El Nino, while raising temps for the year, did not have an major impact on overall global temps.
Anyway, keep on fighting the good fight!
“every realistic dynamical model of the climate” – Just like every working perpetual motion machine proves perpetural motion machines are realistic.
Where’s the dunce cap?
If it wasn’t for the lapse rate, the difference from an non-GHG world to the current one would be double 33C. So all of those factors have already been considered in the calculations. The fact that you are screaming that it hasn’t been done is taking us for chumps.
WHT: The dunce cap in on you who believe the Scientific Method is something to be despised and denigrated. You can hold your religious beliefs in your failed climate models and Lysenkoist dogma that facts don’t matter, but it doesn’t make them reality. The truth is that your “climate models” are as accurate as the ‘end is nigh’ of your fellow religious fanatic.
“Why should I make the data available to you, when your aim is to try and find something wrong with it.”
“The two MMs have been after the CRU station data for years. If they ever hear there is a Freedom of Information Act now in the UK, I think I’ll delete the file rather than send to anyone.”
“I can’t see either of these papers being in the next IPCC report. Kevin and I will keep them out somehow – even if we have to redefine what the peer-review literature is !”
Call me when you have proof of your Imminent! Global! Disaster! based on real science and real evidence from people who have real scientific ethics. Until then, I will continue to be amused by your perpetual motion machines…
WebHubTelescope: So all of those factors have already been considered in the calculations. The fact that you are screaming that it hasn’t been done is taking us for chumps.
The standard texts present lapse rate models based on the equilibrium assumption, and have no account of the non-radiative heat transport by dry and wet thermals, where the lapse rate calculations are clearly in error. If you can direct me to calculations of the non-radiative heat transport in the literature I would appreciate it.
My repetitions are in a quiet voice. You direct a lot of pejorative phrases toward the people and points that you can’t answer. How’s that working out for you?
Matthew Marler, for climate, it is the mean that matters. The lapse rate contains the upper warming influence of extra surface moisture in warmer climates. It is the average over long periods, as is needed for the average radiation.
Jim D: for climate, it is the mean that matters.
That is a proposition I regularly challenge. No process is governed by the spatio-temporal mean temp, humidity, etc; calculations based on means are inherently inaccurate, and I have shown how some are biased.
You are prudent, Rational Lukewarmer, and James Hansen’s predictions too are prudently conditioned.
For blogo-sphere purposes (and to shame the skeptics), FOMD’s Bayesian over-under odds are hereby announced as:
• 5-to-2 odds that 2014 and/or 2015 will witness a year-averaged NOAA ‘Land & Ocean Temperature’ record,
• 6-to-1 odds that the energy-balance climate-change worldview is fundamentally correct.
NOTE The main Bayesian prior driving the above odds is a presumption that radiation transport theory and the Four Laws of thermodynamics *both* are correct. These two fundamental-physics priors are further weighted by global observations of ocean heating, sea-level rise, and polar ice-melt … all proceeding without pause or obvious limit.
To *MOST* folks — meaning a strong majority of scientists, mathematicians, conservationists, farmers, young voters, rational conservatives, and religious leaders — the favored horse is easy to pick!
That’s obvious to *EVERYONE*, eh Climate Etc readers?
Skeptics, post your over-under odds and the Bayesian priors that drive them!
Seriously, “spews”? No need to be a condescending jerk; just say you disagree like an educated adult.
I don’t understand what’s so difficult about being cautious about declaring definitively the “pause” is over until another year or so. As I mentioned before, see the 2010 Land and Ocean record temps from that year’s El Nino.
I said that I agree with a majority of what you post and Dr. Hansen’s projection, and I wasn’t being sarcastic.
The word “spews” was inadvertent (and wrong, the word “posts” was intended); this was an error solely on my part, that I hereby acknowledge, and for which I sincerely apologize.
Rational Lukewarmer, please let me say that I appreciate and respect the rationality, the clarity, and the gracious good manners of your comments.
Please keep it up!
A fan of *MORE* discourse: NOTE The main Bayesian prior driving the above odds is a presumption that radiation transport theory and the Four Laws of thermodynamics *both* are correct.
On what grounds do you exclude non-radiative transport and clouds?
An example of an misguided scientific mind. There are only 4 fundamental forms of interaction — gravitational, electromagnetic, strong nuclear, and weak nuclear
http://en.wikipedia.org/wiki/Fundamental_interaction
So the only way that the earth can significantly exchange energy externally is through radiative EM.
So again, please explain the 33C discrepancy without going through radiative theory and GHGs. And the fact that it is ALREADY compensated by lapse rate, etc. You can not count that twice !
WebHubTelescope: So the only way that the earth can significantly exchange energy externally is through radiative EM.
So again, please explain the 33C discrepancy without going through radiative theory and GHGs. And the fact that it is ALREADY compensated by lapse rate, etc. You can not count that twice !
First point yes, but the nonradiative rate of transfer of energy from the surface to the upper troposphere can be increased more than is accounted for in the models.
Second point, the lapse rate is an equilibrium calculation, not accurate to the Earth’s non-equilibrium dynamics. The equilibrium calculations ignore the wet thermals, which do not conform to the calculated lapse rates. I am not ignoring GHGs and radiative transfer, and I am not explaining how the Earth became this warm. I am exploring what the science has to say about what will happen when the CO2 concentration doubles in the future, given the system as it is now.
According to Lacis, the warming offset would be closer to 66C than the 33C it is now due to the presence of an isothermal GHG. But since there is a lapse rate in effect, this value is actually 33C.
It would take an extended analysis on YOUR part to convince someone that another doubling of CO2 isn’t going to make a dent in a further increase from 33C.
Anyone experienced in physics understands that these effects don’t “stop on a dime” just because someone says that “anything could happen”.
In fact that’s how I breezed through 3 college degrees — I answered all my exam questions with my own theory, and then added “Anything could happen!” and all the professors gave me smiley marks and 100/100 scores.
Sarcasm off {….. } Sarcasm back on. OMG, where do they find these people?
WebHubTelescope:Anyone experienced in physics understands that these effects don’t “stop on a dime” just because someone says that “anything could happen”.
I have not written that anything “[stops] on a dime” or that “anything could happen”. I have written that approximations are in error by amounts large enough to vitiate any predictions, and that specific processes documented in the literature might be affected in unknown amounts by CO2 changes. Anybody who has experience in reading understands that.
Well, I can tell you what will actually happen — a TCR of 2C per doubling and ECS of 3C per doubling of atmospheric CO2. This is what the observational evidence says up to now, and no no reason that it won’t continue.
A fan of *MORE* discourse: Judith Curry asks “Can we trust climate models?”
With respect, that is the wrong question Judith Curry!
A better question — for young scientists, family-starting voters, conservationists, religious leaders, and politicians — is this: “What Bayesian evidence can we find to diminish our confidence that ‘James Hansen’s climate-change worldview is fundamentally correct‘?
I have to disagree with your first assertion: a most important question is whether the models can be trusted to guide policy. This is not a debate about whether James Hansen’s world view is “fundamentally correct” — especially when the “fundamental” parts can be chosen after we learn which of his many predictions have turned out wrong. It is a debate about how much scientific evidence is needed to support a government policy of mandating huge divestments from fossil fuels and redirecting them toward more costly ways of generating electricity and transporting and manufacturing good. Even if the GCMs and Hansen’s world view are “fundamentally correct” for some a posteriori selection of “fundamentals”, the forecasts and projections show them to be at best inaccurate and unreliable, and arguably seriously incomplete.
What exactly do you mean by the unusual phrase “Bayesian evidence”? Probably you want a formal statement of priors, likelihoods and posterior distributions, but as Samaniego pointed out in “A comparison of frequentist and Bayesian methods of estimation”, there is generally no good case that a fully derived posterior distribution is a more accurate representation of the shared world than a Fisherian fiducial distribution, because there is seldom a case that the prior distribution is sufficiently accurate. The case for Bayesian inference is that it is more logically coherent than Fisherian fiducial inference, but that case itself rests on dubious assumptions.
I agree with Fan.
In the face of a sudden and unprecedented elevation of CO2 levels to highs not seen for 15 million years and the realization that the Earth has faced ‘hell and high water’ in the distant past, it is incumbent on all to seek out convincing evidence that continued carbon pollution is safe.
Until climate scientists can tighten the models and unless those models then show a tight and robust low climate sensitivity result below 1C, there remains an unacceptably plausible cataclysm that requires mitigation.
I would urge politicians to stay on the ball as the ball is still in court, and growing bigger and more dangerous all the time.
lolwot: I agree with Fan.
Do you know what “Bayesian evidence” is? Do you have reasons for ignoring non-radiative transfer and clouds?
Reader’s Digest translation of lolwot’s comment –
We’re all gonna die!!!
Seriously Matthew R Marler, based strictly on the scientific evidence, what over-under odds would *YOU* recommend to a young climate-science researcher (and voter)?
Numbers good, bafflegab bad. The world awaits a sensible answer!
a fan of *MORE* discourse: Numbers good, bafflegab bad. The world awaits a sensible answer!
Quote some bafflegab.
Seriously Matthew R Marler, based strictly on the scientific evidence, what over-under odds would *YOU* recommend to a young climate-science researcher (and voter)?
Over-under what? Over what time span? For a voter, I would recommend supporting flood control and irrigation enhancements, avoiding large scale wind farms and solar farms.
I think one thing we can do, is characterize by physical measurements the radiative properties of the surface to space under clear skies. Once we know the thermal properties we can compare this to average weather, this can give us a proportion of clear/calm conditions to cloudy conditions.
Accuracy on an annual to decadal timescale, perhaps not so much but who seriously questions the abilities of Western school teachers to model the future 30 to 50 years out? Besides, all we have to do is look out our windows at all the bad weather to know humanity has effected the climate, right?
The warmists have done a magnificent PR job in convincing the world that the models are useful, and it is up to the skeptics to show that they are not.
That is exactly the wrong way round. The models should be assumed to be useless until our hostess and the rest of the warmists have shown that they are useful.
Jim, it may be logically the wrong way round, but it a political fact to be felt with. Nobody said the average voter was mindful of the correct null hypothesis. The pause helps show the models are wrong. Now we add a couple of alternatives, make some predictions that prove the models even more wrong, and we start to get somewhere against the warmunists.
Meanwhile, it helps to elect a Congress that can get CO2 removed as a defined pollutant no matter what the models say. Amend the CAA.
Rud, you write “and we start to get somewhere against the warmunists.”
I agree. And if this were a political blog, then I would say no more. But our hostess is a scientist, and I always hoped she would see what is the correct hypothesis. I still hope so, even though there are no signs she has got the message yet.
Objectively speaking, reality has been saying for last 17 years that global warming does not exist.
The problem is that the models still say it will. Until those are falsified, then politically nullified, warmunists have the momentum.
That’s not true at all, the only metric that does not show warming is RSS, and anyone want to take a guess as to what RSS measures?
Even UAH show warming at 0.94 C per decade since 1997, though uncertainties are high as the period is still to short to reject either warming or cooling.
How far off are the models? 0.3 degrees or so? About the valley to peak of a mid sized El Nino? Record temp for May by GISS mean anything?
A first course computer programmer could model a series of coin flips, but it is impossible to predict the oscillations inherent in the El Nino La Nina, PDO, AMO oscillations very far out into the future.
Judith, if you disagree with the 2 to 4 C temperature rise for the 21st century, what is your position. 0-6 C?
Keep on pushing those tired, old IPCC myths–e.g.,
~Bob Carter (Quadrant Online)
I wasn’t pushing any of those arguments.
I’ll just touch on the Precambrian argument, you can research it yourself, but the Precambrian period had both extreme ice ages and complete absence of ice caps, so any Precambrian arguments go both ways and comparing our current climate to the Precambrian one is fraught with pitfalls.
Better would be to argue what the climate was last time CO2 was at 400 ppm, and what the ice cap situation was then, perhaps you can research it yourself.
I’m so skeptical I doubt most of what even I think is true, let alone computer simulations of highly complex and poorly understood systems like our planet’s climate. I was amused at the use of the metaphor upthread about knowing that sure death resulting from falling from a tall building is a given. Observations tell us that just isn’t true,
Wikipedia has an interesting article about a related phenomenon, the so-called High Rise Syndrome, that shows us even in something supposedly so easy to understand and agree on there are confounding factors that disturb our understanding of the phenomenon.
“Studies done on cats that have fallen from 2 to 32 stories, and still alive when brought to a veterinarian clinic, show that the overall survival rate is 90 percent of those treated.[unreliable source?][2] In a study performed in 1987 it was reported that cats who fall from less than six stories, and are still alive, have greater injuries than cats who fall from higher than six stories.[3][4] It has been proposed that this might happen because cats reach terminal velocity after righting themselves (see below) at about five stories, and after this point they relax, leading to less severe injuries in cats who have fallen from six or more stories.
Be very skeptical, both of those who demand action and of those who say none is needed, When those assertions emerge from controversial, cutting-edge massaging of numerical data clouds be on heightened alert.
I am a little surprised that anyone seriously mentions models and falsifiability in the same breath. Models are and will always be less than theory. Models, even conceptual models, are just tools and their results can never establish a fact, i.e., any model prediction will always be less than a fact. Models are implementations based one of more theories that perform quantitative or qualitative calculations. It is fine and even essential to address a model’s predictive power from all sides, but to talk about any model or model results in terms of its falsifiability is muddles the distinctions between facts, theories, and models. Hmmm, maybe I am not surprised after all.
What ever happened to “all models are wrong”?
Agreed, I personally talk about model ‘fitness for purpose’. I mentioned ‘falsifiability’ since HA had a section on this. In this context, falsifiability is arguably a pseudonym for fitness for purpose
There is a section on falsifiability and that merits bringing it up. ‘Fitness for purpose’ is a much better term/concept here. One certainly should have a better prospect for negotiating a useful agreement on the fitness of a climate model than falsifiability.
Even more important (to me) is correctly framing and describing the issue when the process is so visible to the public with and lasting impacts to both society and science follow. Also there are big differences in terms of meaning and practical implementation between falsifiability and demonstrating ‘fitness for use’.
mwgrant, you write “I am a little surprised that anyone seriously mentions models and falsifiability in the same breath.”
Why? It is often possible to validate models with measured data. Engineers do it routinely.
Hi Jim Cripwell,
“Why?”
The reason is simple. I am surprised to see falsifiability mentioned because as I suggested above it does not have a direct role in the construction or use of models. Models are tools used to explore assertions, hypotheses, and theories [where falsifiability may apply] but should not be confused with or substituted for them. As for your comment
” It is often possible to validate models with measured data.”
Indeed, it is essential to validate a model before any use. However, validation is a very different matter than falsifiability in a number of respects. The one caveat I would put on that is that it just may not be possible to validate some models to specifications consistent with their intended use. This has to be dealt with in a transparent way. Sometimes it may be that only qualitative discussion of the uncertainties is possible. Still the discussion must be fully developed and documented. (Examples here might be structural uncertainties, lack of observation.)
Validation is a process that applies to models and does not uniquely speak to the falsifiability of the underlying theories and facts Falsifiability is a characteristic of more fundamental entities than models–assertions and hypotheses. Validation works on a sliding scale and part of the process is determining where to set the scale. One’s expectation is that if a model ‘reasonably’ captures reality in some regime then that model is valid in that regime. The practical thrust of the validation is to show a specified degree of correctness, i.e., one documents ‘how good is good enough?’ An assertion or hypothesis is called falsifiable “if it is possible to conceive an observation or an argument which proves the statement in question to be false.” [Cut and paste from Wikipedia]. There never is a question of how false.
All of that said–IMO validation is paramount for climate models and ‘true’ falsifiability is of little consequence and really, way out in the weeds, i.e., its mention contributes nothing while thoroughly hindering the needed discussion. Finally I note that validation is clearly a part of a determination of any ‘fitness for purpose’.
Good points mw. “Can We *Trust* Climate Models” represents ignorance of the fundamental purpose of models in general. More appropriate questions should be framed in terms of problem solving.
For example:
What do the models tell us about our current understanding of the climatic system?
What and where do models indicate we need to collect more and/or new types of field data.
What do models tell us about their inadequate characterization of ocean circulation, aerosol interactions, cloud development, etc, etc, etc.
Jim Cripwell: You are right, of course. However, the earth’s climate is not a widget with limited features and degrees of freedom. There is not enough computing power to construct a model to accurately simulate the climate system beyond one week using the current modeling software.
Howard: “Can We *Trust* Climate Models” represents ignorance of the fundamental purpose of models in general.
That may be true, but model outputs are being used to support policy changes. “Can they be trusted for that purpose?” is not an empty question.
Hi Howard,
“More appropriate questions should be framed in terms of problem solving.”
Sharpening that a little…More appropriate questions should be framed in terms of the specific ‘problem’ being addressed. Doing science and making policy impose different roles, constraints and requirements on models that may be employed.
and hi Matthew R Marler.
“Can they be trusted for that purpose?” is not an empty question.”
Maybe the practical question is determining how much the models can be trusted in working thru policy options. In particular we can not dismiss time as a significant factor in all of this and that at some point leads to potentially having to act on policy with incomplete understanding and information. As I’ve said before, that, in theory at least, is why the supposed big kahunas get the big bucks.
We all know it: Everything else remaining the same, adding CO2 should warm the globe, how much nobody knows.
They’ve gotten terribly lost in the thicket of ‘Everything else remaining the same’. We know it doesn’t, so why keep pretending?
=================
If you run your car into a brick wall, all else being equal, there is no damage.
~Bob Tisdale (from his book, “Climate Models Fail”) quoting what a PBS narrator said on the subject, “Is the Discovery of Global Warming Our Greatest Science Achievement?”)
That such an inanity is attributable to a PBS narrator is disappointing, but … sadly, far from surprising.
It would be quite amusing, I suspect, to have this PBS narrator explain to us all how on Gaia’s green earth s/he (or anyone for that matter) can know that any model is “realistic” – not to mention the simulations.
But, more importantly, I’d like to hear from this PBS narrator how s/he arrived at the rather astounding conclusion that “models and simulations themselves create … knowledge” of any kind!
Then again, perhaps – not unlike “trick” and other assorted words which, evidently, take on a whole new meaning in the context of “climate science” – “knowledge” is just one of those unfortunate words which has fallen into similar disrepute.
As even Keith Briffa had acknowledged, as early as 2005:
Create or, er, conjure up climate ‘knowledge?’
Hole lotta pres-tidi-git-ation goin’ on.
What is happening in the sciences demonstrates the kind of people who become academics. There is accountability in a capitalist economy: there is a reckoning; bad ideas go away; and, in the business world, when good money is spent after bad it’s not a problem because it’s not your money that has been wasted without your consent.
Ideological dementia by Wagathon, plain truths by FOMD.
Typical Leftist response… ad hom attack, no substance.
Wag, that is unfair. Many of us do science for the love of exploration, the joy of design and the hope we can make the world a better place.
Christian Schlüchter has awakened to more than the fact back then the Timberline, in the Alps 4,000 and 2000 years ago, was much higher than it is today. Schlüchter is now convinced: Our society is fundamentally dishonest. The promise of universal education has been a failure.
Hi Wagathon
We developed language to let us lie to each other.
“Given the spread of model results at the local scale, the issue is not so much one of falsification, but rather that current models do not provide much of a guide as to future climate change.”
vs.
“Probably the most iconic and influential result arising from climate models is the prediction that, dependent on the rate of increase of CO2 emissions, global and annual mean temperature will rise by around 2–4∘C over the 21st century. We argue that this result is indeed credible, as are the supplementary predictions that the land will on average warm by around 50% more than the oceans, high latitudes more than the tropics, and that the hydrological cycle will generally intensify.”
So the models do not provide much of a guide to future climate change, but they are credible as to predicting future climate with respect to CAGW.
Are you sure Mosher didn’t write this?
I have taken a close look at the arguments made by global warming academics and their supporters. After developing a keen understanding of the thought processes through which these academics resolve issues and develop policies to protect the rest of us from ourselves, it is obvious that these academics believe they’re a lot smarter than the rest of us. I know they look down their effete snob noses at us but we’re so lucky to have them running our economy and doing all of our thinking for us.
Sensitive dependence and structural instability are humbling twin properties for chaotic dynamical systems, indicating limits about which kinds of questions are theoretically answerable. They echo other famous limitations on scientist’s expectations, namely the undecidability of some propositions within axiomatic mathematical systems (Gödel’s theorem) and the uncomputability of some algorithms due to excessive size of the calculation. http://www.pnas.org/content/104/21/8709.full
This is the trajectory of feasible solutions for a model.
http://rsta.royalsocietypublishing.org/content/369/1956/4751/F2.expansion.html
Pick one arbitrarily and pretend it represents the future.
The other ‘plausibility criteria’ of McWilliams is the representation of physical processes.
The researchers used a climate model, a so-called coupled ocean-atmosphere model, which they forced with the observed wind data of the last decades. For the abrupt changes during the 1970s and 1990s they calculated predictions which began a few months prior to the beginning of the observed climate shifts. The average of all predictions for both abrupt changes shows good agreement with the observed climate development in the Pacific.
“The winds change the ocean currents which in turn affect the climate. In our study, we were able to identify and realistically reproduce the key processes for the two abrupt climate shifts,” says Prof. Latif. “We have taken a major step forward in terms of short-term climate forecasting, especially with regard to the development of global warming. However, we are still miles away from any reliable answers to the question whether the coming winter in Germany will be rather warm or cold.” Prof. Latif cautions against too much optimism regarding short-term regional climate predictions: “Since the reliability of those predictions is still at about 50%, you might as well flip a coin.” http://www.sciencedaily.com/releases/2013/08/130822105042.htm
It seems an epic fail all round. My brand new blog – even my Daisy has a website these days – has an expanded and rewritten version of something posted here a few weeks ago. This seems to be a fundamental reason why climate fails to behave like models.
‘In a scientific problem as potentially complicated as climate, there is another modeling practice that is increasingly important: AOS models are open-ended in their scope for including and dynamically coupling different physical, chemical, biological, and even societal processes.
The rationales for coupling are to investigate potentially significant feedbacks (e.g., radiative properties for different airborne crystalline ice structures, changes in air and water inertia due to suspended dust and sediments, and water and other material exchanges with plants and biome evolution) and to achieve ever fuller depictions of Earth’s fluid envelope. Besides adding to the overall complexity of AOS models, coupling increases the number of processes with a nonfundamental representation (i.e., similar to a parameterization), because, for the most part, the governing equations are not well determined for the model components other than fluid dynamics. When adding a new coupling link, there is no a priori guarantee of seeing only modest consequences in the AOS solution behavior…
For many purposes that are well demonstrated with present practices, AOS models are very useful even without the necessity of carefully determining their precision compared with nature. These models are structurally unstable in various ways that are not yet well explored, and this implies a level of irreducible imprecision in their answers that is not yet well estimated. Their value as scientific tools is undeniable, and the theoretical limitations in their precision can become better understood even as their plausibility and practical utility continue to improve. Whether or not the irreducible imprecision proves to be a substantial fraction of present AOS discrepancies with nature, it seems imperative to determine what the magnitude of this type of imprecision is.’ op.cit
We seem from my perspective to be still talking about the wrong issue. The model issues all seem so obvious – but essentially not germane to the central social, economic and environmental realities.
Today I am about to put some effort into a policy initiative that seems far less sexy for some reason than these endless rounds of assertion and counter assertion – but that seems the practical way forward. It is the based on the Copenhagen Consensus analysis of the Millennium Development Goals. I will post it on my site – and see if Judy wants it.
As such, the climate sensitivity to a doubling of CO2 has only a very limited meaning because it varies with altitude. Models are fundamentally erroneous as based on the independence of the temperature gradient relative to radiative phenomena.
This curious premise has no theoretical basis and is (mis) justified, sometimes the relationship is too difficult to solve, sometimes by reference to a ghost blended adiabatic lapse rate that no one knows how to calculate and even define.
For me an interesting historical question is what scientific basis would have lead Hansen for example to suppose that GCM’s applied to climate would be worth the huge investment. I suspect the answer is that those who got this ball rolling weren’t very well trained in fluid dynamics. I also suspect this was a very controversial thing from the beginning.
What a lovely thread. I’d not heard the name Hawkmoth Effect before although the concept is familiar as I’ve referred to it often in comments. I don’t hold out much hope that Bayesian statistical models will help anymore than dynamical models for the same reason, mathematical chaos.
What Bayesian prior is there for cloud cover and cloud albedo variations? What prior Bayesian is there for biosphere response to increased CO2?
Just two examples.
One can make up all sort of best guess assumptions for dynamical models and you’ll have to make the same sort of prior assumptions for the Bayesian models. But none of them will lead to long term trustable regional predictions that will make a precautionary principle approach to climate related disasters cost effective.
We already know what to do, don’t build on flood plains without suitable flood relief systems, don’t build close to the sea and cliffs without suitable anti erosion constructs, learn the lessons of the 3 little pigs else the big bad wolf will blow your house down. If effort was put into implementing these lessons instead of spending billions tilting with windmills, a great deal more good will be achieved at relatively little extra cost.
Mosher and other self-styled lukewarmers:
Regardless of what physics tells us CO2 should warm the planet there are no physics at all to explain the expected heat rise disappearing into the deep ocean without first being detected in the top 700 metres. If you are happy to violate every law of thermodynamics then don’t seek to lecture the rest of us about simple physics!
The trouble with too many people is that they believe precisely what they want to believe. Well some of us just trust reality! Nature very often does the opposite of what is expected of it. That’s why we have to test and test again. And if the experiment of adding a very large slug (+50%) of CO2 to the biosphere results in a temperature plateau then you should accept that your simplistic linear, two-variable theory of climate is incorrect and that the real climate driver is still most likely the sun.- as has been the consensus among learned people since time immemorial.
The climate models give the outputs reflecting the assumptions made in the inputs. Wrong results means wrong assumptions: Natural variability is not in decline – as skeptics have been trying to tell the climate clique all along.
“Now to the most amusing attempt of “climate skeptics” to wish these scientific results away. Their argument goes like this: It is not possible that warming of the deep ocean accelerates at the same time as warming of the upper ocean slows down, because the heat must pass through the upper layer to reach the depths. A German journalist put it this way:
“Winds can do a lot, but can they beam warm surface waters heated by carbon dioxide 700 meters further down?”
This argument reveals once again the shocking lack of understanding of basic physics in “climate skeptic” circles. First the alleged problem is lacking any factual basis – after all, in the last decades the upper layer of the oceans has warmed faster than the deeper (even if recently not quite as fast as before). What is the problem with the heat first warming the upper layer before it penetrates deeper? That is entirely as expected.”
– See more at: http://www.realclimate.org/index.php/archives/2013/09/what-ocean-heating-reveals-about-global-warming/#sthash.tDOD7Hto.dpuf
“And if the experiment of adding a very large slug (+50%) of CO2 to the biosphere results in a temperature plateau then you should accept that your simplistic linear, two-variable theory of climate is incorrect and that the real climate driver is still most likely the sun.”
In the last 20 years sunspots have plummeted to early 1900 century levels. How much cooling do you think that caused?
Let me put some words in your mouth seeing as you think the Sun is driver. It’s a big drop so I reckon you think since 2000 the drop in solar activity has caused 0.3C cooling.
So why has global temperature not fallen 0.3?
Answer: CO2. Continued and accelerating warming from CO2 over just 10 years is capable of completely overwhelming what should be a cooling back to 1950s levels.
You are not being scientific at all! Realclimate.org had told us to expect a lot of warming by now – it’s right there in their own archives. A little humility because reality did not match their models it seems is beyond them!
If you actually believed in classical physics then you would not hide from the results of the experiment. ie If you add CO2 in large amounts and then predict large temperature rises from it (because man was supposed to be dominating climate) you have to recognise that if it didn’t happen your theory is invalid. Man does not dominate the climate! It really is that simple.
And of course you should know, because Judith has told you often enough, that there is not enough data to say that the deep ocean is warming or cooling and even if the ocean is absorbing it then basic physics also says that it is never coming back! Note that I didn’t say it was impossible – I was reminding you that sometimes nature does the opposite of what basic physics predicts and basic physics undeniably says that the deep ocean cannot accept warmth without it being detected in the top 700m. Now ok If you want to consider more complex thermodynamics that does the opposite of the basics then do not seek to lecture the rest of us on sticking to the basics. That is the only lesson I am trying to get through to any scientific thinkers!
In any event any postulated ocean uptake is merely natural variation by another name (or unexpected negative feedback if you prefer) and is yet another thing that tells us climate physics is not so simple as has been previously assumed. Further, I don’t believe Gavin believes the ocean warming hypothesis anyway – certainly his old boss wanted to push aerosols as the cause for the pause.
Finally, there is actually a very good correlation between Solanki’s solar reconstruction and temperatures in the Arctic and the US48; the only two datasets we know for sure have been either free of, or corrected for, the urban heat island effect. Seek and ye shall find!
About “annual mean temperature will rise by around 2–4∘C over the 21st century. ”
JC says “They do not justify this statement in the main text”
I would say that while not fleshed out, the justification for that statement might be summarized as ‘multiple lines of evidence. That is the justification doesn’t come from any work on models itself (validation, falsification or’fit for purpose) but from the fact that the climate models are in agreement with other sources of evidence completely outside the modelling field.
Even A&H can’t bring themselves to discount the upper end of the project warming from climate models (4C warming in 86 years? 0.45C per decade? That is just nuts!).
If even the most reasonable among consensus climate scientists can’t deal effectively with evolving reality (and A&H are a lot more reasonable than most), then it seems to me the entire field risks becoming irrelevant, at least as far as public policy goes….. and maybe scientifically as well. They certainly are making themselves an easy and well deserving target for budget cuts. Where are the climate scientists who will accept reality and state the obvious: the models appear much too sensitive to radiative forcing. I am shocked that reality seems to have so little influence on the field.
“Even A&H can’t bring themselves to discount the upper end of the project warming from climate models”
How can you? Where has this sense of certainty come from?
“0.45C per decade? That is just nuts!”
So is < 1C climate sensitivity. Doesn't stop many of you from taking it seriously though.
The certainty comes from observations, not models.
Some may think 1C per doubling is plausible, but I think they are mistaken, just as I think those who believe 5C per doubling is plausible are mistaken. The observational evidence is most consistent with an ECS a bit under 2C and a transient sensitivity near 1.3C per doubling. Anything over 3C ECS is very unlikely, as is anything under 1.4.
The certainty comes from observations, not models.
And what of infuence derived by observing natural variabilty?
no, the observational evidence is most consistent with an ECS of 3C and a transient sensitivity around 2C per doubling.
Those who undershoot that by a lot are the same double counting white-collar criminals who keep two sets of books. Who knows where they are coming up with those numbers, but they are.
WebHubTelescope: no, the observational evidence is most consistent with an ECS of 3C and a transient sensitivity around 2C per doubling.
If the CO2 concentration doubles from its present concentration by 2150 (which it will according to estimates of the rate of increase), how much temperature increase due to CO2 will there be by 2150? Your model projects an increase of 2C, due to CO2, by that time. How much more time will elapse before the temperature increases by an additional 0.9C?
Warming since 1850 is about 0.85C; assuming all of that warming is due to rising GHG’s places an upper bound on the GHG driven warming effect. The IPCC estimated GHG forcing (all types of GHG’s) is about 3.1 watts per sq meter. Ocean heat uptake (Argo) plus ice melt and land heat uptake estimates is about 0.6 watt per sq meter. The IPCC best estimate for aerosol effects is 0.9 watt per sq meter. So the extra heat currently being lost to space due to an average surface temperature increase of 0.85 C is about (3.1 – 0.6 – 0.9) = 1.6 watts per sq meter. The sensitivity is then: 0.85/1.6 =0.52 C/Watt/M^2, or 1.97C per doubling (a doubling of CO2 is about 3.71 Watts per sq meter of forcing). The greatest uncertainty is in aerosol effects, both direct and indirect. Climate modelers bring their temperature hindcasts in line with historical temperature records by assuming high aerosol effects (eg, GISS modelers assume that half of all historical GHG forcing was offset by aerosols). High sensitivity is simply inconsistent with the IPCC’s best estimate of aerosol effects. It is a glaring inconsistency which appears to be studiously ignored. Reality will not be forever denied.
Rud can’t be wrong. He already wrote it up for his next book, and books are never wrong, right?
And to what Fitz said:
which is right on the mark for the observational evidence of TCR.
See how skeptics can close the consensus gap?
“The sensitivity is then: 0.85/1.6 =0.52 C/Watt/M^2, or 1.97C per doubling (a doubling of CO2 is about 3.71 Watts per sq meter of forcing).”
—–
This is a nifty bit of “back of the napkin” math which I don’t disagree with in principle, though there are some issues with the set of assumptions , namely:
1) that the response will be linear, without jumps or feedbacks causing nonlinearities. This can cut both ways, to dampen or accelerate warming, but with such a rapid introduction of GH gases, nonlinearities should be expected. The paleoclimate data would still seem to push sensitivity toward the 3C side of things, and this would favor positive feedbacks providing a nonlinear response. Dragon-kings and other nonlinearities seem to be indicated by paleoclimate data to take us from the 2C to the 3C or over range.
2) the real amount of deeper ocean warming. ARGO doesn’t yet measure this (though floats down to 6000m are planned). If actually even more heat is being stored at abyssal levels (as some research indicates), this would alter your initial math. This is heat that is not yet being released to the atmosphere, but is still being accumulated in the climate system,
Curry: And finally, someone needs to get serious and discuss alternative climate model structural forms – not just adding more chemistry to the models, but rethinking the structural form of the dynamical core.
Here is the fodder for alternatives…
1) the OAGCMs (see Fig 1C blue curve – heat content derived from volcanic forcing) http://www.image.ucar.edu/idag/Papers/Gleckler_Krakatoa.pdf (h/t to Gates for the link)
According to the pooled model averages, heat uptake lost through volcanic forcing in the ocean persists undisturbed for at least 80 years. This is a modeled property. In simple terms the model indicates that heat going into the ocean stays there for more than 80 years. On might even argue that Krakatoa 1883 is the root cause of the current pause.
Regardless, the latency of heat uptake and return between atmosphere and ocean emphasizes the role of OGCM in the OAGCM
2) water is 8x stronger a GHG than CO2. It also produces positive feedback but unlike CO2 condenses. Turbulance and variations in the atmospheric stucture and lateral distribution of water vapor could overwrite any constant minor positive feedback from the lesser CO2 GHG.
Condensing water is the main GHG by a factor of 8. Weather and climate is set by the evaporation, condensation, distribution and advection of this principal component. The temperature distributions are to be found by the advecting and fluctuating patterns of water.
Repeat: Water is the main component. You need your models to describe this accurately before you can put confidence in the outcome of the small CO2 asymmetrical bias. …. Why? The fluctuations in the air structure and and water distribution etc *is* the reason we have weather and climate
Unless the ”NORMAL” laws of physics are acknowledged, all the GLOBAL warming ”predictions” are misleading crap.
unless is acknowledged that: oxygen &nitrogen / the horizontal & vertical winds are cooling and regulating the overall planet’s temp – all ”models” are a misleading crap.
some places those winds are cooling from midday to midnight / in 12h by 7C, other PLACES as in the desert by 25C, in 12h – talking that those same winds cannot cool 2C extra in 100y, is the biggest lie since the homo-erectus invented language: http://globalwarmingdenier.wordpress.com/climate/
Can we trust climate models? We can certainly trust them to consume a lot of public funds, and we can be certain they will continue to diagnose high sensitivity, at least until major funding cut are threatened. But trust them to make policy relevant projections about future warming? Not even a little bit; they are worthless WRT public policy.
Can climate models be trusted?
Not by me, but I would have a lot MORE trust in their long term relevance if someone could convince me that the modelers were provided with a list of:
a. all factors affecting the climate.
b. the relative importance of their influence.
c. the sign of their influence.
d. a method of predicting how each factor that affects the climate will vary over the time span for which the model is to be run.
e. how variations in any one factor affects the magnitude and sign of other factors.
I DON’T have any confidence in models based on the premise that atmospheric CO2 is the dominant influence on the temperature of the earth nor on their prediction that anthropogenic CO2, at any concentration that we are capable of producing, can cause the overall temperature of the earth to rise catastrophically when the slope of the temperature of the earth trend line has varied in both sign and magnitude during the period for which we have CO2 data, while CO2 has risen monotonically with a curve that is concave upward.
As kim pointed out above, when models are based on the premise of ‘if all else remains equal’, when all else demonstrably does NOT remain equal, they are useless. Or, considering what the models are being used for, rather than being useless, they pose an active threat far more dangerous than any change to the ‘Temperature of the Earth’ caused by adding anthropogenic CO2 to the atmosphere.
Ain’t you never heard of analysis-paralysis, Matthew R. Marler?
The questions asked are Fermi Questions. Which you are welcomd to practice here!
Keep in mind Matthew R Marler … young researchers have plenty of respect for a well-informed explicitly numerical Fermi Estimate … ideological bafflegab not so much!
a fan of *MORE* discourse: • 5-to-2 odds that 2014 and/or 2015 will witness a year-averaged NOAA ‘Land & Ocean Temperature’ record,
• 6-to-1 odds that the energy-balance climate-change worldview is fundamentally correct.
I declined the first bet. I bet that the yearly trend of 2014 would be indistinguishable from 0 (i.e, 0 will be close to the middle of the 95% confidence interval.) I’d bet the same for the 2 years, jan 2014 to Dec 2015.
For the second, you have some vaguenss: which details of the science do you include/exclude from the fundamental energy-balance climate-change? Anything lacking an accurate model for cloud cover changes I would say is not fundamentally correct.
I wrote out a related “bet” the other day:In the meantime, I’d summarize my current opinion by betting that from Jan 2014 to Dec 2023, the slope of a linear least squares fit will be less than 0.05C per decade.
Why?
The recent trend is nearly flat, with no reason to expect an upturn.
The Tsonis et al model does not have much upturn very soon.
The ENSO predictions do not look like they forecast a high peak or strong El Nino.
I think that “climate sensitivity to a doubling of CO2″ has been over-estimated.
On the whole, I think increases in summertime daytime cloud cover are more likely than decreases.
Whatever the effect of short-term solar variation, we look to be in for reduced warming for a decade, rather than increased.
I don’t expect much decrease in aerosols of Asian industrial origin, or generally.
Meanwhile, any examples of “bafflegab”? I doubt that “the world wonders”, but maybe it all baffled you.
It’s not quite clear what you mean by “a yearly trend” Matthew R Marler.
How would May 2014 is the hottest on record count against that prediction?
The world wonders!
fan: It’s not quite clear what you mean by “a yearly trend” Matthew R Marler.
The slope of the least squares regression line over the full year.
So actual +/-56% actual is a good guess is it FOMD.
From the population of the US I can state that based on average, you have one testicle and one ovary.
a fan of *MORE* discourse: Ain’t you never heard of analysis-paralysis, Matthew R. Marler?
What paralysis do you refer to here? I never advocate standing pat. I just think your bets are poorly formulated.
If we don’t know to practical reliability the climate sensitivity to water, it is dubious that we can know the climate sensitivity to CO2 better.
Raving, ”climate sensitivity” is well known, for the people interested in the truth: look at Brazil and compare it with Sahara – if you don’t know what is good climate, ask the trees. One oak-tree has much more knowledge about the climate than ALL the commentators here combined…
re: the authors say, “annual mean temperature will rise by around 2–4∘C over the 21st century. We argue that this result is indeed credible” [emphasis added]
It may seem like a quibble, but for any intellectual or academic output, even as an “opinion” piece, to claim that you “argue” for a point means much more than assert, declare, opine, and/or bloviate.
The claim that there is an argument, in an academic context, should mean that there are logically linked, rigorous premises and conclusion displaying the relevant data and analysis, at least in summary terms. To fall short of any standard of rigorous argument really turns it into “he said, she said” without a rational claim upon a reader’s attention.
Water is 8 time more potent a GHG than CO2. Moreover it condenses and is a fast transient. Water creates a positive feedback. The time space and flow transients across the whole planet make the climate and weather.
This story is mainly about water.
If a person is arguing for CAGW the emphasis in put on CO2 … A trace gas with mild (1/8 of water) activity.
It is easy for the argument to mislead
If you think water vapor is the most important greenhouse gas you have the dog wagging the tail.
CO2 is the most important greenhouse gas. Without CO2 the Earth would be too cold to support all the water vapor it does.
1) The Earth has never been without CO2-equivalent, in fact the original atmosphere was mostly methane.
2) Water didn’t exactly arrive on the planet’s surface in it’s frozen form
3) The Earth’s surface was originally molten rock
Ergo, the Earth never needed CO2 to arrive in order to warm up sufficiently to support water vapour.
what astounding logic
…to which you have no answer
how does one answer gibberish?
Question: how does one answer gibberish? ….
Answer: … By quoting NASA and A. Lacis
Although carbon dioxide gets most of the bad publicity these days as the critical greenhouse gas, the warming effect of carbon dioxide is minuscule compared to that of water vapor http://earthobservatory.nasa.gov/Features/Iris/
The significance of the non-condensing GHGs is that once they have been injected into the atmosphere, they remain there virtually indefinitely because they do not condense and precipitate from the atmosphere, their chemical removal time ranging from decades to millenia. http://pubs.giss.nasa.gov/abs/la06400p.html
@ Skiphil
“It may seem like a quibble, but for any intellectual or academic output, even as an “opinion” piece, to claim that you “argue” for a point means much more than assert, declare, opine, and/or bloviate.”
Rather than quibble, you have in fact reached the heart of the matter. Progressivism–and you will note that CAGW is entirely the child of progressives–ALWAYS, on EVERY subject and not just CAGW, operate according to the linguistic philosophy of Humpty Dumpty:
““When I use a word,’ Humpty Dumpty said in rather a scornful tone, ‘it means just what I choose it to mean — neither more nor less.’
’The question is,’ said Alice, ‘whether you can make words mean so many different things.’
’The question is,’ said Humpty Dumpty, ‘which is to be master — that’s all.””
A simple example is the ‘Privacy Notice’ forms that you have to sign every time you visit your doctor, which assures you that your medical records will be held in strict confidence, releasable only to you, those you specifically designate, and others ‘as required by law’. What it means in practice is that your medical history is now a public document, available to any government department that decides that it wants to see it, without a warrant and without anyone notifying you that it was accessed or for what purpose it was accessed. ‘Private information’ now means ‘available to the government on demand’. Mr. Dumpty would be proud.
Wow , what a nice thought provoking thread. Thanks all!
It seems many seem to know the role of clouds and cloud formation in climate with much certainty.
Unfortunately, you really don’t. Hence, the physics misconception of CO2’s role in a dynamic environment. This is not a controlled lab environment Mosh.
Good news is I finally figured out who FOMD really was. Analytics work.
That certainly puts things in perspective.
Do tell. Hansen?
I don’t think this is too off topic
Scientist who faked AIDS research indicted
http://www.usatoday.com/story/news/nation/2014/06/19/fake-aids-research/10899589/
This ain’t too off-topic either, eh DocMartyn … May 2014 is the hottest on record.
The FOMD over-under Fermi Estimate is looking pretty solid!
Unless of course climate-science is all a gigantic conspiracy … as the increasingly strong science is forcing increasingly many die-hard denialists to proclaim!
Prediction Increasingly hot planet; increasingly fervent conspiracy theories.
*BOTH* trends are obvious to *EVERYONE*, eh Climate Etc readers?
There is only so much fraud that one can take.
Here is a representative image of ‘temperature’ changes in the Continental USA.
http://assets.inhabitat.com/wp-content/blogs.dir/1/files/2012/06/Continental-US-state-warming-e1339665745108.jpg
Note the heating of New Mexico, almost 2 degrees F.
You FOMD state that this May was the hottest May in US history.
I looked at Roswell, as that is a local that has little rain, I chose
ROSWELL INDUSTRIAL AIR PARK, NM US
http://www.ncdc.noaa.gov/cdo-web/datasets/GHCND/stations/GHCND:USW00023009/detail
I compared May 1950 and May 2014. Using a t-Test, Two-Sample Assuming Equal Variances, I compared Tmax (p=0.853), Tmin (p=0.577) and Tave(p=0.9).
The chances of me picking an unchanged station that missed a greater than 1 degree F temperature rise in New Mexico is essentially zero.
I do not believe the ‘official’ numbers any more as they are fudged to the point of fakery.
FoMD Prediction > … increasingly fervent conspiracy theories.
These would be the ludicrous conspiracies claming that state-funded climate scientists are not working in the interest of their paymaster, but are instead secretly rebelling and doing honest science.
I’ve seen comments from Robert Brown claiming that when you look at SINGLE runs of GCMs their qualitative properties are very different from the actual climate, e.g. they show much more interannual variability than the actual climate. His argument is that the failure of single runs to “look like” the actual climate–not in terms of mean temperature, but of spatiotemporal statistics–casts doubt on the credibility of their underlying mimetic properties. Assuming that his factual claim is true, that strikes me as a compelling argument, since the physics touted as causally explaining the model runs should not require averaging to match these qualitative features.
State-of-the-art fluid dynamics climate models are showing that the climate is deterministic, set by rather periodic forcing perturbations:
http://contextearth.com/2014/05/27/the-soim-differential-equation/
This is bad news for the skeptics, who were hoping for more uncertainty to feed their pet monster.
Stevepostrel, RGB is quite correct. There are some recent non paywalled papers that look at the main GCMs single run outputs compared to things like actual temperature and actual ice in clouds (important because ice is transparent to SW but opaque to LW, so ice in clouds warms. None match observation. Absolute temps to high to too low by 2-3C. Means dew points, ice formation, everything is just off. Clouds are even worse (of course). The temp error are hidden in anomalies. Cloud errors are compensated by tuning. Wrote an essay to the next book that graphically illustrates these points.
Rud,
You are mistaken. Ice particles in clouds scatter visible light in much the same way as do liquid water droplets. Both water and ice are strong absorbers in the infrared wavelength range. There are smallish differences in the optical properties, but small ice particles (think snow) scatter visible light very efficiently. Clouds that consist of ice particles most certainly are not transparent to visible (SW) light; they in fact look much like, well, like clouds.
There is a lot of non-model evidence for 2-4 C too. You only have to look as far as Lovejoy or Muller or Bengtsson and Schwartz to find observation-derived estimates that support that range. It is wrong to attribute this value to models and ignore the observational evidence. The simple change from 1950 to now of temperature and CO2 gives 2 C per doubling as a transient value, which would be a lower bound on the equilibrium value. 2 C per doubling predicts a 0.7 C rise since 1950, and the rise since 1950 is actually 0.7 C (pause and all).
Then again the actual change from 1944 – prior to decadal cooling – to 2013 is 0.34 degrees C.
Spose he’d convinced himself there?
======
Funny then that the IPCC places the lower bound at 1.5C per doubling. Maybe they are considering evidence that supports a lower plausible sensitivity, evidence which you are apparently ignoring.
You only need to adjust the lower bound down from 2 C if you don’t consider that natural variability like the PDO or sun could figure into the recent lull. Studies that ignore these factors, while being too simple, also give numbers that some people like to quote. Even then, 1.5 C sensitivities fall far short of the warming that has occurred and imply a net positive natural variability since 1950, which it is hard to find an excuse for. 2 C is the number you get for transient sensitivity with zero net natural variability since 1950. Some might consider this a lower bound since aerosols have increased globally and solar forcing has decreased towards the end.
Jim D,
You apparently did not see my calculation above showing the best estimated equilibrium sensitivity (usually called “the effective sensitivity”) is a bit under 2C per doubling.
Natural variability cuts both ways, and the pattern of warming over the past 100+ years is consistent with cyclical increases in the rate of warming from about 1915 to 1945 and from about 1975 to about 2005, and cyclical decreases in warming from 1945 to 1975 as well as 2005 to present…. if this pattern continues, then the rate of warming will likely be reduced for the next 20 years or so relative to what it would have been in the absence of cyclical contributions. Which is not to say that anyone knows for certain the contribution from long term internal variability, only that explaining the slow recent warming with natural variability automatically calls into question how much of the relatively rapid warming between 1975 and 2005 was due to the flip side of that natural variability.
Which is why I think a simple energy balance calculation, like I did above, is a more rigorous approach. Some will (of course) invoke ‘missing heat in the deep ocean’ (for which there is virtually no evidence) and/or greater sensitivity in the future (non-linear sensitivity), in spite of the nearly linear response of most GCM’s, and insist that sensitivity must be higher than 2C per doubling. I would suggest the evidence does not support much higher sensitivity. A good argument can be made for both modest climate sensitivity AND a measured public energy policy which reflects that sensitivity. But I have seen no good argument for very high sensitivity and the need for drastic public policies to eliminate fossil fuel use.
We will only be able to properly assess the models and their underlying “CO2 Control-knob” hypothesis, once we can reliably measure the balance of radiation in and out of the earth system in absolute energy terms.
As things stand, we are still nowhere near achieving that.
The next-best thing would be robust data on total ocean heat. And again, that is still a long way off.
The only robust data we do have is surface temperature, which is but a small part of the picture, and right now flies in the face of the models.
We cannot trust the models as they have smoothed out the multidecadal oscillation. They have replaced the whole of the instrumental record by two lines. One line goes until 1965 at a warming rate of 0.06 deg C per decade and the other from 1965 onwards at a warming rate of about 0.18 deg C per decade as shown here:
http://bit.ly/1nqO6YC
Where is the multidecadal oscillation?
As the climate models don’t represent the observed multidecadal oscillation of the global mean temperature, they are useless.
Welcome back to CE Girma.
My opinion about GCMs: They are not intended to model exactly the earth’s climate. They try to simulate a climate with all dynamics equations and properties of earth’s surface and atmosphere. When looking at output, it is warm where it should be warm, rainy where on earth there is a lot of rain. Clouds occur in reasonable patterns, the patterns of the jetstream look similar to what we see on earth.
But it is not earth. It is a simulation. When the properties of this model, say CO2, are doubled over time, the average temperature has changed by one degree when the model runs over 100 years. So, scientists think, hmmm…this is a lot like earth. Maybe this will happen on earth too and we should be worried.
I see no reason to distrust the models for what they are. I do support new approaches to models because there is this nagging worry in the back of my mind…the CO2 is going up faster than it was put in the ground.
@rmdobservations
It would be nice if this was true, but models are just not that good.
GISS 1
GISS 2
Curry, you say: “And finally, someone needs to get serious and discuss alternative climate model structural forms – not just adding more chemistry to the models, but rethinking the structural form of the dynamical core. Otherwise, we are spinning some very expensive wheels and potentially misleading decision makers.”
Why not simply scrap the whole ‘model’ thing and go back to actually looking at real-world data?
The CO2 scare lies ONLY in the models, not in the real world. That’s why this is perhaps your most pertinent point, Curry:
“Hargreaves and Annan say 100 years for ‘falsification’, and then dismiss the idea since we can’t wait in terms of decision making.”
We all see what they’re doing and why they’re doing it … Let’s end this farce.
rmdobservations says, June 21, 2014 at 5:56 am:
“It is a simulation. When the properties of this model, say CO2, are doubled over time, the average temperature has changed by one degree when the model runs over 100 years. So, scientists think, hmmm…this is a lot like earth. Maybe this will happen on earth too and we should be worried.”
That’s the whole problem. It’s NOT a lot like earth. It’s just a result of the modellers’ OPINION about how the world SHOULD work. It’s completely disconnected from the real world. It’s not science. It’s pseudo-science.
Jo Nova has the new David Evans notch filter model which is worth a look if you want something new it looks very interesting and with David’s background brings a different perspective, Tallbloke also has some interesting perspectives on Evan’s work.
This looks very interesting. One strong point is that it is derived from measurements. Actual data. Not estimations.
I’m at bit amused that those who continually call for skeptics to come up with their own model aren’t over at Jo Nova’s place asking questions. Just sayin’.
He hasn’t yet revealed his trick for getting all the century’s net warming out of his collection of filters and delays when the underlying solar signal is fairly flat. Maybe that is in the next part.
Climate models are an agglomeration of measurements (climate data), approximations (too large grid cells), estimations (parameters), and calculations. Not ready for prime time.
OT, but here’s something for those who continually call others conspiracy theorists. Communists ARE behind environmental groups.
From the article:
…
Russian agents are secretly working with environmental campaigners to halt fracking operations in the UK and the rest of Europe, the head of Nato warned yesterday.
Vladimir Putin’s government has ‘engaged actively’ with green groups and protesters in a sophisticated operation aimed at maintaining Europe’s reliance on energy exports from Moscow, said Nato Secretary-General Anders Fogh Rasmussen.
He said the Russians had mounted a highly developed disinformation campaign to undermine attempts to exploit alternative energy sources such as shale gas.
Moves to start fracking in the UK have been disrupted following a sustained campaign by environmentalists that has created fears about its impact.
…
http://www.dailymail.co.uk/news/article-2663075/Putin-plotting-halt-UK-fracking-warns-Nato-chief-Secretary-General-claims-agents-working-campaigners-make-sure-Europe-continues-reliance-Russian-energy-exports.html
We can take an example from another field to illustrate the problem. In mechanics, the breaking point for vertical loadings or bending of many materials have been determined experimentally. “Simple” physics one might assume, but if we try to extrapolate these knowns to earthquake prediction, we get nothing, no predictability, just some general concepts about how different types of quake zones may be behaving. This is partly due to the unknowable complexity of the rock structures, and partly the unknowability of the exact equations to use. I submit that the problem is similar for the ocean/atmos dynamics. Just because fluid flow is a tractable problem sometimes (nice uniform conditions, a soliton moving up a canal) does not mean it is so universally.
@Steven Mosher | June 21, 2014 at 12:15 am | said:
Jim2
The basis is physics. Watch Linden last talk
For the details.
Our best physics says more co2 will warm the planet
Not cool it.
Pretty simple.
(end of quote)
I take exception to that. A small part of basic physics implies that more CO2 will heat the planet. But it’s far from the entire planetary story. It is entirely possible it won’t heat the planet at all because more is in play than Earthly influences.
At any rate, I will give you that it will probably heat the planet some probably small amount.
Is there a note of irony in the title? Certainly the motto for the Church of Climatology could easily be “In GCMs We Trust”.
Some would consider that the raw data over the last half century supports the models. Others prefer not to think about what the raw data is telling them.
http://www.woodfortrees.org/plot/hadcrut4gl/from:1950/mean:12/plot/esrl-co2/from/scale:0.01/offset:-3.3/plot/hadcrut4gl/from:1970/trend
Jim D > Some would consider that the raw data over the last half century supports the models. Others prefer not to think about what the raw data is telling them.
There’s name for them : Pause Deniers.
@Jim D,
Except that is processed and homogenized data which is far from raw.
jimd
but if we go back seven times as far as that, the raw data suggests something else.
http://wattsupwiththat.com/2013/05/08/the-curious-case-of-rising-co2-and-falling-temperatures/
go back another 500 years and the present warm period looks even less remarkable
tonyb
The effective temperature of the Earth can be calculated from fundamental principles.
(4πR^2 ) σTe^4 = (1 – a) πR^2So – solve for Te – the surface temnperature.
R is the planetary radius – So the solar constant, a albedo and σ the Boltzmann constant.
This can be compared with a measured mean surface temperature and the difference is some 33 degrees C.
This represents a balance between radiative warming and convective cooling. In the real world there are many things that can’t be captured by a static equation. Solar irradiance changes with effects propagating through the system in changes in ocean and atmospheric circulation. These change albedo and convection. One of the ways is through changing rainfall regimes influencing convection, vegetation cover, snow and ice cover, wildfire and dust.
It is regime like structures that is the key idea in hydrology and climate – and not the simple energy balance.
‘The fractionally dimensioned space occupied by the trajectories of the solutions of these nonlinear equations became known as the Lorenz attractor (figure 1), which suggests that nonlinear systems, such as the atmosphere, may exhibit regime-like structures that are, although fully deterministic, subject to abrupt and seemingly random change.’
http://rsta.royalsocietypublishing.org/content/369/1956/4751.full
Work that out with a pencil.
Robertson asked if the IPCC’s assessment of natural variability was adequate.
Do you think the models are reliable? Lindzen: “Of course not!”
https://ipccreport.wordpress.com/author/etzpcm/page/4/
Fernando Leanme | June 21, 2014 at 5:27 am |
AT THIS TIME the energy imbalance appears to be much lower than predicted by the global climate models.
Only “appears” to be? Do we know what it is or don’t we ?
If we can’t put a reliable figure of watts to this, what answers are models to checked against ?
Even more bascially, how could tell whether or not the energy imbalance moves in line with CO2 levels or not ? (Or, put another way, whether the whole idea of AGW is the serious matter it is being out to be or not).
There can be no energy imbalance, outside the polar regions, everyday a local influx of solar radiation warms the surface, causes a liquid to gas transition of water and also warms the air, the warm moist air move vertically and poleward, at night the surface and atmosphere cool, gaseous water undergoes a phase transition and become water or ice. Heat from the sun is transferred to the poles, via sensible heat in the air and the ocean surface, and also in the form of latent heat, in the form of gaseous water. Heat is radiated from the Earth constantly, but a particular local can radiate more or less energy, over the course of a year, than it receives from the sun and the atmosphere; heat is pumped from the equator toward the poles via air and ocean surface currents. At the poles there is always more energy radiated than received from solar and atmospheric sources; warm surface saline waters cool, become dense, and plunge to the ocean depths, slowly making their way back to the equator, making up up evaporated water removed from the surface of the oceans.
The time scales of these changes are across at least 10,000 orders of magnitude; the 24 hour cycle of daily warming and cooling can be compared with the millennial movements of cold salines sinking at the poles and then making their way back to the equatorial surface.
GCM’s are used to diagnose an “inferred” energy imbalance of about 0.5 W/m^2 or about 0.1% of the incoming solar radiation, right? CMIP5 model biases/spread in the major components of the energy budget are of the order of 10’s of W/m^2. The known errors and unknowns are something like 50 to 100’s of Hiroshima bombs every second.
The big pieces don’t fit the puzzle but they are forced in. Every modeling center suggests a different way of doing it. Many of them have one area that looks plausible, but get it wrong elsewhere. If one piece doesn’t fit, some of the rest must be wrong too. If most major pieces don’t fit, then the tiny pieces aren’t anywhere near right (unless by a huge coincidence).
Anyway… we look at that tiny 0.1% piece, believe it must be correct, and spend $1 billion every day for decades on trying to change it. Without being able to verify it made any difference.
Everything will lead to more funding, including whether one can trust the models, which is really just another grant proposal rather than a question.
If the problem is changed to whether scientists ought to be interested in curiosity-based science, of course they should. It’s very satisfying, but it doesn’t pay the bills. Go where the funding is.
Having worked long ago on the NS equations and the problem of closure, no, you can’t trust the patches and fixes people will have put in. None of them work. I tried them all.
Skepticism with loud voices also leads to more funding to try to resolve differences. Agreement doesn’t. Go skeptics. It helps fund more research even after the question has been answered.
> Skepticism with loud voices also leads to more funding to try to resolve differences.
In an honest-science environment it would. But government-funded climate science takes every effort to suppress, ignore and ostrasize, so as to prop up the politically-motivated ‘consensus’ it serves.
As to “Can we trust climate models?” I would be inclined to say, no, but I’m more concerned that this area of science has really been dominated by a focus on what the models can and cannot tell us, and there seems to be a dearth of falsifiable hypotheses set forth in this field. I don’t think we’ve gotten to the point where asking “Can we trust climate models” is ready to be asked.
It seems to me that the fundamental questions have not been answered. I just want to address one here.
Pre-industrial CO2 was about 280 ppm (0.03%) in the atmosphere, and the number is now about 395 ppm (0.04%). Some of that increase may be natural and some man-made. I do not think we know the breakdown.
The mere fact that human activities produce CO2 is not sufficient to conclude that the increase is solely due to human activity. Even if the increase seems dramatic and coincides with industrial development, it does not demonstrate causation. Remember, it seemed obvious to most of the population in 1492 that the world was flat, but that didn’t make it true.
In some ways, science is the antithesis of belief in the obvious. The scientific method demands that other reasons for the CO2 increase be put forth in a falsifiable hypothesis, then one by one demonstrated to be incorrect (if, indeed they all are). It seems to me that the persons most invested in CAGW are trying to “prove” they are right, rather than openly, objectively and sincerely examining all the other alternatives and proving them wrong, which is the way things proceed in most other scientific disciplines.
Tweaking a model to demonstrate consistency with past and present conditions is not useful for answering the fundamental questions:
1. Is the recent increase in CO2 solely anthropogenic?
2. How does an increase in atmospheric CO2 impact GMT? (i.e., if increased CO2 always results in a positive feedback loop to temperature, we would have fried a long time ago – there are clearly mitigating factors, or competing negative feedback effects).
That’s all – just had to say that I don’t think we’re even at the point of asking if the models are useful.
Thanks for the forum!
If you add up emissions, they have provided twice the amount needed to account for the CO2 increase in the atmosphere. Meanwhile the ocean acidified too, so we can guess where the rest went. It seems fairly obvious.
I believe one of my main points was that “it’s obvious” is not scientific evidence.
Climate model madness started with James Hansen in 1988 when he produced three curves – A, B, and C – that he had calculated on his IBM mainframe. Curve A was “business as usual” or what should happen in the real world. By now we have lived through most of the period for which his predictions were meant and know that his predictions have been all wrong – much too high compared to what really happened. By now the modelers have switched to supercomputers and are writing million line code for their climate models. Heck, why not, if there are billions available for climate research today. A supercomputer costs upwards of 50 million, which is nothing in this environment, and there are several dozens in operation by now. They have had 26 years to polish their software since Hansen first came out with it and you would think that with superior equipment and time to get it going they would by now have a good grip on what the climate is doing. But no such luck. A look at a CMIP5 house duster will tell you that their output is no better than Hansen’s was 26 years ago and in some cases much worse. In particular, every single one of them, and there are dozens, completely misses it in coverage of the warming pause in the twenty-first century. The real problem seems to be that carbon dioxide greenhouse warming is built into them by fiat and as a result they all point upward and not horizontally when the pause has already been horizontal for the last 17 years. And this drivel gets presented to politicians who need to know what is coming up in order to plan societal responses. I could make suggestions for improvement but clearly the system is out of control and deaf to reality. It is time to admit that climate modeling is not working, is unredeemable, and should be shut down to stop the extravagant waste of research money on worthless projects.
> clearly the system is out of control and deaf to reality.
This stems from its political funding. Reality must needs give way to political correctness.
“Curve A was “business as usual” or what should happen in the real world.”
No curve B was expected. Actual emissions followed scenario C.
http://tamino.files.wordpress.com/2014/03/forcing.jpeg
Pingback: Weekly Climate and Energy News Roundup | Watts Up With That?
Climate models have been tuned to the past, get the present incorrect, but will be accurate in the future? I’d like 2 climate models, the Brooklyn Bridge, and a hot dog with relish please. I’m still waiting for California to break away into the Pacific so my investment in beach property in Arizona pays off.
The problem, as I see it, is that there is theory that changes in ocean heat transport can change global temperatures and there is evidence that it may have changed enough to explain up to all the warming of the modern era. How you can possibly determine what was caused by man and what was caused by nature without first understanding ocean currents is beyond me. Climate models are currently nothing better than very expensive ouija boards.
“Climate models are currently nothing better than very expensive ouija boards”
I think this is an insult to ouija boards!
“What the team found surprised them: When participants were asked, verbally, to guess the answers to the best of their ability, they were right only around 50 percent of the time, a typical result for guessing. But when they answered using the board, believing that the answers were coming from someplace else, they answered correctly upwards of 65 percent of the time. “It was so dramatic how much better they did on these questions than if they answered to the best of their ability that we were like, ‘This is just weird, how could they be that much better?’” recalled Fels. “It was so dramatic we couldn’t believe it.” The implication was, Fels explained, that one’s non-conscious was a lot smarter than anyone knew.”
http://www.smithsonianmag.com/history/the-strange-and-mysterious-history-of-the-ouija-board-5860627/?all&no-ist
“And finally, someone needs to get serious and discuss alternative climate model structural forms – not just adding more chemistry to the models, but rethinking the structural form of the dynamical core.”
The dynamical core is natural variability, the alternative is that it is solar forced at the noise level, at the scale of weather, rather than it being internal stochastic phenomena. Evidence of such forcing would necessitate an additional solar metric to TSI to be acknowledged, particularly if it could explain shifts in atmospheric teleconnections (AO/NAO), and oceanic modes (ENSO & AMO), at a variety of scales.
That would knock a very large hole in the assumed magnitude of CO2 forcing.
From the top level down, I can present a huge volume of consistent evidence of a short term planetary ordering of solar activity, with correlations to monthly CET deviations from normals back to 1659, and English written records back to the 12th century. Like it or not, this is the direction weather and climate science us going to take, as it explains the past at the noise level, and can forecast at the noise level, which is the only way climate can be forecast with any certainty, as climate is the sum of the noise. It’s only a matter of time for the dots to be joined in establishing which solar metrics are working at such scales, and how they interface with the atmosphere, there’s no doubt that the solar wind is important.
Model development is an iterative process in which the model developer nas to continuously strive for better agreement with reality. There is little evidence of this process by which we can judge progress.This is all a part of the model designer writing up his design to make it as transparent as possible. Sadly the IPCC seems to have ignored this principle.
Model validation should also be a part of this continuos process, requiring each physical process to be separately validated. The whole model is only as strong as its weakest link so this is an important step.
Dogged by the ‘science is settled’ argument, the IPCC seems to be unsure whether it is a scientific research organisation, or not. This needs to be settled.
A figure showing the magnitudes of CMIP5 model spread in the major planetary energy budget components. For starters, the range of incoming solar radiation among the models, a major input value, is about five times larger than the inferred energy imbalance (~0,5W/m^2) they’re used to diagnose.
According to Forster et al. 2013 the models show pre-industrial energy imbalances of up to +/- 3 W/m^2. What’s up with that?
https://ilmastotiede.files.wordpress.com/2014/06/energybalance.png
The values are from here, running text and Table 3. Errors, incorrect terminology, etc. in my synthesis are very possible.
Personally, every time I try to discuss model viability with the climate crowd, I get mocked and run off. I come at it as a mathematician and an expert in finite element modelling for engineering problems. I ask for proofs or numerical studies as opposed to simple conclusions (my least favorite: “even if we can’t solve for the precise weather, we can solve climate because statistics are computable”). Oh yeah? Says who? Oh, I see, says you…
As a result, my impression is that the climate community is essentially hostile to the field of mathmatics (despite their love of our equations). My conclusion therefore, is that climate scientists are ‘tinkerers’ in the field of simulation and math. Yet they have the arrogance to slam those that are expert.
Hence I have little to no trust in their results.
A real mathematician working with nonlinear systems … Rare breed and worthwhile asset. Its a shame you aren’t appreciated more.
who have you tried discussing it with, specifically?
Not to mention that my experince working at these institutions trying to gain any of the climate guys interest in state of the art massively parallel adaptive finite element solvers essentially resulting in a blank wall and the eventuall exodus of myself and the other experts wanting to help bring the 1970’s codes into the light of day.
These guys love their old spaghetti finite difference codes, hard wired to the grid size and absolutely unverifiable.
stay tuned, post on model structural form coming next (later today or tomorrow)
So nickels, what you are saying is that you are a loser that likes to give up when you sense resistance to your ideas?
So as to set you straight — you are perfectly free to present your.alternate approaches in whatever communication medium you choose. Good luck.
“So nickels, what you are saying is that you are a loser that likes to give up when you sense resistance to your ideas?”
Classy.
How acute Web is!
=============
Dang, missed by a minute. Web, this is one of your best. Can I get an autographed copy?
=======
WebHubTelescope: So nickels, what you are saying is that you are a loser that likes to give up when you sense resistance to your ideas?
What do you think of his main points: that (most, lots of, etc) climate scientists are behind the times in computing and resistant to new information. How he chose to invest his efforts after learning this has no relevance.
“How acute Web is!”
Chronic I believe is the correct diagnosis
Webby prefers to stick with the traditional climate science curve fitting methodology. Go with what you know.
More fun was reading the dialog at SkS when nickels demanded “proof” of certain characteristics of climate models which are based on the Lorenz equation.
Nickels also linked to a paper by ClaesJohnson who is the skyyydragon that also claims classical theories of flight are wrong,and that he would reject scientific awards out of principal because of disagreements with any award committe that would honor him.
Cuckoo land all around. You skeptics can stew in your juices all you want.
Not that WHT deserves the respect of a reply given its ad hoc personal attacks, but since the results were so interesting:
The claim on SKS was that even though the Lorenz equations are not predictable, “Statistical properties of the equations are predictable”. At which point I requested literature. Given no such literature I did a few simple experiments.
I computed the average x value of the solution for 2000 time units with varying time steps. According to SKS this should be predictable as it is a statistical quantity.
The results:
Wild divergence of the x average based on the timestep.
Which means the claim on sks was a lie. The quantity is simply not computable.
This link is a plot of the resulting x_avg for a variety of timesteps. The smaller the timestep, the more variance in the result. No convergence.
Feel free to try it yourself.
https://onedrive.live.com/redir?resid=7B5D6E112B4BDA77!352&authkey=!AIUBHTroXHgI5fg&v=3&ithint=photo%2c.jpg
Why would you try to solve the Lorenz equation, when a hydrodynamic formulation based on a Mathieu equation would apply to the problem space better?
This formulation is also much better behaved and has stable regimes that match the empirical observations.
It’s hard to believe your side is losing, Web, with stellar communicators such as yourself on board. So persuasive, downright compelling.
====================
Exactly. On this site WHT is fighting uphill, but on sks when you get ten of his type coming at you its simple cyberbullying.
The experience with that bullying is exactly what got me fired up about stepping back into this issue and fighting what I see as a massive wave of scientific bullying hiding under the name of ‘consensus’.
nickels, you can’t even hold your own here.
Confirming that ENSO hydrodynamics can be simplified
http://contextearth.com/2014/06/25/proxy-confirmation-of-soim/
“How he chose to invest his efforts after learning this has no relevance”.
Exactly, thanks. I like my new job and it has nothing to do with climate! Pays more too.
To be fair (maybe too much so), the institution I worked at had a decent mathematics section 15 years ago. Today, that section is gone. The general lack of funding in science in general is a big part of the cause.
Nickels
Many on this site had the same initial reaction to Web and his compatriots as you did. When I realized this attitude was pervasive in the climate community, I knew there was something terribly wrong. The normal rigorous debate and a premium on scientific inquiry is nearly absent. The nastier they are the closer to the truth you are.
Not really. I don’t work on climate science in an official capacity. But I do know how to simplify the physics of seemingly complex systems by applying first-order models. A guy like nickels is very worried about people that can simplify climate because he loses the leverage of FUD. The FUD he is peddling is the great unknown of models that he claims to have insight into. Once his private insight is opened up, the FUD dissipates and he is left with vapor. It really is all smoke and mirrors with the deniers.
New thread! http://judithcurry.com/2014/06/25/model-structural-uncertainty-are-gcms-the-best-tools/
Agreed. I used to work at a climate lab. I hated it. I was marginilized in ways that have never occured in any other part of my highly varied career. I too realized there was something very wrong.
WHT (Paul P.) and his type of bullying brought me on board! No more sitting in the bleachers watching the lies!
Nickels, You are saying something important. Finite element methods are vastly better. We’re you at NCAR by any chance?
Web is as annoying as he is nasty. Name calling and insults are just noise.
Cat out of the bag, I guess; yes, NCAR.
Funding, term positions, dissolution of the math group, finite difference fortran codes, etc, etc…
PTSD
They had a massively parallel HP-adaptive finite element framework right under their noses and they expressed less than interest, hostility.
Funding or culture? Probably both.
Shame. Beautiful campus!
v. interesting, now I’ve made the connection
thanks, btw.
Are the models sufficiently wrong that we should anticipate reality falling outside the range of model results?
That has happened for decades. That is not anticipate, that is history.
It would be helpful to establish the principle that no model can validate another, much less 2+ models collude to self-validate.