'Let’s face it. The climate has never been more boring.'

RGB, love your posts, but really, heads will roll? Simple honesty? Sheesh, you are assuming these folks are like you in some way? These are double down pathological liars who have executive orders and MSM and large universities protecting their grant money. One bristlecone pine, please.? You have no idea who and what you are dealing with. The “I” in IPCC stands for one thing, access to your pocketbook by folks who own the guns. Truth is not even part of the discussion and never has been.

Taking the main post by rgbatduke and all subsequent comments by all commenters on this thread including also rgbatduke’s follow up comments, can one reasonably find the primary fault that is both necessary and sufficient (N&S) for the entire 20+ years of the IPCC that culminated in the AR5? Is it within the environmentalist community or within the certain types of media or within a certain sets of politicians or within certain departments within academia’s halls or somewhere within science’s community or is it somewhere else entirely?
I find the N&S primary fault to be within the portion of the scientific community that supports a philosophy of science thats views science as primarily an adaptable expedient in support of a predetermined climate answer where the answer is supplied by an ideology. Without their complicity I do not see how the IPCC AR5 could exist.
John

Gary Hladik says:
November 4, 2013 at 1:57 pm (Edit)
—————————
There is nothing wrong about Mr Mosher. He is rigorous in his approach and unconcerned with being popular. Many may not agree with him but he always fights his corner and I have never seen him run away or try and defend something when he has been proved wrong. A right royal PITA and the better for all of us that he is.

There is no need to disprove GW – it has been happening recently (since end of LIA) just as GC has happened many times in the past.
I believe, if Dr. Brown is specifically talking about proving or disproving the models, then his assertions seem correct.
I was specifically referring to proving or disproving the CAGW hypothesis, which is contingent upon the AGW hypothesis. That is, that humans are causing global warming (not that GW has or is occurring) and the contingent proposition that the GW caused by us Anthropes will be Catastrophic.
Those are specific hypotheses, and in the time frame of 2100 disproof would be no observed global warming anthropogenic or otherwise by 2100, where proof would be observed global warming by 2100 AND the positive connection of that warming to human activity. Ditto “catastrophic” AGW. — disproof is the failure of any GW at all, any proven anthropogenic connection to whatever GW occurs, and of course, the failure for any GW (anthropogenic or not) to be catastrophic. Proof would be catastrophe caused by GW that has been sufficiently strongly connected to human activity by experiments.
rgb

Kudos for a many-faceted, realistic overview of the situation. In light of the demands of your duties at Duke, unless you type at 1000wpm, it leaves me wondering how you find time for essay-length comments. I only wish you had said “recurrent” instead of “periodic” climate cycle in your presentation,.which minor inaccuracy deflects attention away from the major issues.
My friends and colleagues have sometimes referred to me as the “rgbbot”, suspecting that I’m actually a beowulf-based AI.
And you are quite right — recurrent would be a far better and more precise term than periodic. It expresses the reservations that I’ve been (of course) holding in my own mind and when I often introduce this as numerology, not science per se.
rgb

Having digested R.M. Goody’s “Atmospheric Radiation” and John A. Dutton’s
“The Ceaseless Wind” many decades ago, I dispute the basis of such
conclusions. And I doubt that “Grant Perry’s book”, which is nowhere
identified on the web, would persuade any field-experienced geophysicist
otherwise. That there always is a radiative exchange between surface and
atmospheric matter is as boring as the exchange between the floor and the
insulated attic in a heated room. It’s the heat source and the net
transfer by all mechanisms, not just directional radiative intensity, that
really matters. That was the thrust of my remarks.
Sorry. should have included the link:
http://www.amazon.com/First-Course-Atmospheric-Radiation-2nd/dp/0972903313
If you are a geophysicist (and hence physicist) then I doubt that you will argue with much of the book. I teach graduate-level electrodynamics and quantum mechanics from time to time (at the moment I’m teaching an endless series of intro physics but that’s the way our department rotates teaching) and the physics seems quite sound. In fact, it seems quite elementary. Nowhere in the book does Petty make a concrete assertion about e.g. the total climate sensitivity, BTW. It does, however, contain the TOA and BOA spectrographs that to any physicist are rather concrete proof of the GHE.
As for heat transfer mechanisms, I certainly don’t disagree that heat transfer by all mechanisms is important, but the point of the GHE is that without the atmospheric opacity in the LWIR bands, the troposphere would descend more or less to the surface of the Earth. But I’m guessing that we agree on all of this. I just don’t think there is a smoking gun disproof of CAGW and also don’t think that one can likely point to any particular subroutine in a GCM and say “aha, this is the wrong physics!” The errors are more subtle than that — omission, incorrect initialization, inadequate granularity, and a touch of good old fashioned code tuning to get a desired result.
rgb

Dr. Brown, could you answer the question of whether the Climate Models are projections for the different storylines and scenarios purposed by Ged Davis. If the ensemble is not for business as usual but for an array of futures including a ” much reduced greenhouse gas emissions alternate scenario” then that spaghetti graph is a monumental lie since none of the ‘Projections’ fit reality including their desired scenario.
Hi Gail,
I suppose the best answer I could give would be “No”. Both no I can’t answer the question and no the GCMs are not really for that purpose AFAIK, although there have been figures published in the past for 0 emissions, continued emissions, increased emissions that probably are. But the GCMs (however they were USED in AR5) are for the purpose of modelling the climate. They can be judged on the basis of how well they modelled the climate over the trial period that began as soon as the models were initialized in such a way that they did well over the training period. Since they were initialized to show excessive CO_2-driven warming in what was likely a natural spike that had little to do with CO_2, as soon as the natural process responsible for that warming shifted the models all rapidly diverged from nature. That’s by far the simplest interpretation, and one that at this point I think a lot of even the most ardent warmists would agree with.
The solution is simple enough. In the short run, throw out all of the models that failed to reproduce this shift, as obviously they cannot actually predict what nature does, let alone what nature plus CO_2 does. This alone would significantly lower climate sensitivity estimates, and we are starting to see papers that are doing precisely that. We will see more, every year the climate fails to warm, because every year without significant warming further constrains feedbacks and overall sensitivity and even though the researchers are struggling to overcome a really serious bias towards their Bayesian prior assumption of huge sensitivity, data talks, bullshit walks, and those estimates must come down eventually in the face of the data.
In the longer run, build better models. The idea of building good climate models is not itself flawed. The problem has been building climate models that politically interpreted the late 20th century warming as being somehow different from the nearly identical early 20th century warming and blaming the former entirely on CO_2. This was then built right into the models in the form of maximum direct CO_2 forcing and strong water vapor feedback, and compensating for this by cranking up the (mostly unknown) contribution of aerosols until they got a good fit to the 1970s-1990s warming.
The problem with multivariate systems is there is usually more than one way to get a model to agree with training data, but those different ways usually diverge once you get outside of the training set, and at most one of them will be actually correct. Discovering the actually correct climate description almost certainly requires more than a stretch of monotonic warming to use as a training set.
If they trained on data from 1940 to 2013, OTOH, they’d have at least two periods of neutral to negative temperature changes AND a period of positive change (assuming, of course, that the various GASTA/GAST estimates are reasonably correct over this interval, which is far from certain). A model built that could correctly describe those changes would likely be more robust than the ones built and initialized to produce monotonic warming. None of this is really surprising — except that the writers of AR5 and its SPM would completely neglect mentioning it and indeed, would treat this divergence almost as if it is a problem with the actual global temperature instead of a problem with the models.
And it may not be a problem with the models — it is important to remember that this is a possibility too. They could still be right. In a year, two years, global temperature COULD spike up a half a degree C or whatever and the curves could rejoin one another. The hard thing about prediction is that it is about the future, and it isn’t terribly easy to prove or disprove a prediction except by waiting, and imprecise or multivariate predictions are difficult to check even by waiting. It worked over and over again for Biblical prophets — make an obscure prophecy and wait — sooner or later an even occurs that you can point at as proof that your prophecy was correct. So if I prophecy “There will be an enormous storm, the strongest storm in a 100 years” I have zero chance of being proven incorrect, and every chance of being proven correct. Eventually. If I say “There will be an enormous storm this year and it will strike central Florida“, well, only really stupid prophets would prophecy that. Too easy to get wrong, too “final” when you do.
So it really doesn’t matter how many years the climate stays flat and boring. It could always get exciting next year, and since the GCM “prophecies” are non-specific, any excitement at all can be counted as evidence that they are right, or will eventually be. All we can say is it hasn’t warmed the way they predict yet. We could say more, of course, if we used probability theory and statistics theory to assess the models as null hypotheses — we could then say that the probability that the models are correct is smoothly decaying in time as long as the climate remains neutral or fails to warm as they describe. How long should we wait before we reject the null hypothesis — or to put it more precisely, given the probability that this model is correct AND we have the observed climate different from its predictions, how small does that probability have to get before we conclude that the theory given the data is very, very implausible? The usual rule (lacking Bayesian priors that bias the estimate) is less than 0.05, but really this is pretty arbitrary. Less than 0.01 is pretty safe — it is asserting that one would observe the prediction AND the data only one Universe in 100 that were nearly identically prepared, so we’d have to be pretty unlucky to be in that one. Less that 0.001 and there is only one chance in a thousand that the model is right, given the data.
However, given many independent models (not independent runs of a single model), all of them too warm, the odds change. It becomes much less stringent to reject a model because you have so many chances to get the climate right and the models themselves have some noise and variance — you’d expect even a bad model to have a chance of getting the observed climate right at least some of the time. If you roll a 20 sided coin looking for 0.05 acceptance 20 times, you can’t be surprised if you get it around one time and GETTING it one time doesn’t mean you should accept the model, it means you should reject it!
rgb

I would argue that they cannot
store more energy than their miniscule total mass allows. To the extent
that dry convection can heat the vastly greater atmospheric bulk, GHG’s
strike me as hardly indispensible in maintaining a tropopause at an
altitude roughly equal to maximum convective reach. Furthermore, LWIR
atmospheric backscattering is not on equal thermodynamic footing with
high-entropy insolation. Its low-entropy radiation is almost totally
absorbed in the surface skin of the oceans, thereby sustaining
temperature-lowering evaporation. Instead of being an external “forcing,”
“backradiation” merely recirculates energy within the system.
Not sure what you mean by a “more appropriate forum”, so I’ll reply here. First, the mass fraction of the GHG’s is almost totally irrelevant to the GHE. The simplest single-slab model of the GHE doesn’t have a stated heat capacity for the atmosphere at all — the atmospheric mass/heat capacity determines the relaxation rates but not the steady state temperature. What matters is the opacity of the atmosphere in certain bands of LWIR. Second, nobody suggest that LWIR backscattering is on an “equal thermodynamic footing” with insolation — the latter is the actual energy source for the system, just as outer space at 3 K is the actual energy sink, within more or less irrelevant perturbations. However, I fail to see your point.
Again, in my house in winter the heat source is burning natural gas, the heat sink is the cold outside air, but insulation in between the two, while producing no energy whatsoever, can and does substantially modify the achievable temperature difference between the inside of the house and the outside by modulating the rate of energy flow from the inside (once delivered) to the outside. In the case of the Earth, the sun delivers energy at the TOA at some rate peaked at SW wavelengths. In a single slab model, this energy is partially absorbed by the atmosphere, hits the surface, is partially reflected (with the reflected component again being partially absorbed by the atmosphere on the way out), is partially absorbed by the surface, is reradiated by the surface as LWIR, is partially absorbed by the atmosphere on ITS way out. The atmosphere gains energy from direct insolation and LWIR surface emission and symmetrically reradiates it, with half escaping and half being returned to the surface. All radiation occurs at rates proportional to temperature to the fourth. This constitutes a system that one can easily solve for the surface equilibrium temperature as a function of insolation rate, albedo, and the atmospheric SW and LW absorptivity. In the limit that albedo is zero, SW absorptivity is zero and LW is unity, one gets the usual factor between the steady state temperature of the surface with a perfect LW absorber shell and the temperature without any shell at all. I don’t see what the “footing” of LW vs SW radiation has to do with an energy balance formula. Energy is conserved, period.
Can one dress this up and make it more complex? Sure. That’s almost precisely what GCMs are, models that are more complex. And no, they do not neglect LW absorption at the surface of the ocean, surface evaporation and heat transport in the form of water vapor, the substantial GHE from water vapor alone, and so on. They may or may not do the physics of this all correctly, they may or may not have the correct parametric values for the various rates involved, but in fact one of several reasons that the models run too warm is that they do incorporate a substantial water vapor GHE as a net positive feedback, leading to climate sensitivity some 2x (or more) what one expects from CO_2 alone. I think there is some evidence that this is incorrectly treated in models, that water vapor feedback is anywhere from net neutral to slightly negative, because water vapor is tied to a complex water cycle involving latent heat transfer of surface energy vertically, modulating albedo at different vertical heights, and losing heat to space similarly at different heights and wavelengths and rates (after being largely lofted through the CO_2 “barrier”). But it is not the case that models do not have this stuff going on at all, that they leave the physics out.
If you want to look at this to see for yourself, there is an open source GCM here:
http://www.cesm.ucar.edu/models/atm-cam/docs/description/
The documentation carefully details all of the physics accounted for in the model, how it is initialized, how it is run, and so on. Since the source is open (although a total PITA to build — it is terribly organized and requires modules from multiple sources/sites with separate — but still open — licensing) you can check the physics, add your own physics, modify the code in any way you like, change the parametric initialization, publicly criticize and/or correct any errors you find or wish to assert, and see if you can do better. As can I. With the catch, of course, being that neither of us have time to do so, or the computational resources handy required to integrate the model forward over decades in five minute timesteps. But you look at the documentation and tell me — what is being omitted? There are some real candidates for omission — GCR modulation of albedo, for example — but note well that the model explicitly treats the (single slab) ocean and sea ice. Should they use a multiple slab ocean model? Probably. With ARGO, one might eventually have the data needed to construct one, and of course the ocean is where some climate scientists are looking for the “missing heat” (and if so, good luck getting it out again as the ocean is a heat sink with century-scale heat capacity at the rate of the supposed net radiation anomaly assuming that there is NOT sufficient negative feedback at the surface to eliminate most of its heating). But note well, this is an issue of the model being broken or inadequate or incorrectly parametrized, not the model failing to include the effect.
Third, I do not understand your remark that backradiation “merely recirculates energy within the system”. As noted above, in one sense this is quite irrelevant to the question of whether or not it is a substantial source of power delivered to the surface. If you can measure 200+ W/m^2 in downwelling atmospheric radiation integrated over all wavelengths at nighttime (no sun at all in the sky) the fact that this energy originally came from the sun and is downwelling only after following a complex path is utterly immaterial to the fact that the energy it delivers partially cancels radiative energy outflow from the surface (and, in the event that there is an inversion, can actually MORE than cancel outflow). Nobody is asserting that the atmosphere “produces” this energy, but nobody could possibly argue that this energy is not a substantial factor in the rate that the surface cools and hence in the average surface temperature.
Finally, although I did not directly quote you above, you do remark that often people only look at dry air TOA/BOA spectrographs to demonstrate the GHE. Petty certainly presents several figures of this sort, because without the complication of water vapor one can most easily and cleanly see the relevant absorptive bands associated with e.g. CO_2 and Ozone as well as the “windows” in LWIR where cooling does occur. But Petty also presents several wet air spectrographs and sure, they show a much stronger GHE. All that this means is that the GHE is most certain real, that water vapor is the most important GHG and (as noted) the bulk of the effect from CO_2 was obtained when it first became saturated so that the atmosphere became optically opaque in its primary absorptive bands. It doesn’t mean that CO_2 is unimportant — the Earth teeters along in this particular ice age on the edge of a return to glaciation, and without CO_2 in the atmosphere at all I think there is little doubt that the Earth would devolve to being a permanent iceball everywhere except possibly a narrow band near the equator in a matter of centuries and never emerge. Water vapor alone has a dangerous instability on the cold side and water ice has a huge surface albedo. Even with CO_2 the Earth is at best borderline unstable as evidenced by the glacial record over the last 3 million years.
Do the GCMs get the physics of this sort of feedback and the additional multichannel feedback from aerosols as they play the dual role of nucleation sites for cloud formation and a high-albedo component to the atmosphere in their own right correct? Quite possibly not. This is a really difficult problem. Petty avoids making any positive statements about it — explaining the physical arguments for line broadening and the various methods used to approximate what happens in the absorptive bands and how they might vary with concentration and so on, but this is quite possibly a place where the GCMs don’t do the physics correctly, where they use a smoothed, separable approximation that for one reason or another breaks down in reality because it is non-separable. One cannot just say that they do make this or any other particular error, however, without building a GCM and changing the physics or parameters in a defensible way and showing that one gets a better description of the climate when one does. All once can do without this is what I am doing — note that the agreement between model predictions and actual climate outside of the era used to initialize the models and determine their parametrization is poor, so that it is probable that each GCM in poor agreement contains at least one error responsible for the difference. And IMO, almost certainly more than one. It is a hard problem.
I don’t always agree with Steve Mosher — I think he is too uncritical of the GCMs, and too quick to jump to their defense when there is little benefit in doing so, when the only way to improve them is to acknowledge the fact that they probably aren’t working accurately enough to be used to divert trillions of dollars and negatively impact every human being alive right now on the basis of their predictions of future catastrophe when they manifestly are not doing a good job on the present. But I do agree with him that if it weren’t for the political burden and the fact that an entire community of scientists have essentially bet their careers, their reputations, and a hell of a lot of everybody’s money on what should originally have been considered a very, very, preliminary conclusion, we’d be saying that GCMs aren’t terribly conceived or implemented and that in some ways they do surprisingly well given the complexity of the system.
But surprisingly well is not well enough to trust them to direct the course of civilization itself, especially not if they are in error by a factor of two or even more in their estimates of climate sensitivity, as I think that the mounting evidence suggests that they are. As I point out above, there are real human lives — millions of them — being last as we dick around with “wood burning electricity generation”. We might as well pile our money up and shovel it into the furnaces and use that to make electricity when there is comparatively cheap and abundant coal to use instead, when lowering energy prices by using the least expensive resources affects human health, wealth and happiness worldwide. The only reason not to burn coal to generate electricity in the short run (over the next 20 to 40 years) is the GCM’s prediction of global catastrophe, which then becomes a highly nonlinear, Pascal’s Wager style hidden cost.
Otherwise we’d just burn coal while gradually improving the cost-benefit and safety of nuclear including Thorium, while working on fusion, while perfecting e.g. PV solar and while developing efficient energy storage and transport mechanisms so that we can store solar and use it over 24 hours or more, so that we can transport solar generated electricity from dry tropical and subtropical deserts to moist wet snowy temperate and polar zones. Inside a decade (or at most two), PV solar is going to become one of the least expensive energy resources for a broad band of the Earth’s surface, and if we figure out how to store PV solar energy efficiently to buffer solar energy over (say) 48 to 72 hours (or even more) then nobody will have to advocate solar power as world-saving policy, people will implement it out of simple human greed, a desire to have the cheapest possible energy. By 2050, burning coal will disappear not because it produces dangerous CO_2 but because we have cheaper, cleaner ways of making electricity.
This sort of technological discontinuity is almost completely predictable, although people rarely do make this sort of prediction. Thirty years ago, nearly every household in America and Europe had a rooftop television antenna. The entire planet radiated substantial amounts of electromagnetic energy in certain bands, and we started to look for distant civilizations by means of looking for the “signal” of electromagnetic radiation in those bands, reasoning that any advanced civilization would have TV. Look how well that assumption turned out. no houses have external TV antennae at all any more, and the satellite dishes that many houses do have receive far weaker signals directed straight down. I’ve read proposals to look for remote civilizations by looking for some sort of “pollution signature” in the atmospheres of exoplanets when and as resolution permits, but of course any such signature would also very likely be a short term transient as the civilization evolves to cleaner steady state resources.
I actually think it is perfectly lovely to encourage our global civilization along these pathways, to cleaner energy that doesn’t involve burning coal or oil, since in the long run both of these substances will be far more valuable to us unburned and not used to make energy we can get in other ways without facing long term scarcity issues. We shouldn’t burn diamonds, either. We shouldn’t use Helium in kid’s balloons, or treat Thorium as a toxic waste byproduct of mining rare earth metals. However, the Earth has 7 billion people and is a social and political powderkeg as long as 3 billion of those people are still living in the 17th century, 2 billion of them are living in the 20th century, and only 2 billion of them get the full benefits of living in the 21st century as far as wealth, access to resources, comfort, health, happiness, freedom and all that are concerned. We must choose a path to a steady state civilization that does not advance on the scarred backs and corpses (mostly children) of the poorest people in the world. That means that we should implement political measures that directly raise the cost of energy worldwide only when compelled by near certainty that the only alternative is an even greater catastrophe than the ongoing catastrophe of living in the most peaceful, wealthiest period in the history of mankind, with almost unimaginable wealth and knowledge to bring to bear on any given global issue, and then expend all of this wealth not to bring the gift of civilization to all of the world’s people but rather to prevent it, to keep them energy poor and in the invisible chains of ignorance, poverty, disease, misery, and totalitarian religious beliefs.
rgb
rgb