Set Phasers on Stun
March 29th, 2009 by Roy W. Spencer, Ph. D.
I’ve been receiving a steady stream of e-mails asking when our latest work on feedbacks in the climate system will be published. Since I’ve been trying to fit the material from three (previously rejected) papers into one unified paper, it has taken a bit longer than expected…but we are now very close to submission.
We’ve tentatively decided to submit to Journal of Geophysical Research (JGR) rather than any of the American Meteorological Society (AMS) journals. This is because it appears that JGR editors are somewhat less concerned about a paper’s scientific conclusions supporting the policy goals of the IPCC — regulating greenhouse gas emissions. Indeed, JGR’s instructions to reviewers is to not reject a paper simply because the reviewer does not agree with the paper’s scientific conclusions. More on that later.
As those who have been following our work already know, our main conclusion is that climate sensitivity has been grossly overestimated due to a mix up between cause and effect when researchers have observed how global cloud cover varies with temperature.
To use my favorite example, when researchers have observed that global cloud cover decreases with warming, they have assumed that the warming caused the cloud cover to dissipate. This would be a positive feedback since such a response by clouds would let more sunlight in and enhance the warming.
But what they have ignored is the possibility that causation is actually working in the opposite direction: That the decrease in cloud cover caused the warming…not the other way around. And as shown by Spencer and Braswell (2008 J. Climate), this can mask the true existence of negative feedback.
All 20 of the IPCC climate models now have positive cloud feedbacks, which amplify the small about of warming from extra carbon dioxide in the atmosphere. But if cloud feedbacks in the climate system are negative, then the climate system does not particularly care how much you drive your SUV. This is an issue of obvious importance to global warming research. Even the IPCC has admitted that cloud feedbacks remain the largest source of uncertainty in predicting global warming.
Significantly, our new work provides a method for identifying which direction of causation is occurring (forcing or feedback), and for obtaining a more accurate estimate of feedback in the presence of clouds forcing a temperature change. The method involves a new way of analyzing graphs of time filtered satellite observations of the Earth (or even of climate model output).
Well…at least I thought it was new way of analyzing graphs. It turns out that we have simply rediscovered a method used in other physical sciences: phase space analysis. This methodology was first introduced by Willard Gibbs in 1901.
We found that by connecting successively plotted points in graphs of how the global average temperature varies over time versus how global average radiative balance varies over time, one sees two different kinds of structures emerge: linear striations, which are the result of feedback, and spirals which are the result of internal radiative forcing by clouds.
But such a methodology is not new. To quote from Wikipedia on the subject of ‘phase space’:
“Often this succession of plotted points is analogous to the system’s state evolving over time. In the end, the phase diagram…can easily elucidate qualities of the system that might not be obvious otherwise.”
Using a simple climate model we show that these two features that show up in the graphs are a direct result of the two directions of causation: temperature causing clouds to change (revealed by ‘feedback stripes’), and clouds causing temperature to change (revealed by ‘radiative forcing spirals’).
The fact that others have found phase space analysis to be a useful methodology is a good thing. It should lend some credibility to our interpretation. Phase space analysis is what has helped us better understand chaos, along with its Lorenz attractor, strange attractor, etc.
And the fact that we find the exact same structures in the output of the IPCC climate models means that the modelers can not claim our interpretation has no physical basis.
And now we can also use some additional buzzwords in the new article…which seems to help from the standpoint of reviewers thinking you know what you are talking about. The new paper title is, “Phase Space Analysis of Forcing and Feedback in Models and Satellite Observations of Climate Variability”.
It just rolls of the tongue, doesn’t it?
I am confident the work will get published…eventually. But even if it didn’t, our original published paper on the issue has laid the groundwork…it would just take awhile before the research community understands the implications of that work.
What amazes me is the resistance there has been to ‘thinking out of the box’ when trying to estimate the sensitivity of the climate system. Especially when it has been considered to be ‘thinking in the box’ by other sciences for over a century now.
And it is truly unfortunate that the AMS, home of Lorenz’s first published work on chaos in 1963, has decided that political correctness is more important than the advancement of science.
Phase Space – the bain of my control engineering days.
Why, that must mean it is 100 year old physics. Or math. Or both.
So when Keeling et al. measure the CO2 in dry air, but do not tell you the amount of water vapour they remove for each sample, are they actually overestimating the CO2 by as much as 5%, or nearly 20ppm, and their quality control process may be rejecting some accurate data?
Unfortunately Spencer has been making the same mistake as the environmentalists are doing in another field of science. Religion should not distort one’s science and the natural world.
Because of that, we get these activists who bring up Spencer’s religious beliefs (creationism/intelligent design) to discredit his climate research despite the number of religious views (chakras, spirits, Atlantis, ayurveda, homeopathy, etc) they hold that would not be taken seriously in the fields of biology, chemistry and medicine.
I’ll offer a contrarian view on the displacement question. CO2 is made up of carbon and oxygen. Unless your CO2 is the vocanic kind, the oxygen came from the atmophere to be combined in a forest fire, human body, decaying dead animal, etc. It displaced nothing, ’cause the oxygen is still there.
Watched a show on Science Channel last night called “Snowbal Earth”. They made a claim that “rain cleanses the air of CO2.” They then showed a graphic of a browish air with a rain drop passing through it with blue air in the path behind the falling rain drop (I suppose to create the image of “cleansed air”).
Is this an oversimplification of the process or is it just wrong?
Ohioholic (20:49:25) :
> Also, as a personal curiosity, which gas is CO2 displacing when it is added to the atmosphere?
I’m a little surprised at the responses. What I would have said is:
Oxygen. Burning coal takes in oxygen and releases CO2. However, given that we have 380 ppmv CO2 and 200,000 ppmv O2, the recent reduction in O2 is minuscule. It has been measured, though.
Given the number of other answers, I guess I should extend things a bit:
CO2 doesn’t have to displace anything. Hmm, I replied to “replace”. John’s comment is pretty good, but he should note that adding 100 ppmv would increase the average atmospheric pressure by a minuscule amount, but by more than what you would get by adding nitrogen and oxygen becase CO2 is heavier.
Some CO2 is added without consuming O2. Natural gas wells often have a lot of CO2. Some natural gas wells come up “dry” and are nearly all CO2. Those are the main source of commercial CO2 for soft drinks, CO2 fire extinguishers, dry ice, etc.
The key thing to keep in mind is that CO2 is trace gas and has little effect on the concentration of other gases. (Note that there is enough CO2 to saturate the IR window, so it’s adding to the greenhouse effect about as much as it possibly can, but that’s a different issue.)
> Would I be wrong to assume that the heaviest gas is displaced downward, and the lighter gas displaced upward at a proportional rate to their atomic mass?
The viscosity of the atmosphere means that heavier gases would take a very long time to settle to the bottom. The convection in the atmosphere means that the various gases stay well mixed. Diffusion means that even if things did settle the boundaries between the different gases would still be a bit blurred.
Re: Raven (21:13:07) :
“I suspect they avoid it because they think chaos is noise and cancels over ‘climatically significant’ timeframes. ”
That is actually a very common misconception of chaos – many interpret chaos as meaning randomness or complete disorder (i.e. noise). It is NOT, it is “sensitivity to initial conditions” which means predictability of the system is low over long periods of time. In the case of weather/climate, predictability becomess very low after a couple of weeks ahead.
A climate modeller will tell you that because they are looking at long periods of time – “the average state” of the climate is still predictable. That is simply a trivializing of the complexity of the actual system and a very linear way of thinking, because most chaotic systems exhibit scale-invariant behavior. Scale-invariance means the statistics of the system at short time intervals is similar to the statistics at long time intervals. Therefore, the same problem of predictability applies to both scales.
Lorenz and Mandelbrots work on chaos and complexity is near 50 yrs old, yet many in the climate community ignore it as an interesting study in mathematics. They don’t get that the actual world behaves in exactly the way that Lorenz and Mandelbrot found.
Ohioholic,
“what Gas is displaced”?
According to AGW theory, the increase in CO2 is primarily due to man’s use of fossil fuels. This means that we are burning (oxidizing) fuels containing carbon which is being combined with oxygen from the environment.
Actually you would have to do a mass balance of the before and after conditions, but if everything else were to remain constant, the total mass of gas comprising the atmosphere would be increasing by the amount of carbon being added from the fuel, but the oxygen in the CO2 would come from the atmosphere, so % O2 should be dropping.
In reality, there is so much flux of CO2 due to the natural carbon cycles that it is impossible to know what is really happening.
New paper debunking AGW? Ok by me.
Just don’t cite Wikipedia on it.
Well, after all, a chicken is an egg’s way of making another egg.
Case in point: The Asian brown cloud is renown for interfering with the formation of rain clouds due to surface shading, increasing drought patterns elsewhere in Asia. If the resulting decrease in cloud cover lends to greater heat retention (ground-level humidity would still be high, lending to ever warmer nighttime temperatures), then there might be an unforeseen aspect of global warming that contradicts the assumption of a net cooling effect from aerosol shading (oft cited as masking global warming).
In other circumstances there’s a different interzonal effect. In a less-pronounced anomaly trans-Pacific soot & SO2 plumes actually seed megastorms during winter. Likewise tropical forest fires (from slash & burn agriculture) have been observed seeding thunderstorms downwind from the source point of airborne soot.
But these massive brown aerosol clouds are lending to a net decrease in cloud cover then instead of aerosol clouds masking global warming via surface shading, they could be leading to a net increase in temperatures (instead of a net cooling effect). This would explain the decrease in cloud cover coming before the warming, not the other way around.
On a tangential point the Aqua satellite has yet to demonstrate the extra water vapor that climate models predict, perhaps showing a negative feedback (Marohasy)., Likewise with the Argo floats search for the anticipated extra heat sink in the seas, the heat bucket effect is nowhere nearly as pronounced as the climate models predicted. Once again the warmist position has been that aerosols are causing unanticipated shading, but are we to then also implicate aerosols in the lower water vapor levels as well? Ahah! A contradiction lurks: Blame aerosols for Aqua’s lower water vapor data but don’t blame them for decreased cloud cover? The data are telling us something, should we listen?
GCMs are still being developed to model the net effect of aerosols, with field data from V. Ramanathan showing an unexpected net heating effect – not cooling – within brown clouds all around the world. Another surprise has been the net warming effect from airborne soot in the Arctic as well as sootfall on the ice, pushing soot’s role in the sesquicentennial boreal thaw well past what’s modeled for GHG’s.
Then there’s the problem of intermeriodonal cloud system banding (Lindzen, Spencer, Cristi? ) in the tropics where westerly (or easterly in the S.H.) cloud systems create functional rain shadows of drier cloudless bands in adjacent latitudes.
Climate models have modeled water vapor and relative humidity as evenly distributed throughout the atmosphere – in both latitude and altitude – than is actually the case.
Modeling fluid dynamics in a vat is hard enough, but it seems to me climate models have a long way to go before they’re ready to spit out results consistent with actual field data.
So more clouds would result in more warming, but more warming would cause fewer clouds? But more cloud cover would result in higher albedo (surface shading) and faster heat exchange into space while they also enhanced their dissipation via enhanced warming?
Seems to me that somehow a simple principle has been obscured: Clouds are not the same thing as humidity. Surely an increase in cloudless humidity would add to warming, but cloud cover would seem a net negative feedback (as in the case of the net cooling from wintertime storms seeded by transocean soot…).
Likewise couldn’t warmer adjacent inflow regions near storm systems render the storms more energetic, thereby driving a faster heat exchange process, ejecting more heat into space?
Makes me really wonder whether a zero-sum feedback in increased cloud & storm systems been obscured by other compounded feedback loops manifest in GCM’s. This is a hazard in any computer model, that each subsystem needs to be parametrically unit tested, validated with real-world input ranges, the output results validated unit by unit and then case-matched against all the other subsystem’s expected inputs (with stop limit error traps for outside cases).
Congratulations and thanks are due to Dr Spencer for his perseverence against the odds, the system and political correctness.
Ohioholic (20:49:25) :
It doesn’t displace any gas but adds to the volume. Fossil fuel is made of bonded Carbon and Hydrogen atoms. Although it depends on the exact chemical make up of the fuel and how it is burned, the main gases released are CO2 and H2O. The Oxygen atoms are already in the atmosphere so in a perfect world one Carbon and two Hydrogen atoms are released but they are bonded with three Oxygen.
While I lean towards the idea that CO2 from burning fossil fuels doesn’t displace anything in the atmosphere, other than perhaps grabbing oxygen to bond with the carbon, but that oxygen is still in the atmosphere from which it came. Part of me is thinking that maybe a CO2 molecule released from combustion might cause the ocean to release CO2 more slowly. Yes, the atmosphere is open ended, but there is still pressure exerted on the ocean surface and temperature of both the atmosphere and the ocean will determine how much CO2 will be released. If I burn a gallon of gasoline in my SUV and add CO2 to the astmosphere, I’ve increased the atmsophere’s weight upon the ocean. If I don’t burn that gallon, the ocean, depending upon temperatures and pressure, will release a molecule of CO2 regardless.
Well, Roy, I wish you hadn’t told the zealots which journal to start ratcheting up their pressure campaign against publishing your new article “for the good of the human race” or somesuch.
Re: Allan M R MacRae (00:27:14)
Yes I know it – looks like the experimental confirmation of the “constant opacity” or “self-regulating greenhouse effect” theories above. Lets see whether those trends will change with present cooling.
Oh my gosh!!!
The great Rushbo just mentioned Anthony’s Blog!!
Woohoo ;-D
This makes a lots of sense to me.
Christopher Booker’s article in today’s Daily Telegraph “Rise of sea levels is ‘the greatest lie ever told'” (link provided by Satellite Lover (20:37:05)) is 5th on the Telegraph’s ‘most viewed today’ list. Yes!!!
That is two more nails in the AGW coffin today.
Aron (03:05:44) :
You are the one bringing up this ad hominem attack on Dr. Spencer claiming his religious beliefs are distorting his scientific findings. You have brought no evidence of that. Your post is a descpicable attempt to deprecate a valiant effort to find the truth about what drives our climate.
Ohioholic (21:48:45) :
CO2 reacts with water vapour, in a endothermic reaction, removing heat, and forming H2CO3, carbonic acid, which in turn form carbonates, and so on. Wherever there is CO2 around there is cold around.( think the icecreams 🙂 )
All this GW nonsense is just a pretext or justification to promote political measures which will suposedly make freer, more equal and more “open” societies (“”,after the George Soros model). My theory is that this is a continuation of the “french revolution”. intended to democratize the world.
They, whomsoever they are, think of themselves a the saviors of humanity. I think of them just as plain fools not taking into account the role played by nature itself which can abort such an “altruistic” agenda.
As a retired Geologist, Geophysist, Engineer and Professional Environmentalist I’ve run across too many cases where the theory sounded good but crashed on “Ground Truthing”. To my limited mind the the IPCC climte models have consistantly failed the Ground Truthing test and needed to be revised or abandoned.
I’ve peer reviewed a number of papers that were deficient in the QA/QC of the input data, and/or came to conclusions that were only scantily supported. The papers were published mainly, I believe, because the conclusions conformed to “common knowledge”.
I’ve been involved in cases where the obvious wasn’t. One case involved Minamata Disease (Mercury poisioning) in an aborigional population in northern Quebec. Initially it was blamed on the local mines and pulp mills. We insisted that a proper baseline study be done. The expanded study discovered a pristine drainage system which was a major source of the native’s fish was highly (and naturally) contaminated with mercury.
Another case involves Selenium leaching from mine dumps resulting in “toxic “levels of Selenium to fish downstream from the mines. Fortunately the trout are unable to read and thrive in river and the mine’s sedimentation control works. The river is classified as “trophy water” and is a mecca for fly fishers.
I’ve constructed geostatistical models, most of which failed the ground truthing tests due to faulty assumptions or a lack of adequate data.
In summary nature does not care what the public or politicians believe is the whole truth.
Keep up the good work. Good science is run by skeptics.
http://online.wsj.com/article/SB123837276242467853.html
This is the link to a Wall Street Journal editorial: “Cap and Trade War.” Its main stream view is extremely useful.
Anthony:
A typo in Dr. Spencer’s comments: In paragraph 6, the word “about” should probably be “amount”.
Hi Gordon,
It’s a chicken & egg problem as well. Without good data how does one construct a valid model? And without a good model how does one interpret good data?
When the bulk output from a series of functions keeps exceeding the demonstrated real-world data, there has to be a data flaw in the inputs or range & feedback flaw in the process. Gee, ten years of stable temperatures, who would’ve thought?
The climatology field should take a look at the risk analysis problems in the financial meltdown. The best experts can get burned by hidden functional problems in their analytic process.