Simple Question, Simple Answer… Not
Guest commentary by Spencer R. Weart, American Institute of Physics
I often get emails from scientifically trained people who are looking for a straightforward calculation of the global warming that greenhouse gas emissions will bring. What are the physics equations and data on gases that predict just how far the temperature will rise? A natural question, when public expositions of the greenhouse effect usually present it as a matter of elementary physics. These people, typically senior engineers, get suspicious when experts seem to evade their question. Some try to work out the answer themselves (Lord Monckton for example) and complain that the experts dismiss their beautiful logic.
The engineers' demand that the case for dangerous global warming be proved with a page or so of equations does sound reasonable, and it has a long history. The history reveals how the nature of the climate system inevitably betrays a lover of simple answers.
The simplest approach to calculating the Earth's surface temperature would be to treat the atmosphere as a single uniform slab, like a pane of glass suspended above the surface (much as we see in elementary explanations of the "greenhouse" effect). But the equations do not yield a number for global warming that is even remotely plausible. You can't work with an average, squashing together the way heat radiation goes through the dense, warm, humid lower atmosphere with the way it goes through the thin, cold, dry upper atmosphere. Already in the 19th century, physicists moved on to a "one-dimensional" model. That is, they pretended that the atmosphere was the same everywhere around the planet, and studied how radiation was transmitted or absorbed as it went up or down through a column of air stretching from ground level to the top of the atmosphere. This is the study of "radiative transfer," an elegant and difficult branch of theory. You would figure how sunlight passed through each layer of the atmosphere to the surface, and how the heat energy that was radiated back up from the surface heated up each layer, and was shuttled back and forth among the layers, or escaped into space.
When students learn physics, they are taught about many simple systems that bow to the power of a few laws, yielding wonderfully precise answers: a page or so of equations and you're done. Teachers rarely point out that these systems are plucked from a far larger set of systems that are mostly nowhere near so tractable. The one-dimensional atmospheric model can't be solved with a page of mathematics. You have to divide the column of air into a set of levels, get out your pencil or computer, and calculate what happens at each level. Worse, carbon dioxide and water vapor (the two main greenhouse gases) absorb and scatter differently at different wavelengths. So you have to make the same long set of calculations repeatedly, once for each section of the radiation spectrum.
It was not until the 1950s that scientists had both good data on the absorption of infrared radiation, and digital computers that could speed through the multitudinous calculations. Gilbert N. Plass used the data and computers to demonstrate that adding carbon dioxide to a column of air would raise the surface temperature. But nobody believed the precise number he calculated (2.5ºC of warming if the level of CO2 doubled). Critics pointed out that he had ignored a number of crucial effects. First of all, if global temperature started to rise, the atmosphere would contain more water vapor. Its own greenhouse effect would make for more warming. On the other hand, with more water vapor wouldn't there be more clouds? And wouldn't those shade the planet and make for less warming? Neither Plass nor anyone before him had tried to calculate changes in cloudiness. (For details and references see this history site.)
Fritz Möller followed up with a pioneering computation that took into account the increase of absolute humidity with temperature. Oops… his results showed a monstrous feedback. As the humidity rose, the water vapor would add its greenhouse effect, and the temperature might soar. The model could give an almost arbitrarily high temperature! This weird result stimulated Syukuro Manabe to develop a more realistic one-dimensional model. He included in his column of air the way convective updrafts carry heat up from the surface, a basic process that nearly every earlier calculation had failed to take into account. It was no wonder Möller's surface had heated up without limit: his model had not used the fact that hot air would rise. Manabe also worked up a rough calculation for the effects of clouds. By 1967, in collaboration with Richard Wetherald, he was ready to see what might result from raising the level of CO2. Their model predicted that if the amount of CO2 doubled, global temperature would rise roughly two degrees C. This was probably the first paper to convince many scientists that they needed to think seriously about greenhouse warming. The computation was, so to speak, a "proof of principle."
But it would do little good to present a copy of the Manabe-Wetherald paper to a senior engineer who demands a proof that global warming is a problem. The paper gives only a sketch of complex and lengthy computations that take place, so to speak, offstage. And nobody at the time or since would trust the paper's numbers as a precise prediction. There were still too many important factors that the model did not include. For example, it was only in the 1970s that scientists realized they had to take into account how smoke, dust and other aerosols from human activity interact with radiation, and how the aerosols affect cloudiness as well. And so on and so forth.
The greenhouse problem was not the first time climatologists hit this wall. Consider, for example, attempts to calculate the trade winds, a simple and important feature of the atmosphere. For generations, theorists wrote down the basic equations for fluid flow and heat transfer on the surface of a rotating sphere, aiming to produce a precise description of our planet's structure of convective cells and winds in a few lines of equations… or a few pages… or a few dozen pages. They always failed. It was only with the advent of powerful digital computers in the 1960s that people were able to solve the problem through millions of numerical computations. If someone asks for an "explanation" of the trade winds, we can wave our hands and talk about tropical heating, the rotation of the earth and baroclinic instability. But if we are pressed for details with actual numbers, we can do no more than dump a truckload of printouts showing all the arithmetic computations.
I'm not saying we don't understand the greenhouse effect. We understand the basic physics just fine, and can explain it in a minute to a curious non-scientist. (Like this: greenhouse gases let sunlight through to the Earth's surface, which gets warm; the surface sends infrared radiation back up, which is absorbed by the gases at various levels and warms up the air; the air radiates some of this energy back to the surface, keeping it warmer than it would be without the gases.) For a scientist, you can give a technical explanation in a few paragraphs. But if you want to get reliable numbers - if you want to know whether raising the level of greenhouse gases will bring a trivial warming or a catastrophe - you have to figure in humidity, convection, aerosol pollution, and a pile of other features of the climate system, all fitted together in lengthy computer runs.
Physics is rich in phenomena that are simple in appearance but cannot be calculated in simple terms. Global warming is like that. People may yearn for a short, clear way to predict how much warming we are likely to face. Alas, no such simple calculation exists. The actual temperature rise is an emergent property resulting from interactions among hundreds of factors. People who refuse to acknowledge that complexity should not be surprised when their demands for an easy calculation go unanswered.
8 September 2008 at 7:13
Gilbert’s figure of 2.5 thought, I think, a very good starting point.
The question being asked is “how does greenhouse gasses affect the earths temperature?”. And this one answers “If you consider only CO2 (which we are responsible for), 2.5 degrees per doubling.”
Now if someone wants to gainsay that, they are no longer asking the question, are they. They are asking for an answer that they like.
Now you can just tell them “well, you do the maths then and we’ll look over the result, but by default, this is what would happen and you must take in any changes that occur whether they increase sensitivity or reduce it”. Or just ignore them and let them know that if they wanted a different answer, they should have asked a different question.
8 September 2008 at 7:36
I’m trying to do some laymen-level research on various change models predicting climate winners and losers in North America, particularly related to areas to retire to in, say, 20 years. Where do you think I could find some understandable middle of the road graphics, etc?
8 September 2008 at 7:46
Thanks! This is a great (and eloquent) description of the incrementing complexity required of models to go from the interactions between atoms and photons to those between Earth system components.
As someone who’s used to working with complex (= simple; compared to reality!) models of biogeochemistry, I’ve gotten so used to requiring simulations to explore behaviour that I no longer give it a second thought. This has given me an institutionalised mindset that, at times, blinkers my understanding of why others misunderstand. This article is a helpful, if obvious, reminder that not everyone else is in the same boat!
Best regards,
Andrew.
8 September 2008 at 8:19
#3 Andrew: I’m trying to figure out whether your question is off-topic or not! To get reliable predictions for climate on regions as small as a US state will take more computing power and understanding of climate processes than we have at present. So far, experts are pretty sure that the Southwest from Calif. to Texas will have worse heat waves and drought (=forest fires), which may have started already. This is pretty robust; people have been predicting it for decades. Most other US regions aside from Alaska don’t seem too sensitive to near-term climate change… ironically, the country most responsible for the greenhouse gases in the atmosphere is going to feel the effects later than many innocent countries. We do expect everywhere the generalities of climate change: hotter summer nights and later frost, more variability such as more intense droughts and rainstorms. And don’t get beachfront property near high tide if you are looking more than a couple of decades ahead. All that was predictable from simple hand-waving considerations, and indeed has been predicted for half a century.
If you have access to Science magazine, see the report Aug. 15 p. 909 of an article on this in press in Geophysical Research Letters, with a nice graphic.
8 September 2008 at 8:23
The degree of complexity takes the issue into a region where people not intimate with the complexity are simply outsiders. Since there are consequences to the issue that are the opposite of “academic”, those who want to stir the world to action are on the horns of a dilemma. What they want in the way of response and what’s possible for them to communicate to the layman are at such a remove that it will be a miracle if anything actually happens to stem climate change. In an ideal world, “Trust us” coupled with a well-organized primer (like the IPCC reports) would be sufficient, but as we all know and as Fate would have it, there are Other Interests involved. And as Upton Sinclair heartrendingly pointed out, “It’s hard to get a man to understand something when his paycheck depends upon him not understanding it.” And there are lots and lots of paychecks involved.
8 September 2008 at 8:28
As an engineer, and one that has worked his entire career on issues involving energy and the environment, and most specifically on air/atmospheric issues, I take some exception to the characterization of “engineers.”
While I realized long ago that simplistic equations and approaches can get you nonsense as an answer, I now tell people who attempt this that only when they’ve gone through the process line by line, layer by layer will they have some sense of what is involved. As noted, many will simply give up realizing that they have not the scientific prowess or time to go through the exercise.
That said, I do often point out the most straight forward of the simple laws that govern the processes involved and point out that the theory does not rely on correlations of measurements to create GHG theory. Rather, I point out that the correlations support the underlying physics and are not just some random correlations without some underpinning science to them. One need only go to such sites as http://www.venganza.org/about/open-letter/ and look at the relationship between global temperature and the number of pirates.
As an engineer attempting to deal with the consequences, I like to know what future I am planning for.
8 September 2008 at 8:33
Thanks, Spencer, for a very lucid explanation. Most senior engineers, however, should understand this level of complexity and the confidence one must have in solutions obtained by numerical integration using digital computers. Otherwise, we never could have launched a single missile or space vehicle.
8 September 2008 at 8:33
“Physics is rich in phenomena that are simple in appearance but cannot be calculated in simple terms. Global warming is like that. People may yearn for a short, clear way to predict how much warming we are likely to face. Alas, no such simple calculation exists. The actual temperature rise is an emergent property resulting from interactions among hundreds of factors. People who refuse to acknowledge that complexity should not be surprised when their demands for an easy calculation go unanswered.”
Ah! But then people who offer “simple” solutions to this complex problem shouldn’t be so surprised when they are asked to provide the simple proof, should they? It is a double edged sword, this complexity. If you are asking for change, shouldn’t you be required to accurately predict what your change will produce? But apparently that isn’t possible.
8 September 2008 at 9:47
> If you are asking for change, shouldn’t you be required to accurately predict
> what your change will produce?
You’ve stated the precautionary principle. But as Dr. Weart points out, accurate prediction was available.
8 September 2008 at 9:50
Would it be possible to have an actual senior engineer present their (presumably, mainstream) views of anthropogenic climate change and of the use of models?
As a (senior? and scientifically trained) engineer myself, I can guess what Mr Weart is aiming at, but he’s still using a language that brings down no barrier. For example, a statement such as
“Gilbert N. Plass used the data and computers to demonstrate that adding carbon dioxide to a column of air would raise the surface temperature”
will and does definitely make people suspicious.
You see, I have seen dozens, and I am sure there are out there hundreds of thousands of designs that have been “demonstrated” in a computer only to fail miserably when put into practice.
In fact, one point that I don’t think Mr Weart realizes (and likely, it’s all part of the miscommunication) is that it’s the engineers that have to deal with the actual world out there, and all its complexity, starting from but having to go beyond what calculations (formulae and/or models) suggest.
It really is the job of engineers to understand the complexity of the real world, and to make things work within that complexity.
There is little point in arguing to your manager that, say, in the computer your revolutionary design of a car needs only 2 gallons per 100 miles, when the actual thing is measured as drinking much more than that.
The one rule common to all engineered system is, the more stuff you put in, the higher the chances that something will go wrong. In Mr Weart’s case: the more factors need to be made to interact using models and supercomputers to calculate “global warming”, the higher the chance that the computed answer won’t be the right one.
Therefore, rather than accusing engineers of looking for simple answers (likely, misunderstanding them), Mr Weart should try to bridge the gap.
An example of another scientific endeavor, apart from climate change, where extremely complex, just-made modelling has been successfully applied as-is into an engineering project, would definitely be a good starting point.
8 September 2008 at 9:54
http://www.timesonline.co.uk/tol/news/environment/article4690900.ece
8 September 2008 at 10:04
For those who want a “simple” place to start, the work of the 1920’s on understanding the structures of stellar interiors is worth review. A number of approximations, many developed by Eddington, made calculation tractable with the methods of those days, often a room full of woman calculators. Much more detailed opacities are used now and the atmospheres of stars above their photospheres also get detailed treatment. It turns out that the physics is not controversial, just complicated. It is a fun study, but give it a few months to start to get a grip on it.
8 September 2008 at 10:15
Michael (#8): Change of some kind or other is going to happen– it always does; it’s not a choice between ‘keeping things the same’ and ‘changing them’. Why do those who advocate a reduction in our GHG emissions owe you a ’simple’ proof of what the consequences of their proposal would be, while you don’t owe them a simple proof that continuing our GHG emissions won’t have disastrous consequences? Are you foolish enough to just say, ‘well, there’s no disaster yet’, so maybe it’s OK? If so, consider this old joke: a man falling from the top of the Empire State building falls past the 20th floor and someone yells at him, ‘how are you’– he answers, ‘alright so far’… The evidence that serious problems are already developing, and that much worse are yet to come, is very strong.
8 September 2008 at 10:21
Michael (#8), just because you can’t predict something in a page doesn’t mean you can’t “accurately predict” it.
Consider an example from pure mathematics: Most results that are at all deep haven’t (and likely can’t) be proven in one page or ten or in many cases even a hundred. But that doesn’t undermine the validity of the proof, say, of the solvability of groups of odd order (a famous example from 1963 of a very simple, very important result that took about 250 pages of very dense reasoning and hasn’t been materially condensed in the four decades since then).
8 September 2008 at 10:27
Michael #8: It would appear that we have what is actually a pretty simple argument here, though based on complex diagnostics: The patient says, “Doctor, Doctor, it hurts when I raise CO2 levels.”
To which the Doctor riplies: “Don’t do that.”
Once we have established a causal link between CO2 and climate–and by any reasonable standard, we have–if we remove the cause, we avoid the effect. It would be much more difficult to justify any sort of geoengineering aproach without a huge level of effort by your standards, but reducing CO2 emissions is a no-miss solution.
8 September 2008 at 10:28
Reading the paper I came to think of an experience I had, a (large) number of years ago, with what then was called micro simulation. A number of social scientists attached probabilities to each individual in a region as to for instance their demographic and economic behaviour and made assumptions as to how they all would react to the behaviour of all the others. (Obviously very complicated calculations, far beyond the grasp of a simple central bureaucrat = me.) What made me suspicious (and here we have my link to the present paper) was that the scientists said that they after each computation checked the aggregate results against macro data on the region in question and when they got “monstrous” and “unrealistic” descriptions of the future region they changed their assumptions untill the result was “reasonable”. This “method” made me very doubtful. Do you see the parallel?
8 September 2008 at 10:28
Mauritzio, #10.
OK, so please furnish us with an Earth Mk II so we can avoid having to use a mathematical simulation.
Or do you have any explanation as to WHY the simulations are incorrect?
Or are you just looking for an answer you’re happy with?
8 September 2008 at 10:30
Great post! Does anyone know where can one obtain a fairly detailed laypersons description of modern bio-geo-chemo-climate models anywhere that lists all the positive and negative feedbacks?
8 September 2008 at 10:51
If you are asking for change, shouldn’t you be required to accurately predict what your change will produce? But apparently that isn’t possible.
In this country, we’re terribly familiar with acting upon insufficient information about a potential catastrophe. Our current bill for that “1% possibility” will reach more than $1 trillion dollars before the last veteran maimed by the war is laid to rest.
Compare the transparency of the science with the political shenanigans which prefaced the war. Yes, there’s complexity with the science of AGW, but complexity is a “mote” compared with the “beam” of the 3 Card Monty game involved in manipulating intelligence.
Much of the change encouraged by scientists ought to dovetail nicely with the necessity brought about by Peak Oil and the good involved in extricating our political choices from the swamp of Mid Eastern grievances. The solution to all these things looks remarkably similar: burn a lot less carbon.
8 September 2008 at 11:02
Maurizio Morabito, I would not attach great importance to Spencer’s use of “engineer” as questioner. He is merely taking an engineer as an example of an educated professional not acquainted with details of climate science. As indicated by debacles in the APS Forum on Physics and Society and elsewhere, one could equally take “physicist,” “statistician,” or even “meteorologist.”
The point stands that without concerted effort to understand details of the science, even an educated person will come to incorrect conclusions. Nobody is saying that the models reproduce reality, but rather that they give us insights into how the real world is working as long as we have prepared ourselves by decades of study.
8 September 2008 at 11:06
“You don’t need a weatherman to know which way the wind blows”
Is there a similar phrase for climatology?
Maybe, ‘You don’t need a climatologist to tell which way the heating goes’
Doesn’t quite have the rhythm and poetry of Bob Dylan. We all want elegant and simple statements that reveal truth about change.
8 September 2008 at 11:20
#4: Oops. I don’t think I asked a question. I was just shamelessly offering praise.
I thought that your piece was a very clear articulation of why one can’t simply jump from a handful of first-order equations to a prediction of the consequences of a doubling of CO2. Aside from the plethora of processes that even complicated climate models exclude, there’s a big chunk of simulation necessary to tease out the all of the emergent properties necessary for answering even seemingly simple climate questions.
Anyway, great article.
Andrew.
8 September 2008 at 11:45
I’d like to see the models take into account Hurricanes. These are vast dissapative systems. But they may be the least of them.
8 September 2008 at 12:02
Following up on #11. If modeling problems are so complex that they defy closed-form solution, a sound engineering response, in my experience, is to establish sufficient correlation of the model to reality before using it to project new scenarios. If it is not capable of predicting reality, the usual request is to go look for the missing variables or relationships until it does correlate sufficiently. 3 questions then:
(1) How well have climate models of 10-25 years ago predicted climate-scale change over the last 10-25 years?
(2) Do climate models contain the physical models/factors that recently resulted in the toggle of the PDO and resulting projections by some that we might see a delay in global warming for 10 years or so? And if so, why did they not predict it?
(3) If these factors are not present, what efforts are ongoing to address this?
Can you point to any papers or discussions on this topic if the answer is too involved to give here?
8 September 2008 at 12:10
A question about the basic physics explanation:
Earth is in radiative equilibrium, at least over the long term. In a basic-physics explanation, it is better to describe the present warming trend as a transient response (as the system seeks out a new equilibrium at higher temperature) or a steady-state response (as a new value of opacity enters into the radiative balance)?
George
(forgive me if this is a duplicate posting - I’m having trouble with the captcha)
8 September 2008 at 12:12
Spencer Weart, it’s a good question: how can you calculate global warming mathematically?
Here’s my contribution: a link to an essay that I published recently for nonscientists, “The Case for Modern Anthropogenic Global Warming”. It’s a reply to the contrarian arguments advanced by Alexander Cockburn in 2007 on the CounterPunch website and in the pages of The Nation.
http://monthlyreview.org/080728farley.php
Because the essay was written for a nontechnical audience, all equations have been banished to endnotes.
In this essay, endnotes 12 and 15 give some simple calculations.
It IS possible to calculate, using the Stefan-Boltzmann equation, the temperature of a hypothetical Earth with no greenhouse gases in the atmosphere. It’s below the freezing point of water. This shows that the greenhouse effect is a BIG effect.
People are often surprised that global warming, caused by the enhanced greenhouse effect, is large enough to cause a problem, even though carbon dioxide is only present in the atmosphere at a level of hundreds of parts per million. Instead, the real surprise is that global warming is a small as it is!
-John Farley
Professor of Physics, UNLV
8 September 2008 at 13:40
Re Michael @8: “If you are asking for change, shouldn’t you be required to accurately predict what your change will produce? But apparently that isn’t possible.”
And that impossibility is what makes it so risky: we know in general what will happen, but we don’t know exactly what will happen, which is why we can’t rule out the worst case scenario. Sounds like precisely a time to invoke the precautionary principle, no?
8 September 2008 at 13:44
Since the GHG sensitivity estimates are completely based on the results of climate models, a method of testing of how accurate the computer models are must be undertaken.
There is a clear (moral, social, economic and scientific) responsibility to do so given what is at stake.
[Response: That isn’t the case (sensitivity estimates are also made via paleo-climate and modern changes), and secondly, that testing is already being undertaken (read the IPCC AR4 report for an assessment of that work. - gavin]
8 September 2008 at 13:53
as far as simple a proof goes, not being a scientist myself i’m pretty happy with ockham’s razor (though final year of high school physics is also pretty useful). i find it entirely plausible that if you pump 37% more of a potent and important GHG into the atmosphere, then you can expect it to have an impact.
as for engineers, i work with them regularly (civil and mechanical mainly) and they reliably and usefully design structures with a safety margin built in. so, if a beam will be reasonably expected to carry X load, it will actually be designed to carry say a 10X load. this is basically the “just to be on the safe side” factor, because complex systems are necessarily difficult to predict.
the important and entirely reasonable thing here though is that they design things on the safe side because the consequences of not doing so could be devastating.
and this is only for one beam on one floor of one building, we’re talking about the very planet that sustains us.
i’m not particularly worried about super accurate predictions, the stakes are just so high and human behaviour and the climate system so complex, i’d rather play it safe with my future (and my children’s, your’s, your children’s, etc, etc).
the quest for super accurate predictions actually reminds me of a Borges story where a people are so obsessed with making the perfect map they end up with a map as big as the land it maps. we can wait for perfectly accurate predictions but they will be useless because they will only ever be perfectly accurate *after* the event has been observed.
8 September 2008 at 14:01
Jeffrey Davis says: “Compare the transparency of the science with the political shenanigans which prefaced the war.”
Oh, let’s not, shall we? I’d really like to hold science to a higher standard than politics–particularly the politics of a kleptocratic regime.
[Response: Agreed. The war is OT. - gavin]
8 September 2008 at 14:11
Gösta Oscarsson says: “What made me suspicious (and here we have my link to the present paper) was that the scientists said that they after each computation checked the aggregate results against macro data on the region in question and when they got “monstrous” and “unrealistic” descriptions of the future region they changed their assumptions untill the result was “reasonable”. This “method” made me very doubtful. Do you see the parallel?”
Actually, not at all. Big difference between this and the dynamical approach used for climate modeling.
8 September 2008 at 14:13
Re John Farley @26, your Monthly Review critique of Cockburn’s arguments is a thorough and valuable resource, but I fear your rebuttal of Contrarian Claim 3, the lag of CO2 increase behind temperature rise at the end of an ice age, has a serious deficiency in that it does not describe the subsequent role of CO2 as a feedback in adding more warmth, and thus the ability of CO2 to act as both a feedback in the case of deglaciation, and as a forcing when added directly to the atmosphere as at present.
8 September 2008 at 14:24
John Lang Says:
8 September 2008 at 1:44 PM
” Since the GHG sensitivity estimates are completely based on …” and then states his misunderstanding.
John, where did you get this mistaken idea? Did you read it somewhere? Could be you’re misreading a good source, or you’re being fooled by fake “facts” — ??
8 September 2008 at 14:31
“These people, typically senior engineers, get suspicious”.
Please, can we get deeper than “senior engineers” - that really isn’t improving insight. If we want to do that, we need to probe a lot deeper than just “senior engineers”.
Let me offer a speculation, although not yet a serious hypothesis:
1. SPECULATION
Amongst technically-trained people, and ignoring any economic/ideological leanings:
1) Some are used to having
a) Proofs
OR
b) Simple formulae
OR
c) Simulations that provide exact, correct answers, and must do so to be useful
d) And sometimes, exposure to simulations/models that they think should give good answers, but don’t.
2) Whereas others:
a) Are used to missing data, error bars,
b) Complex combinations of formulae
c) Models with varying resolutions, approximations, and that give probabilistic projections, often only via ensemble simulations.
d) Models that are “wrong”, but very useful.
My conjecture is that people in category 1) are much more likely to be disbelieving, whether in science, math, or engineering.
2. ANECDOTAL EXAMPLES:
1) In this thread, a well-educated scientist (Keith) was convinced that climate models couldn’t be useful, because he was used to models (protein-folding) where even a slight mismodel of the real world at one step caused final results to diverge wildly … just as a one-byte wrong change in source code can produce broken results.
See #197 where I explained this to him, and #233 where light dawned, and if you’re a glutton for detail: #66, #75,l #89, #1230, #132, #145, $151, #166 for a sample.
2) See Discussion here, especially between John O’Connor & I. See #64 and #78. John is an EE who does software configuration management. When someone runs a rebuild of a large software system, everything must be *perfect*. There’s no such thing as “close”.
Also in that thread, Keith returned with some more comments (#137) and me with (#146), i.e., that protein-folding was about as far away from climate modeling as you could get.
3) Walter Manny is a Yale EE who teaches calculus in high school. He’s posted here occasionally (Ray may recall him :-), and participated in a long discussion at Deltoid, and has strong (contrarian) views. In many areas of high school/college math, there are proofs, methods known for centuries, and answers that are clearly right or wrong.
4) “moonsoonevans” at Deltoid, in #21 & #32 describes some reasons for his skepticism, #35 is where light dawns on me. He’s in financial services, had experienced many cases where computer simulations done by smart people didn’t yield the claimed benefits. In #35 I tried to explain the difference.
All this says that if one is talking with an open-minded technical person, one must understand where *they* are coming from, and be able to give appropriate examples and comparisons, because many people’s day-to-day experience with models and simulations might lead them to think climate scientists are nuts.
3. A FEW SPECIFIC DISCIPLINES & CONJECTURES
1) Electrical engineers (a *huge* group, of which only tiny fraction are here)
Many EEs these days do logic design, which requires (essentially) perfection, not just in the design, but (especially) in simulation.
Design + input =>(logic simulator) => results
At any step, the design may or may not be bug-free, but the simulator *must* predict the results that the real design would do given the input, exactly, bit for bit. Many test-cases have builtin self-checks, but the bottom line is that every test-case yields PASS or FAIL, and the simulator must be right.
Many people buy simulators (from folks like Cadence or Synopsys), and run thousands of computers day and night simulating millions of test-case inputs. But, with a million test-cases, they’re not looking for an ensemble that provides a distribution, they’re looking for the set of test-cases to cover all the important cases, and for EVERY one to pass, having been simulated correctly. This has some resemblance to the protein-folding problem mentioned above.
Now, at lower levels of timing and circuit design, it isn’t just ones and zeroes (there’s lots of analog waveforms, probabilistic timing issues, where one must guarantee enough margin, etc). When I’d tease my circuit designer friends “Give me honest ones and zeroes”, they’d bring in really ugly, glitchy HSPICE waveforms and say “so much for your ones and zeroes”. (This is more like the molecular “docking” problems that Keith’s colleagues mentioned.)
At these levels, people try to set up rules (”design rules”) so that logic designers can just act at the ones-and-zeroes level.
If one looks at EEs who worry about semiconductor manufacturing, they think hard about yields, failure attribution, and live with time-series. (Standard answer to “We got better yield this month, how do you think it looks?” was “Two points don’t make a trend.”
2) Software engineers
Programs often have bugs, but even a bug-free program can fall apart if you change the wrong one byte of code, i.e., fragile. (I don’t recall the source, but the old saw goes something like: if skyscrapers were like software, the first woodpecker would knock them over.)
Configuration management / software rebuilds are fairly automated these days, and they must be correct. One cannot include the wrong version of code, or compile with incompatible options.
Performance engineering and benchmarking tend to be more probabilistic-oriented, and although a lot of people want to believe in one number (once the mythical “MIPS” rating), we’ve (mostly) fixed that over the last 20 years. Good performance engineers have always given relative performance ranges and said YMMV (Your Mileage May Vary).
3) Mechanical engineers
This, I expect, varies. In some cases, closed-form equations work pretty well. In other cases, one is using big structural dynamics and computational fluid dynamics codes to obtain “good-enough” approximations to reality before actually building something. For example, automobiles are extensively modeled via “crash codes”.
4) Petroleum engineers
It’s been a while, but certainly, people who do seismic analysis and reservoir modeling *start* with data from the real world, analyze it to make decisions, so ought to be a little more accustomed to probabilistic analyses.
5) Financial engineers (Google: financial engineering)
Not having physics to constrain simulations yields some wild results, although at least, some people are very comfortable with risk, uncertainty, and ensemble projections. I especially like Sam Savage’s Flaw of Averages”.
On the other hand, when Nobel Economists lose $B (LTCM), I’m not surprised there is skepticism about climate models.
4. CONCLUSIONS
That’s a speculative start. I do *not* think lumping a large group together as “senior engineers” helps progress, because I have at least anecdotal evidence that the sources of skepticism tend to be attached to the kinds of models and (maybe) simulations that someone does day-by-day. The problem is that many people tend to generalize from their discipline to others, and especially if they have trouble getting useful models, they tend to be suspicious of others’.
At one extreme, people have long-established mathematical proofs, and answers that are clearly right or wrong.
At the other extreme, people have to make decisions based on the best approximations they can get, and if their discipline has good-enough approximations, they tend to think one way, and if the approximations aren’t so good, they may think another about equations and climate models.
8 September 2008 at 14:32
re: 28
“given what is at stake.”
If the likelihood of a ruinous car crash were greater than 2.5% per year, the insurance companies wouldn’t let me in the door. By the time we get to a rise of 2C, I think the issues of precaution and mitigation are going to be moot. One can’t divvy up probabilities exactly, but the IPCC pegs a rise in temps of between 1.5-4.5C at the 95% level of certainty. Do you think you could get car insurance at that level of risk?
8 September 2008 at 14:41
Re 27: Jim Yeager:
“Sounds like the time to invoke the precautionary principle.”
It depends on the cost of the precaution, the magnitude of adverse consequences, and the degree of certainty.
If, to bring down CO2 levels to 275 ppm or so, we need a global WWII type mobilization, rationing, and restriction of freedoms, then we better have one heck of a solid case!
8 September 2008 at 14:44
John Farley, #26. I followed your link, and read your footnote 15 which is of particular interest to those (like myself) who are looking for methods to estimate climate sensitivity without using GCMs:
The problem with this argument is the glacial - interglacial temperature increase is accompanied by a large decrease in the amount of ice and snow covering the earth, and hence in the amount of sunlight reflected back towards space. Today there is relatively little ice left, and what there is is near the poles which contributes little to reflection (the poles are dark for half the year and for the other half the sun is on the horizon so they present very little “cross-section” to the sun). Thus, we would expect a much stronger positive feedback from glacial to interglacial than we would today from a similar increase in forcing.
[Response: That is already factored in since the ice sheets are imposed as a forcing in this kind of calculation. In general, you are probably correct - the existence of extensive snow/ice cover increases sensitivity, but for the ranges of climate change we expect in the next 100 or so years that does not seem to be a big effect. - gavin]
8 September 2008 at 15:17
Re Fred Jorgensen @36: “It depends on the cost of the precaution, the magnitude of adverse consequences, and the degree of certainty.”
Fred, you left out the cost of not taking the precaution, and the magnitude of adverse consequences of not taking the precaution, yet you specifically set the bar at 275 ppm.
Now why would you do that?
8 September 2008 at 16:11
Dear Maurizio Morabito #10,
I think there is actually a more encouraging explanation to Spencer Weart’s observations, which also supports your statement.
Firstly, at our excellent universities we train scientist and engineers in the basics of their respective fields during their undergraduate studies.
Secondly, during their graduate studies we encourage them to go beyond the basics, discover new things, and go beyond the current knowledge and acquire new knowledge on their own. One way to do this is challenging hypothesis, theories, assumptions of complex models and I guess even sometimes the consensus.
That is how progress is made at the universities and in the six sigma breakthrough methodology in the industry.
Thirdly, in industry we also train them to look for simple rules and models that can be implemented with ease in a production process. This is sometimes necessary to make money.
Actually, I think the linear feedback model that was adapted by climate science is just a typical linearization you would apply in industry to model a more complex problem.
I also agree with you, many of them therefore understand the complexities of real world problems.
Therefore, I am very optimistic that with the help of all those excellently trained scientists and engineers mankind will survive the warming.
8 September 2008 at 16:19
Fred Jorgensen (36) — Don’t need any of that, just 1–2% of the world’s gross product for 70–100 years. Probably it can be done for quite a bit less eventually. That shouldn’t stop us from starting now.
8 September 2008 at 16:57
Spencer Weart, thanks for the great writeup.
Gavin,
concerning your last response to mugwump, I guess I assumed incorrectly that Milankovitch forcing was amplified by reflective and GHG feedbacks… but instead ice sheet responses are considered a forcing? I may just be getting tripped on words rather than anything radiatively significant, (I’m assuming that the “forcing” part is a reduction in surface albedo, though the semantics make no sense to me). How exactly are ice sheets to be considered a forcing, unless we consider freshening of the THC or something along those lines? I do agree with the comment though, that under situations of a same radiative forcing, two Earth’s (one with ice, one with little ice) would expect differences in sensitivity. But you’d have better insight to the actual quantitative aspects of that than me.
[Response: It’s the difference between the Charney sensitivity (fast feedbacks included, ice sheets, vegetation change not) and the Earth System Sensitivity (all feedbacks included). Most discussion has focused on the former, and that is the context for the LGM calculation. The ESS is different - and indeed if you try and calculate the ESS to orbital forcing you get extreme values (because the regionality/seasonality of the orbital forcing is so important for the ice sheets). - gavin]
8 September 2008 at 17:16
Doug H. wrote:
> Does anyone know where can one obtain a fairly
> detailed laypersons description of modern
> bio-geo-chemo-climate models anywhere
Yes, the IPCC, or the Start Here links at top of pag> that lists all the positive and negative feedbacks?
Nope, it’s not such a simple question that all the answers are completely known even for a single moment in time, and they will change over time.
Example (you can follow articles like this forward in time to see how the questions get worked on):
from twenty-one years ago (before ocean pH change was noticed as an issue):
http://www.see.ed.ac.uk/~shs/Global%20warming/Data%20sources/Charlson.pdf
8 September 2008 at 17:36
Re 38: Jim Eager:
The cost of not taking the precaution would be the ‘Adverse Consequences’ in the broadest term.
NASA’s Jim Hansen recently set the bar at 350 ppm, but there have been calls for pre-industrial,
historically stable levels of 280 or so.
Re. 36: David B. Benson: But the world’s GDP isn’t the benchmark! China and India want US
standard of living, - with the premium for low CO2 technology paid by the developed world.
Much more than 1-2 % of our GDP I would think! And we ‘may’ only have 10 years or so before the
speculative ‘tipping point’! Looks like an enormous project, so we’d better NOT start down that path
without a very high degree of certainty, since the cost of ‘precaution’ (insurance), could
be used to solve more critical problems elsewhere.
8 September 2008 at 17:57
Hank Roberts Says:
8 September 2008 at 2:24 PM
“John Lang Says:
8 September 2008 at 1:44 PM
” Since the GHG sensitivity estimates are completely based on …” and then states his misunderstanding.
John, where did you get this mistaken idea? Did you read it somewhere? Could be you’re misreading a good source, or you’re being fooled by fake “facts” — ??”
That is what the commentary by Spencer Weart says.
8 September 2008 at 18:04
Fred Jorgensen (43) — Well, when I worked it out, 1–2% was enough to
(1) Deeply sequester, as biochar or torrified wood, enough carbon to offset the annual addictions of excess carbon to the active carbon cycle;
(2) Just enough funding left to do the same with about 1% of the existing excess carbon; hence the need for 70–100 years.
There are many variations on the above theme; it could be combined with some CO2 CCS; some of the torrified wood could directly replace coal; etc.
The idea is simply to show that it can be done; we ought to start doing some of it right away. Moving away from fossil fuels will not alone suffice; just concrete production will surely continue to contribute about 0.4 GtC per year.
8 September 2008 at 18:12
John Mashey: Let me offer another speculation. “Senior Engineers” and other Senior Technical Leaders who are worth their salt, think more broadly and have a knack for telling scat from shinola. The people to whom you refer are stuck in a very narrow line of thinking and are unable to see the broader picture.
If anyone out there thinks we engineers are a tough crowd, watch out for the bakers and steelworkers
8 September 2008 at 18:34
The linked presentation should appeal to engineers,
it explains surface temperature in an understandable way
using only gravity, radiation and convection.
The GHG effect is proved unnecessary.
It would be appreciated if someone could point out the error
or provide a link to the rebuttal.
The Thermodynamic Atmosphere Effect
- explained stepwise
by Heinz Thieme
German Version: http://people.freenet.de/klima/indexAtmos.htm
Using a set of technical models of planets with and without an atmosphere the reasons are
explained for differences in surface temperature of the planet without an atmosphere
compared with the temperature in the ground layer of atmosphere of the other planet. The
differences are caused by thermodynamic characteristics of these gases, which cause the
mass and the composition of the atmosphere, and the atmospheric pressure, which results
from gravity.
Trying to avoid the spam filter, link to english version is at…
http://www.geo(largetowns).com/atmosco2/atmos.htm
[Response: Nonsense, I’m afraid. He assumes that the planet has an adiabatic lapse rate (heating with increasing pressure), a very specially chosen amount of mass and radiation only at the edge of the planet. This might work, but does not resemble the real world in any respect. - gavin]
8 September 2008 at 18:42
Interesting, there is no simple answer. I am curious why you didn’t reference the brilliant math used (2 plus 4 divided by 2 equals 3 degrees) to determine climate sensitivity. Tsonis et al has a paper on dynamical models that isn’t all that simple to understand. I do recommend it as a good read though.
8 September 2008 at 18:45
Thanks for your clear explanation of some of the basic concepts Spencer. Your book is very clear and well written too.
One aspect of the “basic” physics that puzzles me is the lag time of CO2 from burning fossil fuels in the atmosphere.
Archer (2007: 123) suggests, “The bottom line is that about 15-30% of the CO2 released by burning fossil fuel will still be in the atmosphere in 1000 years, and 7% will remain after 100,000 years.”
Why is there “only” a lag of 800 years in the icecore record between the earth coming out of a glacial period and the response of CO2?
Reference: Archer D (2007) “Global Warming: Understanding the Forecast” (Blackwell Publishing).
8 September 2008 at 19:06
re #46: “That is what the commentary by Spencer Weart says.”
Not anywhere that I could see; there is a lot about what is needed to model the greenhouse effect, to be sure, but nowhere a statement about what other evidence may or may not exist.
For a bit more about the other evidence, see Gavin’s inline response to the original query at #28, with more at #37.
8 September 2008 at 19:44
> sensitivity estimates are completely based on
John Lang, you say you read that in the Commentary.
It’s not there. Search the words. Is it an interpretation of something you read? What?
8 September 2008 at 19:54
Chris McGrath (49) wrote “Why is there ‘only’ a lag of 800 years in the icecore record between the earth coming out of a glacial period and the response of CO2?”
That’s what is seen in the Antarctic ice cores, although there is a similar delay from Pacfic Warm Pool data. It is a balance between the warming oceans expressing CO2 and the changes in the terrestrial carbon pool.
My amateur reading of the matter.
8 September 2008 at 22:16
Engineers work in the real world. When we build bridges, we make sure ‘they don’t fall down!’
But the real story is: ‘Probably won’t fall down for 100 years.’,
and then we schedule inspections and maintenance, because in the real world ’sh*t happens!’ in
spite of our elaborate models and building standards.
So when climate scientists and theoreticians show us computer projections and fancy statistics, well,
‘We’re from Missouri!’.
The simple graphs of steady CO2 increase and up and down temperature over the last 100 years or so, has
this engineer slightly sceptical.
Sorry. We’re a tough crowd!
[Response: No. You are impossible crowd (well a crowd of one perhaps). How many times have you been told patiently that the impact of CO2 is not derived from correlations over the last 100 years, or 100,000 years? How many times have you been pointed to the physics of radiative transfer or estimates of climate sensitivity? It’s clearly dozens. You haven’t moved from your incorrect idea one iota. Why should anyone continue to discuss with you? - gavin]
8 September 2008 at 23:06
RE John Mashey #34:
Part of my skepticism does indeed derive from my day job, which involves pretty heavy duty computer modeling, but not of the climate. I know how easy it is to overfit when you snoop the test data. In fact, we don’t consider a model validated until we’ve tested it against completely unseen data. Climate modelers have spent years tweaking heavily parameterized models against a very limited set of data. They are almost guaranteed to have overfit.
Another part of my skepticism is a kind of feedback. If you ask skeptical (but informed) questions of many prominent “mainstream” climate scientist bloggers you are often treated with disdain, or worse. That kind of sneering attitude only serves to make me more skeptical.
A third part of my skepticism arises from the extent to which the global warming issue has been hijacked by environmentalists who seek to use it to further their political goals.
The final part of my skepticism arises from the extent to which any published work that deviates from the “consensus” climate change view is immediately eviscerated by blogs such as this and by a flurry of negative “comment” responses in the open literature. The more rigorous the deviant research, the greater the attacks. In the words of WS himself, too often the lady doth protest too much.
At present, the only rational position (in my opinion, of course), is a skeptical one.
8 September 2008 at 23:39
If you want to see how the other side thinks, see my latest interview in which I interviwed a man who does not believe in global warming. NSFW. But his views represent what a lot of Americans think and feel. It’s good to know the other side. I mean, what we are up against. Forget polar cities. This comes first.
http://northwardho.blogspot.com
And while I am here: a professor named Jurgen Scheffran, a research scientist in the Program in Arms Control,
Disarmament and International Security and the Center for Advanced
BioEnergy Research at the University of Illinois, is among those
raising concerns. In a survey of recent research published earlier
this summer in the Bulletin of the Atomic Scientists, Scheffran
concluded that “the impact of climate change on human and global
security could extend far beyond the limited scope the world has seen
thus far.”
Scheffran’s review included a critical analysis of four trends
identified in a report by the German Advisory Council on Global Change
as among those most possibly destabilizing populations and
governments: degradation of freshwater resources, food insecurity,
natural disasters and environmental migration.
“Most critical for human survival are water and food, which are
sensitive to changing climatic conditions,” Scheffran said.
The degradation of these critical resources, combined with threats to
populations caused by natural disasters, disease and crumbling
economic and ecosystems, he said, could ultimately have “cascading
effects.”
“Most critical for human survival are water and food, which are
sensitive to changing climatic conditions,” Scheffran said.
“Environmental changes caused by global warming will not only affect
human living conditions but may also generate larger societal effects,
by threatening the infrastructures of society or by inducing social
responses that aggravate the problem,” he wrote. “The associated
socio-economic and political stress can undermine the functioning of
communities, the effectiveness of institutions, and the stability of
societal structures.
These degraded conditions could contribute to civil strife, and,
worse, armed conflict.”
“cascading effects.”
In addition to global cooperation, Scheffran believes that those
occupying Earth now can learn a lot about the future by studying the
past.
“History has shown how dependent our culture is on a narrow window of
climatic conditions for average temperature and precipitation,” he
said. “The great human civilizations began to flourish after the last
ice age, and some disappeared due to droughts and other adverse shifts
in the climate. The so-called ‘Little Ice Age’ in the northern
hemisphere a few hundred years ago was caused by an average drop in
temperature of less than a degree Celsius.
“The consequences were quite severe in parts of Europe, associated
with loss of harvest and population decline,” Scheffran said. “Riots
and military conflicts became more likely, as a recent empirical study
has suggested.”
8 September 2008 at 23:44
#47, Gavin’s inline response
Gavin,
The Thermodynamic Atmosphere Effect just uses Ray Pierrehumberts formula 3.8 from his online Climate Book as a model, don’t you think?
Ts = (ps/prad)^(R/cp)*Trad
[Response: That’s the adiabat. But it isn’t the greenhouse effect. - gavin]
9 September 2008 at 1:11
Gavin et al.
Is it possible to make the blog reaction always point to the shortest url: e.g. www.realclimate.org/index.php/archives/2008/09/simple-question-simple-answer-no and not e.g. www.realclimate.org/index.php/archives/2008/09/simple-question-simple-answer-no/langswitch_lang/tk
I think that would be helpful…
[Response: Click on the English flag on the sidebar. This behaviour is due to an intersection between the cache and the language-switch code. You can always remove the langswitch tag and reload. - gavin]
9 September 2008 at 2:03
I am now retired, but in my former career, I used to be in charge of the work done by the US government involved in evaluating computer models of nuclear power plants. These programs modeled core physics, fuel thermal-mechanical behavior, and thermal-hydraulics in the core and in the entire nuclear coolant system. These programs were developed over the past 60 years of the nuclear industry by the various nuclear vendors and by the government, as well, at a cost of billions of dollars.
These programs are quite complex, having to take into account a large number of variables and factors, from basic uncertainties in the underlying physics, to uncertainties in material properties, to uncertainties associated with human performance and manufacturing. A LOT of this work was done at the behest of environmentalists in the early 1970s, who complained that the analytical models used to “prove” the safety of the plants was not complete or understandable. Dr. Weart should be very familiar with this effort, inasmuch as he write a very interesting book on nuclear power and nuclear issues (”Nuclear Fear”) in the 1980s.
All of the models that I used to evaluate, and the ones that I used in the government to evaluate those produced by the vendors were extensively documented. The government developed a strict methodology to allow the quantititave evaluation of the uncertainty in the models, which is now used by all three reactor vendors for their loss-of-coolant accident models. The uncertainty methodology is sufficiently general that it can be applied to a wide range of computer models, including the atmospheric models described above. it starts with an evaluation of the state of understanding of the the physics, and the mathematical models used to describe the physics, and then proceeds to use a strict process to introduce variations in the important parameters and conflate the results of different “runs” to estimate the overall uncertainty of the model.
Use of this method produces documentation that can be scrutinized by other engineers to determine the suitability of the model. I am highly suspect of modelers who say that their methods are too complex to document properly, especially if the consequences are so dire as they say. It is important for these models to be publicly available, on the internet, with full documentation of their underlying physics and all of the runs that have been made to produce the results.
These are extraordinary claims (the end of the world) being made by the modelers, and they demand extraordinary proof.
I am not a climate scientist, but if the climate scientists cannot produce documentation like this, then their claims cannot be believed.
9 September 2008 at 2:36
Re 4. Spencer & 3. Andrew,
the study is
Diffenbaugh, Noah S., Filippo Giorgi, and Jeremy S. Pal, 2008. Climate change hotspots in the United States. Geophys. Res. Lett., 35, L16709, doi:10.1029/2008GL035075, August 30, 2008, online http://www.purdue.edu/eas/earthsystem/Diffenbaugh_GRL_08.pdf (8,52 MB)
9 September 2008 at 2:59
About models please see
Reichler, Thomas, and Junsu Kim, 2008. How Well do Coupled Models Simulate Today’s Climate? BAMS Vol. 89, No 3, pp 303-311, March 2008, online http://ams.allenpress.com/archive/1520-0477/89/3/pdf/i1520-0477-89-3-303.pdf
“…, we note the caveat that we were only concerned with the timemean state of climate. Higher moments of climate, such as temporal variability, are probably equally as important for model performance, but we were unable to investigate these. Another critical point is the calculation of the performance index. For example, it is unclear how important climate variability is
compared to the mean climate, exactly which is the optimum selection of climate variables, and how accurate the used validation data are. Another complicating issue is that error information contained in the selected climate variables is partly redundant. Clearly, more work is required to answer the above
questions…”
About climate sensitivity please see
Rind, David, 2008. The Consequences of Not Knowing Low- and High-Latitude Climate Sensitivity. BAMS Vol. 89, No 6, pp. 855-864, June 2008, online http://ams.allenpress.com/archive/1520-0477/89/6/pdf/i1520-0477-89-6-855.pdf
“Along with the continuing uncertainty associated with global climate sensitivity [2°–4.5°+, for doubled CO2 in the latest Inter-governmental Panel on Climate Change (IPCC) report], we have not made much progress in improving our understanding of the past/future sensitivity of low- and high-latitude climates. Disagreements in paleoclimate interpretations, and diverse results from the IPCC Fourth Assessment Report future climate model simulations suggest that this uncertainty is still a factor of 2 in both latitude regimes. Cloud cover is the primary reason for model discrepancies at low latitudes, while snow/sea ice differences along with cloud cover affect the high-latitude response. While these uncertainties obviously affect our ability to predict future climate-change impacts in the tropics and polar regions directly, the uncertainty in latitudinal temperature gradient changes affects projections of future atmospheric dynamics, including changes in the tropical Hadley cell, midlatitude storms, and annual oscillation modes, with ramifications for regional climates. In addition, the uncertainty extends to the patterns of sea surface temperature changes, with, for example, no consensus concerning longitudinal gradient changes within each of the tropical oceans. We now know a good deal more about how latitudinal and longitudinal gradients affect regional climates; we just do not know how these gradients will change…”
Current state in Climatology in plain English…
[Response: As always, what is your point? And why do you think papers that demonstrate a) that climate models have been getting better over time, and b) another that says that regional latitudinal gradients are hard to predict, have to do with the importance of complexity in explaining what is going on? Throwing in random citations with cherry-picked quotes is a well worn tactic, but please, try and be a little relevant. - gavin]
9 September 2008 at 3:28
Chris, #49.
The simplest answer is that in the ancient past there were no dino drilling platforms. So 10,000 years to sequester 7% of fossil fuels has nothing to do with the past CO2 expression.
9 September 2008 at 3:33
I have a quick-and-dirty way to calculate planet surface temperatures here.
9 September 2008 at 3:33
Fred, #43. WHAT more pressing needs are there for insurance? And are you an economist to know what % of GDP would be needed? Have you worked on a model for it?
And as a % of GDP of the US, what’s spent on securing oilfields annually? Not even fazing the current administration.
9 September 2008 at 3:45
Gosta, #16
No.
The weather doesn’t have free will. It can’t change based on what it thinks will happen in the future.
People can. People do.
9 September 2008 at 3:45
Dr. Farley,
I take on Cockburn here. It’s very reminiscent of shooting fish in a barrel.
9 September 2008 at 3:49
Fred Jorgenson posts:
Fortunately, we don’t need all that — even though we do have one heck of a solid case! A carbon emissions trading scheme and a push for renewable sources of energy will work just fine.
9 September 2008 at 4:06
Guenther Hess says:
Mankind will probably survive, yes. The current civilization may not. Athens had the best philosophers in the world, but it didn’t stop Alexander the Great from annexing it. The Nazis had the most brilliant physicist in the world working for them — Werner von Heisenberg — but their atom bomb program failed miserably. You can probably think of additional examples.
9 September 2008 at 5:45
Dear Mr. Weart,
You say greenhouse gas warming is not possible to predict but one piece of information I picked up watching an Australian scientific meeting mentioned a number of studies, including a University of Chicago study, predicts CO2 has no/little warming ability passed 22ppm (parts per million).
This is based on CO2’s well known infra-red radiative profile, known and solid science for nearly a century.
Namely once CO2 reaches 22ppm in Earths atmosphere it effectively reaches its limits of warming ability as CO2 is only able to reflect/force only certain infra-red wavelengths, that’s achieved/blocked at 22ppm and past that any additional CO2 (to 380ppm or 600ppm) does no additional ‘work’.
Like sunglasses with 2 of say 4 filters, it doesn’t matter how many extra pairs you wear, 2 sections of light always get through and 2 are always blocked. Add 20 new pairs with the same light filters and it makes no difference!
Water vapour, the most powerful GHG, has a very broad infra-red wavelength forcing ability in contrast.
As I say this has been known science for nearly a century and for me is the killer punch, amongst many, that KO’s the global warming hoax. Can you confirm please?
[Response: None of this true. Where is your source and why do you think it’s trustworthy? For more details about why this is wrong, read these previous posts (here and here). - gavin]
9 September 2008 at 6:00
Johhny B, #56.
Put one blanket on the bed. You are now insulated 100% against losses from radiation and convection. So putting a second blanket on is a waste of time, isn’t it? After all, everything that a blanket stops is already stopped by the one blanket you have on now.
9 September 2008 at 6:59
So, Spenser, it’s Science but not as we know it. No assumptions which can be tested from first principles, no equations which can be checked, and nothing which can be tested against experiments or measurements. Prior experiments (Angstrom’s saturation test in 1905) are dismissed as “botched” (by Gavin) but not repeated.
[Response: I said no such thing, and these experiments have been repeated hundreds of time at much higher accuracy (look up HITRAN). - gavin]
Quantum theory is invoked as photons like little silver bullets random-walking their way into space.
At the long end of the spectrum the dominant behaviour in the wave-particle duality is the wave. Absorption is resonant. “We see, to our amazement, that the photon frequency associated with the transition is just an integer multiple of the classical resonant frequency of the harmonic oscillator” from a First Course in Atmospheric Radiation, page 252, by Grant W Petty, University of Wisconsin Madison.”
Not to my amazement, I have to say. I was regurgitating these equations 50 years ago.
Your comment “the air radiates some of this energy back to the surface, keeping it warmer than it would be without the gases” ignores the second law of thermodynamics. Heat cannot pass from a cooler to a warmer body. Check it out. Switch off the fridge and open the door.
[Response: Oh please. The 2nd law is for *net* transfers. We receive radiation from the big bang at a brightness temperature of about 3K. Your claim would preclude us detecting it because energy can’t come from a colder body (deep space) to a warm one. Nonsense. - gavin]
Robert Essenhigh is one senior engineer who rejects AGW out of hand because the dominant radiation absorber in the atmosphere must be water vapour. He calculates the height at which all the radiation which can be absorbed, will be absorbed, finds this to be relatively close to the surface, and concludes that additional CO2 can make no difference.
“The absorption coefficient is 1–2 orders of magnitude higher than the coefficient values for the CO2 bands at a concentration of 400 ppm. This would seem to eliminate CO2 and thus provide closure to that argument.”
[Response: Only because he doesn’t consider what happens higher in the atmosphere where the situation is very different. This idea that somehow radiative transfer is completely wrong is belied everytime you look at a picture derived from a satellite - if we were so wrong, none of those remotely sensed pictures of water vapour, or air pollution, or IR or aerosols or trace gases or ozone or anything would work. More nonsense. - gavin]
Chuck Wiese on Roger Prielke’s web-site reaches the same conclusion:
“The spectral overlap (C02/H2O) is quite severe, and is precisely why the results (of doubling CO2) don’t produce much change.”
[Response: Whether 4 W/m2 is “much change” is a matter of opinion. I consider half the forcing associated with the transition from a glacial to an interglacial quite a lot. - gavin]
Barton Levenson (62) published an elegant paper (The Irrelevance of Saturation) on this site which accepts the premise but refutes the conclusion. He applies conservation of energy across space, an upper and a lower atmospheric layer, and the surface. The upper layer contains negligible water vapour, and is warmed by any additional CO2. Because energy is radiated downwards, the lower layer and the surface have to compensate by increasing their radiation which, via Stefan-Bolzmann, requires warming.
If the relative absorption in the upper atmosphere increases from 0.5 to 0.6 Barton generates an increase of 3 degrees C at the surface.
But if the relative absorption is a far more realistic 0.01 (zero water vapour, very low pressure), doubling the CO2 produces a temperature increase of only 0.2 degrees at the surface, which is not really measurable.
Elsewhere on the site another of your contributors summarised neatly the (unquantified) “higher is colder” argument – “And as adding greenhouse gases to the atmosphere increases the optical thickness or depth of the medium, it raises the height from which photons escape, and assuming a roughly constant lapse rate this will imply a warmer surface”.
Possibly, but what evidence is there for a “roughly constant lapse rate”?
The UAH data from 1978 quotes:
Lower Troposphere - 1.3 degrees per century
Mid Troposphere - 0.5 degrees per century
These figures are not compatible with AGW, particularly since most of the lower troposphere warming took place between December 1999 and January 2002. Current temperatures are back to 1978 levels, themselves the trough of changes since the previous peak, in the forties.
I am, incidentally, pleased to see that the rise in temperature from the Little Ice Age to the forties is now accepted on the site, without any attribution to AGW.
[Response: “now”? please find some cites that indicates that our explanations have changed in any major respect. - gavin]
9 September 2008 at 7:36
“These people, typically senior engineers, get suspicious when experts seem to evade their question.”
Before I became a statistician I was a chemical engineer and then a biomedical engineer.
It is perhaps the evasiveness that enginees are responding to. Engineers tend to be distrustful of pat answers. But I would have thought physicists would have been more of a problem in that they prefer to see an eplanation reduced to first principles. (Admitedly much of the physics engineers encounter at university is of this variety.)
Engineers are not strangers to complexity - although this may vary by dscipline. Much of what they do cannot be reduced to first principles and is ultimately empirical. Chemical engineers (& others) often work with dimensionless equstions. Relationships of quantities that have been expeimentally derived, and since they are dimensionless, can be scaled to the situation at hand. Furthermore engineers often resort to computer models in the design of chemical reactors, structures, circuits etc. They could no more summarise one of their own complex systems in 6 pages than you can summarise AGW.
I suspect it is a culture clash that is causing the problem, not an incapacity to accept that a simple derivation is all that is required. For most engineers, Gavin’s “The CO2 problem in 6 easy steps” *should be* and excellent start.
9 September 2008 at 7:44
I see that my point (#10) is more or less repeated by other commentators (eg #53 and #58).
Gavin: you reply to #53: “Why should anyone continue to discuss with you?”
[Response: It wasn’t rhetorical, it was serious. I spend way too much time replying to comments because I do genuinely want people to understand what it is we are talking about and why. But even my time is limited, and so one needs to prioritise. People that just want to declaim, rather than learn, just create noise and responding to their pseudo-questions is of limited use (other than to point out that they rarely make any sense). I judge the worth of engaging with people by their willingness to listen and modify their position. If they don’t change at all, why bother? I’m not doing this for my health. - gavin]
If you really want to communicate, then you better find a way of communicating. If on the other hand you don’t want to communicate, there is little point in replying to comments, really.
In fact you remind me of those English-speaking tourists arriving back home in frustration, convinced that the locals they visited are brainless idiots, after having shouted, yelled, huffed and puffed to make themselves understood…by people that simply do not speak English.
If you or Mr Weart want to speak to engineers, or anybody else, then you both better speak in a way that engineers can understand. And if they don’t appear to have understood, you cannot simply jump to saying “why are you people so slow to understand?”…the only sensible option is to see where the miscommunication is (yes, it can be with you too), and to work to fix that.
I have provided a few suggestions already.
People do have various degrees of skepticism in the nature and dangers of anthropogenic global warming. How difficult is it to recognize that? If you instead poo-poo their thoughts whenever expressed, you will win nobody’s mind. Fine by me, but then what’s a blog for?
9 September 2008 at 8:12
I forgot to add the most significant scientific reason behind my skepticism at #54: the uncertainty in forcing due to aerosols and ocean heat uptake is almost as large as the forcing from human GHGs itself.
Every climate model from a simple “single-box” through to global circulation models with multiple layers and interactions has built into the denominator of climate sensitivity the actual forcing - ie the difference between CO2 forcing and the offsets due to aerosols and ocean. Since the offsets are highly uncertain, the denominator of climate sensitivity has error bars that come close to encompassing the origin (zero), which means the climate sensitivity estimates themselves have error bars that approach infinity (in the extreme).
It is this enormous range in estimated climate sensitivity that is invoked by the likes of Stern and Al Gore (and the alarmist industry in general) to claim that there is a non-negligible probability of total climate catastrophe, and that therefore we must act now to drastically curb CO2 emissions.
But the reality is much more sanguine. We simply cannot attach a probability to wildly high estimates of climate sensitivity from climate models, borne as they are out of an uncertainty in the denominator.
To see this, consider an extreme boundary case: there is a small (but non-zero) probability that the net forcing has in fact been zero over the 20th century, once you account for aerosols, ocean heat uptake, clouds, etc. In that case the sensitivity required to reproduce the 20th century temperature increase is infinite. But we know there is zero probability of infinite climate sensitivity (otherwise we wouldn’t be here discussing this), so clearly it is not valid to convert uncertainty in forcing into a probability for climate sensitivity based on the output of climate models.
[Response: You touch on a real point, but you are not correct in your assessment. The issue of the forcing uncertainty is a problem for the twentieth century, and it does mean that high sensitivities cannot be ruled out from considering that period alone (Forest et al, etc). However, this isn’t anything to do with climate models. They have very well defined sensitivities since you can impose whatever forcing you want. But even better is to take a situation where the forcing is unambiguously non-zero - the last glacial period is the best bet for this and actually rules out with high probability both the very low and very high numbers (as we have frequently discussed). - gavin]
9 September 2008 at 8:22
#34 John Mashey,
Perhaps some of the difference between these two categories is in the numbers of components the sysystems that they deal with have and how crucial individual components can be.
Category 1 deal with what could be described as serial systems. A few crucial components have many opportunities to affect the system and the effects of individual components and processes are not swamped by the effects of other components and processes. Components are often in one of a few discrete states. The system is a dictatorship with the crucial components as dictators.
Category 2 deal with what could be described as parallel systems. Many components simultaneouly affect the system. The effects of any single component can be swamped by the affects of many other components. Thus even if you have mad an error about the effects of an individual component your estimates will usully be fairly robust. Components usually are in one of many states or the states have continuous values. The outcome is decided by a vote among many components.
9 September 2008 at 8:28
mugwump at #54,
The last 3 of the 4 reasons for your skepticism do not relate to the technical accuracy of the scientific work on climate change, but rather to the political and emotional context of the debate. By “emotional” I include your own emotional response.
Surely the issue is whether AGW is a real and present danger. Are the predictions accurate. Are the changes observed in the world’s climate illusionary or ephemeral - eg the huge loss or Arctic sea ice - or are they part of a dangerous trend.
Your first point of skepticism concerns over-fitting of models. A valid question. These are not statistical models that are “fit” as such, but rather ones incorporating the physics of the underlying processes. I have no doubt that, despite claims to the contrary, some parameters are “tweaked” at the margins, as the relevant physics is not fully understood. But these are additive processes and for the most part, leaving these out (eg aerosol effects) results in similar effects - Gavin could possibly confirm this. Older simpler models show little deviation from current ones.
Skepticism becomes denial when you refuse to evaluate your position in light of the available evidence. Your last three points make me wonder if this is the case.
9 September 2008 at 8:28
Gavin at 57,
Yes I know, however I guess it confuses a few that e.g. don’t blog themselves…
9 September 2008 at 8:32
Re 68. Johnny B, about Water vapour, the most powerful GHG, two interesting papers have been published:
1. Huang, Yi, and V. Ramaswamy, 2008. Observed and simulated seasonal co-variations of outgoing longwave radiation spectrum and surface temperature. Geophys. Res. Lett., 35, L17803, doi:10.1029/2008GL034859, September 4, 2008
“We analyze the seasonal variations of Outgoing Longwave Radiation (OLR) accompanying the variations in sea surface temperature (SST) from satellite observations and model simulations, focusing on the tropical oceans where the two quantities are strikingly anti-correlated. A spectral perspective of this “super-greenhouse effect” is provided, which demonstrates the roles of water vapor line and continuum absorptions at different altitudes and the influences due to clouds…”
2. Dessler, Andrew E., P. Yang, J. Lee, J. Solbrig, Z. Zhang, and K. Minschwaner, 2008. An analysis of the dependence of clear-sky top-of-atmosphere outgoing longwave radiation on atmospheric temperature and water vapour. J. Geophys. Res. – Atm., 113, D17102, doi:10.1029/2008JD010137, September 3, 2008
“We have analyzed observations of clear-sky top-of-atmosphere outgoing longwave radiation (OLR) … We also analyze the sensitivity of OLR to changing surface temperature Ts, atmospheric temperature Ta, and atmospheric water vapor q. We find that OLR is most sensitive to unit changes in Ta when that change occurs in the lower troposphere. For q, the altitude distribution of sensitivity varies between the midlatitudes, subtropics, and the convective region… In the tropical convective region, a rapid increase in q in the midtroposphere leads to a dramatic reduction in OLR with increasing Ts, which has been termed the “super greenhouse effect”.”
9 September 2008 at 8:54
Re 72. mugwump, about aerosols please see
Rosenfeld, Daniel, Ulrike Lohmann, Graciela B. Raga, Colin D. O’Dowd, Markku Kulmala, Sandro Fuzzi, Anni Reissell, and Meinrat O. Andreae, 2008. Flood or Drought: How Do Aerosols Affect Precipitation? Science Vol. 321, No 5894, pp. 1309-1313, September 5, 2008
“Aerosols serve as cloud condensation nuclei (CCN) and thus have a substantial effect on cloud properties and the initiation of precipitation. Large concentrations of human-made aerosols have been reported to both decrease and increase rainfall as a result of their radiative and CCN activities. At one extreme, pristine tropical clouds with low CCN concentrations rain out too quickly to mature into long-lived clouds. On the other hand, heavily polluted clouds evaporate much of their water before precipitation can occur, if they can form at all given the reduced surface heating resulting from the aerosol haze layer…”
About climate sensitivity please see first
Kiehl, Jeffrey T., 2007. Twentieth century climate model response and climate sensitivity. Geophys. Res. Lett., 34, L22710, doi:10.1029/2007GL031383, November 28, 2007
“Climate forcing and climate sensitivity are two key factors in understanding Earth’s climate. There is considerable interest in decreasing our uncertainty in climate sensitivity. This study explores the role of these two factors in climate simulations of the 20th century. It is found that the total anthropogenic forcing for a wide range of climate models differs by a factor of two and that the total forcing is inversely correlated to climate sensitivity. Much of the uncertainty in total anthropogenic forcing derives from a threefold range of uncertainty in the aerosol forcing used (i.e. tuned) in the simulations…
[20] Finally, the focus of this study has been on anthropogenic forcing. There is also a range of uncertainty in natural forcing factors such as solar irradiance and volcanic aerosol amount. It would of value to reduce uncertainties in these forcing factors as well.”
And finally
Allen, Myles R., and David J. Frame. Call Off the Quest. Science Perspective Vol. 318, No 5850, pp. 582-583, October 26, 2007, online http://www.eci.ox.ac.uk/publications/downloads/frame07-sensitivity.pdf
“…An upper bound on the climate sensitivity has become the holy grail of climate research. As Roe and Baker point out, it is inherently hard to find. It promises lasting fame and happiness
to the finder, but it may not exist and turns out not to be very useful if you do find it. Time to call off the quest.”
Well, time to call off the quest, and follow the suggestion of the authors:
“In reality, of course, our descendants will revise their targets in light of the climate changes they actually observe.”
9 September 2008 at 9:26
RE Gavin @ 73:
Unfortunately, those pushing the alarmist agenda are using the higher sensitivity estimates from the models to further their political goals.
It also goes to the question of how likely we are to overfit when tuning the climate models. When you have knobs on your black box that generate such widely varying behaviour when rotated through relatively plausible ranges, reliable extrapolation becomes problematic.
Fair enough. Do you have a canonical RC reference? (There are a *lot* of posts on RC to wade through).
Without expecting an inline response here (I can read whatever you point me to), I am particularly curious about the very low numbers - what constitutes “very low” and how they are ruled out (very low for the IPCC seems to be 2C but from my reading I am starting to think that 2C is at the upper end of the plausible range).
[Response: Start here (more here). - gavin]
9 September 2008 at 9:34
#77–
Sounds like more bad news for the Lindzen/Spencer postulate that water vapor will provide negative feedback, yes?
9 September 2008 at 9:34
You do engineers a disservice we can understand the science, even if it is not our speciality. Physics is a key subject for engineers.
What engineering teaches us is to reconcile the theory with the concrete. Many “Climatologists” do not seem to check their hypotheses with the observed data. Take the IPCC chart in AR4 that gives the projection of temperature against CO2.It starts at 280ppm and by the time CO2 gets to today’s 385ppm it indicates the temperature should be in range +1.0C to +2.2C, but the actual temperature rise is only +0.6C. An Engineer would ask, “Why”!
It looks like the models are wrong. Perhaps you need some engineers to ask the “Why” question.
Can you explain why the projections do not match the actual data?
[Response: But they do! (Fig SPM.4). I don’t know what other figure you are talking about (reference), but the two factors that might be missing is the lag in the ocean response to forcing, and the net effect of all the forcings (incluiding other GHGs, aerosols and land use etc.). - gavin]
9 September 2008 at 9:52
Bruce Tabor @ #75:
I elaborated on the scientific grounds for my skepticism at #73.
Leaving out aerosols most assuredly does change the output.
The models generate widely varying estimates of climate sensitivity from 20th century observations. I always wondered why, given that as you say they are all supposedly modeling the same physical processes. It was only recently that I looked into it sufficiently to realize that you have to divide by net forcing to get sensitivity (either explicitly as in a box-model, or implicitly via the underlying physics in a GCM), and the error bars on net forcing come perilously close to the origin.
9 September 2008 at 10:13
Mugwump has been told this before, as recently as a couple of weeks ago on Deltoid, but continues to make the “overfitting claim” regardless.
The first time is forgivable ignorance. Continuing to repeat it is [edit]
9 September 2008 at 10:37
rxc above wrote wishing for climate models to be as precisely documented as nuclear power plant models.
I’m sure the climate modelers would like to do that.
Each piece put into a nuclear power plant is specified in advance and you know what it’s made of.
I’m sure the climate modelers would like comparable information.
Suppose you were trying to model a nuclear power plant you couldn’t take apart?
But you vary one factor you’ve calculated will make a change in its behavior and it’s changing?
Willy Ley: “analysis is all very well but you can’t tell how a locomotive works by melting it down and analyzing the mess” — aren’t you glad fission plant modelers aren’t often faced with trying to do that? Do you understand why climate modelers have a difficult task? Do you understand why climate modelers raise concerns when they see CO2 increasing and can predict the climate system will be going outside its known safe performance parameters?
What’s the worst thing that can happen, with either physical system?
Right. Precautions are appropriate before fiddling with the inputs. Restrain those who would meddle, eh?
9 September 2008 at 10:45
mugwump wrote: “Unfortunately, those pushing the alarmist agenda are using the higher sensitivity estimates from the models to further their political goals.”
With all due respect, that statement — “pushing the alarmist agenda to further political goals” — is a dead giveaway that you are a pseudo-skeptic, someone who concludes that the overwhelming consensus of the scientific community that anthropogenic global warming is real, and dangerous, MUST be wrong because you dislike what you imagine to be the “political” consequences of that reality.
9 September 2008 at 10:52
Re 80. Kevin, well, Lindzen and Spencer are not studying water vapor, instead please see first Lindzen’s latest
Rondanelli, Roberto, and Richard S. Lindzen, 2008. Observed variations in convective precipitation fraction and stratiform area with sea surface temperature. J. Geophys. Res. – Atmos., 113, D16119, doi:10.1029/2008JD010064, August 29, 2008
“This paper focuses on the relation between local sea surface temperature (SST) and convective precipitation fraction and stratiform rainfall area from radar observations of precipitation…
Although a dependence on temperature such as the one documented is consistent with an increase in the efficiency of convective precipitation (and therefore consistent with one of the mechanisms invoked to explain the original Iris effect observations) this is but one step in studying the possibility of a climate feedback. Further work is required to clarify the particular mechanism involved.”
[edit - limit your random quotes to at least peer reviewed papers]
9 September 2008 at 11:04
One of the difficulties for sceptics (like me) is that, even if the complexities are understood and believed, when the analysis finally gets to the end, the conclusions turn into very simple and exact assertions with absolute certainty. The final simple answers are usually accompanied with words like certainty, absolute, consensus, incontrovertible, irrefutable, undeniable, etc. As an example, Spencer’s commentary, IMO, explained the reality of the science very well (though I agree he unfairly characterizes engineers — but that’s a nit point); then the very first post said, in essence, ‘and so X gives you exactly Y’ (followed by a ‘and don’t give me any back talk’ — again a minor point) as did a number of following posts. It seems like gaining an understanding of the complexity of the science is not desirable per se but just a necessary evil only if the simple conclusions are called into question. To a sceptic this is not very convincing.
As a bit of mitigation, I have to leave the above as a “difficulty” and can’t quite raise it to a “criticism.” Because: 1) I’m not sure if there is any other realistic and practical way to do it. I mean a scientist can’t just end his/her analysis with a “nobody knows anything”. 2) While scientists in the field are guilty (IMO) of the above, it’s considerably more evident in the scientist wannabes, and who are also much less likely to backpedal when called than are the climatologists. 3) Some of my fellow sceptics (by happenstance, not choice) are guilty of the same thing. I don’t have any helpful suggestions to offer so I just have to live with my difficulty. But it is a real difficulty none-the-less.
9 September 2008 at 11:19
ps there is another factor that I think not pure science but none-the-less rational and expected. If a climatologist makes his/her best analysis with no more than a, say, 50-50 probability, but the 50% probability causes disastrous effects bordering on Armageddon, it’s understandable (maybe even necessary) that the scientist will push that scenario. Though this really has no relevance to this discussion…
[Response: No it isn’t. It’s appropriate to say what the odds are. (though frankly, I wouldn’t get on a plane with 50:50 odds of crashing). - gavin]
9 Septe