Friday, 12 September 2008

An even more complex answer

Spencer Weart has commented on my previous post. Since I rather went off topic there, here is a second attempt, hopefully including some new ideas.

Spencer, you wrote: "Apologies to engineers! I only meant to report that people who emailed me and self-identified as engineers (often "experienced" or "retired") were a surprisingly large number of the people who demanded to know why they couldn't find a page on the Web with a simple calculation of greenhouse warming. I don't think they are typical of engineers in general and regret the implied insult. Engineers are great!"

Don't apologise. It gave me an excuse to slag off the mathematicians and the physicists who write the climate models :-)

I wonder if "experienced" is just a euphemism for "retired". I don't like admitting that I am retired, and have even thought of calling myself an "emeritus" student :-) On the other hand, "experienced" could mean that he has learnt from his mistakes. Scientists don't seem to make them, since science is always right!

You continued:
"As you know, I agree with you that computer models are problematic. But I think you deeply undervalue the very important successes they have had in actual prediction. Given that they predict -- not a certainty to be sure -- but a large risk of bad stuff, and that many things outside of climate models point in the same direction, it's shortsighted to dismiss the modelers as fools."

Actually I did not know that you thought that the models were problematical, but even so you are not facing up to the fact that they do not reproduce abrupt climate change. We know that it happens, so the models do not reflect reality. They can reproduce the climate over last 10,000 years, but not the 100 before that when the glacial Younger Dryas switched into the Holocene interglacial. Should I give the models credit for the 10,000 that they can get right, or fail them for the 100 they get wrong in the past and the 100 they are about to get wrong in the future?

I do not consider the modellers to be fools. Most have PhD’s in maths or science. I struggled to get an ordinary degree in engineering. But it is their success which has been their undoing. They are not used to making mistakes, so when they do then they cannot believe it. In engineering, the criteria is whether a design works. In science the criteria is whether the theory explains the outcome. In other words, engineers work using heuristics, rules of thumb, or simple models. That is why they are surprised there is not a simple model for the greenhouse effect.

Scientists build on the success of their predessors. As Newton said “If I have seen further it is by standing on the shoulders of giants.” When their model does not work, then they just build on a bit more. No wonder there is not a simple model now!

But why did their simple model not ework? The reason is that a mistake was made by a giant of astrophysics Robert Emden. In 1913 he applied the equations for internal solar radiation derived by his brother in law, another giant of astrophysics Karl Schwarzschild, to the earth’s atmosphere. The radiation scheme is based on Schwarzschild's equation. However, we have a planetary atmosphere not a stellar one!

The climate modellers have have been dazzled by the prestige of the astrophyicists such as Eddington, Einstein, Milne and Chandrasekar. But these giants were considering high temperature electronic radiation from atomic plasma at phenomenal pressures, not vibrationsl radiation from molecules at room temperatures and pressures or lower. No-one now considers the fundamentals of atmospheric radiation. They consider that those problems were solved in the dim and distant past.

Spencer, it is not, as you suggests, that I think the models are problematical. It is that they have a fundamental flaw in the way they handle greenhouse gases. Otherwise they are fine. I was informed personally by a professor that if I could see the way the model outputs reproduce the global climate then I would find it difficult to distinguish that form satellite imagery. True? Only an hour later a speaker informed us that the models produce a double ITCZ. It would have been too embarrassing and counter productive to have confronted the professor with the fact that satellites do not show two bands of clouds circling the equator.

If the current model was correct, then it would not produce a double ITCZ, nor would the predictions of warming in the upper troposphere be found to be false by the radiosonde measurement. Philipona would not be reporting that it is boundary layer water vapour that is melting the Alpine glaciers, and there would not have been the gross under estimation of the speed of the melting of the Arctic sea ice. The problem with handling clouds would have been solved, and abrupt climate change explained.

Dr Iain Stewart signed off from episode 1 of his TV program “Earth: the climate wars” saying that in the next episode he would describe how the sceptics won. They have won. They have prevented any action being taken by claiming the models are wrong. The scientists have lost by arguing against them, instead of fixing their models! If they had, then they would have found that the greenhouse effect is more effective than currently believed, and they would have been able to predict the next abrupt climate change.

You concluded:
"Clearly I'm not going to change your mind on this, however."

Dale Carnegie wrote in "How to win friends and influence people" that you can never win an argument. Even if your opponent concedes he will still believe he is correct. We cannot bear to lose face. So no doubt you mind, like mine, has not been changed either, but hopefully I have given you some food for thought :-)

Cheers, Alastair.

4 comments:

Penguindreams said...

Alastair: One thing you get wrong is to think that engineers, or at least engineering mindset, aren't to be found in climate modelling. Since my undergraduate degree was engineering, and I was active in the EE/CS student group, I had more than passing introduction to the people and the thought processes. They're present in the climate world too.

One of the quite common comments -- from modellers -- is to quote "All models are wrong, some models are useful." Or, a comment I heard "We have to beat on the model to see how bad it is." Again, from a climate modeller.

The question is never whether the model is perfect. It isn't. We all know that before we start with one. The question is whether the model is useful -- does it answer the questions we have.

Scientists are quite aware that they make mistakes. Again, that isn't the question. One of the most memorable Seligman Crystal (top award in glaciology) was given, I'm told, by Hans Weertman. His speech was to list off the mistakes he'd made over the years.

The thing is, the mistakes they make are seldom the ones they're accused of in climate. Every few seconds, someone says that scientists don't know about the urban heat island, or that Mauna Loa is a volcano, or that the models don't include water vapor (or clouds, or ...), or the models fix the lapse rate, or ... any of a large number of errors that the scientists, in fact, did not make.

On a flip side, it is strange to see you comment about engineers not using complex models. That doesn't square at all with my experience with engineering and engineers. The chemical engineers certainly had quite involved systems, as much or more so than atmospheric chemists have been using and putting in to climate models. Mechanical and Aerospace engineers certainly have fluid dynamic models about as complex as those used in general circulation models. And so on.

On the other hand, there are still people who try to do climate with simple models. Richard Lindzen is one. He'd built his career on being able to find simple models for apparently complex problems. That said, he also serves as a warning. In spite of his well-developed talents at this, he has not been able to construct a simple model of earth climate that is successful. Some problems, whether it's that industrial scale chemical reactor or the climate system, really are fundamentally complex and just don't have a good simple model.

Even if all that climate people did was to slavishly follow astrophysicists, which they don't, they still wouldn't have the problem you think. Astrophysics involves much more than high pressure/temperature/density. Hence Lyman Spitzer's book, which I'll recommend to you, Physical Processes in the Interstellar Medium. The astrophysical case ranges from far above to far below terrestrial pressure/temperature/density conditions, including fully ionized to fully neutral, and everything in between on all counts.

Alastair said...

Bob,

That is as if I had written "Men are taller than women" and you had replied "But my wife is taller than me." It is a fact that men are taller than women but as in Earth Science there are always exceptions.

Mathematics is different. Euclid's theorems first published over 2000 years ago are still true today. Meanwhile sea level in the Mediterranean has risen AND fallen. See the frontpiece to Lyell's Principles.

The problem that I have when explaining my ideas to professional scientists is that they immediately look for the exception to the rule I am advocating, find one and dismiss my ideas. It is as if I state that men are taller than women and they soon find an exception and say I am wrong. But as you know men are taller than women. That's why Ms Palin wears high heels!

Thus I am fundamentally correct when I say that the models are based on a radiation balance at the top of the atmosphere between incoming and outgoing radiation. The mistake in the modellers are making is to assume that the balance is maintained by adjustments to the outgoing longwave radiation (OLR). In fact it is controlled by changes to the net incoming shortwave radiation (ISR) viz. by alterations of the albedo. This is not a smooth process and hence abrupt climate change.

Just as there are tall women, there are also changes OLR when the surface temperature changes. But the simple engineering model leaves those out. The complex engineering model includes them along with bandwidth broadening, water vapour feedback, cloud blackbody radiation, convection, ozone, etc.

I'll start putting some of the de Saussure stuff on the web today. Once that is done I can explain my ideas in a bit more depth.

Cheers, Alastair.

Tim Curtin said...

Hi Alastair

You may be interested to iknow that I was not able to post text below at RC, as Gavin has now deemed anything I say is unacceptable, no doubt because he got caught out at my last accepted posting #77 on his Greenspan & Einstein thread.

TC to real Climate 3 Nov 08, 1355pm

I note that Gavin has yet to reconcile the difference between his view of the Airborne Fraction of CO2 and that of Hansen & Sato PNAS 2004 or Hansen et al 2008. [see my #77 at RC]

More generally, I would again like to call into question the high regard enjoyed by the Friedlingtein at 28 al paper of whom 7 were contributing authors of ch.7 of WG1 of AR4 2007, in Journal of Climate vol.19 15 July 2006, 3337-3353. Its Figs. 2e and f show massive reductions in land and ocean uptakes with changes in surface temperature, based only on its models with NO reference to any observed evidence to that effect. This paper was endorsed - nay adopted - by Stern, by IPCC loc.cit of course (surprise!), and by the Garnaut report here in Australia, ch 4, p.103. The paper presents no real world evidence at all, offering instead only the ensembled "findings" of the C4MIP models (recommended to me by Gavin) that rising Temperature will reverse the growth of natural uptakes. Naturally F et 28 al. ignore any paper offering statistical analysis of real world field experiments showing that rising temperature reinforces CO2 fertilization over a wide range of the former. For example, although citing Nowak et al 2004 in New Phytologist 162 253-280, they omit any reference to Norby and Luo in the SAME issue of that journal (281-293), who however very tiresomely (for F et al) reported the results of their controlled experiments using red and sugar maple trees at Oak Ridge TN, which showed that without elevated CO2, higher temperature would reduce growth of dry mass, but with elevated CO2, growth was not much less than with ambient temp. and elevated CO2 (Fig.5). One can see why F et 28 al dislike Norby & Luo, because the latter say "models to forecast future [climate] changes need data support to be useful, and data-model fusion has become essential in global change research", precisely what F et 28 al are determined to avoid, with the support of Stern, IPCC,and Garnaut. The Garnaut report at Fig2.7 accepts F et 28 al as evidence for its curves showing declining future natural uptakes for which it has not a shred of evidence in the historical record. Nor does it offer any plausible a priori argument that improvements in varieties of all major food crops will cease, resulting in reduced growth of natural uptakes.

Regards

Tim

Anonymous said...

To get started winning at an online casino the first thing that you need to do is find one that suits your needs. After all, if you do not have an internet casino that you can wager with, you will never be able to reach your goals in the end. In order to find a casino all you have to do is start your search online. This will allow you to find several options to choose from, and also research each one. In turn, you will never have any problems finding the one that will best match your needs.
[url=http://www.casinolesoleil.com]casino[/url] http://www.casinolesoleil.com casino If you want to have the fun of a casino without leaving home, online casinos can be just as exciting and as hopefully profitable as brick and mortar casinos. Whether your new to internet gambling, just need a refresher, or are looking to maximize your profits, our short guide can help you get started!
[url=http://www.redlotuscasino.com]casino[/url] http://www.redlotuscasino.com casino Remember, before you visit any sites like these, check your local laws. Make sure that gambling and online gambling are allowed in your area. Even in different areas that allow online gambling, the rules may differ, so be sure to check the laws and rules thoroughly. Also, read all the terms, conditions, and rules of an online casino before signing up. Thanks for reading this guide! [url=http://www.triumphcasino.com]casino[/url] http://www.triumphcasino.com casino http://www.affrewards.com [url=http://www.affrewards.com]casino affiliate program[/url] Bingo is a beautiful game of luck, which involves patience, rapid coordination between hearing and searching out the numbers and lots of fun. [url=http://www.bingokisses.com]bingo[/url] [url=http://www.bingoeuphoria.com]bingo[/url] bingo bingo http://www.bingokisses.com http://www.bingoeuphoria.com What you have to do is, coordinate the numbers that are called out by the host with what you have on our sheet, aiming at filling up a pattern reading bingo. The pattern would be announced at the beginning of the game, so you have to be careful to know what pattern you are looking out for. The cards have the b-i-n-g-o letters on rows and numbers running as a column. A host would call out a letter-number combination, which you would have to mark on your sheet. For example if I-34 is called, you will have to look at the I column down to search for number 34. In case you have it, you place a market on it, and so on.
http://www.casinolesoleil.com http://www.redlotuscasino.com http://www.triumphcasino.com http://www.bingokisses.com http://www.bingoeuphoria.com http://www.pulsebet.com http://www.affrewards.com http://www.thehighrollerclub.com
http://www.buyphentermine1.com [url=http://www.buyphentermine1.com]buy phentermine[/url] http://www.norxbuy.com buy phentermine