Philosopher Karl Popper claimed in his book ‘The Logic of Scientific Discovery’ that a hypothesis, proposition or theory is scientific only if it is falsifiable:
“Logically, no number of positive outcomes at the level of experimental testing can confirm a scientific theory, but a single counterexample is logically decisive: it shows the theory, from which the implication is derived, to be false. Popper’s account of the logical asymmetry between verification and falsifiability lies at the heart of his philosophy of science.”
It seems that whatever happens in the climate system is consistent with climate model predictions. Warmer, colder, less ice, more ice, droughts, floods, more hurricanes, less hurricanes, stronger hurricanes, weaker hurricanes, and so on.
An recent example from the media:
Cold wave in India attributed to global warming
Mumbai: The recent cold wave sweeping across Mumbai and other parts of India could be attributed to global warming, experts said on Tuesday here at an environmental conference.
Would the observed mid-troposphere warming of less than the 2 to 3 times increase over surface warming predicted by climate models represent falsification? Or would a prolonged period of global cooling do the job?
So, what event or observation, or series of events/observations over what timescale are required to falsify the climate modelled hypothesis of CO2 driven climate change or global warming?
This post was inspired by a couple of blog posts over at Prometheus:
The Consistent-With Game: On Climate Models and the Scientific Method
Climate Model Predictions and Adaptation
Serious answers to a serious question, please.
Jennifer Marohasy BSc PhD is a critical thinker with expertise in the scientific method.
This is pretty desperate Paul – an “authoritative” comment from who again??
“former Union minister for power and environment Suresh Prabhu ”
Relevance ? Never heard of him? Might as well be someone down the pub.
And how exactly would greenhouse warming stop a cold air outbreak? If you thought about it for 30 seconds it doesn’t even get at how greenhouse would work.
Indeed Cold Air Outbreaks still occur in future climate simulations of a greenhouse world. http://igloo.atmos.uiuc.edu/CAO/
Too subtle?
“Certain regions, however, show relatively small changes and others actually experience more CAOs in the future, due to mean atmospheric circulation shifts and decadal variability that counter the thermodynamic tendency from greenhouse forcing. These areas generally experience greater near-surface wind flow from the north or of more continental origin during the 21st century and/or are especially prone to atmospheric blocking events.
Simulated reductions in CAOs are smallest in western North America, the North Atlantic, and in central-southern regions of Asia.” !!!!
http://ams.confex.com/ams/87ANNUAL/techprogram/paper_117372.htm has more.
I didn’t say it was authoritive – it obviously isn’t – but it is the most recent example I have seen.
Anyway, back to the real question:
So, what event or observation, or series of events/observations over what timescale are required to falsify the climate modelled hypothesis of CO2 driven climate change or global warming?
While I’ve heard mumblings from time to time that we’ve moved on from Kuhn’s “The Structure of Scientific Revolutions” I find the basic outline trustworthy. A paradigm once accepted provides researchers the world over with a set of heuristics, classic experiments, crucial features of the paradigm, predictions derived from the theory, etc.
The researcher then devises hypotheses to verify the paradigm at its periphery. They are responsible to publish their protocols and data. Then a second, independent, tier takes up reproducing their results possibly not as ingenious but technically competent.
At the end of this, the test may be found, inconclusive, invalid, or better, either to support the paradigm or be chalked up as a failure.
A good example of a paradigm that has been forming for a quarter century or more without actually entering the practice of science is String Theory. The first prediction to be made is yet to appear. While climate science has made numerous predictions, their failures in actualization have seemingly not endangered the AGW paradigm–until 2007. Moreover, the practitioners have been reluctant to release protocols, data even detailed, rigourous heuristics. While not still-born, the ‘science’ appears to be failing.
A warm event doesn’t prove the hypothesis either, but we still have to listen to Goreing nonsense about the 2003 European heatwave, for example.
If the current climate is within the limits of natural variability for the Holocene, then the enhanced greenhouse effect is more likely to be insignificant. Hang on, here comes a hockey stick!
So, what event or observation, or series of events/observations over what timescale are required to falsify the climate modelled hypothesis of CO2 driven climate change or global warming?
Following a late post on the ‘Observations on January’s Temperatures’ thread, “a natural heat vent at the equator which regulates global temps” and ignoring the NASA ref first up I found this gem.
“A major climate catastrophe ushered in the Dark Ages around AD440. Snow fell during the summer in Southern Europe and the trees stopped growing. John of Ephesus wrote the sun became dark and the darkness lasted for 19 months. In AD800 the Black Sea froze. Savage storms swept the world. Following the climate driven famines, plague devastated Europe. The bubonic plague killed 25 million people. Barbarian tribes from the East attacked Western Europe.
Cold = bad”
http://palosverdesblog.blogspot.com/2006_11_01_archive.html
Note; the Orbiting Vaporizer !
Serious: We can go back to Cloudy days ahead for GCM’s and find the “heat vent” supposedly missing from models then.
“This heat vent works as follows: As the surface of the ocean heats up, the formation of cirrus clouds – i.e. high altitude clouds of ice crystals – diminishes. This is because, according to Richard Lindzen of MIT, “with warmer sea surface temperatures beneath the cloud, the coalescence process that produces precipitation becomes more efficient [and] more cloud droplets form raindrops and fewer are left in the cloud to form ice crystals”
and
“Indeed, the researchers believe this new finding alone could reduce by 2/3rds the projected increase of global temperatures – which all in all might be disappointing news to those of us counting on warmer days ahead!”
http://www.cfact.org/site/view_article.asp?idCategory=4&idarticle=270
Re vertical heat transfer: While the skeptics focus on clouds as Venetian blinds falsification is going nowhere
Recommended watching – Thunderheads ABC
http://www.abc.net.au/tv/guide/netw/200802/programs/NH0501V001D18022008T213500.htm
“The ICE team is particularly interested in the ‘ice clouds’ left behind by thunderstorms. Thunderheads push so high that the water coming out of the top is snap-frozen into tiny ice crystals. The size of these crystals determines whether the ice-clouds, also known as cirrus clouds, reflect the sun’s heat or trap it.
And what of vast thunderstorm systems such as the Indian Monsoon? May it be changing? Is the weather system that dominates the lives of 1 billion people from India to China – growing more ferocious, and wetter? Or is the Monsoon weakening, with potentially catastrophic consequences in terms of many people’s reliance on water?
Employing the latest in space technology – such as the recently launched CloudSat and Calipso satellites – the TWP-ICE experiment is the beginning of answers to these monumental questions.
Thunderstorms are the key. The ICE experiment hopes to develop a far better idea of how thunderstorms function, how they affect other clouds and ultimately how they help regulate the temperature of the Earth’s atmosphere. ”
If anyone does want to watch it and has missed it I may be able to help.
Sure modelling this vertical heat vent will be difficult enough if clouds are the only major variable of concern. An atmospheric model without oceans and the other regulator carbon involved is quite frankly meaningless. Ender’s example, tropical monsoons are not all about temperature differences between continents and oceans.
http://en.wikipedia.org/wiki/Monsoon#Processes
This link to India and its National Institute of Oceanography gives us an insight to other background studies.
http://drs.nio.org/education/orientation/Ocean/session_I.jsp
and specifically on ecosystem modeling
drs.nio.org/drs/handle/2264/860
Let’s suggest the final heat vent model will look more like a figure eight with CO2 at the aperture.
Paul, I think you question is ill-posed.
Models are not a ‘hypothesis’, they are a ‘guess’.
The models are not the theory or hypothesis.
I don’t know, but I would expect the answer to your question is:
No, they’re not falsifiable in the sense you mean (by comparison to observations), but so what?
However, I suppose you could falsify them by proving that parts of the programming is wrong, or by demonstrating that the assumptions in a model is wrong.
Still making little sense Gavin, I see.
Models aren’t hypotheses Mr T? They are just a “guess”. Then why should we take any notice of what they predict? I know, they don’t make predictions. I am amazed at the convoluted thinking that goes on.
Surely the point of a model is to incorporate the hypothesis and associated theory and make predictions. Works for gravity and all sorts of engineering modeling.
Bad predictions = hypothesis no good or associated theory wrong or inappropriate. Of course it could be the programming is wrong. I have severe doubts that any complex software developed over 20+ years by at least dozens of different people most of whom aren’t actually professional programmers actually does what anyone thinks it does particularly when there appears to be no good way to test the outputs. Isn’t that what Gavin Schmidt said over at RC?
“Thunderheads” wasn’t too bad I thought. They are probably still using a gadget I built in the Dimona.
Eyrie, Paul is mixing up his thoughts. He is demanding that models be proved (or at least falsified), but that’s just not feasible.
Models are a useful tool, that’s all.
Do you take notice of economic modelling?
I think you also forget that every time we think about what is going to happen in the future, we use a model (albeit one in our heads). Do you ignore your own future plans?
Do you think the models have predicted badly? They seem to have been pretty close up until about 2002-2003. Maybe future models will be improved?
Models work for gravity etc. because they’re pretty simple (are they linear?) I don’t think you can compare them.
I don’t pay too much attention to Europe, so maybe one of you guys can tell me, does it snow in Greece alot?
They’re talking about trolley lines being shut down, airports closed, people told to stay home out of the weather. More than 100 villages on the island of Crete snowed-in.
Is this unusual? It seems to me that Crete, being in the middle of the Med, makes it somewhat like us California’s waking up to discover San Diego snowed-in. San Diego doesn’t get much snow.
Except for last week I mean. lol
You ever figure heard of HTML, Jen?
Here are the links that didn’t go through on the last try.
Heavy snow brings Greece to standstill –
http://www.ana.gr/anaweb/user/showplain?maindoc=6157497&maindocimg=6154941&service=6
Surprise Snowstorm in San Diego –
http://www.10news.com/news/15306349/detail.html
James and others.
Why are you discussing the weather here? I though this post was about the falsifiability of climate models.
Mr T – James would be well advised to watch this excellent presentation. (Needs permission to install Webex player plugin).
http://ams.confex.com/ams/87ANNUAL/wrfredirect.cgi?id=6473
I’ll watch that tonight Luke. Looks interesting. I think James would say “but, but, but it’s cold! It can’t be global warming”
Well Mr T it goes to the heart of understanding anything on climate. And why half of these dudes aren’t even on the page. Anything other than a straight line causes an outbreak of cognitive dissonance fever.
How to give AGW the big test. Well give it a decade – if there has been no major decrease in solar radiation and no major volcanism – if global temperatures have not increased that’s that.
Funny that the Southern Ocean seems to have warmed at depth though. All that Tasman Sea juice racing past Tassie.
As one of the respondents to Professor Pielke’s question at RealClimate put it:
“It is very hard to envision a development that would significantly alter our understanding of greenhouse forcing unless our whole understanding of climate is radically wrong, and that seems unlikely”
Quite.
The surface temperature records are a ‘dog’s breakfast’ – a standard of ‘evidence’ that would not be accepted in any other branch of science, I venture to say. Radiosonde and satellite are the only records of any value, which go back to the late 50s, hardly long enough to establish a strong CO2/temperature correlation, even if it existed.
The only laboratory capable of testing and therefore falsifying the AGW hypothesis is the atmospheric temperature itself and the climate is far too complex and chaotic for anything like a value for the anthropogenically-caused radiative forcing of CO2 to be accurately determined.
The hypothesis that anthropogenic CO2 was the main driver of long term global warming in the 20th century (which preceded any firm evidence), is not falsifiable, as is the case for any other non-solar (or should I say non-terrestrial based) hypothesis – they admit it themselves.
….(or should I say terrestrial based)…
Chrisgo,
I think you have missed the point somewhat. You can’t falsify AGW, but you CAN falsify the components of it. AGW is based on assumptions all of which are falsifiable. Your mission (should you choose to accept it) is to falsify them
Here are two assumptions I can think of off hand, but there are more.
1) CO2 absorbs IR radiation (or whatever is the proper physical description of this)
2) Humans are the cause of the recent CO2 excursion
Falsify these assumptions and it falls.
Every now and then there is a big cold snap in Greece – in Italy it freezes the olive trees and they die.
Mostly around Athens you get some slush – this is the real deal.
What they are suffering from in Europe is shortness of memory, they were so used to warmer winters and associated propaganda put out by Brussels along with farm payments for social and environmental activities (commonly known as “subsidies”)
“Surely the point of a model is to incorporate the hypothesis and associated theory and make predictions”
Instruction, illustration, institutionalize etc but I don’t expect Eyrie to follow
“Often when an engineer analyses a system or is supposed to control a system, he uses a mathematical model. In analysis, the engineer can build a descriptive model of the system as a hypothesis of how the system could work, or try to estimate how an unforseeable event could affect the system. Similarly, in control of a system the engineer can try out different control approaches in simulations.
A mathematical model usually describes a system by means of variables. The values of the variables can be practically anything; real or integer numbers, boolean values, strings etc. The variables represent some properties of the system, for example, measured system outputs often in the from of signals, timing data, counters, event occurence (yes/no), etc. The actual model is the set of functions that describe the relations between the different variables” – Wiki.
From house plans to areoplanes we also make models of many things to help our own thinking to develop around some concept or other. In engineering I worked with many concepts and made drawings before we had computers. Processes and measurements could be made simple enough with a bit of immagination even in 2D diagrams. Boolean helps with switch logic and so on.
Chemical reactions require all the above and some.
Electro magnetic radiation shifts us smartly into argand diagrams and 3D immagery we can’t touch however I still visualised complex communications in 2D or 3D for analysis and solution.
I can say with all modeling our imagination is the only limiting factor.
Luke; chrisgo can’t see let alone use the “tools”
Model evaluation – wiki
“An important part of the modelling process is the evaluation of an acquired model. How do we know if a mathematical model describes the system well? This is not an easy question to answer. Usually the engineer has a set of measurements from the system which are used in creating the model. Then, if the model is built well, the model will adequately show the relations between system variables for the measurements at hand. The question then becomes: How do we know that the measurement data is a representative set of possible values? Does the model describe well the properties of the system between the measurement data (interpolation)? Does the model describe well events outside the measurement data (extrapolation)? A common approach is to split the measured data into two parts; training data and verification data. The training data is used to train the model, that is, to estimate the model parameters. The verification data is used to evaluate model performance. Assuming that the training data and verification data are not the same, we can assume that if the model describes the verfication data well, then the model describes the real system well”.
Not specificly mentioned above is “operator” training. In the real world we learn the “process” step by step. With out some imagination it takes quite a while.
In practice most records are a “dog’s breakfast” Estimating amplitude on the first sign of a swing is an art form if you are tuned in.
AGW climate models, without exeception, assume a climate sensitivity of 2 x CO2 ==> 1.5 to 5 Kelvin in mean earth surface temperature.
This has been empirically contradicted by measurement.
Hence Paul’s question is answered in the affirmative.
Popper’s philosphical definition doesn’t meet its own criteria for a theory. Ergo it’s not scientific. It was easier to tell the gals from the guys in the good old days but its trickier these days.
Pinxi
Popper’s philosophical definition was never a theory? Do you actually understand what you post here?
Good heavens, we have girl clowns to join the boy clowns led by Luke
Rather obviously models make predictions that are falsifiable.
Prediction:
http://www.realclimate.org/index.php/archives/2004/12/why-does-the-stratosphere-cool-when-the-troposphere-warms/
Falsification:
http://www.agu.org/pubs/crossref/2007/2005JD006881.shtml
Prediction:
http://www.realclimate.org/index.php/archives/2007/01/el-nino-global-warming-and-anomalous-winter-warmth/
Falsification:
http://wattsupwiththat.wordpress.com/2008/02/19/january-2008-4-sources-say-globally-cooler-in-the-past-12-months/
These errors are intrinsic to the model and follow from the non-physical myth of back-radiation. The manifestly slipshod data gathered by GISS in the last link’s comparison might be the source of some faulty model training, certainly, but model is a faithful attempt to implement the paradigm. The paradigm therefore needs an adjustment.
Louis I present you with a choice of 2:
1. explain clearly and logically in your own words why and how it is a falsifiable theory (ie satisfies its own standards) & propose a test for falsification (for bonus marks, relate to complexity science) or
2 tell us again how spirits habitate outer space in big evolving rocks and oil appears from subterranean mystery pots but it’s not possible for humans to affect the atmosphere. Keep us all entertained with your crack theories of miracles.
Or your preferred option:
3. do a runner now it’s got too hot
On topic:
In the real world empirical observation will trump or support a prediction based on a theory.
Each January the Met Office, in conjunction with the University of East Anglia, issues a forecast of the global surface temperature for the coming year. This year, the press release contained the following PREDICTION / FORECAST:
“Prof. Chris Folland from the Met Office Hadley Centre said: “Phenomena such as El Niño and La Niña have a significant influence on global surface temperature and the current strong La Niña will act to limit temperatures in 2008. However, mean temperature is still expected to be significantly warmer than in 2000, when a similar strength La Niña pegged temperatures to 0.24 °C above the 1961-90 average.”
http://www.metoffice.gov.uk/corporate/pressoffice/2008/pr20080103.html
This prediction / forecast is testable, and can falsified by empirical observation thus:
Barring extraneous influence (i.e. volcano) or a “super” La Nina, given that the Ninas are similar, then:
▪ If the temperatures in 2000 are higher than those in 2008 the theory is falsified.
▪ If the temperatures in 2000 are just below those in 2008 – the theory will be somewhat supported.
▪ If the 2008 temperatures are significantly higher than those in 2000, the theory will be strongly supported.
I will be following the development closely. I have a plot that will display the above and included the thresholds that need be met vis-à-vis the above (2000 average temperatures and 2000 average temperatures + 0.12C (an amount that will in my opinion be “significant” in that it will recreate a decadal trend of 0.16C – the bottom of the IPCC projections). I have also included the MEI plots in those years to see if there is a significant difference in the La Ninas.
http://i31.tinypic.com/13yrv6.jpg
11 months to go – the temperatures in 2008 are off to an unfavourable start for the theory.
Cheers
Arnost
Briggs Debunks Beck
http://wmbriggs.com/blog/2008/02/20/an-excuse-i-hadnt-thought-of/
An excuse I hadn’t thought of
February 20th, 2008
A few weeks ago I speculated what would happen if human-caused significant global warming (AGW) turned out to be false. There might be a number of people who will refuse to give up on the idea, even though it is false, because their desire that AGW be true would be overwhelming.
I guessed that these people would slip into pseudoscience, and so would need to generate excuses why we have not yet seen the effects of AGW. One possibility was human-created dust (aerosols) blocking incoming solar radiation. Another was “bad data”: AGW is true, the earth really is warmer, but the data somehow are corrupted. And so on.
I failed to anticipate the most preposterous excuse of all. I came across it while browsing the excellent site Climate Debate Daily, which today linked to Colby Beck’s article “How to Talk to a Global Warming Sceptic“. Beck gives a list of arguments typically offered by “skeptics” and then attempts to refute them. Some of these refutations are good, and worth reading.
His attempt at rebutting the skeptical criticism “The Modelers Won’t Tell Us How Confident the Models Are” furnishes us with our pseudoscientific excuse. The skeptical objection is
There is no indication of how much confidence we should have in the models. How are we supposed to know if it is a serious prediction or just a wild guess?
and Beck’s retort is
There is indeed a lot of uncertainty in what the future will be, but this is not all because of an imperfect understanding of how the climate works. A large part of it is simply not knowing how the human race will react to this danger and/or how the world economy will develope. Since these factors control what emissions of CO2 will accumulate in the atmosphere, which in turn influences the temperature, there is really no way for a climate model to predict what the future will be.
This is as lovely a non sequitur as you’re ever likely to find. I can’t help but wonder if he blushed when he wrote it; I know I did when I read it. This excuse is absolutely bullet proof. I am in awe of it. There is no possible observation that can negate it. Whatever happens is a win for its believer. If the temperature goes up, the believer can say, “Our theories predicted this.” If the temperature goes down, the believer can say, “There was no way to know the future.”
What the believer in this statement is asking us to do, if it is not already apparent, is this: he wants you to believe that his prognostications are true because AGW is true, but he also wants you to believe that he should not be held accountable for his predictions should they fail because AGW is true. Thus, AGW is just true.
Beck knows he is on thin ice, because he quickly tries to get his readers to forget about climate forecasts and focus on “climate sensitivity”, which is some measure showing how the atmosphere reacts to CO2. Of course, whatever this number is estimated to be means absolutely nothing about, has no bearing on, is meaningless to, is completely different than, is irrelevant to the context of, the performance of actual forecasts.
It is also absurd to claim that we cannot know “how the human race will react” to climate change while (tacitly or openly) simultaneously calling for legislation whose purpose is to knowingly direct human reactions.
So, if AGW does turn out to be false, those who still wish to believe in it will have to work very hard to come up with an excuse better than Beck’s (whose work “has been endorsed by top climate scientists”). I am willing to bet that it cannot be done.
I have worked with computers for 30 years. Currently I work as a developer/analyst to a large government department. There about 40 people in my section. The business section asks for implementation of functionality as requested by government. The business analysts write up the specifications and the developers then implement them. This software is then tested to see if it complies to the specs.
The fundamental thing needed for any large computer application is definite testable rules. You have this with business systems and engineering but not with the soft sciences. I think the question is wrong anything is falsifiable but GCMs are fundamentally false. I have looked for information about how they are produced. My opinion is they are developed by an isolated few Boffins. They are not developed by software engineering team in any sense. They need very powerful computers and use Fortran. Performance is a huge issue so what you do is switch off bug checking once it seems to work okay after that errors can be very subtle. I would have much more consideration if they were developed in the same way as the Space Shuttle software. The initial work took 400 people 5 years. Their work is open published for all to see and obviously tested. We are getting the modelers opinions though a computer, why any one believes them is the real question. They are a joke.
What about the IPCC’s FUTURE PROJECTED temperature models?
The ones AGW’s often get excited about that is far into the future.Those 50-100 year trends must be awfully hard to validate.
Can they be falsifiable?
LOL
DHMO you shared your opinion and your rules of thumbs that you think should be met, but your claim that GCMs are fundamentally false doesn’t follow from those points. Try to directly substantiate your core premise
DHMO assumes a decent model is always complex and sets of rules are inflexible.
I once modelled radio spectrum users at prime sites round our major cities with a gate like drawings that resembled sets of meshing false teeth then got them all to change channels and squeeze up. However hardly any body involved could acquire the mandatory sharper equipment at the time. A long time cobber integrating metro rail control for several new but independent operators had half a budget to work with. Meantime somebody had lost the key to the old system.
Let’s go back to basics for a mo. The Tee Square was a very useful tool for drawing perpendicular lines but in functionality v pure thinking in design it had certain limitations. I passed up Fortran 4 etc when I realised The Fox 2 and the PDP 11 behind the ASR 33 were still only a box of switches.
Expectations lead us up the garden path not models.
Paul’s comment this morning suggesting that uncertainty in future human actions makes climate science unfalsifiable is way off the mark.
The IPCC has described about 40 different pathways for possible future emissions and the climate consequences have been calculated.
The “falsifiability” test is to see which of these the world follows and compare the calculated temperatures to what actually happens.
This, of course means waiting a while.
At present, what we can do is look at how the model projections from the 1980s (and a few from the 1970s) have shaped up.
The other thing to note is that climate models get checked against a lot more things than just the global average temperature.
DHMO – you need to get out more. Model builders are continually involved with validation of all manner of processes – err – um – the actual point? Ever talked to any climate modellers – of course not. Back to COBOL for you.
Paul – it fundamentally does matter how humanity reacts to CO2 – there’s probably some difference between doing something or nothing in what CO2 concentrations will be? Pretty darn basic. A duh.
Better pray the warming doesn’t pick up guys – all this will be simply history. Funny that that Southern Ocean is warming. Hmmm….
Ian – “The other thing to note is that climate models get checked against a lot more things than just the global average temperature.”
Agreed – if anyone bothered to watch the program that I posted you would have found out that one of the biggest reasons the 7 aircraft and hundreds of people were assembled for 7 weeks to study thunderstorms is to help model clouds better. Scientists fully understand the limitations of models and are working in the field to improve them. This is what scientists do. To suggest that they falsify data is complete misunderstanding of the role of computer models in science.
OT,
Just wondering. Anyone think the Arctic ice extent has peaked a bit early? Or is this just a wobble and it’ll pick back up in early March?
http://arctic.atmos.uiuc.edu/cryosphere/
Look at the ice-extent graph.
Pinxi my core premise is that without definitive testable rules the GCM will not tell you anything. They are false in the sense that the output is not verifiable. It could be positive or negative so a waste of time.
Gavin I am not saying you model something with a computer but I am saying you need definite testable rules. You obviously had that for your model. So your statement about my assumptions deliberately obfuscates the point. To understand think about producing a model of all children. It should tell you exactly what the child will be doing in 5 years.
Luke you are a waste of space I have met more intelligent boulders.
DHMO, of course a model should be able to be quantified in terms of it’s accuracy. However it is a mistake to think that a model will ‘predict’ results. No model of anything will predict exactly what anything will be doing. If that was possible we could actually predict the future and make a fortune on the stock market. However it will give you a sense of what is more likely to happen.
It is also a mistake to think that demonstrating a model is wrong disproves something. For exmaple physicists model the likely solar cycle. If you prove the model wrong it doesn’t mean there is no such thing as solar maximum and minimum. It doesn’t mean there is no solar cycle.
Proving a model wrong (and they all are, don’t forget that) just shows there are flaws in the model (which everyone already knows). However if the flaws are too big you chuck the model out (which they do).
Mr T,
” For exmaple physicists model the likely solar cycle. If you prove the model wrong it doesn’t mean there is no such thing as solar maximum and minimum.”
It does mean the model is a worthless predictor of future TSI.
“It doesn’t mean there is no solar cycle.”
Do you make habit of missing the point like this?
Gavin the blog gremlin stole a word, I meant..
Gavin I am not saying you can’t model something with a computer but I am saying you need definite testable rules. You obviously had that for your model. So your statement about my assumptions deliberately obfuscates the point. To understand think about producing a model of all children. It should tell you exactly what the child will be doing in 5 years.
Thanks Jan for making the obvious point about Mr T’s comment. If you can not define definite verifiable rules and can not test the result how does such a model have any value? You might as well use a wheel of fortune with various outcomes. Certainly it would be cheaper.
Jan, no it doesn’t necessarily make it worthless. As it may only be slightly wrong. There is still worth in trying and failing.
Also remember that the person who makes the model expects it to be ‘wrong’, you can’t ever model the real world precisely.
The last comment was more in reference to modelling vs AGW.
Maybe you should read the thread before jumping to conclusions.
DHMO, that’s a silly comment. Of course they have testable rules. They also test their models for stability etc. Go and look at ClimatePrediction.net they articulate what parameters they check and what rules they use to discard models.
Modelling is an essential tool, used in everything from engineering to economics. You just need to understand why people model – it’s not to “predict” an outcome, but rather see what sort of outcomes are expected.
And no model would ever do what you suggest: “producing a model of all children. It should tell you exactly what the child will be doing in 5 years.”
Such a model could never exist. Unpredictability is part of our universe.
On obfuscation; I doubt any one here has spent more time offering a bottom up view or reverse engineering white elephants. The children thing tells me a lot too. I was listening to an ABC debate on prison design. Let’s have a “model” prison to sort a few social problems. Outcomes probably depend on what we perceived was wrong with odd bods in the first place.
Too much time with Cobol and C ++ people hey.
Back to basics: Sure I tested every T square by flipping it over on the straight edge of the drawing board to check the difference in perpendiculars but I know that won’t help build my house. Out on the block we run diagonal tape checks on every rectangle and true corners with a plumb line suspended from a decent beam framed up roughly parallel with the horizon.
Tools don’t ever replace creative thinking.
“Harry Collins’ central argument about experimental practice revolves around the thesis that facts can only be generated by good instruments but good instruments can only be recognized as such if they produce facts. This is what Collins calls the experimenters’ regress. For Collins, scientific controversies cannot be closed by the ‘facts’ themselves because there are no formal criteria independent of the outcome of the experiment that scientists can apply to decide whether an experimental apparatus works properly or not.
No one seems to have noticed that the debate is in fact a rehearsal of the ancient philosophical debate about skepticism. The present article suggests that the way out of radical skepticism offered by the so-called mitigated skeptics is a solution to the problem of consensus formation in science”
http://www.sciencedirect.com
“Collins wondered if one can acquire a full domain language of a specialized scientific community without having passed all the technical and mathematical entry examinations”
http://www.dpu.dk/site.aspx?p=8649&newsid1=4678
DHMO – I’m aghast – you’re talking through your hat. Validation is the main activity that GCM modellers do. What a wonderfully theoretical blog it is.
Jan, no it doesn’t necessarily make it worthless.
Useful for propaganda purposes only if not thrown out and for “that didn’t what else can we try” if it is. Now don’t you go grumbling that others are jumping to conclusions when you do your own context dropping. I said it made it a worthless predictor of future cycles. It might have some worth if and only if the errors are known.
The real problem with your statement is the idiotic:
“It doesn’t mean there is no solar cycle.”
as if anyone is suggesting that faulty climate models means there is no climate.
Jan, guess you don’t like my analogy.
Anyway, the point I was making is that showing a model is faulty is not the same as showing AGW theory is faulty. Which is the point of this post.
“when you do your own context dropping.” I don’t know what you mean by this.
The problem I had with your post is that this statement
“It does mean the model is a worthless predictor of future TSI.”
taken on it’s own is pretty pointless. It’s a statement of the obvious and doesn’t add to this discussion. So I had assumed you meant something more significant, I guess I was wrong and all you can do is state the obvious.
Luke, give up this Mr T pretense willya? – you are only confusing yourself (you should keep note of diff identities style and characteristics – you are getting them muddled up – you would be hopeless on stage)
Hey, who is this dude?
No idea, does he speak Greek?
No but he knows how to make a grecian urn
Well Rog, you’ll never know… you see, WE ARE LEGION!
You are powerless to stop the AGW ALARMISTS! GIVE IT UP NOW
WE OWN ALL YOUR THEORY!
LOL,
I see that even a scientist does not want to answer my question.
Could it be because it would make him realize that it in itself invalidate a cornerstone of the paradigm that the IPCC obviously stands on?
“Anyway, the point I was making is that showing a model is faulty is not the same as showing AGW theory is faulty.”
That’s better but it doesn’t show it’s right either and depending on unverifiable/unfalsifiable models to verify AGW doesn’t help matters.
“I guess I was wrong and all you can do is state the obvious.”
What is less obvious is the distraction from the fact that if you can’t verify or falsify the models you also can’t use them to verify or falsify the AGW theories on which they are supposedly based as so many try to do.
Now Mr T cant remember his own name, which shouldnt be too hard eh Luke.
“T” not as in Tango as in Time out dude.
Remember, the sun always rises in the east and always sets in the west.
Jan, the models don’t verify AGW. AGW can be falsified by falsifying the assumptions it is based on.
It seems we are probably in agreement?
I don’t recall the models ever being used to verify AGW. Would be interested in seeing someone try. Can you remember any of the scientists or modellers (or models) that try to verify AGW?
Oh, I think I get it now.
It seems computer climate model results are fed back into the computer models from which the assumptions were derived and so on, in order to approximate the empirical data, maybe a bit like this http://en.wikipedia.org/wiki/Image:DrawingHands.jpg
Climate models mean AGW which means climate models.
When the empirical data refuses to conform to the model, it’s the data which is called into question first, then the model is adjusted based on the same fundamental assumptions – epicycles on epicycles on epicycles etc.
Climate sensitivity, a fundamental assumption, is poorly understood.
Since their initial trend projection in 1990, the IPCC have had to adjust downward in order, I assume, to more closely approximate the observed data. You don’t need enormous computer power to do that.
There is little public confidence in the computer climate models. People say they are concerned about ‘Global Warming’, but it does not affect their investment or spending decisions.
For instance, private property in the ski resorts or with direct river, estuary, bay or ocean access, remain at a premium.
Handy hints here about T squares and house frames though – but “creative thinking” (IMHO) is the last thing climate science needs right now.
Rog!!!! You cannot disclose my secret, we will vanish in a puff of CO2, this creating an even worse calamity of Greenhouse proportions!
On a more serious note,
why do you think Luke and me are the same?
I thought he called me troll the other day (think now he actually meant Jan now).
Have you modelled the luke/Mr T similarities?
Is the “Mr T and Luke are the same person” a valid hypothesis? Is it falsifiable?
Mr T (not Luke, he’s a fraud)
Chrisgo, you are kind of right (I believe).
Although you are still thinking AGW depends on models – it doesn’t it is an hypothesis in it’s own right.
Yes, climate sensitivity is difficult. Can you think of a better way to determine it?
(think now he actually meant Jan now).
Took you a while to work that out you might be Luke after all.;)
“I don’t recall the models ever being used to verify AGW”
They are though mainly in the popular press. It’s not in this case a scientific issue but a political one. Nevertheless as early as 1979 Ramanathan in the conclusion of his peer reviewed “seminal” (though I would say largely heuristic) paper on a one dimensional EBM claimed validity due to agreement with the GCMs.
Little has changed and then Gerry North last year said the EBMs were sanity checks for the big boys (GCMers). So we have a fully circular case that falsifies or verifies nothing.
Jan,
I would say that you can’t validate AGW using models.
I don’t know what an EBM is, nor do I know who Gerry North is.
EBM = energy balance model
Gerry North:
http://www.gulfbase.org/person/view.php?uid=gnorth
“I would say that you can’t validate AGW using models.”
Good
So next time tries to suggest we can you’ll slap em down .
Interesting comment
“Prior to the Industrial Revolution, the primary climate forcings were changes in volcanic activity and in the output of the Sun, but the strength of these forcings is not very well known. In contrast, the increasing concentrations of greenhouse gases in the atmosphere over the past century are consistent with both the magnitude and the geographic pattern of warming seen by thermometers” – Gerald R. North, Ph.D.
before the
Subcommittee on Oversight and Investigations
Committee on Energy and Commerce
U.S. House of Representatives
July 19, 2006
Guess he believes in AGW and climate change
“Dr. Gerald North says advances in satellite and computer technology over the past 30 years have allowed scientists to accurately identify global warming and predict it’s future”
Fox news 2008
Guess he believes in AGW and climate change
He does and is a major player in the game that the noisy Mr t did not know.
while this is an interesting comment
“In contrast, the increasing concentrations of greenhouse gases in the atmosphere over the past century are consistent with both the magnitude and the geographic pattern of warming seen by thermometers”
Did he actually supply some supporting numbers?
Mr T
I looked at ClimatePrediction.net it was very instructive. Quote “scientists are trying to reduce the complex behaviour of the climate down to a set of mathematical equations, in the hope that they can then begin to understand the processes that are going on” to me it says a lot. They also say the weather obeys the laws of physics but that it is chaotic. As for testing I found “The equations are tweaked, within reasonable boundaries, so that the model does as well as possible at producing past and current climates”. I looked at their team about twenty people. It does not look like a software engineering team about to take on the task on the largest engineering project ever.
I am glad you agree with me about the fact it is not possible to model a child. You also said “Unpredictability is part of our universe”.
About child modelling could we not reduce the complex behaviour of a child down to a set of mathematical equations in the hope that we can begin to understand the processes that are going on? Children’s behaviour is quite chaotic but definitely obeys the laws of physics. For instance very few or possibly none have been seen to levitate. The equations would need to be tweaked, within reasonable boundaries, so that the model does as well as possible at producing past and present behaviour. Having read this paragraph please compare it to the first one.
I think there is about a snowflakes chance in hell to produce a child model that is worth anything and even less chance for a GCM. The ideas on ClimatePrediction.net can be used to model anything. If it is not possible to model a child then also it is not possible to model the weather so as to produce a GCM. Unpredictability is part of our universe.
BTW I went looking for on their site for rule and after that test I found nothing. I found nothing because the search button searches for itself. That right 18800 hits on the site for the word “search” hope the GCM is better programmed.
So Mr T it was not a silly comment. Unfortunately you and most of those who get their science from page 3 of the Daily Telegraph have no idea about how computer software is developed and what computers can and can’t do. Computers are vastly less capable than most of the chattering classes think.
Luke You said of me
I’m aghast – you’re talking through your hat. Validation is the main activity that GCM modellers do. What a wonderfully theoretical blog it is.
I am sorry about the comment that I had met more intelligent boulders. I am sorry I will apologize to the boulder.
Jan it seems faculty members of atmospheric sciences at Texas A&M University are a chatty lot on climate change in general but I doubt their local industry funds enough honest research to make waves.
Jerry North himself is chairman of American NAS Panel (IIRC) that makes him a serious player even if he isn’t personally involved in serious research.
Jan Pompe
Is this the same North that spoke at the US senate inquiry into the Wegman report on MBH 98? I thought his name was Gerry North.
DHMO – “So Mr T it was not a silly comment. Unfortunately you and most of those who get their science from page 3 of the Daily Telegraph have no idea about how computer software is developed and what computers can and can’t do. Computers are vastly less capable than most of the chattering classes think.”
I have a pretty good idea of how computers work and how software is developed and I am wondering if you really have any idea of modern software.
Computers will do what you ask them to do. It would seem that you have the romantic notion that climate modellers program in a whole heap of parameters and then expect the oracle of the model to tell them about the future. I am not sure whether you actually believe this or are trying to make this bit of misinformation the truth.
GCMs are like any software model. They are something you can play around with and do experiments on to find out where the right answer might lie. Scientists understand this and most of them use them correctly. The popular press is of the opinion, that you are trying to reinforce, that scientists think that GCMs can predict the future and of course they cannot do this therefore they are somehow deficient.
Again you are unwillingly (or wittingly – I don’t know) contributing to the doubt campaign. This is where sowing doubt substitutes for scientific investigation.
The truth is that software used to make design predictions has to be used in the correct manner. If you are designing a feed hopper you could build 20 prototypes and maybe find the best design or you can model the flow of ore in a model, test thousands of different types and build 1 or 2 prototypes from the results of the optimisation that you did in the model. No engineer would claim that his/her software model was perfect and no company would replace an engineer with a software model.
Similarly climate researchers use models to extend their own knowledge however the real brains is the experience and knowledge of the modeller that is using the tool. They also spawn experiments to check and gather data where the inputs to the models is unknown or deficient.
So before you sling off at other people about their computer knowledge how about you stop contributing to the FUD campaign of the deniers and start posting some real science.
DMHO
He signs his name to posts and email “Jerry” s as many whose name is ‘Gerald’ do. If I used my first name “Gerrit” which is the Dutch protesant version of ‘Gerrard” I would probably do the same since the name is unpronounceable by most native English speakers.
Ender
” Computers will do what you ask them to do.”
Perecisely – Garbage in garbage out. They are too dumb to make an independent assessment of the quality of software and data.
Jan I was just trying to determine if it was the same person. Do you know the senate inquiry I referred to? I have listened to the whole four hours and would recommend if you have not. The Wegman report which was discussed is also very interesting to read.
DHMO
“Jan I was just trying to determine if it was the same person.”
It is. I wasn’t sure if you were challenging the ID or not. I’ve heard of it but haven’t listened to it read some transcripts. I have read the Wegman report and a clarifying letter.
Wegmans report is an important one to read for oneself as it is often misquoted to put a different spin on it.
Ender are you deliberately missing the point or don’t you understand what computers can and can not do? Mr T claimed that my proposition there was not definitive testable rules was a silly statement and directed me to ClimatePrediction.net. So I went the and endeavoured to find such. All I could find was a lot of PR and what I quoted. I paraphrased what they say they are doing there and applied it to a model of a child quite easily. You conveniently ignore this. I really do have a very good idea what modern software can and can’t do and have seen a litany of cons that depend the lack of knowledge exhibited by most of the populace.You read my original post why ignore the what I said my experience is and try to denigrate me? But I do know I and suggest you stop because it shows you are not at all concerned with my argument that building a GCM is waste of time because they are not definitive or testable.
I will say it again science has deduction, induction and discovery test. Engineering models as you say are very worthwhile it saves a lot of expense. You can model a jet engine or a nuclear power station. Each of these obey the laws of physics in an orderly way they are not chaotic. Still the jet engine is physically tested to destruction. Why would that be, because humans are fallible and make mistakes (as Hansen and Mann discovered) large computer applications are fallible and have bugs always. You can not have computer model that predicts a 90% chance the gulf stream will stop if the CO2 rises to 550ppm and then test it. There are many who purport themselves to be in job of running GCM who have joined the media. They produce a press release about what their GCM has told them and that we are all doomed.
Doubt and skepticism is the corner stone of science without it science becomes a religion so do not talk to me about “real science”. Do you expect me to produce an experiment to disprove the quasi paradigm of climate models. The damage that is being done by the fear, uncertainty and doubt of those who say we are doomed because what the GCMs predict, do far more damage to science.
Lastly I slung of mildly at Mr T because he slung of at me an I will at you also if you continue to do the same.
Jan
Yes I have read it not a reference to it the lack of independent review and stats people was very interesting. As a person who knows a lot about computing I must say your idea that they are dumb is wrong. They are way below that, a computer is programmed to do everything by humans. Without that they do not even make a good doorstop. So dumb is sort of a compliment. I have been having a hell of a time over the last month making a computer do what I want it do. In the process I thought of quite a few words to describe it but dumb was not one of them.
DHMO,
” As a person who knows a lot about computing I must say your idea that they are dumb is wrong.”
I’m being nice I like my mac.
“In the process I thought of quite a few words to describe it but dumb was not one of them.”
You weren’t perchance reaching the conclusion the beast had a mind of it’s own?
I’ve had my interesting adventures starting with a you beaut 64K machine that took up the entire bottom floor of a wing at Wollongong University. Eat you heart out Brainiac.
I’ve had a fair bit of experience with micro-controllers and assembly language could I help?
Thanks for the offer but no. What I am working on is but a part of a larger system. I should not complain, I am paid well (6 figures) for the frustration. At the heart of it I enjoy the challenge of solving problems and producing a testable result. The frustration is one of process and communication between people. Defining the problem is a real issue and then getting adequate data to unit test alterations. I can spend a day just getting data into the test machine and then a few minutes to fix a bug. We currently are in the final days of a major release and the pressure is really on. The Business area defined the problem 18 months ago, the Business Analysts have written up the specifications, I have analysed them and changed the necessary parts of a quite large application. The testers are still testing and it goes live 5th of March. If we get it wrong you will see it in the Australian press it would embarrass a federal minister. Potentially a million Australians could be told they owe money to the Government.
On the overall scale of things though what I am on working is relatively simple. All up I guess there is about 50 to 60 people implementing this function. The developer cost is about three million per year. So the overall salary cost is at least 15 million.
The reason I am explaining this is that since you appear to have had or still have university inside knowledge do you have any knowledge of development done in this way in universities or other bodies who do climate research. To model the climate of a planet is not a small task. The Space Shuttle software took 2000 years of effort so we would be looking for something in the order of at least that perhaps a 100 times bigger. The defenders of GCMs on this blog are talking about toys to help scientists to think. That is fair enough but what goes along with that is that we are guilty and must change our way of life totally.
Sorry it was not my intention to write such an extensive reply. I am glad you like your mac most in my profession have a somewhat cynical ambivalent attitude to their computers. All of mine (I have about 8 and use number at work) are too slow and increasingly have stupid annoyances. The ones I own I call FRED and add a number. The server on my home network is called ALFRED. You probably know what these acronyms mean so I won’t explain and get snipped.
DMHO
I don’t have much inside knowledge about how software research is done in universities. I’ve worked as configuration manager in a software house and for Lucent (the latter involved hardware as well of course) so I do know something about software development costs. I’m not all that sure that the software will be bigger and the real cost is in the high power computers needed to do the job.
I have done some linear and non linear programming, basically algebra that we need computers to solve in less than two lifetimes. It might only take a few hours to write the program for a non-linear case that runs for days on a PC to produce a result. Something like Navier-Stokes equations would take a very long time on ordinary computers.
Having said that you can now hear my grip. The climate modellers get the funding and the expensive toys. While cancer research has to make do with volunteers making their (linux) computers available on the internet to do some the multiprocessing in the background. Having completed my degree while working night shift in a hospice I think the money could be better spent.
DHMO – “But I do know I and suggest you stop because it shows you are not at all concerned with my argument that building a GCM is waste of time because they are not definitive or testable.”
How would you know this from a cursory look at Climateprediction.net. Have you actually looked at the peer reviewed literature about climate modelling and how it is tested? Until then you are speaking from a position of ignorance. Why not email Gavin Schmitt, who is a climate modeller, and ask him for a list of reading – I am only too sure that he would be more than willing to help.
“Each of these obey the laws of physics in an orderly way they are not chaotic.”
Are you serious here? The flow over an aircraft’s wing is one of the most apparent demonstrations of chaotic behaviour as the boundary layer is tripped in the transition from laminar flow to turbulent. Planes crash because of this transition happening at the wrong time.
“The Space Shuttle software took 2000 years of effort so we would be looking for something in the order of at least that perhaps a 100 times bigger.”
No the Space Shuttle used IBM AP100s and the reason that it took so long is that the Shuttle kept changing weights every other day. Also this software has eight lives depending on it. CGMs have no such constraints. You are comparing apples and oranges.
You may find this resource as fascinating as I did when I read it (warning geek level extreme)
http://www.hq.nasa.gov/office/pao/History/computers/Part1.html
“Doubt and skepticism is the corner stone of science without it science becomes a religion so do not talk to me about “real science”. Do you expect me to produce an experiment to disprove the quasi paradigm of climate models. The damage that is being done by the fear, uncertainty and doubt of those who say we are doomed because what the GCMs predict, do far more damage to science.”
And that is alive and well in the scientific community if you would care to look. GCMs predict nothing. They give a range of scenerios that might occur if so and so happens. I do not know why you think that the entire AGW problem is invented by GCMs. The concern of AGW started long before GCMs from solid atmospheric science that the GCMs now help to quantify along with the thousands of other scientist working in the field.
The idea that the only evidence for climate change from AGW is from computers running AGM is false and part and parcel of the FUD campaign started by vested interests.
Ender
“Are you serious here? The flow over an aircraft’s wing is one of the most apparent demonstrations of chaotic behaviour as the boundary layer is tripped in the transition from laminar flow to turbulent. Planes crash because of this transition happening at the wrong time.”
So extract an exception and focus on that, So invent the idea I was referring to aircraft design and extract an exception to something I did not say. My statement was “You can model a jet engine or a nuclear power station. Each of these obeys the laws of physics in an orderly way they are not chaotic.” Your rebuttable “The flow over an aircraft’s wing is one of the most apparent demonstrations of chaotic behaviour as the boundary layer is tripped in the transition from laminar flow to turbulent” is very unsound. So you expect them to build a computer model of this and apply it to aircraft manufacture without further testing. Would you travel on that plane?
Your statements about the Space Shuttle do not contribute anything and neither does your link to the site you found fascinating. The hardware that was used for various does not contribute anything to this.
I think you know full well why GCMs are so important in this whole question.
The views you express show you see yourself on the side of good, fighting off the evil doubters who don’t think as you do. Further this is because they are being swayed by vested interests. So Ender you are an Evangelist and we are actually engaged in a religious exchange.
Finally
“There are many methods for predicting the future. For example, you can read horoscopes, tea leaves, tarot cards, or crystal balls. Collectively, these methods are known as ‘nutty methods.’ Or you can put well-researched facts into sophisticated computer model, more commonly referred to as ‘a complete waste of time.'” Scott Adams.
No doubt you will find who wrote this and do the usual ad hominem attack but it is fairly accurate explanation of GCMs that is what you should address. As well the phrase ‘a complete waste of time’ describes you as well. I learnt to not argue with the faithful there is no point to it.
Jan
This will be my last post on this thread. I have finally accepted I was involved in a pointless religuous discussion. I just wanted to give you a tip. You need to link any research you want to do to global warming that way you will get your “Deep Thought” and plenty of funds. As for cancer research here is the link http://news.bbc.co.uk/1/hi/health/1161895.stm
In fact anything but anything can be linked so get with it and join the band wagon. Wonder if requires prayer as well?
A small point of order Ender
“Also this software has eight lives depending on it. CGMs have no such constraints”
IMHO Spaceship Earth has quite a bit hanging onto it
DHMO – “My statement was “You can model a jet engine or a nuclear power station. Each of these obeys the laws of physics in an orderly way they are not chaotic.” Your rebuttable “The flow over an aircraft’s wing is one of the most apparent demonstrations of chaotic behaviour as the boundary layer is tripped in the transition from laminar flow to turbulent” is very unsound.”
No it is completely to the point. Yes you can model a jet engine however they like any other modelling are chaotic making all computer models a simplification. You implied in this statement that modelling a jet engine or nuclear power station is completely predictable whereas it is not. There are elements of chaos in all of this. As you do not seem to get it a compressor or turbine blade is a wing that will stall unpredictably given the right conditions and the engine will flame out. Part of the modelling that is done is to ensure that when the engine is operated normally it stays in the linear region.
“Your statements about the Space Shuttle do not contribute anything and neither does your link to the site you found fascinating. The hardware that was used for various does not contribute anything to this.”
No but it is incredible fascinating and confirms to me that you are not really interested in learning. Anyone that was fascinated by technology and also wanted to learn could not help themselves to read this.
“So Ender you are an Evangelist and we are actually engaged in a religious exchange.”
Yeah right. I am as sick of it as you so go ahead and think what you want why should I care.
“”There are many methods for predicting the future. For example, you can read horoscopes, tea leaves, tarot cards, or crystal balls. Collectively, these methods are known as ‘nutty methods.’ Or you can put well-researched facts into sophisticated computer model, more commonly referred to as ‘a complete waste of time.'” Scott Adams.”
Yes and that would be true if GCMs were used to predict the future which they are not.
Gavin – “IMHO Spaceship Earth has quite a bit hanging onto it”
I am not sure if you read the link however in the case of shuttle software it was in direct control of the spacecraft. Really the pilots tell the software what they want to do – there is no direct connection between the pilots input and the final control output. One bug and they all die.
Fortunately the GCMs do not control the climate in the same way. Wise people will always treat the output of GCMs with a grain of salt and apply their own experience and knowledge to intepreting what the models say. Exactly as is done with aircraft design.
Ender: To test your statement” Yes and that would be true if GCMs were used to predict the future which they are not.” I went straight to this wiki page and noticed it was last updated on the 15th Feb. IMO anyone following this end of the thread should go right through this wiki lot. Note the discussion on nested models and spectral v grid etc.
http://en.wikipedia.org/wiki/Global_climate_model
It’s been my view all the while if GCM’ can’t be used in some way to project / predict future outcomes; why bother building them in the first place?
BTW I knew guys who were employed here in tracking the moon shots and other near space exploration. Most were highly skilled in eng cals on the slide rule, many were ex Woomera, a few went to the US for retraining at mission control. Although I guess the only common ground we had then was around pressure vessel monitoring I was most aware they retained team spirit across the globe long after those spectacular events.
http://en.wikipedia.org/wiki/Project_Apollo
Industry went down a slightly different path and reading short term deviation on a dopey control recorder became the sole chore for a tiny handful of isolated commissioning agents.
Looking back, let’s ask where would we be without the likes of Texas Instruments? The microchip had just come into calculations at the end of the 1960’s. I reckon we had very little help with any process modeling before the expensive HP-35 pocket calculator cheaper TI-30 in the mid 70’s.
Gavin – “t’s been my view all the while if GCM’ can’t be used in some way to project / predict future outcomes; why bother building them in the first place?”
What would you have scientists do? I don’t think that we can create a convenient planet to test things on so we have to do something. Computers offer a way to experiment. They can explore what the future might hold exactly as designers of aircraft use numerical modelling to rough out and refine designs without building lots of hideously expensive prototypes.
No-one in aircraft design would think that a numerical flow model can completely predict the performance of an aircraft but the best of them get withing a few percent in the normal performance envelope. They are not very good outside this however even then you can make sure that normal control inputs do not cause gross excursions in to the non-linear part of the envelope. When this happens the aircrafts behavior is in the hands of the gods. It is a tribute to the designer’s skills that when it happens most planes cope better than can be expected.
In the same way GCMs are tools that can be used to at least see where the future might lie. They clearly show that CO2 warming will cause some degree of climate change. This is not ironclad as there are many things that we do not know. However as they get better and computers get faster they do a better and better job.
Ender: Seems you missed my point, IMO we can use GCM’s to “predict” whatever.
I’m also trying to point out to those non GCM believers that we built aircraft and flew into space largely without depending on modern computer “models”. Although the slide rule was common practice in most engineering some (and I) used a log book for all technical maths. Regardless, that put the emphasis back on obtaining empirical data for the critical stuff.
Faith grows from direct measurements including fatigue testing of wing structures, accelerated aging of moving parts, impacts, skins and coating performance etc. Obtaining good data from turbulent external events however took a while longer.
Sometimes we have to use guinea pigs not instruments where life is concerned. Our sensitivity to a changing environment in the end comes down to the numbers of winners or losers along the development path. Science generally records both groups.
gavin – “I’m also trying to point out to those non GCM believers that we built aircraft and flew into space largely without depending on modern computer “models”. Although the slide rule was common practice in most engineering some (and I) used a log book for all technical maths. Regardless, that put the emphasis back on obtaining empirical data for the critical stuff.”
And I was not going to disagree with you. What computers allow us to do is do all the calculations that go into an aircraft design so quickly that we can now do them over a thousand times with slightly different parameters to see if the performance would be better. Slide rule people could take years to do the calculations where a computer has done it before you can take your finger off the Enter key.
I don’t know if you have read Slide Rule by Neville Shute however in it there is a really good description of the months taken for one stress calculation in the design of the R100. (Neville Shute Norway was an engineer who drifted into aircraft design)
For the climate we could spend a thousand years doing the flow calculations that would model the climate if we did not have computers. I guess someone would do it anyway as it yields interesting lines of research.