Bishop Hill Sceptics are from Mars and warmists are from Venus
Jan 5, 2015 Climate: Models Here is a fascinating article from a blog I haven't come across before called "A Chemist in Langley":
...the vast majority of the warmist community have a worldview that stresses Type I [false positive] error avoidance while most skeptics work in a community that stresses Type II [false negative] error avoidance. Skeptics look at the global climate models and note that the models have a real difficulty in making accurate predictions. To explain, global climate models are complex computer programs filled with calculations based on science's best understanding of climate processes (geochemistry, global circulation patterns etc) with best guesses used to address holes in the knowledge base. The models are "trained" by looking at historical data and seeing whether they can replicate what has occurred in the past. In a simplistic description they are trained to interpolate data and once they get good enough at interpolating data they are then used to extrapolate future conditions. Since the global climate models are works in progress they still do not do a great job at extrapolating, yet. In particular, these models have failed to predict the "pause" in the surface temperature data that has lasted for (depending with whom you talk) somewhere around 15 - 18 years. Essentially the model predictions and the measured temperatures have diverged.
Skeptics see the poor extrapolations and suggest a need to refine the models to address the divergence. From a Type II error avoidance viewpoint, given the relatively poor quality of the model predictions, putting limited resources into addressing potentially faulty predictions seems like a poor choice. Instead, resources should be allocated to improving the models and any additional monies spent on other "demonstrably real" problems out in the world. Warmists, on the other hand, point out that the models are the best tools we have to date and to ignore their predictions, just because they are imperfect, is a big mistake. Warmists point out there is a real risk that a lack of action now could result in a low-probability, high-cost outcome (a fat tail on the uncertainty distribution of the outcomes). So monies should be spent on avoidance immediately while we continue to refine the models.
(The link came to me by a circuitous route, but I believe the hat-tip is for Richard Tol)
Reader Comments (112)
To be fair the models only failed to predict your famous pause because there hasn't been one. [Snip - manners]
http://www.skepticalscience.com/climate-models-intermediate.htm
" Warmists, on the other hand, point out that the models are the best tools we have to date and to ignore their predictions, just because they are imperfect, is a big mistake...."
Would any bookie continue to employ a soothsayer who has a proven track record of failure?
@Onbyaccident : Er. The UK Met office (a.k.a. Mystic Met) says there has been and still is a pause...
http://www.metoffice.gov.uk/research/news/recent-pause-in-warming
But back on topic, the warmists seem to love the precautionary principle, which dictates than no risk is too small when it comes to a potential catastrophe. They need to appreciate and respect that not everyone buys into their world view (belief) and from the 'skeptical' side of the argument they seem to be 'chicken littles' worrying about the sky falling, when it clearly isn't and there is no evidence that it will fall either.
66 reasons and counting...
@ Farley R
...and a quote 3 paragraphs in says
"The first paper shows that a wide range of observed climate indicators continue to show changes that are consistent with a globally warming world, and our understanding of how the climate system works."
[Snip - manners]
Onya, I don't know as much about sites as you seem to, but I believe we are nowhere near Hansens scenario C prediction, never mind his BAU scenario.
It's not just the models that have failed
Onbyaccident
I have had to edit two of your comments for bad manners and have removed others entirely. I'll block you completely if there is any repetition.
Ok not sure where lapogus off to with that. Email exchange. I guess it links into your conspiracy theories but whilst posts interesting I still come back to why you would post.
Now EO - potentially more interesting. Sounds like you may have read something. You would like to expand as naturally I disagree with you? Hansen's '88 work was actually pretty good.
Here we go again with the canard of describing models as merely "imperfect" as opposed to utterly useless for policy.
Interesting.
If I recall, BH commenter Raff used to maintain that we should pay attention to the prediction of models because they are the best thing we have. They fact that they are useless did not discourage Raff from coming back repeatedly and insisting that it still made sense to act on what they said, because there was nothing better.
"The first paper shows that a wide range of observed climate indicators continue to show changes that are consistent with a globally warming world, and our understanding of how the climate system works."
Meaning what, precisely?
"Consistent with" presumably means "might show or then again might show something different".
"our understanding of how the climate works" sounds impressive but since the current temperature plateau (as accepted by mainstream climate science, Met Office, et al — numerous references available; sorry you're too lazy to find them — and only disputed by the terminally desperate, Skeptical (hohoho) Science for example) more than amply demonstrates that we don't have a clue "how the climate works" you're up a blind alley.
And perhaps you had better explain what a "climate indicator" is. It sounds suspiciously like yet another sciencey phrase designed to confuse rather than enlighten. Lot of them in climate "science", I'm afraid.
Come back when you know what you're talking about and we can have a chat maybe.
Unbelievable and you know it BH.
I am deeply curious as to why people hold the views you do. Which is why I post here occasionally. There is rarely any substance to the comments put on here. You seem to have taken offence as my calling you deniers but that is what is going on here. You are denying effectively physics 101 and myself as a modeller/mathematical physicist wonders why.
It would be entirely telling to ban me but that is your choice.
[I only ask that you observe good manners. If all you have to contribute is name-calling then you are no loss to the site. And anyway, your suggestion that I am "denying physics 101" suggests you need to read quite a lot more on this site before attempting another comment.BH]
Oba
"It would be entirely telling to ban me but that is your choice."
But presumably not significant that SkS does that routinely, and without a polite warning first..?
It's good to see another lukewarmer blog saying interesting and reasonable things.
Though I'm not sure about his comments about academia having strong procedures for avoiding type 1 errors - if that were the case we wouldn't be where we are today.
His previous posts on renewables are worth reading, as are About that climate "consensus" and My Lukewarmer post.
On twitter he is @BlairKing_ca
With people like OBA I always wonder what their purpose posting is. If it's to convince us we are wrong, the approach fails. If it's to show his hairy friends what a good little soldier he is, then none of them read here, it's a waste of time. If it's just trolling, then he needs to join the queue, we get bucket loads here, and they all think they're original.
The psychology of trolldom has always escaped me, though.
...Warmists point out there is a real risk that a lack of action now could result in a low-probability, high-cost outcome .... So monies should be spent on avoidance immediately while we continue to refine the models.....
The referenced article suggests that warmists do have a point to make, but that they are not approaching the problem of low-probability, high impact issues in the way that sceptics do.
Actually, that's not the case. Such an argument would make sense when considering, for instance, a large meteorite strike on a town. The chances are unknown, possibly very low, but the impact would be great. There can be no question of that.
In contrast, the warmists have asserted that human-generated CO2 will produce a great danger. But there is no definite evidence of that - it's just an assertion. Unlike a typical low-probability danger, where the chance is low but the impact would be large, EVERY part of the warmists' hypothesis is just that, a hypothesis (or guess).
And furthermore, as far as I can tell, every single prediction/concern they have uttered has turned out to be false. To treat 'climate change' as an example of a known, though low-probability danger is giving it far too much credibility. It is much closer to a religion.
Jan 5, 2015 at 2:44 PM | Onbyaccident
"Hansen's '88 work was actually pretty good."
You reckon? Only if one is prepared to accept every one of his scenarios is warmer (or should that be HOTTER!!!) than observed reality.
Truth be told James Hansen's A, B & C scenario projections, upon which so much has been laid are flat out wrong, and that doesn't do AGW theory many favours either.
http://stevengoddard.wordpress.com/2014/10/15/giss-still-below-hansens-zero-emissions-scenario-c/
The philosophical implications of that piece of garbage are scary
global climate models are complex computer programs filled with calculations based on science's best understanding of climate processes
1/. No, they are not, they are based on the most complicated approximations that the computer (programmer)s can handle, not on the best understanding.
Warmists, on the other hand, point out that the models are the best tools we have to date and to ignore their predictions, just because they are imperfect, is a big mistake.
2/. Ho hum. On what other basis would one ignore them?
3/. On what basis are they judged to be the best tools we have to date???
4/. The Jehovah's Witnesses knocked on my door claiming they have the 'best tools we have to date' for combatting human death. Should I convert to Christianity 'just in case'?
The quoted article says:
" they (the GCM models) still do not do a great job at extrapolating, yet. "
But, the writer fails to recognize that if the overarching hypotheiss is hopelessly flawed these models can never ever do a good job and the wholel exercise is a waste of time., energy and money.
In any field -- other than self-stuled climate science -- such uselless models (whic do no thave a single forecast to their credit) would long ago have been abandioned. You cannot fix an approach that is inherenly flawed.
Wow. What an amazing thread. Pre-historic in its thinking.
As regards Global Climate Models, how about the following conjecture:-
"The Earth's atmosphere is a chaotic system of such infinite complexity that it is utterly impossible to model its behaviour for more than a few days or weeks."
With a little bit of work I'm sure we could turn that into a hypothesis then a theorem and ultimately a law. This would take far far less time than getting the GCMs to "work".
Ah, Onbyaccident (now there is a misnomer!): obviously in denial about the plateau in temperatures. And you have the temerity to revile others as "deniers". How quaint.
Question if it was not so easy to run a series of models until you get the results they need.
Do you think climate 'scientists'; would but more or less effort into their 'guesses ' before they run these models?
Or has greater computing power lead to poorer foundation work in the fundamentals of that which this power is being applied against.
I have to think that if these people needed work it out with pen and paper than than just push in a few numbers on a keyboard , then either then be much more careful over what data they use or they think a lot harder about their 'guesses' given the cost of being wrong a regular basis ,has they are now , would be so much higher while the ability to 'tweak' their models would be a lot less.
Sceptics are from Mars and warmists are from Venus.
Climate models are based on a small planet somewhere near Alpha Centuri. Allegedly, quite accurately.
Meanwhile, back on planet Earth.....
@lapogus hits the mail on the head. We can only live in hope.
"What if climate change appears to be just mainly a multidecadal natural fluctuation? They'll kill us probably..." Tommy Wils, 2007. ClimateGate1 email 1682. Source: http://di2.nu/foia/foia2011/mail/1682.txt
I largely agree with the article except I'm not sure what warmists and sceptics can communicate to each other. Our differences are as fundamental as our DNA. I understand that warmists think and feel the way they do, I just don't agree with the way they see the same information and would chose to do different things about it even if I agreed with warmist views. We've reached a point where we're waiting for the global temperature to do something significant and nothing either side can say will alter that.
Any guesses what Onbyaccident's new year resolution was?
I have a question...
Do they have a model to predict say next year (T+1) using today's climatic "state" and they calculate 2 years ahead by feeding the year1 output into the input.
OR
do they take the inputs project them forward to year 2 and run the model function
?
Either way is prone to large errors and assumptions.
Onbyaccident (Jan 5, 2015 at 2:55 PM) said "You are denying effectively physics 101 and myself as a modeller/mathematical physicist wonders why."
Please be aware that most of us do not deny the physics of the basic green-house effect, which results from the absorption/radiative properties of CO2 gas molecules. We simply question the evidence that the Earth's climate system is amplifying this basic green-house effect, via the domination of positive feed-backs, to the point where it represents a catastrophic threat to mankind... and thereby justifies draconian changes to the world's socio-economic system in order to rapidly remove its use of fossil fuels.
As far as I can tell, even the IPCC have no empirical evidence to support the 'catastrophic anthropogenic global warming' (CAGW) hypothesis and the results from computer models serve merely as a poor substitute. Moreover, as I understand it, the concept of 'sensitivity' can be used as a measure of this amplification, where ~1C would indicate zero amplification, >3C may be of some concern (depending upon your definition) and >6C catastrophic by all definitions. Therefore, given that the latest IPCC report (WG1AR5_Chapter10, p.922) says ‘Estimates based on observational constraints indicate that it is very likely that the equilibrium climate sensitivity is larger than 1.5°C with a most likely value between 2°C and 3°C’, I think there's more than enough evidence to deny the reality of CAGW.
Just give up and stop wasting money.
It is mathematically impossible to model a closely coupled, poorly defined, multivariate, chaotic system using numerical time step integration. Errors are cumulative. Any idiot knows that.
Jan 5, 2015 at 2:28 PM | FarleyR
===================================
The trouble is, we are just as bad at implementing the precautionary principle as we are at modelling the climate. Ebola demonstrates that.
@onbyaccident..
RSS says you are wrong as well as the Met office. "On by accident". Damn right
http://wattsupwiththat.files.wordpress.com/2015/01/clip_image002.png
We were able to analyze the temperature estimates of CMIP5 models and compare them with HADCRUT4 (1850 to 2014), as well as UAH (1979 to 2014).
The models estimate global mean temperatures (GMT) backwards from 2005 to 1861 and forwards from 2006 to 2101.
The details are here: http://wattsupwiththat.com/2014/12/22/cmip5-model-temperature-results-in-excel/
Bottom Line:
In the real world, temperatures go up and down. This is also true of HADCRUT4.
In the world of climate models, temperatures only go up. Some variation in rates of warming, but always warming, nonetheless.
The best of the 42 models according to the tests I applied was Series 31. Here it is compared to HADCRUT4, showing decadal rates in degrees C periods defined by generally accepted changepoints.
Periods HADCRUT4 SERIES 31 31 MINUS HADCRUT4
1850-1878 0.035 0.036 0.001
1878-1915 -0.052 -0.011 0.041
1915-1944 0.143 0.099 -0.044
1944-1976 -0.040 0.056 0.096
1976-1998 0.194 0.098 -0.096
1998-2013 0.053 0.125 0.072
1850-2014 0.049 0.052 0.003
In contrast with Series 31, the other 41 models typically match the historical warming rate of 0.05C by accelerating warming from 1976 onward and projecting it into the future. For example, while UAH shows warming of 0.14/decade from 1979-2014, CMIP5 models estimates average 0.215/decade, ranging from 0.088 to 0.324/decade.
For the next future climate period, 2006-2035, CMIP5 models project an average warming of 0.2C/decade, ranging from 0.97 to 0.375/decade.
The longer the plateau continues, the more overheated are these projections by the models.
Oba says that "I post here occasionally". Does that mean he comes from Truro..?
Okay, suppose the medical industry has a model for assessing a particular risk -- say, testicular cancer -- and your personal physician has run the test on you, and says you "fail". IF this test is prone to Type I errors, you will "fail" even if you have no cancer. If the test is prone to Type II errors, then other people who may be at real risk will not be detected -- their test results appear to have "paused" -- although later measurements with more data may show the damage and indicate high risk.
Okay, with that scenario in mind -- Do you personally accept the test as "good enough, the best we have, nobody has a better way" a forecast of scenarios in your personal future? Do you accept the treatment of the "risk" -- not the disease, but of the RISK! -- that involves some drastic sacrifices in your current lifestyle? Or, do you accept a "wait and see" attitude hoping for either more conclusive indicators, or a better test? Or do you begin contributing generously to a new course of research attempting to develop a new test that brackets the distribution even though it may be just as imperfect, and perhaps prone to Type II errors?
Leo Smith - I think that their justification is based on the Precautionary Principle.
As the PP appears to have been adopted by the UN and the EU (amongst others), regardless of any criticisms we might offer in respect of models, etc., aren't they saying flawed or not it's the best we have and we must stick with them and the Precautionary Principle until such time as we can explain fully the science behind global warming and all its other incarnations.
Criticisms of the PP come with lack of cost benefit analysis - isn't this where Richard Tol comes in to explain that economically the precautionary costs are too high coupled the uncertainties of the science??
I may be in cloud cuckoo-land today being topped up with heavy duty painkillers for injuries sustained from global warming induced cold weather events............
Onbyaccident,
As a modeller/mathematical physicist stop wasting your time here and go and write the first GCM that produces accurate results.
Happy New Year
'as accepted by mainstream climate science, Met Office, et al — numerous references available'.
I thought we roasted that old chestnut ages ago on here, when analysing the Met Office's catalogue of excuses as to why the weather, & thus climate, were so difficult to predict when they had to report to the DECC to explain the wet spring & summer of 2012 at the end of that year! They basically hadn't a clue!
I'm not sure what kind of effect the 'Geochemistry' has to the predictions? I'm sure there is a huge amount of complexity in the models that is superfluous to the linear increasing temperatures with CO2 which appears to be the only significant output from them.
Jan 5, 2015 at 3:45 PM | Unregistered CommenterJohnOfEnfield
Yes, the current models are completely useless at the moment must probably because the theory used to program them is wrong. There is no reason though that the complexity of the atmosphere is not reproducible with a decent model though. Perhaps say, 90% of the 'global temperature' is due to decadal and longer ocean cycles and cosmic rays. If you had an understanding of these phenomena then a really simple model might have very good predictive ability. (This was just an example with a hypothesis different to what the current models are.)
pouncer at the moment the warmists are waving a rusty cleaver and saying "Let's just chop everything off, just in case".
Got to be Stoat, hasn't it? No-one else combines complete wrongness with such obnoxious charm.
As far as I can see, the evidence so far supports the sceptical position.
As time goes by, how can the warmists reconcile the increasing gap between the output of their models and reality? How can they continually attribute events such as bush fires and weather events to global warming when there has been no global warming this century?
Arguing that the science is sound does not improve its credibility when it is clear that the models are flawed. The science is clearly not sound. One could apply the precautionary principle to every whim of the science community. Think of the mess that would produce. Our climate is very complex and we do not understand very much about it.
Only By Accident, would be the best hope of a climate model getting something right.
It has happened yet.
Call me a conspiracy theorist or just a cynic
Climate Change competing for the headlines along with everything else.More reason to prove their spin wrong.
If we were supposed to believe them about weapons of mass destruction in Iraq why should we believe them about the weather.
or Seth Blatter or LIBOR or Peado Politicians clubs or disc jockeys and Royal underage sex scandals.
Joe Public:
Would any bookie continue to employ a soothsayer who has a proven track record of failure?
So long as the soothsayer was pronouncing on dichotomies, one who is reliably wrong is as valuable as one who is reliably right. Many years ago in the dawn of the commercialisation of knowledge based systems as aids to algorithmic trading, I came across a firm who cited just such an input into a forex trading model they built for a bank - who had a trader who was almost infallibly wrong about market direction.
With respect to the models the article says (emphasis mine):-
We may have to accept that climate models may never provide accurate predictions, as climate may be to some extent chaotic. Better approximations of the climate in terms equations will lead to greater complexity by orders of magnitude, and also a shortening of the period for which the models are predictive. As a result, like with economic models, the greater the deviation from historical trend, the less useful will be the model. For this reason the predictive ability of economic models is inversely related to the usefulness - they were good for when predicting no chnage in trend, but were rubbish at predicting the Credit Crunch.
Alternatively, the more unrealistic the climate model in relation to historical data, the more often they will be wrong in the short-run. The question then is how does one know that the underlying relationships are correct? This is when you get an evasive answer, or a PR bouncer stepping in reminding us of the truths that the models are correct and the experts know what they are doing.
@ Idau 5:28
Touche.
However, "... one who is reliably wrong is as valuable as one who is reliably right." only if there are just two variables - say heads or tails.
That phrase: "a fat tail on the uncertainty distribution of the outcomes" sounds pretty impressive, until you realize that it is (at least in IPCC:s case) entirely due to using an unphysical prior in their Bayesian analysis. Whether this was due to ignorance or design I don't know, and perhaps it is unimportant since Bayesian statistics is inherently subjective in any case.
There is at least one notable exception to the general principle outlined by the Langley Chemist. From one of the papers that had statistical foundations described by the Bish as a "swamp of misrepresentation, deceit and malfeasance" (I always liked that line), Amman & Wahl, Climatic Change 2007:
"MBH and WA argue for use of the Reduction of Error (RE) metric as the most appropriate validation measure of the reconstructed Northern Hemisphere temperature within the MBH framework, because of its balance of evaluating both interannual and long-term mean reconstruction performance and its ability thereby to avoid false negative (Type II) errors based on interannual-focused measures …."
Perhaps a better maxim for the Chemist would be something like the following: "the vast majority of the warmist community have a worldview that alternatively stresses Type I [false positive] error avoidance, and Type II [false negative] error avoidance, according to which result better fits the tale they are trying to tell.
It seems to me that Chemist in Langley may have the Warmist position backwards. Or perhaps I do. I recall that in medicine 30 years ago Bayesian statistics made us rethink medical tests and treatments. Treating patients with false negative results was decidedly bad, so test procedures' evolution had long been focused on better detection by minimizing these. Increasing "sensitivity" however often brought lower "selectivity", i.e. generating more false positive results.
So long as the treatments for false positives were benign and cheap, things were okay, but some treatments, for nonexistent diseases, could be harmful and expensive. Thus a new campaign was undertaken to address this matter.
It seems to me that the warmist (fear mongers) are willing to accept false-positive global warming "tests" as justification to implement "treatments" to reduce C02, even though there it's not certain there is an actual "disease" present (environmentally catastrophic warming) that warrants the "treatment".
In essence the warmists don't seem to be in the least concerned that they may have a "false positive" (rising temps 1980-1998 that may have been coincidental to, not caused by rising C02, for which complete revamping of the Western economy may not only be harmful in its own right, but ineffective in modulating the climate.
There is a better analogy than testicular cancer as it is a current real issue. The PSA test for prostate cancer produces about 70% false positives. Consequently it is no longer recommended by most doctors. A positive result leads to invasive biopsies and patient stress. Doctors do not want people to be unduly alarmed as the positive PSA results are more than likely to be wrong. Climate activists on the other hand do want people to be alarmed by the model results despite the inadequacies of the models.
It's a good site worth reading, but I think he's being overgenerous to the modelers. As others have pointed out, not only are the models works in progress that "do not do a great job at extrapolating, yet", but there are good reasons to think that they will not ever, and cannot ever, do a good job at extrapolating. Just looking at recent climate, pine cones, tossing a coin, or watching how many leaves the badgers are taking in, would be a lot cheaper.
The reasons for a 'several day limit' in weather forecasting is well understood. But is swept under the carpet and ignored in CliSci. I never even see this attempted to be justified. If there is some turning point in time where the unpredictable weather magically becomes predictable climate then we should be told.
Hiding behind the semantics of "projections, not predictions" will avail them not when pleading forgiveness with the electorate, never mind us 'pub bores'.