Discussion > Hypothesis testing in climatology
SandyS, Martin A
Why am I aghast?
Let me tell you about my friend Albert ( changed to spare his blushes). Albert and I fly models together and have helped each other out of various holes over the years.
Albert is a schizophrenic, controlled by medication. Because of his illness he has difficulty distinguishing reality from fantasy. For example, one evening while discussing spaceflight he described Star Trek as fiction and Babylon 5 as documentary. I dug out some astronomy magazines and showed him the state of the art. I think he found Soyuz and the Shuttle rather a disappointment.
Talking to Albert I often feel a disconnect between his perception and reality. I am not suggesting that you are insane, but I am starting to feel that same disconnect between your perception and reality.
Entropic man
Entropic, you don't know the meaning of 'hypothesis'. You are better off not using it.
Which edition of ABH are you using? Can't seem to find it in the book editions available online.
shub
Shub
From Dictionary.com
hypothesis
[hahy-poth-uh-sis, hi-] Spell Syllables
Examples Word Origin
noun, plural hypotheses [hahy-poth-uh-seez, hi-] (Show IPA)
1.
a proposition, or set of propositions, set forth as an explanation for the occurrence of some specified group of phenomena, either asserted merely as a provisional conjecture to guide investigation (working hypothesis) or accepted as highly probable in the light of established facts.
noun, plural theories.
1.
a coherent group of tested general propositions, commonly regarded as correct, that can be used as principles of explanation and prediction for a class of phenomena:
Einstein's theory of relativity.
Synonyms: principle, law, doctrine.
2.
a proposed explanation whose status is still conjectural and subject to experimentation, in contrast to well-established propositions that are regarded as reporting matters of actual fact.
Synonyms: idea, notion hypothesis, postulate.
Note that usage differs between science and lay usage. Laymen often use "theory" as in defininition 2, where a scientist would use "hypothesis". I follow scientific usage, defining "theory" as in definition 1 and distinct from " hypothesis".
Entropic man
Your understanding of such terms is as sterile and superficial as the dictionary meaning for 'denier' - another of your favourite epithets - you linked earlier.
shub
Shub
This is the same problem I'm having with SandyS and MartinA. Our world views are so different that it is becoming difficult to hold a meaningful conversation.
Entropic man
Testing of hypotheses is something that has concerned philosophers and scientists for centuries. So why not learn from their wisdom? Fifty years ago physicist Richard Feynman gave a televised lecture on the scientific method. He started off by giving a very blunt definition of falsificationism. He then went on to state something highly relevant to climatology.
You cannot prove a vague theory wrong. If the guess that you make is poorly expressed and the method you have for computing the consequences is a little vague then ….. you see that the theory is good as it can’t be proved wrong. If the process of computing the consequences is indefinite, then with a little skill any experimental result can be made to look like an expected consequence.
Kevin Marshall
"It is from a Brief History of Time. The exact wording is:-
"Progress in science consists in replacing a theory which is wrong with one that is more subtly wrong."
So Stephen Hawking heard it somewhere and used it in 1988 without (I assume) crediting David W. Hawkins who was credited in 2006 with having used the aphorism around 1981. Or perhaps Stephen came up with it independently. Or perhaps both of them heard it from somebody else.
Entropic Man - Let me point out, if you had not already realised it, that it's an unpleasant weasel mannerism to liken people to (for example) someone who is essentially mad and then to say "I am not saying this applies to you" having just very clearly implied it. And then repeating it by talking about discrepancy between those people's perceptions and reality. Don't do it.
The observation of black swans PROVED the hypothesis that "Not all swans are white."
There must be many physical situations that can be mapped to a description in discrete mathematics (which includes integer arithmetic, for example) and which can then be proved beyond question.
I once had a Chinese student and I pointed out to him that limit cycles in otherwise stable integrated circuit recursive digital filters could not exceed a certain amplitude and therefore there could only be a finite number of distinct limit cycles for a given filter (because there are only a finite number of filter states with less than that amplitude).
To my amazement, a week later, he arrived with a stack of paper, having analysed every possible limit cycle for a particular filter. Thus he had *proved* definite results about a physical system. No question about it. Proved conclusively.
If you can't make any sense of what we are saying here, instead of suggesting we are borderline mad, why don't you ask questions to clarify what you can't make sense of and perhaps see what we are getting at?
Some of us who comment here have had experience and track record in dealing with complicated physical realities, including being in situations where saying "We simply can't say" was extremely unwelcome to very forceful and powerful people who wanted definite statements. So saying that we are out of touch with reality may make you feel less uncomfortable but it leaves us just feeling a bit sad (and not for ourselves).
Martin A
Maybe applying the thoughts of a physicist is a bit unfair. Climate is incredibly complex. Any relationships that exist will not be precise. Look at the same sets of data over different time periods, or different estimates over the same time period, (e.g. surface temperature anomalies) and very different interpretations can be obtained. This is without polarized views that have emerged due to very different belief systems. But all is not lost. We can draw upon another subject with similar complexity in the subject matter and even more polarized views - economics. In this area I am a methodological pluralist. However, one of the leading methodologists of last century proposed a way of differentiating between good theory and bad. A good theory should be simple to understand (and therefore very unrealistic) but make bold predictions that come true. Milton Friedman's essay defined post war economics. Fifteen years later he put forward a very simple theory that explained why unemployment and inflation were both rising and would continue to rise. The predictions turned out to be largely correct.
But apply this criteria to climatology and it has spectacularly failed. Accelerating warming, polar ice melt and sea level rise have all not happened despite expert models predicting they would. Heat waves did not become more common post 2003; hurricanes did not become more frequent post Katrina; the Arctic Ocean was not largely ice free in 2013 and my children are growing up not just knowing what snow is, but getting sick of the site of the stuff. Any reasonable person would not expect all the short-run predictions to be come true, but a consistent record of exaggeration shows a lack of understanding of the subject.
Kevin Marshall
Kevin Marshall
Most of the projections you mention are expected to take place over decades or centuries. Most of them are rate or frwquency changes. Expecting them to show immediately in a noisy system is premature.
Remember that at the long term rate of 0.6+/-0.2C/century it takes 40 years + for the warming signal to become statistically significant and the record includes periods of considerably faster warming and flat periods when it appeared to stop warming altogether.
Part of the problem may be that AGW is not one hypothesis, it is a collection of related hypotheses. In the nature of hypotheses , some predictions may be confirmed immediately , some later and some will be wrong.
Entropic man
Martin A
It is a staple of political debate to describe ones opponent as mad something I try to avoid.
I chose Albert as the closest in my experience to a completely alien viewpoint to my own, not as a suggestion that you are insane.
My problem is that I am trained to think along standard scientific lines. This includes using theory and hypothesis as I defined ( anathema to shub ), statistical tests for quality and trends in data (anathema to nial) and the conditional use of may, probabably etc which makes me sound uncertain ( anathema to SandyS ).
I am talking to people who seem to have no objective criteria for judging science beyond opinion, belief and gut feeling. You, as an electronic engineer, operate in a world of precise measurement and mature physical theories. You do not have experience of the complex stochastic systems common in biology and climatology; nor do you find yourself wporking wirh large unxertainties or developing science. In suc areas your intuition is correspondingly uncertain.
We look at what should be common concepts in such different ways that we sometimes fail to communicate at all. The longer I spend here, the more I realise that we are like fencers standing at opposite ends of the mat, not even close enough to properly engage.
Even more depressing, you are at the the most moderate end of the sceptic spectrum. From my viewpoint many others are even more alien.
Entropic man
"You do not have experience of the complex stochastic systems common in biology and climatology; nor do you find yourself wporking wirh large unxertainties or developing science."
EM - you make it clear that have absolutely no idea of the range of areas I have worked in and of my accomplishments.
Martin A
Entropic man
What is an anathema to me is using incomplete data sets which in themselves are a subset of a larger set for which we have virtually no real data and using that and incomplete models to predict anything with confidence.
I echo Martin's
EM - you make it clear that have absolutely no idea of the range of areas I have worked in and of my accomplishments.
Rather than teaching theories to inexperienced adolescents, I spent most of my working life reliability/failure investigations. Every week perfectly good theories would be shot down by the evidence of tests and continued field failures. Many years were spent resolving issues with early generations SRAMS and DRAMs. an employment creation scheme in the early days. Lots of problems were from totally unexpected sources, one in particular took almost two years to track down, and caused Khrushchev's shoe-banging like incidents at meetings with the supplier, who was absolutely certain there was nothing wrong with their product until we were able to prove otherwise.
Perhaps as a result of a life's experience I am quite open to any oddball theory* (including CO2) for the cause(s) of continued climatic changes. However I do think putting everything down to it being solely CO2 is completely misguided.
*I quite like the Lapse Rate theory as a partial explanation as it includes external factors to Earth.
I googled one problem with ceramic packaging and wasn't totally surprised to find people making a career on a similar issue 35 years later. SOFT ERROR ISSUE AND MPORTANCE OF LOW ALPHA
SandyS
Martin A, SandyS
I do indeed have no idea of the extent of your accomplishments and experience. What I can see is the areas where boundary conditions constrain your thought processes.
There are concepts you are unwilling to consider. Martin A has a blind spot for imperfect data. SandyS is uncomfortable with uncertainty, nial rejects statistics and shub would seem to dislike cold objectivity.
My blind spot? Until depression got me I was a rather cold fish. I tend to badly underestimate how much emotion and belief influence opinion forming and decision making. My delusion is that logical argument alone is enough to change someone's mind.
Entropic man
Entropic man
If I may
SandyS is uncomfortable with using partial data and incomplete models as certainty.
I would say that you can't accept the possibility that you are wrong or even not totally correct. Which is interesting in terms of Hypothesis testing in climatology. As a result the thought process is not "How do I make this fail?", but "how do I confirm what I thought?" A test/reliability engineer is always looking for better and better ways to achieve the former objective.
SandyS
"You do not have experience of the complex stochastic systems common in biology and climatology; nor do you find yourself wporking wirh large unxertainties or developing science."
Martin A - are you the Martin A who worked at Bell Labs in the 1980's on the development of new methods for the analysis of large scale stochastic systems?
ex MTS
Sorry, Entropic Mann, but ALL the “projections” (such a sweet term… avoiding the prospect of "predictions" being wrong) that Kevin Marshall highlighted were given with a sense of immediacy: “Our children will not know what snow is!” being a good example; “The Arctic will be ice-free by 2013,” being another (unless you expect there to be another 2013 in a few decades’ or centuries’ time).
Your world-view is completely at odds with reality: “…it can’t be proved wrong.” If that is the view of Richard Feynman, then proving is as valid a term as disproving; unless you think that an object thrown in the air is not accelerating downwards but merely decelerating upwards until it stops and then accelerates downwards (and I have met people who cannot accept any other concept, so you will not be alone). Are you, or are you related to, “Paul from Clerkenwell”?
My problem is that I am trained to think along standard scientific lines.Your absolute and utter arrogance is encapsulated in that one sentence; you have comprehensively put down every other scientist, scientifically-trained and scientifically-thinking person on this site – indeed, every other such person in the entire WORLD with whom you may not agree! One thing that you never even consider, in your vainglorious pomposity, is that, be that as it may be, you clearly demonstrate that a lot of your training has been wasted; in your little mind, on this virtual parade-ground, you are the only one who is in step.
Radical Rodent
My problem is that I am trained to think along standard scientific lines.
My delusion is that logical argument alone is enough to change someone's mind.
Martin A has a blind spot for imperfect data
etc etc - including copy/pasting dictionary definitions when challenged on his understanding of a term.
RR - It's not arrogance (although that's certainly how it comes across). The quoted sentences above confirm what has always been pretty clear - EM is borderline Asperger's (or suffers from ultra nerdism, if you prefer), with very poor insight into other people's thought processes and how they will react to what he says.
And despite all his talk about having been 'trained to think along standard scientific lines', when he gets something in his head, it becomes his reality. His imaginings about my areas of expertise, stated as if they were fact, are just one example.
That's just how he is.
Martin A
EM
I largely agree with your comment of Nov 28, 2014 at 12:08 AM . Many of the signals of warming will take decades to become apparent. But the so-called experts do not recognize this, and nobody within the scientific community calls them to account when they make such alarming predictions that are not based on any rigorous science. But policy is based these alarming predictions, not on the historical data of temperatures rising at 0.2 degress a decade, or sea levels rising at 20-30cm a century.
Kevin Marshall
Kevin Marshall:
Many of the signals of warming will take decades to become apparent.That is assuming that no other factors have any influence on the climate; as the records do show that warming can be positive or negative (i.e. warming or cooling), and we do not have enough information to determine why this should be, it is a strange assumption. The over-weening opinion, nowadays, is that it is all the fault of CO2 – or, more specifically, man-made CO2 (which, presumably, is far more malignant than natural CO2). Quite what was at fault for the other variations is conveniently ignored.
Sorry, Martin A, but you are being far too generous; it is arrogance, nothing but totally contemptuous arrogance – “I am right and you are all wrong. You are idiots if you cannot see that.”
Radical Rodent
RR - I have to concede that, if it walks and, in particular, if it quacks like a duck, it's not utterly unreasonable to call it a duck.
Martin A
EM Wrote:
"My problem is that I am trained to think along standard scientific lines. This includes using theory and hypothesis as I defined ( anathema to shub ), statistical tests for quality and trends in data (anathema to nial)"
I have no problem with statistical tests, what I have a problem with is 'science' like.....
"The average centred on 1998 was 0.44 the average for the five years centred on 2011 is 0.58 and the 2009- 2014 average looks likely to be 0.61."
Despite the fact that the 1998 average contains pre-plateau years?
This is O level maths stuff.
"95% confidence limits. In the context of Hadcrut4, they define their global temperature averages with an uncertainty quoted as +/- 0.1C (+/-2Standard Deviations). 95% of these averages should be within 0.1C of the actual average temperature.
This is a completely incorrect. You can't take a confidence level from a derived average and just apply it to the trend of that average.
A level perhaps?
"Ice volume on land measured by GRACE and Cryosat is decreasing by 500 cubic kilometres per year.
The latent heat of fusion of water is 3.34*10^5 J/kg. 500 cu. Km.of ice is 5*10^14Kg."
500 cu KM is too suspicioulsy round a figure to be accurate. Where is your error analysis in the calculation
that follows?
"Surface area of the Earth is 5.1*10^14M^2. The imbalance is 1.54*10^15 / 5.1*10^14 = 0.3W/M^2."
+/- how much?
"My own hypothesis is that tthere has been s a 0.6C/ century warming trend since 1880 due to increasing CO2"
CO2 wasn't increasing in 1880, what triggered the warming then?
"My problem is that I am trained to think along standard scientific lines"
My father did a physics degree and PhD then ended up in academic engineering. He said that science graduates come out of university thinking they know everything, engineering graduates come out realising they know nothing.
You haven't lost the arrogance of a fresh science graduate.
Nial
EM wrote:
"Most of the projections you mention are expected to take place over decades or centuries. Most of them are rate or frwquency changes. Expecting them to show immediately in a noisy system is premature."
But yet we're supposed to react with immediate panic to an un-remarkable period of warming over a 30 year period?
"My delusion is that logical argument alone is enough to change someone's mind."
You could always give it a shot and see what happens?
Nial
Radical Rodent /Martin A/Nial
I agree definitely the arrogance of a fresh science graduate.
Working in education is the perfect way to keep that particular bubble from being burst by experience.
SandyS
SandyS: …... Ouch!
Radical Rodent
.............. I give up............ You win.