Thursday, June 27, 2013

The IPCC and the Flat Earth Society may merge now

Anthony Watts has highlighted an article in,
Actually, even the Flat Earth Society believes in climate change.
The president of the International Flat Earth Society, Daniel Shenton, has endorsed the climate hysteria.

Even Shenton realizes that it makes sense for others to make fun of his society (whose views are "unorthodox", using his language) and to present it as an example of things that are absurd. When someone believes preposterous, patently and demonstrably false propositions, we sometimes compare him to the Flat Earth Society. Shenton is kind and doesn't take these references personally.

We sometimes use the Flat Earth analogy to humiliate the climate alarmism as well except that in the case of the climate alarmism, the similarity between the two examples of bad science goes well beyond their being bad science. The climate alarmists and the Flat Earthers build on extremely similar if not isomorphic misconceptions, flawed assumptions, mistakes in their reasoning, and on analogous demagogic verbal tricks to fool others and themselves.

The basic similarity is the two societies' dependence on excessive linear extrapolations taken well beyond the self-evident point where the extrapolation could be justified. What do I mean?

Well, it's simple. The Flat Earthers look around their headquarters and they see that the Earth is apparently flat. So they extrapolate this observation: the whole Earth and every place where we can walk or drive must be flat, too. Everywhere. Indefinitely.

The climate alarmists are doing the same thing. They see an apparent increasing trend in the graph of the global mean temperature since the late 1970s so they promote it to a law of physics. Such a trend must surely have a cause that will last indefinitely. It will produce warming everywhere and at all times.

In fact, the degree of unjustified extrapolation is worse in the case of climate alarmists. The Flat Earthers think that the Earth is flat but the plane into which the surface of the Earth belongs is an arbitrary plane that is not claimed to be parallel to another plane you may have thought about – such as the ecliptic.

On the other hand, the climate alarmists assume that before the humans began to influence the climate, the global mean temperature was not only flat but constant. The slope was equal to zero. Needless to say, there is overwhelming evidence that the graphs of the global mean temperature were never flat for extended periods of time and lots of evidence that the Earth isn't flat everywhere, either. But none of the two groups cares because the opinion that "it's right for the Earth or the Earth's climate to be flat" is a dogma that is more important for them than any direct or indirect empirical evidence, than any contradiction one may find in their belief system.

To discuss these issues with some real-world quotes in mind, let me copy some seemingly semi-intelligent sentences offered by the president of the Flat Earth Society, Daniel Shenton:
Sticking your head in the sand might make you feel safer, but it’s not going to protect you from the coming storm.
Can't you just dig through our planet to the other side of the planar Earth where no storm is coming now?

Oops. This wasn't even semi-intelligent but it wasn't Shenton's sentence, either. It came from the president of the U.S. called Barack Obama. The quote by Shenton I want to discuss is this one:
I accept that climate change is a process which has been ongoing since beginning of detectable history, but there seems to be a definite correlation between the recent increase in world-wide temperatures and man’s entry into the industrial age. If it’s a coincidence, it’s quite a remarkable one. We may have experienced a temperature increase even without our use of fossil fuels since the Industrial Revolution, but I doubt it would be as dramatic as what we’re seeing now.
I hope everyone will agree that I am not unfairly humiliating the climate alarmists because the structure of Shenton's propositions looks civilized and the content is pretty much identical to what we often hear from other climate alarmists, too. Still, it's completely wrong and crucially rooted in the Flat Earth reasoning.

Why it's wrong?

First of all, science has accumulated overwhelming evidence of upticks and dips not only in recent decade or centuries but also in recent thousands, tens of thousands, hundreds of thousands, millions, hundreds of millions, and billions of years, too. We are sure that there have been significant temperature changes not only before the industrial age but before the humans were walking on the Earth (whether it was flat or round), too.

So if you try to quantify the correlation based on extended periods of time which is really needed because you have to compare the industrial age with a non-industrial age to back a statement similar to Shenton's, you will inevitably conclude that there is no correlation here, at least not a statistically significant one.

Moreover, it's not true that the fossil fuel-driven climate change theory predicts that the temperatures start to rise as soon as someone declares the industrial age to have started. Instead, the global mean temperature should be correlated with the CO2 concentration in the atmosphere. These two quantities seem correlated in the last 100 years or so but the bulk of the correlation is explained by the fact that both quantities saw some overall rise during that time – much like thousands of completely different quantities that are surely not causally connected to the temperature rise.

There's no detailed high-frequency correlation between the two functions of time; at the level of interannual temperature variations, the claimed CO2-temperature correlation surely breaks down. One may say that in every century, it's almost certain that one sees either an increasing trend or a decreasing trend – the linear regression is extremely unlikely to produce a vanishing slope – and because the sign of the slope of the CO2 concentration is well-defined in pretty much every century as well (surely in the 20th century in which the rise of CO2 was dominated by our activities), there's a 50% probability that these two signs agree. A 50% confidence-level argument is much less than 1-sigma evidence. It's no evidence at all. And if we acknowledge that the anticorrelation could be abused as well (and yes, some people have been hysterical about a new coming ice age in the 1970s, too), we would be 100% guaranteed to see this kind of evidence (correlated signs between the CO2 rise and temperature rise in a given temperature) even if we assume that there exists no causal relationship between the quantities whatsoever.

So the strength of this "evidence" is really strictly equal to zero. The "remarkable" observed pattern has to be there regardless of the validity of the theory about fossil-fuel-driven temperature changes.

Both the president of the Flat Earth society and the majority of the climate alarmists also suffer from the aforementioned "wild extrapolation" disease as well as the fear of proper quantitative comparisons of numbers.

Shenton talks about a recent uptick of the global mean temperature which has more or less continued for a century or so (although it's easy to find a recent 15-year period without any warming and 30-year periods in the middle that display no warming, too). For Shenton, it's already a "coincidence" that he's willing to promote to a permanent law of physics.

As I already said, this unmasks his characteristic Flat Earth reasoning, something that most climate alarmists share with him. When a quantity increases for a century, a pretty long time, it still doesn't mean that the rise is gonna be permanent. It is just circumstantial evidence that the typical periodicity of some of the Fourier components that importantly contribute to the function of time is probably longer than a century. But if the function \(\sin(\omega t)/\omega\) changes its slope by less than by \(O(100\%)\) per century, it doesn't imply that \(\omega=0\) i.e. that the function is \(t\), a linear function of time. Instead, it just means that \(\omega\) is smaller than the inverse century or so.

There may exist – and at least in some contexts, there undeniably exist – many cycles and processes whose characteristic timescale is longer than a century. The observation that the trend has a more or less uniform sign over some period of time simply doesn't imply that a constant such as \(\omega\) is strictly zero or that its inverse \(1/\omega\) is strictly infinite (again, it would be an isomorphic mistake to the Flat Earthers' flawed "derivation" that the curvature radius of the Earth has to be infinite because it seems very large relatively to our bodies). The observation of "slow changes" only means that we can suspect that an inequality for \(\omega\) holds. We're measuring the upper bound on \(\omega\) that dominates in certain kinds of changes of the temperature.

Shenton also mentions that he "doubts" that the increase of the temperature would be "as dramatic as what we see now". Again, this is a carbon copy of statements you may hear from hundreds of his fellow climate alarmists. This statement is also completely wrong; and it is utterly irrational. The irrationality has two main components encoded in the words "doubt" and "dramatic". Concerning the word "doubt", Shenton seems to suggest that the temperature change without fossil fuel would be lower than the observed one because he "doubts" it would be equally large. But he doesn't offer any argument or a calculation of the "fossil-fuel-free temperature change" so the word "doubt" only means that he predecided what he wanted to believe. This whole line of propositions pretending to be an argument is a classical case of circular reasoning. He doubts that fossil fuels could be unimportant which is why he also has to doubt that the natural temperature change could be comparable to the observed one and he adjusts all other opinions (and sometimes facts) to his preconceptions, too. But there's actually nothing whatsoever to support his beliefs.

What about the word "dramatic"? What does it mean for an increase of temperatures in 100 years to be "dramatic"? If you're rational, it means that it's greater than a threshold that you choose to divide the "dramatic" and "not dramatic" intervals. Now, a rational person must ask – without irrational prejudices – how high this threshold is? There are various benchmarks (and classes of benchmarks) that you might choose to define such a threshold.

Needless to say, our high-frequency data showing how the global mean temperature changes annually from one year to another or between one decade and the following one only exist in the recent 100 years. We really don't possess any geological data that could reliably reconstruct the annual temperatures from individual years or decades well before 1900. The older, e.g. geological proxies always tend to smear and homogenize the temperatures over timescales comparable to a century or longer so the reliable information about the magnitude of faster-than-centennial temperature oscillations in the distant past is lost.

What does it imply? For a rational person, it simply means that what we have observed in the last 100 years should define our expectation about what is the natural, normal change (or dynamics) of the global mean (or other) temperature in one century – to say the least, it's in the ballpark of the normal changes and it defines the right order of magnitude of such centennial changes. We just don't have any better numbers if want to arrive to a justified expectation and the Earth's climate system is just too complicated for us to calculate this slope from the first principles. At least, it seems impossible if we want the error of such calculations to be much smaller than 50 or 100 percent.

So we don't have any other century that the more or less known change of the global mean temperature in the 20th century could be compared with! That's why there exists no empirical basis for the assertion that the 20th century temperature change was "dramatic". Dramatic relatively to what? It was the first century from which we could have learned what is "normal" but some people have learned nothing. To "doubt" that the observed change was comparable to the normal, natural centennial temperature changes means to be biased.

We may also compare the 20th century temperature change with some other temperature changes – not necessarily those of the Earth's global mean temperature. The LHC accelerates the protons from the 1.9 K (colder than the cosmic microwave background) needed to operate the magnets to 40,000,000,000,000,000 K which is the temperature associated with the proton-proton collisions. OK, you don't want to hear about the LHC but even more mundane situations are connected with large temperature swings.

The temperature on the Moon changes between 100 K and 400 K within half a day (it's the Sun, and not CO2, that is responsible for these swings). That's about 400 times wider interval of temperatures than the claimed "dramatic" temperature change that the Earth needed to be accumulating for 100 years. Even on Earth, temperature changes that are just 10 times smaller than those on the Moon may occur within a day or two (Czechia experiences temperatures that are 25 °C lower than those we saw a week ago – and be sure that I remember such changes from my childhood as well, they were always possible). Are you really serious that 0.7 °C temperature change per century is "dramatic"? Write a drama about this "dramatic" change and check whether it will beat William Shakespeare's dramas.

Most of us can't feel a 0.7 °C temperature change of the air surrounding us even if it occurs within a fraction of a second. Now, slow down this negligible temperature change so that it is diluted to a whole century. Will you feel it or is it more likely that you will die before you feel anything related to the temperature?

The adjective "dramatic" for the temperature change 0.7 °C is just complete rubbish. The right adjective is "negligible" relatively to every change that matters and "by definition normal" relatively to the centennial temperature change we should expect based on the empirical evidence – and our expectations should be empirically rooted, after all.

I think that many climate alarmists must know that phrases such as "remarkable coincidences" or "dramatic changes" are lies. But they have almost no problem to emit these lies because these words are "qualitative" and not "quantitative" so it's not really possible to rigorously prove that they're lies. I can convince any sensible and honest person that the 0.7 °C temperature change seen in the 20th century is negligible but those who claim that the change was "dramatic" may always construct a contrived interpretation or argument that will justify the otherwise indefensible adjective. By these demagogic words, they are changing the gullible people's perceptions but they're also protected against rock-solid proofs that they're liars because their statements aren't really well-defined.

They're still dishonest to neglect the actually crucial empirical and other scientific data and to pronounce similar untrue albeit fuzzy adjectives but honesty isn't their top priority. They're obsessed with something that they consider more important than the truth. That's the ultimate reason why Daniel Shenton denies the obvious and elementary proofs that the Earth isn't flat (just use a percent of the Flat Earth Society's budget to buy an air ticket around the Earth to check that it's round!) and why he and his fellow climate alarmists continue to spread the preposterous, indefensible, insultingly stupid myth that a dangerous global climate change is underway.

And that's the memo.


  1. Just reading an article on the Science & Vie saying that the gaps between summer and winter temperatures are dropping in the Artic. They say that the North of Sweden could be like the South of France in 2091. You were right Lubos, Science is folkloric ;-).

  2. I just used this graph , , from the .ppt I'm putting together on , well , "how to calculate the temperature of a radiantly heated colored ball" to perhaps put a dent in the true belief of a talented programmer friend . He couldn't believe that's all this hysteria is over a near noise level slope of 0.01 . I had to point him to Richard Lindzen's great graph , , to show that , yes , the total change in generally asserted mean temperature has only been about 0.3% at most .

    One of the first things which indicated to me the pathetic state of even the most basic mathematical analysis by "climate scientist" was no mention or measurement of the profound effect of CO2 on our diurnal , and equatorial-polar , variance by transferring surface heat back and forth to the whole atmosphere . I still have never seen an analysis of this most obvious effect .

  3. Hi Lubosi.

    I do not like that likening "flat Earth" to global warming. I always have seen flat earth as in principle wrong, yet in practice useful approximation. People who "work" on projects smaller than 1 km, and never walk farther than 20 km from their place (which until recently - 1700 - were almost all people) could assume flat earth and get pretty accurate results. That is the reason why the idea lasted so long.

    On the other side, I see no scale or setting where AGW would give you similarly accurate, or even useful at all, results (I do not count grabbing taxpayers money for AWG grants and similar criminal activities). So at the end, compared to Flat earth theory, AGW is much more wrong and it should strive to improve and achieve FE accuracy.At that point they still will be wrong, but not quite as obviously as now.