Help needed! What's difficult about fractal noise?

Lost and Confused SignpostI need help. I’ve many times tried to get a discussion going on 1/f noise so that I can understand how to put over this subject so that it is understandable.
And let me be blunt, if only I could get people to understand this, they would never again concern themselves with “manmade global warming”.

Yet, no one ever seems to comment.

I don’t get any idea where anyone is having problems. But again when I saw a chance to bring up this subject using the similarity of the Antarctic Ice and Central England Record, to show that they are both 1/f type noise, no one commented – I thought this time someone might get it and I would have some kind of response.
So, what’s wrong with this article:

Global Warming disproved: Bad posts, CET, Antarctic Ice and Fractal Noise

I mean it. Global Warming disproved. – at least “Global warming” in the sense used by the warmest as something “manmade” or a change that is recent. If only people could understand it and the implications, then it’s all over and we can all go home.
But never have I had anyone comment on what I’ve said.
And, it’s not like I didn’t put it prominently in my submission to parliament after Climategate:

I’ve no idea what it is that people are missing.
I can understand why the issue arises. I used to design very sensitive analogue circuits where this type of noise was very important. So I was dealing with these types of things all the time, and it just seems second nature to me. So what might be obvious to me might be intractable gobble-de-hook to someone else. And likely if I simplify it in the wrong way it just gets worse.
But unless I get some comments, I have no idea where I need to improve. What is wrong with the way I’m presenting this. I’m completely stumped and really do not know how to approach this subject so that people understand.

This entry was posted in 1/f, Climate. Bookmark the permalink.

32 Responses to Help needed! What's difficult about fractal noise?

  1. TinyCO2 says:

    Did you ever see the BBC documentaries about chaos?

  2. Thanks – I will watch tomorrow.
    Do people understand that saying a climate is predictable – is the same as saying it has no chaos?
    So, it was particular ironic to watch the first few minutes where they implied all natural systems have chaos.
    … it’s chaotic – but in an entirely predictable way!

  3. manicbeancounter says:

    There might be some semblance of order possible in even chaotic data. For instance, it might be possible to devise a betting system that will “beat the odds” at horse racing. The problem is then to discern between the genuine successful system, and the compulsive gambler who believes they have a superior system. In betting, the test of the successful system is one that at least breaks even over a large number of bets, despite possibly failing in the majority of individual cases. The compulsive gambler will always find special reasons after the event why in a that particular case the system failed, but will not accept that that the system as a whole does not work. It is the same for the most devoted England football supporters.

  4. I would prefer to put it this way: it is very predictable when a flow will become turbulent, it is not possible to predict each piece of turbulence.
    Likewise, if a system has a natural “resonance” at around 60years, then completely random noise will manifest itself as an apparent 60year cycle – which is in fact entirely random (within the band at which it resonates).
    However, the key thing when trying to work out whether a system is behaving “abnormally” is to first work out how it behaves “normally”. This is usually achieved by producing a generalised model of the natural variability. For climate, this tells us that we see large long term trends and apparent cycles, diminishing in size as we go to shorter periods. And this is what we would expect to see because 1/f or fractal noise is as one person said “ubiquitous” – it’s seems to be almost everywhere in nature.
    The question we should ask is not: “is climate 1/f noise”, but instead, “why isn’t it 1/f noise”? Or even “why have climate academics of all fields of study chosen to reject the most common type of noise in physical systems when they have no evidence to do so.”

  5. tchannon says:

    A starting point might be an article I produced 2013 (comments are closed). Not very good but might do. Looks like I have already done CET for you and I indicate an available tool.
    http://tallbloke.wordpress.com/2013/02/07/hurst-exponent-a-long-omitted-subject-from-the-talkshop/
    I have more planned, time, energy and so on…
    Given the low interest is more likely on my own blog.

  6. Derek Alker says:

    Mike, to answer your question it might be useful to ask yourself do you still believe there is a greenhouse effect?
    ie, the people will continue to unquestioningly believe untill the proper higher authority tells them otherwise, regardless of the facts to the contrary they may be presented with by other mere mortals……

  7. Great! We are talking about the same group of phenomenon!
    I have seen this general approach to explaining the subject used several times – presumably by those from a statistical background, but whilst people like Anthony Watts obviously warms to it over my own way of presenting, I feel it is so disconnected from the climate model that even for those like the Met Office, it has no impact.
    Also, another fundamental problem with a statistical approach is that there is no real understanding of the mechanism or mechanisms that give rise to this type of, 1/f. fractal, flicker, pink/red, autocorrelation noise/variation.
    And this makes hard statistics damned near impossible in some aspects.
    So, e.g. one of the fundamental problems of the subject is this: “given a certain size of climate signal – what is an abnormal trend”. However, in order to answer that question you need to have a generalised theory of the mechanism acting in order to work out the probability of this or that trend.
    Another approach I’ve considered is to do that statistics in frequency space. (In other words do a Fourier transform).
    My hunch is that the test is now much simpler because it is just the sum over all the frequencies of the probability of there being signal rather than noise at that frequency. The only assumption one then needs is the “normal” level of signal at each frequency and either this is assessed using a frequency analysis of a “control” period without the signal or we use a generalised model of 1/f^n.
    I have numerous times tried to find papers on the “statistical analysis of 1/f noise”. I have also tried to find “statistics in frequency space”.
    Either I’m using the wrong terminology, or it just isn’t possible, or no one has done it, because I’ve failed to find anything.
    PS. Doesn’t “Guassian noise” refer to the shape of the probability distribution. So, irrespective of whether it is normal(white) or autocorrelated noise (1/f), it can be Guassian or not. For example, if we have a particle detector which gives a “blip” each time it detects a particle. The noise is white noise, but the probability distribution is two peaks “blip” and “baseline”.

  8. if by greenhouse we mean literally the effect in a greenhouse then obviously not.
    if however we mean “radiative absorption and emission”, then does CO2 interfere with IR – that’s for sure. Will that tend to block IR transmission, again that is a sure thing. Will that tend to increase the temperature of the earth’s atmosphere. Then yes.
    However … CO2 is also an emitter of IR. At temperatures higher than the sink (space), it will also tend to promote IR loss by acting as a vector (intermediary) being heated by the surrounding atmosphere, and then flinging this to space. So it will cool.
    I’ve not explicitly seen this in a model – I would hope it is – but when so much else has been wrong I don’t know.
    So, CO2, is both a greenhouse and “anti-greenhouse”. And the effect will vary according to prevalence of CO2, the type of upgoing and downgoing radiation and the temperature.
    In theory, with a sufficiently big database like HITRAN, I could do that calculation. I expect I would agree with the experts with a static analysis like that.
    However. The atmosphere is not static. Any heating of the atmosphere will cause an imbalance resulting in dynamic effects, upward movement for IR heating and downward movement for IR cooling. That again could be modelled, but then we have all that stuff called water. Now water vapour is also a “greenhouse gas” and also an “anti-greenhouse gas”. So, again we have all the same problems but 10x worse. However, to add to the misery, we have all those fluffy things in the sky which for some odd reasons scientists haven’t yet found a hugely complex latin name and so refer to them as “clouds”.
    That, might just be modellable. However, now we need to think about what all that rising and falling and heating and cooling air does to air pressure and consequently to temperature. And. there are undoubtedly good arguments showing that the temperature rise to the earth’s surface is what we might expect to get from the change in air pressure. These are particularly annoying, because i know I should have an interest in thermodynamics but honestly I don’t. So, how could I judge?
    Which leaves me in the situation of not knowing.
    However, looking at the climate signal, I do know the earth is particularly stable in an inter-glacial, so no matter what the effect of CO2, there are clearly large negative feedbacks that will never allow us to significantly warm the world just by releasing CO2.
    Which means that even if we take the worst case scenario of 1.2C warming proposed by the worst eco-zealot academics, the worst case scenario for warming is less than ~1C.
    And, in the realm of climate, warmer appears to be better and 1C warming appears to be such a small change that whilst it is scientifically interesting, it does not appear to me to be worth any real change in our behaviour.
    So, do I believe in the greenhouse effect? It certainly has not been disproven, there is very strong evidence for the radiative effects of CO2, and when experts look at this they all seem to agree that this will cause warming. So it is certainly believed by a lot of people to exist.
    So, there is good laboratory evidence and a general consensus of experts (even those sceptical) so this appears to be very well supported.
    In contrast, the positive feedbacks are just codswallop.
    So, where do I focus my efforts? On an area, where even if there were a problem, it would be highly problematic to persuade people it existing
    … or on pushing the noses of the climate academics in the codswallop of positive feedbacks until they have the honesty to admit it is codswallop?

  9. Derek Alker says:

    2nd paragraph needs a rethink….
    This planet’s atmosphere is not rigidly contained, as the diurnal bulge proves beyond any reasonable doubt.
    ie, CO2 merely scatters IR with no heating effect upon the atmosphere. Latent heat release by water vapour condensation does release heat higher in the troposphere. The two often being misunderstood / represented…..

  10. I think I was careful to say that there was emission and absorption. Which of those will dominate will depend on the temperature of the CO2 and the frequency profile of the incoming radiation.
    To me it feels like a massive line of Chinese whispers – that even if each step appears to be remarkably simple, the overall combination of all the different layers may be startling different from what a simple analysis might suggest.
    So when someone says “is there a CO2 greenhouse effect” I can only say that I imagine that they’ve done calculations and that in the same circumstances I might have used their approach and came to the same conclusion – but then again, given the clear complexity involved, I might not – and what past history shows is a huge bias toward anything that suggests warming, so all I can really say with any certainty is that I am more likely to come to a lower figure than they would.

  11. Derek Alker says:

    Mike you were careful to say CO2 absorbing IR causes warming of the atmosphere. It is a questionable assumption / assertion, as I have hopefully pointed out, because the atmosphere is not rigidly contained. That fact changes “everything”, somewhat inconveniently..

  12. If CO2 absorbs, then it warms, if it emits it cools.
    I’d need to see primary data to see what actually happens and develop a model that fits that data to go much further.

  13. TinyCO2 says:

    The documentary is good as explaining how chaos forms patterns and what might seem like a simple machine, isn’t. Like the orary, climate seems simple but the imperfections of nature don’t allow the handle on the climate models to turn many times before it deviates from reality. When you think about chaos you realise that the temperature stations don’t come anywhere near enough to reality to be used for the starting conditions. We’re not measuring a fraction of the things that are significantly destabilising the simple system, let alone the tiny things.
    I don’t know if they’ve tried it but evolutionary computers seem to be the only possible answer, where the computer decides the climate rules, not the programmer.

  14. Richard Mallett says:

    The Hadley Centre CET annual average from 1850 has increased by 0.62 C per century. From 1659-1850 it increased by 0.18 C per century. From 1772 (when they started daily readings) to 1850 it increased by 0.37 C per century. That suggests that, since industrial activity started around 1850, there has been a further warming effect of 0.25 to 0.44 C per century.

  15. Robertv says:

    How do we want to understand the atmosphere if we don’t understand the oceans. We know there is a +/- 800 year lag between atmospheric temperature rise and CO2 rise. Does that mean that the MWP is the reason CO2 is rising like it does by releasing the gas it absorbed 800 years ago.
    But is CO2 a greenhouse gas? Would the dry desert stay warmer at night with CO2 at 800 ppm?

  16. Derek Alker says:

    If you would like to see primary data Mike, try the CO2 in a bottle experiment, but as Carl Brehmer did put a hole in the lid of the container. CO2 only container warms no more and no less than the air only container, WHEN not rigidly contained. Basic principle is that as CO2 is heavier than air if rigidly contained (ie, no hole) then the more mass of CO2 in the same volume will heat more than just air. This is NOT CO2 absorption causing warming, it is more mass rigidly contained that causes warming. Two very different things that are usually confused / spun…..
    Radiatively able gases that absorb IR expand, which is not the same thing as heat up, unless rigidly contained.
    Have you read Robitaille’s little heat engine yet?
    http://www.researchgate.net/publication/258883304_The_Little_Heat_Engine_Heat_Transfer_in_Solids_Liquids_and_Gases

  17. Derek Alker says:

    Robertv you may be interested in a question I asked myself a little while back.
    ie,
    Is CO2 a negative feedback within earth’s climate system?
    http://www.globalwarmingskeptics.info/thread-2239.html

  18. Derek Alker says:

    Excerpt –
    “Why is it important that the oceans, water vapour and CO2 ARE negative feedbacks?
    The IPCC ignored water in the atmosphere as much as it could (for example water vapour [THE radiatively able gas of the atmosphere] is not given a greenhouse gas value compared to CO2 by the IPCC) to avoid the issue that they are obviously looking at earth’s climate system from the wrong view point. In short, they are using the wrong paradigm. That is why the IPCC et al tries to portray water vapour and CO2 in particular, as positive feedbacks.
    The oceans and the water cycle are the dominant and negative feedback within earth’s (stable) climate system, AND, CO2 is also a negative feedback and coolant within earth’s climate system. The GH “theory” based paradigm that AGW fears are based upon has failed and it is already way past the time it should have been abandoned.”

  19. it reminds me of my experiment trying to freeze ice on a cloudless day.
    In theory, the sky being below zero, any surface exposed to it and sheltered from the sun should freeze. (The Romans used this method) I therefore set up an open top wide-neck flask with a suitable thermometer. Far from freezing, I got no measurable change. It then dawned on me that the very small air current was just blowing any cold air that developed away.
    I realised that the energy needed to “scour” out my flask was a lot less than the 10m pits that the Romans used – and that they probably set them up in places with no wind (not Scotland where the wind almost never stops).
    Likewise, doing this CO2 experiment on a bottle looks to me like a complete waste of time.
    if I did the experiment, I would want to quantify the effect of CO2 and pressure etc.

  20. manicbeancounter says:

    Derek,
    Your question is a distraction and essentially unscientific. If we wish to understand the “real world” it must be from a combination of hypotheses and observations. If we try to impose a particular set of beliefs (or rejection of beliefs) then that can influence the outcome. For each individual this is unavoidable. Without structure, existing knowledge and selection of data, then the hypotheses will be empty. But only thinking along mutually-accepted lines will give prominence over rationalising anomalies and defending current beliefs rather than pursuing alternative hypotheses. It is once these alternatives have been specified and compared with the data that we can compare and contrast the different approaches. The outcome might be to clearly reject one hypothesis. More often there can be a synthesis of the different hypotheses, or modified hypotheses being derived.
    Being able to pursue and develop hypotheses, no matter how strange, is fundamentally important. Even if that pursuit leads up blind alleys, it can lead to new insights or perspectives on existing science. Being consistent with exsiting group or personal beliefs goes against this exploratory process.

  21. Richard Mallett says:

    Yes, for example, plotting CO2 against temperature since 1880 gives only a 75% correlation (longer periods give even lower correlation) so CO2 on its own is not the driver of temperature that some would have us believe.

  22. catweazle666 says:

    As I understand it, the mechanism by which the GHG effect works does not involve increase in temperature of the GHG molecule, this is a red herring.
    When a CO2 molecule “captures’ a photon of the appropriate wavelength (~14 microns) there is no thermalisation of the molecule, as the photon raises the internal energy level of the molecule. This molecule is on average travelling away from the Earth’s surface, which is therefore cooled, and were it not captured, would be re-radiated back into space. Subsequently, the photon is released, but in a random direction, thus 50% of the re-radiated photons will be emitted downwards towards the Earth’s surface, causing it to be re-warmed. This process may be repeated a number of times.
    Eventually of course all the photons will be re-radiated to space, but its path length and residence time in the atmosphere is increased.
    All experiments purporting to measure increased temperature of vessels containing CO2 are in fact entirely bogus, for whatever reason.

  23. catweazle666 says:

    Anyone who has not read James Gleick’s excellent work “Chaos” needs to do so as a matter of extreme urgency, attempting to debate climate science without at least the level of knowledge in that work is futile in the extreme!
    http://www.amazon.co.uk/Chaos-Making-Science-James-Gleick/dp/0749386061
    Ironically, it was Edward Lorenz, a climatologist, whose research revealed much of our understanding of non-linear dynamics AKA chaos theory, and explained precisely why any attempt to model the climate on a computer – no matter how powerful – would be doomed to failure. All the increased computing power does is gives you the wrong answer quicker.

  24. Robertv says:

    Fractals

  25. Derek Alker says:

    Manicbeancounter, thanks, I think. LOL.
    Fair point, however, when was the hypothesis that is “greenhouse effect theory” proven, or even tested? A failed hypothesis is just that, and at present a politically correct blind alley, that is very useful for taxing and increasing control over the population purposes. A heady (politically convenient) mix, but NOT scientific in the slightest..
    My own understanding is that the failed hypothesis led me back to a thermodynamics approach, WITHOUT the use of “black body”. This later point is important because how do you cure a sick unicorn? The answer is any way you want to, BECAUSE it is an imaginary beast… The point is simple, black body is imaginary, BECAUSE it is a set of unphysical assumptions. Therefore we KNOW, but usually refuse to admit that, a black body explanation (in part or whole) DOES NOT explain what happens in actual reality.
    The bare earth model is a black body model, bad start.. AND, GH “theory” is a black body explanation. GH “theory” employs the bare earth model, AND P/4, which we also know is unphysical.
    In the scientific sense GH “theory” is zombi science. It was never proven, and it refuses to die. Mike’s answer to my comments on this thread illustrate this very well, AND his follow up post is even more illustrative of how belief can become so entrenched in the common sense of ordinary people as if it were the truth and the only truth of the matter. Even to the point were everyone has their own pet version of the “theory”……..

  26. Surely an important thing you are missing is that even the longest time periods you looked at don’t include known major climate events like ice ages. A graph of global temperature that includes the step-changes around ice ages would not be fractal in the way you suggest?

  27. A good point. The best way to explain that is that the ice ages are a resonant amplification of the noise.
    An example of such an effect is when a tree or blade of grass sways too and fro. The rate of swaying is determined by the elasticity of the plant and its mass. But the scale of that rocking is determined by the scale of the “noise” in the shape of the wind.
    note: the ice ages are more complex than that.

  28. But surely if a series (any series) isn’t fractal in the longest time frame then it isn’t fractal overall, even if episodes within it are. And if it isn’t fractal then it isn’t fractal noise?

  29. catweazle666 says:

    Milanković cycles.

  30. Tomorrow was a bit later than planned, but I can see the connection between Turin’s work and climate models.

Comments are closed.