Another academic shows they are clueless on climate.

Over at “The End of Physics” we find a gem of a piece from this academic who says:

This one is particularly irritating because it’s essentially a silly circular argument. “Yes, we’ve warmed. However, we can’t tell, using a statistical model, if that warming was natural or not. Also, the only way we can tell if it is statistically significant is using a statistical model. Therefore we don’t know if it is natural or not.”

You’d like to think that people might eventually be embarrassed to have made such an argument.

On the positive side, he seems to have grasped the basic theory that one requires a model of “what is normal” in order to tell what is “abnormal”, but on the negative side, instead of admitting this means we do not have any certainty of human causation, he instead cannot bring himself to accept this, so turns the argument on its head to claim: “because we must be able to know it’s caused by mankind … it’s silly for skeptics to demand that we know what is normal …” before he jumps in like the big twit he is to say “of course global temperature is abnormal … I know because I’m a omnipotent academic and therefore the basic laws of physics and statistics must be wrong.

It’s a basic requirement that in order to know what is abnormal, we must first know what is normal.

Lets take a simple example. We measures the heights of children and finds that they are 1.01m, 1.02m, 1.04m, 1.05,1.06m   and 1.35m. Is the 1.35m kid “within the normal distribution”? It certainly looks skewed, but what if I were to say the class were a mixed year class with more younger children and that there is a distinct sex difference in height. Now would the kid 30cm higher than the rest be “abnormal”?

Of course, instead of a simple “normal” distribution, we are now dealing with a complex probability curve with 4 possible peaks (2 age groups, 2 sexes). Now simple “standard deviation” and simple “Noddy statistics” don’t help us much – at least not unless we do a lot of ground work, measure a lot of children and work out what is … normal for children of both sexes of this age group.

But no, The End of Physics, can’t possibly accept that he has to understand what is normal for a complex sample like this. He demands that we just apply the Noddy Statistics he was taught and make a decision. And he is derisory to those who say:

We can’t tell, using a statistical model [on the available evidence], if that kid is normal or not…. therefore we don’t know.

He knows he knows that the kid is “abnormal” and is contemptuous of any who suggest doing a more detailed analysis to understand the probability distribution of this complex sample. Instead he blurts out:

You’d like to think that people might eventually be embarrassed to have made such an argument.

Let us take another example, closer to the climate issue. We measure the level of a water against a dam even 30 minutes and obtain readings of: 3.15m, 3.12m, 3.19m & 3.55m. Again, the last appears to be abnormal, but is it? …. well I forgot to tell you the dam was a dam against sea water and that I was taking the measurements around low tide. Suddenly, the dramatic rise in the last reading is no longer “abnormal” but starts to seem quite normal.

Even if we didn’t know the time of the tide or even how much it rose of fell, if we had taken some measurements over the last few hours, we would have got a feeling for how much the water rose and fell naturally.

Likewise, just as the tide is a long term trend on the water level which superimposes itself on the shorter wave-to-wave variation, so we know that the climate has long term variations of many forms from Atlantic Multidecadel Oscillation to El Nino, through little ice-ages to actual ice-ages. These variations are part of the normal, natural variation of the climate. And, like the tide, we do not need to know precisely when the tide is coming or even exactly how big it is to know that we expect considerable change in the water level over a day. So, even a rise in water of several meters would not (without other factors) be a cause for alarm.

And just like the tide, if you don’t have a model of “natural variation” in the climate which includes natural long term changes, you will be misled into believing something is “abnormal” with the climate when all you are seeing is natural long-term variations. So, unless and until there is a model for natural long-term change, one cannot know whether a particular change is normal or abnormal.

One cannot know whether it is human or natural.

And to suggest you know for certain, either shows you are corrupt, or are just pig ignorant of the real behaviour of real systems with natural variation.

And this is where “The end of physics” just hasn’t a clue. Like so many of the clueless academics who don’t seem to ever have encountered noise in real life situations, he thinks the only way to understand natural variation is to “understand” what is causing it – to build some kind of massive model – filled will all kinds of silly assumptions – of which the biggest is that natural variation is not present. And he thinks that by “understanding” through models and “consensus” (or in reality failing to understand) what causes the climate to vary, that he can then assess how much natural variation there will be.

This is complete nonsense – it is the nonsense you get from ivory tower academics who haven’t a clue how to deal with real-world systems.

Any instrumentation engineer knows that it is almost trivial to characterise a system’s noise/natural variation. It doesn’t take any “understanding” of the cause of the noise to build up a model of what is “normal” variation within a system.

 

This entry was posted in climate. Bookmark the permalink.

9 Responses to Another academic shows they are clueless on climate.

  1. Richard Mallett says:

    I have taken a stab at that by determining :-

    (a) Hadley Centre Central England Temperatures have been rising since 1850 at 0.62 C per century.
    (b) From 1659-1850, Central England Temperatures rose by 0.18 C per century.
    (c) From 1772-1850 (when daily records started) Central England Temperatures rose by 0.37 C per century.
    (d) Therefore, since 1850, the rate has increased by somewhere between 0.25 and 0.44 C per century.

  2. That’s interesting – although I don’t quite understand how (d) follows from the rest.

    Something that I would like to do – would be to get an idea of probability distribution of trends. So, e.g. I would like to see the spread of decadel trends, bidecadel & tri-decade.

    From this one should be able to say “the probability of 3 decades of warming in a row of more than this trend is …”

    • Richard Mallett says:

      I just subtracted the pre-1850 trend from the post-1850 trend. Is there anything wrong with that ?
      Since 1850 globally, there have only been two warming periods – 1911-1944 1.95 C/century and 1976-1998 3.43 C/century, so three warming decades in a row would be very unusual.

      • Now I understand – as written it suggests the rate is changing from 1850 onward whereas you mean the post 1850 trend is between 0.25 and and 0.44 larger than the pre 1850 trend.

        And technically this would not capture a stepwise up/down change in 1850.

        Since 1850, there have been 16 decades so, 14 possible sequences of three decades. The chances of getting three in a row is 1 in 8, so you would expect roughly 14/8** runs of three warming decades so we expect ~1.75 since you can only have an integer number, the most likely number is 2 – which is precisely what we have got.

        **It’s more complicated than that as they can overlap.

        • Richard Mallett says:

          Yes, I only made the split at 1850 because that’s when I think the industrial era is assumed to have started.

          • I think it’s more that 1850 is a nice round number and roughly when there was reasonable coverage of temperature.

            However, it would be interesting to see what happens if you take it back a decade or two!

            • Richard Mallett says:

              Will do. I will use the record from 1772, since people seem to agree that it is reliable from that time.

            • Richard Mallett says:

              For the Berkeley Earth Global Land average temperature (this is published month by month, so I had to multiply the trend by 12) :-
              1753-2013 +0.36 C/century
              1753-1850 -0.60 C/century
              1850-2013 +0.84 C/century
              Therefore, since 1850, the rate has increased by 1.44 C/century.

  3. Richard Mallett says:

    I don’t see the data that I thought I posted this afternoon, so here they are (possibly again) :-

    CET 1659-1850 +0.18 C/century; 1772-1850 -0.005 C/century; 1850-2013 +0.62 C/century.
    CET 1772-1840 -0.02 C/century; 1840-2013 +0.58 C/century.
    CET 1772-1830 C/century -0.11 C/century; 1830-2013 +0.53 C/century.

    A few comments :-

    The temperature :-
    cooled by 1.20 C from 1659 to 1698;
    warmed by 2.84 C from 1698 to 1733;
    cooled by 1.93 C from 1733 to 1755 (with an anomalously cold year of 1740 at 6.84 C compared to 9.20 C and 9.30 C in 1739 and 1741)
    then was much less variable after 1755, so I suspect that from 1659-1755 it was still recovering from the Little Ice Age.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s