September 28, 2008, 5:39 am News


The most dishonest statement in the IPCC "Climate Change 2007 is as follows:
“A common confusion between weather and climate arises when scientists are asked how they can predict climate 50 years from now when they cannot predict the weather a few weeks from now. The chaotic nature of weather makes it unpredictable beyond a few days. Projecting changes in climate (i.e., long-term average weather) due to changes in atmospheric composition or other factors is a very different and much more manageable issue."

The Greek philosophers laid down the basis for modern scientific thought by insisting that absolute truth is impossible. Science can only discover information about the world that has a degree of inaccuracy. Improvements in science may lead to less accuracy, but they can never eliminate it altogether.

It is common knowledge that however many times we try to measure any quantity we always get a different answer. In the 19th century it was usual to give the "range" of results obtained by different observers. Although there may have been opinions on who might be the most accurate, there was no satisfactory way of deciding which, if any, of the figures was preferable.

The Scottish philosopher, David Hume (1711-76), suggested an improvement. He showed that induction was unreliable. If some phenomenon is known to occur frequently, there is no certainty that it will occur again, only a probability that it might.

In the 1930s a group of scientists, notably R A Fisher, suggested  that probability could be modelled by mathematics . Several different formulae were developed, based on different sets of assumptions. If a particular situation complied approximately with one of these sets of assumptions, then the mathematical formula could be used to assess the best estimate of a quantity and the extent of its variability.

Scientific results nowadays always use one or other of these models  to claim the value and accuracy of their results. Or should I say that they almost invariably choose only one of these models, the Gaussian distribution, or "Bell curve" because it is the simplest, and its calculations are available on every computer spreadsheet and "scientific" calculator. There is widespread ignorance, however, of the necessary conditions before this model should be used, or its results believed. Amongst these conditions are: that the sample should be completely representative, that the results should be symmetrical about a "mean" value, and there should be a sufficiently large number of observations where the observations have been made in identical conditions.

We are familiar with the example of political opinion polls which adopt the Gaussian model to forecast the most likely result of an election and its reliability, which is usually quoted as one part in 20 accuracy (the “margin of error”). The assumptions needed to justify these results are not always met, but at least we find out eventually if they are wrong.

The IPCC is riddled with examples of the use of statistical models for which the necessary assumptions are ignored. Representative samples are rare, symmetry is often absent and sufficient observations carried out under identical conditions are usually non existent.

"Chaos" theory is attributed to Edward N Lorenz (1917- 2008). Here is the beginning of his "Wikipedia" biography

"Lorenz built a mathematical model
of the way air moves around in the atmosphere .
As Lorenz studied weather patterns
he began to realize that they did not always change as predicted. Minute variations in the initial values of variables in his twelve variable computer
weather model (c. 1960)
would result in grossly divergent weather patterns
This sensitive dependence on initial conditions came to be known as the butterfly effect .
[3 "

It is, unfortunately, quite often, that a scientist finds that his model does not work. We usually try to keep it quiet, rather than build an entire career on this unfortunate occurrence, like Lorenz has done.

His crucial paper was "Deterministic Nonperiodic Flow" 1963 Journal Of  the Atmospheric Sciences Volume 20 pages 130-141. He found that existing mathematics did not accurately predict the weather, and that
"prediction of the sufficiently distant future is impossible by any known method  unless the present conditions are known exactly. In view of the inevitable inaccuracy and incompleteness of weather observations, precise, very long range forecasting would seem to be non existent"

Note  the qualifications. "Any known method".
Has the science of mathematics come to a halt? Is it impossible that we could actually discover a successful method which is currently unknown?
Then , what is "exactly". We all know that nothing can be known "exactly", so perhaps we cannot predict the weather "exactly". Why does that matter?

Then, can we live with "imprecise" or "not very long range" forecasting which he seems to agree is possible.

 He spoils his argument completely when he next says
"There remains the very important question as to how long is ‘very long range’. Our results do not give the answer for the atmosphere, conceivably it could be a few days or a few centuries"

 So "very long range" might be "a few days". !!!

Lorenz has done little more than confirm the ancient Greeks, that all measurements and forecasts are bound to be inaccurate. He is unable to say how much, for any example.

The weather is a complex system and weather forecasters have been struggling with its complexity in order to provide useful forecasts ever since the system was established in the early 19th century. Most of us would be agreed that they are making progress, and that more progress is likely.

Followers of Lorenz are prone to misinterpret his work to imply that there is a limit to future progress in forecasting the weather. Nowhere in Lorenz's own work does he say any such thing.

The above quotation from the IPCC, while admitting the complexity and the difficulty of forecasting weather, have fooled so many people in claiming that by merely changing the word "weather" to "climate", suddenly, all the "chaos" disappears and they can confidently tell us with high probability, what the climate is going to be like 100 years from now. They have persuaded us to make drastic changes in our current lifestyle in the belief that we must act now to avoid their confidently predicted future. They are always careful, of course, never to make a prediction sufficiently far ahead that somebody will be able to check whether it is right.


Vincent Gray
75 Silverstream Road
Crofton Downs
Wellington 6035
New Zealand
Phone/fax (0064) 4 973 5939
"The desire to save humanity is
always a false front for the urge to rule it"
H L Mencken

Next Post Previous Post