At New Scientist I found this: Climate Myths: Chaotic Systems Are Not Predictable (May 16, 2007) by Michael Brooks
In it he starts out with
“While weather and to some extent climate are chaotic systems, that does not mean that either are entirely unpredictable…” followed by an assertion of a demonstration whose link is a dead link.
I agree with the statement that – as science’s capacity currently stands – weather is chaotic. And yet, within that chaos, our scientists do a great job of dealing with weather, on its short time scale, in a manner that is both quite accurate and real-world functional. It is functional on the short scale to the pint of almost being a branch of engineering instead of science, per se. Engineering is what is called “applied science.” To “apply” a specific science, the mechanics, chemistry, materials science, electronics, and physics of it need to be known to within close approximations and then well defined for producers of products and designers, so that one group knows what to provide and the other knows what they will be dealing with and be able to produce a next-level product with characteristics known to within close parameters.
In other words, once a process or material passes to the point of engineering it is very nearly the exact opposite of a chaotic system.
Climate? If weather is chaotic and we can deal with it to the precision we do, then climate – being an averaged weather over at least 11,000 days – cannot be asserted to be chaotic, not to the level that nothing can be predicted.
In engineering, “predicted” is not used. One would say “expected”, not predicted. It seems that, in climate, researchers throw up their hands and claim “Chaos! Therefore our hands are tied!”
But if the short-term is handle-able, then it is not really chaotic. And if the short-term wild fluctuations can be mostly predicted, and those can be fed back into the analysis for >11,000 day periods and averaged then the degree of chaos in the system should be much less for the climate than for weather. After all, averaging waters down extremes, both high and low, so with the lessened variability the chaos MUST be lower.
“Climate scientists sometimes refer to the effects of chaos as intrinsic or unforced variability: the unpredictable changes that arise from the dynamic interactions between the oceans and atmosphere rather than being a result of “forcings” such as changes in solar irradiance or greenhouse gases.”
This is gobbledegook obfuscation and untrue, to boot – and on more than one front.
First of all, he knows full well that the oceans and atmosphere are not the only parts of the system. The land is also part of the system, since it absorbs solar energy and transfers it to the atmosphere, as do the oceans.
But being NOT first-order elements in the system, the “dynamics” between the ocean and atmosphere (and land) are internal forcings. THAT is what makes it dynamic, that elements that are AFFECTED then turn around and affect other elements. He cannot argue this by saying, “Well the sun’s energy is an external forcing, and so are greenhouse gases.
Why not? Two reasons: MOST greenhouse gsaes are themselves elements within the system (e.g., natural CO2, natural methane, and most of all water vapor), AND the Sun cannot be separated from the rest of the system; without it there IS no system, no weather, no climate.
He argues that the dynamic interactions between oceans and atmosphere are separate from the forcings within the system. “Every knows” that the biggest factor in our climate is the energy arriving from the Sun. The SUn is what gives the atmosphere and the oceans theirTherefore to claim that the dynamic interactions between ocean and atmosphere are causing the “chaotic effects” is
“rather than” arriving solar energy “or” only ONE other factor that might influence climate. NO FORCING can influence climate without going through WEATHER, i.e., without affecting the short term. It is IMPOSSIBLE to influence long term effects without affecting short term effects.
He is also claiming that only the “changes” in forcings and the “variability” of interactions makes for climate. Not so. If all the forcings and variabilities were always static, we would still have climate, at some fixed level.
Weather is the interactions of internal elements, which, however, DO continue to change. They do so because the other elements around them are themselves changing. These other elements, in turn are changed by the changing elements around them (some of those first ones just mentioned).
It makes for a very COMPLEX system and very COMPOUND system.
But is that system really chaotic – or is that just an excuse? Do they call it chaotic because then their failure to understand it and predict with it on the long term scale can be looked at as “Well, it’s just too damned HARD”?
The VERY odd part of this is something called reduction. In science reduction is the philosophy – approach – of breaking compound things into their smallest component parts, and that, by studying those individual parts, they can figure out what is going on with the whole. Reduction can be likened to having a machine and taking it apart, all the way down to the nuts and bolts, and then by determining the functions of all the parts they can discover the way the machine works. It makes sense, and reductionism works – they understand LOT by utilizing that approach.
But reductionism has not succeeded in some areas of study. I won’t go into others, but in climate there are so many bits of data – and sometimes the data are uncertain and/or sporadic – that the sheer volume of it all is daunting and more. In climate reductionism has not brought success. At present there are glimmers of understanding, but with the quantity of and the level of completeness of the “parts” conclusions have not been reliable. Code written for computer climate models have failed to be able to “hindcast” – they cannot reconstruct the recent (known) past. They also cannot so far predict into the future with any solid reliability. For example, the global temperatures have been essentially flat since about 1996, even while the CO2 being added by humans has continued to rise at an almost constant rate. This divergence in all likelihood suggests that somewhere in the computer code and the formulae behind the code there are either:
- coding errors or
- invalid formulae or
- invalid assumptions about constants or
- missing factors
- irrelevant factors included that shouldn’t be there
Now the question I have is this:
Why does averaged-out chaotic evidence (data) make for a LESS reliable predictability than non-averaged-out (more wildly swinging) short-term evidence (data)?
It seems to me that climate should be substantially less chaotic, if for no other reason than that it is smoothed-out weather.
And in the first and last analysis, isn’t that what climate IS – smoothed-out weather?
And one more thing, before I go. . . (shades of Columbo) . . .
I think that of all the tings that science studies with it, reductionism SHOULD be able to give us a handle on climate. After all, reductionism looks at the smallest elements making up a while, and by studying those elements, shouldn’t reductionism be able to look at all the vast number of individual sea water molecules and atmospheric molecules? And then to discover how the individual parts interact/function?Isn’t that what reductionism DOES? Is it a case of not enough computer power, then? After all, each molecule can only have so many ways it can be affected. With so few, wouldn’t it NOT be chaotic? Compound YES. Complex YES. Chaotic? If so, in what way?
So, is the failure so far a matter of not enough brain power (in humans and computers)? Or a lack of understanding of the relevent causes and effects going on? Or a lack of adequate evidence to plug in?
I suspect that we are too new at this climte study thing, and haven’t yet glimpsed all the factors that are involved, which means we re drawing conclusions prematurely – leading us to be using (so far) inadequate formulae, code and assumptions.
I don’t thik this is an insurmontable problem – just that we need the science to mature, by identifying and quantifying every one of the reelevant cactors, and then using enough computers to churn the vast amounts of data with the proper formulae. Without the right formulae, any conclusions being spit out and onto our monitors and graphs are just not going to be correct.
Yes, they are currently spitting out such graphs, but we are also learning that the graphs – even though they look good – are off enough that the models are not matching our real reality. Ergo, they aren’t correct.
I think the climateologists just need do more due diligence – keep plugging away to acquire the fundamental evidence to create better overall pictures of what is going on, which will allow them to derive better formulae – really correct formulae.