"Will" wrote in message ...
================================================== ==================
This posting expresses the personal view and opinions of the author.
Something which everyone on this planet should be able to do.
================================================== ==================

Yes it is chaos since we cannot observe SSTs etc to the resolution required for
weather forecasting. Hence errors creep into the analysis which leads to chaos.
The atmosphere itself is also chaotic due to non-linear feedback processes.
Remember chaos can arise out of extremely *simple* equations which have
application in the macro world.

Will.


/snip/

Hi Will and all,

I agree, Will, that it's the errors in determining what's happening
NOW which generate the errors in the forecasting, and not our
understanding (or lack of it) of the atmosphere. We understand fluuid
dynamics, thermodynamics, etc very well, and when we know, as near as
damn-it, the current situation, such as in a lab experiment, we can
forecast what's going to happen (e.g. heating a confined gas with a
light piston in the box, and the expanding gas pushing the piston out
a forecastable distance).

Then look at the global situation - try telling the computer program
what every single atom/molecule of the atmosphere is doing now
(impossible anyway due to the Heisenberg Uncertainty principle), and
then getting it to accurately forecast 5-10 days, and more ahead.

Even if people have all these other methods, such as for example,
pattern-matching, they must still have to match the current situation
EXACTLY with an historical one (bearing in mind that an analysis is
based on computer output as well as observations) before making a
forecast.

The atmosphere isn't unpredicatable per se, it's just chaotic, as Will
says, and even the smallest initial error will cause major variations
in forecasts.


cheers,

Paul K.