Terry Fields wrote:
The Natural Philosopher wrote:

I am a bit curuous as to where the data from the years 2010-2050 comes
from...

Roger suggested inverting the data from 1950 - 2005 and adding it on
from 2006, to give a mirror image of the datatset based about 2005.
The idea was to try to identify as soon as possible where any downturn
could be confirmed. The second tranche of graphs was that exercise.

Oh, I see. Using pictures rather than maths to understand the filter
characteristics.

Well you have completely reinvented the wheel, and discovered that
predicting the future needs a model. That can only be proven to be
correct (or not) once the future has happened.

The triumph of science is to pick models other than simple polynomials,
sine waves and the like, that reflect what seems to be the underlying
*mechanisms*. And in this case thats a definite plural.

Which is what climate change science is all about.

Just peering at the data and applying random simplistic filters wont
actually get you very far at all.

As I said, if what you are looking for is sine waves, what you will get
is sine waves.

I don't have to do all this: I did it years ago when working on software
for a digital sampling and storage oscilloscope: when we got up towards
the sampling frequency, we had various filters we could apply: the sine
interpolation filter attempted to fit a sine wave to the data points. It
always managed to do precisely that, irrespective of how sinusoidal the
original signal might have been, whilst a moving average more or less
wiped out te data at thse frequencies completely. It was better, in that
it wasn't subject to such wild extrapolations, but it was worse, in that
the bandwidth was totally lost.

In short, you cant do prediction with these sorts of brute force curve fits.

In terms of the actual climate change models so far proposed, there are
two very very conflicting mechanisms at leats: Pollution from CO2 and
methane, acting to raise temperatures, and pollutin from particulate
emissions like carbon soot from diesel and coal, acting in reverse.

Leaving aside amplification, this would tend to make periods of rapid
carbon based economic expansion, initially cooling in effect, as the
greater amount of atmospheric soots is a net cooler: once into
recession, these will wash out, leaving the longer term gaseous
pollutants well able to increase temperatures rapidly. So watch out for
a sharp upswing as the global recession bites deeper.

With current La Nina type conditions likely to come to an end, that
should be a triple whammy.