Friday, August 14, 2009

Signal and Noise

So much of what passes as the results of global warming is due to
aliasing. Aliasing occurs when a signal is sampled too frequently or not
frequently enough, when the samples are used to recreate the original
signal you get the wrong answer, such as a hockey stick fitted to a sine wave.

The real temperature signal is a sine wave, or several sine waves superimposed
on one another. A long term glaciation signal, a short term multi-decade oscillation
and probably several others in between. Unfortunately the most common sample
of temperature data is a human lifetime. We remember that when we were young,
it was hotter or colder, but now it is different, and it is worse. It's such a common
experience that it is enshrined in jokes ("back in my day we walked two miles in
the snow to school, uphill both ways").

Some examples of this effect:
We used to see several hurricanes per year way back in the 1930's, but now we see
The temperature at the airport used to be much colder than in the city,
now it's the same. We used to have zero sunspots at the solar minimum, now
there are several even though they are not visible without electronics.

All of these measurements that seem to be proving global warming could just
as easily be just proving that our measurements are now much better, or they
are much worse than they used to be, but they are not the same.

More measurements, more science and less demogogery are what's needed now.
Instead we're getting the inverse.

No comments: